UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular phenotype of human CD4+CD25+ T regulatory cells Crellin, Natasha K. 2007

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

Item Metadata

Download

Media
831-ubc_2007-267035.pdf [ 9.84MB ]
Metadata
JSON: 831-1.0100385.json
JSON-LD: 831-1.0100385-ld.json
RDF/XML (Pretty): 831-1.0100385-rdf.xml
RDF/JSON: 831-1.0100385-rdf.json
Turtle: 831-1.0100385-turtle.txt
N-Triples: 831-1.0100385-rdf-ntriples.txt
Original Record: 831-1.0100385-source.json
Full Text
831-1.0100385-fulltext.txt
Citation
831-1.0100385.ris

Full Text

MOLECULAR PHENOTYPE OF HUMAN CD4+CD25+ T REGULATORY CELLS by NATASHA K . CRELLIN B . S c . , Un ive r s i t y o f Saskatchewan, 1997 M . S c , Un ive r s i t y o f Saskatchewan, 2000 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y i n T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Exper imenta l M e d i c i n e ) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A A P R I L , 2007 © N A T A S H A K. CRELLIN, 2007 Abs t rac t Suppression by T regulatory cel ls (Tregs) is a major mechanism by w h i c h the immune system controls responses to se l f and innocuous foreign proteins. A l t h o u g h there are many different types o f T reg cel ls , the best characterized are those that const i tut ively express c e l l -surface I L - 2 R a ( C D 2 5 ) . The a i m o f this research was to; increase understanding o f the molecula r phenotype o f human C D 4 + C D 2 5 + Tregs, and was addressed us ing 3 different approaches. I) Cer ta in pathogens are k n o w n to be capable o f modula t ing the funct ion o f C D 4 + C D 2 5 + Tregs, but it was unclear i f this was v i a a direct or indirect mechan ism. The expression o f T L R 5 o n human C D 4 + T cel ls was studied, and it was observed that both C D 4 + C D 2 5 " T effectors and C D 4 + C D 2 5 + Tregs express T L R 5 at funct ional ly significant levels . Further, the T L R 5 l igand f l age l l in is a co-st imulatory molecu le for C D 4 + C D 2 5 " T effector cel ls , and f lage l l in can increase the suppressive capacity o f C D 4 + C D 2 5 + Tregs i n the absence o f antigen presenting cel ls . II) In an effort to understand the mechan i sm under ly ing the unique functional phenotype o f C D 4 + C D 2 5 + Tregs, i nc lud ing hypo-responsiveness to T ce l l receptor ( T C R ) mediated act ivat ion and l ack o f cytokine product ion, T C R - m e d i a t e d s ignal ing i n pure populat ions o f ex v i v o human C D 4 + C D 2 5 + Tregs was investigated. A consistent defect i n A K T act ivat ion i n C D 4 + C D 2 5 + Tregs was observed, w h i c h when reversed, abrogated C D 4 + C D 2 5 + T r eg suppressive capacity. I l l ) C l i n i c a l applicat ions and experimental manipulat ions have been hampered by the lack o f a unique c e l l surface marker for C D 4 + C D 2 5 + Tregs that is not also an act ivat ion marker for C D 4 + C D 2 5 " T effector cells . M i c r o a r r a y analysis o n single ce l l -der ived T reg and T effector clones was performed as an in i t i a l screen, f o l l o w e d by quantitative R T - P C R va l ida t ion on p o l y c l o n a l populat ions o f human C D 4 + C D 2 5 + Tregs. It was observed that human C D 4 + C D 2 5 + Tregs specif ical ly •• • i i express the membrane channel protein A q u a p o r i n 9. In summary, the research described w i t h i n is o f considerable import to the study o f human C D 4 + C D 2 5 + Tregs and their control o f immune homeostasis, and represents a significant contr ibut ion to the f i e ld o f i m m u n o l o g y . i i i Table of Contents A B S T R A C T . ii T A B L E O F C O N T E N T S .. iv LIST O F T A B L E S vi LIST O F F I G U R E S ..vii LIST O F A B B R E V I A T I O N S viii A C K N O W L E D G E M E N T S ix C O - A U T H O R S H I P S T A T E M E N T x 1. I N T R O D U C T I O N A N D L I T E R A T U R E R E V I E W 1 1.1. HUMAN CD4+T REGULATORY CELLS ..: 1 1.1.1 Naturally occurring CD4+CD25+ Tregs 2 1.1.2. Antigen induced CD4+CD25+ (iTreg) Treg cells 3 1.1.3. Trl cells.... '. 4 1.1.4. Infectious Tolerance '. T 4 1.2. TREGS AND TOLL LIKE RECEPTORS (TLRS) 5 1.2.1 Innate immune stimuli: linking innate and adaptive immunity 6 1.2.2 TLRs indirectly control the function of CD4+CD25+ Tregs 7 1.2.3 Can TLRs directly control the function of CD4+CD25+ Tregs? 8 1.3. TCR SIGNALING IN TREGS 10 1.3.1 Overview of TCR signaling 11 1.3.2 T cell anergy ,.' 13 1.3.3 Signaling in CD4+CD25+Tregs 14 1.4. THE SEARCH FOR A TRUE CELL SURFACE MARKER FOR CD4+CD25+ TREGS 16 1.4.1. Microarray analysis of CD4+CD25+ Tregs 16 1.5. HYPOTHESES 19 1.6. REFERENCES ' '. 20 2. R O L E O F T L R 5 O N CD4 + CD25 + / T C E L L S 28 2.1 INTRODUCTION 28 2.2 MATERIALS AND METHODS 30 2.3 RESULTS 35 2.4 DISCUSSION 50 2.5 REFERENCES 56 3. I N T R A C E L L U L A R S I G N A L I N G IN C D 4 + C D 2 5 + T R E G S , 60 •3.1 INTRODUCTION '. •. 60 3.2 MATERIALS AND METHODS 62 3.3 RESULTS.. : 67 3.4 DISCUSSION 82 3.5 SUPPLEMENTARY FIGURES 86 3.6 REFERENCES 90 4. C D 4 + C D 2 5 + T R E G S E X P R E S S A Q U A P O R I N 9 95 4.1 INTRODUCTION 95 4.2 MATERIALS AND METHODS 96 4.3 RESULTS 102 4.4 DISCUSSION '. 116 4.5 REFERENCES 119 . DISCUSSION AND CONCLUSIONS 5.1. REFERENCES List of Tables T A B L E 4 . 1 . P R I M E R S E Q U E N C E S U S E D F O R Q U A N T I T A T I V E R T - P C R 1 0 0 T A B L E 4 . 2 . C D 2 5 E X P R E S S I O N O N C D 4 + T C E L L C L O N E S ' • 1 0 3 T A B L E 4 . 3 . M I C R O A R R A Y A N A L Y S I S O F H U M A N C D 4 + C D 2 5 + T R E G A N D C D 4 + C D 2 5 " T E F F E C T O R C L O N E S 1 0 7 v i List of Figures F I G U R E 1.1. S U B S E T S O F T R E G U L A T O R Y C E L L S , 2 F I G U R E 1.2. C A N PAMPs I N T E R A C T D I R E C T L Y W I T H T C E L L S ? 8 F I G U R E 1.3. O V E R V I E W O F T C R S I G N A L I N G 1 2 F I G U R E 1.4 T R E G - A S S O C I A T E D M O L E C U L E S P R O P O S E D A S C E L L S U R F A C E M A R K E R S , 1 7 F I G U R E 2 . 1 . P U R I T Y O F C D 4 + T C E L L S U B S E T S 3 1 F I G U R E 2 . 2 . C D 4 + C D 2 5 + T R E G C E L L S E X P R E S S M O R E T L R 5 A N D T L R 4 M R N A I N C O M P A R I S O N T O C D 4 + C D 2 5 " T C E L L S v 3 7 F I G U R E 2 . 3 . C D 4 + T C E L L S E X P R E S S P H Y S I O L O G I C A L L Y R E L E V A N T L E V E L S O F T L R 5 P R O T E I N 3 9 F I G U R E 2 . 4 . C H A N G E S I N T L R 5 E X P R E S S I O N F O L L O W I N G A C T I V A T I O N 4 2 F I G U R E 2 . 5 . F L A G E L L I N D O E S N O T A L T E R C D 4 + C D 2 5 + T R E G C E L L H Y P O R E S P O N S I V E N E S S B U T IT I N C R E A S E S C D 4 + C D 2 5 - T C E L L P R O L I F E R A T I O N A N D I L - 2 P R O D U C T I O N 4 4 F I G U R E 2 . 6 . E F F E C T S O F F L A G E L L I N O N T H E S U P P R E S S I V E C A P A C I T Y O F C D 4 + C D 2 5 + T R E G C E L L S I N T H E P R E S E N C E O R A B S E N C E O F A P C S 4 7 F I G U R E 2 . 7 . F L A G E L L I N I N C R E A S E S F O X P 3 E X P R E S S I O N I N C D 4 + C D 2 5 + T R E G C E L L S F O L L O W I N G T C R -M E D I A T E D A C T I V A T I O N 4 9 F I G U R E 2 . 8 . M O D E L O F H O W F L A G E L L I N M A Y I N T E R A C T W I T H C D 4 + C D 2 5 + T R E G C E L L S D U R I N G A N I M M U N E R E S P O N S E ; 5 4 F I G U R E 3 . 1 . S I N G L E C E L L A N A L Y S I S O F M A P K A C T I V A T I O N I N E X V I V O H U M A N C D 4 + C D 2 5 + T R E G S A N D C D 4 + C D 2 5 " T C E L L S F O L L O W I N G T C R A C T I V A T I O N 6 8 F I G U R E 3 . 2 . H U M A N C D 4 + C D 2 5 + T R E G S H A V E A R E D U C E D C A P A C I T Y T O P H O S P H O R Y L A T E A K T F O L L O W I N G T C R S T I M U L A T I O N 71 F I G U R E 3 . 3 . D I M I N I S H E D A K T P H O S P H O R Y L A T I O N I N C D 4 + C D 2 5 + T R E G S R E S U L T S I N D E C R E A S E D A C T I V A T I O N O F D O W N S T R E A M E F F E C T O R S •. 7 4 F I G U R E 3 A. E N F O R C E D A C T I V A T I O N O F AKT R E V E R S E S T H E S U P P R E S S I V E C A P A C I T Y O F C D 4 + C D 2 5 + T R E G S . C D 4 + C D 2 5 " , G H O R C D 4 + C D 2 5 " T C E L L S W E R E T R A N S D U C E D W I T H L E N T I V I R U S E N C O D I N G A N I N D U C I B L E A K T - E R O R C O N T R O L ( P C C L ) V E C T O R 7 6 F I G U R E 3 . 5 . E N H A N C E D A K T A C T I V I T Y IN C D 4 + C D 2 5 + T R E G S D O E S N O T R E D U C E E X P R E S S I O N O F F O X P 3 , C T L A - 4 , C D 2 5 , O R G R A N Z Y M E S A O R B 7 9 F I G U R E 3 . 6 . E N H A N C E D A K T A C T I V I T Y I N C D 4 + C D 2 5 + T R E G S R E S T O R E S T H E I R C A P A C I T Y T O P R O D U C E I F N - Y , T N F - a , I L - 4 , A N D I L - 1 0 , B U T N O T I L - 2 81 F I G U R E S 3 . 1 . C D 4 + C D 2 5 + T R E G S H A V E D E F E C T I V E A K T P H O S P H O R Y L A T I O N IN R E S P O N S E T O S T I M U L A T I O N V I A T H E T C R IN T H E A B S E N C E O R P R E S E N C E O F C O S T I M U L A T I O N V I A C D 2 8 8 6 F I G U R E S 3 . 2 . C D 4 + C D 2 5 + T R E G S A N D C D 4 + C D 2 5 " T C E L L S H A V E E Q U I V A L E N T P I 3 K F U N C T I O N A N D L E V E L S O F T H E P H O S P H A T A S E S P T E N A N D S H I P 8 7 F I G U R E S 3 . 3 . D E F E C T I V E A C T I V A T I O N O F P H O S P H O - A K T A N D S 6 I N C D 4 + C D 2 5 + T R E G C E L L L I N E S 8 8 F I G U R E S 3 . 4 . T R A N S D U C T I O N E F F I C I E N C Y A N D R E L A T I V E P U R I T Y O F T R A N S D U C E D T C E L L L I N E S 8 9 F I G U R E 4 . 1 . S I N G L E C E L L C L O N I N G O F C D 4 + T C E L L S 1 0 2 F I G U R E 4 . 2 . S C R E E N I N G O F H U M A N C D 4 + C D 2 5 + T R E G A N D C D 4 + C D 2 5 " T E F F E C T O R S I N G L E C E L L C L O N E S . ... 1 0 5 F I G U R E 4 . 3 . Q U A N T I T A T I V E P C R V A L I D A T I O N O F M I C R O A R R A Y R E S U L T S 1 1 0 F I G U R E 4 . 4 . H U M A N C D 4 + C D 2 5 + T R E G S E X P R E S S M O R E A Q P 9 M R N A A S C O M P A R E D T O C D 4 + C D 2 5 " T E F F E C T O R S 1 1 1 F I G U R E 4 . 5 . A Q P 9 is N O T A N A C T I V A T I O N M A R K E R F O R H U M A N C D 4 + C D 2 5 " T E F F E C T O R C E L L S 1 1 2 F I G U R E 4 . 6 . H U M A N C D 4 + C D 2 5 + T R E G S E X P R E S S M O R E A Q P 9 P R O T E I N A S C O M P A R E D T O C D 4 + C D 2 5 " T E F F E C T O R S .' : 1 1 3 F I G U R E 4 . 7 . L E N T I - V I R A L T R A N S D U C T I O N O F H U M A N C D 4 + C D 2 5 " f E F F E C T O R S W I T H A Q P 9 1 1 5 v n List of Abbreviations antigen ( A g ) A q u a p o r i n 9 ( A Q P 9 ) C y t o k i n e B e a d A s s a y ( C B A ) estrogen receptor ( E R ) extracellular signal-related kinase ( E R K ) hemagglu t in in ( H A ) 4 -hydroxy tamoxi fen ( 4 H T ) M e a n Fluorescence Intensity ( M F I ) mi togen activated protein kinase ( M A P K ) M A P kinase kinase ( M E K ) m a m m a l i a n target o f rapamycin ( m T O R ) Pathogen associated molecule pattern ( P A M P ) , Pattern recogni t ion receptors ( P R R ) , p lecks t r in h o m o l o g y ( P H ) P H doma in leucine-r ich repeat protein phosphatase ( P H L P P ) phosphoinosit ide-dependent protein kinase 1 ( P D K 1 ) phosphat idy l inosi t ide-3 kinase ( P I 3 ' K ) phosphat idyl inositol-3,4,5-triphosphate (PIP3) . prote in kinase C ( P K C ) prote in phosphatase 2 A ( P P 2 A ) phosphatases and tensin homologue deleted o n chromosome 10 ( P T E N ) T c e l l receptor ( T C R ) T regulatory cel ls (Tregs) src h o m o l o g y 2 (SH2)-conta in ing inos i to l phosphatase ( S H I P ) Acknowledgements I gratefully acknowledge R o s a G a r c i a for technical assistance throughout, and O m e e d Hadisfar for assistance w i t h q P C R i n Chapter 4. Support and supervis ion was p rov ided by D r . M e g a n L e v i n g s , for w h i c h I am extremely appreciative. F inanc ia l support for m y studies was p rov ided by the Transplant Trainee F e l l o w s h i p and M S F H R Senior D o c t o r a l award. C o - A u t h o r s h i p Statement I designed and performed a l l experiments described w i t h the except ion o f those performed under m y supervis ion by R o s a G a r c i a or O m e e d Hadisfar . I wrote the manuscripts under the guidance o f m y supervisor, D r . M e g a n L e v i n g s . x 1. Introduction and Literature Rev iew T regulatory cel ls (Tregs) are a specific class o f T cel ls that are required for immune homeostasis and peripheral tolerance. A l t h o u g h in i t i a l ly described earlier, the study o f T regulatory cel ls has undergone a renaissance i n the last 10-15 years as a result o f i m p r o v e d techniques for detection and pur i f ica t ion o f Tregs. In 1995, Sakaguch i et a l . reported that i m m u n o l o g i c tolerance was maintained by a sub-populat ion o f T cel ls expressing I L - 2 R c t ( C D 2 5 ) , wi thout w h i c h a number o f autoimmune diseases w o u l d develop i n m ice 1 . It was then demonstrated that the suppressive effect o f C D 4 + C D 2 5 + Tregs cou ld be reproducibly observed i n vi t ro 2 . F i n a l l y , the existence o f a funct ional ly defined, regulatory subset o f T cells was p roven when it was observed that the t ranscript ion factor F O X P 3 was expressed i n C D 4 + C D 2 5 + Tregs and was required for their development, and that the absence o f F O X P 3 i n either mouse or m a n was characterized by a syndrome o f severe autoimmune disease 3 - 5 . The study o f human Tregs has tremendous c l i n i c a l relevance, as their manipu la t ion may be relevant for a number o f autoimmune diseases, cancer, and as a cell-based therapy to prevent rejection o f organ transplants 6 . The therapeutic potential o f Tregs has further fueled investigations into their characteristics and mode o f action, both i n v i v o and i n v i t ro . 1.1. H U M A N C D 4 + T R E G U L A T O R Y C E L L S In humans, there are at least three types o f T reg cells w i t h i n the C D 4 + subset: the natural ly occur r ing C D 4 + C D 2 5 + Tregs, antigen-induced C D 4 + C D 2 5 + T r e g ( iTreg) cel ls , and type 1 T reg ( T r l ) cells . A l t h o u g h further discuss ion w i l l focus on natural ly 1 occurr ing C D 4 C D 2 5 Tregs, a b r i e f d iscuss ion o f a l l three subsets is benef ic ia l , and is summar ized i n F igure 1.1. Thymus CD25 Figure 1.1. Subsets of T regulatory cells. Thymically derived natural Tregs (nTregs) in the periphery supress T effector cells (Te). Te may differentiate into cytokine secreting Trl cells, or induced Tregs (iTregs). nTregs may promote Trl differentiation in suppressed Te. 1.1.1 Naturally occurring C D 4 + C D 2 5 + Tregs C D 4 + C D 2 5 + Tregs arise i n the thymus dur ing normal thymocyte development, and al though they have a po lyc lona l T C R repertoire equivalent to effector T cel ls 7 , they appear to be h igh ly enriched for self-antigen specific T C R s 8 . These cel ls were o r ig ina l ly defined by their constitutive expression o f C D 2 5 , but more recent data indicate that the transcript ion factor F O X P 3 may be a more specific marker 9 . In human peripheral b lood , C D 4 + C D 2 5 + Tregs are h igh ly enriched i n the 1-2% CD25-br ightes t cel ls w i t h i n the C D 4 + T ce l l p o o l 1 0 " , al though a significant number also exist i n the C D 2 5 d i m subset l 2 . Considerable effort has been put into def ining the mechanism by w h i c h C D 4 + C D 2 5 + Tregs potently suppress effector T c e l l responses both i n v i t ro and i n v i v o . In v i t ro , the mechan i sm is cell-contact-dependent and does not require cytokines; i n fact, one o f the def ining characteristics o f these cel ls is failure to produce most cytokines 1 0 . In v i v o , however , the situation is more controversial since i n some situations TGF-(3 and/or I L - 1 0 can have a major ro le i n media t ing their suppressive effects 1 3 ' 1 4 . W i t h respect to T G F - | 3 , more recent data indicate that the product ion o f this cytokine by C D 4 + C D 2 5 + Tregs themselves is not required, since TGF- |31-def ic ien t C D 4 + C D 2 5 " Tregs are funct ional ' . It has also been proposed that granzyme mediated perforin-independent cy to tox ic i ty may mediate suppression 1 7 , a f ind ing w h i c h remains to be corroborated. 1.1.2. Antigen induced CD4 +CD25 + (iTreg) Treg cells It is n o w clear C D 4 + C D 2 5 + Tregs that are phenotypica l ly and funct ional ly ident ical to the natural ly occur r ing subset can be generated f rom C D 4 + C D 2 5 " T cel ls 1 8 . I n humans, this process appears to be stochastic rather than developmenta l ly predetermined, since upon po lyc lona l or A g - s p e c i f i c act ivat ion o f naive or memory C D 4 C D 2 5 " T cel ls , a significant number o f cel ls remain C D 2 5 b r i g h t i n t h e r e s t i n g p h a s e and acquire the characteristics o f the natural ly occur r ing subset l 9 . S i m i l a r l y , there are n o w examples i n murine models where C D 4 + C D 2 5 " Tce l l s develop into C D 4 + C D 2 5 + i T r e g s . 2 0 In both mic e and humans, this process m a y be enhanced by co-s t imula t ion w i t h TGF-(3 2 1 w h i c h may be related to the capacity o f this cytokine to stimulate and/or main ta in h i g h levels o f F O X P 3 expression 2 2 . Recent evidence, ind ica t ing that i n the mouse a smal l propor t ion o f C D 4 + C D 2 5 " T cel ls consti tutively express F O X P 3 and are suppressive, h ighl ights the need to clar i fy whether this is t ruly de novo development or 3 s imp ly expans ion o f a pre-exist ing subset 2 3 . I f C D 4 + C D 2 5 + i T r e g cel ls can t ruly arise de novo , then Ag- spec i f i c cel ls cou ld be generated for targeted immunotherapy. 1.1.3. TM cells T r l cel ls derive f rom naive C D 4 + T cel ls when they repeatedly encounter their cognate A g o n tolerogenic A P C s and/or i n the presence o f I L - 1 0 2 4 ' 2 5 . In v i v o , T r l -induc ing D C s m a y be long to a dedicated tolerogenic lineage (eg specia l ized subsets found i n mucosa l tissues or the l iver ) , and/or they may be modulated by s t imul i such as pathogens o r apoptotic cel ls w h i c h swi tch the balance o f cy tokine product ion to l o w I L -12 and h i g h I L - 1 0 2 4 ' 2 6 . S ince T r l cel ls derive f rom naive C D 4 + T ce l ls , they have the potential to be specific for any A g . Curren t ly there are no k n o w n T r l - c e l l specific markers; a l though they transiently upregulate C D 2 5 f o l l o w i n g act ivat ion, they do not retain const i tut ively h igh levels l ike C D 4 + C D 2 5 + T reg cel ls . In addi t ion, expression levels o f F O X P 3 are equivalent to those i n effector T cells 2 4 . Thus , despite m u c h effort to define more precise markers, T r l cel ls can currently on ly be defined by their unique cy tokine product ion profi le : I L - 1 0 + , IL-4" , I L - 2 l o w , I F N - y + . In contrast to C D 4 + C D 2 5 + Treg cel ls , they suppress v i a a process that is cy tokine mediated and requires product ion o f I L - 1 0 and T G F - p \ Recent ly , it has been proposed that they may also be direct ly cy to toxic to their targets by expression o f granzyme B . 