Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The molecular pathology of lecithin : cholesterol acyltransferase deficiencuy Hill, John Stuart 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1994-894164.pdf [ 7.79MB ]
Metadata
JSON: 831-1.0088143.json
JSON-LD: 831-1.0088143-ld.json
RDF/XML (Pretty): 831-1.0088143-rdf.xml
RDF/JSON: 831-1.0088143-rdf.json
Turtle: 831-1.0088143-turtle.txt
N-Triples: 831-1.0088143-rdf-ntriples.txt
Original Record: 831-1.0088143-source.json
Full Text
831-1.0088143-fulltext.txt
Citation
831-1.0088143.ris

Full Text

THE MOLECULA R PATHOLOG Y O F LECITHIN:CHOLESTERO L ACYLTRANSFERASE DEFICIENC Y by JOHN STUAR T HIL L B.M.L.Sc, Th e Universit y o f British Columbia , 198 8 M.Sc, Th e Universit y o f Britis h Columbia , 199 0 A THESIS SUBMITTE D I N PARTIAL FULFILMEN T O F THE REQUIREMENT S FO R TH E DEGRE E O F DOCTOR O F PHILOSOPH Y in THE FACULT Y O F GRADUATE STUDIE S (Department o f Pathology ) We accep t thi s thesis a s conformin g to the require d standar d THE UNIVERSIT Y O F BRITIS H COLUMBI A March, 199 4 ©John Stuar t Hill , 199 4 In presentin g thi s thesi s i n partia l fulfilmen t o f th e requirement s fo r a n advance d degree a t th e Universit y o f Britis h Columbia , I  agre e tha t th e Librar y shal l mak e i t freely availabl e fo r referenc e an d study . I  furthe r agre e tha t permissio n fo r extensiv e copying o f thi s thesi s fo r scholarl y purpose s ma y b e grante d b y th e hea d o f m y department o r b y hi s o r he r representatives . I t i s understoo d tha t copyin g o r publication o f thi s thesi s fo r financia l gai n shal l no t b e allowe d withou t m y writte n permission. (Signature) Department o f The Universit y o f Britis h Columbi a Vancouver, Canad a Date H a c c K 4  ,  I  °f\ 4 -- ""' V •• • DE-6 (2/88 ) ABSTRACT Epidemiological studie s hav e indicate d tha t decrease d level s o f plasma HD L are associated wit h a n increased ris k o f coronary arter y disease . However , ther e ar e several rar e familia l disorder s o f HD L metabolis m i n which, despit e ver y lo w level s of HD L cholesterol , affecte d individual s d o no t appea r t o be a t an increase d ris k fo r premature atherosclerosis . Example s o f such disorder s includ e familia l LCA T deficiency an d fish eye disease whic h ar e rar e autosoma l recessiv e disease s associated wit h inherite d defect s withi n th e gen e codin g fo r lecithin:cholestero l acyltransferase (LCAT) . I n both cases , cornea l opacitie s an d a  severe HD L deficiency ar e characteristi c feature s o f the disease . However , unlik e fis h ey e disease, familia l LCA T deficienc y i s associated wit h sever e lipoprotei n abnormalitie s and additiona l clinica l symptom s includin g hemolyti c anemia , proteinuri a an d a progressive rena l insufficiency . Th e basi s fo r this segregatio n i s believed t o b e a result o f functional difference s associate d wit h the LCA T enzyme . The mai n purpos e o f this thesi s wa s t o define th e molecula r basi s fo r th e clinical an d biochemica l heterogeneit y observe d fo r differen t geneti c defect s o f LCAT. T o achiev e this , the expressio n o f recombinan t LCA T (rLCAT ) i n mammalia n cell cultur e wa s establishe d a s a model to analyse th e propertie s o f huma n plasm a LCAT. Subsequently , a  series o f natura l mutation s associate d wit h familia l LCA T deficiency an d fish ey e disease wer e re-create d an d expressed i n both monke y kidney COS- 1 an d bab y hamste r kidne y cel l lines . To determine th e functiona l significance o f each mutation , th e specifi c activit y o f different mutan t enzyme s wa s analysed usin g bot h syntheti c an d natura l substrates . A wide rang e o f functional abnormalitie s wer e identified : (i ) defects o f ii secretion, (ii ) los s o f activit y agains t HDL , (iii ) los s o f activity agains t al l lipoprotein s and (iv ) variabl e reactivitie s fo r al l lipoproteins . Fo r defect s whic h occu r i n the homozygous form , the propertie s o f mos t mutan t rLCAT s wer e consisten t wit h th e biochemical phenotyp e observe d i n the plasm a o f affected probands . Th e re -creation an d analysi s o f single mutation s associate d wit h compoun d heterozygou s genotypes reveale d tha t the product s o f each allel e pai r ha d differen t characteristics . In addition, th e result s suggeste d tha t the presenc e o f only a  single allel e codin g fo r a partiall y activ e LCA T i s sufficient t o maintai n norma l rate s o f cholestero l esterification. Fro m these biochemica l analyses , i t appears tha t LCA T ha s a t leas t two functionall y importan t domains , th e catalyti c cente r an d a  recognition sit e fo r HDL substrates . In conclusion, i t seems tha t the origina l familie s identifie d wit h familia l LCA T deficiency an d fish ey e disease represen t th e clinica l an d biochemica l extreme s o f a wide rang e o f defects o f LCAT function . Consequently , thes e disorder s shoul d no t be classified a s separate disease s bu t thought o f as par t o f a large r grou p o f LCA T deficiency syndromes . in TABLE O F CONTENT S Page ABSTRACT j j TABLE O F CONTENT S i v ABBREVIATIONS i x AMINO ACI D DESIGNATION S xi i LIST O F TABLE S xii i LIST O F FIGURE S x v ACKNOWLEDGEMENTS xvi i 1 INTRODUCTIO N 1 1.1 HD L Structur e an d Compositio n 2 1.2 HD L Metabolis m 5 1.2.1 Revers e Cholestero l Transpor t 6 1.2.2 HD L Remodellin g 1 1 1.2.3 Relationshi p Betwee n Apolipoprotei n Compositio n and the Metabolis m o f HD L 1 3 1.3 LCA T Biochemistr y 1 5 1.3.1 Physica l Propertie s o f LCAT 1 6 1.3.2 Th e Catalyti c Mechanis m o f LCA T 1 8 1.3.3 Substrate s o f LCA T 2 2 1.3.3.1 Monomeri c Substrate s 2 3 1.3.3.2 Aggregate d Substrate s 2 5 1.3.4 Activatio n o f the LCA T Reactio n 3 0 1.3.5 Measuremen t o f LCA T Activit y an d Cholestero l Esterification Rat e 3 1 1.4 Clinica l Finding s i n LCAT Deficienc y an d Fis h Ey e Diseas e 3 2 1.4.1 LCA T Deficienc y 3 3 1.4.1.1 Cornea l Change s 3 3 IV 1.4.1.2 Hematologica l Abnormalitie s 3 4 1.4.1.3 Rena l Diseas e 3 6 1.4.2 Fis h Ey e Diseas e 3 8 1.4.2.1 Cornea l Change s 3 8 1.5 Laborator y finding s i n LCA T Deficienc y an d Fis h Ey e Diseas e 3 9 1.5.1 Standar d Lipi d Profil e 3 9 1.5.2 Specialize d Laborator y Test s 3 9 1.5.3 Lipoprotei n Abnormalitie s i n Familia l LCA T Deficienc y 4 2 1.5.4 Lipoprotei n Abnormalitie s i n Fish Ey e Diseas e 4 4 1.6 Heterozygote s o f LCAT Deficienc y an d Fis h Ey e Diseas e 4 5 1.6.1 Laborator y Finding s i n LCA T Deficienc y 4 5 1.6.2 Laborator y Finding s i n Fis h Ey e Diseas e 4 6 1.7 Pathophysiolog y 4 6 1.7.1 LCA T Deficienc y 4 6 1.7.1.1 Lipoprotei n Abnormalitie s 4 6 1.7.1.2 Rena l Diseas e 4 9 1.7.1.3 Cornea l Change s 5 1 1.7.2 Fis h Ey e Diseas e 5 1 1.7.2.1 Lipoprotei n Abnormalitie s 5 1 1.8 Coronar y Arter y Diseas e i n HD L Deficienc y Syndrome s 5 3 1.9 Treatmen t o f Familia l LCA T Deficienc y 5 3 1.10 LCA T Gen e Structur e an d Expressio n 5 5 1.11 Gen e Defect s Associate d wit h LCA T Deficienc y an d Fish Ey e Diseas e 5 7 1.12 Mutagenesi s an d Expressio n o f Recombinant LCA T 5 8 1.13 Rational e 6 1 1.14 Specifi c Aim s 6 2 2 MATERIAL S AN D METHOD S 6 3 2.1 Material s 6 3 2.2 Molecula r Biolog y Method s 6 4 2.2.1 Growt h an d Transformation o f E. Coli  6 4 2.2.2 Purificatio n o f DNA 6 5 2.2.2.1 Smal l Scal e Plasmi d Preparatio n 6 5 2.2.2.2 Larg e Scal e Plasmi d Preparatio n 6 6 2.2.3 Oligonucleotide-directe d Mutagenesi s 6 7 2.2.4 Isolatio n o f cDNA Fragment s 6 8 2.2.5 DN A Sequenc e Analysi s 6 8 2.2.6 Constructio n o f Expression Vecto r 7 0 2.3 Eucaryoti c Cel l Culture 7 0 2.3.1 Transien t Transfectio n o f COS-1 Cell s 7 0 2.3.2 Stabl e Transfectio n o f BHK Cell s 7 1 2.3.2.1 Tunicamyci n Treatmen t o f BHK Cell s 7 2 2.3.3 Endogenou s Radiolabelin g o f Recombinan t LCA T 7 2 2.3.3.1 Immunoadsorptio n o f Recombinant LCA T 7 2 2.4 Protei n Analysis 7 3 2.4.1 Enzymati c Deglycosylatio n o f LCAT 7 3 2.4.2 SDS-Polyacrylamid e Ge l Electrophoresi s 7 3 vi 2.4.3 Immunoblo t Analysi s 7 4 2.4.4 Purificatio n o f Recombinan t LCA T 7 4 2.4.5 Quantitatio n o f LCA T Protei n 7 4 2.4.6 Preparatio n o f Plasma an d Lipoprotein s 7 5 2.4.7 Measuremen t o f LCA T Activit y 7 6 2.4.7.1 Exogenou s Substrate s 7 6 2.4.7.2 Endogenou s Substrate s 7 6 3 RESULT S 7 8 3.1 Expressio n o f Recombinan t LCA T i n BH K Cell s 7 8 3.1.1 Propertie s o f Recombinan t LCA T 7 8 3.1.2 Productio n an d Purificatio n o f Recombinan t LCA T 8 0 3.1.3 Enzymati c Deglycosylatio n o f Plasma LCA T an d Recombinant LCA T 8 3 3.1.4 Tunicamyci n Treatmen t o f BHK Cell s 8 4 3.1.4.1 Quantitatio n o f Secreted LCA T Protei n an d Activity 8 4 3.1.4.2 Immunoadsorptio n an d Western Blottin g o f Recombinant LCA T 8 5 3.2 Biochemica l Analyse s o f Selected Familie s wit h Inherite d Disorders o f LCAT Deficienc y 8 7 3.3 Transien t Transfectio n o f COS-1 Cell s 8 9 3.3.1 Endogenou s Radiolabelin g o f Wild Typ e an d Mutan t LCAT Specie s 8 9 3.3.1.1 Endoglycosidas e H  Digestion o f Intracellula r Recombinant LCA T 9 2 3.3.2 Establishmen t o f Enzyme Assa y Condition s fo r Wil d Type Recombinan t LCA T Secrete d b y Transfecte d COS-1 Cell s 9 3 vii 3.3 Quantitatio n o f Secreted LCA T Activit y an d Protei n fo r a Series o f Mutan t Recombinan t Enzyme s 9 5 3.4 Stabl e Transfectio n o f BHK Cell s 9 7 3.4.1 Endogenou s Radiolabelin g o f Wild Typ e an d Mutan t LCAT Specie s 9 7 3.4.2 Establishmen t o f Enzyme Assa y Condition s fo r Wild Type Recombinan t LCA T Secrete d b y Transfecte d BHK Cell s 9 9 3.4.3 Lipoprotei n Substrat e Specificit y o f Wild Typ e and Mutan t LCA T 10 1 4 DISCUSSIO N 10 4 4.1 Expressio n o f Recombinan t LCA T i n Mammalian Cel l Cultur e 10 4 4.2 Compariso n o f the I n Vivo an d I n Vitro Biochemical Characteristics fo r a  Series o f LCAT Mutation s 10 6 4.2.1 Pro10-»Le u 10 6 4.2.2 Ala93^Th r an d Arg158-^Cys 10 7 4.2.3 Thr123^ll e 11 0 4.2.4 Arg135^Tr p an d lle375-*Frameshif t 11 1 4.2.5 Leu209-»Pr o 11 2 4.2.6 Met252^Ly s an d Asn391-Ser 11 3 4.2.7 Thr123^ll e an d Thr347->Me t 11 4 4.2.8 Leu30 0 Deletio n 11 5 4.3 Geneti c an d Biochemica l Heterogeneit y o f LCA T Deficiency Syndrome s 11 6 4.4 Propose d Structure-Functio n Relationship s o f LCA T 12 0 4.5 Perspective s fo r Futur e Stud y 12 3 5 REFERENCE S 12 7 viii ABBREVIATIONS ACAT apo BHK cDNA CE CER CETP CHO COS dATP dCTP dGTP dTTP DEAE DHFR DMEM DMSO DNA DPPC EDTA Endo H FA acyl coenzym e A:cholestero l acyltransferas e apolipoprotein baby hamste r kidne y complementary DN A cholesteryl este r cholesterol esterificatio n rat e cholesteryl este r transfe r protei n Chinese hamste r ovar y transformed monke y kidne y cel l deoxyadenosine triphosphat e deoxycytidine triphosphat e deoxyguanosine triphosphat e deoxythymidine triphosphat e diethyiaminoethyl dihydrofolate reductas e Dulbecco's modifie d Eagl e mediu m dimethysulfoxide deoxyribonucleic aci d dipalmitoylphosphatidylcholine ethylenediamine tetra-aceti c aci d endoglycosidase H fatty aci d FBS FED HDL HEPES HL IDL LAT LCAT LDL LP LPL mRNA PC PCR PL PLTP PNP POPC rHDL rLCAT RNA RNase SDS TC fetal bovin e seru m fish ey e diseas e high densit y lipoprotei n N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic aci d hepatic lipas e intermediate densit y lipoprotei n lysolecithin acyltransferas e lecithin:cholesterol acyltransferas e low density lipoprotei n lipoprotein lipoprotein lipas e messenger RN A phosphatidylcholine polymerase chai n reactio n phospholipid phospholipid transfe r protei n p-nitrophenol palmitoyloleoylphosphatidylcholine reconstituted hig h densit y lipoprotei n recombinant lecithin:cholestero l acyltransferas e ribonucleic aci d ribonuclease sodium dodecylsulfat e total cholestero l X TG triglycerid e UC unesterifie d cholestero l VHDL very  hig h densit y lipoprotei n VLDL ver y lo w density lipoprotei n XI AMINO ACI D DESIGNATION S Amino Aci d Thre e Lette r Singl e Lette r Glycine Gl y G Alanine Al a A Threonine Th r T Serine Se r S Tyrosine Ty r Y Tryptophan Tr p W Aspartic aci d As p D Glutamic aci d Gl u E Glutamine Gi n Q Asparagine As n N Phenylalanine Ph e F Arginine Ar g R Leucine Le u L Proline Pr o P Methionine Me t M Histidine Hi s H Lysine Ly s K Valine Va l V Isoleucine H e I Cysteine Cy s C XII LIST O F TABLE S Description Pag e Protein, lipi d an d apolipoprotei n compositio n o f L p A-l, Lp A-l/A-ll , Lp A-IV, an d L p A-I/A-IV. 4 Percentage o f total recovere d weigh t fo r apolipoproteins A-l , A-ll, and A-IV. 5 Molecular substrate s o f LCAT. 2 4 A summar y o f the majo r clinica l an d biochemica l feature s o f homozygotes o f familial LCA T deficienc y an d FED . 3 3 Abnormalities o f composition an d function o f erythrocytes obtaine d from patient s wit h familia l LCA T deficiency . 3 6 Tissue localizatio n o f abnormal cel l types foun d i n familia l LCAT deficiency . 3 7 Laboratory finding s i n representative homozygote s fo r familia l LCAT deficienc y an d FED . 4 0 Laboratory finding s i n familial LCA T deficienc y an d FED : specialized tests . 4 1 Homology o f human LCA T wit h differen t specie s o f LCAT . 5 7 Natural mutation s o f the LCA T protein . 5 9 Oligonucleotides use d fo r site-directed mutagenesi s o f an LCA T cDNA . 6 8 Secretion o f rLCA T protei n an d activit y fro m BHK-LCA T cell s into serum-fre e medium . 7 9 Purification o f rLCAT secrete d b y transfected BH K cells . 8 3 Effect o f tunicamycin o n the secretio n o f rLCAT. 8 5 Structural change s an d biochemica l analyse s fo r a  group of molecula r defect s associate d wit h LCA T deficienc y syndromes . 8 8 Measurement o f LCAT protei n an d specifi c activit y fo r differen t mutant rLCA T specie s secrete d fro m transfecte d CO S cells . 9 7 XIII 17. Specifi c activit y o f wild type rLCA T secrete d fro m BH K cells fo r selecte d substrates . 10 2 18. Specifi c LCA T activit y o f a  series o f mutant rLCA T specie s for selected substrates . 10 3 XIV LIST O F FIGURE S Figure Descriptio n Pag e 1. Schemati c diagra m o f the proces s o f reverse cholestero l transport . 7 2. Schemati c diagra m o f HDL remodelling . 1 2 3. Schemati c diagra m o f the LCA T reaction . 1 6 4. Structura l feature s o f human LCAT . 1 7 5. Expressio n o f rLCAT . 7 9 6. Activatio n o f plasma LCA T an d rLCA T a s a  function o f apo A-l concentration . 8 0 7. Effec t o f Zn2+ io n concentration o n the secretio n o f rLCAT . 8 1 8. Elutio n profil e o f phenyl-Sepharose CL-4 B chromatography . 8 2 9. Compariso n o f the effect s o f enzymatic deglycosylatio n o f human plasm a LCA T an d rLCAT . 8 4 10. Effec t o f tunicamycin o n the processin g o f rLCAT . 8 6 11. Immunoadsorptio n o f wild typ e an d mutan t LCAT . 9 0 12. Immunoadsorptio n o f two mutan t LCA T species . 9 1 13. Endoglycosidas e H  treatment o f intracellula r wil d typ e recombinant LCAT . 9 2 14. Enzym e activit y o f wild type recombinan t LCA T secrete d b y COS-1 cell s fo r heat-inactivate d plasma . 9 4 15. Enzym e activit y o f wild typ e recombinan t LCA T secrete d b y COS-1 cell s fo r LDL . 9 5 16. Immunoadsorptio n o f intracellula r an d secreted form s o f a series o f mutan t LCA T species . 9 9 17. Enzym e activit y o f wild typ e recombinan t LCA T secrete d b y BH K cells fo r LDL . 10 0 18. Enzym e activit y o f wild typ e recombinan t LCA T secrete d b y BH K cells fo r VLDL. 10 1 xv 19. Compariso n o f the amin o aci d sequenc e o f a segment o f LCA T among si x differen t species . 10 9 20. Th e pathophysiolog y o f LCA T deficienc y syndromes . 11 8 21. Th e functiona l heterogeneit y o f LCAT mutations . 11 9 XVI ACKNOWLEDGEMENTS There hav e bee n man y peopl e wh o hav e contribute d t o the successfu l completion o f this thesis . I n particular, Dr . Roge r McLeod , Dr . Xingbo Wan g an d Carolyn Robbin s provide d invaluabl e assistanc e an d instructio n i n the molecula r biology an d cel l culture aspect s o f this work . Also , I  wish to thank Dr . Andras Lack o for providin g polyclona l antibodie s fo r human LCAT , Dr . John McLea n fo r the gif t o f an LCA T cDN A an d Dr . Ros s MacGillivra y fo r supplying th e pNU T expressio n vecto r used i n these studies . I  am indebte d t o Dr . Karmi n O  for bot h technica l an d intellectual contribution s a s well a s he r continuou s encouragemen t throughou t thes e studies. A  specia l thank s t o Dr . Lind a Perit z wh o mad e hersel f availabl e fo r helpfu l discussions o n a  wide rang e o f topics. I  wish t o offe r m y sincere appreciatio n t o Dr . Haydn Pritchar d who , a s m y researc h supervisor , provide d suppor t an d guidanc e a s well a s the opportunit y t o develop a s an independen t scientist . I  acknowledge al l members o f the Pritchar d laboratory : Lid a Adler, Elizabet h Donnachie , Sta n Shaw , Elizabeth Cramb , Janin e Senz , Dr . Francoi s Bowde n an d Dr . Bet h Allison fo r creating a  stimulating an d productiv e workin g atmosphere . Finally , I  acknowledge the persona l suppor t o f my family especiall y m y mothe r an d father wh o provide d a home fo r m y retreat . XVII 1 INTRODUCTIO N Atherosclerosis i s a slowly progressiv e diseas e characterize d b y the accumulation o f lipi d withi n th e intim a o f larg e elasti c an d muscula r arteries . Fo r many years , i t has bee n recognize d a s the primar y caus e o f myocardia l an d cerebral infarction . On e o f the majo r determinant s fo r the development o f atherosclerosis i s the concentratio n o f lipid , particularl y cholestero l an d it s esters, circulating i n plasma. Muc h effor t ha s bee n devote d t o the stud y o f the specifi c mechanisms o f lipi d an d lipoprotei n metabolism . Lipoprotein s ar e complexe s o f lipi d and protei n whic h functio n a s carriers o f lipi d i n the blood , interstitia l flui d an d lymph . These particle s consis t o f a  neutra l lipi d cor e surrounde d b y a  surface monolaye r containing phospholipids , unesterifie d cholestero l an d variou s apolipoproteins . Th e protein component s o f these particle s hav e bot h pola r an d nonpola r propertie s (amphipathic) an d to a  large exten t determin e th e structur e an d metabolis m o f different lipoproteins . A  variety o f functional role s hav e bee n attribute d t o apolipoproteins suc h a s cofactors fo r enzymes o f lipi d metabolism , facilitator s o f lipi d transfer reactions , an d a s ligand s fo r specifi c cel l membran e receptors . Traditionally, plasm a lipoprotein s ar e characterized accordin g t o thei r relativ e density: ver y lo w densit y lipoprotei n (VLDL , d<1.00 6 g/ml) , lo w densit y lipoprotei n (LDL, d=1.006-1.06 3 g/ml ) an d hig h densit y lipoprotei n (HDL , d=1.063-1.2 1 g/ml) . The synthesi s an d secretio n o f VLDL b y the live r i s recognize d a s the initia l even t o f a complex pathwa y describin g th e transpor t an d metabolis m o f endogenous lipids . The primar y functio n o f LDL , generated b y the metabolis m o f VLDL, involve s th e "forward" deliver y o f cholestero l fro m the live r t o periphera l tissues . Apolipoprotei n B (apo B) , located o n the surfac e o f LDL , i s recognized b y specific receptor s whic h 1 mediate th e uptak e an d distribution o f LDL cholestero l t o periphera l cell s (Brow n and Goldstein , 1986) . The associatio n o f elevated concentration s o f LD L wit h a n accelerated developmen t o f atherosclerosis ha s bee n wel l established b y clinica l and experimenta l observation s (Goldstei n an d Brown , 1987) . B y contrast , hig h level s of plasma HD L appea r t o be protective agains t coronar y arter y diseas e (Gordo n et al., 1977) . Additional evidenc e supportin g a  link betwee n HD L and atherogenesi s has bee n show n i n studies associatin g lo w o r absent HD L with prematur e atherosclerosis i n families wit h inherite d defect s o f HD L metabolis m (Schaefer , 1984). A  larg e bod y o f evidence ha s bee n accumulate d whic h indicat e tha t HD L ca n stimulate th e efflu x o f cholesterol fro m periphera l cell s (Badimo n et  al.,  1992) . Thi s "reverse" deliver y o f cholestero l ou t o f cells int o the plasm a i s considered t o b e the first ste p i n the transpor t o f cholesterol fro m periphera l tissue s bac k t o the live r fo r excretion. Unlik e th e forward transpor t o f cholesterol , th e specifi c mechanism s o f reverse cholestero l transpor t ar e no t wel l understood . Th e complexit y o f these reactions i s due i n par t to the considerabl e heterogeneit y associate d wit h HD L particles. Consequently , a  detailed understandin g o f the distinc t structural , functional, an d metaboli c propertie s o f different subtraction s o f HD L i s required. 1.1 HD L Structur e an d Compositio n By definition , HDL s ar e composed  o f lipoproteins wit h densitie s betwee n 1.063 an d 1.2 1 g/m l (Have l et  al.,  1955) . Within thi s densit y range , particle s wit h differences i n size (7.0-12. 0 nm) , hydrate d density , apolipoprotei n an d lipi d composition, an d the presenc e o f additiona l protei n constituent s hav e bee n identifie d (Fruchart et  al.,  1993) . A variety o f techniques includin g ultracentrifugation , 2 polyanion precipitation , polyacrylamid e gradien t ge l electrophoresis , agaros e electrophoresis, an d immunoaffinit y chromatograph y hav e bee n use d to analys e different subpopulation s o f HDL. Ultracentrifugation enable s th e separation  o f two majo r densit y classes , HDL 2 (1.063-1.125 g/ml ) an d HDL 3 (1.125-1.2 1 g/ml ) (Patsc h et  al.,  1980 ) whil e nondenaturing polyacrylamid e gradien t ge l electrophoresis differentiate s a t leas t fiv e subclasses, tw o o f lowe r densit y (HDL 2a and HDL 2b) an d three o f highe r densit y (HDL3a, HDL 3b, an d HDL 3c) (Blanch e et  al.,  1981) . Differentia l dextra n sulfate -magnesium chlorid e precipitatio n o f serum ha s als o bee n use d t o isolat e HDL 2 an d HDL3 subfraction s simila r t o those obtaine d b y ultracentrifugatio n (Barstei n et  al., 1989). Afte r agaros e ge l electrophoresis , th e majorit y o f HD L particle s migrat e wit h cr-mobility (cr-HDL ) wherea s onl y a  smal l fraction (approximatel y 5 % o f total HDL ) has preft-mobilit y (preR-HDL ) (Kunitak e et  al., 1985) . Furthe r analysi s o f these particles b y polyacrylamide gradien t electrophoresi s i n a second dimensio n separates preR-HD L int o three subfraction s (preG r, preG 2-, and preB 3-HDL) an d cr-HDL int o two (cr-HDL 2 and cr-HDL 3) (Castr o an d Fielding , 1988 ; Fieldin g et  al.,  1991) . More recently , th e classificatio n o f HDL s base d o n their apolipoprotei n composition ha s found favou r sinc e thes e particle s appea r t o hav e distinc t metabolic, functiona l an d clinica l significance . Sequentia l immunoaffinit y chromatography ha s bee n use d to isolat e fou r types o f particles : L p A-l, Lp A-l/A-ll . Lp A-I/A-IV, an d L p A-IV (Fruchar t et  al.,  1993) . Particle s wer e define d b y thei r major apolipoprotei n conten t suc h tha t L p A-l an d L p A-IV contai n a  single majo r apolipoprotein. Despit e a  more comple x protei n composition , L p A-l/A-ll an d L p A-l/A-IV hav e ap o A- l a s the majo r apolipoprotein . Tabl e 1  describes th e lipi d an d 3 apolipoprotein compositio n fo r each lipoprotei n particle . Al l particle s containe d a similar proportio n o f protein bu t showed difference s i n cholesterol an d triglycerid e (TG) content . L p A-l an d L p A-l/A-ll ha d highe r cholestero l bu t lowe r triglycerid e levels compare d t o L p A-I/A-IV an d L p A-IV. Table 1 . Protein, lipid , and apolipoprotei n compositio n o f Lp A-l, Lp A-l/A-ll, L p A-IV, an d L p A-I/A-IV particles . Adapte d fro m Fruchar t et  al.  (1993). Mass (% ) Proteins Lipids TC UC TG PL Apolipoproteins A-l A-l I A-IV C-lll Lp A-l 62.0±7.4 11.8±4.1 2.8±1.1 5.0±2.2 21.2±4.4 97.6±2.5 -2.1±2.4 0.3±0.3 Lp A-l/A-l l 65.2±7.3 11.5±3.9 1.6±1.1 3.8±1.8 19.5±4.3 67.4±13.3 31.4±13.5 1.0±1.9 1.2±0.2 Lp A-IV 70.2±20.1 4.2±3.5 0.4+0.2 9.4±6.8 16.2±10.5 -0.4±0.7 99.6±0.8 0.1±0.1 Lp A-I/A-IV 61.3±9.4 8.0±3.8 1.2±1.1 9.6±5.7 21.1±6.9 57.8±14.4 10.2±2.7 31.5±21.0 0.4±0.1 Lp, lipoprotein ; TC , total cholesterol ; UC , unesterified cholesterol ; TG , triglycerides; PL, phospholipids . Al l values ar e given a s mea n ±  standard deviation . Table 2  indicates th e percentag e o f total recovere d weigh t fo r apolipoprotein s A-l, A- l I, and A-IV for each lipoprotei n particle . Th e majorit y o f apo A-l an d ap o A-l I are found i n L p A-l/A-ll particle s an d greate r tha n 80 % of apo A-IV i s contained i n Lp A-IV particle s (Fruchar t et  al.,  1993) . 4 Table 2 . Percentage o f total recovere d weigh t fo r apolipoprotein s A-l , A-ll, and A-IV. Compile d dat a from Duverge r et  al. (1993) an d Fruchar t et  al.  (1993) . Apolipoproteins A-l A-ll A-IV Lp A-l 25% 0% 3% Lp A-l/A-l l 65% 98% 5% Lp A-IV 0% 0% 83% Lp A-I/A-I V 2% 2% 9% When isolate d particle s wer e subjecte d t o SDS-polyacrylamide ge l electrophoresis followe d b y immunoblotting , severa l additiona l protein s includin g lecithin:cholesterol acyltransferas e (LCAT) , cholestery l este r transfe r protei n (CETP) , and ap o D  were identifie d wit h eac h type o f isolated particle . I n the presenc e o f apolipoproteins, LCA T utilize s HD L phospholipi d an d unesterifie d cholestero l t o generate cholestery l este r wherea s CET P facilitate s th e exchang e o f HD L cholesteryl este r fo r triglyceride containe d withi n VLD L an d LD L particles . Th e measurement o f LCA T activit y associate d wit h eac h subtractio n o f HD L indicate d that L p A-IV particle s ha d the highes t activity , followe d b y L p A-I/A-IV an d L p A- l whereas L p A-l/A-ll particle s ha d the lowes t level s o f activity (Fruchar t et  al.,  1993) . In addition, CET P wa s mainl y associate d wit h L p A-l, Lp A-I/A-IV, an d L p A-IV bu t not with L p A-l/A-ll . 1.2 HD L Metabolis m Current evidenc e indicate s tha t the synthesi s o f plasma HD L i s derived fro m the live r an d intestine . Thes e newl y formed  o r "nascent " HD L appea r a s bot h smal l spherical an d discoidal  particle s whic h ar e protein-ric h bu t lipid-poo r an d exhibi t 5 preft-electrophoretic mobilit y o n agarose gels . The lipi d compositio n o f these particles consist s mainl y o f phospholipi d an d smal l amount s o f unesterifie d cholesterol an d triglyceride . Th e intestin e produce s nascen t HD L which contai n ap o A-l an d ap o A-IV bu t no t apo A-l l (Brewe r et  al.,  1990) . B y contrast , hepati c nascen t HDL consis t o f bot h L p A-l an d L p A-l/A-ll particle s (Fort e et  al.,  1989) . I n addition t o the direc t secretio n o f HD L particle s b y tissues, discoida l HD L particle s ric h i n phospholipid, ap o A- l an d ap o C s can b e generated extracellularl y fro m th e production o f excess apolipoprotei n an d phospholipi d surfac e component s create d by the actio n o f lipoprotein lipas e (LPL ) o n triglyceride-rich lipoprotein s (Tal l and Small, 1978) . 1.2.1 Revers e Cholestero l Transpor t Although al l cell s ar e capabl e o f cholestero l synthesis , onl y th e hepatocyt e has the abilit y t o degrade cholesterol . Thus , a  mechanism(s) i s required fo r the transport o f cholesterol fro m periphera l cell s t o the live r where cholestero l i s further metabolized o r excreted fro m th e bod y i n the form o f bil e an d bil e acids . Th e process o f "reverse cholestero l transport " was  firs t propose d b y Glomset (1968 ) an d since the n ha s bee n studie d b y numerou s investigators . Despit e intensiv e analysis , the characteristic s o f this pathway , particularl y it s putativ e anti-atherogeni c role , remain poorl y understood . Figur e 1  reflects th e mos t curren t concept s o f this hypothetical pathwa y whic h i s composed o f severa l distinc t elements . 1) Cholesterol  efflux  from  cells Several differen t model s o f cholestero l efflu x hav e bee n proposed : 1 ) The passive diffusion  o f cholesterol molecule s throug h th e aqueou s phas e surroundin g 6 the cel l (Phillip s era/. , 1980) ; 2 ) cholestero l translocatio n an d efflu x promote d b y the binding o f HDL to cel l surface receptor s (Oram , 1990) ; 3 ) acceptor retroendocytosi s (Schmitz et  al, 1985) ; 4 ) apolipoprotein-mediate d cholestero l efflu x (Rothbla t et  al. 1992). Th e presen t dat a sugges t tha t cholestero l efflu x fro m differen t cel l types ca n not b e explained b y one mechanism , bu t rathe r severa l mechanism s ma y operat e simultaneously dependin g o n the existing conditions . CELL MEMBRANE LIVE R Figure 1 . Schematic diagra m o f the proces s o f reverse cholestero l transport . The dar k arrow s indicat e th e flux o f unesterifie d cholestero l (UC ) an d th e ligh t gre y arrows indicat e th e transpor t o f cholestery l este r (CE) . CETP , Cholestery l este r transfer protein ; LCAT , lecithin:cholestero l acyltransferase . One o f the mos t efficien t acceptor s o f membrane cholestero l ha s bee n show n to b e smal l HD L enriched i n phospholipid an d protei n an d relativel y deplete d o f 7 cholesterol (Castr o an d Fielding , 1988) . The compositio n an d structur e o f these HDL, termed preB-HDL , ar e very simila r t o nascen t HD L synthesize d b y the live r o r intestine o r formed b y the lipolysi s o f triglyceride-rich lipoproteins . I n most cases , apo A- l i s the onl y apolipoprotei n presen t o n these particles . Studie s b y Segres t et al. (1992) predic t tha t the conformation o f the ap o A-l molecul e containe d withi n small disc s would enabl e a  hinged regio n o f the molecule t o extend int o the aqueous phase . I t has bee n postulate d tha t the extended portio n o f the ap o A- l molecule woul d interac t wit h specifi c lipi d domain s an d ancho r th e lipoprotei n t o the plasma membran e thu s facilitating th e diffusion o f cholestero l int o the accepto r lipoprotein (Rothbla t et  al.,  1992) . Th e accumulatio n o f unesterified cholestero l withi n the disc ma y modif y th e conformation o f the apo A-l t o regulate it s dissociation fro m the membrane . Suc h a  phenomenon woul d no t necessaril y b e restricted t o apo A-l, but coul d b e mediated  t o differing extent s b y amphipathic helice s o f othe r apolipoproteins. A  recen t stud y ha s demonstrated tha t th e siz e o f discoida l reconstituted HD L (rHDL) , determine d b y the numbe r o f apo A- l molecule s present , correlates positivel y wit h cholestero l efflu x bu t negativel y wit h cholestero l transfe r (Agnani an d Marcel , 1993) . A second mode l describes a n HDL receptor-mediate d cholestero l translocation an d efflu x (Oram , 1990) . I n this model , specifi c HD L apolipoprotein s (A-l an d A-IV) ar e recognize d b y a cel l surface receptor , whic h initiate s a  signal-transduction pathwa y stimulatin g th e translocation o f intracellula r cholestero l t o the membrane (Steinmet z et  al.,  1990 ; Mende z et  al.,  1991) . However , th e lo w affinit y and hig h capacit y o f these receptor s demonstrate d i n some system s ar e no t indicative o f classic recepto r metabolism . Althoug h th e balanc e o f evidence woul d 8 indicate tha t thi s pathwa y i s no t responsibl e fo r the bul k o f cholestero l efflux , it s specific quantitativ e rol e i n cholesterol homeostasi s remain s t o be determined. 2) Creation  of  a cholesterol  concentration  gradient Pulse-chase studie s usin g [ 3H]cholesterol-loaded fibroblast s incubate d wit h human plasm a hav e examine d th e transfe r an d esterification o f cell-derive d unesterified cholesterol . A s mentione d above , cellula r unesterifie d cholestero l i s first taken u p b y small discoidal , ap o A- l containin g HD L with preS-mobility . Fro m thes e preBrHDL, unesterifie d cholestero l i s rapidly transferre d t o large r preB 2-HDL an d subsequently t o preB 3-HDL t o be esterified i n the latte r (Castr o an d Fielding , 1988 ; Francone etal,  1989) . Immunochemica l studie s hav e indicate d tha t LCA T an d CETP comigrat e wit h preU 3-HDL bu t were absen t i n the two smalle r component s (Francone et  al.,  1989) . Th e rapi d an d specifi c transfe r o f unesterifie d cholestero l among differen t HDL s coul d no t b e the resul t o f simple diffusio n bu t ma y resul t fro m a precursor-produc t relationship . More recently , th e transfe r o f cellular unesterifie d cholestero l ha s bee n show n to continu e throug h cr-HD L and subsequentl y t o LD L (Huan g et  al.,  1993) . Mos t o f the LDL-derive d unesterifie d cholestero l i s redistributed t o cr-HDL to b e esterified b y LCAT (Miid a et  al.,  1990 ; Fieldin g et  al.,  1991) . I n addition, th e transfe r o f unesterified cholestero l fro m LD L to HD L i s LCA T dependen t sinc e inhibitio n o f the enzyme decrease d no t onl y th e productio n o f cholestery l este r bu t also th e remova l of unesterifie d cholestero l fro m LD L (Huan g et  al., 1993) . As a  result , cel l membranes appea r t o b e only a  minor dono r o f unesterifie d cholestero l t o the LCA T reaction. I n fact, mos t unesterifie d cholestero l esterifie d b y LCA T ha s bee n show n t o originate fro m ap o B-containin g particle s (Fieldin g an d Fielding , 1981 ; Park et  al., 9 1987; Cheun g an d Wolf , 1989 , Huan g et  al.,  1993) . These result s indicat e tha t a cholesterol gradien t betwee n cel l membranes an d HD L can b e established throug h at leas t two mechanisms : (i ) The esterification o f unesterified cholestero l b y LCA T i n HDL an d (ii ) the transfer o f unesterified cholestero l t o LD L vi a HD L followed b y it s uptake b y hepatic receptors . Thes e propose d mechanism s woul d provid e a n effective mean s o f revers e cholestero l transpor t b y maintaining a n LCAT -independent mechanis m o f cholesterol transpor t t o the live r an d secondly , throug h the redistributio n o f unesterifie d cholestero l t o a  large r HD L pool  (or-HDL) , cholesterol esterificatio n i s maximized. 