1.1.4. Infectious Tolerance O n e o f the many unsolved questions i n the T reg f ie ld is numer ica l : h o w can these cel ls exert long- term and systemic effects w h e n they on ly represent a smal l fraction o f peripheral C D 4 + T cel ls? C D 4 + C D 2 5 + T r eg cel ls are extremely potent, w i t h suppression 4 evident at ratios as l o w as lTreg^OOreffectors i n v i t ro . H o w can we account for this g iven a cel l-contact dependent mechanism o f action? A n emerging concept is that T reg cel ls operate not o n l y by inh ib i t ing effector responses, but also by s imultaneously conferr ing suppressive properties on their target T cel ls - a process termed infectious tolerance 2 1 . T h i s use o f the term refers to infectious tolerance o n a ce l l to ce l l basis, rather than the prev ious ly described 2 8 transfer o f tolerance between animals. T r l cel ls l i k e l y mediate this effect v i a product ion o f I L - 1 0 and T G F - | 3 w h i c h act to promote tolerogenic D C s 2 4 . The mechan i sm by w h i c h C D 4 + C D 2 5 + nT reg cel ls achieve this effect is u n k n o w n al though it has been suggested that TGF-(3 is required 2 9 . Interestingly, the "suppressed" targets o f human C D 4 + C D 2 5 + T r e g cel ls can become IL -10 -p roduc ing T r l cel ls 3 0 ' 3 1 , ind ica t ing that we should no longer th ink o f i nd iv idua l subsets o f T reg cel ls i n isolat ion, but rather as networks. The phenomenon o f infectious tolerance offers an explanat ion for the systemic and long-las t ing effects o f T reg cel ls that can be adopt ively transferred through many generations o f mice . T h i s regulatory ampl i f ica t ion cascade also accounts for the observat ion that the T reg cel ls do not need be at the site o f in f l ammat ion i n order to exert a suppressive e f f e c t 3 2 . 1.2. T R E G S A N D T O L L L I K E R E C E P T O R S (TLRs) Tregs are essential for the induc t ion and maintenance o f tolerance to both se l f and foreign antigens (Ags ) . M u l t i p l e factors control the balance between a product ive immune response, in f lammat ion , and homeostasis. T reg cel ls , i n particular, have been the subject o f intense recent research due to their therapeutic potential as regulators o f tolerance. The general phenotypes and functions o f T reg cel ls i n a variety o f diseases have been extensively discussed i n many recent reviews 2 5 ' 3 3 . Here we w i l l focus o n evidence that 5 supports the concept that T reg cel ls are key integrators o f innate and adaptive immune s t imul i . O f particular interest is h o w pathogens and the innate and adaptive i m m u n e systems have co -evo lved to regulate the balance between effector T cel ls and different subsets o f T r e g cel ls . There is a steady increase i n the number o f examples o f foreign organisms that promote Treg ce l l function. In some cases this phenonemon is h igh ly benef ic ia l to the host, for example i n regulat ion o f responses to commensa l bacteria 3 4 , whereas i n others it appears to be deleterious 3 5 . Innate immune s t imul i c lear ly influence the development and function o f Treg cel ls at mul t ip le levels, and s tudying mechanisms o f tolerance, w h i c h have evo lved over m i l l i o n s o f years, w i l l provide the basis for manipu la t ion o f tolerance i n the c l i n i c a l setting. 1.2.1 Innate immune stimuli: linking innate and adaptive immunity In contrast to the lock and key method o f receptor-antigen recogni t ion that is used by the adaptive i m m u n e system, the innate immune system employs a set o f ge rm l ine encoded pattern recogni t ion receptors ( P R R s ) that recognize repeating elements termed pathogen-associated molecula r patterns ( P A M P s ) . The T o l l - l i k e receptors ( T L R s ) have been extensively studied i n this context, and their capacity to provide a 'danger ' s ignal and 36 stimulate the maturation and function o f A P C s is w e l l characterized . H o w e v e r , T L R s are on ly one component o f a large fami ly o f proteins that transduce s t imul i o f the innate immune system, and molecules such as the N O D s , the mannose-6-phosphate receptor, dect in-1, and var ious other scavenger receptors have s imi lar functions 3 7 ~ 3 9 . V i a their capacity to stimulate A g presentation and co-s t imulat ion, the P A M P / A P C interaction is c o m m o n l y thought to provide the cr i t ica l l i n k between the innate and adaptive immune responses. H o w e v e r , this s impl is t ic mode l fails to expla in h o w the i m m u n e system can 6 discriminate between pathogenic and commensal organisms, which also encode PAMPs. Recent data from our own and several other laboratories lead us to propose that a PAMP/Treg cell interaction may represent an additional and key cellular link between the innate and adaptive immune systems. 1.2.2 T L R s indirectly control the function of C D 4 + C D 2 5 + Tregs TLRs potently stimulate APCs to mature, produce cytokines and upregulate a variety of costimulatory molecules. Together, these factors lead to vigorous stimulation of T cells, such that they are no longer subject to regulation by Tregs. The induction of resistance to suppression appears to be cytokine-mediated. One characterised pathway involves the secretion by TLR-activated DCs of a number of cytokines, such as IL-6 and IL-1 (3, that act on CD4 + T effector cells to stabilize IL-2 mRNA 4 0 4 1 and prevent CD4 + CD25 + nTreg cell-mediated suppression. Interestingly, when dendritic cells are stimulated via TLRs, they also begin producing IL-2 4 2 . Although the resulting high local concentrations of IL-2 prevent suppression, at least transiently, they likely also promote the CD4 + CD25 + Tregs themselves, since this cytokine is essential for their survival and regulatory function 4 3 . The capacity of TLRs to regulate expression of co-stimulatory molecules such as GITR-Iigand may also be important in controlling the balance between effector and CD4 + CD25 + Treg cells. Although ligation of GITR renders CD4 +CD25" T effector cells resistant to suppression44, it also provides a potent co-stimulatory signal for Treg cells 4 5 . Together, these findings suggest that an unrecognized role of TLR stimulation may be the long-term enhancement of Treg cell numbers and/or function. In vivo evidence that further supports this hypothesis comes from analysis of TLR2 deficient mice. In the absence of TLR2, mice display an increased resistance to C. albicans infection 4 6 with 7 higher levels o f I F N - y and reduced I L - 1 0 compared to controls. A s discussed be low, the f inding that T L R 2 deficient mice have lower levels o f c i rcula t ing C D 4 + C D 2 5 + Tregs suggest that T L R s may also direct ly influence the development and/or function o f T reg cel ls . 1.2.3 Can TLRs directly control the function of CD4 +CD25 + Tregs? In addi t ion to evidence that T L R s control the function o f C D 4 + C D 2 5 + n T r e g cel ls indi rec t ly v i a A P C s , several recent reports suggest that these molecules may also di rect ly affect T r e g cel ls themselves, summar ized i n Figure 1.2. Netea et a l . found that s t imulat ion o f T L R 2 signif icantly increases the surv iva l o f C D 4 C D 2 5 Tregs . Moreove r , T L R 2 - / - mice have reduced numbers o f C D 4 + C D 2 5 + Tregs, further support ing their i n vitro f indings. Figure 1.1. Can PAMPs interact directly with T cells? PAMPs bind to TLRs on APCs, modulating presentation and production of cytokines. An unresolved question is whether PAMPs can directly effect Treg and Te growth and function. M u r i n e C D 4 + C D 2 5 + nTreg cells express more T L R 4, 5, 7 and 8 m R N A i n compar ison to C D 4 + C D 2 5 " effector T cel ls 4 7 . Importantly, i n vi t ro exposure to L P S , a T L R 4 l igand, has been reported to enhance surv iva l and prol i fera t ion o f murine 8 C D 4 + C D 2 5 + Tregs. T h i s does not appear to be a robustly reproducible observat ion, however , poss ib ly depending on the source o f L P S 4 8 , and/or the presence o f contaminants (eg pept idoglycan) i n some batches o f L P S 4 9 . A c ruc ia l addi t ional question is whether T L R s may direct ly influence the suppressive capacity o f C D 4 + C D 2 5 + nT reg cel ls . Caramalho et a l . reported that L P S increases the suppressive capacity o f murine C D 4 + C D 2 5 + T r eg cel ls 4 7 . Further experiments by Sutmul ler et a l revealed that T L R 2 co-s t imula t ion induced T reg prol i ferat ion, and was associated w i t h a transient loss i n suppressive capaci ty that returned after resting i n .the absence o f T L R 2 s t imulat ion 5 0 ' 5 1 . S i m i l a r l y , T L R 2 -dependent H S P 6 0 pre-treatment increases the suppressive capacity o f Tregs . It seems quite l i k e l y that these and s imi la r mechanisms might underl ie the host's tolerance o f commensa l bacteria i n the intestine. There is also evidence that expression o f T L R s may be a more general phenomenon, and that the capacity o f P R R s to direct ly influence T - c e l l responses is not l imi t ed to T reg cel ls . C D 4 + T effector cel ls can express T L R 2, 4, 5 and 9 4 7 > 5 3 - 5 5 . Expres s ion levels o f the other T L R s remain to be investigated. Interestingly, T L R 2 and 5 appear to funct ion as co-st imulatory receptors on activated human C D 4 + T ce l ls , and may participate i n the development and maintenance o f T ce l l memory 5 3 ' 5 5 . T L R 2 s t imulat ion increased the prol i fera t ion and I L - 2 product ion o f effector T cel ls , rendering them transiently unsusceptible to suppression 5 0 . In addi t ion C p G D N A , a T L R 9 l igand, promotes surv iva l o f activated C D 4 + T cel ls 5 4 . M u c h w o r k remains to be done i n this exc i t ing f i e ld , as, due to the lack o f appropriate antibodies most studies have, as yet, only reported levels o f T L R m R N A . In addi t ion, a thorough compar i son o f express ion levels 9 and funct ion i n C D 4 + T cel ls versus t radi t ional T L R - r e s p o n d i n g cel ls (i.e. monocytes and D C s ) is required. O f part icular interest is h o w certain pathogens may expedite tolerance and act ively promote a pro-tolerogenic environment. F o r example, it has already been demonstrated that A P C s exposed to Trypanosoma brucei brucei induce a tolerogenic m i l i e u that abrogates the development o f E A E 5 6 . In addi t ion treatment o f mice w i t h a k i l l e d Mycobacterium vaccae-suspension, induced al lergen-specific T reg ce l l s w h i c h confer protect ion against a i rway in f lammat ion 5 7 . Semina l research conducted by B e l k a i d et a l . has demonstrated that infect ion w i t h L e i s h m a n i a induces a potent T r e g popula t ion , w h i c h maintains both a parasite reservoir, and i m m u n o l o g i c a l memory and tolerance to repeat infections 3 5 - 5 8 - 6 0 . A better understanding o f the molecular basis o f these phenomena is urgently required, g iven the potential therapeutic applicat ions o f b l o c k i n g or harnessing mechanisms o f tolerance that have evo lved over m i l l i o n s o f years. Unders tanding h o w bacteria manipulate immune responses w i l l not on ly enable the design o f better defenses against pathogenic organisms, but also point to nove l immunomodu la to ry therapies for a variety o f human diseases. 1.3. TCR SIGNALING IN TREGS One o f the def ining characteristics o f C D 4 + C D 2 5 + Tregs is their i n vi t ro anergy. C D 4 + C D 2 5 + Tregs are hypoprol i ferat ive to p o l y c l o n a l s t imul i , and fa i l to proliferate i n the presence o f I L - 2 alone, al though the combined signals o f T C R s t imula t ion and exogenous I L - 2 are able to overcome anergy 1 0 . T h i s aberrant prol iferat ive response to p o l y c l o n a l T C R st imulat ion suggests that altered s ignal ing downstream o f the T C R and/or I L - 2 receptor must exist i n C D 4 + C D 2 5 + Tregs. 10 1.3.1 Overview of T C R signaling A n o v e r v i e w o f the s ignal ing events downstream o f the T C R is depicted i n F igure 1.3. T C R s ignal ing is init iated by A g - b i n d i n g , and results i n act ivat ion o f protein tyrosine kinases o f the Src , S y k and Tec famil ies , and assembly o f scaffolds o f adaptor molecules 6 1 . Phosphory la t ion o f these adaptors subsequently leads to act ivat ion o f downstream effectors, i nc lud ing serine/threonine kinases such as the mi togen activated prote in kinases ( M A P K s ) and protein kinase C ( P K C s ) , and phosphat idyl inosit ide-3 kinase (PI3'In-dependent serine/threonine kinases such as A K T . A c t i v a t i o n o f these cascades results i n cytoskeletal rearrangements, cy tokine product ion , c e l l cyc l e progress ion and engagement o f T ce l l effector functions. 11 Figure 1.2. Overview of TCR signaling. TCR binding with CD28 co-stimulation results in the recruitment of kinases and adaptor molecules to the plasma membrane, allowing complex formation and clustering to occur. The major downstream consequences are the nuclear translocation of transcription factors, and the modulation of proteins controlling cell survival and proliferation. The act ivat ion o f P I 3 ' K is greatly enhanced by C D 2 8 co-s t imula t ion 6 2 ' 6 3 . The product o f P I 3 ' K act ivat ion is phosphat idyl inositol-3,4,5-triphosphate (PIP3) . PIP3 is a b io log ica l ly significant molecule i n that it binds to pleckstr in homology domains ( P H ) , and thus recruits P H domain-conta in ing kinases and adaptor proteins to the membrane 6 2 . The act ion o f P I 3 K is reversed by phosphatases and tensin homologue deleted on chromosome 10 ( P T E N ) and src homology 2 (SH2)-conta in ing inosi to l phosphatase 12 ( S H I P ) . Muta t ions or deletions o f P T E N are c o m m o n i n many types o f tumors b \ and T ce l l specif ic P T E N knockouts develop severe lymphoprol i fera t ive disease 6 5 . A K T has mul t ip le targets, both direct and indirect, whose effects co l l ec t ive ly push the c e l l towards surv iva l and prol i ferat ion. F u l l act ivat ion o f A K T requires phosphoryla t ion at two different sites: Thr308 and Ser473 6 6 ' 6 7 . The act ivat ion or regulatory loop o f A K T is control led by phosphoryla t ion at Thr308 by a wel l - s tudied 1 66 67 kinase, phosphoinosit ide-dependent prote in kinase 1 ( P D K 1 ) . ' . Ser473 phosphoryla t ion is required for funct ion o f the kinase d o m a i n o f A K T , and is phosphorylated by the elusive P D K 2 6 6 ' 6 7 . The molecula r identity o f P D K 2 has been intensely debated, w i t h Integrin l i nked kinase ( I L K ) 6 8 " 7 0 , D N A - P K 1 1 , and even autophosphorylat ion 7 2 proposed as candidates for this kinase. Recent evidence has strongly suggested that w h e n complexed w i t h r ictor, m a m m a l i a n target o f r apamyc in ( m T O R ) phosphorylates Ser473 o n A K T , and thus is P D K 2 7 3 . Howeve r , when complexed w i t h raptor, m T O R remains a target o f A K T , and thus m T O R can be simultaneously up-stream and down-stream o f A K T ac t iva t ion 7 3 . A l t h o u g h it was prev ious ly thought that phosphoryla t ion at both Thr308 and Ser473 was required for A K T function, it has n o w been suggested that phosphoryla t ion at either site can occur independently, and that a separation exists that splits the act ion o f A K T depending on w h i c h site is phophorylated 7 3 " 7 7 . 1.3.2 T cell anergy T ce l l anergy is a term coined to describe the lack o f responsiveness characteristic o f a T ce l l that had received signal 1 without s ignal 2; that is to say s t imulat ion through the T C R without a co-st imulatory s ignal through C D 2 8 or the I L - 2 receptor . T effector 13 cel ls can be rescued f rom this anergy by the addi t ion o f I L - 2 dur ing T C R and C D 2 8 co-s t imulat ion . The b ind ing o f antigen to the T C R results i n the act ivat ion o f numerous s ignal ing molecules , as depicted i n F i g . 1.3, and culminate i n the release o f intracel lular Ca++ and the nuclear translocation o f N F A T 7 8 . Co-s t imula t ion w i t h C D 2 8 results i n the format ion o f A P - 1 and nuclear translocation o f N F k B 7 8 . N F A T complexes w i t h A P - 1 , and induces fu l l T ce l l act ivat ion. H o w e v e r , the presence o f N F A T without A P - 1 78 initiates a different transcriptional program, and the result is an anergic T ce l l . T h i s anergy p rogram appears to invo lve the up-regulat ion o f several E 3 ub iqu i t in ligases, par t icular ly I T C H , cbl -b , and G R A I L 7 8 " 8 1 . Further evidence for the importance o f these molecules can be found i n the observat ion that cb l -b deficient T effector cel ls are resistant to suppression by Tregs 8 2 , suggesting that an abi l i ty to respond to anergy signals i n effector T cel ls is necessary for suppression. 1.3.3 Signal ing in C D 4 + C D 2 5 + Tregs 1.3.3.1 Developmental Signaling Requirements Deve lopmenta l ly , it appears that C D 4 + C D 2 5 + Tregs have unique s ignal ing requirements. It has l ong been recognized, that a requirement for the t hymic development o f Tregs is I L - 2 8 3 . B o t h I L - 2 and C D 2 5 deficient mice fa i l to develop Tregs, and this is associated w i t h severe lymphoprol i fera t ive diseases 8 3 ' 8 4 . Indeed, there is an essential requirement for funct ional S T A T 5 s ignal ing i n Tregs to mediate I L - 2 signals 8 5 . It has been previous ly reported 2 ' 8 6 ' 8 7 that Treg development requires s ignal ing through the T C R . Further experiments have suggested that a very specif ic interaction through L A T ( l inker for act ivat ion o f T cells) and P L C y is required for the development 14 o f F O X P 3 + C D 4 + C D 2 5 + Tregs 8 8 . Interestingly, this interaction is required for T C R mediated C a 2 + f lux and N F A T act ivat ion, suggesting an involvement between N F A T 8 8 levels and F O X P 3 expression . 1.3.3.2. IL-2R and TCR signaling Recent investigations into the intracel lular s igna l ing o f C D 4 + C D 2 5 + Tregs have largely concentrated on events downstream o f the I L - 2 R . Bensinger et a l first described the altered mur ine T reg response to I L - 2 , demonstrating that cel ls do not beg in to d iv ide , but enlarge i n size and up-regulate anti-apoptotic molecules such as b c l - x L 8 9 . Investigations into downstream s ignal ing events determined that mur ine Tregs have a fu l ly funct ional J A K - S T A T pathway, but that a defect exists i n the P I 3 K pathway 8 9 . Further investigations focused on the P I 3 K pathway, and revealed that aberrant regulat ion o f the phosphatase P T E N was responsible for the b lock i n T r e g prol i ferat ion, as targeted delet ion o f P T E N restored Treg prol iferat ion i n response to I L - 2 9 0 . Downs t r eam o f the T C R , experiments performed using human cord b l o o d der ived ce l l l ines describe a defect i n E R K act ivat ion and an increase i n p 2 7 k i p l that prevents ce l l cyc l e progression 9 1 . Interestingly, no defect i n A K T act ivat ion was observed i n these experiments 9 1 . A d d i t i o n a l l y , experiments i n ex v i v o human Tregs also revealed a decrease i n Z A P - 7 0 and S L P - 7 6 phosphoryla t ion, result ing i n defective V A V recruitment 9 2 . A defect i n E R K act ivat ion downst ream o f the T C R was also observed i n murine Tregs , poss ib ly as a result o f d imin ished P L C y act ivi ty 9 3 . C o l l e c t i v e l y , these studies suggest mul t ip le s ignal ing alterations exist i n Tregs as compared to T effector cel ls , downstream o f both the I L - 2 R and the T C R . E luc ida t i ng 15 these alterations w i l l be very important for the understanding o f under ly ing T reg b io logy , and very benef ic ia l for any c l i n i c a l applicat ions requi r ing the ex v i v o expansion o f Tregs. 1.4. T H E S E A R C H F O R A T R U E C E L L S U R F A C E M A R K E R F O R C D 4 + C D 2 5 + T R E G S T h e great therapeutic potential for C D 4 + C D 2 5 + Tregs has revealed an urgent need for tools that w o u l d a l l o w detection and pur i f ica t ion o f homogeneous populat ions o f Tregs . A s discussed above, C D 4 + C D 2 5 + T reg cel ls were in i t i a l ly defined by their consti tutive expression o f C D 2 5 , a ce l l surface marker t radi t ional ly used to identify activated effector T cel ls . The expression o f C D 2 5 o n activated effector T cel ls is transient, rather than the const i tut ively h igh expression associated w i t h C D 4 + C D 2 5 + Tregs 9 4 , 9 5 . In general, C D 4 + C D 2 5 + Tregs have a cell-surface phenotype that is indis t inguishable f rom that o f activated effector T cel ls . F O X P 3 represents a more specific marker, but i ts ' intracel lular expression renders it useless for i so la t ion o f l ive cel ls . In humans, enriched, but not pure, populat ions o f C D 4 + T reg cel ls exist w i t h i n the - 1 - 2 % brightest C D 2 5 + cel ls , and we have recently shown that the majori ty o f these cel ls also express F O X P 3 1 0 . Howeve r , even these h igh ly pur i f ied populat ions are s t i l l contaminated w i t h non-regulatory T effector ce l ls , w h i c h invar iab ly ou tgrow Tregs dur ing i n vi tro culture. 1.4.1. Microarray analysis of C D 4 + C D 2 5 + Tregs The search for a specific ce l l surface marker o f C D 4 + C D 2 5 + Tregs has been undertaken by a number o f groups, and the molecules proposed as candidate have been summar ized i n F igure 1.4. Fo r example , Sakaguchi et al took the approach o f us ing C D 4 + C D 2 5 + T r eg cel ls themselves as an immunogen , and generated a series o f ant i -Treg 16 monoclonal antibodies (mAbs) 9 6 . This process lead to the discovery of GITR, a molecule that is highly expressed on Treg cells, but unfortunately is also expressed by activated T cells. Gavin et al. performed cDNA microarrays comparing murine CD4"CD25" T cells and CD4+CD25+ Tregs, and identified that among others, OX-40, GITR, and CTLA-4 were more highly expressed in Tregs, as were inhibitors of signaling such as SOCS-1 9 1 . cDNA microarrays on murine CD4+CD25" T cells and CD4+CD25+ Tregs were also performed by McHugh et al, and compared gene expression profiles either at rest or after activation with aCD3 and IL2 8 4 . At rest, GITR, OX-40, CTLA-4 and 4-IBB were again observed to be upregulated in Tregs, as was CD 103, an integrin expressed by intraepithelial lymphocytes . However, as was previously observed, GITR, OX-40, CTLA-4 and 4-IBB were all up-regulated in CD4+CD25" T effector cells following activation84. CD 103 was the exception in that it was not up-regulated on T effector cells following activation, however separation on the basis of CD 103 expression was not OA found to identify suppressive capacity ; Further, CD 103 does not appear to be expressed on human CD4+CD25+ Tregs25. Resting CD4 + CD4 + Treg Activated CD4 + T effector T effector Figure 1.3 Treg-associated molecules proposed as cell surface markers. Comparison of expression of potential Treg cell surface markers on resting and activated CD4+ T effector cells with CD4+CD25+ Tregs. 17 M o r e recent efforts have suggested that neuropil in-1 ( N R P l ) , a receptor i n v o l v e d i n axon guidance and angiogenesis, as w e l l as T ce l l act ivat ion, is a specif ic ce l l surface marker o f C D 4 + C D 2 5 + Tregs 9 8 . Bruder et a l . again used sorted mur ine p o l y c l o n a l C D 4 + C D 2 5 " T e and C D 4 + C D 2 5 + Tregs, and inc luded antigen specific T effectors and Tregs i n the compar ison as w e l l . Bruder et a l looked at genes regulated i n a s imi la r manner to F o x p 3 pre- and post act ivat ion, and found 2 ce l l surface proteins: neuropi l in-1 and k i l l e r c e l l l ec t in- l ike receptor G l ( K L R G 1 ) 9 8 . A s on ly 3 % o f Tregs expressed K L R G 1 o n the c e l l surface, Brude r et a l focused o n neuropi l in -1 . Neu rop i l i n -1 is not up-regulated o n activated T effectors, and selection on the basis o f C D 4 + N R P h l g h produces a popula t ion expressing h igh levels o f F O X P 3 and hav ing potent suppressive capacity 9 8 . A l t h o u g h very p romis ing , further research into neuropil in-1 has been hampered by the . l ack o f good ant i-human neuropi l in-1 m A b s and thus its relevance i n the human cannot . be conf i rmed. A d d i t i o n a l l y , it has n o w been suggested that neuropi l in-1 is also expressed on F O X P 3 " , non-suppressive anergic T cel ls , suggesting it is not a specif ic marker for C D 4 + C D 2 5 + T r e g s 9 9 1 0 0 . Recen t ly , i t has been suggested that the expression o f C D 127 ( I L - 7 R a ) may be useful i n d i sc r imina t ing between activated effector T cel ls and Tregs. L i u et a l have demonstrated that C D 1 2 7 expression inversely correlates w i t h F o x P 3 expression 1 0 1 . Se lec t ion o n the basis o f C D 4 + C D 2 5 h i g h C D 1 2 7 l 0 W produces a popula t ion that is h igh ly enr iched for F O X P 3 + Tregs 1 0 1 . H o w e v e r , expression o f C D 127 on C D 4 + C D 2 5 " T effector cel ls decreases f o l l o w i n g act ivat ion, and thus contaminat ion w i t h activated T effector cel ls remains a concern , 0 ° . 18 Past studies are inherently f lawed for 2 reasons. Firs t , as l ong as the input popula t ion is a bu lk popula t ion o f C D 4 + C D 2 5 + T r eg cel ls , contaminat ing effector T cel ls c o u l d s ignif icant ly skew the data. M o r e o v e r , it is essential that candidate molecu les be r igorous ly selected based on h igh specific expression i n both the resting and activated state. 1.5. H Y P O T H E S E S A l t h o u g h Tregs are k n o w n to be cr i t ica l regulators o f peripheral tolerance, they are poor ly understood at a molecular level . O v e r a l l , the purpose o f this research was to investigate further the molecular phenotype o f human C D 4 + C D 2 5 + Tregs, and the relat ionship between this phenotype and the b io logy o f Tregs; par t icular ly in vitro anergy and suppression. The hypotheses were as fo l lows : 1. S igna l i ng through T L R s w i l l affect the proliferat ion and suppressive capacity o f human Tregs. 2. K n o w l e d g e o f the unique intracel lular s ignal ing pathways operational i n Tregs w i l l elucidate functional differences w i t h T effector cel ls . 3. H u m a n Tregs express a unique set o f proteins not found o n C D 4 + T effector cel ls . 19 1.6. R E F E R E N C E S I. Sakaguchi S, Sakaguchi N , A s a n o M , Itoh M , T o d a M . Immuno log ic self-tolerance mainta ined by activated T cel ls expressing I L - 2 receptor alpha-chains ( C D 2 5 ) . B r e a k d o w n o f a single mechan i sm o f self-tolerance causes var ious autoimmune diseases. J I m m u n o l . 1995;155:1151-1164. . 2. Thorn ton A M , Shevach E M . C D 4 + C D 2 5 + immunoregulatory T cel ls suppress p o l y c l o n a l T ce l l act ivat ion i n v i t ro by inh ib i t ing inter leukin 2 product ion. J E x p M e d . 1998;188:287-296. 3. B r u n k o w M E , Jeffery E W , Hje r r i l d K A , et a l . D i s rup t ion o f a new forkhead/winged-he l ix protein, scurfin, results i n the fatal lymphoprol i fera t ive disorder o f the scurfy mouse. N a t Genet. 2001;27:68-73. 4. H o r i S, N o m u r a T , Sakaguchi S. C o n t r o l o f regulatory T ce l l development by the t ranscript ion factor F o x p 3 . Science. 2003;299:1057-1061. 5. Kha t t r i R , K a s p r o w i c z D , C o x T , et a l . T h e amount o f scurfin protein determines peripheral T . c e l l number and responsiveness. J I m m u n o l . 2001;167:6312-6320. 6. ' Ronca ro lo M G , G r e g o r i S, Bat tag l ia M , Bacchet ta R , Fleischhauer K , L e v i n g s M K . Interleukin-10-secreting type 1 regulatory T cel ls i n rodents and humans. I m m u n o l R e v . 2006;212:28-50. 7. Fu j i sh ima M , H i r o k a w a M , Fu j i sh ima N , Sawada K . T C R a l p h a b e t a repertoire divers i ty o f human naturally occur r ing C D 4 + C D 2 5 + regulatory T cel ls . I m m u n o l Lett . 2005;99:193-197. 8. H s i e h C S , L i a n g Y , T y z n i k A J , S e l f S G , L igg i t t D , Rudensky A Y . Recogn i t i on o f the peripheral se l f by naturally ar is ing C D 2 5 + C D 4 + T c e l l receptors. Immuni ty . 2004;21:267-277. 9. O c h s H D , Z ieg le r S F , Torgerson T R . F O X P 3 acts as a rheostat o f the immune response. I m m u n o l R e v . 2005;203:156-164. 10. L e v i n g s M K , Sangregorio R , Sartirana C , et a l . H u m a n C D 2 5 + C D 4 + T suppressor ce l l clones produce transforming growth factor beta, but not inter leukin 10, and are distinct f rom type 1 T regulatory cel ls . J E x p M e d . 2002;196:1335-1346. I I . V i g l i e t t a V , B a e c h e r - A l l a n C , W e i n e r H L , Haf ler D A . L o s s o f funct ional suppression by C D 4 + C D 2 5 + regulatory T cel ls i n patients w i t h mul t ip le sclerosis. J E x p M e d . 2004;199:971-979. 20 12. Roncador G , B r o w n P J , Maestre L , et a l . A n a l y s i s o f F O X P 3 protein expression i n human C D 4 ( + ) C D 2 5 ( + ) regulatory T ce l l s at the s ingle-cel l leve l . E u r J I m m u n o l . 2005;35:1681-1691. 13. N a k a m u r a K , K i t a n i A , Fuss I, et al . T G F - b e t a 1 plays an important role i n the mechan i sm o f C D 4 + C D 2 5 + regulatory T ce l l act ivi ty i n both humans and mice . J I m m u n o l . 2004;172:834-842. 14. A s s e m a n C , R e a d S, Powr i e F . Co l i togen ic T h l cells are present i n the antigen-experienced T ce l l poo l in normal mice : control by C D 4 + regulatory T cel ls and I L - 1 0 . J I m m u n o l . 2003;171:971-978. 15. Fah len L , Read S, G o r e l i k L , et a l . T cel ls that cannot respond to T G F - b e t a escape control by C D 4 ( + ) C D 2 5 ( + ) regulatory T cel ls . J E x p M e d . 2005;201:737-746. 16. P i c c i r i l l o C A , Letter io JJ , Thorn ton A M , et a l . CD4(+)CD25(+) regulatory T cel ls can mediate suppressor funct ion i n the absence o f transforming growth factor be ta l product ion and responsiveness. J E x p M e d . 2002;196:237-246. 17. G o n d e k D C , L u L F , Quezada S A , Sakaguchi S, N o e l l e R J . Cut t ing Edge : Contact-M e d i a t e d Suppression by C D 4 + C D 2 5 + Regula tory C e l l s Involves a G r a n z y m e B-Dependent , Perforin-Independent M e c h a n i s m . J I m m u n o l . 2005;174:1783-1786. 18. B u c k n e r J H , Z ieg le r S F . Regula t ing the immune system: the induc t ion o f regulatory T cel ls i n the periphery. Ar th r i t i s Res Ther. 2004;6:215-222. 19. W a l k e r M R , Carson B D , N e p o m G T , Z ieg le r S F , B u c k n e r J H . D e novo generation o f antigen-specific C D 4 + C D 2 5 + regulatory T cel ls f rom human C D 4 + C D 2 5 - cel ls . P roc N a t l A c a d S c i U S A . 2005;102:4103-4108. 20. M u c i d a D , Ku tchukh idze N , E r a z o A , Russo M , La fa i l l e JJ , Curot to de La fa i l l e M A . O r a l tolerance i n the absence o f naturally occurr ing Tregs. J C l i n Invest. 2005. 21 . C h e n W , J i n W , Hardegen N , et a l . C o n v e r s i o n o f peripheral C D 4 + C D 2 5 - naive T cel ls to C D 4 + C D 2 5 + regulatory T cel ls by T G F - b e t a induct ion o f t ranscript ion factor F o x p 3 . J E x p M e d . 2003;198:1875-1886. 22. M a r i e J C , Let ter io JJ , G a v i n M , Rudensky A Y . T G F - b e t a l maintains suppressor funct ion and F o x p 3 expression i n C D 4 + C D 2 5 + regulatory T cel ls . J E x p M e d . 2005;201:1061-1067. 23. Ze lenay S, Lopes -Ca rva lho T , Caramalho I, Moraes-Fontes M F , Rebe lo M , Demengeot J . F o x p 3 + C D 2 5 - C D 4 T cel ls constitute a reservoir o f commit ted regulatory cel ls that regain C D 2 5 expression upon homeostatic expansion. P r o c . N a t l A c a d S c i U S A . 2005;102:4091-4096. 21 24. Sa lva t i V M , M a z z a r e l l a G , Gianfrani C , e t a l . Recombinant human in ter leukin 10 suppresses g l i a d i n dependent T ce l l act ivat ion i n ex v i v o cultured coel iac intestinal mucosa . Gut . 2005;54:46-53. 25 . L e v i n g s M K , Roncaro lo M G . Phenotyp ic and functional differences between human C D 4 + C D 2 5 + and type 1 regulatory T cel ls . C u r r T o p M i c r o b i o l I m m u n o l . 2005;293:303-326. 26. L e v i n g s M K , Bacchet ta R , S c h u l z U , Ronca ro lo M G . The role o f I L - 1 0 and T G F - b e t a i n the differentiation and effector funct ion o f T regulatory cel ls . Int A r c h A l l e r g y I m m u n o l . 2002;129:263-276. 27. W a l d m a n n H , B e m e l m a n F , C o b b o l d S. Tolerance induct ion w i t h C D 4 monoc lona l antibodies. Nova r t i s F o u n d S y m p . 1998;215:146-152; discussion 152-148, 186-190. 28. W a l d m a n n H , A d a m s E , F a i r c h i l d P , C o b b o l d S. Infectious tolerance and the long-term acceptance o f transplanted tissue. I m m u n o l R e v . 2006;212:301-313. 29 . Z h e n g S G , W a n g J H , G r a y J D , Souc ie r H , H o r w i t z D A . Na tu ra l and induced C D 4 + C D 2 5 + cel ls educate C D 4 + C D 2 5 - cel ls to develop suppressive act iv i ty : the role o f . I L -2, T G F - b e t a , and I L - 1 0 . J I m m u n o l . 2004;172:5213-5221. 30. D i e c k m a n n D , Bruett C H , Ploettner H , L u t z M B , Schuler G . H u m a n C D 4 ( + ) C D 2 5 ( + ) regulatory, contact-dependent T cel ls induce in ter leukin 10-producing, contact-independent type 1 - l i ke regulatory T cel ls [corrected]. J E x p M e d . 2002; 196:247-253. 31 . Jonulei t H , Schmit t E , K a k i r m a n H , Stassen M , K n o p J , E n k A H . Infectious . tolerance: human C D 2 5 ( + ) regulatory T cel ls convey suppressor act ivi ty to convent ional C D 4 ( + ) T helper cel ls . J E x p M e d . 2002;196:255-260. 32. D e n n i n g T L , K i m G , Kronenbe rg M . Cut t ing edge: C D 4 + C D 2 5 + regulatory T cells impa i red for intestinal h o m i n g can prevent col i t i s . J I m m u n o l . 2005;174:7487-7491. 33. C o o m b e s J L , R o b i n s o n N J , M a l o y K J , U h l i g H H , P o w r i e F . Regula tory T cel ls and intestinal homeostasis. I m m u n o l R e v . 2005;204:184-194. 34. P o w r i e F.- Immune Regula t ion i n the Intestine: A Ba lanc ing A c t between Effector and Regula tory T C e l l Responses. A n n N Y A c a d S c i . 2004;1029:132-141. / • • 35. B e l k a i d Y , Rouse B T . Na tu ra l regulatory T cel ls i n infectious disease. N a t I m m u n o l . 2005;6:353-360. 36. Pasare C , M e d z h i t o v R . Tol l -dependent control mechanisms o f C D 4 T c e l l act ivat ion. Immuni ty . 2004;21:733-741. 37. A b r e u M T , Fuka ta M , A r d i t i M . T L R s ignal ing i n the gut i n health and disease. J I m m u n o l . 2005;174:4453-4460. 22 38. M u k h o p a d h y a y S, Herre J , B r o w n G D , G o r d o n S. The potential for T o l l - l i k e receptors to collaborate w i t h other innate i m m u n e receptors. Immuno logy . 2004 ; 112:521-530. 39. Takeda K , A k i r a S. T o l l - l i k e receptors i n innate immuni ty . Int I m m u n o l . 2005 ; 17:1-14. 40 . K u b o T , Hat ton R D , O l i v e r J , L i u X , E l s o n C O , Weaver C T . Regula tory T c e l l suppression and anergy are differential ly regulated by proinf lammatory cytokines produced by T L R - a c t i v a t e d dendri t ic cel ls . J I m m u n o l . 2004;173:7249-7258. 4 1 . Pasare C , M e d z h i t o v R . T o l l pathway-dependent blockade o f C D 4 + C D 2 5 + T c e l l -mediated suppression by dendrit ic cel ls . Science. 2003;299:1033-1036. E p u b 2003 Jan 1016. 42 . Rogers N C , S lack E C , Edwards A D , et a l . Syk-dependent cytokine induc t ion by Dect in-1 reveals a nove l pattern recogni t ion pathway for C type lectins. Immuni ty . 2005;22:507-517. 43. Thorn ton A M , D o n o v a n E E , P i c c i r i l l o C A , Shevach E M . Cut t ing edge: I L - 2 is c r i t i ca l ly required for the i n vi t ro act ivat ion o f C D 4 + C D 2 5 + T ce l l suppressor function. J I m m u n o l . 2004;172:6519-6523. 44. Stephens G L , M c H u g h R S , Whit ters M J , et a l . Engagement o f g lucocor t ico id- induced T N F R family-rela ted receptor o n effector T cel ls by its l igand mediates resistance to suppression by C D 4 + C D 2 5 + T cel ls . J I m m u n o l . 2004;173:5008-5020. 45 . K a n a m a r u F , Y o u n g n a k P , H a s h i g u c h i M , et a l . Cos t imula t ion v i a g lucocor t i co id -induced T N F receptor i n both convent ional and C D 2 5 + regulatory C D 4 + T cel ls . J Immuno l . 2004;172:7306-7314. 46. Ne tea M G , Sutmul ler R , H e r m a n n C , et a l . T o l l - l i k e receptor 2 suppresses immuni ty against C a n d i d a albicans through induc t ion o f I L - 1 0 and regulatory T cel ls . J I m m u n o l . 2004;172:3712-3718. 47. Caramalho I, Lopes -Ca rva lho T , Ostler D , Ze lenay S, Haury M , Demengeot J . Regula tory T cel ls selectively express to l l - l i ke receptors and are activated by l ipopolysacchar ide . J E x p M e d . 2003;197:403-41 L . 48. Fehervar i Z , Sakaguchi S. C o n t r o l o f F o x p 3 + C D 2 5 + C D 4 + regulatory ce l l act ivat ion and funct ion by dendri t ic cells . Int I m m u n o l . 2004;16:1769-1780. 49 . Ne tea M G , V a n der M e e r J W , K u l l b e r g B J . T o l l - l i k e receptors as an escape mechan i sm f rom the host defense. Trends M i c r o b i o l . 2004;12:484-488. 50. L i u H , K o m a i - K o m a M , X u D , L i e w F Y . T o l l - l i k e receptor 2 s ignal ing modulates the functions o f C D 4 + C D 2 5 + regulatory T cel ls . P r o c N a t l A c a d S c i U S A . 2006 ; 103:7048-7053. 23 51. Sutmul le r R P , den B r o k M H , K r a m e r M , et a l . T o l l - l i k e receptor 2 controls expansion and funct ion o f regulatory T cel ls . J C l i n Invest. 2006; 116:485-494. 52. Z a n i n - Z h o r o v A , Caha lon L , T a l G , M a r g a l i t R , L i d e r O , C o h e n I R . Heat shock prote in 60 enhances C D 4 + C D 2 5 + regulatory T ce l l funct ion v i a innate T L R 2 s ignal ing. J C l i n Invest. 2006;116:2022-2032. 53. K o m a i - K o m a M , Jones L , O g g G S , X u D , L i e w F Y . T L R 2 is expressed o n activated T cel ls as a cost imulatory receptor. P roc N a t l A c a d S c i U S A . 2004;101:3029-3034. 54. G e l m a n A E , Z h a n g J , C h o i Y , T u r k a L A . T o l l - l i k e receptor l igands di rect ly promote activated C D 4 + T c e l l surv iva l . J I m m u n o l . 2004;172:6065-6073. 55. C r e l l i n N , G a r c i a R V , A l l a n S E , Steiner T S , L e v i n g s M K . H u m a n C D 4 + T ce l l s express T L R 5 and its l igand f lage l l in enhances the suppressive capacity and expression o f F O X P 3 i n C D 4 + C D 2 5 + T regulatory cells , submitted. 56. W a l l b e r g M , Harr i s R A . Co- in fec t ion w i t h Trypanosoma brucei brucei prevents experimental autoimmune encephalomyel i t is i n D B A / 1 mice through induc t ion o f suppressor A P C s . Int I m m u n o l . 2005;17:721-728. 