3) Transport  of  cholesterol  to  the liver Once cholestero l i s esterified i n HDL i t can b e transferred t o LD L an d VLDL i n exchange fo r triglyceride throug h th e actio n o f CETP. VLDL-remnant s an d LD L ca n then b e cleared fro m th e plasm a b y hepati c ap o B/ E receptors . Thi s pathwa y i s particularly prominen t i n species wit h elevate d CET P activity . However , fo r specie s which lac k CETP , alternativ e pathway s ma y b e utilized . Fo r example , HD L containing ap o E  may b e recognized an d internalize d directl y b y apo E  receptors i n the ra t live r (Glas s et  al.,  1985) . Additiona l evidenc e ha s demonstrated tha t HD L ca n interact wit h cel l membrane s o f hepatocytes suc h tha t cholestery l ester s ca n b e transferred int o the cel l through th e actio n o f a specific accepto r protei n (Rinninge r et al.,  1988) . I n addition, i t has recentl y bee n demonstrate d i n perfused ra t liver s tha t hepatic lipas e (HL ) i s required fo r the hepati c uptake  o f HDL triglycerides an d cholesteryl ester s an d that the uptak e o f cholesteryl ester s was  no t dependent o n the hydrolysi s o f HDL triglyceride bu t on the hydrolysi s o f HD L phospholipi d (Kadowki et  al.,  1992) . 10 1.2.2 HD L Remodellin g The interconversio n o f HD L particle s consist s o f a  complex cyclin g o f lipid s and apolipoprotein s associate d wit h differen t size s an d shapes o f HD L particle s (Figure 2) . HDL-like particle s an d free apolipoprotein s (ap o A-l, apo A-ll , and ap o A-IV) hav e bee n identifie d i n interstitia l flui d (Lefevr e era/. , 1988) . Sinc e ap o A-l , apo A-ll an d ap o A-IV ar e no t commonly synthesize d b y periphera l tissues , thei r presence i n interstitia l flui d probabl y reflect s diffusio n o f free apolipoprotein s and/o r the filtration o f plasma HD L acros s th e endothelium . I n addition, th e observatio n tha t plasma level s o f preB rHDL wer e maintaine d onl y i f plasma was  incubate d wit h various cel l types suggest s tha t preG rHDL ca n b e formed b y the interactio n o f HD L components an d cel l membrane s (Miid a et  al.,  1992) . Indeed , HD L i n interstitia l flui d have prefi-mobilit y an d appea r t o b e effective acceptor s o f cellular cholesterol . However, i n the sam e manne r a s pref ^ and2 -HDL, the y d o no t appear t o b e good substrates fo r cholestero l esterificatio n (Won g et  al.,  1992) . The conversio n o f plasma HD L discs  int o spherica l HD L i s initiall y mediate d b y the esterificatio n o f surface unesterifie d cholestero l b y LCA T an d the migratio n o f cholestery l este r t o the cor e formin g HDL 3. The releas e o f unesterifie d cholestero l an d phospholipi d a s a consequence o f the actio n o f LP L o n triglyceride-rich lipoprotein s an d the continuou s action o f LCA T result s i n the transformation o f HDL 3 to large r HDL 2 particle s (Patsc h et al.,  1978) . I n contrast, i n vitro studie s hav e show n tha t through th e synergisti c effects o f CETP, HL , and VLDL, spherica l HD L los e cholestery l ester , triglyceride,and phospholipid , an d becom e progressivel y smalle r resultin g i n the dissociation o f apo A- l an d the formation o f discoidal , preG-HD L (Newnha m an d Barter, 1990 ; Cla y et  al.,  1991 , 1992; Kunitak e et  al.,  1992) . Th e observatio n tha t 11 the concentratio n o f plasma HDL 2 correlate s positivel y wit h th e posthepari n activit y of LP L (Patsc h et  al,,  1987 ) bu t negativel y wit h H L (Kuus i et  al.,  1980 ; Patsc h et  al., 1987) i s evidence tha t these transformation s als o occu r i n vivo. Thus , HD L appea r to cycle fro m discs- * HDL 3-* HDL2-^ HDL 3^ > discs enablin g th e continuou s movemen t of lipid s t o and from differen t lipoproteins . CELL MEMBRANE ENDOTHELIUM Figure 2 . Schematic diagra m o f HD L remodelling . UC , unesterifie d cholesterol ; PL phospholipid ; TG , triglyceride; CETP , cholestery l este r transfe r protein ; HL , hepatic lipase ; LPL , lipoprotei n lipase ; LCAT , lecithin.cholestero l acyltransferase ; FA, fatty acid . 12 Yet anothe r plasm a protei n ha s bee n show n t o modif y HD L size . The putativ e HDL conversio n factor , no w identifie d a s the phospholipi d transfe r protei n (PLTP) , can conver t HD L o f homogeneou s siz e int o two subpopulation s o f smaller an d large r particles (Jauhiaine n et  al.,  1993) . The mechanism s involve d o r the specifi c physiological rol e o f PLT P ha s no t bee n determine d bu t it s action i n plasma ma y contribute t o the suppl y o f phospholipi d t o HD L for the LCA T reaction . Therefore, change s i n the size , shape , lipi d an d apoprotei n compositio n o f HDL particle s ar e modifie d b y severa l plasm a factor s includin g LCAT , CETP , LPL , HL and PLTP . Th e specifi c influenc e o f these differen t component s o n the structur e and function o f different HDL s i s regulated b y other lipoprotein s an d possibl y additional plasm a constituents . Consequently , th e metabolis m o f HD L consist s o f a highly comple x serie s o f events involvin g th e transfe r an d exchang e o f apolipoproteins an d lipid s amon g cel l membrane s an d plasm a lipoproteins . 1.2.3 Relationshi p Betwee n Apolipoprotei n Compositio n an d th e Metabolism o f HDL As describe d above , HD L ca n b e divided int o differen t subpopulation s o n the basis o f thei r apolipoprotei n composition . I n most cases , studie s hav e focuse d o n two majo r HD L populations : thos e containin g ap o A- l bu t no t apo A-l l (L p A-l ) an d those containin g bot h ap o A- l an d ap o A-l l (L p A-l/A-ll) . Specifi c interes t i n these two types o f particles ha s bee n emphasize d b y a  clinica l stud y indicatin g tha t level s of Lp A-l, but no t L p A-l/A-ll, wer e predictiv e fo r coronary arter y diseas e (Puchoi s et al., 1987) . Also , plasm a L p A-l level s wer e highe r an d L p A-l/A-ll lowe r i n octogenarians (Lu c et  al.,  1991) . Thes e finding s hav e le d man y investigator s t o determine i f metabolic difference s exis t betwee n thes e tw o particles . Ther e ha s 13 been conflictin g dat a regardin g th e abilit y o f these particle s t o stimulate efflu x o f cholesterol fro m culture d cell s (Barki a et  al.,  1991 ; Johnson et  al.,  1991) . I t i s likely that differen t metaboli c characteristic s ma y exis t fo r differen t cel l types whic h coul d contribute t o these discrepancies . I n a recent repor t b y Fruchar t et  al.  (1993) , bot h Lp A-l an d L p A-l/A-ll promote d th e efflu x o f cholestero l fro m the membran e o f bovine aorti c endothelia l cell s t o a  similar extent . However , onl y L p A-l coul d stimulate th e efflu x o f intracellula r cholesterol . Analysis o f the i n vivo kinetic s o f L p A-l an d L p A-l/A-ll particle s ha s reveale d that L p A-l i s catabolized mor e rapidl y tha n L p A-l/A-ll (Rade r et  al.,  1991) . Thus , both the increase d abilit y t o mobilize intracellula r cholestero l an d it s faster rat e o f catabolism may  indicat e a n increased anti-atherogeni c potentia l fo r L p A-l. In several instances , th e metabolis m o f apo A-ll-containing particle s i n vitro has bee n show n t o be distinct fro m L p A-l. Fo r example, th e preferentia l conversio n of Lp A-I/A-II-HDL2, rathe r tha n L p A-I-HDL2, to HDL 3 was associate d wit h th e presence o f apo A-ll , which enhance d th e HL-mediate d lipolysi s o f these particle s (Mowri et  al.,  1992) . However , th e presenc e o f apo A-l l i n either discoida l rHD L o r spherical nativ e HDL 3 inhibite d thei r conversio n int o smalle r particle s mediate d b y CETP (Ry e et  al.,  1992 ; Lagrost , 1992) . Despit e thi s observation , i t has bee n demonstrated tha t bot h L p A-l an d L p A-l/A-ll particle s ar e capable o f generatin g preS-HDL (Henness y et  al.,  1993) . There i s now significan t evidence  tha t these tw o HD L subtraction s diffe r wit h respect t o a  number o f biological activitie s includin g cholestero l efflux , membran e binding, an d metabolism . However , give n th e variabilit y o f result s obtaine d s o far , further investigatio n o f the specifi c role s o f these differen t particle s i s required. 14 1.3 LCA T Biochemistr y Lecithin:cholesterol acyltransferas e (LCAT ) i s the enzym e responsibl e fo r the esterification o f cholestero l i n plasma. I t is unique i n that a  single polypeptid e sequence catalyze s sequentia l reactions . Thus , b y combining bot h phospholipas e A 2 and acyltransferas e activities , LCA T catalyze s th e transfe r o f an sn-2  fatty aci d fro m phosphatidylcholine t o the 3-hydroxy l grou p o f cholestero l formin g lysophosphatidylcholine an d cholestery l este r (Figur e 3) . LCA T i s synthesized i n the liver an d i s secreted int o the plasm a wher e i t associates primaril y wit h HDL , th e major sit e o f cholestero l esterificatio n (Fieldin g an d Fieldin g 1971 ; Chen an d Albers , 1982). As describe d above , LCA T create s a  cholesterol concentratio n gradien t between periphera l cell s an d HD L and thereby influence s th e concentration o f unesterified cholestero l an d cholestery l este r associate d wit h plasm a lipoprotein s and tissues. Th e significanc e o f it s rol e i n lipid metabolis m i s demonstrated b y the clinical an d biochemica l patholog y observe d i n inherited defect s o f the LCA T gen e (Norum et  al.,  1989) . The physiologica l importanc e o f LCAT , combine d wit h th e general interes t fo r enzymes o f lipoprotein metabolism , le d to work o n the purification, an d it s subsequent physica l an d chemica l characterization . 15 LECITHIN LYSOLECITHI N CHO - saturated  fatty  acid  CH O -  saturated  fatty  acid 2 I I CHO- unsaturated  fatty  acid  CHO H I O  I  O I I I I  u CHO - p  - cholin e CH O -  p  - cholin e OH O H Phospholipase A 2 - LCA T • Acyltransferase CHOLESTEROL CHOLESTERY L ESTE R unsaturated fatty acid Figure 3 . Schematic diagra m o f the LCAT reaction . 1.3.1 Physica l propertie s o f LCAT The primar y structur e o f human LCA T ha s been determine d b y the cloning and sequencing o f the LCA T cDN A (McLea n et  al., 1986a) . A 24 amino aci d hydrophobic leade r sequenc e precede s th e mature protein , which consist s o f a single polypeptid e o f 416 amin o acid s wit h a  calculated molecula r weigh t o f 47,090. LCAT i s glycosylated an d contains abou t 25 % carbohydrate increasin g it s molecular weight to about 65,00 0 t o 69,000 (Marcel , 1982) . Chemica l analysi s o f LCAT peptides hav e identifie d tw o disulfide bridge s betwee n Cys-50-Cys-7 5 an d Cys-313-Cys-356 whil e the two remaining cystein e residue s (Cys-3 1 an d Cys-184) hav e fre e sulfhydryl group s (Yan g et  al., 1987 ) (Figure 4). Analysis o f the amin o aci d sequence predict s fou r potentia l N-linke d glycosylatio n site s a t Asn residue s 20 , 84 , 272 and 384. Based o n the Chou-Fasman structura l algorithm , the proportion o f a-16 helix, R-shee t an d rando m coi l structure s i n LCA T ar e 21 , 24, and 55% , respectivel y (Yang et  al.,  1987 ) an d ar e simila r t o those determine d b y circular dichrois m (Chon g et al.,  1983) . Analysis o f the helica l hydrophobi c moment s indicate s severa l region s of possible amphipathi c structur e bu t onl y th e sequenc e betwee n Glu154-I_ys17 3 could potentiall y for m a n a-helix (McLea n era/. , 1986b) . Thus , th e lac k o f cr-helica l structures an d increase d hydrophobicit y associate d wit h LCA T distinguis h i t from th e lipid bindin g propertie s o f the solubl e apolipoproteins . Ser NK COOH Figure 4 . Structural feature s o f human LCAT . Th e relativ e position s o f the four re-linked glycosylatio n site s ar e indicate d abov e th e lin e whereas th e bridge d an d fre e cysteine residue s ar e indicate d belo w th e line . The catalyti c Ser-18 1 i s marked b y the vertica l box . LCAT i s known t o hav e multipl e isoelectri c point s rangin g fro m p i 3. 9 t o 4.4. After th e remova l o f sialic acid s b y neuraminidase digestion , thes e multipl e band s shift towar d highe r p H region s an d converge int o a  single ban d afte r extensiv e treatment (p i 5.2 ) (Do i and Nishida , 1983) . B y counting th e numbe r o f bands presen t in the isoelectri c focusin g gel , there appea r t o b e a total o f 1 6 sialic acids/molecul e of LCAT . Desialylatio n o f LCA T reduce s th e apparen t molecula r weigh t b y 3,00 0 and i s associated wit h a n increase i n LCA T activity , suggestin g tha t the remova l o f additional negativ e charge s enhance s th e LCAT-substrat e interactio n (Do i an d 17 Nishida, 1983) . Analysis o f the distributio n o f LCAT i n plasma ha s indicate d tha t abou t 90 % is associated wit h HD L bu t smal l amount s remai n boun d t o LD L (Che n an d Albers, 1982). Separatio n o f HD L particle s b y gel filtration indicate s tha t LCA T i s mor e commonly associate d wit h large r particle s (Che n an d Albers, 1982 ; Par k et  ai, 1987). After th e isolatio n o f Lp A-l an d L p A-l/A-ll particles , mos t o f plasma LCA T protein (70±15% ) was  detecte d i n L p A-l wherea s a  smaller portio n (16±7% ) was detected i n Lp A-l/A-ll (Cheun g et  ai,  1986) . 1.3.2 Th e Catalyti c Mechanis m o f LCA T Comparative amin o aci d sequenc e analyse s hav e indicate d tha t althoug h LCAT ha s littl e overal l sequenc e identit y wit h othe r lipases , i t does contai n a  Gly-X -Ser-X-Gly moti f characteristi c o f the activ e sit e o f severa l othe r lipase s (Brenner , 1988). Francon e an d Fieldin g (1991b ) hav e use d site-directe d mutagenesi s an d i n vitro expression t o investigat e th e rol e o f Ser-181 an d Ser-216 containe d withi n different lipas e consensu s sequence s o f human LCAT . Thi s stud y provide d stron g evidence tha t Ser-18 1 form s par t o f the activ e sit e o f LCA T a s the exchang e o f Ser -181 to eithe r threonine , glycin e o r alanine resulte d i n a complete los s o f LCA T activity. B y contrast , th e substitutio n o f Ser-216 wit h these sam e amin o acid s di d no t abolish LCA T activit y bu t resulte d i n significantly differen t specifi c activitie s observe d for each mutant . A s a  result , i t appears tha t Ser-21 6 doe s no t form par t o f the activ e site bu t retain s a n important , albei t indirec t role , i n the catalyti c reactio n o f LCAT . The three-dimensiona l structur e o f pancreati c lipas e indicate s tha t the activ e Se r contained withi n thi s consensu s sequenc e i s part o f a  catalytic Asp-His-Se r tria d 18 (Winkler et  al.,  1990) . I t is likely tha t a  similar conformatio n exist s fo r LCA T sinc e a monoclonal antibod y raise d agains t LCA T inhibit s th e enzym e activit y o f severa l phospholipases (Khali l et  al.,  1986) . Jauhiainen an d Dolphi n (1986 ) hav e demonstrate d tha t the chemica l modification o f Ser , Hi s and Cy s residue s inhibite d LCA T activity . Specifically , th e modification o f one serine residu e b y phenylmethanesulfony l fluorid e o r one histidin e residue b y diethyl pyrocarbonat e inhibite d bot h the phospholipas e an d acyltransferase activitie s o f LCAT . Fro m the result s obtaine d wit h sulfhydry l inhibitors, the y propose d tha t the two free cysteine s o f LCA T (Cys-3 1 an d Cys-184 ) were essentia l fo r activit y throug h thei r participatio n i n the formation o f thioeste r bonds prio r t o transfer o f the acy l grou p t o cholesterol . I n a subsequent study , Dolphin an d co-worker s demonstrate d tha t the free cystein e residue s o f LCA T ar e vicinal t o one anothe r an d within th e catalyti c sit e (Jauhiaine n et  al.,  1988) . However, th e result s o f i n vitro site-directe d mutagenesi s studie s demonstrate d tha t the substitutio n o f free Cy s residue s di d no t affec t LCA T reactivit y an d tha t th e action o f sulfhydry l inhibitor s was  attribute d t o steric inhibitio n (Francon e an d Fielding, 1991a ; Q u et  al.,  1993) . Site-directed mutagenesi s ha s als o bee n applie d t o produce LCA T specie s i n which th e fou r potentia l individua l attachmen t site s fo r N-linke d glycosylatio n chain s are destroyed (Francon e era/. , 1993) . Fro m these studies , i t was conclude d tha t elimination o f the glycosylatio n consensu s sequenc e a t position s 20 , 84 o r 38 4 ha d only mino r effect s o n either th e phospholipas e o r acyltransferase activitie s o f LCAT , whereas th e oligosaccharid e moiet y a t position 27 2 ha d an essential rol e i n acyltransferase activity . Studie s i n our laboratory , conducte d i n paralle l wit h thos e 19 reported i n this thesis , provid e a  more detailed analysi s o f the rol e o f N-linke d carbohydrate i n LCAT functio n ( O et  al.,  1993) . We definitivel y identifie d tha t al l four potential N-glycosylatio n site s i n LCAT ar e occupied an d tha t th e los s o f carbohydrate a t each sit e ha s diverse effect s o n enzyme function . Interestingly , th e loss o f carbohydrate a t position 38 4 was associate d wit h a  two-fold increas e i n enzyme specifi c activit y suggestin g tha t the oligosaccharid e a t this positio n ha s a n inhibitory effec t o n LCAT activity . I n addition, a  quadruple mutan t i n which al l four sites wer e eliminate d was  secrete d poorl y (10 % of wild type ) an d it s specifi c activit y was decrease d t o 5% o f wild type. Additional studie s wil l b e necessary t o defin e further th e specifi c rol e o f carbohydrate i n the reactivit y o f LCA T fo r nativ e substrates. Based o n the curren t knowledg e obtaine d fro m the physica l an d biochemica l analysis o f LCAT , th e catalyti c step s o f the enzyme reactio n ca n b e divided int o si x separate steps : (i ) reversibl e bindin g o f the enzyme t o the lipoprotei n surfac e (ii ) apolipoprotein activatio n (optimall y ap o A-l) facilitatin g acces s o f lipi d t o the activ e site (iii ) deacylatio n o f the phospholipi d substrat e an d releas e o f lysophosphatidylcholine (iv ) formation o f a Ser-O-acyl intermediat e (v ) subsequen t transfer o f the acy l chai n t o cholestero l (vi ) releas e o f cholestery l este r produc t int o the cor e o f the lipoprotein . Th e inhibitio n o f the LCA T reactio n b y lysophosphatidylcholine ha s bee n reporte d b y several investigator s an d i s thought t o be mediated  eithe r b y an end-product inhibitio n o r it s detergent-like propertie s (Fielding era/. , 1972 ; Smith an d Kuksis , 1980) . However , th e majorit y o f lysolecithi n is believed t o exis t a s a  complex wit h albumi n and , as a  result , i t may no t hav e a profound physiologica l role . 20 Although cholestero l i s the preferre d acy l acceptor , LCA T ha s also bee n shown t o carry ou t the acylatio n o f lysophosphatidylcholine t o refor m phosphatidylcholine (Subbaia h et  al.,  1982) . The catalyti c mechanis m fo r th e lysolecithin acyltransferas e reactio n (LAT ) appear s t o be the reversa l o f the phospholipase A 2 action o f LCAT . I n vivo, the increase d affinit y o f LD L fo r lysophosphatidylcholine confine s LA T activit y almos t exclusivel y t o the LD L fraction. In fact, the acylatio n o f lysophosphatidylcholin e ha s bee n show n t o b e dependent o n the presenc e o f LD L bu t no t ap o A-l (Subbaia h et  al.,  1982) . Specifically , ap o B present i n LDL was essentia l fo r enzyme activatio n sinc e trypsi n treatmen t o f LD L abolished it s abilit y t o activat e LA T (Subbaia h et  al.,  1985) . Furthe r studie s indicate d that th e specifi c conformatio n o f apo B  was affecte d b y the concentratio n o f LD L triglyceride which , i n turn, influence d th e magnitud e o f LA T activit y (Li u et  al.,  1992) . The LA T activit y o f LCA T was  als o demonstrate d i n proteoliposomes containin g eg g phosphatidylcholine an d radiolabele d lysophosphatidylcholine . Apolipoprotein s C- l and E  were respectively , 70 % and 40% as effective a s ap o A-l i n the activatio n o f LAT (Li u an d Subbaiah , 1993) . Thes e result s sugges t tha t i n addition t o end-produc t inhibition, lysophosphatidylcholin e ma y compet e wit h unesterifie d cholestero l a s an acyl acceptor . Th e physiologica l significanc e o f the LA T reaction , however , remain s to b e determined. It has als o bee n reporte d tha t LCA T i s capable o f hydrolysi s o f cholestery l ester t o produc e unesterifie d cholestero l (Sorci-Thoma s et  al.,  1990) . However , th e rate o f releas e o f unesterifie d cholestero l wa s onl y abou t 10 % of the forwar d rat e o f esterification. I n addition, ther e was  n o evidence tha t the fatt y aci d obtaine d fro m th e cholesteryl este r coul d b e transferred t o lysophosphatidylcholin e t o form 21 phosphatidylcholine. Thus , the two "forward " activities o f the LCA T reactio n ar e eac h independently reversibl e bu t no t i n succession. 1.3.3 Substrate s o f LCA T Due t o the heterogenou s natur e o f plasma HDL , i t i s difficul t t o study th e different bindin g an d enzyme reactivit y o f LCAT fo r differen t HD L subtractions . A s a result, i t has bee n necessar y t o synthesize chemicall y an d physicall y define d substrates t o investigat e th e propertie s o f the LCA T reaction . The firs t syntheti c substrate s t o be used were unilamella r vesicle s containin g phosphatidylcholine, cholestero l an d exogenous apolipoprotein s whic h forme d particles terme d "proteoliposomes " (H o an d Nichols , 1971) . Subsequently , th e sodium cholat e dialysi s metho d was  develope d t o create discoidal  complexe s o f similar compositio n (Mat z an d Jonas, 1982) . The stabilit y an d uniformit y o f these complexes ha s enabled th e investigatio n o f the substrat e specificity , apolipoprotei n activation an d kineti c propertie s o f LCAT . The know n molecula r substrate s o f LCAT an d thei r physica l stat e i n aqueou s solutions ar e liste d i n Table 3 . Although non-physiological , th e monomeri c acy l donors ca n b e helpfu l i n the analysi s o f the enzyme kinetic s an d mechanis m i n the absence o f a lipid/wate r interface . Th e physiologica l substrate s o f LCAT exis t i n aggregated for m eithe r i n native lipoproteins , syntheti c lipi d vesicle s o r reconstitute d HDL (rHDL ) (Jonas , 1986) . The chemica l an d physica l propertie s o f these differen t substrates determin e thei r relativ e reactivit y fo r LCAT . 22 1.3.3.1 Monomeri c Substrates Bonelli an d Jonas (1989 ) hav e use d p-nitropheno l (PNP ) ester s t o investigat e the effect s o f substrate chai n length , io n concentration an d apolipoprotei n composition o n the enzym e kinetic s o f LCA T i n solution. However , i t should b e noted tha t thes e studie s asses s onl y th e phospholipas e activit y o f LCAT . The y reported tha t wit h increasin g chai n length , bot h K m and V max decreased, bu t the rati o of Vmax/Km increased . O n the on e hand , from thes e studies , i t was conclude d tha t th e binding o f molecules wit h lon g acy l chain s t o the activ e sit e o f LCA T i s determine d by hydrophobi c interaction s suc h tha t thei r affinit y increase s wit h increasin g chai n length. O n the othe r hand , the decreasing V max values ma y aris e from a  restricte d mobility o f the lon g chain substrate s i n the activ e sit e that coul d reduc e th e frequency o f successfu l binding . 23 Table 3 . Molecula r substrate s o f LCAT . Adapted fro m Jona s (1991) . Substrates Acyl donor s Phospholipids Cholesteryl ester s Arachidonyl Co A PNP acy l ester s DPybcPC Acyl acceptor s Sterols (cholesterol ) Lysophospholipids Alcohols (lon g chain ) Water Physical Stat e Aggregate Aggregate Micelle Monomer Monomer Aggregate Aggregate Aggregate Monomer Reference Aron et  al.,  197 8 Sorci-Thomas et  al.,  199 0 Jauhiainen an d Dolphin , 198 6 Bonelli an d Jonas , 198 9 Bonelli an d Jonas , 199 2 Kitabatake et  al.,  197 9 Subbaiah et  al.,  198 2 Kitabatake et  al.,  197 9 Aron et  al.,  197 8 PNP, p-nitrophenol ; DPybPC , 1,2-bis[4-(1-pyreno)-butanoyl]-sn-glycero-3 -phosphocholine The influenc e o f surfactants o n LCA T reactivit y wit h wate r solubl e substrate s has als o bee n studied . Ioni c detergent s i n the concentratio n rang e o f 1CT 4 to 10 3M inhibited LCAT , wherea s tw o nonioni c detergent s activate d th e enzym e i n a simila r concentration rang e (Bonell i an d Jonas , 1993) . Fro m the analysi s o f enzyme kinetic s and the pattern s o f inhibition , i t was propose d tha t SD S bind s cooperativel y t o LCA T 24 and cause s inhibitor y structura l changes , wherea s laurat e an d cholat e appea r t o bind to specifi c site s eithe r a t the activ e sit e o r negativ e effecto r site s elsewhere . B y contrast, nonioni c detergent s ma y create a n interfacia l activatio n o f the phosphoiipase reactio n nea r thei r critica l micella r concentrations . Also , i n contrast t o aggregated substrates , th e presenc e o f anions an d free apolipoprotein s ha d n o specific effec t o n LCA T activit y i n solution (Bonell i an d Jonas , 1992) . Therefore , these studie s woul d see m t o indicat e tha t the activatio n o f LCA T b y apolipoprotein s and it s modulatio n b y anions ar e associate d wit h interaction s a t the lipid/wate r interface, an d ar e no t a  resul t o f direct effect s o n the enzyme . 1.3.3.2 Aggregate d Substrate s As mentione d above , th e firs t step s o f the LCA T reactio n fo r aggregate d substrates includ e interactio n o f the enzym e wit h a  lipid interface , followe d b y binding o f a  lipi d substrat e t o the activ e site . The characteristic s o f these initia l events ar e influence d t o varying degree s b y several interrelate d component s including lipi d an d apolipoprotei n composition , apolipoprotei n conformation , an d th e size an d shap e o f the lipoprotei n substrate . Apo A- l i s recognized a s the mos t poten t activato r o f the LCA T reactio n (Fielding et  al.,  1972) . However , othe r apolipoprotein s an d eve n syntheti c amphipathic peptide s ar e know n t o activat e LCA T t o varying degree s (Jonas , 1991) . This topi c wil l b e discussed i n detail  i n a late r section . The compositio n o f phospholipids hav e multipl e effect s o n LCA T reactivit y which include : 1 ) the relativ e fluidity , lipi d packing , an d hydratio n o f the lipi d interface t o which LCA T binds ; 2 ) the conformatio n o f apolipoproteins i n the 25 membrane; an d 3 ) their interactio n wit h th e activ e site . It has bee n establishe d tha t phospholipid s whic h increas e membran e fluidit y also enhanc e cholestero l esterificatio n b y LCA T (Pownal l et  al.,  1985) . Fo r syntheti c dispersions o f phosphatidylcholine  an d cholesterol , maxima l enzym e activit y i s observed a t a  phosphatidylcholine/cholestero l rati o o f 4:1 i n the presenc e o f apo A- l (Fielding et  al.,  1972) . The creatio n o f rHDL particle s containin g eithe r palmitoyloleoylphosphatidylcholine (POPC ) o r dipalmitoylphosphatidylcholin e (DPPC) exhibi t distinc t an d characteristi c siz e distributions . Consequently , differences i n both the numbe r an d conformation o f apo A- l molecule s boun d t o the surface wil l affec t activatio n o f the LCA T reaction . Whe n usin g rHD L substrates , phosphatidylcholines wit h tw o saturate d chain s containin g 1 6 carbons o r les s wer e excellent substrate s fo r LCA T bu t poo r LCA T reactivit y was  observe d whe n bot h chains wer e longe r (Pownal l et  al.,  1987) . However , th e compositio n o f surroundin g phosphatidylcholines ca n influenc e th e reactivit y fo r a  single particula r phosphatidylcholine type , an d thus, the acy l chai n specificit y o f LCA T observe d fo r artificial substrate s i s difficult t o relat e t o the function o f LCA T i n vivo (Jona s et  al., 1986). I n plasma, th e phosphatidylcholin e preferenc e i s largely determine d b y the availability o f the phosphatidylcholin e specie s i n the plasm a lipoprotein s (Subbaia h and Monshizadegan , 1988) . These studie s indicate d tha t i n native plasma , LCA T prefers 16:0>18:1>18: 0 acy l group s a t the 1-positio n an d 18:2>18:1>22:6>20: 4 acy l groups a t the 2-position . Analysi s o f the positiona l specificit y o f huma n LCA T ha s indicated tha t greate r tha n 90 % of the acy l group s use d fo r cholestery l este r synthesis ar e derived fro m th e sn-2  positio n o f mos t o f the naturall y occurrin g phosphatidylcholine substrate s (Subbaia h era/. , 1992) . However , abou t 75 % of the 26 cholesteryl este r specie s wer e derive d fro m th e sn- 1 positio n whe n phosphatidylcholines containin g 20: 4 o r 22:6 acy l chain s a t the sn- 2 positio n wer e used a s substrates . Thi s observation wa s no t influence d b y membrane fluidity , apolipoprotein activator , o r b y the unesterifie d cholesterol/phosphatidylcholin e rati o in the substrate . I t i s possible tha t the bulk y 20: 4 an d 22: 6 group s i n the phosphatidylcholine d o no t fi t efficiently int o the activ e sit e and , as a  result , th e sn- 1 acyl grou p which i s no t usuall y a  preferred substrate , i s more favourably oriented toward th e activ e site . However , a  finding tha t i s inconsisten t wit h thi s hypothesi s i s that, whil e the positiona l specificit y i s altered i n the presenc e o f 14:0-20: 4 phosphatidylcholine, 16:0-20: 4 phosphatidylcholine , an d 16:0-22: 6 phosphatidylcholine, i t i s altered t o a  much lesse r exten t i n the presenc e o f 18:0 -20:4 phosphatidylcholine . Th e reaso n fo r thi s exceptio n i s no t clear , bu t i t appears that 18: 0 acy l chain s ar e eve n les s preferre d tha n 20: 4 group s b y the enzym e (Subbaiah et  al.,  1992) . Particles o f different siz e hav e bee n show n t o hav e significantl y differen t reactivities wit h LCAT . I n general, smalle r HD L (HDL 3) ar e bette r substrate s fo r LCAT compare d t o large r HDL , particles (Fieldin g an d Fielding , 1971 ; Jahani an d Lacko, 1982) . Fo r syntheti c rHD L disc s containin g POPC , thes e difference s hav e been attribute d t o change s i n the conformatio n o f apo A- l (Jona s et  al.,  1989) . However, i n rHDL particle s prepare d wit h ap o A- l an d DPPC , th e conformatio n o f apo A- l was  quit e simila r i n particles wit h diameter s rangin g fro m 9. 7 t o 18. 6 nm , but a 20-fold differenc e i n LCA T reactivit y was  observe d betwee n th e smalles t an d largest particle s (Hefel e Wal d et  al.,  1990) . Thus , i n certain conditions , th e siz e o r surface curvatur e o f the particl e ma y independentl y influenc e it s abilit y t o interac t 27 with LCAT . In general, discoida l HD L ar e mor e reactiv e wit h LCA T whe n compare d t o native spherica l HD L (Marce l et  al.,  1980) . As mentione d above , change s i n apolipoprotein conformatio n and/o r lipi d packin g ma y b e responsible fo r thes e differences. Studie s performe d wit h syntheti c disc s indicat e tha t LCA T react s bes t with larg e discs  containin g 4  molecules o f apo A-l , followed b y progressivel y smalle r discs containin g 3  and 2  molecules o f apo A-l , respectively (Men g et  al.,  1993) . I n addition, LCA T activit y i s also proportiona l t o disc circumferenc e whic h ma y reflec t the bindin g o f more amphipathi c cr-helices/particl e increasin g th e stabilit y o f apo A- l and thus LCA T reactivit y (Men g et  al.,  1993) . I n addition, th e increase d abilit y o f synthetic discs  t o store cholestery l este r compare d t o vesicles o f a  simila r composition ma y explai n th e highe r reactivit y o f these particle s compare d t o spherical substrates . (Mat z an d Jonas , 1982) . However , Barte r et  al.  (1985) , reported a n invers e relationshi p betwee n plasm a HDL 3 particle s siz e an d substrat e reactivity tha t wa s independen t o f cholestery l este r content . Th e preparatio n o f spherical rHD L with differen t cholestery l este r t o triglyceride ratio s i n their core s indicated tha t triglycerid e correlate d negativel y an d cholestery l este r positivel y wit h the V max o f LCAT , wherea s th e apparen t K m remained unchange d (Spark s an d Pritchard, 1989) . However , th e surfac e compositio n o f the rHD L influence d enzym e kinetics suc h tha t th e K m was positivel y relate d t o protei n conten t an d particl e siz e but negativel y wit h phospholipi d an d unesterifie d cholestero l content . Fro m thes e studies, i t i s apparent tha t the surfac e an d cor e lipi d composition , an d the proportio n of lipi d t o apolipoprotei n affec t particl e siz e an d shape . I n turn, the siz e an d shap e o f the particl e influence s th e structur e o f the apolipoprotein . A s a  result , th e 28 interdependence o f these parameter s mus t b e considered whe n determinin g th e properties o f substrates fo r optima l LCA T reactivity . It has bee n reporte d tha t LCA T ca n als o interac t directl y wit h LD L to esterif y cholesterol (Barter , 1983 ; Barte r et  al.,  1984 ; Knippin g et  al.,  1986) . When th e formation o f cholestery l este r wa s analyse d i n incubated plasm a devoi d o f CET P activity, 73 % o f the esterifie d cholestero l wa s i n HDL , 25% i n LD L an d onl y 1 % in VLDL (Rajara m an d Barter , 1985) . Additional studie s usin g purifie d LCA T an d isolated lipoprotei n fraction s fro m bot h huma n an d pi g plasm a confirme d th e abilit y of LCA T t o esterify cholestero l i n LD L (Knippin g et  al.,  1986) . Th e remova l o f non-apo B  proteins fro m LD L reduce d bu t did no t eliminate thi s activity . Sinc e pi g plasm a lacks CET P activity , th e appearanc e o f cholestery l este r i n specific lipoprotein s ca n be attributed t o the direc t actio n o f LCAT. Afte r incubatio n o f pig plasm a a t 37°C , the greatest amoun t o f cholestery l este r (60-70% ) wa s consistentl y foun d i n the LD L fraction suggestin g tha t pi g LCA T ma y hav e a  preference fo r LDL , eve n i n the presence o f physiologica l concentration s o f HD L (Knippin g et  al.,  1987) . The substrat e specificit y o f LCAT fro m a  number o f different specie s hav e been compare d indicatin g differin g abilitie s t o transfer acy l groups dependin g o n their lengt h an d degre e o f a  saturation (Grov e an d Pownall , 1991) . As a  result , th e authors suggeste d tha t the conformatio n o f the activ e site s o f these enzyme s ar e not the same . However , th e us e o f synthetic substrate s i n this stud y may  no t b e representative o f their actio n i n vivo. O n the othe r hand , there i s in vivo evidenc e that ra t LCA T ha s a  greater specificit y fo r arachidonat e (20:4 ) acy l group s compare d to huma n LCA T (Subbaia h et  al.,  1992) . The purificatio n o f ra t LCA T ha s enable d the compariso n o f it s characteristics wit h it s huma n counterpar t (Furkaw a et  al., 29 1989). Th e affinit y o f the ra t enzyme t o ra t apo A-l-containing vesicle s was  stronge r when compare d t o the interactio n o f the huma n enzym e wit h huma n ap o A- l (Furkawa et  al.,  1992) . Ra t apo A- l was  abl e to activate huma n LCA T a t onl y 18 % of the rat e obtaine d wit h huma n ap o A-l . By contrast , th e rate s o f esterification wer e more tha n two-fold highe r fo r the huma n ap o A-l activatio n o f ra t LCA T compare d t o human LCAT . 