57. Z u a n y - A m o r i m C , S a w i c k a E , M a n l i u s C , et a l . Suppression o f a i rway eos inophi l i a by k i l l e d M y c o b a c t e r i u m vaccae-induced al lergen-specific regulatory T-ce l l s . N a t M e d . 2002;8:625-629. 58. B e l k a i d Y . The role o f C D 4 ( + ) C D 2 5 ( + ) regulatory T cel ls i n Leishma'nia infect ion. Exper t O p i n B i o l Ther . 2003;3:875-885. 59. B e l k a i d Y , P i c c i r i l l o C A , M e n d e z S, Shevach E M , Sacks D L . C D 4 + C D 2 5 + regulatory T cel ls control L e i s h m a n i a major persistence and immuni ty . Nature . 2002;420:502-507. 60. B e l k a i d Y , Sun C M , B o u l a d o u x N . Parasites and immunoregula tory T cel ls . C u r r O p i n I m m u n o l . 2006. 61 . Can t re l l D A . T - c e l l antigen receptor s ignal transduction. Immuno logy . 2002;105:369-374. 62 . K a n e L P , We i s s A . The PI-3 k ina se /Ak t pathway and T c e l l ac t ivat ion: p le io t ropic pathways downstream o f P I P 3 . I m m u n o l R e v . 2003;192:7-20. 63. O k k e n h a u g K , B i l a n c i o A , E m e r y J L , Vanhaesebroeck B . Phosphoinos i t ide 3-kinase i n T c e l l ac t ivat ion and surv iva l . B i o c h e m Soc Trans. 2004;32:332-335. 64. V a z q u e z F , Devreotes P . Regula t ion o f P T E N funct ion as a P IP3 gatekeeper through membrane interaction. C e l l C y c l e . 2006;5:1523-1527. 24 65. S u z u k i A , Y a m a g u c h i M T , O h t e k i T , et a l . T cel l -specif ic loss o f P ten leads to defects i n central and peripheral tolerance. Immuni ty . 2001;14:523-534. 66. Seminar io M C , Wange R L . L i p i d phosphatases i n the regulat ion o f T c e l l act ivat ion: l i v i n g up to their P T E N - t i a l . I m m u n o l R e v . 2003;192:80-97. 67. Woodget t JR . Recent advances i n the protein kinase B s ignal ing pathway. Cur r O p i n C e l l B i o l . 2005;17:150-157. 68. B r a v o u V , K l i r o n o m o s G , Papadaki E , Taraviras S, V a r a k i s J . I L K over-express ion i n human c o l o n cancer progression correlates w i t h act ivat ion o f beta-catenin, down-regula t ion o f E-cadher in and act ivat ion o f the A k t - F K H R pathway. J Pathol . 2006;208:91-99. 69. L i u E , S inha S, W i l l i a m s C , et a l . Targeted delet ion o f in tegr in- l inked kinase reveals a role i n T - c e l l chemotaxis and surv iva l . M o l C e l l B i o l . 2005;25:11145-11155. 70. N h o R S , X i a H , K a h m J , K l e i d o n J , D i e b o l d D , Henke C A . R o l e o f in tegr in- l inked kinase i n regulat ing phosphoryla t ion o f A k t and fibroblast su rv iva l i n type I co l lagen matrices through a beta l integrin v i ab i l i t y s ignal ing pathway. J B i o l C h e m . 2005;280:26630-26639. 71. F e n g J , Park J , C r o n P , Hess D , H e m m i n g s B A . Identif ication o f a P K B / A k t hydrophob ic m o t i f Ser-473 kinase as D N A - d e p e n d e n t protein kinase. J B i o l C h e m . 2004;279:41189-41196. E p u b 42004 Ju l 41115. 72. T o k e r A , N e w t o n A C . Ak t /p ro t e in kinase B is regulated by autophosphorylat ion at the hypothet ical P D K - 2 site. J B i o l C h e m . 2000;275:8271-8274. 73. Sarbassov D D , Guer t in D A , A l i S M , Sabatini D M . Phosphory la t ion and regulat ion o f A k t / P K B by the r i c t o r - m T O R complex . Science. 2005;307:1098-1101. 74. Y a n g Q , Inok i K , Ikenoue T , G u a n K L . Identification o f S i n l as an essential T O R C 2 component required for complex format ion and kinase act ivi ty . Genes D e v . 2006;20:2820-2832. 75. Sh io ta C , W o o JT , L i n d n e r J , Shel ton K D , M a g n u s o n M A . M u l t i a l l e l i c d isrupt ion o f the r ic tor gene i n mice reveals that m T O R complex 2 is essential for fetal g rowth and v i ab i l i t y . D e v C e l l . 2006;11:583-589. 76. Fr ias M A , Thoreen C C , Jaffe J D , et a l . m S i n l is necessary for A k t / P K B phosphoryla t ion , and its isoforms define three distinct m T O R C 2 s . C u r r B i o l . 2006; 16:1865-1870. 77. Jacinto E , L o e w i t h R , Schmid t A , et a l . M a m m a l i a n T O R complex 2 controls the actin cytoskeleton and is r apamycin insensit ive. Na t C e l l B i o l . 2004;6:1122-1128. 78. M a c i a n F . N F A T proteins: key regulators o f T - c e l l development and function. N a t R e v I m m u n o l . 2005;5:472-484. 25 79. Anandasabapathy N , F o r d G S , B l o o m D , et a l . G R A I L : an E 3 ub iqu i t in ligase that inhibi ts cy tok ine gene transcript ion is expressed i n anergic C D 4 + T cel ls . Immuni ty . 2003;18:535-547. 80. Heissmeyer V , M a c i a n F , I m S H , et a l . Ca l c ineu r in imposes T ce l l unresponsiveness through targeted proteolysis o f s ignal ing proteins. Na t Immuno l . 2004;5:255-265. 81 . Seroogy C M , Soares L , R a n h e i m E A , et a l . The gene related to anergy i n lymphocytes , an E 3 ubiqui t in l igase, is necessary for anergy induct ion i n C D 4 T cel ls . J I m m u n o l . 2004;173:79-85. 82. Wohl fe r t E A , Ca l l ahan M K , C l a r k R B . Resistance to C D 4 + C D 2 5 + regulatory T cel ls and T G F - b e t a i n C b l - b - / - mice . J I m m u n o l . 2004;173:1059-1065. 83. Sakaguchi S, O n o M , Setoguchi R , et a l . F o x p 3 + C D 2 5 + C D 4 + natural regulatory T cel ls i n dominant self-tolerance and autoimmune disease. I m m u n o l R e v . 2006;212:8-27. 84. M c H u g h R S , Whit ters M J , P i c c i r i l l o C A , et a l . CD4(+)CD25(+) immunoregula tory T cel l s : gene expression analysis reveals a functional role for the g lucocor t ico id- induced T N F receptor. Immuni ty . 2002;16:311-323. 85. A n t o v A , Y a n g L , V i g M , Ba l t imore D , V a n Parijs L . Essent ial role for S T A T 5 s ignal ing i n C D 2 5 + C D 4 + regulatory T ce l l homeostasis and the maintenance o f self-tolerance. J I m m u n o l . 2003;171:3435-3441. 86. Fontenot J D , Rasmussen JP , W i l l i a m s L M , D o o l e y J L , Farr A G , Rudensky A Y . Regula tory T c e l l l ineage specif icat ion by the forkhead transcript ion factor foxp3. Immuni ty . 2005;22:329-341. 87. K l e i n L , K h a z a i e K , v o n Boehmer H . In v i v o dynamics o f antigen-specific regulatory T cel ls not predicted f rom behavior i n vi t ro . Proc N a t l A c a d S c i U S A . 2003;100:8886-8891. 88. K o o n p a e w S, Shen S, F lower s L , Z h a n g W . L A T - m e d i a t e d s ignal ing i n C D 4 + C D 2 5 + regulatory T c e l l development. J E x p M e d . 2006;203:119-129. 89. Bens inger S J , W a l s h P T , Z h a n g J , et a l . D i s t inc t I L - 2 receptor s igna l ing pattern i n C D 4 + C D 2 5 + regulatory T cel ls . J Immunol . .2004;172:5287-5296. 90. W a l s h P T , B u c k l e r J L , Z h a n g J , et a l . P T E N inhibits I L - 2 receptor-mediated expans ion o f C D 4 + C D 2 5 + Tregs. J C l i n Invest. 2006;116:2521-2531. 91 . L i L , Godfrey W R , Porter S B , et a l . C D 4 + C D 2 5 + regulatory T - c e l l l ines f rom human cord b l o o d have functional and molecular properties o f T - c e l l anergy. B l o o d . 2005;106:3068-3073. 92. Tsang J Y , Camara N O , E r e n E , et a l . A l t e r e d p r o x i m a l T ce l l receptor ( T C R ) s ignal ing i n human C D 4 + C D 2 5 + regulatory T cel ls . J L e u k o c B i o l . 2006;80:145-151. 26 93. H i c k m a n S P , Y a n g J , Thomas R M , W e l l s A D , T u r k a L A . Defect ive ac t ivat ion o f protein kinase C and R a s - E R K pathways l imi t s I L - 2 product ion and prol i fera t ion by C D 4 + C D 2 5 + regulatory T cel ls . J I m m u n o l . 2006;177:2186-2194. 94. Sakaguchi S. Na tura l ly ar is ing Foxp3-express ing C D 2 5 + C D 4 + regulatory T cel ls i n i m m u n o l o g i c a l tolerance to se l f and non-self. N a t I m m u n o l . 2005;6:345-352. 95. Shevach E M . Regulatory/suppressor T cel ls i n health and disease. Ar th r i t i s R h e u m . 2004;50:2721-2724. 96. S h i m i z u J , Y a m a z a k i S, Takahashi T , Ishida Y , Sakaguchi S. S t imula t ion o f C D 2 5 ( + ) C D 4 ( + ) regulatory T cel ls through G I T R breaks i m m u n o l o g i c a l self-tolerance. Na t I m m u n o l . 2002;3:135-142. 97. G a v i n M A , C l a r k e S R , N e g r o u E , Ga l l egos A , Rudensky A . Homeostas is and anergy o f C D 4 ( + ) C D 2 5 ( + ) suppressor T cel ls i n v i v o . Na t Immuno l . 2002;3:33-41. 98! Bruder D , Probs t -Kepper M , Wes tendor f A M , et al . N e u r o p i l i n - 1 : a surface marker o f regulatory T cel ls . E u r J I m m u n o l . 2004;34:623-630. 99. K n o e c h e l B , L o h r J , Z h u S, et a l . Func t iona l and molecular compar i son o f anergic and regulatory T lymphocytes . J I m m u n o l . 2006;176:6473-6483. 100. Shevach E M . F r o m v a n i l l a to 28 flavors: mul t ip le varieties o f T regulatory cel ls . Immuni ty . 2006;25:195-201. 101. L i u W , Pu tnam A L , X u - Y u Z , et a l . C D 127 expression inversely correlates w i t h F o x P 3 and suppressive function o f human C D 4 + T reg cel ls . J E x p M e d . 2006;203:1701-1711. 27 2. Ro l e of T L R 5 o n C D 4 + C D 2 5 + / _ T ce l l s 1 2.1 INTRODUCTION G e r m - l i n e encoded pattern recogni t ion receptors ( P R R s ) , such as T o l l - l i k e receptors ( T L R s ) , p rovide a c r i t i ca l l i n k between the innate and adaptive immune systems. T h i s l i n k is general ly be l ieved to i nvo lve antigen-presenting ce l l s ( A P C s ) w h i c h become capable o f s t imulat ing effector T c e l l responses f o l l o w i n g l iga t ion o f one or more P R R s 1 . Interestingly, there is g r o w i n g evidence to suggest that, i n addi t ion to their w e l l characterized indirect effects mediated by A P C s , s t imulat ion v i a P R R s may also direct ly regulate the funct ion o f T cel ls 2 " 4 . T h i s interaction cou ld be meaningful i n the context o f host-pathogen interactions, and i n intestinal homeostasis. Immune responses are t ight ly contro l led by a subset o f T cel ls k n o w n as C D 4 + C D 2 5 + T regulatory (Treg) cel ls , w h i c h have a central role i n establishing and main ta in ing peripheral tolerance and immune homeostasis 5 " 7 . Ev idence for the role o f C D 4 + C D 2 5 + T r e g cel ls i n prevent ing au to immuni ty and al lergy, and promot ing allograft acceptance is c o m p e l l i n g 6 " 8 . In addi t ion, they have a key role i n regulat ing i m m u n e responses to a var iety o f pathogenic and non-pathogenic infectious agents. F o r example, C D 4 + C D 2 5 + T r eg cel ls can prevent the development o f col i t i s induced by the transfer o f C D 4 + effector T cel ls 9 1 0 or by Helicobacter hepaticus " , and even reverse established col i t i s 1 2 . C D 4 + C D 2 5 + T r eg cel ls are also required for the persistence o f several parasites such as L. major 1 3 , M. Tuberculosis, 1 4 , and C. albicans 1 5 . Thus the interactions between C D 4 + C D 2 5 + T r e g cel ls and 1 M a t e r i a l presented i n this chapter has been publ i shed as " H u m a n C D 4 + T cel ls Express T L R 5 and Its L i g a n d F l a g e l l i n Enhances the Suppressive Capaci ty and Expres s ion o f F O X P 3 i n C D 4 + C D 2 5 + T regulatory C e l l s . " Natasha K . C r e l l i n , R o s a V . Ga rc i a , O m e e d Hadisfar , Sarah E . A l l a n , Theodore S. Steiner, and M e g a n K . L e v i n g s . Journal o f Immuno logy , 2005, 175:8051-8059. Copyr igh t 2005 The A m e r i c a n Assoc i a t i on o f Immunologis ts , Inc. 28 microorganisms , and the mechanisms that control tolerance versus i m m u n i t y i n this setting are o f considerable interest. Recent reports suggest that direct s t imulat ion o f C D 4 + C D 2 5 + T r eg cel ls v i a T L R s c o u l d modulate immune regulat ion. F o r example , mice l ack ing T L R 2 have decreased numbers o f peripheral C D 4 + C D 2 5 + T r eg cel ls , indicat ing that this molecu le may influence expans ion and/or maintenance o f suppressive T c e l l populat ions l 6 . Further, murine C D 4 + C D 2 5 + T r e g cel ls express h i g h levels o f m R N A for T L R 1, 2, 4, 5, 7 and 8, and L P S , w h i c h acts through T L R 4 , has been shown to enhance their suppressive capaci ty 4 . F i n a l l y , mice l ack ing M y D 8 8 , an adaptor molecu le cr i t ica l for m u c h o f T L R - m e d i a t e d s ignal ing, exhibi t increased susceptibil i ty to dextran-sulphate induced col i t i s l 7 , consistent w i t h the concept that P R R s may also have ant i- inf lammatory effects i n the intestine. T L R 5 is one o f two k n o w n T L R s w h i c h are activated by proteins 1 8 . Current ly , the on ly characterized l igand for T L R 5 is f lage l l in , a w e l l k n o w n pathogen associated molecular pattern ( P A M P ) protein that autopolymerizes to f o r m the f lagel la o f mot i le bacteria 1 9 . F l a g e l l i n has an important role i n regulat ing virulence o f intestinal bacteria 2 0 ' 2 ! , and is required for the development o f Sa lmonel la - induced col i t i s 2 0 . Interestingly, f l age l l in is also a dominant antigen i n Crohns disease 2 2 , suggesting that this protein m a y direct intestinal immune responses at mul t ip le levels. S ince C D 4 + T ce l l subsets encounter f l age l l in i n the intestinal environment , we investigated whether human C D 4 + C D 2 5 + T r e g and/or C D 4 + C D 2 5 " T cel ls can direct ly respond to this bacterial element. Our results demonstrate that s t imulat ion v i a T L R 5 can direct ly influence the phenotype and funct ion o f both effector and suppressive T cel ls . T h i s 29 capacity o f adaptive immune system cel ls to direct ly respond to P A M P s may represent an important new mechanism by w h i c h pathogens regulate immune responses. 2.2 M A T E R I A L S A N D M E T H O D S Cell purification. Per ipheral b l o o d was obtained f rom healthy volunteers f o l l o w i n g in formed consent and upon approval o f the protocol by the U B C C l i n i c a l Research Eth ics B o a r d . P B M C s were isolated by density centrifugation over F i c o l l (Stemcel l Technologies) , and C D 4 + T cel ls were subsequently pur i f ied by negative selection w i t h magnet ic beads ( M i l t e n y i B i o t e c h or S temcel l Technologies) . A s T L R ligands are k n o w n to potently activate A P C s , w h i c h cou ld be an erroneous source o f cytokines i n some experiments, we r igorous ly ensured the puri ty o f our isolated T ce l l populat ions. F o l l o w i n g pur i f ica t ion, C D 4 + T cel ls were 9 5 - 9 8 % pure and free o f detectable C D 1 4 + monocytes or C D 1 9 + B cel ls (Figure 2.1 A). C D 2 5 + ce l ls were subsequently pur i f ied by posi t ive selection ( M i l t e n y i B io t ech ) , and were passed over 2 M S columns to ensure 9 0 - 9 5 % purity (Figure 2.IB). The f low-through f rom the C D 2 5 + pos i t ive selection was retained and used as C D 2 5 " cel ls . 30 A. C D 4 + T c e l l purity 11 ' 1 10' "072/ o 0.3 21 • #97.7 8 ">'*97:4 CD4 B. Treg cell purity CD4* CD25" T cells CD4* CD25' T cells I 103 u> 10* 8l0< .0 10.0 .k*' 85.7 .1.5 94.0 1.1'' •V: *J 3.4 id' i o 3 i o 3 . . id 1 10s 103 CD 4 -*• Figure 2.1. Purity of GD4 + T cell subsets. (A) CD4+ T cells were purified by negative selection from PBMCs and analysed for expression of CD4, CD14, and CD19. (B) CD4+CD25+ Treg cells were isolated by positive selection and purity was assessed by analysis of CD4 and CD25 expression. Results are representative of more than 20 individual purifications. In some cases, A P C s were prepared i n paral le l by deplet ion o f C D 3 + T cel ls f rom P B M C s us ing magnetic beads (S temcel l Technologies) . M o n o c y t e - d e r i v e d D C s were generated f rom the adherent fraction o f P B M C s by culture i n the presence o f I L - 4 ( lOng /ml ) ( R & D Systems) and G M - C S F (50 ng/ml) ( R & D Systems) as prev ious ly descr ibed 2 3 . C e l l s were either left immature, or matured w i t h L P S (100 ng/ml) for 48 hrs. Pur i ty and maturation were ve r i f i ed by moni to r ing expression o f C D l a , C D 14, C D 8 3 and H L A - D R ( B D Pharmingen) . C D 1 4 + monocytes were pur i f ied by posi t ive selection us ing C D 14 magnetic beads (S temcel l Technologies) . T ceil lines and clones. C D 4 + T cel ls were stained for C D 4 ( B D Pharmingen) and C D 2 5 ( M i l t e n y i B io tech) and F A C S sorted into C D 2 5 h i (top 2-3%) and , o w fractions o n a B D F A C S A r i a . T - c e l l l ines were generated by st imulat ion w i t h a n t i - C D 3 / 2 8 coupled beads (Dyna l ) at a 1:1 ratio o f cel ls to beads i n T ce l l m e d i u m ( X v i v o - 1 5 , 5% human serum [North 31 B i o ] , l x pen ic i l l in / s t rep tomycin [Invitrogen], l x G lu t amax [Invitrogen]) i n the presence o f r h I L - 2 ( l O O U / m l ) (Chi ron) . Al te rna t ive ly , sorted cel ls were c loned by l i m i t i n g d i lu t ion as prev ious ly described 2 4 . C D 4 + C D 2 5 + T - c e l l l ines and clones were moni tored at the end o f every cyc le to ensure preservation o f their suppressive capacity. Quantitative R T - P C R . F o r quantitative R T - P C R , amounts o f T L R 4 , T L R 5 and F O X P 3 m R N A were determined us ing custom T a q m a n probe and pr imer sets ( A p p l i e d Biosys tems) . O l i g o s were as fo l lows : h T L R 4 sense: 5' T C C A T G A A G G T T T C C A T A A A A G C 3' ; h T L R 4 antisense: 5' C C A G C G G C T C T G G A T G A A 3' ; H T L R 5 sense 5' G C C T T G A A G C C T T C A G T T A T G C 3* ; h T L R 5 antisense 5' C C A A C C A C C A C C A T G A T G A G 3' ; F O X P 3 sense 5' T C A C C T A C G C C A C G C T C A T 3' ; h F O X P 3 antisense 5' T C A T T G A G T G T C C G C T G C T T 3'. Probe sequences were as fo l lows : h T L R 4 : 5' A A A G G T G A T T G T T G T G G T G T C C C A G C A 3'; h T L R 5 : 5' C A G G G C A G G T G C T T A T C T G A C C T T A A C A G T G 3'; F O X P 3 : 5' T G G G C C A T C C T G G A G G C T C C A 3*. A m o u n t s o f G A P D H m R N A were determined using A s s a y o n D e m a n d real-t ime P C R kits ( A p p l i e d Biosys tems) . A l l P C R reactions were performed w i t h T a q M a n Mas te r M i x (Qiagen) on an A B I 5700 R e a l T i m e P C R machine . A l l samples were run i n tr iplicate, and relative expression o f F O X P 3 , T L R 5 or T L R 4 was determined by no rma l i z ing to G A P D H i n order to calculate a fo ld -change i n value. Source of Flagel l in and L P S . F l a g e l l i n was pur i f ied as prev ious ly described 1 9 . In brief, recombinant His- tagged f lage l l in was expressed i n B L 2 1 ( D E 3 ) p L y s S bacteria and pur i f ied under native condi t ions by metal affinity chromatography. Contamina t ing L P S was removed 32 by pur i f ica t ion over a p o l y m y x i n B c o l u m n ( D e t o x i - G e l ; Pierce) . The resul t ing f lage l l in was tested i n a l i m u l u s assay (Cambrex) and found to have <0.06 E U o f L P S w h e n tested at a concentrat ion o f 25 fxg/ml. B i o l o g i c a l act ivi ty was ver i f ied by s t imulat ion o f I L - 8 release f rom C a c o cel ls , as p rev ious ly descr ibed 1 9 . T o further rule out possible effects o f contaminants such as L P S or T L R 2 l igands i n the f lage l l in preparation, a non- inf lammatory variant o f the E. coli H I 8 f lage l l in generated by random mutagenesis 1 9 was tested i n paral le l and shown not to effect prol iferat ion of, or cytokine product ion by, C D 4 + C D 2 5 " T cel ls (data not shown). U l t r a pure L P S extracted accord ing to the methods o f Manthey and V o g e l 2 5 was purchased f rom L i s t B i o l o g i c a l Labora tor ies , Inc. T h i s L P S was b i o l o g i c a l l y active and resulted i n potent maturation o f monocyte-der ived immature D C s . In some experiments, a less pure f o r m o f L P S (Sigma) w h i c h is l i k e l y contaminated w i t h other bacter ial ly der ived molecules was also tested. F low cytometry. Express ion o f T L R 5 o n the c e l l surface was detected by incubat ion w i t h a 1/200 d i l u t i on o f an a - T L R 5 A b ( A l e x i s ) f o l l o w e d by addi t ion o f a F ITC-con juga t ed rabbit ct-goat secondary A b (Sigma) on non- f ixed non-permeabi l ized cel ls . To ta l T L R 5 expression ( intracel lular and extracellular) was determined f o l l o w i n g staining w i t h a 1/50 d i lu t ion o f an c t - T L R 5 m A b direct ly conjugated to F I T C (Imgenex I M G - 6 6 3 F ) after the cel ls were fixed w i t h 2 % formaldehyde i n P B S and permeabi l ized by saponin. Exp re s s ion o f F O X P 3 was detected us ing P E anti-human F O X P 3 staining k i t (ebiosciences) accord ing to the manufacturer's instructions. Samples were read o n a B D F A C S C a n t o ™ and analyzed wi th F C S Express V 2 ( D e N o v o software). 