1.3.5 Activatio n o f the LCA T Reactio n Initially, Fieldin g et  al.  (1972) demonstrate d tha t egg-phosphatidylcholin e an d cholesterol vesicle s require d ap o A- l t o activate th e LCA T reaction . Sinc e then , a number o f investigator s hav e show n othe r exchangeabl e apolipoprotein s als o activate LCA T bu t to a  lesse r exten t (Souta r et  al.,  1975 ; Steinmet z an d Utermann , 1985). Severa l studie s hav e attempte d t o localiz e th e domain(s ) withi n ap o A- l responsible fo r it s LCA T activatin g propertie s b y using syntheti c peptid e analogue s (Pownall et  al.,  1984 ; Anantharamaiah et  al.,  1990 ) an d monoclona l antibodies (Banka et  al.,  1991) . I n most cases , th e activatin g efficienc y o f peptides a s well a s other apolipoprotein s ha s a n intrinsi c uppe r limit , approximatel y 30 % of apo A- l (Anantharamaiah et  al.,  1990) . Therefore , th e disruption o f the lipid/wate r interfac e by amphipathic helixe s ma y b e necessary fo r LCA T activatio n bu t i t i s clearly no t sufficient. Ther e i s evidence indicatin g tha t a  specific regio n o f apo A- l (residue s 66 -120) i s associated wit h a  predominant LCA T activatin g domai n o f apo A-l . The identification o f apo A-l-specific monoclona l antibodies  whic h inhibi t LCA T activit y and the us e o f synthetic peptide s ha s defined th e regio n spannin g residue s 95-12 1 as critica l fo r activatio n o f the LCA T reactio n (Bank a et  al.,  1991) . Segres t an d 30 colleagues (1992 ) hav e propose d tha t a  unique positionin g o f Glu residue s o n the non-polar fac e o f two helixes , no t presen t i n the othe r apolipoproteins , i s responsibl e for the uniqu e activatin g abilitie s o f apo A-l . However , poin t mutation s i n this regio n do no t significantl y affec t activatio n (Jona s et  al.,  1991 ; Minnich et  al.,  1992) . Therefore, a  series o f amino acid s rathe r tha n individua l residue s appear s t o b e responsible fo r the ap o A- l activatio n o f LCAT . I n addition, i t i s technically very difficult t o assess th e relativ e reactivit y o f synthetic peptide s an d apolipoproteins . Thus, th e definitio n o f specific activatin g domain s o f apo A- l o r the mechanis m b y which suc h a  region coul d enhanc e LCA T reactivit y remain s elusive . Hopefully , site -directed mutagenesi s an d i n vitro expressio n o f recombinan t A- l wil l b e helpfu l i n this regard . 1.3.5 Measuremen t o f LCAT Activit y an d Cholestero l Esterificatio n Rat e Two method s ar e routinel y use d t o assess th e abilit y o f a  patient's plasm a t o esterify cholesterol . Firstly , th e measuremen t o f LCAT activity  i s determined i n the absence o f plasma constituent s i n order t o reflec t th e amoun t o f active enzym e i n the plasma . Therefore , i n the absenc e o f any defects , LCA T activit y shoul d b e proportional t o plasm a LCA T protei n levels . I n this assay , a n excess o f artificiall y prepared liposome s containin g know n amount s o f [ 3H]cholesterol, phospholipi d an d apo A- l ac t a s the substrat e fo r LCA T presen t i n a smal l sampl e o f plasma . Th e activity i s determined b y measuring th e rat e o f conversion o f [ 3H]cholesterol int o [3H]cholesteryl este r an d the resul t i s expressed a s nmole s o f cholestero l converte d to cholestery l este r pe r hou r pe r millilitr e o f plasma . Th e endogenou s cholesterol esterification rate  (CER ) measure s th e rat e a t which cholestery l ester s ar e 31 synthesized i n whole plasm a i n the absenc e o f any exogenous substrate . I n this assay, th e endogenou s poo l o f unesterified cholestero l i n the patient' s plasm a i s radiolabeled b y equilibration wit h [ 3H]cholesterol a t 4°C. Subsequent incubatio n o f the patient' s plasm a a t 37°C an d determination o f the rat e o f synthesis o f [3H]cholesteryl este r provide s a n estimate o f the ne t (o r endogenous) synthesi s rate . As a  result , th e measuremen t o f CER i s influenced b y both the amoun t o f enzym e present an d the natur e o f plasma lipoprotei n substrate s a s well a s cofactors an d other plasm a constituents . Thus , the CE R i s no t necessaril y proportiona l t o the concentration o f LCA T protein . Th e unit s o f activity ar e identica l t o those state d fo r LCAT activit y (nmole s o f cholesterol esterifie d pe r hou r pe r millilitre ) bu t the tw o assays ar e no t equivalent . 1.4 Clinica l Finding s i n LCAT Deficienc y an d Fis h Ey e Diseas e It has bee n nearl y 3 0 years sinc e th e firs t patien t wit h familia l lecithin:cholesterol acyltransferas e (LCAT ) deficienc y was  describe d b y Kar e Noru m and collaborator s (Noru m an d Gjone , 1967 ; Gjon e an d Norum , 1968 ; Torsvik etai, 1968). Sinc e tha t time , the stud y o f LCA T deficienc y ha s receive d a  great dea l o f attention fro m clinica l an d basi c scientist s throughou t th e world . These researc h efforts hav e provide d informatio n o n the clinica l presentatio n an d laborator y finding s in affected individual s whic h hav e le d to a  better understandin g o f the pathologica l changes tha t ar e commo n i n this rar e yet instructiv e diseas e process . As expected , a  functional o r absolute deficienc y o f this enzym e i n plasm a leads t o the failure t o synthesize norma l amount s o f cholestery l esters . Thus , i t is not surprisin g tha t suc h defect s lea d t o the formation o f abnormal lipoprotein s tha t 32 are enriche d i n unesterified cholestero l an d phosphatidylcholine . Th e dat a presente d in this sectio n ar e derived fro m summarie s fro m ove r 5 0 patients wit h familia l LCA T deficiency an d 1 1 patients wit h fish eye disease (FED ) wh o hav e bee n reporte d i n the literature . Th e majo r clinica l an d biochemica l feature s o f LCA T deficienc y an d FED ar e summarize d i n Table 4 . Table 4 . A summary o f the majo r clinica l an d biochemica l feature s o f homozygotes o f familial LCA T deficienc y an d FE D Clinical Feature s LCAT deficiency Corneal opacitie s Mild hemolyti c anemi a Proteinuria Progressive rena l insufficienc y Renal failur e Fish eye  disease Corneal opacitie s Biochemical feature s 1 LCA T activit y t unesterifie d cholestero l 1 phosphatidylcholine I cholestery l este r HDL deficienc y an d i  ap o A- l I LCA T activit y HDL deficiency an d i  ap o A- l Normal rati o o f unesterifie d t o esterified cholestero l 1.4.1 LCA T Deficienc y 1.4.1.1 Cornea l Change s Corneal opacitie s develo p i n al l patient s wit h familia l LCA T deficienc y i n early childhood an d ar e easil y detectabl e i n the secon d decad e o f life . The entir e cornea l stroma contain s numerou s greyis h dot s givin g th e appearanc e o f a  diffuse cloudin g of the cornea . A  muc h dense r accumulatio n o f these dot s exist s i n the cornea l 33 periphery resultin g i n an arcus-like appearance ; it s border , however , i s les s sharpl y defined tha n classica l arcu s lipoide s senilis . Initia l ultrastructura l examinatio n o f sections obtaine d b y superficia l keratotom y (Bro n et  ai,  1975 ; Bethel l et  ai,  1975 ) revealed numerou s vacuole s i n both Bowman' s laye r an d the anterio r stroma , som e of which containe d electro n dens e o r "membranous" deposits . Chemica l analysi s demonstrated hig h concentration s o f phospholipi d an d unesterifie d cholestero l compared t o the norma l corne a (Winde r et  ai, 1978) . I n subsequent studies , th e staining o f cryostat section s o f the corne a confirme d th e presenc e o f lipi d infiltratio n (Winder etai,  1985 ) an d unesterifie d cholestero l (Coga n et  ai,  1992 ) throughou t the strom a whil e electro n microscop y indicate d multilamina r inclusio n bodies . Corneal opacitie s hav e als o bee n identifie d i n other geneti c disorder s associated wit h HD L deficiency . Fo r example , simila r finding s hav e bee n observe d in patients wit h Tangie r diseas e (Ch u et  ai,  1979) , HD L deficienc y wit h plana r xanthomas (Gustafso n etai,  1979) , combine d ap o A-l/C-ll l deficienc y (Noru m etai, 1982), ap o A-I/C-III/A-IV deficienc y (Schaefe r et  ai,  1985 ) an d i n a frameshif t mutation o f the ap o A- l gen e (Funk e et  ai,  1991a) . 1.4.1.2 Hematologica l Abnormalitie s Most patient s wit h homozygou s familia l LCA T deficienc y presen t wit h a  mil d normochromic anemi a whic h fro m radioisotop e kineti c dat a an d bon e marro w studies appear s t o b e due to a  moderate hemolysi s combine d wit h a  reduce d compensatory erythropoiesi s (Gjon e et  ai,  1968 ; Cheve t et  ai,  1978) . I n addition, abnormalities i n the appearance , lipi d compositio n an d function o f erythrocytes hav e been observe d (Tabl e 5) . Examinatio n o f periphera l bloo d smear s ofte n indicate s a n 34 increased proportio n o f target cell s an d occasiona l anisocytosis , poikilocytosi s an d stomatocytes. Measuremen t o f the erythrocyte lipi d compositio n ha s indicate d increased concentration s o f unesterified cholestero l an d phosphatidylcholine , bu t decreased amount s o f phosphatidylethanolamine an d sphingomyelin . Tota l phospholipid concentrations , however , remai n norma l (Noru m et  al.,  1970 ; Godi n et al., 1978 ; Cheve t et  al.,  1978) . I n some cases , a  marked decreas e i n osmoti c fragility o f erythrocytes ha s bee n reporte d (Godi n et  al.,  1978 ; Shojani a et  al., 1983). Additiona l functiona l abnormalitie s observe d includ e a  decreased sodium influx an d reduced  acetylcholinesteras e activit y (Murayam a et  al.,  1984a) . I t is unclear i f abnormalities als o exis t fo r othe r bloo d cells . However , platele t functio n and lipi d compositio n appea r norma l (Nordo y an d Gjone , 1971) . Analysis o f bone marro w ofte n reveal s targe t cell s an d a  smal l numbe r o f foam cell s (Tabl e 6) . Giemsa stai n ha s identifie d "sea-blu e histiocytes " i n both th e bone marro w an d splee n o f many patient s (Jacobse n et  al.,  1972) . Ultrastructura l studies hav e demonstrate d tha t the histiocyt e granule s ar e composed  o f membrane s in a lamella r arrangemen t suggestiv e o f hig h concentration s o f unesterifie d cholesterol an d phosphatidylcholine . 35 Table 5 . Abnormalities o f composition an d function o f erythrocytes obtaine d from patient s wit h familia l LCA T deficiency . Erythrocyte Abnormalitie s t unesterifie d cholestero l t phosphatidylcholin e I phosphatidylethanolamin e I sphingomyeli n t mechanica l fragilit y I osmoti c fragilit y I sodiu m influ x i acetylcholinesteras e activit y 1.4.1.3 Rena l Diseas e Proteinuria i s a common findin g i n these patient s an d i s frequently detecte d i n the secon d an d third decad e o f life . I n addition, erythrocyte s an d hyalin e cast s ar e present i n the urin e o f many patients . Electrophoresi s o f urinary protei n indicate s that albumi n i s the predominan t protei n althoug h a r an d cr 2-migrating protein s ar e also present . Th e quantit y o f protein i n the urin e o f the patien t remain s moderat e (0.5 t o 1. 5 mg/ml ) an d persist s fo r man y years . Indice s o f rena l function suc h a s serum urea , seru m creatinin e an d creatinin e clearanc e usuall y remai n norma l i n the first three decade s o f life . However , late r i n life , deterioration o f rena l functio n ma y occur suddenl y an d develop int o a  rapid progressio n o f rena l insufficienc y wit h increasing proteinuri a an d hypertension . I t i s interesting t o not e tha t ther e ar e several patient s wit h complet e LCA T deficienc y who , despit e increasin g age , hav e not developed proteinuri a o r impaire d rena l function (Gjon e et  al.,  1974 ; Borysiewic z era/., 1982 ; Murayam a et  al.,  1984b) . Light microscop y o f kidney biopsie s (Hovi g an d Gjone , 1973 ) hav e identifie d 36 foam cell s i n the glomerul i (Tabl e 6) . These kidne y lesion s ar e characterize d b y the deposition o f phospholipi d membrane s an d membrane-boun d particle s i n the glomeruli, mesangiu m an d arterioles . Lipi d materia l ha s bee n found i n the subendothelial spac e a s well a s i n the basemen t membran e o f rena l arterie s an d arterioles. Analysi s o f the lipi d compositio n o f rena l tissues ha s indicate d elevate d concentrations o f unesterified cholestero l an d phospholipi d (Stokk e et  al.,  1974) . Electron microscop y ha s demonstrated th e narrowin g o f capillary lumens , th e detachment o f endothelia l cell s fro m th e basemen t membrane , fusio n o f endothelia l foot processe s an d the presenc e o f membrane-surrounded particle s i n both th e subendothelial an d subepithelia l region s (Hovi g an d Gjone , 1973) . Immunohistolog y has reveale d tha t immunoglobulin s an d complement-component s wer e onl y weakl y positive. However , indirec t immunofluorescenc e fo r ap o B  showed stron g stainin g along th e glomerula r capillar y wall s a s well a s i n mesangial region s (Oht a et  al., 1986). Table 6 . Tissue localizatio n o f abnormal cel l types foun d i n familial LCA T deficiency Cell Typ e Target cell s Foam cell s Sea-blue histiocyte s Multilaminar inclusio n bodies Tissues peripheral blood , bone marro w bone marrow , spleen , kidne y bone marrow , splee n peripheral blood , bone marrow , cornea, spleen , kidne y 37 1.4.2 Fis h Ey e Diseas e 1.4.2.1 Cornea l Change s FED was  initiall y reporte d i n two Swedis h familie s an d was  characterize d b y a severe HD L deficienc y an d extensiv e cornea l opacitie s (Carlso n an d Philipson , 1979; Carlson , 1982) . The latte r was  responsibl e fo r the unusua l nam e o f the disease, sinc e th e appearanc e o f the eye s o f affected individual s was  sai d t o b e similar t o that o f boiled fish . I n contrast t o LCA T deficiency , ther e hav e bee n muc h fewer report s o f patients wit h FE D with a  total numbe r o f 1 1 published patient s fro m 6 families originatin g fro m Sweden , Germany , Algeri a an d the Netherlands . Ther e i s also on e cas e fro m Canad a whos e biochemica l descriptio n closel y resemble s FE D (Frohlich et  al.,  1987) . Corneal opacitie s hav e bee n observe d a s early a s the secon d decad e an d are simila r i n appearance t o those describe d i n familial LCA T deficiency . The y consist o f minute greyis h dot s throughou t th e entir e cornea l strom a wit h increasin g density a t the peripher y revealin g a  diffuse bilatera l arcus . However , th e degre e o f opacification i n FE D appear s t o b e greater tha n familia l LCA T deficienc y a s visua l acuity i s often impaire d a t an earlier age . Analysis o f the cornea l dis c o f one patien t obtained a t keratoplast y reveale d hig h concentration s o f unesterifie d cholestero l an d the presenc e o f vacuoles i n the stroma containin g membrane-lik e materia l i n the periphery (Carlson , 1982) . Other tha n th e cornea l opacities , ther e ar e n o other clinica l symptom s suc h as anemia , proteinuri a o r rena l insufficienc y commonl y foun d i n familial LCA T deficiency. 38 1.5 Laborator y Finding s i n LCAT Deficienc y an d Fis h Ey e Diseas e 1.5.1 Standar d Lipi d Profil e The standar d lipi d profil e fo r bot h familia l LCA T deficienc y an d FE D i s presented i n Table 7 . I n each case , there i s a significant reductio n o f HD L cholesterol an d ap o A-l , but tota l cholestero l level s ar e usuall y withi n th e norma l range. A  moderat e hypertriglyceridemi a i s common i n both disorders . However , levels ca n b e significantly elevate d i n LCAT deficienc y afte r th e onse t o f rena l insufficiency. Sinc e th e lipi d profile s o f these disorder s a s well a s other HD L deficiencies ma y b e similar , additiona l specia l test s ar e require d fo r a  differentia l diagnosis. 1.5.2 Specialize d Laborator y Test s The determinatio n o f the cholestero l esterificatio n rat e (CER ) an d th e measurement o f LCA T enzym e activit y a s described previousl y usin g bot h syntheti c and endogenou s lipoprotei n substrate s i s necessary t o distinguish LCA T deficienc y from FED . As indicate d i n Table 8 , LCA T deficienc y i s associated wit h ver y lo w concentrations o f plasma cholestery l este r an d a  virtual absenc e o f activit y fo r bot h synthetic an d endogenou s substrates . I n contrast, ther e i s only a  mild reductio n i n the percentag e o f esterified cholestero l i n the plasm a o f FE D patient s an d th e endogenous cholestero l esterificatio n rat e ma y b e within norma l limit s o r onl y slightl y reduced. However , whe n LCA T activit y i s measured usin g a  syntheti c proteoliposome substrate , th e abilit y t o esterify cholestero l i s usuall y les s tha n 10 % of normal . Although ther e i s considerable variatio n i n plasma LCA T protei n level s i n LCAT deficienc y (Alber s et  al.,  1981) , they ar e usuall y lowe r whe n compare d t o those concentration s observe d i n FED. 39 Table 7 . Laboratory finding s i n representative homozygote s fo r familia l LCA T deficienc y an d FED . Familial LCA T deficiency (n=5) Fish ey e disease (n=4) Normals (n=8) Total Cholesterol (mmol/L) 3.56±1.14 5.11±0.40 5.39±0.85 HDL Cholesterol (mmol/L) 0.14±0.11 0.05±0.06 1.48±0.31 Triglycerides (mmol/L) 2.74±1.30 2.80±0.71 0.97±0.27 Apo A- l (g/L) 0.44±0.06 0.28±0.09 1.44±0.2 Apo B (g/L) 0.35±0.01 0.83±0.11 1.00±0.2 References Frohlich et  ai, 198 8 Funke et  ai, 199 3 Funke et  ai. 1991 b Kastelein et  ai,  199 2 Frohlich et  ai,  198 8 Table 8 . Laboratory finding s i n familial LCA T deficienc y an d FED : specialized tests . Familial LCA T deficiency (n=5) Fish ey e disease (n=4) Normals (n=7) % Cholestery l esters 12±2 56±3 68±7 LCAT Activity (nmol/h/ml) 0.6±0.8 0.6±0.3 30±2.6 Cholesterol esterification rat e (CER) i n plasm a (nmol/h/ml) 0 58±13 94±27 LCAT Concentration (//g/ml) 0.09±0.13 2.30±0.21 4.83±0.2 References Frohlich et  al.,  198 8 Funke et  al.,  199 3 Funke et  al.,  1991 b Kasteiein et  al.,  199 2 Funke et  al.,  1991 b 1.5.3 Lipoprotei n Abnormalitie s i n Familia l LCA T Deficienc y Analysis o f the plasm a fro m LCA T deficien t patient s reveal s multipl e lipoprotein abnormalitie s an d a  considerable heterogeneit y o f plasma lipi d concentrations. However , th e mos t consisten t biochemica l finding s ar e hig h plasm a concentrations o f unesterifie d cholestero l an d phosphatidylcholine  an d lo w concentrations o f plasma cholestery l este r an d lysolecithin . Th e very  lo w amounts o f plasma cholestery l este r tha t ar e presen t ar e likel y t o originate fro m th e intestin e a s a resul t o f the actio n o f the intracellula r enzym e acy l coenzym e A:cholestero l acyltransferase (ACAT) . Th e highe r proportion s o f palmiti c an d olei c acid s i n comparison wit h linolei c aci d presen t i n cholesteryl ester s ar e consisten t wit h th e pattern normall y produce d b y ACAT activity . Plasma lipoprotein s hav e bee n isolate d an d analyse d b y a  number o f differen t methodologies includin g ultracentrifugation , ge l filtration, electrophoresi s an d electron microscopy . Ver y lo w densit y lipoprotein s (VLDL ) o f d<1.006 g/m l ar e usually elevate d i n concentration an d contai n highe r concentration s o f unesterifie d cholesterol an d phosphatidylcholin e relativ e t o triglyceride an d protei n (Glomse t et al., 1973) . Apoprotein analysi s ha s indicate d increase d amount s o f apo C- l an d ap o E, bu t decreased level s o f apo C-l l an d apo C-ll l (Glomse t et  al.,  1980) . Afte r fastin g overnight, th e VLDL fractio n stil l contains a  large numbe r o f particles tha t ar e greater tha n 9 0 n m i n diameter. However , dietar y studie s (Glomse t et  al.,  1975 ) have demonstrate d tha t th e proportio n o f these larg e lipoprotein s ca n b e markedl y reduced whe n th e patient s consum e fat-fre e diets , suggestin g tha t these particle s are i n fact chylomicrons . Abnormalities ar e als o presen t i n intermediate densit y lipoprotein s (IDL , 42 d=1.006-1.019 g/ml) . Thes e particle s hav e bee n show n t o be heterogeneou s i n nature an d ric h i n unesterified cholestero l an d triglyceride (Glomse t et  al.,  1973) . One possibl e caus e o f the triglyceride enrichmen t o f ID L may aris e fro m th e decreased abilit y o f hepati c lipas e t o hydrolys e triglycerid e i n these particle s (Murano et  al.,  1987) . Gel filtration o f the lo w densit y lipoprotei n fractio n (LDL , d=1.019-1.06 3 g/ml ) usually yield s thre e relativel y wel l define d subtractions , wherea s onl y on e fraction i s obtained fro m norma l LDL . The relativ e proportio n o f each fractio n differ s amon g individuals an d i n a similar manne r t o the VLDL fraction , the quantit y o f the large r molecular weigh t LD L particle s appear s t o be relate d t o the fa t conten t i n the die t (Glomset et  al.,  1975 ; Borysiewic z et  al.,  1982) . Th e larges t subtractio n contain s multilaminar particle s whic h ar e 9 0 nm i n diameter consistin g mostl y o f lipi d wit h a high mola r rati o o f unesterified cholestero l t o phospholipi d (2:1) . The smal l amoun t of protein presen t i n these particle s i s mostly albumin . A secon d subtractio n o f patient LD L i s made u p of particles whic h rang e i n diameter fro m 3 0 to 8 0 nm and lik e the large r particles , hav e n o obviou s counterparts i n norma l plasma . However , th e characteristic s o f these particle s ar e similar t o the abnorma l lipoprotei n LP- X presen t i n cholestasis (Hamilto n et  al., 1971). Thes e particle s ar e ric h i n unesterified cholestero l an d phosphatidylcholin e and hav e a  disc-shaped appearanc e whe n examine d b y electron microscopy . Ap o C, mainly  C-l , make s u p the primar y protei n componen t o f these particles . Spherica l particles resemblin g norma l remnant s o f VLDL an d chylomicron s ar e als o presen t i n this fraction . The thir d subfractio n o f patien t LD L ar e spherica l an d hav e a  simila r siz e 43 compared t o norma l LD L (20-2 2 nm) . The majo r protei n i s apo B 100 and th e proportion o f core lipi d t o surface lipi d an d protei n i s normal . However , th e lipi d composition i s markedly differen t wit h 1. 5 t o 3  times a s much unesterifie d cholesterol an d phosphatidylcholin e an d 13-time s a s much triglyceride pe r m g o f protein (Noru m era/. , 1971 ) Like th e othe r lipoprotein s associate d wit h familia l LCA T deficiency , hig h density lipoprotein s (HDL , d=1.063-1.2 1 g/ml ) ar e heterogeneou s i n natur e an d consist o f abnormal shape s an d lipi d composition . Man y o f the particle s ar e disc -shaped whil e other s remai n spherical , bu t unusuall y small . Disc-shape d particle s mainly consis t o f unesterifie d cholestero l an d phosphatidylcholine . Som e contai n mixtures o f apo A-l , apo A-ll , apo A-IV an d the C  apolipoproteins (Torsvik,  1972 ; Soutar et  al.,  1982 ) whil e othe r discs  remai n ric h i n apo E  (Mitchell et  al.,  1980) . Th e small spherica l HD L ar e abou t 6  nm i n diameter an d contai n unesterifie d cholesterol, phosphatidylcholine , a  smal l amoun t o f core lipi d an d 2  molecules o f apo A- l (Che n et  al.,  1984) . 1.5.4 Lipoprotei n Abnormalitie s i n Fish Ey e Diseas e Results fro m the analysi s o f the compositio n an d structur e o f the plasm a lipoproteins i n FE D were obtaine d i n large par t from the origina l Swedis h patient s (Carlson, 1982 ; Carlso n an d Holmquist , 1983 ; Fort e an d Carlson , 1984) . Th e concentrations o f VLDL cholestero l an d triglyceride ar e increase d five-fol d althoug h the triglycerid e t o cholestery l este r rati o remain s normal . VLDL particle s hav e a larger mea n diamete r du e to a  much broade r rang e o f particle siz e no t presen t i n controls. 44 Both th e ID L and LD L fractions ar e enriched i n triglyceride, particularl y LDL , in which th e triglyceride t o cholestery l este r rati o i s ten-fold greate r tha n normal . Th e LDL from FE D patient s ar e mostl y homogeneou s roun d structure s wit h a  smalle r mean diamete r tha n contro l LDL . Occasionally , a  few larg e structure s (5 0 to 8 5 n m diameter) ar e seen i n the patien t LD L fraction, bu t ar e absen t fro m norma l LDL . The HD L fraction o f these patient s demonstrat e th e greates t abnormalities . The tota l mas s o f HD L i s only 10 % that o f controls an d the percentag e o f unesterified cholestero l (60% ) i s about three-fol d greate r compare d t o contro l HDL . There i s an enrichment o f apo E  and a  reduction o f the A  apolipoproteins b y 90% with a  relativel y greate r reductio n o f apo A-l l compare d t o apo A-l . (Carlson an d Holmquist, 1983) . Th e patien t HD L containe d severa l particl e morphologies . Th e tw o major component s ar e smal l roun d particle s wit h a  diameter o f 7.6 n m an d discoida l particles wit h a  thickness o f 4.4 n m and a n average diamete r o f 17. 4 n m (Fort e an d Carlson, 1984) . Ther e ar e also large r particle s presen t whic h rang e fro m 4 0 to 9 0 nm i n diameter . 1.6 Heterozygote s o f LCA T Deficienc y an d Fis h Ey e Diseas e 1.6.1 Laborator y Finding s i n LCAT Deficienc y The measuremen t o f the endogenou s cholestero l esterificatio n rat e doe s no t differentiate heterozygote s fro m unaffecte d famil y member s o r norma l subjects . However, LCA T activit y i s reduced t o 50% of norma l an d ca n b e use d t o assig n genotype i n non-obligate heterozygotes . Despit e thes e change s i n LCA T activity , n o major abnormalitie s o f plasma lipoprotein s hav e bee n observe d i n heterozygou s LCAT deficiency . However , abnormalitie s i n erythrocyte compositio n an d functio n 45 have bee n describe d i n several heterozygote s (Godi n era/. , 1978 ; Jai n et  al.,  1982) . In addition, ther e hav e bee n report s o f hyperlipidemia , especiall y hypercholesterolemia, i n several differen t kindred s (Torsvi k et  al.,  1968 ; Cheve t et al., 1978 ; Borysiewic z et  al.,  1982) . A mor e detailed repor t o f one family indicate d tha t heterozygote s ha d significantly highe r fastin g plasm a triglycerides , ap o B , an d lowe r HD L cholestero l and ap o A- l tha n th e unaffecte d famil y member s (Frohlich  et  al.,  1988) . I n addition, HDL cholestery l este r was  decrease d b y approximately 50% . Differences i n these values wer e intermediat e betwee n thos e fo r norma l an d homozygou s famil y members. Therefore , i t seems likel y tha t mos t o f these abnormalitie s occu r a s a result o f reduce d LCA T activity . 1.6.2 Laborator y Finding s i n Fish Ey e Diseas e As wit h LCA T deficiency , heterozygote s o f FE D ca n onl y b e distinguished b y the measuremen t o f LCA T activit y an d no t the CER . I n FED, LCA T activit y i s reduced t o 50 % of normal . Although no t necessaril y a  consistent finding , HD L cholesterol level s ar e ofte n decrease d i n heterozygotes . I n contrast, LCA T protei n levels an d endogenous cholestero l esterificatio n rate s ar e usuall y withi n th e norma l range. 1.7 Pathophysiolog y 1.7.1 LCA T Deficienc y 1.7.1.1 Lipoprotei n Abnormalitie s From the analysi s o f the lipoprotei n abnormalitie s i n LCA T deficien t plasma , i t appears tha t th e failure t o esterify cholestero l cause s a  chain reactio n o f event s 46 which affect s th e norma l metabolis m o f al l lipoprotei n classes . Th e change s i n the composition, structur e an d metabolis m o f these lipoprotein s giv e ris e t o the tissu e pathology tha t ultimatel y lea d to clinica l complications . T o determine th e mechanisms associate d wit h thes e differen t abnormalities , severa l studie s hav e conducted i n vitro experiment s utilizin g patien t lipoprotein s an d norma l LCAT . Incubation experiment s wit h partiall y purifie d LCA T an d patien t lipoprotein s (Glomset et  al.,  1970 ) hav e indicate d tha t the enzym e react s primaril y wit h HD L an d to a  much lesse r exten t wit h lipoprotein s o f lower density . Thes e result s indicate d that th e lipoprotei n abnormalitie s presen t i n familial LCA T deficienc y wer e no t caused b y the inabilit y o f these particle s t o reac t with LCAT , bu t were directl y related t o the LCA T deficiency . Late r studie s demonstrate d tha t LCA T converte d patient HD L from a  highl y heterogeneou s mixtur e o f disc-shaped particle s an d smal l spherical particle s t o materia l tha t resemble d norma l HDL 2 and HDL 3 (Glomse t et al., 1980) . The time cours e o f this reactio n implie d tha t the smal l spherica l HD L were converte d int o HDL 3 an d the discs int o HDL 2. These result s ma y indicat e tha t HDL3 an d HDL 2 originat e fro m differen t nascen t particle s an d that th e rol e o f LCA T is no t simpl y a  conversion o f HDL 2 to HDL 3. There i s considerable evidenc e tha t disc-shaped HD L represen t newl y formed  o r nascen t HDL . Structure s o f simila r siz e and shap e hav e bee n identifie d i n perfused ra t liver s (Hamilto n et  al.,  1976) , huma n hepatic venou s bloo d (Turne r et  al.,  1979 ) an d from the huma n hepatoblastom a cel l line, HepG 2 (Thrif t et  al.,  1986) . In vitro experiment s hav e demonstrate d tha t the actio n o f LCA T o n patien t HDL indirectl y affect s th e characteristic s o f both LD L an d VLDL (Glomse t et  al., 1970). Severa l studie s hav e indicate d tha t the sourc e o f unesterifie d cholestero l an d 47 phosphatidylcholine consume d b y the LCAT-HD L comple x i s derived mostl y fro m LDL an d VLDL (Fieldin g an d Fielding , 1981 ; Park era/. , 1987 ; Cheun g an d Wolfe, 1989; Huan g et  al,  1993 ) an d that the majorit y o f cholestery l ester s synthesize d ar e transferred t o these lipoprotein s i n exchange fo r triglyceride b y plasm a cholestery l ester transfe r protei n (CETP ) (Chaje k an d Fielding , 1978) . Th e combine d effec t o f these lipi d transfe r reaction s i n patient plasm a incubate d wit h norma l LCA T result s in the normalizatio n o f LD L an d VLDL wit h a  particular reductio n i n the larg e size d particles an d a n increas e i n cholesteryl este r i n the norma l size d LD L an d VLDL. When patient s wit h familia l LCA T deficienc y consum e fat-fre e diet s fo r severa l days, th e concentratio n o f the larg e VLDL , th e larg e an d intermediat e size d LD L an d in one case th e smal l spherica l HD L al l decreased (Glomse t et  al.,  1975) . Th e possibility tha t these particle s ar e derived fro m chylomicron s i s supported b y i n vitro studies i n which lipoprotei n lipas e (LPL ) incubate d wit h triglyceride-rich lipoprotein s produced particle s whic h wer e ric h i n unesterified cholestero l an d phosphatidylcholine (Deckelbau m et  al.,  1979) . Thus , these lipoprotein s migh t b e formed fro m surfac e remnan t unesterifie d cholesterol , phosphatidylcholin e an d protein whic h woul d otherwis e b e normall y metabolized . Ther e ar e conflictin g report s of the abilit y o f these particles , suc h a s LP-X , t o ac t directl y a s substrates fo r LCA T (Wengeler an d Seidel , 1973 ; Patsc h et  al.,  1977) , bu t there i s evidence tha t LP- X can provid e a  source o f unesterified cholestero l fo r the LCA T reactio n (Ritlan d an d Gjone, 1975) . Thus , i t may b e that the lipid s fro m these surfac e remnant s ar e transported t o LD L an d VLDL particle s whic h i n turn suppl y th e substrat e fo r th e LCAT reactio n i n the HD L fraction. In familial LCA T deficiency , th e relationshi p betwee n tissu e abnormalitie s an d 48 the presenc e o f abnormal lipoprotei n particle s remain s unclear . Result s o f incubatio n studies wit h norma l erythrocyte s an d patien t plasm a sugges t tha t th e abnorma l lipi d composition o f erythrocytes i s related t o the hig h concentrations o f unesterifie d cholesterol i n patien t plasm a (Noru m an d Gjone , 1968) . I t is likely tha t the altere d membrane compositio n an d function o f erythrocytes observe d i n these patient s contributes t o the hemolyti c anemi a normall y presen t i n LCA T deficiency . 1.7.1.2 Rena l diseas e Multilaminar structure s hav e bee n consistentl y foun d i n the plasma , bon e marrow, spleen , corne a an d kidney s o f patients wit h familia l LCA T deficienc y an d appear t o be responsibl e i n par t fo r the patholog y recognize d i n these differen t tissues. I t has bee n suggeste d tha t the accumulatio n o f these membrane s an d membrane-bound particle s i n the glomerula r capillarie s ma y cause damag e t o the endothelium an d underlyin g basemen t membran e (Hovi g an d Gjone , 1973) . Thes e changes coul d b e responsibl e fo r the proteinuria , reduce d filtratio n efficienc y an d increasing rena l insufficienc y observe d i n these patients . Initially , i t was propose d b y Gjone et  al.  (1974) tha t th e larg e molecula r weigh t LD L ma y b e responsibl e fo r th e impaired kidne y function . Fro m the initia l eigh t patient s studied , thi s lipoprotei n was identified i n seven subject s wit h rena l involvement , bu t was no t presen t i n the onl y patient withou t proteinuri a an d norma l rena l function (Gjon e et  al.,  1974) . However , in a subsequent report , thes e particle s wer e foun d i n large quantit y i n two case s o f familial LCA T deficienc y whic h ha d n o proteinuria an d norma l rena l functio n (Borysiewicz et  al.,  1982) . Kidne y biopsie s take n fro m thes e asymptomati c patient s contained widesprea d subendotheliai , subepithelia l an d intramembranou s glomerula r 49 lipid deposit s a s well a s occasiona l foa m cells . I n addition, a  similar lipi d accumulation ha s bee n observe d i n rena l biopsie s take n si x month s afte r transplantation fro m LCA T deficien t patient s wit h normall y functionin g graft s (Flatmark et  al.,  1977) . Therefore , i t would see m tha t th e infiltratio n o f lipi d alon e i s not sufficien t t o initiat e th e onse t o f reduce d rena l function . I t has bee n suggeste d b y Borysiewicz et  al.  (1982) tha t thi s conditio n coul d caus e a  reduced clearanc e o f immune complexe s predisposin g t o the developmen t o f glomerulonephritis , whic h has bee n reporte d i n several patient s wit h LCA T deficiency . From a  review o f the literature , ther e ar e tota l o f 1 0 reported case s o f familia l LCAT deficienc y i n which th e patien t ha d n o proteinuria an d norma l rena l function . Three case s presente d wit h partia l LCA T deficienc y i n which th e percentag e o f plasma cholestery l este r was  greate r tha n 40% (Sakuma et  al.,  1982 ; Gyllin g an d Miettinen, 1992) . I n two othe r cases , th e patient s wer e ver y youn g a t examinatio n and could represen t earl y stage s o f the disease (Frohlic h et  al.,  1978 ; Winder an d Bron, 1978) . There hav e bee n fou r reporte d case s o f females, al l o f which wer e over 5 0 years o f age a t examination (Gjon e et  al.,  1974 ; Borysiewic z et  al.,  1982 ; Murayama et  al.,  1984 ) whic h ha d n o symptoms bu t ha d mal e sibling s whic h demonstrated typica l sign s o f rena l involvement . Also , i n three cases , th e asymptomatic patien t was  a  vegetarian (Murayam a et  al.,  1984b ; Takat a et  al., 1989; Frohlich , unpublishe d results) . A s a  result , th e influenc e o f diet a s wel l a s sex hormones shoul d als o b e considered a s factors whic h ma y affec t th e developmen t of rena l disease . 50 1.7.1.3 Cornea l change s The pathophysiologica l mechanis m responsibl e fo r the developmen t o f corneal opacitie s remain s t o b e determined. A s mentione d earlier , ther e ar e a number o f different disorder s associate d wit h HD L deficienc y an d cornea l lipi d deposits, bu t the degre e o f opacification doe s no t correlat e wit h th e exten t o f HD L deficiency o r the ris k o f premature atherosclerosis . Fro m the analysi s o f the differen t characteristics o f the lipoprotein s i n these disorders , i t i s more likel y tha t th e specifi c composition an d structur e o f the abnorma l HD L particle s wil l reflec t th e exten t an d rate a t which the y accumulat e i n the cornea l tissue . 1.7.2 Fis h Ey e Diseas e 1.7.2.1 Lipoprotei n Abnormalitie s Initial study o f the activit y o f LCAT from FE D plasm a indicate d nea r norma l rates o f esterification whe n th e patient s lipoprotein s acte d a s substrate bu t was  onl y 10-15% o f norma l whe n proteoliposome s wer e use d a s substrate (Carlso n an d Holmquist, 1985a) . Thi s "paradoxica l esterification " o f plasma cholestero l wa s furthe r investigated b y performin g experiment s wit h mixture s o f isolated lipoprotein s an d plasma fraction s fro m patient s wit h FE D and norma l lipoprotein s an d plasm a fractions fro m controls . HD L isolate d fro m FE D patient s wer e foun d t o b e excellen t substrates whe n incubate d wit h norma l LCA T a s presen t i n lipoprotein deplete d plasma fro m health y individual s (Carlso n an d Holmquist , 1985b) . Mos t o f the unesterified cholestero l i n these HD L fractions becam e esterifie d an d shifte d th e size o f these particle s i n the rang e o f that o f norma l HDL . However , lipoprotei n depleted FE D plasm a was  unabl e t o esterify cholestero l i n HD L isolate d fro m eithe r 51 control subject s o r from th e plasm a o f FE D patients . I n addition, th e unesterifie d cholesterol containe d withi n th e combine d fractio n o f LD L an d VLDL was  esterifie d in the presenc e o f lipoprotein deplete d plasm a fro m a  control subjec t o r a  patien t with FE D (Carlso n an d Holmquist , 1985c) . Thes e result s le d to the formulation o f a hypothesis pu t forth b y Carlson an d Holmquis t tha t two differen t LCA T activitie s exist i n norma l plasma . On e o f these activities , was  denote d cr-LCA T whic h esterifies cholestero l presen t i n HD L particle s an d the other , wa s denote d G-LCA T which esterifie s cholestero l presen t i n LDL an d VLDL particles . Furthe r evidence  t o support thi s hypothesi s was  demonstrate d whe n HD L cholestery l este r conten t an d its particl e siz e wer e normalize d afte r smal l amount s o f highl y purifie d LCA T wer e added t o FE D plasm a (Holmquis t an d Carlson , 1988) . According t o this ne w concept, FE D would b e classified a s an cr-LCAT deficiency , i n contrast t o familia l LCAT deficiency , whic h lack s bot h a-  and R-LCA T activities . In norma l plasma , the majorit y o f LCAT i s associated wit h HD L an d probabl y continuously cycle s fro m matur e HD L to newl y forme d nascen t HD L whil e onl y a small portio n o f LCA T remain s associate d wit h LD L (Chun g et  al.,  1982) . I n FED, the bindin g o f LCA T t o HD L ma y b e impaired resultin g i n a shif t i n the distributio n o f LCAT toward s LD L enabling th e maintenanc e o f norma l endogenou s plasm a cholesterol esterificatio n rates . Despit e th e norma l esterificatio n o f cholestero l i n the LDL fraction , abnormalitie s persis t i n the lowe r densit y lipoproteins . Th e accumulation o f triglycerides i n LDL i s probably a  reflection o f the inabilit y o f CETP to carry ou t it s norma l lipi d exchang e o f cholesteryl este r fo r triglyceride i n HDL. 52 1.8 Coronar y Arter y Diseas e i n HDL Deficienc y Syndrome s A causal  relationshi p betwee n decrease d HD L level s an d coronar y arter y disease i s now wel l established . Ther e ar e a  wide variet y o f HD L deficienc y syndromes whic h consis t o f disorders o f different etiolog y an d clinica l phenotypi c expression. Disorder s suc h a s HD L deficiency wit h plana r xanthoma s (Gustafso n et al., 1979) , ap o A-l/C-ll l deficienc y (Noru m et  al.,  1982 ) an d ap o A-I/C-II/A-I V deficiency (Schaefe r era/. , 1985 ) hav e bee n associate d wit h prematur e coronar y heart disease . I n contrast, suc h a n association ha s no t bee n reporte d fo r Tangie r disease (Ch u et  al.,  1979) , HD L deficiencie s secondar y t o LP L deficiency (Brunzell , 1989) o r mutation s o f the ap o A- l gen e (Franceschin i et  al.,  1980 ; von Eckardstei n et al.,  1989 ; Funk e et  al.,  1991a) . I n addition, althoug h som e patient s affecte d wit h either LCA T deficienc y o r FE D hav e develope d coronar y arter y disease , th e clinica l data doe s no t indicat e tha t they ar e a t an increased risk . On e possibl e explanatio n of this apparen t parado x ma y b e relate d t o the presenc e o f smal l pool s o f HD L i n these disorder s unlik e HD L deficiencies cause d b y defects o f the A-I/C-III/A-IV gen e cluster. Thes e smal l nascen t HD L particle s ma y stil l b e able t o induc e a n efflux o f unesterified cholestero l fro m periphera l tissue s an d transpor t cholestero l t o othe r lipoproteins o r directly t o the live r i n order t o maintai n revers e cholestero l transport . This notio n i s supported b y the descriptio n o f an LCAT-independen t mechanis m o f reverse cholestero l transpor t discusse d previousl y (Huan g et  al.,  1993) . 1.9 Treatmen t o f Familia l LCA T Deficienc y Dietary treatmen t i n the for m o f fat restrictio n ha s bee n implemente d a s a conservative approac h i n several patient s i n order t o reduc e th e level s o f the 53 potentially pathogeni c larg e molecula r weigh t lipoprotei n particle s foun d i n the LD L and VLDL fractions . Presently , ther e i s no information o f the effectivenes s o f such a treatment i n either th e preventio n o r amelioration o f reduced rena l function. The firs t reporte d cas e o f whole plasm a o r bloo d transfusion s i n familial LCA T deficiency consiste d o f the administratio n o f 450 m l blood an d 50 0 m l plasma i n a single transfusio n (Noru m an d Gjone , 1968) . Thi s resulte d i n an immediat e increas e in plasma cholestery l este r whic h continue d a t a  slower rat e reachin g a  peak a t 6 days, followe d b y a  decrease t o pre-treatmen t level s i n 2 weeks. Th e i n vivo half-lif e of LCA T i n plasma was  calculate d t o b e 4.6 days . I n a second report , th e plasm a transfusion consiste d o f 2200 m l of plasma ove r a  period o f ten days (Murayam a e f a/., 1984b) . LCA T activit y onl y increase d fro m 9.4 % o f the norma l valu e t o 17.4% . Despite thi s smal l increas e i n LCAT activity , ther e wer e significan t increase s i n cholesteryl este r i n al l lipoprotei n fraction s an d electron microscop y demonstrate d that bot h VLDL an d LD L fractions converte d t o nea r norma l siz e particles . However , there appeare d t o b e n o effect o n the siz e o r shape o f the HD L particle s a s the characteristic disc-shape d particle s stil l remained . Th e effectivenes s o f the treatmen t continued fo r 7  days afte r th e cessatio n o f the infusion . Thes e report s indicat e tha t plasma infusio n i s a relatively inefficien t for m o f treatment a s the administratio n o f large amount s o f plasm a provide d onl y a  modest improvemen t o f the biochemica l findings fo r a  short duratio n o f time. Renal transplantatio n ha s bee n performe d i n severa l patient s wit h familia l LCAT deficiency . Althoug h th e plasm a lipi d an d lipoprotei n abnormalitie s persist , a significant reductio n i n triglyceride level s wa s observe d i n one patien t afte r transplantation (Noru m et  al.,  1989) . As mentione d previously , ther e i s evidence o f 54 early lipi d depositio n i n the transplanted kidneys ; however , norma l rena l function ca n still b e maintained (Flatmar k et  ai,  1977) . Fro m a  biochemica l poin t o f view, a  live r transplant woul d correc t th e underlyin g defec t o f LCA T deficiency ; however , i t would be difficul t t o justify suc h a  high ris k procedure  fo r what woul d otherwis e b e considered a  norma l functionin g liver . 1.10 LCA T Gen e Structur e an d Expressio n The LCA T gen e i s found o n the lon g ar m o f chromosome 1 6 and i s divided into si x exon s spannin g 4.2k b o f genomic DN A (McLea n et  ai,  1986b) . Souther n blot hybridizatio n dat a sugges t tha t ther e i s only on e LCA T gen e i n humans. A n unusual featur e o f the LCA T mRN A i s that the poly(A ) signa l overlap s th e carboxyl -terminal glutami c aci d an d stop codons , an d the 3 ' untranslate d regio n contain s onl y 23 base s (McLea n et  ai,  1986a) . Th e firs t 5 9 nucleotides o f exon 5  share a  66% DNA sequenc e homolog y wit h th e 3 ' terminal codin g regio n o f the ap o E  gene. Thi s appears t o b e the onl y regio n which coul d potentiall y for m a n cr-helix . Functional analysi s o f the 2. 9 k b of the 5 ' flanking sequence s o f the LCA T gene indicate d tha t there was  n o change i n promoter activit y fo r deletions fro m -2900 t o -300 , a  50% reduction afte r deletin g t o position -71 , and a  100 % reductio n after deletin g t o position -4 2 (Meron i et  ai,  1991) . Site-directe d mutagenesi s withi n the 7 1 bp regio n indicate d tha t the activit y o f the promote r was  dependen t o n the integrity o f two putativ e Sp 1 bindin g sites . I n addition, sinc e th e promote r wa s activ e in both hepati c an d non-hepati c cel l lines , the presenc e o f hepato-specifi c regulator y elements wer e no t indicated . The huma n LCA T mRN A ha s bee n detecte d i n the live r an d HepG 2 55 (hepatocyte) cells , bu t no t i n small intestine , splee n pancreas , placent a o r adrena l tissue (McLea n et  a/. , 1986b) . Analysi s o f the tissue expressio n i n the ra t also indicated th e presenc e o f LCA T mRN A i n the brai n an d testes (Warde n et  ai,  1989) . More detaile d analysi s o f the cellula r localizatio n o f LCAT mRN A i n rhesus monkey s demonstrated tha t i n addition t o hepatocyte s an d brai n cells , LCA T mRN A was present i n the basa l cel l laye r o f the dermi s (Smit h et  ai,  1990) . I n the brain , LCA T mRNA wa s synthesize d b y scattered neurons , neuroglia l cells , ependyma l cells , a s well a s a  discrete cel l laye r i n the cerebellum . I n a recent study , th e expressio n o f LCAT i n baboon tissue s was  analyse d usin g a n RNas e protectio n assa y (Hixso n et ai, 1993) . Th e highes t level s o f LCA T mRN A wer e foun d i n the cerebellu m an d live r (33% compare d t o cerebellum) , wherea s onl y trac e amount s wer e identifie d i n the ileum, splee n an d cerebra l cortex . Fro m these results , i t would see m tha t LCA T als o participates i n lipi d transpor t i n the brain , particularl y th e cerebellum . Comparison o f the percen t homolog y o f human LCA T nucleotid e an d amin o acid sequence s wit h th e LCA T sequence s reporte d fo r mous e (Warde n et  ai,  1990) , rat (Meron i et  ai,  1990) , pi g (Pritchar d et  ai, unpublishe d observations ) an d baboo n (Hixson et  ai,  1993 ) i s depicted i n Table 9 . These dat a demonstrat e a  high degre e of homolog y o f LCA T sequenc e betwee n al l species . Th e ra t and mous e sequence s have a  highe r degre e o f homolog y whe n compare d t o one another , 92 % and 90 % at the amin o aci d an d nucleotid e levels , respectivel y (Meron i et  ai,  1990) . Al l o f the major structura l feature s o f LCA T suc h a s the catalyti c Ser , glycosylatio n site s an d Cys residue s ar e conserved amon g al l species . 56 Table 9 . Homology o f huma n LCA T wit h differen t specie s o f LCAT . Species Mouse Rat Pig Baboon Nucleotide 85% 84% 90% 97% Amino Aci d 85% 86% 92% 98% References Warden era/. , 199 0 Meroni et  al.,  199 0 Pritchard e f al., unpublished observation s Hixson et  al.,  199 3 1.11 Gen e Defect s Associate d wit h LCA T Deficienc y an d Fis h Ey e Diseas e Analysis o f the clinica l an d biochemica l phenotype s o f relative s withi n different familie s affecte d wit h eithe r familia l LCA T deficienc y o r FE D clearl y indicat e that thes e disease s ar e inherite d a s an autosomal recessiv e trait . Ove r th e las t 5 years, severa l laboratorie s hav e reporte d gen e defect s tha t appea r t o b e causativ e for LCA T deficienc y an d FED . This ha s bee n don e b y cloning an d sequencin g o f the entire LCA T gen e o r by amplification an d sequencing o f the exo n an d intron/exo n boundaries. Th e sequenc e analysi s o f the LCA T gen e i n patients wit h familia l LCA T deficiency an d FE D ha s reveale d a  large variet y o f mutations sprea d throughou t al l six exons (Tabl e 10) . Only on e o f the mutation s (Gly183->Ser ) appear s t o affec t th e catalytic sit e directl y b y disrupting th e highl y conserve d structura l moti f Gly-X-Ser-X -Gly associate d wit h the catalyti c domai n o f many lipase s (Brenner , 1988) . Th e causative natur e o f these mutation s ha s been establishe d fro m the fact tha t non e o f these defect s hav e bee n identifie d i n a normal populatio n and , i n mos t cases , th e observed mutatio n was  th e onl y on e found o n each allel e analysed . Th e assignmen t of these mutation s t o either LCA T deficienc y o r FE D i s often initiall y base d o n the presence o r absence o f specific clinica l symptoms . However , mor e detaile d 57 biochemical analyse s hav e indicate d tha t suc h a  segregation shoul d b e use d wit h caution. Th e variet y o f LCA T concentration s an d activitie s observe d i n patien t plasma indicat e a  wide rang e o f functional defect s whic h ofte n consis t o f characteristics o f bot h LCA T deficienc y an d FED . The recreatio n an d expression o f selected mutant s o f LCA T ha s bee n use d to investigat e th e functiona l significanc e o f different inherite d defect s o f the LCA T gene . 1.12 Mutagenesi s an d Expressio n o f Recombinan t LCA T The firs t geneti c defec t o f LCA T t o be reported was  als o the first  t o b e expressed i n mammalian cel l cultur e (Taramell i et  al.,  1990) . Expressio n vector s containing eithe r th e wild type o r mutan t LCA T (Arg147^Trp ) genomi c clone s wer e transiently transfecte d int o monke y kidne y COS- 1 cells . Despit e comparabl e level s of mRNA transcribe d fro m norma l an d mutate d alleles , LCA T protei n an d activit y i n the cultur e mediu m coul d onl y b e detected fo r the wild typ e construct . I t was suggested tha t thi s mutatio n ma y caus e a  structural chang e resultin g i n a decrease d stability o f the enzym e insid e o r outside th e cell . 58 Table 10 . Natural mutation s o f the LCA T Codon Structura l Zygosit y Change 10 Pro-*Le u Homozygou s 10 32 83 93 119 123 135 141 144 147 156 158 183 209 228 244 252 293 300 321 347 375 391 Pro->Frameshift Leu->Pro Tyr-^Stop Ala^Thr Gly-*Frameshift Thr-lle Arg-*Trp Gly Insertio n Tyr-»Cys Arg^Trp Tyr-^Asn Arg^Cys Gly^Ser Leu-*Pro Asn-»Lys Arg^Gly Met-*Lys Met->lle Leu deletion Thr-lle Thr^Met lle-^Frameshift Asn^-Ser Homozygous Heterozygous Heterozygous Homozygous Heterozygous Homozygous Heterozygous Homozygous Heterozygous Heterozygous Heterozygous Homozygous Heterozygous Homozygous Homozygous Homozygous Homozygous Homozygous Homozygous Heterozygous Heterozygous Heterozygous Heterozygous LCAT Def , familia l LCA T deficiency ; FED , protein. Phenotype Referenc e FED LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . FED LCAT Def . LCAT Def . FED LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . LCAT Def . FED LCAT Def . FED LCAT Def . FED Skretting an d Pyrdz , 1992 Bujo et  al.,  199 1 McLean, 199 2 Klein et  al.,  1993 a Funke et  al.,  199 3 Gotoda et  al.,  199 1 Funke et  al.,  1991 b Funke et  al.,  199 3 Gotoda et  al.,  199 1 Contacos et  al.,  199 3 Taramelli et  al.,  199 0 Klein et  al.,  1993 a Funke et  al.,  199 3 McLean, 199 2 Funke et  al.,  199 3 Gotoda et  al.,  199 1 McLean, 199 2 Skretting et  al.,  199 2 Maeda et  al.,  199 1 Klein et  al.,  1993 b McLean, 199 2 Klein et  al.,  199 2 Funke et  al.,  199 3 Pritchard et  al., unpublished observations eye disease 59 The applicatio n o f site-directed mutagenesi s an d the i n vitro expressio n o f recombinant LCA T (rLCAT ) ha s provide d ne w insight s int o the catalyti c functio n o f LCAT (Francon e an d Fielding , 1991a ; Francon e an d Fielding , 1991b ; Francon e et al., 1993 ; Qu et  al.,  1993) . These sam e tool s hav e bee n use d to recreat e natura l mutations o f the LCA T gen e i n order t o establish thei r functiona l significance . Th e i n vitro expressio n o f rLCA T containin g a  Tyr156-*Asn substitutio n i n huma n embryoni c kidney-293 cell s indicate d tha t thi s mutan t enzym e was  secrete d inefficientl y an d was onl y partiall y activ e (Klei n et  al.,  1993a) . I t was suggeste d tha t thi s mutatio n may caus e irregula r folding  o f the protei n resultin g i n reduced secretio n and/o r rapi d degradation o f the enzyme . I n addition, i t was speculate d tha t the Tyr156^As n mutation ma y disrup t th e function o f an cr-helix conformatio n predicte d t o spa n residues 156-16 9 o f human LCAT . In a recent repor t b y Klei n et  al.  (1993b), th e expressio n o f rLCA T containin g a Leu30 0 deletio n reveale d norma l mRN A an d intracellula r LCA T concentrations , but onl y very  lo w amount s o f secreted rLCA T whic h ha d a  normal specifi c activity . Thus, th e author s sugges t tha t the FED-lik e symptom s presen t i n the patien t wit h this defec t aris e from a  quantitative rathe r tha n a  functional deficienc y o f the enzyme. The continuin g analysi s o f mutant recombinan t enzyme s shoul d increas e ou r understanding o f the structure-functio n relationship s o f LCA T an d provid e a  basis t o investigate th e specifi c mechanism s responsibl e fo r the biochemica l an d clinica l abnormalities associate d wit h thi s disease . 60 1.13 Rational e The relationshi p betwee n decrease d level s o f plasma HD L an d a n increase d risk o f coronary arter y diseas e i s well established . However , ther e ar e severa l rar e familial disorder s o f HD L metabolis m i n which, despit e ver y lo w level s o f HD L cholesterol, affecte d individual s d o no t appear t o b e at an increase d ris k fo r premature atherosclerosis . Familia l LCA T deficienc y an d fish ey e disease ar e two examples o f such inherite d disorders . Thus , to determine th e basi s o f this paradox , a mor e detaile d understandin g o f lipoprotein metabolis m i n LCAT deficienc y syndromes i s necessary . The applicatio n o f site-directed mutagenesi s an d i n vitro expressio n enable s the recreation , expressio n an d functiona l analysi s o f a variety inherite d defect s o f the LCA T gene . Throug h th e descriptio n o f the molecula r basi s o f the phenotypi c heterogeneity whic h exist s fo r LCA T deficienc y syndromes , w e ca n identif y relationships betwee n specifi c structura l change s an d the functional propertie s o f the enzyme. According t o the hypothesi s o f reverse cholestero l transport , LCA T play s a key rol e i n the deliver y o f excess cellula r cholestero l fro m periphera l tissue s t o the liver fo r excretion . A s a  result , i f we ar e to fully understan d th e rol e o f LCA T i n reverse cholestero l transport , w e mus t determin e th e factors tha t gover n it s abilit y t o esterify cholestero l i n different lipoprotei n classes . Therefore , a  complete understanding o f the relationshi p betwee n LCA T functio n an d HD L metabolis m i s essential fo r the characterizatio n o f the anti-atherogeni c propertie s o f HDL . 61 1.14 Specifi c Aim s (i) T o establish th e expressio n o f recombinan t LCA T i n mammalian cel l culture a s a  model to analyse th e structure an d function o f huma n plasma LCAT . (ii) T o create a  series o f natura l mutation s associate d wit h familia l LCA T deficiency an d fish ey e disease throug h PC R site-directe d mutagenesis, (iii) T o express thes e mutan t cDN A construct s b y transient transfectio n int o monkey kidne y COS- 1 cells , (iv) T o establish th e functiona l significanc e o f each mutatio n throug h th e measurement o f the specifi c activit y usin g syntheti c HDL-lik e substrates an d heat-inactivate d plasma , (v) T o create a  series o f stably transfecte d bab y hamste r kidne y (BHK ) cel l lines which secret e greate r quantitie s o f recombinan t protei n allowin g the analysi s o f the lipoprotei n substrat e specificit y fo r each mutan t enzyme usin g syntheti c HDL-lik e substrates , LD L an d VLDL. 62 2 MATERIAL S AN D METHOD S 2.1 Material s A ful l lengt h LCA T cDN A containe d withi n pUC1 9 wa s kindl y supplie d b y John McLean , Genentech , Inc. , San Francisco . Restrictio n an d modificatio n enzymes fo r the manipulation  o f DNA sequence s wer e purchase d fro m Bethesd a Research Laboratories  (BRL , Burlington , Ont.) , Pharmacia-LK B Biotechnolog y (Bai e d'Urfe Que.) , an d Boehringe r Mannhei m Corporatio n (BMC , Laval , Que.) . Enzyme s and reagent s fo r DN A sequencin g wer e fro m Unite d State s Biochemica l (USB , Cleveland, Ohio) . Geneclea n an d MERmai d kit s (Bio101 , L a Jolla, Ca. ) wer e use d for the purificatio n o f double strande d DN A fragments . Radiolabele d product s suc h as [ 35S]-dATP, [ 35S]methionine, [ 3H]cholesterol an d [ 14C]methylated protei n molecula r weight marker s wer e obtaine d eithe r fro m Ne w Englan d Nuclea r o r Amersha m Canada Ltd . Before autoradiography , fixe d gel s wer e equilibrate d wit h Amplif y (Amersham Canada  Ltd. ) an d exposed o n X-Omat A R fil m fro m Eastman-Kodak . Oligonucleotides wer e prepare d i n the Oligonucleotid e Synthesi s Laboratory, Department o f Biochemistry , UBC . Th e pNU T vecto r (Palmite r et  al.,  1987 ; Fun k et al., 1990 ) an d BH K cell s wer e kindl y provide d b y Dr . Ros s MacGillivray , Departmen t of Biochemistry , UBC . COS-1 , a n SV40-transformed Africa n Gree n Monke y kidne y cell line , was obtaine d fro m the American Typ e Cultur e Collectio n (Rockville , Md.) . All tissue cultur e reagent s includin g feta l bovin e seru m (FBS) , Dulbecco' s Modifie d Eagle Mediu m (DMEM ) an d reduce d seru m mediu m (Opti-MEM ) wer e supplie d b y Gibco-BRL (Mississauga , Ont. ) Methotrexat e (Cyanami d Canada , Inc. , Montreal , Que.) fo r selectio n mediu m was  obtaine d i n sterile salin e fro m th e University Hospital Pharmacy . 63 Endoglycosidase H , N-glycanas e an d tunicamycin wer e supplie d b y Boehringer Mannhei m Corporatio n wherea s neuraminidas e (Clostridium  perfringens, type V) wa s obtaine d fro m th e Sigm a Chemica l Co . (St . Louis , Mo. ) Polyclonal goa t anti-huma n LCA T antibodie s wer e kindl y provide d b y Dr . Andras Lacko , Texa s Colleg e o f Osteopathic Medicine , Universit y o f Nort h Texas , Fort Worth . Agarose-immobolized Protei n G  used fo r the immunoadsorptio n o f recombinant LCA T was  supplie d b y Pharmacia-LK B Biotechnology . Electrophoresi s grade reagents , protei n G  conjugated t o horseradis h peroxidas e (Protei n G-HRP ) and a  protein assa y ki t were obtaine d fro m BioRa d (Richmond , Ca.) . The concentratio n o f unesterified cholestero l i n plasma an d isolate d lipoproteins wa s determine d enzymaticall y b y a  reagent ki t (Boehringe r Mannheim) . All othe r chemical s wer e o f reagen t grad e o r bette r an d were purchase d eithe r fro m Sigma Chemica l Co . or from BD H Inc . (Vancouver , B.C. ) 2.2 Molecula r Biolog y Method s 2.2.1 Growt h an d Transformatio n o f E.  Coli E. Coli  strains DH5c r o r MC106 1 wer e maintaine d i n LB (10g/L tryptone , 5g/ L yeast extract , 10g/ L NaCI) . Froze n bacteria l stock s wer e prepare d i n 20% glycero l from lat e log-phas e brot h culture s an d were store d a t -70°C. Competent E.  Coli  were prepare d b y inoculatin g a  single colon y int o a  20 m l TYM brot h (2 0 g/L tryptone , 5  g/L yeas t extract , 5.8 4 g/ L NaCI , 1.2 0 g/ L MgS0 4). After th e growt h o f the cell s ha d achieved midlo g phas e (OD 600= 0.2-0.8) , the y wer e added t o a  10 0 m l o f TYM. After a n OD 600 o f 0.5-0.9 was  reached , th e cultur e was diluted wit h TY M t o 50 0 ml . When a n OD o f 0.6 wa s reached , th e cultur e was 64 rapidly coole d o n ice , pellete d a t 5000 x  g  for 1 5 min, and resuspende d o n ice i n 100 m l o f cold 30m M KAcetate , 50m M MnCI 2, 100m M KCI , 10m M CaCI 2 and 15 % glycerol. Th e mixtur e was  pellete d a t 5000 x  g for 1 0 min and resuspende d i n 20 m l of cold 10m M NaMOPS , 75m M CaCI 2, 10m M KC I and 15 % glycerol. Aliquots o f 0. 1 and 0. 5 m l were froze n i n liquid nitroge n an d stored a t -70°C. Transformation was  accomplished  b y gently mixin g competen t cell s wit h plasmid DN A i n an ice bath . Afte r 3 0 min , the mixtur e was  hea t shocke d a t 37°C for 5 min an d then dilute d 1:1 0 i n LB and incubate d wit h gentl e agitatio n a t 37°C fo r 9 0 min. Mixtures  wer e plate d ont o LB-aga r plate s containin g 100//g/m l ampicilli n an d incubated inverte d a t 37°C fo r 16h . All plasmid s use d i n these studie s containe d th e B-lactamase gen e allowin g growt h o f the transformants i n the presenc e o f ampicillin. 2.2.2 Purificatio n o f DNA 2.2.2.1 Smal l Scal e Plasmi d Preparatio n Aliquots o f 1. 5 m l of L B broth containin g ampicilli n (10 0 //g/ml) wer e inoculated wit h a  single bacteria l colon y an d were incubate d a t 37°C for 16 h with vigorous agitation . Bacteri a wer e pellete d b y microcentrifugation an d resuspende d i n 100//I o f 25mM Tris-HC I p H 8.0 , 50m M glucose , 10m M EDTA . Subsequently , th e cells wer e lyse d with 200  //I 0.2M NaOH , 1 % SDS an d neutralize d wit h 15 0 //I of 5M KAcetate p H 4.8. Cellula r debri s was  remove d b y microcentrifugatio n an d th e supernatant wa s extracte d wit h a n equal volume o f phenohchloroform (1:1) . Afte r centrifugation t o separate phases , th e uppe r aqueou s laye r i s collected an d th e plasmid DN A precipitate d b y the additio n o f 2  volumes o f absolute ethanol . Afte r microcentrifugation, th e pelle t i s washed wit h 70 % ethanol t o remov e coprecipitatin g 65 salts. Finally , th e pelle t i s air-dried an d dissolved i n 50 //I of sterile wate r o r TE (10mM Tris-HC I p H 8.0 , 0. 1 mM EDTA) . 2.2.2.2 Larg e Scal e Plasmi d Preparatio n To obtain large r quantitie s o f plasmid DN A suitabl e fo r eucaryoti c cel l transfection a  5  ml overnight cultur e wa s use d t o inoculat e 25 0 m l o f selective L B broth. Afte r a  16 h incubatio n wit h vigorou s agitation , cell s wer e pellete d a t 2800 x  g and resuspende d o n ic e i n 5 ml of 25mM Tris-HC I p H 8.0 , 5 0 mM glucose , 1 0 mM EDTA. Afte r 2 0 min , the mixtur e wa s transferre d t o sterile 3 0 m l polycarbonat e screw ca p tubes befor e cell s wer e lyse d with 1 0 ml o f 0.2M NaOH , 1 % SDS b y gentle inversion . Th e lysat e was  the n treated wit h 7. 5 m l of 5M KAcetat e p H 4.8. After 1 0 min, the mixtur e i s centrifuged a t 25,000 x  g for 1 5 min and the supernatan t transferred t o a  50 m l polypropylene scre w ca p tube . Thi s mixtur e i s digested wit h 50/yg o f ribonucleas e fo r 2 0 min a t 37°C. The DN A solutio n i s precipitated wit h 2 volumes o f ice-col d 95 % ethanol a t -20°C for 2 0 min . The DN A i s pelleted b y centrifugation an d washed wit h 70 % ethanol befor e i t i s resuspended i n 2 m l of sterile water . Th e mixtur e i s transferred t o a 1 5 ml polypropylene scre w ca p tub e and extracte d twic e wit h phenokchlorofor m (1:1 ) an d onc e wit h chlorofor m befor e transferring 1. 6 m l o f DN A solutio n t o a  1 5 ml silanized glas s tub e (Corex) . Afte r th e addition o f 0.4 m l o f 5M NaC I and 2. 0 m l of 13 % polyethylenegiycol (PEG-8000) , th e plasmid DN A i s precipitated o n ice for 60 min . Following centrifugatio n (12,00 0 x  g, 15 min), the pelle t i s redissolved i n 200-500 // I of TE and quantitated b y ultraviole t absorbance a t 260 nm . 66 2.2.3 Oligonucleotide-directe d Mutagenesi s A serie s o f mutagenic oligonucleotide s wer e synthesize d whic h correspon d t o natural mutation s identifie d withi n th e LCA T gen e (Tabl e 11) . Usin g a  mutageni c oligonucleotide a s primer , eac h mutatio n was  introduce d int o the LCA T cDN A contained withi n th e pUC1 9 vecto r b y the polymeras e chai n reactio n (PCR) . Firs t round amplificatio n wit h PC R wa s carrie d ou t b y usin g 1 0 pg of template DNA , 10 0 pmole o f mutagenic oligonucleotid e a s a forward primer , an d a  10 0 pmol e o f the reverse pU C prime r whic h hybridize s t o the vecto r downstrea m fro m the LCA T cDNA sequence . PC R was  carrie d ou t fo r 3 0 cycles , wit h ste p cycle s o f 95°C fo r 3 0 sec, 55° C fo r 3 0 sec, an d 74° C for 60 sec. Afte r th e thirtieth cycle , th e reactio n wa s extended a t 74°C for 5  min . After purificatio n o f the PC R product , a  second roun d o f PCR was  performe d whic h containe d 2 0 pg of the LCA T cDN A templat e i n pUC19, 10 pmole o f universa l primer , an d the purifie d firs t roun d amplificatio n produc t a s the other primer . PC R condition s wer e th e sam e a s those outline d i n the firs t round , except tha t th e extensio n tim e for eac h cycl e was  increase d t o 90 sec. 67 Table 11 . Oligonucleotides use d fo r site-directed mutagenesi s o f an LCA T cDNA. Structural Chang e Pro1(M_eu Ala93->Thr Thr123^lle Arg135-*Trp Arg158^Cys Leu209-Pro Met252->Lys Leu300 deletio n Thr347^Met He375-FS Asn391^Ser Mutagenic Prime r 5'-CTTCCCCC TGCACACCACGC-3' 5'-TGTCCAACACCCCTGGT-3' 5'-CCTGCACATACTGGTGC-3' 5'-GCTACGTG TGGGACGAG-3' 5'-AGTACTAC rGCAAGCTC-3' 5'-CATCTCTCCTGGGGCTC-3' 5'-CCCCTGGAAGTTTCCCT-3' 5'-GTCACGTGACCTGGCAGGAC-3' 5'-TGATGACATGGTGGCGA-3' 5'-CACGGGATAACAGCATC-3' 5'-GCACATCAGTGCCATCC-3' The sequenc e modification s o f the mutageni c primer s ar e printe d i n italic . Th e altered codo n i s underlined . FS , frameshift . 2.2.4 Isolatio n o f cDNA Fragment s PCR product s o r fragments obtaine d fro m the restrictio n enzym e digestio n o f plasmids wer e purifie d b y agarose ge l electrophoresis . Specifi c band s wer e excise d from th e ge l and DN A was  recovere d accordin g t o the protocol s outline d fo r Geneclean o r MERmai d (Bio101) . Th e quantit y o f produc t was  estimate d b y visualization o f the DN A i n ethidium-stained agaros e gels . 2.2.5 DN A Sequenc e Analysi s The sequencin g o f the differen t mutan t LCA T cDNA s was  carrie d ou t t o 68 confirm th e presenc e o f the mutatio n an d the absenc e o f any othe r anomalou s mutations. Doubl e strande d sequencin g was  performe d b y the enzymati c chai n termination metho d (Sange r era/. , 1977 ) usin g a  modified T 7 polymeras e (Sequenase, Unite d State s Biochemical) . Onl y thos e fragments generate d b y the PCR wer e subjecte d t o sequencing . Reaction s wer e performe d accordin g t o the manufacturers recommendations . Approximatel y 2  //g o f plasmid DN A wa s incubated i n 20 fj\ o f 0.2M NaO H an d 0.2m M EDT A fo r 1 5 min a t roo m temperature . The mixtur e wa s neutralize d b y the additio n o f 3// I o f 3M NaAcetat e an d the n precipitated wit h 75// I of absolute ethanol . After th e pelle t was washe d an d dried , i t was resuspende d wit h 5  pmol o f sequencing prime r i n Sequenase reactio n buffer . The mixtur e was  incubate d a t 37°C for 20 min an d subsequentl y a t roo m temperature fo r 1 0 min. The anneale d prime r was  the n simultaneousl y extende d an d labeled wit h Sequenas e i n the presenc e o f 1.5//M dGTP, dTTP , dCT P an d 2.5//C i o f cr-[35S]-dATP ( 5 min a t roo m temperature) . Chai n terminatio n was  the n achieve d b y removing 2.75// I o f the labelin g reactio n mixtur e int o 1.25// I of a prewarmed solutio n containing dideoxynucleotides , fou r reaction s correspondin g t o the fou r type s o f bases. Terminatio n reaction s wer e performe d a t 37°C fo r 5  min an d stopped b y the addition o f Stop Solutio n (95 % formamide, 20m M EDTA , 0.05 % Brompheno l blue , 0.05% xylen e cyanol e FF) . Labeled product s wer e separate d o n 6% polyacrylamide gel s containin g 1  x TBE (89m M Tris , 89m M bori c acid , 2m M EDTA ) an d 8 M urea . At the completio n o f separation, th e ge l was drie d ont o Whatman 3M M chromatograph y pape r an d exposed t o autoradiographic fil m fo r 16-7 2 hours . 69 2.2.6 Constructio n o f Expression vecto r The amplifie d DN A was  purifie d an d digested wit h specifi c restrictio n enzymes t o generate DN A fragment s whic h encompasse d th e desire d mutation . These fragment s wer e inserte d a s a  cassette i n the wild typ e LCA T cDN A containe d within th e pUC1 9 vector . Th e mutan t cDNA s wer e excise d fro m positiv e clone s an d then transferre d t o the mammalia n expressio n vecto r pNU T (Palmite r et  al.,  1987) . In one case , a  construct containin g tw o mutation s a t codons 9 3 and 15 8 was mad e through th e selectio n o f restrictio n enzyme s t o create tw o fragments eac h containin g a different mutatio n whic h wer e subsequentl y ligate d int o the pNU T expressio n vector. Al l final construct s wer e sequence d t o confirm th e presenc e o f the desire d mutation. Th e pNU T vecto r contain s SV4 0 an d mous e metallothionei n promoter s a s well a s a  mutant for m o f the dihydrofolate reductas e (DHFR ) gen e permittin g th e selection o f cells stabl y transfecte d wit h the plasmi d DN A b y their surviva l i n hig h concentrations o f methotrexate (Fun k et  al.,  1990) . 