33 ELISAs. T o determine amounts o f I L - 2 , capture E L I S A s ( B D Pharmingen) were performed on supernatants after act ivat ion w i t h the indicated concentration o f i m m o b i l i z e d a C D 3 alone or w i t h soluble a C D 2 8 ( lp ,g /ml) , i n the absence or presence o f f lage l l in (10-1000 ng/ml) or L P S (1 ng/ml) for 24h . Western Blotting. T o determine relative amounts o f T L R 5 expression, who le ce l l lysates f rom pur i f ied populat ions o f cel ls were prepared by sonicat ion i n lys is buffer conta in ing 1% S D S , l O m M H E P E S , and 2 m M E D T A ( p H 7.4). A p p r o x i m a t e l y 150 \xg o f protein was loaded per lane, and subjected to S D S - P A G E electrophoresis. F o l l o w i n g transfer to ni t rocel lulose membranes, expression o f T L R 5 was determined by immunob lo t t ing w i t h an a - T L R 5 m A b (Imgenex I M G - 6 6 4 ) used at a 1/500 d i lu t ion , f o l l owed by goat c tmouse -HRP (Dako) . The specif ici ty o f the antibody was conf i rmed w i t h 293T cel ls transfected w i t h a T L R 5 - e n c o d i n g expression vector (data not shown) . Membranes were stripped and re-probed w i t h a - p 3 8 A b s (Santa Cruz ) to assess load ing equivalency. Activation, Proliferation and Suppression of T cells. T o activate C D 4 + C D 2 5 + Tregs or C D 4 + C D 2 5 " T cel ls v i a T C R st imulat ion, cel ls were st imulated w i t h a - C D 3 / 2 8 - c o u p l e d beads ( D y n a l , 1 bead/2.5 cel ls) , or the indicated concentration o f i m m o b i l i z e d a - C D 3 m A b s i n the presence or absence o f soluble a - C D 2 8 . m A b s ( lp ,g /ml) . T o test the prol iferat ive capacity o f C D 4 + C D 2 5 + Tregs, cel ls were plated (8,000 ce l l s /we l l i n 200 JAI) i n 96 w e l l round bot tom plates, and stimulated w i t h a - C D 3 / 2 8 beads (1 bead/6 cel ls) . T o test for suppressive capacity i n the absence o f A P C s , C D 4 + C D 2 5 " T cel ls were plated at 8,000 ce l l s /we l l ( in 200 pl ) i n 96 w e l l round bot tom plates, and st imulated w i t h 34 a - C D 3 / 2 8 coupled beads ( lbead/6ce l l s ) i n T - c e l l m e d i u m (op t imum bead/cel l ratio varies s l ight ly per lot o f beads). C D 4 C D 2 5 T reg cel ls were added i n decreasing amounts, starting at a ratio o f 1:1. Pro l i fe ra t ion was assessed at day 6, after addi t ion o f [ 3 H ] t h y m i d i n e ( l u C i per w e l l , A m e r s h a m ) . T o test for suppressive capacity i n the presence o f A P C s , C D 4 + C D 2 5 " T cel ls (50,000 ce l l s /we l l i n 200 u.1) were st imulated w i t h soluble a n t i - C D 3 m A b s ( O K T 3 , l L i g / m l ) i n the presence o f A P C s (CD3-dep le ted P B M C s , irradiated 5000 Rads, 50,000 ce l l s /wel l ) . C D 4 + C D 2 5 + T r eg cel ls were added i n decreasing amounts, and suppression was assessed after 4 days by determining the amount o f [ 3 H]- thymid ine incorporat ion. F o r a l l prol i ferat ion and suppression experiments, f l a g e l l i n or L P S was added at the in i t i a l t ime o f culture i n amounts indicated i n figure l e g e n d s . Statistical Analysis. A l l analysis for statistically significant differences was performed w i t h the Student's paired t test, p values less than 0.05 were considered signif icant . A l l cultures were performed i n triplicate and error bars represent the S D . 2.3 RESULTS CD4+CD25+ Treg cells express high levels of TLR5 mRNA. W e investigated whether human C D 4 + C D 2 5 + T r eg or C D 4 + C D 2 5 " T cel ls expressed T L R 5 us ing quantitative R T - P C R . A l t h o u g h T L R 5 m R N A was detected i n both populat ions, the amounts i n C D 4 + C D 2 5 + T r eg cel ls were s ignif icant ly higher than i n C D 4 + C D 2 5 " T cells (Figure 2.2A, p=0.0048). Since p o l y c l o n a l populat ions o f human C D 4 + C D 2 5 + ce l ls contain a mixture o f activated effector T cel ls and Treg cel ls 2 4 , we conf i rmed this f ind ing at the c lona l l eve l . S i m i l a r to freshly isolated cel ls , C D 4 + C D 2 5 + T r eg c e l l clones were found to express 35 higher levels o f m R N A for T L R 5 than their non-suppressive counterparts ( F i g u r e 2.2A). W e were also interested to k n o w whether human C D 4 + C D 2 5 + T reg cel ls expressed h i g h levels o f m R N A for T L R 4 , as previous ly described i n the mouse 4 . A s shown i n F i g u r e 2.2B, both ex v i v o isolated C D 4 + C D 2 5 + T reg cel ls , and T c e l l clones expressed more T L R 4 m R N A than their C D 4 + C D 2 5 " counterparts, however i n ex v i v o cel ls this was variable and not statistically significant. 36 A. TLR5 < g 2.5-| ce £ E S 2 H i— "S 1-5 > 0 1 c 0.54 1 H Ex vivo CD25 CD25+ , T cell clones f^r7 Treg B. TLR4 CC c P ° "«t o Ct ~ _1 T3 h- 0) Q) £ > (0 11 CC 2 90 80 70 60 50 40 30 20 10 Ex vivo CD25- CD25+ 70 -, 60 50 40 30 20 10 o T c e l l clones Th Treg CD25- CD25+ CD14 + Imm Mat DC DC Figure 2.2. CD4+CD25+ Treg cells express more TLR5 and TLR4 mRNA in comparison to CD4+CD25 T cells. (A) RNA was isolated from ex vivo purified CD4+CD25+ Treg and CD4+CD25'T cells or suppressive (Treg) and non-suppressive (Th) CD4+CD25+ T cell clones. Expression of TLR5 was determined by quantitative RT-PCR. (B) Levels of TLR4 mRNA were tested in parallel. (C) Levels of TLR5 mRNA in CD4+CD25" T cells and CD4+CD25+ Treg cells were compared to those in ex vivo CD14+ monocytes and DCs in an immature (Imm) or mature (Mat) state. Data are expressed as relative amounts compared to the 37 CD4+CD25" T cells (A and B) or mature DCs (C) and are normalized to amounts of GAPDH. For A&B, each point represents an individual experiment done with different donors or clones, and the horizontal line represents the average. For C, results are representative of 3 experiments with different donors. W e next investigated whether the levels o f T L R 5 m R N A i n C D 4 + T cel ls were comparable to those i n innate immune cel ls k n o w n to be h igh ly T L R 5 responsive. Rela t ive levels o f T L R 5 m R N A i n both C D 4 + C D 2 5 + and C D 4 + C D 2 5 " T cel ls were w i t h i n the range o f those i n C D 1 4 + monocytes , and s l ight ly higher than those i n immature and mature dendrit ic cel ls (Figure 2.2C). Thus , T L R 5 m R N A appears to be expressed i n human C D 4 + T cel ls at phys io log i ca l l y relevant levels . TLR5 protein is expressed on the cell surface of CD4+ T cells at physiologically relevant levels. W e next investigated the relative levels o f T L R 5 expression at the prote in l eve l . A l t h o u g h some T L R s , par t icular ly those that b ind to nucle ic acids, are held i n intracel lular reservoirs 2 6 , it has been reported that T L R 5 must be at the ce l l surface for s ignal ing to occur 2 1 . W e therefore determined the leve l o f T L R 5 surface expression by f l o w cytometry. Interestingly, w e d i d not observe any significant difference i n T L R 5 protein expression between C D 4 + C D 2 5 " effector T cel ls and C D 4 + C D 2 5 + T r eg cel ls (Figure 2 . 3 A ) . T h e average difference between the mean fluorescence intensity ( M F I ) o f the secondary alone control versus cel ls stained w i t h T L R 5 was 4.4 ± 0.3 for C D 4 + C D 2 5 " T cel ls compared to 3.9 ± 0 . 8 for C D 4 + C D 2 5 + T r e g cel ls (p=NS) . The amount o f ce l l surface T L R 5 i n both populat ions o f T cel ls , however , was comparable to that o n monocytes , immature and mature D C s (Figure 2 . 3 A ) . 38 A. Extracellular TLR5 C. CD25- CD25+ CD14+ 10' 10 2 10 3 10' 10 z 10 3 — CD4 • — T L R 5 — • Figure 2.3. CD4+ T cells express physiologically relevant levels of TLR5 protein. (A) TLR5 expression was determined by flow cytometry on CD4+CD25+ Treg or CD4+CD25" T cells, CD14+ monocytes, and immature or 39 LPS-matured DCs. (A) represents levels of cell-surface TLR5, whereas (B) represents total TLR5 expression as determined by intracellular staining. The thin dotted line represents secondary antibody alone controls, whereas the thick line represents a-TLR5 staining. (C) Whole cell lysates from CD4+CD25+ Treg, CD4+CD25" T cells, and CD14+ monocytes were resolved by SDS-PAGE and immunoblotted with cc-TLR5 mAbs. The membrane was reprobed with a-p38 Abs to assess equivalency of loading. (D) CD4+ T cells were stained, gated on FOXP3+ and - populations, and amounts of total TLR5 were compared between the two populations. The black line represents FOXP3+ cells, and the grey line represents FOXP3" cells. Results are representative of 4 individual experiments for A&B and 2 for C&D. Faced w i t h the lack o f correlat ion between cell-surface T L R 5 protein and m R N A levels i n C D 4 + C D 2 5 T reg cel ls and C D 4 X D 2 5 " cel ls , we asked whether intracel lular stores o f T L R 5 prote in might differ between these populations. W e therefore performed intracel lular staining on the same populat ions o f cells us ing an antibody raised against the intracel lular por t ion o f T L R 5 . A s shown i n Figure 2.3B, there was no signif icant difference between the total amount o f T L R 5 i n C D 4 + C D 2 5 + T reg cel ls compared to C D 4 + C D 2 5 " T cel ls . S i m i l a r to our f indings w i t h extracellular staining on ly , the amounts o f total T L R 5 i n T cel ls were comparable to those i n monocytes and D C s . The s imi la r i ty i n the l eve l o f T L R 5 expression between T ce l l subsets was also conf i rmed by western blot t ing, w h i c h further conf i rmed that the levels o f expression were comparable to those i n innate immune cells (Figure 2.3C). It is n o w w e l l accepted that the F O X P 3 transcript ion factor is a more specif ic marker than C D 2 5 for T reg cel ls 2 8 ' 2 9 . W e therefore repeated the intracel lular s ta ining o n total C D 4 + T cel ls gating o n the F O X P 3 posi t ive and negative populations. C D 4 + F O X P 3 + T cel ls and C D 4 + F O X P 3 " T cel ls expressed equal levels o f T L R 5 (Figure 2.3D). S i m i l a r results were obtained when ex v i v o C D 4 + T cel ls were gated o n the C T L A - 4 + or C T L A - 4 " populat ions (data not shown). Thus , despite higher levels o f m R N A , there was no difference i n T L R 5 protein expression between C D 4 + C D 2 5 " T and C D 4 + C D 2 5 + T reg cel ls . W e also attempted to examine the cell-surface expression o f T L R 4 on C D 4 + T ce l l subsets, but found these levels to be very l o w and often undetectable T L R 4 (data not shown). 40 T h i s is consistent w i t h a previous report w h i c h found that T L R 4 was present o n the surface o f C D 4 T cel ls on ly after act ivat ion w i t h a C D 3 m A b s and I F N - a . Expression of TLR5 on CD4+CD25+Treg and CD4+CD25 T cells following activation C D 4 + C D 2 5 + T r eg cel ls const i tut ively express h igh levels o f markers such as C D 2 5 , H L A -D R , G I T R , and C T L A - 4 , w h i c h are also transiently upregulated on activated effector T cel ls 2 4 ' 3 0 . It was therefore important to determine the levels o f T L R 5 f o l l o w i n g T C R - m e d i a t e d act ivat ion, i n order to determine i f T L R 5 is another act ivat ion marker for C D 4 + T cel ls . C D 4 + C D 2 5 + T r eg cel ls or C D 4 + C D 2 5 " T cel ls were activated w i t h c<CD3/CD28 m A b s , and R N A was isolated after 0, 24, or 48 hours. L e v e l s o f T L R 5 m R N A dropped dramat ical ly i n both C D 4 + C D 2 5 + T reg (average decrease 9 5 % ± 6, n=3, p=0.0006) and C D 4 + C D 2 5 _ T cells (average decrease 8 9 % ± 1 0 n=3, p=0.002) f o l l o w i n g act ivat ion (Figure 2.4A). 41 A. Hr post activation O.CD3/28 mAbs 0 15 24 39 48 Hr post activation a£D'3/28 mAbs Figure 2.4. Changes in T L R 5 expression following activation. (A) CD4+CD25+ Treg or CD4+CD25- T cells were activated with ct-CD3/28-coupled beads and RNA was collected after 0,24, and 48 h. Amounts of TLR5 mRNA were determined by quantitative RT-PCR and normalized to GAPDH. Data are expressed as relative amounts compared to the lowest sample. (B) Total CD4+ cells were activated with ct-CD3/28-coupled beads and cell surface or total (intracellular) expression of TLR5 was determined by flow cytometry at the times indicated. Numerical values indicate the change in MFI between secondary only controls and TLR5 staining. Results are representative of 3 individual experiments for A and 4 for B. We also examined whether TCR-mediated activation altered the expression of cell-surface and/or total T L R 5 protein in these T cell subsets. Amounts of total T L R 5 expression 42 transiently increased (between 12 and 24 hours) and then rapidly decl ined by 48-72 hours, w i t h the exact k inet ics depending o n the donor (Figure 2.4B), i n both T c e l l subsets. In contrast, ce l l surface T L R 5 expression gradually dec l ined over the t ime course i n both T ce l l subsets. Together w i t h the decrease i n m R N A expression, these data suggest that f o l l o w i n g T ce l l act ivat ion, the amount o f functional T L R 5 on C D 4 + T cel ls , and thus their capacity to respond to f lage l l in , is decreased. Flagellin fails to reverse hyporesponsiveness of CD4+CD25+ Treg cells, but is a co-stimulatory molecule for CD4+CD25" T effector cells O n e o f the def in ing characteristics o f C D 4 + C D 2 5 + Tregs is their hypo-responsiveness to po lyc lona l s t imul i 2 4 . W e therefore investigated whether co-s t imula t ion w i t h f lage l l in , the on ly k n o w n l igand for T L R 5 , affected the proliferat ive capacity o f C D 4 + T cells . C D 4 + C D 2 5 + T r eg or C D 4 + C D 2 5 " T cel ls were stimulated w i t h a - C D 3 / 2 8 m A b coupled beads, i n the absence or presence o f increasing amounts o f f lage l l in (Figure 2.5A). A d d i t i o n o f f l age l l in over a w ide range o f concentrations fa i led to break the anergy o f C D 4 + C D 2 5 + Treg cel ls . 43 25" •TO C D 20-X E O 15" cr h- 10-X o 5-0 • CD25 • CD25 + con LPS 10 100 1000 Flagellin OCD28 Figure 2.5. Flagellin does not alter CD4+CD25+ Treg cell hyporesponsiveness but it increases CD4+CD25-T cell proliferation and IL-2 production. (A) Ex vivo isolated CD4+CD25+ Treg or CD4+CD25- T cells (8,000 cells/well) were stimulated with a-CD3/28-coupled beads in the absence or presence of flagellin or LPS (1 f.tg/ml) and amounts of 3H-thymidine incorporation were measured after 6 days. (B) CD4+CD25" T cells (100,000 cells/well) were stimulated with immobilized aCD3 mAbs alone or in combination with aCD28 mAbs (1 ug/ml), in the absence or presence of flagellin or LPS (1 ug/ml), and amounts of 3H-thymidine 44 incorporation were measured after 3 days. (C) Supernatants were collected in parallel after 24 hours, and IL-2 was measured by ELISA. Note that the scale is different between the left and right panels. A single representative experiment of at least 3 performed is depicted. In contrast, f lage l l in s ignif icant ly enhanced the prol i ferat ion o f C D 4 + C D 2 5 " T cel ls i n response to s t imulat ion w i t h a - C D 3 m A b s , i n a dose-dependent manner (Figure 2.5B). M a x i m a l co-s t imula t ion occurred at 100 n g / m l o f f l age l l in w i t h c t - C D 3 m A b s ( 1 . 7 9 ± 0 . 2 2 fo ld increase p=0.026, n=3 compared to a - C D 3 [ lp,g/ml] alone), and prol i ferat ion reached levels that were equal to those i n the presence o f c i - C D 2 8 m A b s (compare to control i n right panel). M o r e o v e r , addi t ion o f f l age l l in further enhanced prol i ferat ion i n response to a - C D 3 m A b s i n the presence o f a - C D 2 8 m A b s ( 1 . 5 2 ± 0 . 0 9 fo ld increase p=0.04, n=3 as compared to c c - C D 3 / 2 8 alone). T h i s strong pro-prol iferat ive effect o f f l age l l in suggested that, l ike T L R 2 2 ' 1 6 , T L R 5 may act as a co-s t imulatory molecule . W e also compared the b io log i ca l effects o f f lage l l in s t imulat ion to those o f h igh ly pur i f ied L P S . W e found that L P S had no effect o n the prol i ferat ion o f C D 4 + C D 2 5 + T reg cel ls , and tended to inhibi t the prol i ferat ion o f C D 4 + C D 2 5 " T cel ls (Figure 2.5A and B). W h e n less pure L P S , that was potential ly contaminated w i t h other T L R l igands, was used, C D 4 + C D 2 5 " T cel ls d isp layed an increased prol i fera t ion on ly i n the presence o f h igh concentrations (10ug/ml) . H o w e v e r even this less-pure preparation o f L P S d i d not affect the prol i fera t ion o f C D 4 + C D 2 5 + T r e g cel ls (data not shown). W e next investigated the effects o f f l age l l in on I L - 2 product ion. C D 4 + C D 2 5 + T reg or C D 4 + C D 2 5 " T cel ls were st imulated w i t h a - C D 3 m A b s i n the presence or absence o f or a - C D 2 8 m A b s , alone, or w i t h increasing concentrations o f f l age l l in or L P S . Af te r 24h the amounts o f I L - 2 i n supernatants were determined (Figure 2.5.C). Consistent w i t h its failure to break anergy, f l age l l in had no effect on the inabi l i ty o f C D 4 + C D 2 5 + T r eg cel ls to produce I L - 2 (data not shown). Howeve r , co-s t imulat ion w i t h f lage l l in s ignif icant ly enhanced 45 product ion o f I L - 2 by C D 4 C D 2 5 " T cel ls i n a dose dependent manner. U p o n s t imulat ion w i t h a - C D 3 , addi t ion o f f lagel l in (100 ng/ml) increased I L - 2 product ion an average o f 2.2 ± 0.6 fo ld (n=4, p=0.027). In combinat ion w i t h c t - C D 2 8 co-s t imulat ion, addi t ion o f f lage l l in further increased I L - 2 product ion an average o f 2.1 ± 0.2 fo ld (n=4, p=0.02). The drop i n I L -2 detectable w i t h 1000 n g / m l f l age l l in i s l i k e l y due to ce l lu lar consumpt ion . Therefore T L R 5 , but not T L R 4 , acted as a cost imulatory molecule and enhanced I L - 2 produc t ion and prol i fera t ion i n effector T cells . Flagellin enhances the suppressive capacity of CD4+CD25+ Treg cells The most important functional characteristic o f Treg cel ls is their abi l i ty to suppress effector T ce l l responses. It has previous ly been shown that T L R 4 - s t i m u l a t e d D C produce cytokines w h i c h act o n effector T cel ls and make them resistant to the suppressive effects o f C D 4 + C D 2 5 + T r eg cel ls 3 1 ' 3 2 . Howeve r , i n these studies it was di f f icul t to determine whether T L R - l i g a n d s may. also direct ly act o n the C D 4 + C D 2 5 + T r eg cel ls themselves. W e therefore developed a modi f i ed vers ion o f the suppression assay first reported by V i g l i e t t a et a l . to rule out any indirect A P C - i n d u c e d effects and investigate whether f l age l l in altered the functional capacity o f C D 4 + C D 2 5 + T r e g cel ls 3 3 . A c c o r d i n g l y , l o w numbers o f C D 4 + C D 2 5 " effector T cel ls (8,000 ce l l s /wel l ) were stimulated w i t h a C D 3 / 2 8 - m A b coupled beads, i n the absence o f A P C s , for 6 days. A s expected, addi t ion o f increasing numbers o f C D 4 + C D 2 5 + T r e g cells suppressed prol i ferat ion, and significant suppression cont inued to be observed even at a ratio o f 1:81 (Treg:Teffector) (Figure 2.6A). A t a ratio o f 1:81, addi t ion o f f l age l l in increased suppression by an average o f 27 ± 2 . 5 % (n=3, p=0.0005) (Figure 2.6B). Importantly, i n the presence o f f lage l l in , significant suppression cont inued to be observed even at a 1:243 ratio, 46 w h e n it was no longer detectable i n controls. A s expected, addi t ion o f exogenous I L - 2 reversed suppression, even i n the presence o f f l age l l in (data not shown). Figure 2.6. Effects of Flagellin on the suppressive capacity of CD4+CD25+ Treg cells in the presence or absence of APCs. (A&B) CD4+CD25" T cells were stimulated with anti-CD3/28-coupled beads in the presence of increasing numbers of CD4+CD25+Treg cells, and with or without flagellin (10 ng/ml) or LPS (10 ng/ml, similar results obtained with 1 ug/ml). 3H-thymidine incorporation was measured after 6 days. (A) depicts one representative experiment and (B) represents the average increase in suppression in the presence of flagellin or LPS at a ratio of 1:81 (Treg:Teffector), in 3 independent experiments. (C&D) CD4+CD25" T cells were stimulated with APCs (autologous, irradiated CD3-depleted PBMCs) and soluble a-CD3 mAbs (1 ug/ml) in the presence of increasing numbers of CD4+CD25+ Treg cells, and with or without flagellin (100 ng/ml) or LPS (1 Ug/ml). 3H-thymidine incorporation was measured after 4 days. (C) depicts one representative experiment, and (D) represents the average decrease in percent suppression at a 1:2 ratio (Treg:Teffector) from 3 independent experiments. A d d i t i o n o f L P S to these cultures d i d not affect the outcome, and i n on ly 1 o f 3 experiments caused a slight enhancement i n the suppression by C D 4 + C D 2 5 + T reg cells 47 (Figure 2.6B). Moreove r , this appeared to be attributable to the anti-proliferative effects o f L P S o n the C D 4 + C D 2 5 ~ T cel ls as documented i n Figure 2.5B. W e also investigated whether s t imula t ion o f T L R 5 i n the presence o f A P C s altered the suppressive capacity o f C D 4 C D 2 5 T r e g cel ls . C D 4 C D 2 5 " T cel ls were st imulated w i t h irradiated autologous A P C s and soluble a n t i - C D 3 m A b s (Figure 2.6C), i n the absence or presence o f C D 4 + C D 2 5 + T r eg cel ls . A s expected, prol i fera t ion o f effector T cel ls was suppressed by 70 -80% at a 1:2 ratio o f Tregs:Teffectors. H o w e v e r , addi t ion o f f lagel l in resulted i n a consistent decrease i n the capaci ty o f C D 4 + C D 2 5 + T r eg cel ls to suppress prol i fera t ion (Figure 2.6C). A t a 1:2 ratio, f lage l l in decreased the percent suppression by an average o f 15- ± 4 . 7 % (n=3, p=0.015) (Figure 2.6D). In contrast the effects o f L P S on suppression were variable and not significant. T h i s latter result is i n contrast to the f indings o f Pasare et a l 3 I , but consistent w i t h Peng et a l 3 4 , and poss ibly due to the fact that the A P C s i n our experiments were monocytes rather than fu l ly differentiated D C s . Flagellin enhances expression of FOXP3 in activated CD4+CD25+ Treg cells H u m a n C D 4 + C D 2 5 + T r e g cel ls express h igh levels o f F O X P 3 3 5 , and this t ranscript ion factor is essential for their normal development i n mice ' ' . I n an attempt to determine the molecula r basis for the enhanced suppressive capacity o f C D 4 + C D 2 5 + T r eg cel ls upon s t imulat ion o f T L R 5 , we investigated whether flagellin altered the expression o f F O X P 3 . C D 4 + C D 2 5 + T r eg cel ls were activated w i t h a C D 3 / C D 2 8 m A b s , wi thout or w i t h f lagel l in or L P S for 48 hours (Figure 2.7). A s expected, C D 4 + C D 2 5 + T r eg cel ls expressed ~100- fo ld more F O X P 3 than C D 4 + C D 2 5 " T cel ls 3 5 . F o l l o w i n g act ivat ion, expression o f F O X P 3 m R N A i n C D 2 5 + C D 4 + T r e g cel ls was consistently decreased by ~4 fo ld . In contrast, when C D 4 + C D 2 5 + Tregs cel ls were activated i n the presence o f flagellin, levels o f F O X P 3 m R N A 48 did not decrease, but rather increased by 6 ± 2.6 fold (n=5, p=0.006) as compared to the 48-hr activated control (Figure 2.7B). Flagellin did not alter the minimal activation-induced expression of FOXP3 in CD4+CD25" T cells (Figure 2.7A). Addition of LPS had no significant effect on levels of FOXP3 mRNA in either CD4 + CD25 + Tregs cells or CD4+CD25" T cells. These data indicate that flagellin may enhance the suppressive capacity of CD4 + CD25 + Treg cells by influencing expression of this key transcription factor. A 30Ch • CD25 • CD25 Ct. c Flagellin ng/ml aCD3/28 activation B 10, : - 6 ^ 2 FL Figure 2.7. Flagellin increases FoxP3 expression in CD4+CD25+ Treg cells following TCR-mediated activation. Ex vivo isolated CD4+CD25+ Treg or CD4+CD25' T cells were stimulated with a-CD3/28-coupled beads, in the absence of presence of flagellin or LPS (10 u.g/ml) for 48 hrs. RNA was collected and quantitative RT-PCR was performed to determine expression of FOXP3. Data are expressed as relative amounts compared 49 to ex vivo CD4+CD25" T cells and are normalized to amounts of GAPDH. (A) depicts a single representative experiment, and for (B) each symbol represents the fold increase in FOXP3 mRNA in the presence of flagellin (100 ng/ml) compared to the 48hr-activated control in individual experiments. The horizontal line represents the average. • 2.4 D I S C U S S I O N W e have shown that human C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T cel ls express T L R 5 , and that f l age l l in can direct ly influence their function. Importantly, levels o f T L R 5 m R N A and protein on T cel ls were s imi lar to those observed i n monocytes and D C s , ind ica t ing a phys io log i ca l l y relevant l eve l o f expression. A l t h o u g h f lage l l in d i d not overcome the anergy o f C D 4 + C D 2 5 + T r eg cel ls , it acted as a cost imulatory molecule on effector T cel ls . Remarkab ly , f l age l l in potently increased the suppressive capacity o f C D 4 + C D 2 5 + T r eg cel ls , even under condi t ions i n w h i c h prol i ferat ion o f the target C D 4 + C D 2 5 " T cel ls was enhanced. T h e f ind ing that f l age l l in can induce a significant increase i n F O X P 3 expression i n C D 4 + C D 2 5 + T r eg cells suggests that, i n addi t ion to indirect effects mediated by innate immune cel ls , T L R ligands may also have the capacity to direct ly regulate adaptive immune responses. Surpr i s ing ly , al though w e consistently found that levels o f T L R 5 m R N A were s ignif icant ly higher i n ex v i v o pur i f ied C D 4 + C D 2 5 + T reg cel ls , T ce l l l ines and T ce l l clones than i n their C D 4 + C D 2 5 " counterparts, there was no corresponding increase i n protein expression. T h e increased expression o f T L R 4 and 5 m R N A w e observed i n C D 4 + C D 2 5 + Treg cel ls is consistent w i t h previous studies w h i c h d i d not go o n to investigate expression at the prote in leve l 4 . O u r data highl ight the importance o f examin ing both m R N A and protein expression whenever possible. V e r y l i t t le is k n o w n about the correlat ion between T L R 5 m R N A and prote in expression, and it is not clear i f this discordance is a c o m m o n 50 phenomenon. There was also a dispari ty between m R N A and protein expression f o l l o w i n g T C R - m e d i a t e d act ivat ion when there was a transient increase i n total T L R 5 protein expression despite a concurrent dramatic decrease i n T L R 5 m R N A . It is possible that pre-exis t ing T L R 5 m R N A may be subjected to post-transcriptional regulat ion f o l l o w i n g T ce l l ac t ivat ion p r o v i d i n g a template for this transient increase i n protein expression. Al te rna t ive ly , there may be a s imi la r transient increase i n T L R 5 m R N A expression at a t ime point earlier than 24 hr. The overa l l conc lus ion is that f o l l o w i n g act ivat ion, T cel ls l i k e l y rap id ly lose their capacity to respond to f lagel l in , poss ibly as part o f a negative feedback mechanism. Co- s t imu la t ion w i t h f lage l l in , or. L P S , d id not alter the anergic state o f human C D 4 + C D 2 5 + T r e g cel ls . These data contrast w i t h a previous report w h i c h showed that L P S can di rect ly increase the prol i ferat ion o f murine C D 4 + C D 2 5 + T r eg cel ls 4 . T h i s discrepancy may be the result o f inherent differences between mouse and human cel ls , a l though both expressed h i g h levels o f T L R 4 m R N A , o r due to the use o f different sources o f L P S . Other groups have also fai led to f ind a pro-prol iferat ive effect o f L P S on mur ine C D 4 + C D 2 5 + Treg cel ls 2 ' 3 . T L R 5 is k n o w n to fo rm homodimers , but can also heterodimerize w i t h T L R 4 3 8 . Therefore, it is possible that T L R 4 o n C D 4 + T cel ls functions p r imar i l y through the format ion o f heterodimers w i t h T L R 5 . In contrast, f lage l l in direct ly enhanced prol i ferat ion and I L - 2 product ion by C D 4 + C D 2 5 " T cel ls , p rov id ing a co-st imulatory st imulus as strong as that o f C D 2 8 . N o t e that L P S had no effect on the prol i ferat ion o f C D 4 + C D 2 5 " T cel ls , strongly support ing the conc lus ion that f l age l l in direct ly affects C D 4 + T ce l ls , rather than acting indi rec t ly v i a smal l numbers o f contaminat ing cel ls . These f indings are consistent w i t h a recent report that f l age l l in has a st imulatory effect o n m e m o r y C D 4 + T cel ls 3 9 . Together w i t h evidence that 51 l ipopeptide, a T L R 2 l igand, results i n a s imi la r e f f e c t 2 ' 1 6 , these data indicate that T L R s may represent a new class o f co-st imulatory molecules . Importantly, i n the absence o f A P C s , f l age l l in s ignif icant ly increased the capacity o f C D 4 + C D 2 5 + T r eg cel ls to b lock effector T - c e l l prol i ferat ion, and suppression cont inued to be observed even at a - 1 : 2 5 0 (Treg:Teffector) ratio. T h i s effect d i d not appear to be related to alterations i n Treg-cel l-associated surface molecules (data not shown), but d i d correlate w i t h enhanced expression o f F O X P 3 . F l a g e l l i n not on ly prevented act ivat ion- induced d o w n -regulat ion o f F O X P 3 , but also enhanced its expression compared to resting levels . Since f l age l l in d i d not increase the prol iferat ive capacity o f C D 4 + C D 2 5 + T r eg cel ls , it is un l ike ly this increased F O X P 3 was due to a preferential increase i n their numbers relative to contaminat ing effector T cel ls 2 4 . M o r e o v e r , since f l age l l in actually enhanced I L - 2 produc t ion i n effector T cel ls , it is also un l ike ly that T L R 5 s t imula t ion drove the de novo differentiat ion o f C D 4 + C D 2 5 + T r eg cel ls f rom contaminat ing non-Treg cel ls . Thus , we favour the hypothesis that signals downstream o f T L R 5 can interfere w i t h the apparently negative effects o f T C R st imulat ion, to enhance F O X P 3 expression. Considerable evidence is accumula t ing w h i c h suggests that P A M P s regulate C D 4 + C D 2 5 + T r eg ce l l numbers and funct ion i n health and disease. F o r example , mice deficient i n T L R 2 have decreased numbers o f C D 4 + C D 2 5 + T r eg cel ls , and T L R 2 ligands have a protective effect o n their capaci ty for surv iva l i n vi t ro 1 6 . T L R 2 - d e f i c i e n t mice are resistant to infect ion w i t h C. albicans, suggesting that the .normal interaction between T L R 2 and T r e g cel ls promotes immunosuppress ion 1 6 . Further, l iga t ion o f T L R 4 o n murine C D 4 + C D 2 5 + T r eg cel ls enhances their suppressive capacity i n some situations 4 . In contrast, it has recently been reported that T L R 8 l igands act direct ly o n C D 4 + C D 2 5 + T r eg cel ls and 52 reverse their suppressive capacity 3 4 . Thus it appears l i ke ly that different T L R l igands have differential effects on the funct ion o f T reg cel ls . In support o f a role o f T L R s i n regulat ing intestinal in f lammat ion , m i c e deficient i n M y D 8 8 , a c r i t i ca l s ignal ing molecu le downstream o f most T L R s , are more susceptible to dextran-sulfate col i t i s l 7 . In addi t ion, mutations i n the N O D 2 P R R are associated w i t h susceptibi l i ty to Crohn 's disease. These f indings support the hypothesis that the lack o f an appropriate P R R signal can lead to loss o f r e g u l a t i o n 4 0 ' 4 1 . W e therefore propose the f o l l o w i n g mode l (Figure 2.8). Firs t , i n the context o f the early inf lammatory response to invad ing micro-organisms, A P C s w o u l d become activated by P R R s and del iver a st imulatory s ignal to C D 4 + T cel ls w h i c h is further enhanced by direct interactions between C D 4 + T cells and P A M P s . Together, this process w o u l d result i n cy tokine release, strong C D 4 + T c e l l responses w h i c h are resistant to suppression by C D 4 + C D 2 5 + T r eg ce l l s 3 1 ' 3 2 , and a pro- inf lammatory an t i -microb ia l response. Concurrent ly , the same P A M P s w o u l d act direct ly o n the C D 4 + C D 2 5 + T reg cel ls to preserve F O X P 3 expression and their suppressive capacity. W i t h t ime, or i n the presence o f non- invas ive commensa l bacteria, the cytokine m i l i e u w o u l d shift towards a tolerogenic environment, where C D 4 + effector T cel ls are more easi ly suppressed by C D 4 + C D 2 5 + T r eg cells . M o r e o v e r , s ince C D 4 + C D 2 5 + T r eg cel ls can d i rec t ly suppress effector T cel ls 4 2 , it is possible that recently activated C D 4 + C D 2 5 + T r eg cel ls c o u l d interact w i t h and.suppress effector T cel ls i n the absence o f T L R - m a t u r e d A P C s . 53 Early Inflammation -PAMPs > f CD4+effector T cells, not susceptible to suppression t F O X P 3 Late/chronic Inflammation immature DC/M^ suppression of inflammation CD4+ effector T cells susceptible to suppression preserved numbers and function of CD4+CD25+ Tregs Figure 2.8. Model of how flagellin may interact with C D 4 + C D 2 5 + Treg cells during an immune response. We propose a model where during an immune response, CD4+CD25+ Treg cells would be attracted to sites of inflammation and innate stimuli would preserve and possibly enhance their local numbers and/or function. However, early during this response, production of counter-regulatory cytokines, expression of co-stimulatory molecules, and strong activation of effector T (Te) cells would make the response "unsuppressable". With time, and reduction in inflammation, the effector cells would become re-susceptible to suppression, and the Treg cells would be ready to dampen the response. The capacity o f P R R s to direct ly enhance the numbers and funct ion o f C D 4 + C D 2 5 + Treg cel ls ensures a mechanism to reduce the potent ial ly harmful effects o f uncontrol led inf lammat ion to the host. Further evidence to support this mode l comes f rom the f inding that C D 4 + C D 2 5 + T r eg cel ls exert their suppressive effects at the end o f an immune response 5 and 54 that these cel ls are w e l l equipped w i t h specific h o m i n g receptors to t ravel to sites o f in f l ammat ion 4 3 . I f our hypothesis is correct, the capacity o f P R R s to direct ly influence Treg c e l l responses w o u l d represent a new point o f integration between the innate and the adaptive immune systems. 55 2.5 REFERENCES 1. Iwasaki A , M e d z h i t o v R . T o l l - l i k e receptor control o f the adaptive i m m u n e responses. N a t I m m u n o l . 2004;5:987-995. 2. K o m a i - K o m a M , Jones L , O g g G S , X u D , L i e w F Y . T L R 2 is expressed o n activated T cel ls as a cost imulatory receptor. P roc N a t l A c a d S c i U S A . 2004;101:3029-3034. 3. G e l m a n A E , Z h a n g J , C h o i Y , T u r k a L A . T o l l - l i k e receptor l igands direct ly promote activated C D 4 + T ce l l surv iva l . J I m m u n o l . 2004;172:6065-6073. 4. Caramalho I, L o p e s - C a r v a l h o T , Ost ler D , Ze lenay S, H a u r y M , Demengeot J . Regula tory T cel ls selectively express to l l - l i ke receptors and are activated by l ipopolysacchar ide . J E x p M e d . 2003;197:403-411. 5. P o w r i e F . Immune Regula t ion i n the Intestine: A Ba lanc ing A c t between Effector and Regula tory T C e l l Responses. A n n N Y A c a d S c i . 2004;1029:132-141. 6. Shevach E M . Regulatory T cel ls i n au to immmuni ty . A n n u R e v I m m u n o l . 2000;18:423-449. 7. Sakaguchi S. Na tu ra l ly ar is ing C D 4 + regulatory T cel ls for i m m u n o l o g i c self-tolerance and negative control o f immune responses. A n n R e v I m m u n o l . 2004;22:17.11-17.32. 8. W a l d m a n n H , C o b b o l d S. Regula t ing the Immune Response to Transplants: A R o l e for C D 4 + Regula tory C e l l s ? Immuni ty . 2001;14:399-406. 9. M o r r i s s e y P J , Charr ier K , B r a d d y S, L i g g i t t D , Watson J D . C D 4 + T cel ls that express h igh levels o f C D 4 5 R B induce wast ing disease when transferred into congenic severe combined immunodef ic ien t mice . Disease development is prevented by cotransfer o f purif ied C D 4 + T cel ls . J E x p M e d . 1993;178:237-244. 10. P o w r i e F , L e a c h M W , M a u z e S, Cadd ie L B , Cof fman R L . Pheno typ ica l ly distinct subsets o f C D 4 + T cel ls induce or protect f rom chronic intestinal in f l ammat ion i n C . B - 1 7 sc id mice . Int I m m u n o l . 1993;5:1461-1471. 11. C a h i l l R J , F o l t z C J , F o x J G , Dang le r C A , P o w r i e F , Schauer D B . Inf lammatory bowe l disease: an immuni ty-media ted condi t ion triggered by bacterial infec t ion w i t h Hel icobacter hepaticus. Infect I m m u n . 1997;65:3126-3131. 12. Mot te t C , U h l i g H H , P o w r i e F . Cu t t ing edge: cure o f col i t i s by C D 4 + C D 2 5 + regulatory T cel ls . J I m m u n o l . 2003;170:3939-3943. 13. B e l k a i d Y , P i c c i r i l l o C A , M e n d e z S, Shevach E M , Sacks D L . C D 4 + C D 2 5 + regulatory T cel ls control L e i s h m a n i a major persistence and immuni ty . Nature . 2002;420:502-507. 56 14. Z u a n y - A m o r i m C , S a w i c k a E , M a n l i u s C , et a l . Suppression o f a i rway eos inophi l i a by k i l l e d M y c o b a c t e r i u m vaccae- induced allergen-specific regulatory T-ce l l s . N a t M e d . 