2.3 Eucaryoti c Cel l Cultur e COS-1 an d BH K cell s wer e maintaine d i n Dulbecco' s Modifie d Eagl e mediu m (DMEM) containin g 10 % fetal bovin e seru m (FBS) . FB S was heat-inactivate d a t 56°C fo r 3 0 min prio r t o us e a s a  growth supplement . Cel l culture s wer e maintaine d in a humidified incubato r wit h a  5% C0 2 atmosphere . 2.3.1 Transien t Transfectio n o f COS-1 Cell s The pNU T expressio n vecto r containin g eithe r wil d typ e o r a  mutan t LCA T cDNA wa s transientl y transfecte d int o COS- 1 cell s b y DEAE-dextra n transfectio n 70 (Kriegler, 1990) . Subconfluen t COS- 1 cel l monolayers wer e washe d twic e wit h transfection buffe r (25m M Tris-HCI , p H 7.4 , 140m M NaCI , 1m M CaCI 2, 3m M KCI , 0.5mM MgCI 2, 0.9m M Na 2HP04) befor e incubatio n o f the cell s wit h th e DN A solutio n at 37°C fo r 3 0 min . Subsequently , th e DN A solutio n was  remove d an d replace d wit h DMEM containin g 10 % FBS and 80// M chloroquine . Afte r a  3h incubation, transfection wa s complete d b y 3  min incubatio n wit h DME M containin g 10 % DMS O at roo m temperature . Afte r washin g wit h transfection buffer , th e cell s wer e incubate d in DME M containin g 10 % FBS for 12 h and subsequentl y i n serum-free mediu m (Opti-MEM)) fo r 48h. 2.3.2 Stabl e Transfectio n o f BH K Cell s To mediat e th e transfection  o f BH K cells , coprecipitate s o f plasmid DN A an d CaP04 wer e prepare d (Kriegler , 1990) . Plasmi d DN A (20-3 0 //g) i n 0.50 m l of 0.25 M CaCI2 wa s mixe d with 2 X HEPES-buffere d salin e (2 X HBS ) containin g 40m M HBS , pH 6.96 , 280n M NaCI , 10m M KCI , 1.5m M Na 2HP04 and 10m M glucose . Th e calcium phosphate-DN A mixtur e wa s incubate d a t room temperature fo r 3 0 mi n before i t was adde d dropwis e t o a  10 0 m m culture dis h containin g a  50% confluen t BHK monolayer . Followin g a n overnight incubatio n a t 37°C unde r 5 % C0 2, th e transfection mediu m wa s replace d wit h DMEM/10 % FB S for 2 4 hour s befor e transfected cell s wer e selecte d ove r a  period o f 10-1 4 day s i n DMEM/10% FB S containing 50 0 /JM methotrexate. Survivin g colonie s wer e transferre d t o 20m m culture well s an d grown to confluency unde r selecte d conditions . Clone s expressin g maximal quantitie s o f LCA T wer e identifie d b y solid-phase LCA T immunoassay . 71 2.3.2.1 Tunicamyci n Treatmen t o f BH K Ceil s Tunicamycin was  dissolve d i n 95% ethanol t o produc e a  stock solutio n o f 200 ug/ml. Confluen t BH K cel l monolayers wer e incubate d i n serum-free Opti-ME M containing 10-25 0 ng/m l tunicamycin fo r 24h . The cultur e mediu m was  collecte d an d the protei n mas s an d enzym e activit y wer e determined . Cell s wer e remove d fro m the culture dis h with a  rubbe r policema n an d LCA T activit y wa s measure d i n the cellular lysate . 2.3.3 Endogenou s Radiolabelin g o f Recombinan t LCA T BHK o r COS-1 cell s were incubate d i n methionine-free DMEM  (DMEM-Met ) for 2 0 minute s a t 37°C to deplete th e methionin e pool . Th e endogenou s methionin e pool was the n labelle d fo r 3 0 minutes i n DMEM-Me t supplemente d wit h 100-20 0 //Ci/ml [ 35S]methionine. Subsequently , th e labellin g mediu m wa s remove d an d the cells wer e incubate d wit h DMEM/10 % FBS . After a  specified incubatio n period , th e medium was  collecte d an d the cellula r protei n harveste d i n lysis buffe r (50m M Tris -HCI, pH8.0 , 62.5m M EDTA , 1 % Nonidet P40 , 0.4 % sodiu m deoxycholate , 1m M phenylmethylsulfonylfluoride) fo r subsequen t analysis . 2.3.3.1 Immunoadsorptio n o f Recombinan t LCA T Solid phas e immunoadsorptio n wa s use d to detect th e presenc e o f LCA T i n the cultur e mediu m an d the cellula r lysate . Polyclona l goa t anti-huma n LCA T antibodies wer e pre-adsorbe d ont o agarose-immobilize d protei n G  for 3 0 minutes a t 4°C. An aliquo t o f medium o r cellula r lysat e was  adde d t o the antibody-protei n G -agarose suspensio n an d the mixtur e wa s rotated  overnigh t a t 4°C. After 72 centrifugation an d washing, th e adsorbed materia l wa s then elute d fro m the agaros e beads b y heating a t 90°C i n the presenc e o f 2X sodium  dodecy l sulfat e (SDS ) sample buffe r (0.1 M Tris-Hcl , pH6.8 , 2 % SDS, 40% glycerol) . Th e bead s wer e removed b y centrifugation an d the supernatan t recovere d fo r electrophoreti c analysis. 2.4 Protei n Analysi s 2.4.1 Enzymati c Deglycosylatio n o f LCA T Neuraminidase (Clostridium  perfringens,  typ e V, Sigm a Chemica l Co. ) was used to remov e th e siali c aci d residue s fro m LCA T b y adding 3  //I of neuraminidas e (1U/ml) t o 7// I o f culture mediu m o r purifie d huma n LCAT . Sample s wer e incubate d at 37°C fo r 1h . N-glycanase digestio n was  performe d b y adding 1 0 //I of 0.9 M sodium phosphate , p H 8. 7 an d 4  //I of 10 % Nonidet P-4 0 t o 1 5 //I of sample . Th e mixture was  boile d fo r 1 0 min, cooled an d 1  fj\ o f N-glycanas e (250U/ml ) wa s adde d before incubatio n a t 37°C overnight . Endoglycosidas e H  digestion wa s carrie d ou t b y adding 1 0 //I of 0.15M sodiu m acetat e buffer , pH5. 8 an d 4 //I of 10 % Nonide t P-4 0 to 1 5 fj\ o f sample. Th e mixtur e wa s boile d fo r 1 0 min, cooled an d 2  //I o f endoglycosidase H  (1U/ml) was  adde d befor e incubatin g a t 37°C fo r 16h . 2.4.2 SDS-Polyacrylamid e Ge l Electrophoresi s Samples containin g LCA T wer e mixe d 1: 1 wit h 2 X SD S sampl e buffe r containing 10 % S-mercaptoethano l an d 0.1% Bromphenol Blue . The mixtur e was boiled fo r 5  min prio r to loadin g ont o a  10% polyacrylamide ge l ru n a t a  constan t current o f 15mA/ge l fo r 45 min . [ 14C]methylated protei n molecula r weigh t marker s 73 were use d a s standards . Befor e autoradiography , th e fixed gel s wer e equilibrate d with Amplify an d dried ont o Whatman 3M M chromatograph y paper . Autoradiogram s were expose d o n X-Omat A R fil m fo r 16-72h . 2.4.3 Immunoblo t Analysi s Polyacrylamide gel s wer e electroblotte d ont o nitrocellulos e pape r (0.45//M ) a s described b y Towbin et  al.  (1979). Transfe r was  achieve d a t 100 V for 1 h in a  coole d chamber containin g blo t buffe r (25m M Tris , 192m M glycine , 20 % methanol) . Unoccupied nitrocellulos e site s wer e blocke d wit h 5 % non-fa t powdere d mil k i n PBS (30 min) . Membrane s wer e the n incubate d successivel y wit h polyclona l goa t anti -human LCA T antibodie s an d Protei n G-HR P t o detec t th e LCA T protein . 2.4.4 Purificatio n o f Recombinant LCA T Approximately 30 0 m l of culture mediu m containin g rLCA T was  loade d ont o a phenyl-Sepharose colum n (10x15cm ) whic h wa s previousl y equilibrate d wit h 0.005 M sodium phosphate , 0.3 M sodiu m chloride , p H 7.4 . The colum n wa s washe d wit h th e same buffe r unti l the absorbanc e (280nm ) fel l belo w 0.01 . Subsequently, th e rLCA T was elute d with deionize d water . 2.4.5 Quantitatio n o f LCAT Protei n The immunoassa y o f LCAT was carrie d ou t b y using nitrocellulos e membranes (0.45// M por e size ) a s a solid-phase support . Sample s containin g eithe r culture mediu m o r purified huma n recombinan t LCA T standar d wer e boun d t o the membrane i n a BioRad slo t blo t apparatus . Th e membran e wa s subsequentl y 74 blocked wit h 5 % non-fa t dr y mil k i n PB S fo r 3 0 min a t 37°C . Polyclona l goa t anti -human LCA T antibodie s wer e incubate d wit h the membran e overnigh t a t roo m temperature i n PB S containin g 0.5 % non-fa t dry  milk . Th e membran e wa s the n washed thre e time s (5min ) i n PB S containin g 0.02 % Twee n befor e i t was reacte d with Protei n G  conjugated t o horseradis h peroxidas e fo r 60 min . The membran e wa s washed agai n thre e time s an d then developed i n 50 m l of PB S containin g 2 5 m g diaminobenzidine (Sigm a Chemica l Co.) , 1 5 mg CoCI 2 an d 0.01 0 m l of 30% H 202 to visualize th e protein . The blo t was scanne d usin g a  BioRad Vide o Densitometer (Model 620) . The interassa y coefficien t o f variation was  7.3 % fo r a  single protei n measurement i n nin e separate assays . Tota l protei n determination s wer e carrie d ou t with a  BioRad protei n assa y ki t usin g bovin e immunoglobuli n a s the standard . 2.4.6 Preparatio n o f Plasm a an d Lipoprotein s Blood wa s collecte d fro m norma l volunteer s afte r 12 h fasting an d plasm a wa s prepared b y low spee d centrifugatio n (1,200g , 2 0 min) . The differen t lipoprotein s were isolate d b y sequentia l ultracentrifugatio n usin g lipoprotei n fraction s define d b y their densities : d<1.00 6 g/m l fo r VLDL an d 1.006<d<1.06 3 g/m l fo r LDL . Tota l lipoprotein-depleted plasm a wa s prepare d b y ultracentrifugation a t density 1.2 1 g/ml . All lipoprotei n fraction s wer e dialyze d extensivel y a t 4°C agains t 0.0 1 M Tris-HCI (p H 7.4) containin g 0.15 M NaC I and 0.005 M EDTA . Plasm a was  hea t inactivate d a t 56°C fo r 3 0 min to eliminate endogenou s activity . Th e concentratio n o f unesterifie d cholesterol i n these differen t fraction s wa s determine d enzymaticall y b y a  reagen t kit. 75 2.4.7 Measuremen t o f LCA T Activit y 2.4.7.1 Exogenou s Substrate s The enzym e activitie s o f wild type an d mutan t LCA T gen e product s wer e determined usin g HDL-lik e proteoliposome s containin g [ 3H]cholesterol, phosphatidylcholine an d apolipoprotein A-l . The egg yol k phosphatidylcholine:cholesterol liposom e wa s prepare d b y ethanol injectio n according t o Batzr i an d Ko m (1973) . Th e substrat e mixtur e containin g 4.6 6 nmo l [3H]cholesterol (0.03//Ci/nmol) , 18.4 6 nmo l phosphatidylcholine , an d 7. 5 jjg  purifie d human ap o A- l i n 10m M Tris-HC I (p H 7.4)-150mM NaCI-5m M EDT A was preincubated a t 37°C fo r 3 0 min . Subsequently , 5m M B-mercaptoethano l an d 1.5 % bovine seru m albumi n (essentiall y fatt y aci d free) wer e adde d t o the substrat e mixture. Th e reactio n wa s initiate d b y the addition o f either 10 0 //I of cel l cultur e medium fro m transfected COS- 1 cell s o r 50 /J\ of medium fro m transfecte d BH K cells. Th e reactio n was  carrie d ou t a t 37°C fo r 1  h and was terminate d b y the addition o f 4 m l of chloroform.methanol (2:1 ) an d incubate d fo r two hour s a t roo m temperature t o extrac t lipids . Labele d cholestero l an d cholestery l este r wer e separated b y thin-layer chromatograph y o n silica ge l layer s incubate d i n petroleu m ethendiethyl ethenaceti c aci d (70:12:1 ) an d radioactivit y wa s determine d b y liqui d scintillation spectrometry . 2.4.7.2 Endogenou s Substrates The abilit y o f rLCAT t o esterify cholestero l was  als o determine d b y usin g [3H]cholesterol labelle d plasm a o r lipoprotei n fraction s a s a  source o f substrate . Afte r heat-inactivation, radiolabele d plasm a o r lipoprotein s wer e prepare d b y equilibratio n 76 with [ 3H]cholesterol a t 4°C as described b y Dobiasov a et  al.  (1992). An aliquo t o f culture mediu m containin g rLCA T (10 0 //I for transfected COS- 1 cell s o r 5 0 /J\ for transfected BH K cells ) was  adde d t o the radiolabele d substrat e (100-15 0 nmo l o f unesterified cholesterol/ml) , 5m M B-mercaptoethano l an d 1.5 % bovin e seru m albumin (essentiall y fatt y acid-free) . Th e reactio n mixtur e wa s incubate d a t 37°C fo r 4h. Th e reactio n was  terminate d b y adding 2  ml of ethanol . Cholestero l an d cholesteryl este r wer e separate d b y thin-layer chromatograph y an d radioactivit y determined b y liquid scintillatio n counting . The result s wer e expresse d a s nmole s o f cholesteryl este r forme d pe r h  per //g o f rLCAT protein . 77 3 RESULT S 3.1 Expressio n o f Recombinan t LCA T i n BHK Cell s The autoradiogra m i n Figure 5  indicates tha t BH K cell s transfecte d wit h pNUTLCAT secrete d a  predominant protei n tha t was  precipitabl e b y antibodie s specific fo r huma n plasm a LCAT . Thi s protei n migrate d a s a  broad ban d typica l o f glycosylated protein s i n SDS-polyacrylamide gel s an d spanne d a  molecular weigh t of 64,000-67,000. Thi s i s consistent wit h previou s determination s fo r purifie d huma n LCAT (Marcel , 1982) . N o LCA T protei n was  detecte d i n the mediu m fro m thos e cell s transfected wit h th e pNU T vecto r containin g n o LCA T cDN A insert . 3.1.1 Propertie s o f Recombinan t LCA T Aliquots o f serum-free mediu m wer e collecte d a t severa l tim e interval s fro m 35mm cultur e dishe s containin g BHK-LCA T cell s an d assaye d fo r the presenc e o f LCAT protei n an d activity . A s shown i n Table 12 , the level s o f both LCA T protei n and activit y continue d t o increas e durin g a  48h incubatio n period . Th e averag e specific activit y o f the LCA T enzym e calculate d fro m thes e value s o f protei n an d activity was  3.75±0.3 9 nmol/h///g . Th e specifi c activit y o f rLCAT wa s simila r t o plasma LCA T (5.64±0.3 2 nmol/h/^g ) (Funk e et  al.,  1991 ) an d LCA T purifie d fro m plasma (0.75±0.0 8 nmol/h///g ) usin g the sam e substrate . I n addition, th e abilit y o f apo A- l t o activate bot h rLCA T an d huma n plasm a LCA T wer e compared . Bot h enzymes demonstrate d a  typical saturatio n curv e with a  maximal activatio n betwee n 5 and 7.5 //g o f apo A-l/assay (Figur e 6) . After numerou s cel l passages , bot h the activity an d protei n concentratio n o f rLCAT secrete d fro m thi s stabl e BH K cel l lin e have remaine d consistent . 78 69 kDa -46 kD a -30 kDa -Figure 5 . Expressio n o f rLCAT . BH K cell s transfecte d wit h pNU T (control ) o r pNUTLCAT (wt ) plasmid s wer e subsequentl y labelle d wit h [ 35S]methionine fo r a  30 min pulse . Afte r 6 h o f incubation , mediu m wa s collecte d an d rLCA T wa s immunoadsorbed a s described i n Materials an d Methods . Th e immunoadsorbe d protein wa s reduced  an d electrophoresed i n 10 % SDS-polyacrylamide gels . Afte r ge l drying, radioactivit y was  detecte d b y autoradiography. Molecula r weigh t standard s are indicate d i n kilodaltons . Table 12 . Secretion o f rLCA T protei n an d activit y fro m BHK-LCA T cell s int o serum-free medium . Hours o f Incubation 24 30 48 Protein (A/g/ml) 4.36 6.91 10.30 Activity (nmol/h/ml) 17.90 26.39 34.30 Specific Activit y (nmol/h///g) 4.11 3.82 3.33 Confluent BHK-LCA T cell s wer e incubate d i n serum-free Opti-ME M fo r the indicate d time intervals . LCA T activit y an d protei n wer e determine d a s described i n Material s and Methods . Activit y unit s ar e expressed a s nmo l o f cholesterol esterifie d pe r hou r per ml . The value s depicte d ar e the mea n o f two separat e experiments , eac h don e in duplicate. 79 50 4 0 -0 1 2 3 4 5 6 7 8 9 1 0 Apoprotein A-l (ug ) Figure 6 . Activation o f plasm a LCA T an d rLCA T a s a  function o f apo A- l concentration. Assay s wer e performe d b y usin g singl e bilaye r cholesterol:lecithi n (1:4) vesicles . LCA T activit y was  measure d usin g 0.01 5 m l of plasma ( • ) o r culture medium containin g rLCA T (• ) an d expressed a s nmo l o f cholestery l este r forme d per hou r pe r ml . Data point s ar e mean s o f duplicate assays . 3.1.2 Productio n an d Purificatio n o f Recombinan t LCA T The LCA T cDN A i s unde r th e contro l o f the mous e metallothionei n promote r which ca n b e stimulated b y the presenc e o f divalent cation s (Stuar t et  al.,  1984) . Due t o it s lowe r cel l toxicity, th e effec t o f Zn2+ io n concentration o n the secretio n o f rLCAT fro m the BHK-LCA T cell s wa s investigated . A s show n i n Figure 7 , a  Zn2+ io n concentration o f 20//M produce d th e highes t secretio n rat e o f rLCAT. 80 o S 12 10 8 6 4 2 A 20umZn + + / o  25u m Zn+ + / &  15u m Zn + + / //  *  10umZ n + + / / / A  5u m Zn + + 1// /  *  NoZ n + + r x  \  i  1  i  1  .  I . i . 20 4 0 6 0 8 0 Incubation Tim e (h ) 100 Figure 7 . Effec t o f Zn2+ io n concentration o n the secretion o f rLCAT . Differen t concentrations o f ZnS04 wer e adde d t o a suspended cultur e o f BHK-LCAT cell s after th e initiatio n o f incubation i n serum-free medium . Cultur e mediu m wa s analysed fo r LCA T activit y expresse d a s a percentage o f cholesterol esterifie d pe r hour. After a  72 hou r incubatio n i n serum-free Opti-MEM , BH K cell s attache d t o microcarrier bead s secrete d rLCA T a t level s exceedin g lOpg/ml . Th e secrete d rLCAT wa s purifie d b y phenyl-Sepharose chromatograph y wit h th e elution profil e depicted i n Figur e 8 . Followin g elutio n wit h deionize d water , 97 % o f the origina l activity was  recovere d resultin g i n a 29-fold purificatio n (Tabl e 13) . A homogeneou s 81 LCAT protei n was  elute d fro m the colum n a s observed b y a  single ban d followin g SDS polyacrylamid e electrophoresi s (dat a no t shown) . LCAT Protein 0 2 0 4 0 6 0 8 0 10 0 12 0 Column Fractio n Figure 8 . Elutio n profil e o f Pheny l Sepharos e CL-4 B chromatography . A n aliquot (30 0 ml ) o f culture mediu m containin g rLCA T secrete d fro m BH K cell s wa s applied t o a  phenyl sepharos e CL-4 B colum n (10x1 5 cm ) equilibrate d wit h 0.00 5 M potassium phosphat e buffe r (p H 7.4 ) containin g 0. 3 M  sodium chloride . Th e colum n was washe d wit h th e same buffe r unti l the absorbance (280nm ) fel l below 0.0 1 a t which poin t th e rLCA T wa s elute d wit h deionize d wate r (arrow) . Th e protei n absorbance an d LCA T activit y expresse d a s a  percentage o f cholestero l esterifie d per hou r were determine d i n each fraction. 82 Table 13 . Purification o f rLCAT secrete d b y transfected BH K cells . Fraction Culture Medium Phenyl-Sepharose eluent Volume (ml) 350 110 Protein (mg) 149 4.9 Specific Activit y (units/mg) 105 3,090 Purification (fold) 29 Yield (%) 100 97 3.1.3 Enzymati c Deglycosylatio n o f Plasm a LCA T an d Recombinan t LCAT To compar e th e carbohydrat e structur e o f purified huma n LCA T an d rLCAT , each was selectivel y deglycosylate d wit h eithe r neuraminidas e t o remov e siali c aci d residues o r N-glycanas e t o diges t N-linke d carbohydrat e chains . Figur e 9  shows a Western blo t o f an SDS polyacrylamid e ge l which demonstrate s tha t bot h enzyme s have equivalen t molecula r weight s befor e an d afte r enzymati c deglycosylation . However, i t i s apparent tha t rLCA T migrate s a s a  broader  ban d compare d t o plasm a LCAT. Digestio n wit h neuraminidas e was  accompanie d b y a reduction i n molecula r weight whic h was  consisten t wit h previou s dat a (Chun g et  al.,  1979 ; Colle t an d Fielding 1991 ) indicatin g tha t siali c aci d make s u p a significant portio n o f the tota l carbohydrate mass . I n addition, reactio n wit h N-glycanas e reduced  th e apparen t molecular weigh t o f LCA T to about 48,000 , whic h was  comparabl e t o the calculate d molecular weigh t fo r the matur e protei n o f 47,090 (McLea n et  al.,  1986a) . I n addition, th e remova l o f the N-linke d carbohydrat e significantl y decrease d th e siz e heterogeneity o f rLCAT. However , bot h plasm a LCA T an d rLCA T wer e resistan t t o digestion wit h endoglycosidas e H  (data no t shown) whic h onl y remove s N-linke d 83 oligosaccharides tha t terminate wit h mannos e residues . Thes e result s indicat e th e assembly o f either comple x o r hybri d oligosaccharid e chain s i n the Golg i apparatu s prior t o the secretio n o f LCA T fro m the cel l (Tarentin o et  al.,  1974) . 66.2 k D -45.0 k D -1 2 3  4  5  6 Figure 9 . Comparison o f the effects o f enzymatic deglycosylatio n o f huma n plasma LCA T an d rLCAT . Th e molecula r weigh t o f human plasm a LCA T an d rLCAT wer e determine d b y SDS polyacrylamid e ge l electrophoresis . LCA T protei n was blotte d ont o nitrocellulos e membrane s an d detected immunologicall y a s described i n Material s an d Methods . Lan e 1 , untreated rLCAT ; Lan e 2 , untreate d human plasm a LCAT ; Lan e 3 , neuraminidase-treated rLCAT ; Lan e 4 , neuraminidase-treated huma n plasm a LCAT ; Lan e 5 , N-glycanase-treate d rLCAT ; Lane 6 , N-glycanase-treate d huma n LCAT . 3.1.4 Tunicamyci n Treatmen t o f BH K Cell s 3.1.4.1 Quantitatio n o f Secreted LCA T Protei n an d Activit y The effec t o f tunicamycin treatmen t o n the secretio n an d enzym e activit y o f rLCAT i s shown i n Table 14 . Tunicamycin inhibit s th e synthesi s o f the oligosaccharide chai n o n the lipi d carrie r (dolicholpyrophosphate ) thereb y preventin g 84 the additio n o f oligosaccharides t o asparagine residues . Tunicamyci n concentration s of 10-25 0 ng/m l were adde d t o the cultur e medium . Afte r 24 h o f incubation , untreated cell s secrete d 8.5 7 //g/ml o f LCAT protei n bu t les s than 1. 0 //g/ml wa s observed i n tunicamycin-treated cells . Similarly , LCA T activit y fro m th e cultur e medium o f treated cell s wa s 5-12 % o f the activit y observe d fro m the mediu m o f untreated cells . Highe r concentration s o f tunicamycin ( 5 //g/ml) di d no t furthe r reduce th e level s o f either LCA T protei n o r activity (dat a no t shown) . Ther e wa s n o activity associate d wit h the cellula r fractio n o f tunicamycin-treated cells . However , low level s o f LCA T activit y (0.2 9 nmol/h/ml ) coul d b e detected withi n untreate d cells . Table 14 . Effec t o f tunicamycin o n the secretion o f rLCAT . Tunicamycin Concentration (ng/ml) 0 10 25 50 100 250 LCAT Protein (//g/ml) 8.57 1.31 0.84 0.71 0.93 0.50 LCAT Activit y (nmol/h/ml) 2.50 0.30 0.27 0.22 0.23 0.17 Confluent BHK-LCA T cell s wer e incubate d fo r 24h i n Opti-MEM containin g tunicamycin. Esterificatio n rate s were measure d ove r a  period o f 30 min usin g 0.05 0 ml of culture mediu m i n the presenc e o f apo A-l . LCAT protei n was  measure d immunologically a s described i n Materials an d Methods . 3.1.4.2 Immunoadsorptio n an d Western Blottin g o f Recombinan t LCA T Immunoadsorption o f the cellula r lysat e an d cultur e mediu m (Figur e 10 ) 85 indicated tha t unglycosylate d LCA T wit h a n approximate  molecula r weigh t o f 46,000 was accumulatin g withi n tunicamycin-treate d cells . I n contrast, a  partiall y glycosylated intermediat e for m o f LCAT wit h a n estimated molecula r weigh t o f 55,000 wa s detecte d withi n untreate d cells . Onl y full y glycosylate d LCAT , however , appeared i n the cultur e mediu m regardles s o f tunicamycin treatment . However , pretreatment o f the cell s wit h tunicamycin fo r 2 h prio r to a  24 hou r incubatio n eliminated th e activit y an d the protei n i n the cultur e mediu m (dat a no t shown) . o »$ > nJ > t>° 4 > °# o 66.2-45 .0-31.0-21.5-14.4-T i • • -f f I . -.. „ , . .  ~ — i • >^ <$> # ^ ° 4 CELLS MEDIA Figure 10 . Effect o f tunicamycin o n the processin g o f rLCAT . Immunoadsorbe d protein fro m th e cellula r lysat e an d the cultur e mediu m wa s reduce d an d electrophoresed i n 10 % SDS-polyacrylamide gels . After blotting , LCA T protei n was detected immunologicall y a s described i n Materials an d Methods . Th e concentratio n of tunicamycin i n ng/ml i s indicated abov e eac h lane . 86 3.2 Biochemica l Analysi s o f Selected Familie s wit h Inherited  Disorders  o f LCAT Deficienc y The geneti c an d biochemica l analysi s o f published report s o f patients wit h familial LCA T deficienc y an d FE D i s summarized i n Table 15 . I n this grou p o f families, highe r concentration s o f plasma LCA T protei n an d a n increase d cholestero l esterification rat e (CER ) wer e use d t o distinguish th e FE D phenotyp e fro m familia l LCAT deficiency . However , th e rang e o f LCAT activitie s an d concentration s reporte d for LCA T deficienc y syndrome s suggest s tha t neithe r o f these parameter s ca n necessarily provid e a  definitive diagnosis . I n several instances , tw o differen t geneti c defects hav e bee n inherite d simultaneously . Fo r such cases , th e associatio n between eac h mutatio n an d the corresponding biochemica l phenotyp e remain s unclear. I n attempt t o resolv e thes e ambiguities , th e structura l change s indicate d i n Table 1 5 were re-create d individuall y b y site-directed mutagenesi s o f an LCA T cDNA. Th e initia l assessmen t o f the functional significanc e o f these mutation s wa s achieved b y the transient transfectio n o f COS-1 cells . 87 Table 15 . Structural change s an d biochemica l analyse s fo r a  group o f molecular defect s associate d wit h LCAT deficienc y syndromes . Structural Chang e Homozygosity: Pro10-*Le u Homozygosity: Ala93^Th r and Arg158^Cy s Homozygosity: Thr123->ll e Heterozygosity: Ar g 135-»Trp and Ile375-*F S Homozygosity: Leu209->Pr o Heterozygosity: Met252->Ly s and Asn391^Se r Heterozygosity: Thr123-»ll e and Thr347->Me t Homozygosity: Leu30 0 De l LCAT Protein 50% ND 50% 0% 3% 30% 50% 20% LCAT Activity 10% 3% 3% 0% 0% 10% 3% 14% CER 100% 0% 57% 0% 0% 64% 54% 42% Phenotype FED LCAT Def . FED LCAT Def . LCAT Def . FED* FED FED* Reference Skretting an d Prydz , 199 2 Frohlich era/. , 198 7 Funke et  al.,  199 3 Funke et  al.,  199 1 Funke era/. , 199 3 Funke era/. , 199 3 Frohlich et  al.,  198 7 Klein et  al.,  199 2 Klein et  al.,  1993 b LCAT protein , LCA T Activity an d cholestero l esterificatio n rat e (CER ) ar e expressed a s a  percentage o f normal. FED, fish eye disease ; LCA T Def. , LCA T deficiency ; FS , frameshift; ND , no t determined; Del , deletion. FED* , these cases wer e describe d a s ones which resemble d fis h eye disease. 3.3 Transien t Transfectio n o f COS-1 Cell s 3.3.1 Endogenou s Radiolabelin g o f Wild Typ e an d Mutan t LCA T Species Site-directed mutagenesi s was  use d t o recreat e severa l mutation s i n the human LCA T cDN A correspondin g t o the structura l change s liste d i n Table 15 . After the transfection o f each mutan t LCA T cDN A construct , th e synthesis an d secretio n of recombinan t LCA T (rLCAT ) wer e studie d b y a  pulse-chase experimen t wit h [35S]methionine. A  representativ e autoradiogra m depictin g th e expression o f (Thr123-*lle)rLCAT i s indicated i n Figure 11 . COS cell s transientl y transfecte d wit h wild typ e o r mutan t LCA T cDN A secrete d a  major protei n wit h a n apparen t molecular weigh t o f 67,000 whic h wa s comparabl e t o fully glycosylate d plasm a LCAT. Thi s LCA T protei n was  no t seen i n the cultur e mediu m isolate d fro m the cell s transfected wit h vecto r DN A whic h di d no t contain th e LCA T cDN A inser t (control) . In some experiments , a  faint protei n ban d with a  lower molecula r weigh t (47,000 ) was observe d i n the media . However , thi s ban d als o appeare d i n the mediu m fro m the cell s transfecte d wit h th e empty expressio n vector . Furthermore , a n equivalen t band wa s observe d whe n irrelevan t antibodie s wer e use d fo r the immunoadsorption . Thus, i t appears tha t the 47,000-molecula r weigh t protei n doe s no t originate fro m the LCA T cDNA . Immunoadsorptio n o f the cellula r lysat e reveale d a  specific protei n band wit h a  molecular weigh t lowe r tha n the matur e LCA T (52,000 ) i n al l experiments excep t th e cell s transfecte d wit h the expressio n vecto r DN A alone . Th e extent o f glycosylation o f this intermediate  for m o f rLCAT wa s investigate d an d i s described i n the nex t section . I n addition, a  specific protei n ban d o f molecula r weigh t 32,000 wa s observe d i n the cell s transfecte d wit h th e vecto r containin g th e 89 Thr123-»lle substitution . W e speculat e tha t thi s ma y b e a product o f degrade d mutant LCA T protein . 200 -97.4 -69.0 -46.0 -30.0 -14.3 -& w ' T CELLS * 4 MEDIA Figure 11 . Immunoadsorption o f wild typ e an d mutant  LCAT . Th e transfecte d COS-1 cell s wer e pulse-labele d wit h [ 35S]methionine (20 0 //Ci/ml) fo r 3 0 mi n followed b y a  chase incubatio n wit h nonradiolabele d methionin e fo r 4h. Subsequently, protein s i n the cultur e mediu m an d cell s wer e immunoadsorbe d wit h polyclonal anti-huma n LCA T antibodie s an d electrophoresed i n 10 % SDS polyacrylamide gels . The radiolabele d protein s o f wild type an d mutan t LCA T wer e visualized b y autoradiography . Migratio n positio n an d size (i n kilodalton) o f protei n standards ar e indicated . Figure 1 2 compares th e secretio n an d intracellula r form s fo r (Arg135^Trp)rLCAT an d (He375-^FS)rLCAT . Th e LCA T protei n containin g th e Arg135-*Trp mutatio n was  secrete d int o the mediu m an d ha d a  molecular weigh t identical t o tha t o f the full y glycosylate d wil d typ e rLCAT . I n contrast, n o secrete d 90 rLCAT coul d b e detected fo r the protei n containin g th e frameshif t mutatio n followin g codon 375 . However , analysi s o f the cellula r lysat e indicate d a  lowe r molecula r weight partiall y processe d for m o f LCAT fo r bot h mutan t enzymes . Th e intracellula r form o f LCA T correspondin g t o the frameshif t mutan t consistentl y ha d a  lowe r molecular weigh t whe n compare d t o either th e wild type o r the Arg135^Trp mutan t LCAT. Al l othe r mutan t rLCA T protein s expressed  i n COS cell s wer e indistinguishable fro m the wild type. op ^ op & Media Cells Figure 12 . Immunoadsorption o f two mutan t LCA T species . Th e transfecte d COS-1 cell s wer e pulse-labelle d wit h [ 35S]methionine (200//Ci/ml ) fo r 3 0 min followed b y a  chase incubatio n wit h non-radiolabelle d methionin e fo r 4h. Subsequently, protein s i n the cultur e mediu m an d cell s wer e immunoadsorbe d wit h polyclonal anti-huma n LCA T antibodie s an d electrophoresed i n 10 % SDS polyacrylamide gels . The radiolabele d protein s wer e visualized b y autoradiography . Migration positio n an d siz e (i n kilodaltons ) o f protein standard s ar e indicated . 91 3.3.1.1 Endoglycosidas e H  Digestion o f Intracellula r Recombinan t LCA T Digestion o f the intracellula r for m o f wild typ e rLCA T wit h endoglycosidas e H (endo H ) (Figure 13 ) reduce d th e molecula r weigh t t o abou t 47,00 0 whic h wa s comparable t o the calculate d molecula r weigh t o f the matur e unglycosylate d protei n (McLean etai,  1986a) . Thi s resul t indicate s th e presenc e o f high-mannose chain s normally associate d wit h intracellula r processin g i n the endoplasmic reticulu m (Tarentino etai,  1974) . Th e secrete d for m o f wild typ e rLCA T remaine d end o H resistant (dat a no t shown) . — + * LJ Figure 13 . Endoglycosidase H  treatment o f intracellula r wil d typ e recombinan t LCAT. Transfecte d COS- 1 cell s were pulse-labelle d wit h [ 35S]methionine fo r 3 0 min followed b y a chase incubatio n wit h non-radiolabelle d methionin e fo r 4h. Subsequently, protein s i n the cellula r fractio n wer e immunoadsorbe d wit h LCA T antibodies an d incubate d i n either th e presenc e (+ ) o r absence (- ) o f endo H . After polyacrylamide ge l elctrophoresis , th e radiolabele d protein s wer e visualize d b y autoradiography. Migratio n positio n an d siz e o f protein standard s ar e indicated . 92 kD 69 -46 -3.3.2 Establishmen t o f Enzyme Assa y Condition s fo r Wild Typ e Recombinant LCA T Secrete d b y Transfected COS- 1 Cell s The condition s fo r enzym e assay s containin g proteoliposome s wer e no t altered whe n compare d t o those describe d fo r huma n plasm a (Funk e et  al.,  1991b) . Nevertheless, linearit y o f the assa y fo r differen t time d interval s was  confirme d wit h rLCAT. However , specifi c assa y condition s wer e assesse d fo r heat-inactivate d plasma an d LD L substrates . Th e influenc e o f incubation tim e and unesterifie d cholesterol concentratio n o n LCA T activit y fo r heat-inactivate d plasm a i s depicted i n Figure 14 . The cholestero l esterificatio n rat e was linea r durin g 1- 6 h  of incubatio n a t an unesterifie d cholestero l concentratio n o f 40 nmol/ml . When th e rat e o f cholestero l esterification wa s determine d wit h variou s amount s o f plasma cholestero l (15-8 0 nmol/ml plasma) , th e activit y o f wild typ e rLCA T increase d an d then plateaue d afte r reaching concentration s o f unesterified cholestero l greate r tha n 40 nmol/ml . Sinc e this assa y involve s th e additio n o f exogenous rLCA T t o a  heat-inactivated plasm a substrate, i t i s importan t t o not e that th e measure d activit y i s no t directl y comparabl e to those value s reporte d previousl y fo r the cholestero l esterificatio n rat e (CER ) i n patient plasma . 93 0.5 1 2 3  4  5 Incubation tim e (h ) 0 2 0 4 0 6 0 8 0 Unesterified cholestero l (nmol/ml ) Figure 14 . Enzyme activit y o f wild typ e recombinan t LCA T secrete d b y COS- 1 cells fo r heat-inactivate d plasma . A : An aliquo t o f cel l culture mediu m containin g wild typ e rLCA T was incubate d wit h heat-inactivate d plasm a (4 0 nmol/ml ) pre -equilibrated wit h [ 3H]cholesterol. Th e incubatio n was  carrie d ou t a t 37° C for variou s time periods . Enzym e activit y was  determine d a s described i n Materials an d Methods an d was expressed  a s nmo l o f cholesterol esterifie d pe r //g LCA T protein . B: An aliquo t o f cel l culture mediu m containin g wil d type rLCA T wa s incubate d wit h various amount s o f heat-inactivated [ 3H]cholesterol plasm a fo r 5h a t 37°C. Enzym e activity wa s determine d an d expressed a s nmo l o f cholesterol esterifie d pe r h  per //g LCAT protein . The abilit y o f the secreted rLCA T protei n t o esterif y cholestero l i n the LD L fraction wa s als o assesse d (Figur e 15) . [ 3H]cholesterol was  pre-equilibrate d wit h LDL an d the esterificatio n o f cholesterol catalyze d b y rLCA T was  determined . Th e [3H]cholesterol esterificatio n rate s i n LDL catalyze d b y the wild type enzym e wa s very lo w bu t stil l measurable afte r a  3h incubation . Whe n increasin g amount s o f LDL cholesterol (20-10 0 nmol ) wer e adde d t o the assa y mixture , th e enzym e activit y increased bu t bega n t o plateau afte r unesterifie d cholestero l concentration s o f 50 94 nmol/ml wer e reached . However , du e to the limite d activit y observe d i n this assay , i t was no t possibl e t o accurately asses s th e specifi c activit y o f different recombinan t mutant enzymes . 1 3 ^-1 0. 8 £ c 0 '•£ 0. 6 o >c w 0) S °- 4 __ o »». d) | 0. 