2002;8:625-629. 15. B e l k a i d Y , Rouse B T . Na tura l regulatory T cel ls i n infectious disease. N a t I m m u n o l . 2005;6:353-360. 16. Ne tea M G , Sutmul ler R , He rmann C , et a l . T o l l - l i k e receptor 2 suppresses immun i ty against C a n d i d a albicans through induc t ion o f I L - 1 0 and regulatory T cel ls . J I m m u n o l . 2004;172:3712-3718. 17. A r a k i A , K a n a i T, Ishikura T , et a l . M y D 8 8 - d e f i c i e n t mice develop severe intestinal in f l ammat ion i n dextran sod ium sulfate col i t i s . J Gastroenterol. 2005;40:16-23. 18. H a y a s h i F , S m i t h K D , O z i n s k y A , et a l . T h e innate immune response to bacterial f l age l l in is mediated by T o l l - l i k e receptor 5. Nature. 2001;410:1099-1103. 19. D o n n e l l y M A , Steiner T S . T w o nonadjacent regions i n enteroaggregative Esche r i ch i a c o l i f l age l l in are required for act ivat ion o f to l l - l i ke receptor 5. J B i o l C h e m . 2002;277:40456-40461 . 20. Stecher B , Hapfe lmeier S, M u l l e r C , K r e m e r M , S ta l lmach T , Hardt W D . F l a g e l l a and chemotaxis are required for efficient induc t ion o f Sa lmone l l a enterica serovar T y p h i m u r i u m col i t i s i n streptomycin-pretreated mice . Infect Immun . 2004;72:4138-4150. 21 . K h a n M A , K a n g J , Steiner T S . Enteroaggregative Esche r i ch ia c o l i f lagel l in- induced inter leukin-8 secretion requires T o l l - l i k e receptor 5-dependent p38 M A P kinase act ivat ion. Immuno logy . 2004;112:651-660. 22. L o d e s M J , C o n g Y , E l s o n C O , et a l . Bac te r ia l f lagel l in is a dominant antigen i n C r o h n disease. J C l i n Invest. 2004;113:1296-1306. 23.. L e v i n g s M K , G r e g o r i S, T reso ld i E , Cazzan iga S, B o n i n i C , Ronca ro lo M G . Different ia t ion o f T r l cel ls by immature dendri t ic cel ls requires I L - 1 0 but not C D 2 5 + C D 4 + T r cel ls . B l o o d . 2005;105:1162-1169. 24. L e v i n g s M K , Sangregorio R , Sartirana C , et a l . H u m a n C D 2 5 + C D 4 + T suppressor ce l l clones produce t ransforming growth factor beta, but not inter leukin 10, and are distinct f rom type 1 T regulatory cells . J E x p M e d . 2002;196:1335-1346. 25. Man they C L , Perera P Y , Hen r i c son B E , H a m i l t o n T A , Qureshi N , V o g e l S N . Endo tox in - induced early gene expression i n C 3 H / H e J (Lpsd) macrophages. J I m m u n o l . 1994;153:2653-2663. 26. Matsumoto M , F u n a m i K , Tanabe M , et a l . Subcel lu lar loca l i za t ion o f T o l l - l i k e receptor 3 i n human dendrit ic cel ls . J I m m u n o l . 2003;171:3154-3162. 57 27. Wes t A P , Dancho B A , M i z e l S B . Gangl ios ides inhibi t f lage l l in s ignal ing i n the absence o f an effect o n f lage l l in b ind ing to to l l - l i ke receptor 5. J B i o l C h e m . 2005. 28. H o r i S, N o m u r a T , Sakaguchi S. C o n t r o l o f regulatory T ce l l development by the t ranscript ion factor F o x p 3 . Science. 2003;299:1057-1061. 29. Fontenot J D , G a v i n M A , Rudensky A Y . Foxp3 programs the development and funct ion o f C D 4 + C D 2 5 + regulatory T cel ls . Na t I m m u n o l . 2003;4:330-336. 30. L e v i n g s M K , Sangregorio R , R o n c a r o l o ' M G . H u m a n cd25(+)cd4(+) t regulatory cel ls suppress naive and memory T ce l l prol i ferat ion and can be expanded i n v i t ro wi thout loss b f function. J E x p M e d . 2001;193:1295-1302. 31. Pasare C , M e d z h i t o v R . T o l l pathway-dependent blockade o f C D 4 + C D 2 5 + T c e l l -mediated suppression by dendri t ic cel ls . Science. 2003;299:1033-1036. E p u b 2003 Jan 1016. .32. Pasare C , M e d z h i t o v R . Tol l -dependent control mechanisms o f C D 4 T c e l l act ivat ion. Immuni ty . 2004;21:733-741. 33. V i g l i e t t a V , B a e c h e r - A l l a n C , W e i n e r H L , Haf le r D A . L o s s o f functional suppression by C D 4 + C D 2 5 + regulatory T cel ls i n patients w i t h mul t ip le sclerosis. J E x p M e d . 2004;199:971-979. 34. P e n g G , G u o Z , K i n i w a Y , et a l . T o l l - l i k e receptor 8-mediated reversal o f C D 4 + regulatory T c e l l function. Science. 2005;309:1380-1384. 35. W a l k e r M R , K a s p r o w i c z D J , Gersuk V H , et a l . Induction o f F o x P 3 and acquis i t ion o f T regulatory act ivi ty by stimulated human C D 4 + C D 2 5 - T cel ls . J C l i n Invest. 2003;112:1437-1443. 36. Sakaguchi S. The o r ig in o f F O X P 3 - e x p r e s s i n g C D 4 + regulatory T cel ls : thymus or periphery. J C l i n Invest. 2003;112:1310-1312. 37. Kha t t r i R , C o x T, Y a s a y k o S A , R a m s d e l l F . A n essential role for Scur f in i n C D 4 + C D 2 5 + T regulatory cel ls . Na t I m m u n o l . 2003;4:337-342. 38. M i z e l S B , H o n k o A N , M o o r s M A , S m i t h P S , West A P . Induct ion o f macrophage ni t r ic ox ide product ion by Gram-negat ive f l age l l in involves s ignal ing v i a heteromeric T o l l -l ike receptor 5 /To l l - l i ke receptor 4 complexes . J I m m u n o l . 2003;170:6217-6223. 39. C a r o n G , D u l u c D , F remaux I, et a l . D i rec t s t imulat ion o f human T cel ls v i a T L R 5 and T L R 7 / 8 : f l age l l in and R-848 up-regulate prol i ferat ion and I F N - g a m m a product ion by memory C D 4 + T cel ls . J I m m u n o l . 2005;175:1551-1557. 40. O g u r a Y , B o n e n D K , Inohara N , et a l . A frameshift mutat ion i n N O D 2 associated w i t h suscept ibi l i ty to Crohn 's disease. Nature. 2001 ;411:603-606. 58 41 . H u g o t JP , C h a m a i l l a r d M , Z o u a l i H , et a l . Assoc i a t i on o f N O D 2 leuc ine- r ich repeat variants w i t h susceptibi l i ty to Crohn 's disease. Nature . 2001 ;411:599-603. 42. P i c c i r i l l o C A , Shevach E M . Cu t t ing edge: control o f C D 8 + T ce l l ac t ivat ion by C D 4 + C D 2 5 + immunoregula tory cel ls . J I m m u n o l . 2001;167:1137-1140. 43. H u e h n J , S i egmund K , L e h m a n n J C , et a l . Deve lopmenta l stage, phenotype, and migra t ion d is t inguish naive- and effector /memory-l ike C D 4 + regulatory T cel ls . J E x p M e d . 2004;199:303-313; 59 3. Intracellular S igna l ing in C D 4 + C D 2 5 + T r e g s 2 3.1 I N T R O D U C T I O N A c t i v e suppression o f immune responses by T regulatory cel ls (Tregs) is a key mechan i sm for induc t ion and maintenance o f peripheral tolerance I _ 4 . T h e importance o f Tregs in vivo has been demonstrated i n several mouse models : their absence results i n systemic autoimmune disease, wh i l e their presence can inhibi t anti-tumor, anti-allergen, ant i -vi ra l and anti-parasite immun i ty 2 " 5 . K n o w l e d g e o f exact ly h o w Tregs arise, the precise mechanisms that cont ro l their suppressive function, and o f h o w they differ f rom effector T cel ls at the molecu la r l eve l remains largely unknown . A better understanding o f the basic b io log ica l characteristics o f Tregs w i l l lead to nove l therapies for diseases resul t ing f rom immune dysregulat ion. A l t h o u g h there is evidence that T cel ls w i t h a regulatory/suppressor funct ion exist w i t h i n a l l major subsets, most research has been focused on those that are C D 4 + and consti tut ively express h i g h levels o f the I L - 2 R a ( C D 2 5 ) . These C D 4 + C D 2 5 + Tregs have more recently been further defined based on h i g h expression o f the F O X P 3 transcript ion factor 6 ' 7 , and i n humans, re la t ively pure populat ions o f C D 4 + C D 2 5 + F O X P 3 + Tregs can be isolated f rom peripheral b l o o d by sort ing 1-3% o f the brightest C D 2 5 + cel ls w i t h i n the C D 4 + T c e l l subset 8 " 1 0 . C D 4 + C D 2 5 + Tregs possess several characteristics that suggest their intracellular s ignal ing f o l l o w i n g T ce l l receptor ( T C R ) act ivat ion may differ f rom that o f effector T cells . 2 T h i s research was or ig ina l ly publ i shed i n Blood. C r e l l i n N K , G a r c i a R V , and L e v i n g s M K . " A l t e r e d A c t i v a t i o n o f A K T is Requi red for the Suppressive Func t ion o f C D 4 + C D 2 5 + T Regula tory C e l l s . " B l o o d , 2007, M a r l ;109(5) :2014-22. E p u b 2006 O c t 24. C o p y r i g h t the A m e r i c a n Society o f Hemato logy . 60 F o r example , activated C D 4 + C D 2 5 + Tregs fa i l to produce most c lass ica l T ce l l -der ived cytokines , are hypo-responsive i n the absence o f exogenous growth factors, and suppress the functions o f many different ce l l types n ' 1 2 . The molecular changes, however , w h i c h underl ie this unique phenotype remain unknown . T C R s ignal ing is init iated by antigen (Ag ) -b i n d i n g , and results i n ac t ivat ion o f tyrosine kinases o f the Src , S y k and Tec famil ies , and assembly o f scaffolds o f adaptor molecules 1 3 . Phosphory la t ion o f these adaptors subsequently leads to act ivat ion o f downst ream effectors, i nc lud ing serine/threonine kinases such as the mi togen activated prote in kinases ( M A P K s ) and protein kinase C ( P K C s ) , and phosphat idyl inosit ide-3 kinase (PI3 'K)-dependent serine/threonine kinases, such as A K T . A c t i v a t i o n o f these cascades results i n cytoskeletal rearrangements, cy tokine product ion, c e l l cyc l e progression and engagement o f T ce l l effector functions. In contrast, T cel ls w h i c h are funct ional ly hypo-responsive due to ac t ivat ion i n the absence o f cos t imula t ion have a b lock i n act ivat ion o f the R a s / M A P K pathway and/or i n c a l c i u m m o b i l i z a t i o n 1 4 . A detai led analysis o f whether any o f these pathways may be altered i n human T C R -activated C D 4 + C D 2 5 + Tregs has been hampered by the scarcity o f these cel ls , diff icul t ies i n isola t ing pure populat ions due to the lack o f a reliable ce l l surface marker, and their relat ively poor prol i ferat ion i n vi t ro. Nevertheless, there have been several attempts to characterize intracel lular s ignal ing events i n mouse and human C D 4 + C D 2 5 + Tregs. M u r i n e C D 4 + C D 2 5 + Tregs displayed reduced act ivation o f A K T f o l l o w i n g s t imulat ion w i t h I L - 2 l 5 , and f o l l o w i n g s t imulat ion w i t h P M A / i o n o m y c i n , they were reported to have a reduced capacity to activate J N K , but not extracellular signal-related kinase ( E R K ) or p38 M A P K 1 6 . In contrast, i n vi tro expanded C D 4 + C D 2 5 + T r eg ce l l l ines der ived f rom human cord b l o o d showed normal 61 act ivat ion o f A K T , but s ignif icant ly reduced act ivat ion o f Ras , M A P kinase kinase ( M E K ) 1/2 and E R K 1 / 2 , upon T C R - m e d i a t e d act ivat ion 1 7 . Thus , there is currently no consistent evidence for a specific b lock i n one or more s ignal ing pathways i n C D 4 + C D 2 5 + Tregs, and the potential b io log i ca l relevance o f altered s ignal transduction to the unique funct ion o f C D 4 + C D 2 5 + Tregs has not been investigated. In order to overcome the p rob lem o f l im i t ed numbers o f C D 4 + C D 2 5 + Tregs w h i c h can be obtained ex v i v o , and avo id the caveats associated w i t h s tudying molecula r events i n T ce l l l ines expanded i n v i t ro upon super-physiologica l act ivat ion, we took advantage o f recent advances ' i n f l o w cytometry based methods 1 8 - 2 0 to study T C R - m e d i a t e d s ignal ing i n subpopulations o f human C D 4 + T cel ls . W e report here that T C R - m e d i a t e d act ivat ion o f A K T is d imin i shed i n human ex v i v o C D 4 + C D 2 5 + Tregs, as compared to C D 4 X D 2 5 " T cel ls , and that their suppressive capacity is dependent upon this altered s ignal transduction. 3.2 MATERIALS AND METHODS Cell purification. Per ipheral b lood was obtained f rom healthy volunteers fo l lowing informed consent and approva l o f the p ro toco l by the U B C C l i n i c a l Research Eth ics B o a r d . P B M C s were isolated by densi ty centrifugation over F i c o l l (Stemcell Technologies) , and C D 4 + T cells were subsequently pur i f ied by negative selection w i t h magnetic beads ( M i l t e n y i B i o t e c h or Stemcell Technologies). C D 2 5 + cel ls were either pur i f ied by posi t ive selection ( M i l t e n y i B io tech) and passed over 2 M S co lumns to ensure 9 0 - 9 5 % puri ty, or F A C S sorted. The f low- through f rom the C D 2 5 + posi t ive select ion was then passed over an L D deple t ion co lumn to remove any C D 2 5 1 0 cells, and used as the C D 2 5 " fraction. 62 TCR-mediated activation of signaling. F o l l o w i n g pur i f ica t ion, C D 4 + T cel ls or T c e l l l ines were rested overnight i n R P M I (Stemcel l Technologies) w i t h 1% human serum ( N o r t h B i o ) , and then starved i n serum-free R P M I for 2 hours pr ior to s t imulat ion. C D 4 + T cel ls (2-4 x l 0 6 / m l ) were activated by addi t ion o f c i C D 3 ( O K T 3 , 1 p g / m l , Orthoclone) , w i t h or without a C D 2 8 ( lp ,g /ml , B D Biosciences) m A b s , on ice for 15 minutes. C e l l s were washed once, and pr imary m A b s were cross- l inked by addi t ion o f anti-mouse I g G F ( A b ' ) 2 (20 u,g/ml, Jackson Immune Research) at 3 7 ° C for the indicated times. A c t i v a t i o n was arrested by f ixa t ion i n either 2 % formaldehyde or the F O X P 3 - s p e c i f i c fix/perm buffer (ebiosciences). Flow cytometry. F o r signaling experiments, fixed samples were washed, and permeabi l ized w i t h i ce -co ld methanol , and stained w i t h a n t i - C D 4 (conjugated to F I T C , A P C or A P C - C y 7 eBiosciences) , a n t i - C D 2 5 (conjugated to P E [ M i l t e n y i Bio tech] or P E - C y 7 [ B D Pharmingen]), and the indicated A b s against signaling proteins. A n t i - p h o s p h o - A K T (Ser 473 or T h r 308) , a n t i - A K T , a n t i - P T E N , and anti-phospho-S6 R i b o s o m a l Pro te in (Ser 235/236) were obtained f rom C e l l Signaling Technology. F o r non-conjugated A b s , a secondary goat-anti-rabbit A l e x a - 4 8 8 ( Mole c u l a r Probes) was used for detection. An t i -phospho E R K A l e x a -647 ( T 2 0 2 / Y 2 0 4 ) and anti-phospho P38 A l e x a - 4 8 8 ( T 1 8 0 / Y 1 8 2 ) , an t i -Granzyme A ( F I T C ) and an t i -Granzyme B (Alexa-647) were obtained f rom B D Pharmingen . The P I 3 ' kinase product, PI(3 ,4 ,5)P3, was detected by an a n t i - P I P 3 - F I T C antibody obtained f rom E c h e l o n Biosc iences Inc. A l l an t i -SHIP A b s were a generous gift f rom D r . Ge r ry K r y s t a l . F o r detection o f F O X P 3 , cells were fixed and permeabil ized according to the manufacturer 's instructions and incubated w i t h a n t i - F O X P 3 - P E (ebiosciences clone P C H 1 0 1 ) . Intracellular 63 staining for C T L A - 4 was performed as described 1 1 . A l l samples were read o n a B D F A C S C a n t o ™ and analyzed w i t h F C S Express V 3 ( D e N o v o software). Western blotting. H i g h l y pur i f ied C D 4 + C D 2 5 + T reg or C D 4 + C D 2 5 " T cel ls were activated as descr ibed above, and lysed us ing 1% N P - 4 0 lys is buffer 2 1 . Lysates f rom 1-3 m i l l i o n cells were quantitated by B C A (Pierce), separated by S D S - P A G E gel electrophoresis, transferred to P V D F membranes, and immunoblo t ted w i t h a n t i - p h o s p h o - F O X O (S256), or anti-phospho-A K T S473 A b s ( C e l l S igna l ing Techno logy) . Membranes were re-probed w i t h anti-p38 A b s (Santa C r u z Bio tech) to determine equivalency o f loading. L e n t i - v i r a l vec tors . T h e bidi rec t ional l en t i -v i ra l vector ( p C C L . s i n . c P P T . S V 4 0 p o l y A . C T E . m C M V A N G F R . P G K ) 2 2 was mod i f i ed by replacing the P G K promoter w i t h the human E F 1 a promoter to improve transgene expression i n human T cel ls . A c D N A cassette containing a truncated f o r m o f human A K T 1 l a c k i n g its P H domain , fused to the src myr i s tdy la t ion s ignal at the amino terminus, and the steroid b i n d i n g domain o f the estrogen receptor a long w i t h an hemagglu t in in ( H A ) epitope tag at its ca rboxy l terminus 2 3 , was inserted downstream o f the E F l a promoter. The resul t ing control ( p C C L ) and p C C L . A K T - E R ( A K T - E R ) vectors both drive constitutive expression o f A N G F R as a marker gene. V S V - p s e u d o t y p e d third-generation lenti-viruses were produced by transient 22 four -p lasmid co-transfection into H E K 2 9 3 T cel ls and concentrated by ultracentrifugation . Exp re s s ion titers were determined on H E K 2 9 3 T cel ls by l i m i t i n g d i lu t ion and ranged f rom 7 x 1 0 6 to 1.5x10 7 t ransducing uni ts /ml . 64 Generation of T cell lines and lenti-viral transduction. C D 4 + T cells were stained for CD4 and CD25 and FACS sorted into CD25 h i and CD25" fractions on a BD FACSAria as described 1 0 . T cells (150,000/well) were activated by stimulation with aCD3 (1 ug/mi) and APCs (500,000/well irradiated 50 Gy) or by co-culture with CD32 + CD58 + CD80 + L cells (200,000/well irradiated 75 Gy) and ctCD3 (100 ng/ml) 2 1 for 24. hours in T cell medium (Xvivo-15, 5% human serum [North Bio], lx penicillin/streptomycin [Invitrogen], lx Glutamax [Invitrogen]), in the presence of rhIL-2 (lOOU/ml, Chiron). Control pCCL or pCCL A K T - E R encoding lenti-virus was added to the activated cells at a multiplicity of infection of 2. The % of ANGFR+-transduced T cells was monitored after 6 days and was routinely 20-40%. A N G F R + cells were purified with anti-NGFR magnetic beads (Miltenyi Biotech) following the manufacturer's instructions and allowed to expand. Equivalent expression of the A K T - E R protein in the various cell lines was confirmed by flow cytometric analysis using an anti-HA-FITC mAb (Roche) as described 2 4 . Purified CD4 + CD25 + Treg and CD4 +CD25" T cell lines were re-stimulated every 14 days as previous described 1 0 and monitored at the end of every cycle to ensure preservation of their suppressive capacity. Proliferation and suppression of T cells. Transduced T cells were plated at 20,000 cells/well in 96-well plates, and stimulated with soluble aCD3 mAbs (1 ug/ml) in the presence of APCs (CD3-depleted PBMCs, irradiated at 50 Gy, 50,000 cells/well). Proliferation was assessed after 72 hours, by addition of [3H]thymidine (1 uCi per well; Amersham Biosciences) for the final 16 hours of the assay. To test for suppressive capacity CD4 + T cells (40,000 cells/well) were stimulated with aCD3/APCs, in absence or presence of 65 transduced T cel ls (20,000 ce l l s /wel l ) . Exper iments were conducted i n either vehic le alone ( E t O H 0.15%) or 4 -hydroxytamoxi fen ( 4 H T ) at 150 n M (Sigma) , a concentrat ion w h i c h was found to stimulate m a x i m u m prol i ferat ion o f transduced T cel ls . Suppression was assessed by determining [ 3 H]thymid ine incorporat ion after 72 hours. In a l l cases untransduced C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T ce l l l ines were analyzed i n paral le l and found not to be different f rom control vector-transduced T ce l l l ines (data not shown). Cell surface marker expression and cytokine production. Transduced T cel ls (100,000 ce l l s /we l l i n 9 6 - w e l l plates), were st imulated w i t h a C D 3 (plate-bound lp ,g /ml) and c i C D 2 8 (soluble 1 p:g/ml) m A b s . Exper iments were conducted i n either vehic le alone or 150 n M 4 H T . Supernatants were col lected at the indicated t imes and the T h l / T h 2 cytometr ic bead assays ( B D Biosc iences ) was used to determine amounts o f cy tokine . A f t e r 48 hours, the cel ls were col lected for f l o w cytometr ic analysis o f ce l l surface markers. Statistical analyses. A l l analysis for statistically significant differences was performed w i t h the student's paired t test, p values less than 0.05 were considered significant. F o r cytokine analysis, l o g (base 10) values were used for statistical analysis i n order to account for the var iab i l i ty between ce l l l ines. A l l cultures were performed i n tr ipl icate and error bars represent the S D . 