2 0 •c o A -/ / y^ S 1  1  1 1  1  i —s .3 Jc O £ c^ c <o o v . "C o mm^ o »-01 V) a o o 0.30 0.25 0.20 £ 0.1 5 0.10 0.05 0.0 0 1 2 3  4 5  6 Incubation time (h ) 0 2 0 4 0 6 0 8 0 10 0 Unesterified cholestero l (nmol/ml ) Figure 15 . Enzyme activit y o f wild type recombinan t LCA T secrete d b y COS- 1 cells fo r LDL . A: An aliquo t o f cel l culture mediu m containin g wil d typ e rLCA T was incubated wit h [ 3H]cholesterol LD L (8 0 nmol/ml ) a t 37°C for various tim e periods . Enzyme activit y wa s determine d a s described i n Material s an d Method s an d wa s expressed a s nmo l o f cholestero l esterifie d pe r /jg LCA T protein . B : An aliquo t o f cell culture mediu m containin g wil d type rLCA T was  incubate d wit h variou s amount s of labeled LD L for 5 h at 37°C . The enzym e activit y wa s determine d an d expresse d as nmo l o f cholesterol esterifie d pe r h  per //g LCA T protein . 3.3.3 Quantitation o f Secreted LCA T Protei n an d Activity fo r a  Series o f Mutant Recombinan t Enzyme s The dat a obtaine d fro m th e analysi s o f the secretio n o f rLCA T fro m transientl y transfected COS- 1 cell s i s summarized i n Table 16 . Abnormalities i n the secretio n o f rLCAT fro m COS- 1 cell s were detecte d fo r only thre e mutations . Ther e wa s approximately a  50% reduction o f the secreted protei n fo r the (Thr123^-lle)rLCA T 95 compared t o wild type , a  75% reduction fo r (Leu209-*Pro)rl_CA T an d a  complete absence o f secreted LCA T protei n fo r the frameshif t mutatio n followin g codo n 375 . Analysis o f the specifi c activit y o f these mutan t enzyme s fo r HD L analogue s revealed tha t al l mutants , wit h the exceptio n o f the Asn391^»Ser an d Arg158-*Cy s substitutions, ha d specifi c activitie s o f less than 10 % of wild type . LCA T specie s containing th e Ala93-»Thr mutatio n alon e o r i n combination wit h the Arg158^Cy s substitution ha d specifi c activitie s o f less than 10 % of wild type usin g eithe r HDL-lik e proteoliposomes o r hea t inactivate d plasm a a s substrate . I n contrast, th e mutan t LCAT containin g th e singl e amin o aci d exchang e Arg158-*Cy s ha d a  specific activit y of greater tha n 80 % of wild type fo r bot h proteoliposome s an d hea t inactivate d plasma. Fou r o f the mutation s associate d wit h FE D (Pro10->Leu , Th r 123-Mle , Thr347-*Met, Asn391-»Ser ) demonstrate d a  moderate increas e i n the proportio n o f cholesterol esterifie d whe n radiolabele d heat-inactivate d plasm a wa s use d a s substrate, whil e the res t o f the mutant s remaine d inactive . 96 Table 16 . Measurement o f LCAT protei n an d specifi c activit y fo r differen t mutant rLCA T specie s secrete d fro m transfecte d COS- 1 cells . Structural Chang e Wild Typ e Pro10-*Leu Ala93^Thr Ala93-»Thr/ Arg158-*Cys Thr123^lle Arg135^Trp Arg158^Cys Leu209^Pro Met252^Lys Leu300 Deletio n Thr347-*Met Ile375-*FS Asn391-*Ser Secreted Protein 100% 94±6 88±20 91±15 42±5 145±21 75±10 24±6 118±17 107±16 101±20 0 92±23 LCAT Activit y (HDL Analogue ) 100% 6±3 7±3 3±1 0.2±0.1 0 86±11 1±1 0.3±0.3 1±1 3±1 0 15±3 LCAT Activit y (Plasma) 100% 24±6 9±5 8±8 35±7 0.5±0.9 84±29 2±2 3±4 4±4 19±12 0 25±5 Culture mediu m collecte d fro m transfected COS- 1 cell s was  analyse d fo r LCA T protein an d LCA T activit y a s described i n Materials an d Methods . Specifi c LCA T activity usin g eithe r a n HD L analogue o r hea t inactivate d plasm a wa s determine d a s nmol o f cholesterol esterifie d pe r hou r pe r //g o f rLCAT protein . The data i s expressed a s a  percentage o f wild type : LCA T protein , 1.14±0.2 6 //g/ml (n=6) ; Specific LCA T activity : HD L analogue , 2.08±0.4 9 nmol/h/// g (n=6) ; Heat-inactivate d plasma, 0.39±0.0 6 nmol/h/// g (n=5) . Values show n ar e mea n ±  standar d deviatio n for 3-6 differen t experiments . Statistica l compariso n o f enzyme activit y value s o f the wild typ e LCA T an d each mutan t enzym e fo r both substrates indicate d tha t eac h mutant protei n wa s significantl y differen t fro m the wild type (p<0.001 ) wit h th e singl e exception o f the (Arg158^Cys)rLCAT . 3.4 Stabl e Transfectio n o f BHK Cell s 3.4.1 Endogenou s Radiolabelin g o f Wild Typ e and Mutan t LCA T Species With th e exceptio n o f the frameshif t mutatio n whic h faile d t o be secreted b y 97 COS-1 cells , al l o f the mutan t LCA T cDNA s wer e stabl y transfecte d int o BH K cells . The synthesi s an d secretio n o f these differen t mutan t rLCAT s wer e studie d b y pulse-chase experiment s usin g [ 35S]methionine. A s depicted i n Figure 16 , analysis o f the cellula r lysat e indicate d tha t al l o f the mutan t protein s wer e synthesize d an d were indistinguishabl e fro m th e wild typ e construct . A  partiall y glycosylate d LCA T with a n approximate molecula r weigh t o f 55,000 accumulate d intracellular^ . I n addition, a  protein o f about 46,00 0 molecula r weigh t was  presen t i n al l lane s bu t appeared t o be a non-specifi c immunoadsorbe d protei n a s i t was als o associate d with cell s transfecte d wit h the empty expressio n vecto r (control) . Analysi s o f the culture mediu m reveale d tha t al l bu t two o f the mutan t construct s wer e normall y secreted an d ha d a n apparent molecula r weigh t o f 66,000, comparabl e t o full y glycosylated plasm a LCAT . Th e LCA T immunoassa y wa s unable  t o detect an y LCAT protei n i n the mediu m fro m BH K cell s transfected wit h the Leu209-*Pr o mutant cDNA . However , lo w level s coul d b e consistently detecte d fo r the doubl e mutant Ala93->Thr/Arg158-*Cys . Th e quantit y o f rLCAT secrete d afte r a  48h incubation differe d amon g th e various mutan t BH K cel l clones an d range d fro m 2-1 2 //g/ml o f culture medium . 98 97 .4 -69 .0 -46 .0 -# < i • Cells 9 7 . 4 -6 9 . 0 -4 6 . 0 -Media Figure 16 . Immunoadsorption o f intracellula r an d secreted form s o f a series o f mutant LCA T species . Th e transfected BH K cell s wer e pulse-labelle d wit h [35S]methionine (200//Ci/ml ) fo r 3 0 min followed b y a  chase incubatio n wit h non -radiolabeled methionin e fo r 4h. Subsequently, protein s i n the cultur e mediu m an d cells wer e immunoadsorbe d wit h polyclona l anti-huma n LCA T antibodie s an d electrophoresed i n 10 % SDS polyacrylamid e gels . The radiolabele d protein s wer e visualized b y autoradiography. Migratio n positio n an d size (i n kilodaltons ) o f protei n standards ar e indicated . 3.4.2 Establishment o f Enzym e Assa y Condition s fo r Wild Typ e Recombinant LCA T Secrete d b y Transfected BH K Cell s In the sam e manne r a s previously describe d fo r COS- 1 cells , the influenc e o f incubation tim e an d unesterifie d cholestero l concentratio n o n LCA T activit y fo r LD L and VLDL substrate s wa s analysed . Figure s 1 7 and 1 8 indicate tha t a  4h incubatio n with a n unesterifie d cholestero l concentratio n greate r tha n 12 5 nmol/m l wer e sufficient t o provide linea r cholestero l esterificatio n rate s an d to maintai n a n exces s concentration o f LD L an d VLDL substrate . A  simila r kineti c analysi s was  performe d 99 for those mutan t recombinan t enzyme s whic h eithe r ha d norma l o r partia l reactivit y with LDL . Th e relativ e specifi c activite s o f these mutant s remaine d consisten t throughout th e indicate d concentratio n rang e o f unesterifie d cholestero l (0-15 0 nmol/ml). 3 - - v .3 0^  2. 5 s £ c 9 * » • TO O § 1. 5 a; « « o 1 w d) Vt 01 O 0- 5 •c O 0 A , / f i  i i  i _^ J3 ^C ^^ " o E c c o <0 *^  •-a> to o .^ , o ^ M w a> o •c o u.o 0.6 0.4 0.2 0 B / f 1  1 1  1 !  1 1 2 3  4 Incubation tim e (h ) 0 2 5 5 0 7 5 10 0 12 5 15 0 Unesterified cholestero l (nmol/ml ) Figure 17 . Enzyme activit y o f wild type recombinan t LCA T secrete d b y BHK cells fo r LDL . A: An aliquo t o f cel l culture mediu m containin g wil d typ e rLCA T was incubated wit h [ 3H]cholesterol LD L (12 5 nmol/ml ) a t 37° C for various tim e periods . Enzyme activit y was  determine d a s described i n Material s an d Method s an d was expressed a s nmo l o f cholesterol esterifie d pe r /jg LCA T protein . B : An aliquo t o f cell cultur e mediu m containin g wil d typ e rLCA T wa s incubate d wit h variou s amount s of labeled LD L fo r 4h a t 37°C . The enzym e activit y was  determine d an d expressed as nmo l o f cholesterol esterifie d pe r h  per //g LCA T protein . 100 u.o 3 =5 0.2 5 E c^ C 0. 2 o • * • <o o £ 0.1 5 « 0) •S 0. 1 o >. Q) « •2 0.0 5 o •c O A -/ ^ r / / i  i  i  i —-s en .3 « o E c_ c o <«*# <0 o <D M a> • o u. 0) CO o O 0.08-S 0.0 6 -•S 0.0 4 0.02 1 2 3  4 Incubation tim e (h ) 0 2 5 5 0 7 5 10 0 12 5 15 0 17 5 20 0 Unesterified cholestero l (nmol/ml ) Figure 18 . Enzyme activit y o f wild typ e recombinan t LCA T secrete d b y BH K cells fo r VLDL . A : An aliquo t o f cel l culture mediu m containin g wil d type rLCA T wa s incubated wit h [ 3H]cholesterol VLD L (12 5 nmol/ml ) a t 37°C for various tim e periods . Enzyme activit y was  determine d a s described i n Material s an d Method s an d was expressed a s nmo l o f cholesterol esterifie d pe r //g LCA T protein . B : An aliquo t o f cell culture mediu m containin g wil d typ e rLCA T wa s incubate d wit h variou s amount s of labele d VLD L fo r 4h a t 37°C. The enzyme activit y wa s determine d an d expresse d as nmo l o f cholesterol esterifie d pe r h  per jjg LCA T protein . 3.4.3 Lipoprotei n Substrat e Specificit y o f Wild Typ e an d Mutan t LCA T The enzymic activit y o f the secrete d rLCA T protei n was  determine d wit h either proteoliposome s containin g [ 3H]cholesterol (HD L analogues ) o r with isolate d lipoproteins radiolabele d wit h [ 3H]cholesterol. Fo r each activit y assay , th e substrat e was presen t i n excess a s determined b y the measuremen t o f maxima l esterificatio n rates fo r increasin g concentration s o f unesterifie d cholesterol . Th e specifi c activitie s of wild type an d mutan t enzyme s wer e calculate d withi n thes e define d parameters . As show n i n Table 17 , the specifi c activit y o f wild typ e rLCA T secrete d fro m BH K 101 cells wa s nearl y 10-fol d lowe r fo r LD L compared t o HD L analogues an d 100-fol d lower whe n VLD L was  use d a s substrate . Table 17 . Specific activit y o f wild typ e rLCA T secrete d fro m BH K cell s fo r selected substrates . Substrate HDL Analogue LDL VLDL Specific Activit y (nmo\/h/jjg) 3.42±0.45 (n=5 ) 0.52±0.07 (n=5 ) 0.056±0.005 (n=5 ) After a  48h incubation , cultur e mediu m fro m BH K cell s stabl y transfecte d wit h th e wild typ e LCA T cDN A was  collecte d an d analysed fo r LCA T activit y an d protei n a s described i n Material s an d Methods . Specifi c LCA T activit y usin g eithe r HD L analogues, LD L o r VLDL a s substrate was  expressed  a s nmo l o f cholestero l esterified pe r hou r pe r fjg o f rLCAT protein . Th e result s ar e depicted a s mea n ± standard deviatio n (numbe r o f experiments) . Table 1 8 compares th e specifi c LCA T activit y fo r the differen t mutan t rLCA T species usin g thes e differen t substrates . Wit h th e exception o f a  higher reactivit y fo r the Thr123->ll e mutant , th e result s obtaine d fo r the specifi c activit y fo r HD L analogues wit h transfecte d BH K cell s wer e very  simila r t o those reporte d fo r transiently transfecte d COS- 1 cell s (Tabl e 16) . Three o f the mutant s associate d wit h FED ha d averag e specifi c activitie s fo r HD L analogue s rangin g fro m 11-31 % o f wild type (Pro10-*Leu , Thr123-»lle , Asn391-»Ser ) whil e th e remainin g mutation s ha d activities rangin g fro m 0-6 % o f wild type . Difference s i n the specifi c activit y amon g the mutan t rLCA T specie s wer e mor e strikin g whe n LD L acte d a s the substrate . Activities comparabl e t o wild typ e were obtaine d fo r bot h homozygou s mutant s o f FED (ProlO^-Leu , Thr123^»lle) . Intermediat e activitie s wer e observe d fo r th e 102 Asn391-*Ser defec t whil e al l other mutant s ha d activitie s o f les s tha n 10 % of wild type. I n contrast, specifi c activitie s obtaine d fo r VLDL substrate s di d no t excee d 50% o f wild typ e fo r an y o f the mutations . I n some cases , th e relativ e activit y fo r VLDL substrate s was  highe r whe n compare d t o LD L while fo r other s i t decreased. However, i t should b e noted tha t the specifi c activit y o f wild typ e rLCA T fo r VLD L substrates wa s considerabl y lowe r whe n compare d t o LD L (Tabl e 17) . Table 18 . Specific LCA T activit y o f a  series o f mutan t rLCA T specie s fo r selected substrates . Structural Chang e Wild Typ e Pro10-*Leu Ala93^Thr/ Arg158-Cys Thr123-*lle Arg135^Trp Leu209^Pro Met252^Lys Leu300 Deletio n Thr347^Met Asn391^Ser HDL Analogu e 100% 11±3 6±3 31±9 0.1±0.1 0 0.3±0.3 4±1 3±1 18±4 LDL 100% 118±20 5±2 104±10 0.8±0.3 0 3±3 5±1 8±2 53±7 VLDL 100% 33±6 13±2 47±14 4±3 0 9±2 10±2 34±9 22±7 Culture mediu m collecte d fro m transfected BH K cell s was  analyze d fo r LCA T protei n and LCA T activit y a s described i n Material s an d Methods . Specifi c LCA T activit y using eithe r HD L analogues , LD L o r VLDL a s substrate was  expresse d a s nmo l o f cholesterol esterifie d pe r hou r pe r //g o f rLCAT protein . Th e dat a i s expressed a s a percentage o f wild typ e (specifi c activitie s indicate d i n Table 17) . Values show n ar e mean +  standard deviatio n fo r 4-5 differen t experiments . Statistica l compariso n o f enzyme activit y value s o f the wild type LCA T an d each mutan t enzym e fo r al l substrates indicate d tha t each mutan t protei n wa s significantl y differen t fro m th e wil d type (p<0.001 ) wit h th e exceptio n o f the activit y o f (Pro10-*Leu)rLCAT an d (Thr123^lle)rLCAT fo r LD L substrates . 103 4.0 DISCUSSIO N 4.1 Expressio n o f Recombinan t LCA T i n Mammalian Cel l Cultur e The transien t transfectio n o f COS-1 cell s an d the stabl e transfectio n o f BH K cells wit h a n expression vecto r containin g a n LCA T cDN A ha s enabled th e production o f recombinan t LCA T (rLCAT ) whos e propertie s closel y resembl e thos e of plasma LCA T wit h respec t t o molecula r weight , activatio n b y apo A-l , specifi c activity an d carbohydrat e content . However , fo r bot h cel l types, ther e wa s a  greate r size heterogeneit y associate d wit h rLCA T i n SDS polyacrylamid e electrophoresi s when compare d t o LCA T purifie d fro m plasma . Simila r observation s hav e bee n reported fo r recombinan t antithrombi n secrete d fro m BH K an d CH O cell s (Bjor k et a/., 1992) . I n each case , remova l o f the N-linke d carbohydrat e resulte d i n the migration o f a single shar p ban d wit h n o apparent differenc e i n mobility compare d t o the plasm a protei n suggestin g a  greater degre e o f heterogeneit y i n the glycosylatio n of the recombinan t protein . I t i s possible tha t difference s i n the specifi c activit y o f LCAT reporte d fo r differen t cel l line s suc h a s COS-1 cell s (3-1 2 nmol/h///g) , CH O cells (1- 2 nmol/h///g) , an d huma n embryoni c kidney-29 3 cell s (4 6 nmol/h///g ) coul d be attributed i n par t t o differences i n the glycosylatio n pattern s presen t i n each mammalian cel l type. However , i t i s likely tha t a  much greate r sourc e o f variability i s associated wit h th e differen t method s use d fo r substrate preparatio n an d th e measurement o f LCA T activit y a s well a s the variet y o f polyclona l antibodie s employed fo r LCA T protei n analysis . Tunicamycin markedl y reduce d th e secretio n o f rLCA T b y BHK-LCA T cell s even a t concentrations a s lo w as 1 0 ng/ml . Immunoadsorptio n o f the cellula r lysat e and cultur e mediu m reveale d tha t unglycosylate d LCA T wa s no t secrete d bu t 104 accumulated withi n th e cellula r fraction . Additional studie s i n our laborator y hav e demonstrated tha t th e remova l o f al l four N-linke d glycosylatio n site s o f LCA T b y site-directed mutagenesi s significantl y impair s th e secretion o f rLCAT b y bot h COS-1 and BH K cells . However , th e secretio n o f unglycosylated rLCA T protei n b y transfected Chines e hamste r ovar y cell s (CHO ) was  no t affecte d b y tunicamycin (Collet an d Fielding , 1991) . I t is possible tha t these differen t cel l types ma y vary  wit h respect t o their mechanism s o f glycosylation an d protei n transpor t withi n th e cell . However, i n COS-1 , BHK , an d CH O cells , N-linke d glycosylatio n o f LCAT i s necessary fo r norma l LCA T activity . Th e influenc e o f carbohydrates o n protei n expression an d function o f plasma LCA T stil l remain s unclear . A  comparativ e analysis o f the specifi c compositio n an d structure o f the carbohydrat e chain s fo r both rLCA T an d plasm a LCA T wil l b e necessary t o determine th e basi s fo r thes e functional differences . The transient transfectio n o f COS-1 cell s ha s enabled th e functiona l assessment o f a  large numbe r o f LCA T mutations . Defect s i n both the secretio n an d LCAT activit y o f different recombinan t enzyme s ar e easil y detected . However , th e assessment o f the abilit y o f these differen t mutan t enzyme s t o us e endogenou s substrates i s technically mor e demanding . Th e presenc e o f endogenous LCAT , th e necessity fo r heat-inactivatio n an d the presenc e o f a  heterogenous mixtur e o f lipoproteins associate d wit h huma n plasm a limit s th e exten t t o which th e result s o f specific activit y fo r thi s substrat e ca n b e interpreted . I n addition, ofte n th e quantit y o f rLCAT protei n obtaine d fro m transfecte d COS- 1 cell s doe s no t permi t mor e detaile d structural an d functional analyses . T o overcome thes e difficultie s large r quantitie s o f rLCAT wer e obtaine d fro m stabl y transfecte d BH K cells . Thi s permitte d th e 105 measurement o f the reactivit y o f rLCA T fo r isolate d lipoprotei n substrates . A s expected, th e abilit y o f rLCA T t o esterify cholestero l i n lower densit y lipoprotein s was considerabl y reduced , approximatel y ten-fol d lowe r fo r LD L an d 100-fol d lowe r for VLDL whe n compare d t o HD L analogues . The cultur e o f stably transfecte d BH K cell s i n conjunction wit h th e us e o f serum-free mediu m ha s als o facilitate d th e purificatio n o f large r quantitie s o f rLCA T by a  single ste p chromatograph y procedure . Th e simplificatio n o f the productio n an d purification o f rLCAT wil l pav e th e way fo r a  more i n depth analysi s o f the structur e and function o f this enzyme . However , althoug h th e expressio n o f rLCA T i n mammalian cel l culture ha s bee n establishe d a s a suitable mode l fo r the stud y o f human LCAT , th e exten t t o which th e extrapolatio n o f it s properties , especiall y it s cellular processing , t o the characteristic s o f human LCA T i n vivo remain s t o b e determined. 4.2 Compariso n o f the I n Vivo an d I n Vitro Biochemica l Characteristic s fo r a Series o f LCAT Mutations . 4.2.1 Pro10-*Le u Homozygosity fo r a  Pro10->Leu exchang e i n LCA T wa s reporte d fo r th e original FE D patient s i n Sweden (Skrettin g an d Pyrdz , 1992) . Thes e patient s ha d about 50 % of norma l plasm a LCA T protei n an d 10 % of norma l LCA T activity , bu t completely norma l cholestero l esterificatio n rate s (CER ) (Tabl e 15) . There wa s n o apparent abnormalit y i n either th e synthesi s o r the secretio n o f this mutan t rLCA T from transfecte d COS- 1 o r BH K cells . Th e enzym e activit y fo r HD L analogue s o f (Pro10-»Leu)rLCAT secrete d fro m bot h cel l types was  significantl y reduce d an d 106 similar t o those value s previousl y reporte d i n patient plasm a (Carlso n an d Holmquist, 1985b) . I n addition, i n the sam e manne r a s LCA T i n patien t plasma , th e ability o f this recombinan t enzym e t o esterify cholestero l i n LD L was  equa l t o wild type. I n contrast, th e specifi c activit y fo r VLDL substrate s was  reduce d t o onl y one -third o f wild type . Thus , the reactivit y o f (Pro10->Leu)rl_CAT fo r LD L an d VLD L substrates wa s no t equivalent suggestin g tha t the previousl y terme d G-LCA T activit y may b e restricte d t o LD L particle s only . Fro m these results , i t i s apparent tha t th e biochemical characteristic s o f (Pro10->Leu)rLCAT closel y correspon d t o th e properties o f this mutan t for m i n human plasma . 4.2.2 Ala93-*Th r an d Arg158^Cy s Site-directed mutagenesi s an d i n vitro expressio n i s particularly usefu l fo r cases i n which tw o differen t geneti c defect s hav e bee n inherite d simultaneously . I n most cases , thi s woul d occu r i n instances o f compound heterozygosity ; however , there ha s bee n on e repor t o f a patient wit h familia l LCA T deficienc y i n which tw o different mutation s ar e presen t o n each allel e (Ala93-*Thr , Arg158-»Cys ) (Funk e et ai, 1993) . To determine th e causativ e natur e o f these defect s an d to define th e extent t o which eac h contribute s t o the biochemica l phenotype , eac h mutatio n was re-created an d expresse d individuall y an d i n combination i n COS-1 cells . LCA T species containin g th e Ala93-»Thr mutatio n alon e o r i n combination wit h th e Arg158-*Cys substitutio n ha d specifi c activitie s o f less than 10 % of wild typ e fo r bot h HDL analogue s an d heat-inactivate d plasma . B y contrast , th e mutan t LCA T containing th e singl e amin o aci d exchang e Arg158-»Cy s ha d a  specific activit y o f greater tha n 80 % of wild typ e fo r bot h substrat e types . 107 Thus, th e characteristic s o f those mutan t recombinan t enzyme s containin g th e Ala93->Thr mutatio n wer e consisten t wit h th e biochemica l phenotyp e o f LCA T deficiency reporte d fro m the analysi s o f this patient' s plasm a (Funk e et  al.,  1993 ) whereas th e rLCA T containin g onl y th e Arg158-*Cys substitutio n ha d a  specifi c activity simila r t o the wil d typ e enzym e regardles s o f the substrate utilized . As a result, i t appears tha t thi s secon d mutatio n (Arg158-»Cys ) i s a natura l polymorphis m of LCA T whic h ha s littl e effec t o n enzyme function . Sinc e ther e ar e n o known DN A polymorphisms withi n th e LCA T gene , this mutatio n represent s th e firs t reporte d protein polymorphis m o f LCAT . However , th e frequency o f this mutatio n i n the normal populatio n i s presently unknown . As recentl y describe d b y Klei n et  al..  (1993a) , th e predictio n o f the secondar y structure o f LCA T woul d plac e Arg158 a s one o f the residue s o f the hydrophilli c fac e within a  segment o f cr-helix conformatio n spannin g residue s 15 6 to 169 . Compariso n of the complet e amin o aci d sequenc e o f human , baboon , pig , rabbit , mouse , an d ra t LCAT protei n indicate d tha t codo n 15 8 was th e onl y residu e fo r which ther e was  a different amin o aci d presen t fo r each specie s (Figur e 19) . I t should b e note d tha t al l other codon s containin g mutant s reporte d i n this stud y ar e conserve d amon g al l si x species o f LCAT. Ther e ar e a  variety o f amino acid s a t codon 15 8 amon g th e different specie s whic h includ e hydrophobic , hydrophilli c an d positivel y charge d residues. I n addition, th e i n vitro expressio n o f the rLCA T containin g a n Arg158-»Cys exchange ha d simila r propertie s whe n compare d t o the wild typ e LCAT . A s a  result , it would see m unlikel y tha t Arg158 play s a  critical rol e i n maintaining a  hydrophilli c face o f an amphipathic helix . However , anothe r mutatio n associate d wit h LCA T deficiency jus t tw o residue s fro m codo n 15 8 (Tyr156^Asn ) wa s show n t o hav e a 108 significant effec t o n both the secretio n an d specifi c activit y o f rLCAT secrete d b y transfected huma n embryoni c kidney-29 3 cell s (Klei n et  al., 1993a) . In keeping wit h th e findings i n COS-1 cells , the (Ala93->Thr/Arg158-*Cys)rl_CAT secrete d b y transfected BH K cell s wa s relativel y inactive fo r al l substrates . However , unlik e transfecte d COS- 1 cells , this doubl e mutant was  poorl y secreted . I t should b e noted tha t the quantit y o f rLCAT protei n secreted fro m BH K cell s ma y no t onl y b e affected b y the natur e o f the mutation , bu t also b y the numbe r o f plasmid copie s transfecte d pe r cell . A s a  result , cautio n should b e used when interpretin g th e significanc e o f the propertie s o f secretion o f rLCAT fro m cel l cultur e an d it s relationshi p t o what migh t occu r fo r hepatocyte s i n vivo. CodOll 15 3 15 4 15 5 15615 7 Human LCAT  Gi n Baboon LCAT  Gi n Pig LCAT  Gi n RabbitLCAT Gi n Mouse LCAT  Gin Rat LCAT  Gi n Glu Glu Tyr Tyr Glu Glu Tyr Tyr Glu Glu Tyr Tyr Glu Glu Tyr Tyr Asp Gl u Tyr Tyr Asp Gl u Tyr Tyr 158 Arg His Leu Giy Lys Gin 159 16 0 16 1 16 2 16 3 Lys Leu Ala Gly Leu Lys Leu Ala Gly Leu Lys Leu Ala Gly Leu Lys Leu Ala Gly Leu Lys Leu Ala Gly Leu Lys Leu Ala Gly Leu Figure 19 . Comparison o f the amin o aci d sequenc e o f a  segment o f LCA T among si x differen t species . Th e amin o aci d sequenc e o f residue s 153-16 3 i s compared betwee n huma n (McLea n et  al.,  1986a) , baboo n (Hixso n et  al.,  1993) , pi g (Pritchard et  al.,  unpublishe d observations) , rabbi t (Sha w an d Pritchard , unpublishe d observations), mous e (Warde n et  al.,  1990) , an d ra t (Meron i et  al.,  1990 ) LCA T protein sequences . Th e residue s a t codon 15 8 (boxed ) ar e non-conserve d amon g all si x species . 109 4.2.3 Thr123^ll e Homozygosity fo r a  Thr123-»lle mutatio n ha s no w bee n identifie d i n two families o f different ancestr y affecte d wit h FE D (Funk e et  al.,  1991 ; Kastelein et  al. 1992). Homozygote s o f these familie s hav e 50 % of norma l LCA T protei n levels , les s than 5 % o f norma l LCA T activit y an d abou t 60 % of norma l CER . Th e (Thr123-*lle)rl_CAT secrete d fro m transfected COS- 1 cell s displaye d th e sam e molecular weigh t a s the wild typ e protein , bu t it s protei n concentratio n i n culture medium wa s abou t 50 % of the wild type enzyme . Th e occurrenc e o f a  unique intracellular immunoadsorbe d protei n ban d o f approximately 32,00 0 molecula r weight associate d wit h thi s mutatio n suggeste d tha t increase d cellula r degradatio n of this mutan t protei n ma y occur . However , observation s o f this kin d wer e no t observed i n transfected BH K cells . The fac t tha t the (Thr123-*lle)rl_CA T secrete d b y BH K cell s ha d a  highe r reactivity fo r HD L analogue s compare d t o COS-1 cell s indicate s tha t variabilit y o f the characteristic s o f rLCA T secrete d fro m differen t mammalia n cel l types ma y exist . Nevertheless, a  significant reductio n o f the reactivit y o f (Thr123->lle)rLCAT fo r HD L analogues wa s observe d i n both cel l types. Also , i n the sam e manne r a s the origina l FED defec t (ProlO^Leu) , th e Thr123^ll e mutatio n di d no t impai r th e abilit y o f rLCA T to esterif y cholestero l i n the LD L fraction. Thus , these result s ar e consisten t wit h th e hypothesis o f Carlso n an d Holmquis t (1985b ) suggestin g tha t LCA T i n FE D plasm a esterifies cholestero l i n lower densit y lipoprotei n (G-LCA T activity ) an d tha t it s abilit y to esterify cholestero l i n HDL (cr-LCA T activity ) i s significantly reduced . 110 4.2.4 Arg135^Tr p an d lle375-*Frameshif t The plasm a analysi s o f both homozygote s i n this famil y wit h familia l LCA T deficiency reveale d a n absence o f LCA T protein , LCA T activit y an d CER . The y als o had typical cornea l opacitie s an d normochromi c anemia . However , onl y on e proban d developed kidne y disease . Th e mutan t LCA T containin g th e Arg135->Trp substitutio n was secrete d normall y fro m th e transfected COS- 1 cell s whil e th e frameshif t mutation resulte d i n the synthesi s an d partia l intracellula r processin g o f rLCAT , bu t not it s secretion . Th e insertio n o f an adenine followin g codo n 37 5 cause s a  shif t i n the readin g fram e suc h tha t the fina l 40 amino acid s o f the protei n ar e o f differen t composition compare d t o the nativ e sequence . A s a  result , a n N-linke d glycosylatio n site normall y presen t a t Asn384 i s eliminated an d the sto p codo n normall y presen t at positio n 41 7 no w occur s prematurel y a t codon 416 . The reductio n i n the molecular weigh t o f the intracellula r for m o f this mutan t woul d b e consistent wit h th e loss o f a  glycosylation attachmen t site . This i s supported b y additiona l studie s reported b y our laborator y whic h demonstrat e tha t th e los s o f a  single glycosylatio n site cause s a  reduction i n protein size , bu t does no t impai r th e secretio n o f rLCA T from transfecte d CO S cells . Therefore , th e inabilit y o f this mutan t t o b e secrete d may b e relate d t o a  marked conformationa l chang e i n the protei n induce d b y the altered amin o aci d sequenc e containe d withi n th e fina l 4 0 residues . Th e accumulation o f an endoglycosidase H  sensitive intracellula r for m o f rLCA T i s consistent wit h a  rate-limiting ste p determined b y the processin g o f secrete d proteins withi n th e endoplasmi c reticulu m (Ros e an d Doms , 1988) . I n the absenc e of any highe r molecula r weigh t form s o f mutant LCA T o r degradation product s withi n the cell s o r culture medium , i t i s possible tha t th e 37 5 frameshif t mutatio n cause s 111 this protei n t o b e retaine d withi n th e endoplasmi c reticulum . Analysis o f the enzym e activit y o f the (Arg135-*Trp)rLCA T fro m bot h transfected COS- 1 an d BH K cell s indicate d a  virtual absenc e o f activit y fo r eithe r HDL-like proteoliposomes , heat-inactivate d plasma , LDL , o r VLDL. Thus , bot h o f these re-create d mutation s wer e associate d wit h a  complete los s o f LCA T activity , but fo r differen t reasons . Th e frameshif t mutatio n afte r codo n 37 5 resulte d i n the inability o f the LCA T protei n t o be secreted whil e th e Arg135-»Trp substitutio n was associated wit h th e norma l secretio n o f a dysfunctional enzyme . Thi s i s consisten t with th e analysi s o f heterozygot e carrier s o f these tw o mutation s amon g who m n o significant difference s i n the biochemica l phenotyp e coul d b e detected (Frohlic h e f a/., 1988) . However , th e absenc e o f LCA T protei n fro m the plasm a o f homozygote s (Frohlich et  a/. , 1982 , 1988 ) suggest s tha t i f the Arg135-*Trp LCA T protei n i s normally secreted , i t i s probably rapidl y catabolized . Fro m these analyses , i t i s not possible t o determine i f the caus e o f such a  catabolism coul d aris e fro m a  lipi d binding defec t and/o r inactivatio n o f the catalyti c sit e o f the LCA T enzyme . 4.2.5 Leu209^Pr o Analysis o f the plasm a o f the proban d i n this famil y affecte d wit h familia l LCAT deficienc y indicate d onl y 3 % of norma l LCA T protei n level s an d n o detectabl e LCAT activit y o r CE R (Funk e et  ai,  1993) . I n a similar manne r t o the (He375->frameshift)rLCAT, th e (Leu209^Pro)rLCA T wa s synthesize d an d partiall y processed intracellular^ , bu t it s secretion wa s impaired . I t i s possible tha t th e introduction o f a  proline residu e i n the middl e o f the primar y amin o aci d sequenc e o f the LCA T protei n woul d resul t i n a conformational chang e associate d wit h defective 112 secretion and/o r a n increase d susceptibilit y o f intracellula r degradation . Therefore , the ver y lo w level s o f plasma LCA T protei n reporte d i n this famil y ma y b e a  resul t o f impaired secretio n o f mutant LCAT . However , analysi s o f the lo w level s o f (Leu209^Pro)rLCAT obtaine d fro m CO S cell s suggeste d tha t thi s enzym e i s also functionally defective . 4.2.6 Met252-*Ly s an d Asn391-*Se r The proban d o f this family was  initiall y reporte d t o hav e hypoalphalipoproteinemia resemblin g FE D (Frohlich  et  al.,  1987) . Plasm a LCA T protein wa s abou t 30 % of norma l an d plasm a CE R wa s 64 % of norma l bu t LCA T activity was  approximatel y 10 % of norma l suggestiv e o f a  higher specifi c activit y compared t o previou s report s o f FED . The proban d ha s a  bilatera l dens e cornea l arcus bu t no t the characteristi c cloud y cornea l opacitie s identifie d wit h eithe r FE D o r familial LCA T deficiency . I n addition, a n increased numbe r o f target cell s an d decreased osmoti c fragilit y o f the proband' s erythrocyte s wer e simila r t o thos e reported fo r homozygote s o f familial LCA T deficiency . The Met252->Ly s substitutio n wa s initiall y identifie d a s a  homozygous defec t in three o f the origina l Norwegia n LCA T deficienc y familie s (Skrettin g et  al.,  1992) . Although normall y secreted , th e (Met252^Lys)rLCA T obtaine d fro m eithe r transfected COS- 1 o r BH K cell s was  virtuall y unreactiv e fo r al l substrate types . Thi s confirms th e causativ e natur e o f this mutatio n an d i s consistent wit h th e biochemica l changes reporte d i n the Norwegia n LCA T deficien t patients . I n contrast, th e (Asn391->Ser)rLCAT appeare d t o hav e uniqu e characteristic s whic h wer e intermediate compare d t o those observe d i n FE D o r familia l LCA T deficiency . Thi s 113 mutation wa s associate d wit h a  partial reactivit y fo r both HDL-lik e an d LD L substrates. A s a  result , i t is no t surprising tha t a  compound heterozygou s defec t containing thes e mutation s i s associated wit h th e characteristic s o f both FE D an d familial LCA T deficiency . Thus , the natur e o f the clinica l an d biochemica l characteristics observe d fo r this patien t ar e reflecte d b y the differen t propertie s o f each inherite d defect . 4.2.7 Thr123-^l e an d Thr347^»Me t The clinica l an d biochemica l descriptio n o f the proban d wit h thi s compoun d heterozygous defec t wa s almos t identica l t o the report s o f patients wh o ar e homozygous fo r the Thr123->lle defect . Plasm a LCA T protei n i s about 50 % o f normal, LCA T activity , 3 % o f norma l an d CER , 54 % of normal . Despit e efficien t secretion, th e (Thr347^Met)rLCA T wa s associate d wit h a  poor reactivit y fo r al l substrates an d was  simila r t o the characteristic s o f rLCA T containin g th e substitutions Met252-»Ly s o r Arg135^-Trp. Thus, i n a similar manne r t o the Met252-»Ly s an d Asn391->Ser compoun d heterozygote, th e produc t o f one allel e was  associate d wit h a  complete los s o f LCAT activit y whil e the secon d mutatio n demonstrate d eithe r norma l (Thr123->lle ) o r partial (Asn391^-Ser ) S-LCA T activity . A s a  result , thes e dat a demonstrat e tha t th e mutations a t codons 12 3 an d 39 1 presen t o n a  single allel e gover n th e biochemica l and clinica l phenotyp e observe d i n these compoun d heterozygotes . I n addition, these studie s sugges t tha t th e presenc e o f only a  single allel e codin g fo r a  partiall y active LCA T i s sufficient t o maintai n nea r norma l level s o f plasma cholestery l ester . 114 4.2.8 Leu30 0 Deletio n The classificatio n o f 'FED ' i n this famil y was  initiall y base d o n the presenc e o f corneal opacitie s an d the absenc e o f hematologi c an d rena l abnormalitie s (Cler c et al., 1991) . However , i n a subsequent mor e detaile d report , th e analysi s o f the properties o f the mutan t LCA T reveale d severa l uniqu e characteristics . Th e proban d in this famil y ha s lo w level s o f plasma LCA T protei n (20 % of normal ) an d a  reduced CER (42 % of normal ) bu t interestingl y ha d an LCA T activit y o f 14 % of normal. These result s sugges t tha t thi s mutan t LCA T ha s a  highe r specifi c activit y compare d to othe r FE D mutant s (Klei n et  al.,  1993b) . I n addition, supportiv e evidenc e was provided b y the expressio n o f (Leu300 deletion)rLCA T i n huma n embryoni c kidney -293 cell s whic h reveale d a  poorly secrete d bu t functionally norma l enzyme . However, th e expressio n an d analysi s o f (Leu300 deletion)rLCA T fro m transfecte d COS-1 an d BH K cell s doe s no t concur wit h these results . I n both o f these cel l types, the (Leu30 0 deletion)rLCA T wa s secrete d normall y bu t ha d a  poor reactivit y fo r al l substrate type s used . As previousl y mentioned , difference s i n the secretio n an d activity o f rLCA T fro m differen t mammalia n cel l types fo r species containin g th e same mutatio n ha s bee n observed . Thus , difference s i n cel l type a s well a s the methods o f measuremen t o f LCA T activit y an d protei n fo r bot h the i n vivo an d i n vitro studie s reporte d b y Klei n et  al. (1993b) ma y contribut e t o the discrepanc y between thes e results . I t should b e noted tha t the measuremen t o f LCA T activit y fo r the othe r reporte d famil y studie s wa s performe d wit h Batzr i an d Kor n (1973 ) proteoliposomes (vesicula r structures) , th e sam e substrat e use d fo r the analysi s o f rLCAT secrete d b y transfected COS- 1 an d BH K cells . I n contrast, discoida l structures wer e use d b y Klei n et  al.  (1993b) i n their studies . Therefore , i t i s possibl e 115 that som e mutation s o f rLCAT ma y differentiate thei r abilit y t o utiliz e thes e differen t substrates. T o resolv e thi s issu e i t will b e necessary t o analyse th e plasm a LCA T activity o f this proban d usin g vesicula r substrates . 4.3 Geneti c an d Biochemica l Heterogeneit y o f LCA T Deficienc y Syndrome s Familial LCA T deficienc y an d fish eye diseas e ar e rar e autosoma l recessiv e disorders associate d wit h inherite d defect s withi n th e gen e encodin g fo r lecithin:cholesterol acyltransferase . I n both cases , cornea l opacitie s an d a  sever e HDL deficienc y ar e characteristi c feature s o f the disease . However , unlik e FED , familial LCA T deficienc y i s associated wit h sever e lipoprotei n abnormalitie s an d additional clinica l symptom s includin g a  mild hemolyti c anemia , proteinuria , progressive rena l insufficienc y an d rena l failure . Th e basi s fo r this segregatio n i s believed t o b e a  resul t o f functional difference s associate d wit h th e LCA T enzyme . In familial LCA T deficiency , ther e ar e very lo w level s o f plasma cholestery l este r an d a virtua l absenc e o f LCA T activit y fo r bot h syntheti c an d endogenou s substrates . I n contrast, th e CE R an d the concentratio n o f cholestery l este r i n FE D plasm a ar e nea r normal. Therefore , th e abilit y o f some LCA T mutant s t o utiliz e lipoprotein s othe r tha n HDL to synthesize cholestery l ester s i s directly relate d t o the clinica l phenotype . However, th e abilit y o f FE D LCA T t o esterify cholestero l i n HDL-lik e substrate s i s significantly reduced  an d i s similar t o the propertie s o f LCA T observe d i n the plasm a of patients wit h familia l LCA T deficiency . My curren t hypothesi s regardin g th e explanatio n o f the differen t biochemica l and clinica l abnormalitie s observe d i n LCA T deficienc y syndrome s i s depicted i n Figure 20 . I n both complet e an d partia l LCA T deficiency , th e abilit y o f LCA T t o ac t 116 upon prefi-HD L i s los t preventin g th e formation o f mature HD L causin g HD L deficiency. A s a  result , th e smal l spherica l an d discoidal  HD L accumulat e an d penetrate th e vasculatur e o f the corne a wher e ove r time they for m cornea l opacities . To maintai n a  cholesterol gradien t betwee n periphera l cell s an d the plasma , unesterified cholestero l i s redistributed t o LD L an d triglyceride-rich particles . However, i f LCAT i s unable t o esterify cholestero l i n the LD L fraction, the flu x o f lipids associate d wit h the lipolysi s o f chylomicrons an d VLDL particle s t o LD L i s impaired. Consequently , abnorma l lipoprotein s ric h i n triglyceride, unesterifie d cholesterol an d phospholipi d accumulate , deposi t i n the kidne y an d contribut e t o renal disease . O n the othe r hand , i f LCAT i s able to esterify cholestero l i n LDL , a more efficien t cholestero l gradien t ca n b e established throug h th e generatio n an d redistribution o f cholestery l esters . A s a  result , th e sever e lipoprotei n abnormalitie s normally associate d wit h familia l LCA T deficienc y d o no t occu r an d rena l diseas e does no t develop . However , i t i s importan t t o not e tha t i n both scenarios , cholestero l can stil l b e transported fro m periphera l tissue s t o the live r fo r degradation . Thi s ma y explain th e paradoxica l absenc e o f premature atherosclerosi s i n LCA T deficienc y syndromes despit e a  deficiency o f the matur e HD L pool . 117 1 Corneal opacitie s /' " Abnormal ( TG-ric h t U C V^P a r t i c l g^ y — • Rena l UC •  ^ 3 g= ^ (  i  Disease I Pre- B HDL V [Complete LCAT V LCAT  ^ (Deficiency ' "  — — L | V E R Partial LCA T Deficiency i UC J Corneal opacitie s Mature HDL LCAT* Pre-B HDL jr Figure 20 . The pathophysiolog y o f LCAT deficienc y syndromes . UC , unesterified cholesterol ; LCAT , lecithin.cholesterolacyltransferase ; TG ; triglyceride. The origina l hypothesi s tha t two types o f LCA T activit y exis t i n plasma, a-LCAT (actin g o n HDL ) an d S-LCA T (actin g o n LD L an d VLDL) i s supported b y the results fro m the analysi s o f recombinan t enzyme s containin g differen t FE D mutations. However , th e relativel y lo w reactivit y o f rLCAT fo r VLDL substrate s an d the decreased activit y o f rLCA T containin g FE D mutant s fo r VLDL sugges t tha t LD L is the primar y substrat e associate d wit h E-LCA T activity . I n addition, th e result s indicate tha t th e relativ e magnitud e o f a-  and E-LCA T activit y ma y diffe r amon g th e 118 various mutants . A s depicte d i n Figur e 21 , in the presenc e o f a  poor LCA T activit y for HD L analogues , th e abilit y o f mutan t rLCAT s t o esterify cholestero l i n LDL determines th e biochemica l an d ultimatel y th e clinica l phenotyp e observe d i n the patients wit h thes e inherite d defects . WT v. ' of rLCA . Analogi Activity HDL Wild Fund defec Type1 0 0 80 60 4U tional 2 0 ts of LCA T " — — 93/158 135 252 300 347 / 391 j ® ?§® I I  I 20 4 0 6 0 I / > A T n « * i « ; « « « LUA © 123 © 10 ® i i 80 10 0 Activity o f rLCA T fo r LD L Figure 21 . The functional heterogeneit y o f LCAT mutations . Individua l LCA T mutations wer e plotte d accordin g t o their relativ e abilit y t o esterify cholestero l i n LDL and HD L analogu e substrates . The numbe r o f different molecula r defect s identifie d i n the LCA T gen e s o far is consistent wit h the heterogeneit y o f the phenotypi c expressio n see n i n these disorders. I n addition, th e frequenc y o f compound heterozygou s defect s o f LCA T wil l contribute t o the variabilit y o f the characteristic s o f LCA T observe d i n vivo. However , given th e variabilit y o f the clinica l expressio n reporte d withi n som e families , i t is 119 likely tha t the additiona l effect s o f the environmen t an d othe r gene s wil l als o influence th e expressivit y o f the clinica l an d biochemica l phenotypes . Fro m thes e analyses, i t is apparent tha t the absolut e assignmen t o f these mutation s t o eithe r familial LCA T deficienc y o r FE D i s no t straigh t forwar d an d cautio n shoul d b e use d in such a  segregation. Thi s i s especially tru e i n cases wher e th e clinica l phenotyp e is caused b y a compound heterozygosity . I t appears tha t the origina l familie s identified wit h familia l LCA T deficienc y an d FE D represen t th e clinica l an d biochemical extreme s o f a  wide rang e o f defects o f LCA T function . As a  result , these disorder s shoul d no t b e classified a s separate disease s bu t though t o f as par t of a  large r grou p o f LCA T deficienc y syndromes . Fro m a  biochemica l perspective , i t is likely tha t eve n milde r form s o f LCA T deficienc y exist , bu t i n the absenc e o f any clinical signs , thes e defect s woul d remai n undetected . However , i t i s possible tha t the measuremen t o f HD L cholestero l level s i n large populatio n studie s may  resul t i n the discover y o f a  greater numbe r o f cases. 4.4 Propose d Structure-Functio n Relationship s o f LCA T Following th e crystallizatio n o f pancreati c lipas e an d the elucidatio n o f it s three-dimensional structur e (Winkle r et  al.,  1990) , man y investigator s hav e use d thi s protein a s mode l t o predic t an d investigat e th e structure-functio n relationship s fo r lipases wit h simila r amin o aci d sequenc e homolog y (Derewend a an d Cambillau , 1991). Although LCA T doe s no t shar e a  high degre e o f overal l sequenc e homolog y with lipase s i n general , i t does contai n specifi c homologou s region s whic h correspond t o importan t functiona l domain s suc h a s the consensu s sequenc e containing th e catalyti c Ser . Usin g pancreati c lipas e a s a  model , the Asp-His-Se r 120 catalytic triad s o f lipoprotein lipas e (LPL ) an d hepati c lipas e (HL ) hav e bee n proposed an d subsequentl y confirme d b y site-directed mutagenesi s an d i n vitro expression (Emmeric h et  al.,  1992 ; Davi s et  al.,  1990) . A comparison o f the primar y amino aci d sequenc e o f these lipase s wit h LCA T indicate s tha t Asp204, His286 , an d Ser181 o f human LCA T coul d for m a  similar catalyti c triad . Ther e i s much evidenc e to justify th e assignmen t o f Ser181 t o the catalyti c sit e o f LCA T bu t further studie s will b e necessar y t o investigat e th e specifi c role s o f Asp204 an d His286 . Lipase sequenc e compariso n als o indicate s a  second homologou s regio n near Ser21 6 o f LCA T whic h ma y b e at or nea r the hing e o f a  helica l fla p coverin g the catalyti c site . Simila r movabl e segment s hav e bee n found t o b e essentia l fo r th e activity o f several enzyme s tha t function a t lipid/wate r interfaces . Fo r example , i t has been demonstrate d tha t a  phospholipase A 2 contains a  mobile , hydrophobi c fla p which open s upo n interfacia l activation , allowin g th e formation o f a  hydrophobi c channel i n which th e phospholipi d substrat e i s translocated int o the protei n an d the n hydrolysed a t the activ e sit e insid e th e channe l (Scot t et  al.,  1990) . I n addition, ther e is evidence tha t th e carboxy l terminu s o f CETP ma y form a  "flexible tail " whic h moves i n relation t o the res t o f the molecule , creatin g a  hydrophobic surfac e fo r interaction wit h lipid s (Wan g et  al.,  1993) . The substitution o f Ser to Ala within a homologous regio n o f HL , LPL , and LCA T ha s increase d th e specifi c activit y o f these lipolyti c enzyme s (Davi s et  al.,  1990 ; Emmeric h et  al.,  1992 ; Francon e an d Fielding, 1991b) . I t has bee n speculate d tha t suc h a  change i n structure ma y favou r an "open " conformation o f the loo p coverin g th e activ e sit e o f LCA T (Francon e an d Fielding, 1991b) . I n addition, th e enhance d LCA T reactivit y associate d wit h th e elimination o f the carbohydrat e moiet y a t Asn384 suggest s tha t thi s regio n o f the 121 molecule ma y als o influenc e eithe r th e conformatio n o r access t o the catalyti c site . It is apparent tha t the further analysi s o f the functional abnormalitie s o f the LCAT mutation s describe d i n this thesi s wil l b e necessar y t o define th e structure -function relationship s o f LCAT i n more detail . However , som e genera l relationship s between thes e structura l defect s an d LCA T function ca n b e proposed. Fo r example , mutations whic h ar e likel y t o have a  major affec t o n the structur e o f LCA T suc h a s frameshifts, prematur e sto p codon s o r the introductio n o f a  proline residu e appea r t o be associated wit h a  nea r absenc e o f LCA T enzym e protei n i n plasma and , consequently, LCA T activit y i s absent. Evidenc e t o suppor t thi s was observe d i n the secretion defect s associate d wit h (He375-»frameshift)rLCA T an d (Leu209->Pro)rLCAT. B y contrast , singl e amin o aci d change s ar e mor e ofte n associated wit h highe r level s o f LCA T protei n an d activity . Onc e again , this i s consistent wit h th e characteristic s o f rLCAT i n vitro. Mos t mutation s di d no t affec t the synthesi s o r secretion o f rLCAT bu t rathe r th e activit y o f the enzym e wa s mos t sensitive t o structura l changes . FE D mutation s suc h a s Pro10-*Le u an d Thr123->ll e which d o no t affec t th e abilit y o f LCA T t o esterify cholestero l i n LD L indicat e tha t th e structural integrit y o f the catalyti c sit e remain s intact . A s a  result , the abilit y o f LCA T to reac t wit h HD L i n these patient s ma y stem fro m a  lipid bindin g defec t o f LCA T fo r HDL. T o maintai n norma l endogenou s plasm a cholestero l esterificatio n rates , the distribution o f LCA T woul d hav e to shif t fro m HD L toward LDL . Thus , i t i s possibl e that thi s clas s o f mutation ma y preven t LCA T fro m makin g a  conformational chang e which i s necessary fo r norma l bindin g t o HDL , bu t no t LDL . However , fo r thos e mutations whic h exhibite d onl y a  partial o r an absence o f reactivit y fo r differen t substrates, i t i s difficult t o distinguish betwee n lipi d bindin g and/o r abnormalitie s i n 122 catalytic function . Without th e knowledg e o f the three dimensiona l structur e o f LCAT , i t remain s difficult t o resolv e ho w mutation s seemingl y distan t t o one another (e.g. , Pro1CM_e u and Thr123->lle ) hav e almos t identica l biochemica l characteristic s ye t mutation s closely space d t o on e anothe r (e.g. , Thr123-*ll e an d Arg135^-Trp) ar e associate d with dramati c functiona l differences . Th e presenc e o f widely space d mutation s i n human LCA T an d it s hig h degre e o f sequence homolog y wit h othe r specie s o f LCA T suggest tha t it s norma l functio n i s very sensitiv e t o change s throughou t it s primar y amino aci d sequence . Fro m the biochemica l analysi s o f these differen t recombinan t enzymes, i t appears tha t LCA T ha s a t leas t tw o functionall y importan t domains , th e catalytic cente r an d a  recognition sit e fo r HD L substrates . Additiona l studie s wil l b e required t o determine th e specifi c mechanism s responsibl e fo r the abnormalitie s o f enzyme functio n identifie d fo r these differen t LCA T mutations . 4.5 Perspective s fo r Futur e Stud y The recen t histor y o f the investigatio n o f LCAT deficienc y syndrome s woul d indicate tha t man y ne w LCA T mutation s wil l continue t o be discovered wit h a corresponding increas e i n the numbe r o f reporte d familie s affecte d wit h LCA T deficiency. T o increas e ou r understandin g o f plasma lipoprotei n metabolis m i n these individuals, i t will b e necessary t o investigat e th e specifi c compositiona l an d metabolic heterogeneit y o f the HD L poo l i n these differen t disorders . Specifically , th e quantitation o f L p A-l an d L p A-l/A-ll particle s an d thei r functiona l capacit y t o induc e cholesterol efflu x fro m cell s shoul d b e a priority . I n addition, analysi s o f the i n vivo kinetics o f these apolipoprotein s i n affected homozygote s coul d b e helpfu l t o 123 determine th e basi s fo r the absenc e o f increase d atheroscleroti c ris k despit e very low plasm a HD L levels . Fro m a  clinical perspective , muc h remain s t o be learne d regarding th e pathogenesi s o f the progressiv e rena l insufficienc y observe d i n some of these patients . Hopefully , th e creatio n o f suitable anima l model s discusse d belo w will provid e som e opportunitie s fo r study i n this regard . With respec t t o treatment , the productio n an d purificatio n o f large quantitie s o f rLCA T describe d i n this thesi s may allo w fo r a n efficient mean s o f enzyme replacemen t therap y fo r those patient s at ris k fo r developing rena l abnormalities . Also , a  more detailed biochemica l an d clinical analysi s o f the heterozygot e member s o f different kindred s shoul d provid e more informatio n o n the influenc e o f specific geneti c defect s o f LCA T i n vivo. Results fro m thes e studie s wil l no t onl y provid e additiona l informatio n o n the potential pathologica l mechanism s o f LCAT deficienc y bu t wil l als o b e essentia l fo r the understandin g o f HD L metabolis m an d revers e cholestero l transpor t i n norma l individuals. As ne w LCA T mutation s ar e identified , th e applicatio n o f site-directe d mutagenesis an d i n vitro expressio n o f rLCA T shoul d continu e t o provid e ne w information o n the functiona l significanc e o f these differen t defects . However , furthe r studies wil l b e require d t o investigat e th e specifi c lipi d bindin g an d catalyti c properties o f these mutan t enzymes . I n the firs t instance , endogenou s radiolabelin g and purificatio n o f rLCA T coul d b e used t o assess th e relativ e associatio n o f mutan t enzymes fo r differen t lipoproteins . Also , th e us e o f water-soluble substrate s shoul d enable th e investigatio n o f the catalyti c potentia l o f different enzyme s i n the absenc e of a lipid/wate r interface . Hopefully , thes e studie s wil l b e able to identif y specifi c amino acid s o r structura l domain s withi n th e LCA T protei n whic h distinguis h th e lipi d 124 binding an d catalyti c propertie s o f the enzyme . It i s likely tha t site-directe d mutagenesi s wil l also b e used t o establish th e functional significanc e o f specific amin o acid s o r structura l domain s propose d t o have a n importan t rol e i n enzyme function . Fo r example , th e determinatio n o f the Asp-His-Ser catalyti c tria d a s well a s the propertie s o f the putativ e loo p domain o f human LCA T shieldin g th e catalyti c site . I n addition, th e specifi c rol e o f carbohydrates i n substrate specificit y an d conformatio n o f LCA T remain s t o b e clearly defined . Also , difference s i n the molecula r substrat e specificit y o f ra t an d mouse LCA T compare d t o huma n LCA T ma y b e relate d t o differences i n thei r primary amin o aci d sequence . Throug h th e substitutio n o f specific amin o acid s i n human an d mous e rLCA T an d the analysi s o f their reactivit y wit h artificia l substrate s of defined lipi d composition , potentia l change s i n their molecula r substrat e specificit y can b e determined. Th e descriptio n o f the propertie s o f LCA T fro m differen t specie s and thei r interactio n wit h endogenou s lipoprotein s shoul d provid e usefu l informatio n regarding th e factors whic h influenc e th e interactio n o f LCA T wit h variou s substrates. Specifically , th e knowledg e o f mouse LCA T an d it s interaction s wit h lipoproteins wil l b e helpfu l fo r the analysi s o f transgenic huma n LCA T mic e an d mic e genetically modifie d t o eliminate LCA T expression . Thes e anima l model s coul d provide fo r the firs t time the mean s fo r detaile d i n vivo studie s o f the metabolis m o f LCAT an d it s influenc e o n lipoprotei n metabolis m i n both deficien t an d over -expressed states . Finally, th e preparatio n o f large quantitie s o f homogenou s recombinan t enzyme shoul d enable  th e investigatio n o f some o f the physica l propertie s o f this protein. Fo r example , analysi s o f carbohydrate structur e b y fast atom  bombardmen t 125 mass spectrometry , conformationa l analysi s b y circula r dichrois m an d NM R spectroscopy and , quantitative an d qualitativ e analysi s o f the interactio n o f LCA T with surfac e lipi d monolayers . Ultimately , th e crystallizatio n o f LCA T an d X-ra y diffraction analysi s shoul d revea l it s three-dimensional structure . Wit h thi s knowledge, specifi c structure-functio n relationship s o f LCA T ca n b e proposed an d investigated i n more detail . From the abov e discussion , i t i s apparent tha t the availabilit y o f ne w tool s wil l advance ou r knowledg e o f the structur e an d function o f LCA T includin g it s relationship wit h HD L metabolism . Consequently , thi s wil l benefi t ou r understandin g of the mechanism s o f revers e cholestero l transpor t an d the putativ e anti-atherogeni c properties o f HDL. 126 5 REFERENCE S Agnani G  and Marce l Y . 1993 . Cholestero l efflu x fro m fibroblasts t o discoida l lipoproteins wit h apolipoprotei n A- l (L p A-l) increase s wit h particl e siz e bu t cholesterol transfe r fro m L p A-l t o lipoprotein s decrease s wit h size . Biochemistr y 32 : 2643-2649. Albers J J and Uterman n G . 1981 . Genetic contro l o f lecithin-cholestero l acyltransferase (LCAT) : measuremen t o f LCA T mas s i n a large kindre d wit h LCA T deficiency. A m J  Hu m Gene t 33 : 702-708. Anantharamaiah GM , Venkatachalapathi YV , Brouillett e C G and Segres t J P 1990 . Use o f synthetic peptid e analogue s t o localiz e lecithin:cholestero l acyltransferas e activating domai n i n apolipoprotein A-l . Arteriosclerosis 10 : 95-105. Aron L , Jones S  and Fieldin g C J 1978 . Huma n plasm a lecithin-cholestero l acyltransferase. Characterizatio n o f cofactor-dependent phospholipas e activity . J Biol Che m 253 : 7220-7226. Badimon JJ , Fuster V  and Badimo n L . 1992 . Rol e o f hig h densit y lipoprotein s i n the regression o f atherosclerosis . Circulatio n 86:(supp l lll):lll-86-lll-94 . Banka CL , Bonne t DJ , Black AS , Smit h R S and Curtis s L K 1991 . Localization o f an apolipoprotein A- l epitop e critica l fo r activatio n o f lecithin-cholestero l acyltransferase . J Bio l Che m 266 : 23886-23892 . Barkia A , Puchoi s P , Ghali m N , Torpier G , Ailhaud G  and Fruchar t J-C . 1991. Differential rol e o f apolipoprotein A- l containin g particle s i n cholesterol efflu x fro m adipose cells . Atherosclerosis 87 : 135-146 . Barter PJ . 1983 . Evidenc e tha t lecithin:cholestero l acyltransferas e act s o n bot h high-density lipoprotein s an d low-densit y lipoproteins . Biochi m Biophy s Act a 751: 261-270. Barter PJ , Hopkins GJ  and Gorjatschko L . 1984 . Comparison o f human plasm a low -and high-densit y lipoprotein s a s substrates fo r lecithin:cholestero l acyltransferase . Biochim Biophy s Act a 792 : 1-5 . Barter PJ , Hopkins G J and Gorjatschk o L . 1985 . Lipoprotei n substrate s fo r plasm a cholesterol esterification . Influenc e o f particle siz e an d compositio n o f the hig h density lipoprotei n subfractio n 3 . Atherosclerosis 58 : 97-107. Batzri S  and Kor n ED . 1973 . Singl e laye r liposome s prepare d withou t sonication . Biochim Biophy s Act a 1042 : 310-314. Bethell W , McCulloc h C  and Ghos h M . 1975 . Lecithi n cholestero l acy l transferas e 127 deficiency. Ligh t an d electron microscopi c findin g fro m two corneas . Ca n J  Ophtha l 10: 494-501 . Blanche PJ , Gong EL , Fort e TM and Nichol s AV . 1981 . Characterization o f huma n high-density lipoprotein s b y gradient ge l electrophoresis . Biochi m Biophy s Act a 665:408-419. Bjork I , Ylinenjarvi K , Olso n ST , Hermenti n P , Conrad t H S and Zettlmeiss l G . 1992 . Decreased affinit y o f recombinan t antithrombi n fo r hepari n du e to increase d glycosylate. Bioche m J  286: 793-800. Bonelli F S and Jona s A . 1989 . Reactio n o f lecithi n cholestero l acyltransferas e wit h water-soluble substrates . J  Bio l Che m 264 : 14723-14728 . Bonelli F S and Jona s A . 1992 . Continuou s fluorescenc e assa y fo r lecithinxholesterol acyltransferas e usin g a  water solubl e phosphatidylcholine . J  Lipi d Res 33 : 1863-1869 . Bonelli F S and Jona s A . 1993 . Reactio n o f lecithinxholestero l acyltransferas e wit h a water solubl e substrate : effect s o f surfactants. Biochi m Biophy s Act a 166 : 92-98. Borysiewicz LK , Souta r AK , Evan s DJ , Thompson G R an d Ree s AJ . 1982 . Rena l failure i n familial lecithinxholestero l acyltransferas e deficiency . Quar t J  Med 51: 411-426. Brenner S . 1988 . The molecula r evolutio n o f genes an d proteins : a  tale o f two serines. Natur e 334 : 528-530. Brewer HB , Rade r D , Foj o S  and Hoe g JM. 1990 . Frontier s i n the analysi s o f HD L structure, function , an d metabolism . In  Disorders o f HDL . Carlso n LA , editor . London, Smith-Gordo n an d Compan y Limited , p p 51-58. Bron AJ , Lloy d JK , Fosbrook e AS , Winder A F an d Tripath i RC . 1975 . Primar y L.C.A.T.-deficiency disease . Lance t 1 : 928-929. Brown M S and Goldstei n JL . 1986 . A receptor-mediate d pathwa y fo r cholestero l homeostasis. Scienc e 232 : 34-47. Brunzell JD . 1989 . Familia l lipoprotei n lipas e deficienc y an d othe r cause s o f the chylimicronemia syndrome . In  The Metaboli c Basi s o f Inherite d Disease . Scrive r CR , Beaudet AL , Sl y W S an d Valle D , editors . McGraw-Hil l Inc. , Ne w York . pp . 1165 -1180. Bujo H , Kusunok i J , Ogasawara M , Yamamoto T , Ohta Y , Shimad a T , Sait o Y  and Yoshida S . 1991 . Molecular defec t i n familial lecithinxholestero l acyltransferas e (LCAT) deficiency : a  single nucleotid e insertio n i n LCAT gen e cause s a  complet e deficient typ e o f the disease . Bioche m Biophy s Re s Commu n 181 : 933-940. 128 Burstein M , Fin e A, Atger V , Wirbel E  and Girard-Globa A . 1989 . Rapi d metho d fo r the isolatio n o f two purifie d subfraction s o f hig h densit y lipoprotein s b y differentia l dextran sulfate-magnesiu m chlorid e precipitation . Biochimi e 71 : 741-746. Carlson L A and Philipso n B . 1979 . Fish-ey e disease . A  ne w familia l conditio n wit h massive cornea l opacitie s an d dyslipoproteinaemia . Lance t 2 : 922-924. Carlson LA . 1982 . Fis h eye disease : a  new familia l conditio n wit h massiv e cornea l opacities an d dyslipoproteinemia . Eu r J  Clin Inves t 12 : 41-53. Carlson L A and Holmquis t L . 1983 . Studie s o n hig h densit y lipoprotein s i n fish ey e disease. Act a Me d Scan d 213 : 177-182 . Carlson L A and Holmquis t L . 1985a . Paradoxica l esterificatio n o f plasma cholestero l in fish ey e disease . Act a Me d Scan d 217 : 491-499. Carlson L A and Holmquis t L . 1985b . Evidenc e fo r deficiency o f high densit y lipoprotein lecithin : cholestero l acyltransferas e activit y (alpha-LCAT ) i n fish ey e disease. Act a Me d Scan d 218 : 189-196 . Carlson L A and Holmquis t L . 1985c . Evidenc e fo r the presenc e i n huma n plasm a o f lecithin:cholesterol acyltransferas e activit y (B-LCAT ) specificall y esterifyin g fre e cholesterol o f combined pre-B - and G-lipoproteins . Act a Me d Scan d 218 : 197-205 . Castro G R an d Fieldin g CJ . 1988 . Earl y incorporatio n o f cell-derived cholestero l int o pre-fc-migrating high-densit y lipoprotein . Biochemistr y 27 : 25-29. Chajek T  an d Fieldin g CJ . 1978 . Isolatio n an d characterizatio n o f human seru m cholesteryl este r transfe r protein . Pro c Nat l Acad Sc i US A 75 : 3445-3449 . Chen C- H an d Albers JJ . 1982 . Distributio n o f lecithin-cholestero l acyltransferas e (LCAT) i n human plasm a lipoprotei n fractions . Evidenc e fo r the associatio n o f activ e LCAT wit h lo w densit y lipoproteins . Biochem . Biophys . Res . Commun . 107 : 1091-1096. Chen C , Applegate K , Kin g WC, Glomse t JA , Noru m K R and Gjon e E . 1984 . A study o f the smal l spherica l hig h densit y lipoprotein s o f patients afflicte d wit h familia l lecithin: cholestero l acyltransferas e deficiency . J  Lipid Re s 25 : 269-282. Cheung MC , Wolf AC, Lu m KD , Tollefson J H and Albers J J 1986 . Distributio n an d localization o f lecithin:cholestero l acyltransferas e an d cholestery l este r transfe r activity i n A-l-containing lipoproteins . J  Lipid Re s 27 : 1135-1144 . Cheung M C an d Wol f AC. 1989 . I n vitro transformation o f apoA-l-containin g lipoprotein subpopulations : rol e o f lecithin:cholestero l acyltransferas e an d apoB-containing lipoproteins . J  Lipid Re s 30 : 499-509. 129 Chevet D , Rame e MP , L e Pogamp P , Thomas R , Garr e M  and Alcindor LG . 1978 . Hereditary lecithi n cholestero l acyltransferas e deficiency . Repor t o f a  new famil y wit h two afflicte d sisters . Nephro n 20 : 212-219. Chong KS , Jahani M , Har a S  and Lack o AB . 1983 . Characterization o f lecithin-cholesterol acyltransferas e fro m huma n plasma . 3 . Chemica l propertie s o f the enzyme. Ca n J  Bioche m Cel l Bio l 61 : 875-881. Chu FC , Kuwabar a T , Coga n DG , Schaefe r E J and Brewe r H B Jr . 1979 . Ocula r manifestations o f familial high-densit y lipoprotei n deficienc y (Tangie r disease) . Arc h Ophthal 97 : 1926-1928 . Chung J , Abano DA , Fles s G M and Scan u AM . 1979 . Isolation , properties , an d mechanism o f i n vitro actio n o f lecithinxholestero l acyltransferas e fro m huma n plasma. J  Bio l Che m 254 : 7456-7464. Chung J , Abano D , Byrn e R  and Scan u AM . 1982 . I n vitro mass:activit y distributio n of lecithin-cholestero l acyltransferas e amon g huma n plasm a lipoproteins . Atherosclerosis 45 : 33-41 . Clay MA , Newnha m H H and Barte r PJ . 1991 . Hepatic lipas e promote s a  loss o f apolipoprotein A- l fro m triglyceride-enriched huma n hig h density lipoprotein s durin g incubation i n vitro. Arterioscler Throm b 11 : 415-422. Clay MA , Newnha m HH , Forte TM and Barte r PJ . 1992 . Cholestery l este r transfe r protein an d hepati c lipas e activit y promot e sheddin g o f apo A-l fro m HD L an d subsequent formatio n o f discoidal HDL . Biochi m Biophy s Act a 1124 : 52-58 . Clerc M , Dumo n MF , Sess D , Freneix-Cler c M , Mackness M  and Conr i C . 1991 . A 'fish-eye disease ' familia l conditio n wit h massiv e cornea l opacitie s an d hypoalphalipoproteinemia: clinical , biochemica l an d geneti c features . Eu r J  Clin Invest 21 : 616-624. Cogan DG , Krut h HS , Datili s M B and Marti n N . 1992 . Cornea l opacit y i n LCA T disease. Corne a 11 : 595-599. Collet X  and Fieldin g CJ . 1991 . Effects o f Inhibitor s o f N-linke d oligosaccharid e processing o n the secretion , stability , an d activit y o f lecithin:cholestero l acyltransferase. Biochemistr y 30 : 3228-3234. Contacos C , Neel y RDG , Funk e H , Assmann G  and Sulliva n DR . 1993 . Postprandia l lipoprotein metabolis m i n fish ey e disease . [Abstract ] Deckelbaum RJ , Eisenberg S , Fainar u M , Barenhol z Y  and Olivecron a T . 1979 . I n vitro productio n o f human plasm a lo w densit y lipoprotein-lik e particles . A  mode l fo r very lo w density lipoprotei n catobolism . J  Bio l Chem 254 : 6079-6087. 130 Flatmark AL , Hovi g T , Myhr e E  and Gjon e E . 1977 . Rena l transplantatio n i n patient s with familia l lecithin : cholesterol-acetyltransferas e deficiency . Transplan t Pro c 9 : 1665-1671. Davis RC , Stahnk e G , Wong H , Doolittl e MH , Ameis D , Will  H  and Schot z MC . 1990. Hepati c lipase : site-directe d mutagenesi s o f a serine residu e importan t fo r catalytic activity . J  Bio l Che m 265 : 6291-6295. Doi Y an d Nishid a T . 1983 . Microheterogeneit y an d physica l propertie s o f huma n lecithin-cholesterol acyltransferase . J  Bio l Che m 258 : 5840-5846 . Duverger N , Ghalim N , Ailhaud G , Steinmetz A , Fruchar t J- C an d Castr o G . 1993 . Characterization o f apo A-IV containin g lipoprotei n particle s isolate d fro m huma n plasma an d interstitia l fluid . Arterioscle r Throm b 13 : 126-132 . Emmerich J , Beg OU , Peterso n J , Previato L , Brunzel l JD , Brewe r H B J r an d Santamarina-Fojo S . 1992 . Huma n lipoprotei n lipase . Analysis o f the catalyti c tria d by site-directed mutagenesi s o f Ser-132, Asp-156 , an d His-241 . J  Bio l Che m 267 : 4161-4165. Fielding C J and Fieldin g P E 1971 . Purification an d substrat e specificit y o f lecithin-cholesterol acyltransferas e fro m huma n plasma . FEB S Lettr . 15 : 355-358. Fielding CJ , Shore V G an d Fieldin g P E 1972 . Lecithin:cholestero l acyltransferase : Effects o f substrate compositio n upo n enzym e activity . Biochi m Biophy s Act a 270 : 513-518. Fielding C J and Fieldin g PE . 1981 . Regulation o f huma n plasm a lecithin:cholestero l acyltransferase activit y b y lipoprotei n accepto r cholestery l este r content . J  Bio l Chem 256 : 2102-2104 . Fielding PE , Miid a T  and Fieldin g CJ . 1991 . Metabolism o f low-densit y lipoprotei n free cholestero l b y human plasm a lecithin-cholestero l acyltransferase . Biochemistr y 30: 8551-8557 . Forte T M an d Carlso n LA . 1984 . Electro n microscopi c structur e o f serum lipoproteins fro m patient s wit h fis h eye disease . Arteriosclerosi s 4 : 130-137 . Forte TM , McCal l MR , Knowle s B B and Shor e VG . 1989 . J  Lipi d Re s 30 : 817-829 . Franceschini G , Sirtor i CR , Capurs o A , Weisgraber K H an d Mahle y RW . 1980 . A-lMjlano apoprotein . Decrease d hig h densit y lipoprotei n cholestero l level s wit h significant lipoprotei n modification s i n an Italia n family . J  Cli n Inves t 66 : 892-900. Francone OL , Guraka r A  an d Fieldin g CJ . 1989 . Distributio n an d functions o f lecithin:cholesterol acyltransferas e an d cholestery l este r transfe r protei n i n plasm a lipoproteins. Evidenc e fo r a  functional uni t containin g thes e activitie s togethe r wit h 131 apolipoproteins A- l an d D  that catalyze s th e esterificatio n an d transfe r o f cell-derive d cholesterol. J  Bio l Che m 264 : 7066-7072 . Francone O L an d Fieldin g C J 1991 . Effects o f site-directed mutagenesi s a t residue s cysteine-31 an d cysteine-18 4 o n lecithin-cholestero l acyltransferas e activity . Pro c Natl Acad Sc i US A 88 : 1716-1720 . Francone OL , Evangelist a L  and Fieldin g CJ . 1993 . Lecithin-cholestero l acyltransferase: effect s o f mutagenesis a t N-linke d oligosaccharid e attachmen t site s on acy l accepto r specificty . Biochi m Biophy s Act a 1166 : 301-304 . Frohlich J , Hon K  and McLeo d R . 1982 . Detectio n o f heterozygotes fo r familia l lecithin:cholesterol acyltransferas e (LCAT ) deficiency . A m J  Hu m Gene t 34 : 65-72. Frohlich J , Hoag G , McLeod R , Hayde n M , Godin DV , Wadsworth LD , Critchley J D and Pritchar d PH . 1987 . Hypoalphalipoproteinemi a resemblin g fis h ey e disease . Acta Me d Scan d 221 : 291-298. Frohlich J , McLeod R , Pritchar d PH , Fesmir e J  and McConath y W . 1988 . Plasm a lipoprotein abnormalitie s i n heterozygotes fo r familia l lecithin:cholestero l acyltransferase deficiency . Meta b Cli n Ex p 37 : 3-8 . Fruchart J-C , Ailhaud G  and Bar d J-M . 1993 . Heterogenit y o f hig h densit y lipoprotein particles . Circulatio n 87(supp l III) : III-22-III-27 . Funk WD , MacGillivra y RTA , Maso n AB , Brow n S A an d Woodworth RC . 1990 . Expression o f the amino-termina l half-molecul e o f huma n seru m transferri n i n cultured cell s an d characterizatio n o f the recombinan t protein . Biochemistr y 29 : 1654-1660. Funke H , von Eckardstei n A , Pritchar d PH , Karas M , Albers J J and Assmann G . 1991a. A  frameshif t mutatio n i n the huma n apolipoprotei n A- l gen e cause s hig h density lipoprotei n deficiency , partia l lecithin : cholesterol-acyltransferas e deficiency , and cornea l opacities . J  Clin Inves t 87 : 371-376 . Funke H , von Eckardstei n A , Prithchar d PH , Albers JJ , Kastelein JJP , Drost e C  and Assmann G . 1991b . A  molecula r defec t causin g fis h eye disease : a n amino aci d exchange i n lecithin:cholestero l acyltransferas e (LCAT ) lead s t o the selectiv e los s o f a-LCAT activity . Pro c Nat l Acad Sc i US A 88 : 4855-4859. Funke H , von Eckardstei n A , Pritchar d PH , Hornby AE , Wiebusch H , Mott i C , Hayden MR , Dache t C , Jacoto t B , Gerdes U , Faergema n O , Albers JJ , Colleoni N , Catapano A , Frohlic h J  and Assmann G . 1993 . Geneti c an d phenotypi c heterogeneity i n familial lecithin:cholestero l acyltransferas e (LCAT ) deficiency . J  Clin Invest 91 : 677-683. Furukawa Y , Uran o T , Ito h H , Takahashi C  and Kimur a S . 1989 . Isolatio n an d 132 properties o f ra t plasma lecithin-cholestero l acyltransferase . J  Bioche m 105 : 962-967. Furukawa Y , Uran o T , Hid a Y , Ito h H , Takahashi C  and Kimur a S . 1992 . Interaction of ra t lecithin-cholestero l acyltransferas e wit h ra t apolipoprotein A- l an d wit h lecithin -cholesterol vesicles . J  Bioche m 111 : 413-418. Gjone E , Torsvik H  and Noru m KR . 1968 . Familia l plasm a cholstero l este r deficiency: A  stud y o f the erythrocytes . Scan d J  Clin La b Inves t 21 : 327-332. Gjone E  and Noru m KR . 1968 . Familia l serum-cholestero l este r deficiency : Clinica l study o f a  patient wit h a  new syndrome . Act a Me d Scan d 183 : 107-112 . Gjone E , Blomhof f J P an d Skarbovi k AJ . 1974 . Possibl e associatio n betwee n a n abnormal lo w densit y lipoprotei n an d nephropath y i n LCAT deficiency . Cli n Chi m Acta 54 : 11-18 . Glass C , Pittma n RC , Cive n M  and Steinber g D . 1985 . Uptak e o f high densit y lipoprotein associate d apoprotei n A- l cholestery l ester s b y 1 6 tissues o f the ra t i n vivo an d b y adrenal cell s an d hepatocyte s i n vitro. J  Bio l Che m 260 : 744-750 . Glomset JA . 1968 . The plasm a lecithin:cholestero l acyltransferas e reaction . J  Lipi d Res 9 : 155-167 . Glomset JA , Noru m K R an d Kin g W. 1970 . Plasm a lipoprotein s i n familia l lecithin:cholesterol acyltransferas e deficiency : lipi d compositio n an d reactivit y i n vitro. J  Clin Inves t 49 : 1827-1837 . Glomset JA , Nichol s AV , Noru m KR , Kin g W  and Fort e T . 1973 . Plasm a lipoprotein s in familial lecithin:cholestero l acyltransferas e deficiency . Furthe r studie s o f very lo w and lo w densit y lipoprotei n abnormalities . J  Cli n Inves t 52 : 1078-1092 . Glomset JA , Noru m KR , Nichol s AV , Kin g WC, Mitchel l CD , Applegate KR , Gon g E L and Gjon e E . 1975 . Plasm a lipoprotein s i n familial lecithin : cholestero l acyltransferase deficiency : effect s o f dietary manipulation . Scan d J  Cli n La b Inves t 35 (Supp l 142) : 3-30 . Glomset JA , Mitchel l CD , Kin g WC, Applegate KA , Fort e T , Noru m K R an d Gjon e E . 1980. I n vitro effect s o f lecithin:cholestero l acyltransferas e o n apolipoprotei n distribution i n familial lecithinxholestero l acyltransferas e deficiency . An n Ne w Yor k Acad Sc i 348 : 224-243. Godin DV , Gra y G R an d Frohlic h J . 1978 . Erythrocyt e membran e alteration s i n lecithin.cholesterol acyltransferas e deficiency . Scan d J  Clin La b Inves t 3 8 (Supp l 150): 162-167 . Goldstein JL  and Brow n MS . 1987 . The lo w density lipoprotei n pathwa y an d it s 133 relation t o atherosclerosis . Ann u Re v Bioche m 46 : 897-930. Gordon T , Castell i WP , Hjortlan d MC , Kanne l W B an d Dawbe r TR . 1977 . High -density lipoprotei n a s a  protective facto r agains t coronar y hear t disease : Th e Framingham Study . Am J  Med 62 : 707-714. Gotoda T , Yamada N , Muras e T , Sakum a M , Murayam a N , Shimano H , Kozak i K , Albers JJ , Yazaki Y  and Akanuma Y . 1991 . Differential phenotypi c expressio n b y three mutan t allele s i n familial lecithin:cholestero l acyltransferas e deficiency . Lance t 338: 778-781 . Grove D  and Pownal l H J 1991 . Comparative specificit y o f plasma lecithin:cholestero l acyltransferase fro m te n anima l species . Lipid s 26 : 416-420. Gustafson A , McConath y AW , Alaupovic P , Curry M D and Perso n B . 1979 . Identification o f apolipoprotein familie s i n a variant o f human plasm a apolipoprotei n A deficiency . Scan d J  Clin La b Inves t 39 : 377-388. Gylling H  and Miettine n TA . 1992 . Non-cholestero l sterols , absorptio n an d synthesi s of cholestero l an d apolipoprotei n A- l kinetic s i n a Finnish lecithin-cholestero l acyltransferase deficien t family . Atherosclerosi s 95 : 25-33. Hamilton RL , Have l RJ , Kane JP , Blauroc k A E an d Sat a T . 1971 . Cholestasis: lamellar structur e o f the abnorma l huma n seru m lipoprotein . Scienc e 172 : 475-478. Hamilton RL , Williams MC , Fieldin g C J and Have l RJ . 1976 . Discoida l bilaye r structure o f nascen t hig h densit y lipoprotein s fro m perfuse d ra t liver . J  Cli n Inves t 58: 667-680 . Havel RJ , Eder H A and Bragdo n JH . 1955 . The distribution an d chemica l composition o f ultracentrifugall y separate d lipoprotein s i n human serum . J  Cli n Invest 34 : 1345-1353 . Hefele Wal d J , Kru l E S and Jonas A . 1990 . Structur e o f apolipoprotein A- l i n three homogenous, reconstitute d hig h densit y lipoprotei n particles . J  Bio l Che m 265 : 20037-20043. Hennessy LK , Kunitak e ST , Kan e JP . 1993 . Apolipoprotein A-l-containin g lipoproteins, wit h o r without apolipoprotei n A-ll , as progenitor s o f pre-S high-densit y lipoprotein particles . Biochemistr y 32 : 5759-5765. Hixson JE , Driscol l DM , Birnbaum S  and Britte n M L 1993 . Baboo n lecithi n cholesterol acyltransferas e (LCAT) : cDN A sequence s o f two alleles , evolutio n an d gene expression . Gen e 128 : 295-299. Ho WKK an d Nichol s AV . 1971 . Interaction o f lecithin:cholestero l acyltransferas e with sonicate d dispersion s o f lecithin . Biochi m Biophy s Act a 231 : 185-193. 134 Holmquist L  and Carlso n LA . 1988 . Normalizatio n o f high densit y lipoprotei n i n fish eye disease plasm a b y purified norma l huma n lecithin:cholestero l acyltransferase . Lipids 23 : 225-229. Hovig T  an d Gjon e E . 1973 . Familia l plasm a lecithin:cholestero l acyltransferas e (LCAT) deficiency . Ultrastructura l aspect s o f a new syndrom e wit h particula r reference t o lesion s i n the kidney s an d the spleen . Acta Patho l Micr o Scan d Sectio n A, 81 : 681-697. Huang Y , von Eckardstei n A  and Assmann G . 1993 . Cell-derived unesterifie d cholesterol cycle s betwee n differen t HDL s an d LD L fo r it s effective esterificatio n i n plasma. Arterioscle r Throm b 13 : 445-458. Jacobsen CD , Gjon e E  and Hovi g T . 1972 . Sea-blu e histiocyte s i n familial lecithin : cholesterol acyltransferas e deficiency . Scan d J  Haema t 9 : 106-113 . Jahani M  and Lack o A G 1982 . Stud y o f the lecithin:cholestero l acyltransferas e reaction wit h liposom e an d hig h densit y lipoprotei n substrates . Biochi m Biophy s Acta. 713 : 504-511 . Jain SK , Mohanda s N , Sensabaugh GF , Shojania A M an d Shohet  SB . 1982 . Hereditary plasm a lecithin:cholestero l acyltransferas e deficiency : A  heterozygou s variant wit h erythrocyt e membran e abnormalities . J  La b Cli n Me d 99 : 816-827 . Jauhiainen M  and Dolphi n P J 1986 . Huma n Plasm a Lecithin-Cholestero l Acyltransferase. A n elucidatio n o f the catalyti c mechanism . J  Bio l Che m 261: 7032-7043. Jauhiainen M , Stevenson K J and Dolphi n P J 1988 . Huma n plasm a lecithin -cholesterol acyltransferase . Th e vicina l natur e o f cysteine 3 1 and cystein e 18 4 i n the catalytic site . J  Bio l Che m 263 : 6625-6533. Jauhiainen M , Metso J , Pahlman R , Blomqvis t S , van To l A an d Ehnhol m C . 1993 . Human plasm a phospholipi d transfe r protei n cause s hig h densit y lipoprotei n conversion. J  Bio l Che m 268 : 4032-4036. Johnson WJ , Kilsdon k EPC , Van To l A, Phillip s M C and Rothbla t GH . 1991. Cholesterol efflu x fro m cell s t o immunopurifie d subtraction s o f huma n high-densit y lipoprotein: L p A-l an d L p A-hA-ll. J  Lipid Re s 32 : 1992-2000 . Jonas A . 1986 . Syntheti c substrate s o f lecithin:cholestero l acyltransferase . J  Lipi d Res 27 : 689-698. Jonas A , Daehle r J L and Wilson ER . 1986 . Anion effect s o n the reactio n o f lecithin-cholesterol acyltransferas e wit h discoidal  complexe s o f phosphatidylcholines , apolipoproteins an d cholesterol . Biochi m Biophy s Act a 876 : 474-485. 135 Jonas A , Kezd y H E and Hefel e Wal d J . 1989 . Define d apolipoprotei n A- l conformations i n reconstitute d hig h densit y lipoprotei n discs . J  Bio l Che m 264 : 4818-4824. Jonas A , vo n Eckardstei n A , Kezd y KE , Steinmetz A  and Assmann G . 1991. Structural an d functiona l propertie s o f reconstituted hig h density lipoprotei n disc s prepared wit h si x apolipoprotei n A- l variants . J  Lipi d Re s 32 : 97-106. Jonas A . 1991 . Lecithin-cholesterol acyltransferas e i n the metabolis m o f high-densit y lipoproteins. Biochi m Biophy s Act a 1084 : 205-220. Kadowaki H , Patto n GM , Robins SJ . 1992 . Metabolis m o f high densit y lipoprotei n lipids b y the ra t liver : evidence  fo r participatio n o f hepati c lipas e i n the uptak e o f cholesteryl ester . J  Lipi d Re s 33 : 1689-1698 . Kastelein JJP , Pritchar d PH , Erkelens DW , Kuivenhove n JA , Albers J J and Frohlic h JJ. 1992 . Familia l high-density-lipoprotei n deficienc y causin g cornea l opacitie s (fis h eye disease ) i n a  family o f Dutc h descent . J  Inter n Me d 231 : 413-419. Khalil A , Farooqu i J  and Scan u AG 1986 . Antigenic relatednes s betwee n huma n lecithin-cholesterol acyltransferas e an d phospholipase s o f the A 2 family. Biochi m Biophys Act a 878 : 127-130 . Kitabatake K , Pira n U , Kami o Y , Do i Y and Nishid a T . 1979 . Purificatio n o f huma n plasma lecithin:cholestero l acyltransferas e an d it s specificity toward s th e acy l acceptor. Biochi m Biophy s Act a 573 : 145-154 . Klein H-G , Lohs e P , Pritchar d PH , Bojanovski D , Schmid t H  and Brewe r H B Jr . 1992. Two differen t alleli c mutation s i n the lecithin:cholestero l acyltransferas e gen e associated wit h th e fish ey e syndrome . J  Clin Inves t 89 : 499-506. Klein H-G , Lohs e P , Duverge r N , Albers JJ , Rader D , Zech LA , Santamarina-Foj o S and Brewe r H B Jr . 1993a . Tw o differen t alleli c mutation s i n the lecithin:cholestero l acyltransferase (LCAT ) gen e resultin g i n classic LCA T deficiency : LCA T (tyr 83-*stop) and LCAT(tyr 156^asn). J  Lipi d Re s 34 : 49-58. Klein H-G , Santamarina-Foj o S , Duverge r N , Clerc M , Dumo n M-F , Alber s JJ , Marcovina S , Brewe r H B Jr . 1993b . Fis h eye syndrome : A  molecula r defec t i n the lecithin-cholesterol acyltransferas e (LCAT ) gen e associate d wit h norma l cr-LCAT -specific activity . Implication s fo r classification an d prognosis . J  Cli n Inves t 92 : 479-485. Knipping G , Birchbaue r A , Steyre r E , Groener J , Zechner R  and Kostne r G M 1986 . Studies o n the substrat e specificit y o f human an d pi g lecithin:cholestero l acyltransferase: Rol e o f low-densit y lipoproteins . Biochemistr y 25 : 5242-5249 . Kriegler M . 1990 . In  Gene transfe r an d expression . A  laborator y manual . Stockto n 136 Press, Ne w York . pp . 96-101 . Kunitake ST , LaSal a K J and Kan e JP . 1985 . Apolipoprotein A-l-containin g lipoproteins wit h pre-bet a electrophoreti c mobility . J  Lipi d Re s 26 : 549-555 . Kunitake ST , Mende l C M and Henness y LK . 1992 . Interconversio n betwee n apolipoprotein A-l-containin g lipoprotein s o f pre-beta an d alpha electrophoreti c mobilities. J  Lipi d Re s 33 : 1807-1816 . Kuusi T , Saarine n P  and Nikkil a EA . 1980 . Evidenc e fo r the rol e o f hepati c endothelial lipas e i n the metabolis m o f plasma hig h densit y lipoprotein 2 i n man. Atherosclerosis 36 : 589-593. Lagrost L . 1992 . Effec t o f the huma n cholestery l este r transfe r protei n o n the particl e size redistributio n o f specific subpopulation s o f high densit y lipoprotein-3 . Biochi m Biophys Act a 1126 : 143-150 . Lefevre M , Sloop C H and Rohei m PS . Characterization o f dog prenoda l periphera l lymph lipoproteins . Evidenc e fo r the periphera l formatio n o f lipoprotein-unassociate d apo A- l wit h slo w pre-f t electrophoreti c mobility . J  Lipid Re s 29 : 1139-1148 . Liu M , Kru l E S and Subbaia h PV . 1992 . Effec t o f apoprotein B  conformation o n the activation o f lysolecithi n acyltransferas e an d lecithin:cholestero l acyltransferase . Studies wit h subfraction s o f low density lipoproteins . J  Bio l Che m 267 : 5139-5147 . Liu M  and Subbaia h P V 1993 . Activation o f plasma lysolecithi n acyltransferas e reaction b y apolipoproteins A-l , C-l an d E . Biochi m Biophy s Act a 1168 : 144-152 . Luc G , Bar d J-M , Lussier-Caca n S , Bouthillie r S , Parr a HJ , Fruchar t J- C an d Davignon J . 1991 . High-density lipoprotei n particle s i n octogenarians. Metabolis m 40: 1238-1243 . Maeda E , Nak a Y , Matozak i T , Sakuma M , Akanuma Y , Yoshino G  and Kasug a M . 1991. Lecithin-cholestero l acyltransferas e (LCAT ) deficienc y wit h a  missens e mutation i n exon 6  of the LCA T gene . Bioche m Biophy s Re s Commu n 178 : 460-466. Marcel YL , Vezina C , Emon d D  and Suzu e G . 1980 . Heterogeneit y o f huma n hig h density lipoprotein : presenc e o f lipoproteins wit h an d withou t ap o E  and thei r role s as substrates fo r lecithin:cholestero l acyltransferas e reaction . Pro c Nat l Acad Sc i USA 77 : 2969-2973. Marcel YL . 1982 . Lecithin:cholestero l acyltransferas e an d intravascula r transport . Adv Lipi d Re s 19 : 85-136. Matz C E an d Jona s A . 1982 . Micella r complexe s o f huma n apolipoprotei n A- l wit h phosphatidylcholines an d cholestero l prepare d fro m cholate-lipi d dipersions . J  Bio l 137 Chem 257 : 4535-4540. McLean J , Fielding C , Drayn a D , Dieplinge r H , Bae r B , Koh r W, Henze l W  an d Lawn R . 1986a . Clonin g an d expression o f human lecithin-cholestero l acyltransferase cDNA . Pro c Nat l Acad Sc i US A 83 : 2335-2339. McLean J , Wion K , Drayn a D , Fieldin g C  and Law n R . 1986b . Huma n lecithin:cholesterol acyltransferas e gene : complet e sequenc e an d site s o f expression. Nuclei c Acid s Re s 14 : 9397-9406. McLean JW . 1992 : Molecula r defect s i n the lecithin:cholestero l acyltransferas e gene . In High densit y lipoprotein s an d atherosclerosi s III . Miller N E and Tal l AR, editors . Elsevier Scienc e Publisher s B.V. , Amsterdam , pp . 59-65. Mendez AJ , Oram JF , and Bierma n EL . 1991 . Protein kinas e C  as a  mediator o f high densit y lipoprotei n recepto r dependen t efflu x o f intracellula r cholesterol . J  Bio l Chem 266 : 10104-10111. Meng Q-H , Calabres i L , Fruchar t J- C an d Marce l Y L 1993 . Apolipoprotein A- l domains involve d i n the activatio n o f lecithin:cholestero l acyltransferase . J  Bio l Chem 268 : 16966-16973 . Meroni G , Malgarett i N , Magnagh i P  and Taramell i R . 1990 . Nucleotid e sequenc e o f the cDN A fo r lecithin-cholestero l acy l transferase (LCAT ) fro m th e rat . Nuclei c Acid s Res 18 : 5308. Miida T , Fieldin g C J and Fieldin g PE . 1990 . Mechanis m o f transfer o f LDL-derive d free cholestero l t o HD L subfraction s i n huma n plasma . Biochemistr y 29 : 10469 -10474. Miida T , Kawan o M , Fieldin g C J and Fieldin g PE . 1992 . Regulatio n o f the concentration o f pre B high-densit y lipoprotei n i n normal plasm a b y cel l membrane s and lecithin-cholestero l acyltransferas e activity . Biochemistr y 31 : 11112-11117. Minnich A , Colle t X , Roghan i A , Cladara s C , Hamilto n RL , Fieldin g C J and Zanni s VI 1992 . Site-directe d mutagenesi s an d structure-functio n analysi s o f the huma n apolipoprotein A-l . Relation betwee n lecithin-cholestero l acyltransferas e activatio n and lipi d binding . J  Bio l Che m 267 : 16553-16560 . Mitchell CD , Kin g WC, Applegate KR , Fort e T , Glomse t JA , Noru m KR , Gjon e E . 1980. Characterizatio n o f apolipoprotein E-ric h hig h densit y lipoprotein s i n familia l lecithin:cholesterol acyltransferas e deficiency . J  Lipid Re s 21 : 625-634. Mowri H-O , Patsc h W , Smit h LC , Gotto A M J r and Patsc h JR . 1992 . Differen t reactivities o f hig h densit y lipoprotei n 2 subfractions wit h hepati c lipase . J  Lipi d Re s 33: 1269-1279 . 138 Murano S , Shira i K , Sait o Y , Yoshida S , Oht a Y , Tsuchida H , Yamamoto S , Asano G, Chen C H an d Albers J  J. 1987 . Impaire d intermediate-densit y lipoprotei n triglyceride hydrolysi s i n familial lecithin:cholestero l acyltransferas e (LCAT ) deficiency. Scan d J  Cli n La b Inves t 47 : 775-783 . Murayama N , Asano Y , Hosod a S , Maesaw a M , Saito M , Takaku F , Sugihar a T , Miyashima K  and Yawata Y . 1984a . Decrease d sodiu m influ x an d abnorma l re d cel l membrane lipid s i n a patient wit h familia l plasm a lecithin : cholestero l acyltransferas e deficiency. A m J  Hemat 16 : 129-137 . Murayama N , Asano Y , Kat o K , Sakamot o Y , Hosod a S , Yamada N , Kodam a T , Murase T  and Akanuma Y . 1984b . Effect s o f plasma infusio n o n plasma lipids , apoproteins an d plasm a enzym e activitie s i n familial lecithin : cholestero l acyltransferase deficiency . Eu r J  Clin Inves t 14 : 122-129 . Newnham H H and Barte r PJ . 1990 . Synergisti c effect s o f lipi d transfer s an d hepati c lipase i n the formation o f very smal l high-densit y lipoprotein s durin g incubatio n o f human plasma . Biochi m Biophy s Act a 1044 : 57-64. Nordoy A  and Gjon e E . 1971 . Familial plasm a lecithin:cholestero l acyltransferas e deficiency: A  stud y o f the platelets . Scan d J  Clin La b Inves t 27 : 263-268. Norum K R and Gjon e E . 1967 . Familia l plasm a lecithin:cholestero l acyltransferas e deficiency. Biochemical  stud y o f a new inbor n erro r o f metabolism . Scan d J  Clin La b Invest 20 : 231-243. Norum K R and Gjon e E . 1968a . Th e influenc e o f plasma fro m patient s wit h familia l plasma lecithin:cholestero l acyltransferas e deficienc y o n the lipi d patter n o f erythrocytes. Scan d J  Clin La b Inves t 22 : 94-98. Norum K R an d Gjon e E . 1968b . The effec t o f plasma transfusio n o n the plasm a cholesteryl ester s i n patients wit h familia l lecithin:cholestero l acyltransferas e deficiency. Scan d J  Clin La b Inves t 22 : 339-342. Norum KR , Borstin g S  and Grund t I . 1970 . Familia l lecithin:cholestero l acyltransferase deficiency . Act a Me d Scan d 188 : 323-326. Norum KR , Glomse t JA , Nichol s A V an d Fort e T . 1971 . Plasma lipoprotein s i n familial lecithinxholestero l acyltransferas e deficiency : Physica l an d chemica l studie s of lo w an d hig h densit y lipoproteins . J  Clin Inves t 50 : 1131-1140 . Norum RA , Lakie r JB , Goldstein S , Angel A , Goldber g RB , Bloc k WD , Nofz e DK , Dolphin PJ , Edelglass J , Bogorad D D and Alaupovic P . 1982 . Familia l deficienc y o f apolipoproteins A- l an d C-ll l and precociou s coronar y hear t disease . Ne w Eng l J Med 306 : 1513-1519 . Norum KR , Gjon e E  and Glomse t JA . 1989 . Familia l lecithin:cholestero l 139 acyltransferase deficiency , includin g fis h ey e disease . In  The Metaboli c Basi s o f Inherited Disease . Scrive r CR , Beaude t AL , Sl y WS an d Valle D , editors . McGraw -Hill, Inc. , Ne w York . pp . 1181-1194 . O K , Hil l JS, Wang X , McLeo d R  and Pritchar d PH . 199 3 Lecithinxholestero l acyltransferase: Rol e o f N-linke d glycosylatio n o n enzyme function . Biochemical Journal 294 : 879-884 . Ohta Y , Yamamoto S , Tsuchida H , Muran o S , Saitoh Y , Tohjo S  and Okad a M . 1986. Nephropath y o f familial lecithin-cholestero l acyltransferas e deficiency : repor t of a case. Am J  Kidney Di s 7 : 41-46. Oram JF . 1990 . Cholestero l traffickin g i n cells. Cur r Opi n Lipido l 1 : 416-421. Palmiter RD , Behringe r RR , Quaif e CJ , Maxwel l F , Maxwel l I H and Brinste r RL . 1987. Cel l lineag e ablatio n i n transgenic mic e b y cell-specific expressio n o f a toxin gene. Cel l 50 : 435-443. Park M-SC , Kudchodka r BJ , Frohlich J , Pritchard H  and Lack o AG. 1987 . Study o f the component s o f revers e cholestero l transpor t i n lecithin:cholestero l acyltransferase deficiency . Arc h Bioche m Biophy s 258 : 545-554 . Patsch JR , Soutar  AK , Moriset t JD , Gotto A M J r and Smit h LC . 1977 . Lipoprotein-X : a substrate fo r lecithin:cholestero l acyltransferase . Eu r J  Cli n Inves t 7 : 213-217. Patsch JR , Gott o A M Jr , Olivecron a T , Eisenber g S . 1978 . Formatio n o f hig h densit y lipoprotein2-like particle s durin g lipolysi s o f very lo w density lipoprotein s i n vitro. Pro c Natl Acad Sc i US A 75 : 4519-4523. Patsch WG , Schonfel d G , Gotto AM J r and Patsc h JR . 1980 . Characterizatio n o f human hig h densit y lipoprotein s b y zonal ultracentrifugation . J  Bio l Che m 255 : 3178-3185. Patsch JR , Prasa d S , Gotto AM an d Patsc h W . 1987 . Hig h densit y lipoprotein., : Relationship o f the plasm a level s o f this lipoprotei n specie s t o it s composition , t o the magnitude o f postprandia l lipemia , an d to the activitie s o f lipoprotei n lipas e an d hepatic lipase . J  Cli n inves t 80 : 341-347. Phillips MC , McLea n LR , Stoud t G W and Rothbla t GH . 1980 . Mechanis m o f cholesterol efflu x fro m cells . Atherosclerosis 36 : 409-422. Pownall HJ , Gotto A M J r and Sparro w JT . 1984 . Thermodynamics o f lipid-protei n association an d the activatio n o f lecithin:cholestero l acyltransferas e b y syntheti c model apolipopeptides . Biochi m Biophy s Act a 793 : 149-156 . Pownall HJ , Pa o Q  and Masse y JB . 1985 . Acyl chai n an d hea d grou p specificit y o f human plasm a lecithin:cholestero l acyltransferase . Separtio n o f matri x an d 140 molecular specificities . J  Bio l Che m 260 : 2146-2152. Pownall HJ , Pao Q , Brockma n H L and Masse y JB . 1987 . Inhibitio n o f lecithin-cholesterol acyltransferas e b y diphytanoyl phosphatidylcholine . J  Bio l Che m 262 : 9033-9036. Puchois P , Kandouss i A , Fieve t P , Fourrie r JL , Bertran d M , Kore n E  and Fruchar t J -C. 1987 . Apolipoprotein A-l-containin g i n coronary arter y disease . Atherosclerosi s 68: 35-40 . Qu S-J , Fa n H-Z , Blanco-Vac a F  and Pownal l H J 1993 . Role s o f cysteines i n human lecithinxholesterol acyltransferase . Biochemistr y 32 : 3089-3094. Rader DJ , Castro G , Zech LA , Fruchar t J- C an d Brewe r HB . 1991 . In vivo metabolism o f apo A- l o n hig h densit y lipoprotei n particle s L p A-l an d L p A-l, A-ll. J Lipid Re s 32 : 1849-1859 . Rajaram O V and Barte r P J 1985 . Reactivit y o f human lipoprotein s wit h purifie d lecithin:cholesterol acyltransferas e durin g incubation s i n vitro. Biochi m Biophy s Act a 835: 41-49. Rinninger F  and Pittma n RC . 1988 . Mewchanis m o f the cholestery l este r transfe r protein-mediated uptak e o f high densit y lipoprotei n cholestery l ester s b y He p G 2 cells. J  Bio l Che m 264 : 6111-6118. Ritland S  and Gjon e E . 1975 . Quantitativ e studie s o f lipoprotein- X i n familia l lecithin:cholesterol acyltransferas e deficienc y an d during cholestero l esterification . Clin Chi m Acta 59 : 109-119 . Rose J K an d Dom s RW . 1988 . Regulatio n o f protein expor t fro m th e endoplasmi c reticulum. Ann u Re v Cel l Bio l 4 : 257-288. Rothblat GH , Mahlberg FH , Johnson W J an d Phillip s MC . 1992 . Apolipoproteins , membrane cholestero l domains , an d the regulatio n o f cholesterol efllux . J  Lipi d Re s 33: 1091-1097 . Rye K-A , Garret y K H an d Barte r PJ . 1992 . Change s i n the size o f reconstituted hig h density lipoprotei n durin g incubatio n wit h cholestery l este r transfe r protein : Th e rol e of apolipoproteins . J  Lipid Re s 33 : 215-224. Sakuma M , Akanuma Y , Kodam a T , Yamada N , Murat a S , Muras e T , Itakur a H , Kosaka K . 1982 . Familia l plasm a lecithin : cholestero l acyltransferas e deficiency . A new famil y wit h partia l LCA T activity . Act a Me d Scand 212 : 225-232. Sanger F , Nickle n S  and Coulso n AR . 1977 . DN A sequencin g wit h chain-terminatin g inhibitors. Pro c Nat l Acad Sc i US A 74 : 5463-5469 . 141 Schaefer EJ . 1984 . Clinical , biochemica l an d geneti c feature s i n familial disorder s o f high density lipoprotei n deficiency . Arteriosclerosi s 4 : 303-322. Schaefer EJ , Ordovas J , Law S , Ghisell i G , Kashya p L , Srivastava L , Heato n WH , Albers J , Connor W , Lindgre n F , Lemeshe w A , Segres t J  and Brewe r H B Jr . 1985 . Familial ap o A- l an d C-ll l deficiency . Varian t II . J Lipi d Re s 26 : 1089-1101 . Schmitz G , Assmann G , Robenek H  and Brennhause n B . 1985 . Tangie r disease : A disorder o f intracellula r membran e traffic . Pro c Nat l Acad Sc i US A 82 : 6305-6309. Scott DL , White SP , Otwinowsk i Z , Yuan W, Gel b M H and Sigle r PP . 1990 . Interfacial catalysis : th e mechanis m o f phospholipase A 2. Scienc e 250 : 1541-1546 . Segrest JP , Jones MK , D e Loo f H , Brouillett e G , Venkatachalapathi Y V an d Anantharamaiah GM . 1992 . The amphipathi c heli x i n the exchangeabl e apolipoproteins: a  review o f secondary structur e an d function. J  Lipi d Re s 33 : 141-166. Shojania AM , Jain S K an d Shohe t SB . 1983 . Hereditar y lecithin-cholestero l acyltransferase deficiency . Repor t o f 2 new case s an d revie w o f the literature . Cli n Invest Me d 6 : 49-55. Skretting G  and Pryd z H . 1992 . An amin o aci d exchang e i n exon I  of the huma n lecithin:cholesterol acyltransferas e (LCAT ) gen e i s associated wit h fis h ey e disease . Biochem Biophy s Re s Commu n 182 : 583-587. Skretting G , Blomhof f JP , Solheim J  and Pryd z H . 1992 . The geneti c defec t o f the original Norwegia n lecithin:cholestero l acyltransferas e deficienc y families . FEB S Lettr 309 : 307-310 . Smith N B and Kuksi s A . 1980 . Stere o chemica l substrat e requirement s o f lecithin:cholesterol acyltransferas e an d it s inhibitio n b y enantiomeric lysolecithins . Can J  Bioche m 58 : 1286-1291 . Smith KM , Lawn R M and Wilcox JN . 1990 . Cellula r localizatio n o f apolipoprotein D and lecithin:cholestero l acyltransferas e mRN A i n rhesus monke y tissue s b y i n situ hybridization. J  Lipi d Re s 31 : 995-1004. Sorci-Thomas M , Babia k J  and Rude l LL . 1990 . Lecithin:cholestero l acyltransferas e (LCAT) catalyze s transacylatio n o f intac t cholestery l esters . Evidenc e fo r the partia l reversal o f the forward LCA T reaction . J  Bio l Chem 265 : 2665-2670. Soutar AK , Garne r CW , Bake r HN , Sparrow JT , Jackson RL , Gotto A M J r an d Smit h LC. 1975 . Effec t o f the huma n plasm a apolipoprotein s an d phosphatidylcholin e acy l donor o n the activit y o f lecithin.cholestero l acyltransferase . Biochemistr y 14 : 3057-3064. 142 Soutar AK , Knigh t B L and Myan t NB . 1982 . The characterizatio n o f lipoprotein s i n the hig h densit y fractio n obtaine d fro m patient s wit h familia l lecithin : cholestero l acyltransferase deficienc y an d thei r interactio n wit h culture d huma n fibroblasts . J Lipid Re s 23 : 380-390 . Sparks D L and Pritchar d P H 1988 . The neutra l lipi d compositio n an d siz e o f recombinant hig h densit y lipoprotein s regulate s lecithinxholestero l acyltransferas e activity. Biochem . Cel l Biol . 67 : 358-364. Steinmetz A  an d Uterman n G . 1985 . Activation o f lecithin:cholestero l acyltransferas e by huma n apolipoprotei n A-IV . J  Bio l Chem 260 : 2258-2264. Steinmetz A , Barbara s R , Ghali m N , Clavey V  and Fruchar t J-C . 1990 . Huma n apolipoprotein A-I V bind s t o apolipoprotein A-l/A-l l recepto r site s an d promote s cholesterol efflu x fro m adipos e cells . J  Bio l Che m 265 : 7859-7863 . Stokke KT , Bjerv e KS , Blomhof f JP , Oystese B , Flatmar k A , Noru m K R and Gjon e E. 1974 . Familia l lecithin:cholestero l acyltransferas e deficiency : Studie s o n lipi d composition an d morpholog y o f tissues. Scan d J  Clin La b Inves t 3 3 (Supp I 137) : 93-100. Stuart GW , Searl e PF , Chen HY , Brinste r R L and Palmite r RD . 1984 . A 12-base -pair DN A moti f tha t i s repeated severa l time s i n metallothionein gen e promote r confers meta l regulatio n t o a  heterologous gene . Pro c Nat l Acad Sc i US A 81 : 7318-7322. Subbaiah PV , Che n C-H , Albers J J and Bagdad e J D 1982 . Studies o n the cofacto r requirement fo r the acylatio n an d hydrolysi s reaction s catalyze d b y purified lecithin -cholesterol acyltransferase . 45 : 181-190 . Subbaiah PV , Che n CH , Bagdade J D an d Albers JJ . 1985 . Rol e o f lo w densit y lipoprotein i n the activatio n o f plasma lysolecithi n acyltransferas e activity . Effec t o f chemical an d enzymati c modification s o f the lipoprotei n o n enzyme activity . Atherosclerosis 54 : 99-110. Subbaiah PV , Li u M , Bola n P J and Paltau f F  1992 . Altered positiona l specificit y o f human plasm a lecithin-cholestero l acyltransferas e i n the presenc e o f sn-2 arachidonoyl phosphatidy l cholines . Mechanis m o f formation o f saturated cholestery l esters. Biochi m Biophy s Acta . 1128 : 83-92. Takata K , Kajiyam a G , Horiuch i I , Watanabe T , Tokumo H , Hirat a Y . 1989 . A ne w case o f familial lecithin : cholestero l acyltransferas e (LCAT ) deficiency-paradoxica l findings regardin g LCA T mas s an d activit y i n 23 members o f a family. Ja p J  Med 28: 765-771 . Tall AR an d Smal l DM . 1978 . Plasm a high-densit y lipoproteins . N  Engl J  Me d 299: 1232-1236. 143 Taramelli R , Pontogli o M , Candian i G , Ottolengh i S , Dieplinge r H , Catapan o A , Albers J , Vergani C  and McLea n J . 1990 . Lecithi n cholestero l acy l transferas e deficiency: molecula r analysi s o f a mutated allele . Hu m Gene t 85 : 195-199 . Tarentino AL , Plumme r T H an d Male y F . 1974 . The releas e o f intac t oligosaccharides fro m specifi c glycoprotein s b y endo-G-N-acetylglucosaminidase H . J Bio l Che m 249 : 818-824. Thrift RN , Fort e TM, Cahoon B E and Shor e VG . 1986 . Characterizatio n o f lipoproteins produce d b y the huma n live r cel l line , He p G2 , under define d conditions . J Lipi d Re s 27 : 236-250. Torsvik H , Gjone E  and Noru m KR . 1968 . Familia l plasm a cholestery l este r deficiency: Clinica l studie s i n a family. Act a Me d Scan d 183 : 387-391 . Torsvik H . 1972 . Studie s o n the protei n moiet y o f serum hig h densit y lipoprotei n from patient s wit h familia l lecithin:cholestero l acyltransferas e deficiency . Cli n Gene t 3: 188-200 . Towbin H , Staeheli n T  and Gordo n J . 1979 . Electrophoreti c transfe r o f proteins fro m polyacrylamide gel s t o nitrocellulos e sheets : Procedur e an d some applications . Pro c Natl Acad Sc i US A 76 : 4350-4354. Turner P , Mille r N , Chrystie I , Coltar t J , Mistry P , Nicol l A an d Lewi s B . 1979 . Splanchnic productio n o f discoidal  plasm a high-densit y lipoprotei n i n man . Lance t 1 , 645-646. von Eckardstei n A , Funk e H , Henk e A , Altland K , Benninghove n A  an d Assmann G . 1989. Apolipoprotein A- l variants . Naturall y occurin g substitution s o f proline residue s affect plasm a concentratio n o f apolipoprotein A-l . J Clin Inves t 84 : 1722-1730 . Wang S , Wang X , Den g L , Rassar t E , Miln e R W and Tal l AR. 1993 . Poin t mutagenesis o f carboxyl-termina l amin o acid s o f cholestery l este r transfe r protein . Opposite face s o f an amphipathic heli x importan t fo r cholestery l este r transfe r o r for binding neutralizin g antibody . J  Bio l Che m 268 : 1955-1959 . Warden CH , Langne r CA , Gordo n Jl , Taylor BA , McLea n J W and Lusi s AJ . 1989 . Tissue-specific expression , developmenta l regulation , an d chromosoma l mappin g o f the lecithin:cholestero l acyltransferas e gene . J  Bio l Che m 264 : 21573-21581. Wengler H  and Seide l D . 1973 . Doe s lipoprotein- X (LP-X ) ac t a s a  substrate fo r the lecithin:cholesterol acyltransferas e (LCAT) ? Cli n Chi m Acta 45 : 429-432. Winder A F an d Bro n AJ . 1978 . Lecithin:cholestero l acy l transferase deficienc y presenting a s visua l impairment , wit h hypocholesterolaemi a an d norma l rena l function. Scan d J  Clin La b Inves t 3 8 (Supp l 159) : 151-155 . 144 Winder AF , Garne r A , Sheraida h G A and Barr y P . 1985 . Familia l lecithin.cholestero l acyltransferase deficiency . Biochemistr y o f the cornea . J  Lipi d Re s 26 : 283-287. Winkler FK , D'Arc y A  and Hunzike r W . 1990 . Structur e o f human pancreati c lipase . Nature 343 : 771-774. Wong L , Curtiss LK , Huan g J , Mann CJ , Maldonado B  and Rohei m PS . 1992 . Altered epitop e expressio n o f human interstitia l flui d apolipoprotei n A- l reduce s it s ability t o activat e lecithi n cholestero l acy l transferase . J  Clin Inves t 90 : 2370-2375. Yang C , Manoogia n D , Pa o Q , Le e F , Knap p RD , Gotto AM J r and Pownal l H J 1987. Lecithimcholestero l acyltransferase . Functiona l region s an d a  structural mode l of the enzyme . J  Bio l Che m 262 : 3086-3091 . 145 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088143/manifest

Comment

Related Items