66 3.3 R E S U L T S Single cell analysis of intracellular signaling cascades reveals that ex vivo human CD4+CD25+ Tregs and CD4+CD25 T cells have equivalent ERK1/2 and p38 MAPK phosphorylation following TCR activation A major limitation in the molecular analysis of human C D 4 + C D 2 5 + Tregs is the inability to isolate homogeneous populations of live cells ex vivo. Although highly suppressive populations of C D 4 + C D 2 5 + Tregs can be isolated by sorting the brightest 2-3% C D 2 5 + cells, an in vitro expansion step is required to obtain sufficient cells to perform signal transduction assays using traditional western blotting. In vitro expansion of C D 4 + C D 2 5 + Tregs is possible u , but prolonged supra-physiological TCR-mediated activation, and the outgrowth of contaminating non-Tregs represent confounding factors. In order to overcome these limitations, we have used flow cytometry to analyze intracellular signaling cascades in I i n "}C\ 2S 29 heterogeneous populations of C D 4 T cells ' " . This method allows quantitative analysis of signaling within single cells, more accurately replicates the in vivo scenario, and decreases variables associated with cell isolation and handling. The specificity and sensitivity of Abs directed against TCR-activated proteins such as phospho-ERK, -p38 and - A K T have been validated by use of specific inhibitors and parallel Western blot analys is 2 5 ' 2 8 . 67 i. > I • C I Phospho- *. ERK1/2 B. phospho-ERK1/2 Tine (mn) C. phospho p38 4 10 T O H 1 1 1 1 0 20 40 60 80 Time (mil) Figure 3.1. Single cell analysis of MAPK activation in ex vivo human CD4+CD25+ Tregs and CD4+CD25 T cells following TCR activation. (A) Ex vivo CD4 + T cells were left unstimulated or stimulated with cross-linked aCD3/28 mAbs for 10 minutes, and stained for CD4, CD25 and phospho-ERKl/2. Histograms depicting 68 levels of phospho-ERK were generated by gating on subsets of CD4+CD25high or CD4+CD25" T cells. (B) Geometric MFIs of cell populations stained with anti-phospho-ERKl/2 or (C) -p38 Abs were determined over a 60 min time course following activation with aCD3/28 mAbs. The experiment was performed three times with similar results and a representative analysis is shown. The inset is the fold change in MFI from resting to activation (at 10 minutes), with each point representing a separate experiment. G i v e n the well-establ ished role o f M A P K pathways i n T C R - s t i m u l a t e d c e l l d i v i s i o n 1 4 , we investigated the activation o f these kinases i n C D 4 + C D 2 5 + T r eg versus C D 4 + C D 2 5 " T cel ls upon T C R st imulat ion. C D 4 + T cel ls were stimulated w i t h O . C D 3 / C D 2 8 m A b s , fixed and stained w i t h a n t i - C D 4 , - C D 2 5 , and - p h o s p h o E R K A b s . The relat ive amounts o f E R K phosphoryla t ion were compared i n the two subsets by gating on the brightest 2 - 3 % C D 2 5 + cel ls versus C D 2 5 " cel ls (Figure 3.1 A). W e consistently observed equivalent phosphoryla t ion o f E R K i n C D 4 + C D 2 5 + T reg and C D 4 + C D 2 5 " T cel ls f o l l o w i n g T C R act ivat ion (n=4, p=NS) . A n a l y s i s o f these experiments at mul t ip le t ime points showed that there was no difference between C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T cel ls i n the kinet ics o f E R K phosphoryla t ion (Figure 3.1B). S i m i l a r l y , no difference i n the capacity o f C D 4 + C D 2 5 + Tregs to activate p38 M A P K was observed (Figure 3.1C) (n=4, p=NS) . These data indicate that C D 4 + C D 2 5 + Tregs respond to T C R - m e d i a t e d s t imulat ion through the E R K and p38 M A P K pathways and demonstrate that these cel ls do not have a g loba l defect i n T C R - m e d i a t e d s ignal transduction. Human CD4+CD25+ Tregs have a reduced capacity to phosphorylate AKT following TCR stimulation. A c t i v a t i o n o f the P I 3 ' K - A K T pathway, w h i c h can be further enhanced by co-s t imula t ion v i a C D 2 8 3 0 , promotes T ce l l su rv iva l and is required for product ion o f both T h l - and T h 2 -associated cytokines 3 I ' 3 2 . W e observed a profound defect i n the capacity o f C D 4 + C D 2 5 + Tregs to activate A K T upon s t imulat ion v i a C D 3 and C D 2 8 , and after 10 m i n , the increase i n levels o f p h o s p h o - A K T (Ser473) was on ly 1.8 ± 0.17 fo ld i n C D 4 C D 2 5 Tregs compared to 69 3.2 ± 0.58 fo ld i n C D 4 + C D 2 5 " T cel ls O = 3 . 5 2 x l 0 " 6 , n = l l Figure 3.2A). A n a l y s i s over 30 m i n f o l l o w i n g s t imulat ion revealed that the reduct ion i n phosphoryla t ion o f Ser473 was not due to altered kinet ics o f activation. Defect ive A K T phosphoryla t ion i n C D 4 + C D 2 5 + Tregs was also observed us ing tradit ional western b lo t t ing (Figure 3.2B). A s imi l a r defect was found i n the absence o f C D 2 8 co-s t imula t ion (Figure S3.1A). The amount o f total A K T protein was equivalent i n the two c e l l types, ind ica t ing that differential expression o f A K T was not responsible for this phenotype (Figure S3.1B). I n order for A K T to be fu l ly activated, both Ser473 and Thr308 must be phosphorylated 3 3 ' 3 4 . E x a m i n a t i o n o f the phosphoryla t ion state o f A K T at Thr308 f o l l o w i n g T C R act ivat ion showed that C D 4 + C D 2 5 + Tregs were no different from C D 4 + C D 2 5 " T cel ls in this respect (Figure 3.2C). 70 A. phospho-AKT (Ser473) 25i D. phospho-AKT (Ser473) on F0XP3+ or F0XP3 cells • CD4+CD25' • CD4+CD25' 10 15 20 25 30 35 Time^nh) B. phospho-AKT (Ser473) CD4-CD25- CD4+CD25* min: 0 10 pAKT p38 C. phospho-AKT (Thr308) 3 0 , #CD4"CD25 + • CD4+CD25-I 1 * T 20 40 Trne(mir») 60 80 ro Q. 10 10* 10 2 101 10C 1.5% 10 10 10 10 10 CD4 * 0 10 10" 10 10 pAKT »• E. phospho-AKT (Ser473) on F0XP3+ or FOXP3 cells 40 i • CD4 + FOXP3 + O CD4+FOXP3' 10 20 Time (mini) 30 40 Figure 3.2. Human CD4+CD25+ Tregs have a reduced capacity to phosphorylate A K T following TCR stimulation. (A) Ex vivo CD4 + T cells were stimulated with cross-linked aCD3/28 mAbs for the indicated times, and stained for CD4, CD25, and geometric MFIs of cells stained with anti-phospho-AKT (Ser473) were determined in CD4 + CD25 h ' 8 h or CD4 +CD25" T cells. (B) >90% pure C D 4 + C D 2 5 h i g h and CD4 +CD25" T cells were left unstimulated or stimulated with aCD3/28 mAbs for 10 minutes, lysed and analyzed by Western blotting for amounts of A K T (Ser473) phosphorylation. Blots were re-probed with anti-p38 Abs to ensure equivalency of loading. (C) Ex vivo CD4 + T cells were stimulated and analyzed as in A , and MFIs following 71 staining with anti-phospho-AKT (Thr308) were determined. (D'and E) Ex vivo CD4+ T cells were stimulated as in A, then stained for CD4, FOXP3 and phospho-AKT (Ser473). (D) Histograms depicting levels of phospho-AKT were generated by gating on subsets of CD4+FOXP3+ or CD4+FOXP3" T cells, and (E) MFls of cells stained with anti-phospho-AKT (Ser473) were determined in CD4+FOXP3+ or CD4+FOXP3" T cells. A, C &E represent a single experiment with the inset depicting the fold change in MFI from resting to activation (10 minutes) for all experiments. For B & D, a single representative example of 3 experiments performed is depicted. C D 2 5 is a not a true lineage marker for Tregs since it can also be expressed by-activated effector cel ls . Recent data indicate that F O X P 3 may be a more specif ic marker 3 5 , a l though it may also represent an act ivat ion marker i n a subset o f cel ls i n humans 8 ' 2 4 , 3 6 . W e therefore examined the capacity o f C D 4 + F O X P 3 + Tregs to activate A K T . S i m i l a r to C D 2 5 + cel ls , F O X P 3 + cel ls d isplayed a consistent defect i n their capacity to phosphorylate A K T : after l O m i n o f s t imulat ion, phospho-Ser473 i n C D 4 + F O X P 3 + Tregs was on ly 1.86 ± 0.57 fo ld compared to 2.65 ± 0.7 fo ld i n C D 4 + F O X P 3 " T cel ls (p=0.0089, n=3) (Figure 3.2D). T h i s reduced act ivi ty was not due to altered kinet ics as levels o f p h o s p h o - A K T i n C D 4 + F O X P 3 + Tregs remained l o w throughout a 30 m i n t ime course (Figure 3.2E). These data indicate that the altered s ignal ing i n C D 4 + C D 2 5 + putative Tregs was not due to contaminat ing T effector cel ls . T o determine i f reduced A K T phosphoryla t ion was the result o f an upstream defect i n the ac t iv i ty o f P I 3 ' K , w e investigated the levels o f phosphat idyl inositol-3,4,5-triphosphate (PIP3) , the b io log i ca l l y active product o f P I 3 ' K , i n stimulated C D 4 + C D 2 5 + Tregs. A s shown i n Figure S3.2A, no difference was observed between C D 4 + C D 2 5 + Tregs and C D 4 + C D 2 5 " T cel ls i n P IP3 levels f o l l o w i n g T C R act ivat ion, indica t ing that P I 3 ' K funct ion i n these cel ls is equivalent. T h i s f ind ing is consistent w i t h observations made i n the murine system, downstream o f the I L - 2 R 1 5 . 72 W e then investigated whether C D 4 + C D 2 5 + Tregs have h igh basal expression and/or increased act ivi ty o f src homology 2 (SH2) -con ta in ing inosi to l phosphatase ( S H I P ) 1 or phosphatases and tensin homologue deleted o n chromosome 10 ( P T E N ) , the l i p i d phosphatases w h i c h counteract the act ivi ty o f P I 3 ' K 3 1 . A s shown i n Figure S3.2B and C, C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T cel ls d i d not differ i n their expression of, or capacity to phosphorylate, S H I P 1 . A d d i t i o n a l l y , no differences i n the basal expression levels o f P T E N were detected (Figure S3.2D). Diminished AKT phosphorylation in CD4+CD25+ Tregs results in decreased activation of downstream effectors. T o evaluate i f decreased Ser473 phosphoryla t ion was indicat ive o f a drop i n A K T kinase act ivi ty , w e examined the act ivat ion state o f F O X O l ( F R K H L ) and F O X 0 3 a ( F R K H L - 1 ) , proteins k n o w n to be direct ly phosphorylated by A K T 3 8 . A s A b s val idated for f l o w cytometry were not avai lable, we pur i f ied C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T cel ls , activated them w i t h a C D 3 / C D 2 8 m A b s , and performed Western blot t ing o f ce l l lysates (Figure 3.3A). C D 4 + C D 2 5 + Tregs had consistently lower levels o f F O X O phosphoryla t ion as compared to C D 4 + C D 2 5 " T effectors, suggesting that the reduced A K T phosphoryla t ion resulted i n reduced kinase act ivi ty (densitometric analysis 1 .82±1.3 fo ld i n C D 4 + C D 2 5 + Tregs versus 3 . 3 ± 1.3 fo ld i n T effectors at 5 m i n , n=3, p=0.006). In addi t ion, we examined the capacity o f C D 4 + C D 2 5 + Tregs to phosphorylate the S6 r ibosomal protein. S6 is a direct target o f m a m m a l i a n target o f r apamyc in ( m T O R ) , a Ser /Thr kinase that is also a k n o w n target o f A K T 3 9 . E x v i v o C D 4 C D 2 5 Tregs d isplayed s ignif icant ly lower amounts o f phospho-S6 after act ivat ion than C D 4 + C D 2 5 " T cel ls ( 1 . 9 ± 0 . 2 8 fo ld versus 3 . 6 ± 0 . 9 6 fo ld at 73 10 min, n=3, p=0.03) (Figure 3.3B). Thus, the defect in A K T phosphorylation observed in ex vivo C D 4 + C D 2 5 + Tregs results in reduced kinase activity and appears to cause a general blockade of activation of downstream effector molecules. A. phospho-FOXO CD4+CD25+ CD4+CD25" min: 0 5 15 0 5 15 T h e (min) Figure 3.3. Diminished AKT phosphorylation in CD4+CD25+ Tregs results in decreased activation of downstream effectors. (A) >90% pure CD4+CD25high and CD4+CD25" T cells were left unstimulated or stimulated with aCD3/28 mAbs for the indicated times, lysed and analyzed by Western blotting for amounts of phosphorylated FOXOl and FOX03a. Blots were re-probed with anti-p38 Abs to ensure equivalency of loading. Shown is a representative experiment of 3 performed. (B) Ex vivo CD4+ T cells were stimulated with ctCD3/28 Abs for the indicated times, and MFIs following staining with anti-phospho-S6 (Ser235/236) were determined in CD4+CD25h'Bh or CD4+CD25" T cells. A single representative experiment is depicted, and the inset represents the fold change in MFI from resting to activation (10 minutes), for each separate experiment. 74 Enforced activation of AKT reverses the suppressive capacity of CD4+CD25+ Tregs G i v e n the w e l l established role o f the P I 3 ' K - A K T pathway i n ce l l cyc le progression 3 9 , we hypothesized that d imin i shed act ivat ion o f this pathway i n C D 4 + C D 2 5 + Tregs might be the direct cause o f their hyporesponsive and suppressive phenotype. In order to test this poss ib i l i ty we first generated a series o f T ce l l l ines and conf i rmed that expanded C D 4 + C D 2 5 + T r e g ce l l lines d isplayed d imin i shed phosphorylat ion o f A K T and S6 (Figure S3.3). W e then used lent i -v i ra l mediated gene transfer to over-express a cond i t iona l ly active fo rm o f A K T , consis t ing o f the A K T kinase domain fused to an amino- terminal myr i s toy la t ion sequence and the hormone b i n d i n g d o m a i n o f the estrogen receptor ( E R ) at the carboxy terminus ( A K T - E R ) 2 3 . The result ing protein remains i n an inact ive state i n the cy top lasm unt i l 4 -hydroxytamoxi fen ( 4 H T ) is added, whereupon a 17-50 f o l d increase i n kinase ac t iv i ty is induced 2 3 . C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T ce l l s were sorted and infected w i t h cont ro l - ( p C C L ) or A K T - E R - e n c o d i n g ( A K T - E R ) lenti-viruses (Figure S3.4A). The act ivi ty o f the A K T - E R protein was conf i rmed by analysis o f phosphoryla t ion o f a downstream target, S6 , upon addi t ion o f 4 H T (Figure S3.4B). F o l l o w i n g isola t ion and expansion o f transduced cells , their prol iferat ive capacity was tested i n the absence or presence o f 4 H T . Surpr is ingly , a l though the presence o f the A K T - E R resulted i n an increase i n the prol i ferat ive capacity o f the C D 4 + C D 2 5 + Tregs upon T C R - s t i m u l a t i o n i n the presence o f 4 H T ( 5 . 5 ± 2 . 2 5 fold compared to control , n=4, p~0.014) , these cel ls remained hypo-responsive i n compar i son to C D 4 + C D 2 5 " T effector cells (Figure 3.4A). A s expected, act ivat ion o f A K T - E R by addi t ion o f 4 H T i n the A K T - E R + C D 4 + C D 2 5 " T cel ls also resulted i n increased prol i ferat ion ( 1 . 6 4 ± 0 . 3 5 f o l d compared to control , n=4, p=0.017). 4 H T had no effect o n control p C C L - t r a n s d u c e d T c e l l l ines (Figure 3.4A). 75 • vehicle pCCL AKT-ER pCCL AKT-ER CD4+CD25* CD4*CD25-100 • vehicle • .4HT pCCL AKT-ER pCCL AKT-ER CD4-CD25+ CD4+CD25-Figure 3.4. Enforced activation of AKT reverses the suppressive capacity of CD4+CD25+ Tregs. CD4+CD25high or CD4+CD25'T cells were transduced with lenti virus encoding an inducible AKT-ER or control (pCCL) vector. (A) Cells were stimulated with aCD3/APCs in the presence of vehicle alone or 4HT (150 nM). Inset is the fold increase in proliferation induced by 4HT in AKT-ER+ CD4+CD25+ Treg or CD4+CD25" T cell lines, with each point representing a single experiment. (B) CD4+ T cells were stimulated with aCD3 (soluble lu.g/ml) and irradiated APCs in the absence or presence of a 1:2 ratio (T cell lines:target) of control-transduced, or AKT-ER+ CD4+CD25+ Treg or CD4+CD25" T cell lines, in the presence of either vehicle alone or 4HT (150 nM). Data are depicted as % suppression ([l-(Te+Treg/Te alone]*100), negative values are plotted as 0. Proliferation and suppression were assessed by 3H-thymidine incorporation. A single representative experiment of 4 is depicted. W e next determined whether restoration o f A K T activi ty altered the suppressive capacity o f C D 4 + C D 2 5 + Tregs. Responder C D 4 + T cel ls were st imulated w i t h a C D 3 / A P C i n the absence or presence o f C D 4 + C D 2 5 + T r eg or C D 4 + C D 2 5 " effector T c e l l l ines, and i n the 76 absence or presence o f 4 H T . A s expected, i n the absence o f 4 H T , p C C L - t r a n s d u c e d or A K T -E R - expressing C D 4 + C D 2 5 + T reg ce l l l ines suppressed prol i ferat ion by 7 0 - 8 0 % (ratio 1:2 Treg:Teffector) (Figure 3.4B). In contrast, i n the presence o f 4 H T , the capacity o f C D 4 + C D 2 5 + Tregs to suppress effector T ce l l prol i ferat ion was almost complete ly abrogated (percent suppression i n the presence o f 4 H T for A K T - E R T r e g ce l l l ines was 1 2 . 5 ± 1 7 % , compared to 7 6 ± 1 6 % for p C C L controls, n=5, p=0.000954) (Figure 3.4B). The degree o f increase i n prol i ferat ion upon act ivat ion o f A K T i n C D 4 + C D 2 5 + T r e g c e l l l ines (Figure 3.4A) is c lear ly insufficient to account for the lack o f suppression observed i n these assays. C D 4 + C D 2 5 " effector T ce l l l ines d i d not exhibi t suppressive capacity i n either control or 4 H T condi t ions , nor d i d the presence o f 4 H T i t se l f cause T-effector cel ls to proliferate (data not shown). Enhanced AKT activity in CD4+CD25+ Tregs does not suppress expression of FOXP3, CTLA-4, CD25, or Granzyme A or B A l t h o u g h the mechanism o f suppressive act ion o f C D 4 + C D 2 5 + Tregs remains to be ident i f ied, it has been speculated that F O X P 3 6 , C T L A 4 2 , granzymes A or B 4 0 ' 4 1 and possibly, also C D 2 5 i t se l f 4 2 may play a role. W e therefore determined whether expression o f any o f these 5 proteins was altered upon addi t ion o f 4 H T to A K T - E R " C D 4 + C D 2 5 T reg ce l l l ines. A s the control p C C L C D 4 + C D 2 5 + Tregs d id not survive upon s t imula t ion w i t h 4 H T alone for the length o f t ime required for these assays (data not shown) , the T c e l l l ines were activated w i t h a C D 3 / 2 8 for 48 hrs i n the absence or presence o f 4 H T , and analysed by f l o w cytometry. A s expected, both control p C C L and A K T - E R + C D 4 + C D 2 5 + T r eg l ines expressed h i g h levels o f F O X P 3 , C T L A 4 and C D 2 5 compared to C D 4 + C D 2 5 " T ce l l l ines i n the resting state (Figure 3.5). U p o n s t imulat ion w i t h 4 H T , expression o f F O X P 3 and C T L A 4 was 77 unchanged (or slightly increased), and CD25 was increased by 2.84±1.0 fold (n=3, p=0.046) (Figure 3.5). Although previous reports have suggested a role for granzymes in CD4 + CD25 + Treg function 4 0 , 4 ' , we did not observe high levels of granzyme A or B expression on resting or activated CD4 + CD25 + Tregs compared to CD4+CD25" T cells. Addition of 4HT to A K T -E R + CD4 + CD25 + Tregs either did not change, or slightly increased, the minimal expression of granzyme A and B. Therefore, the loss of suppressive activity upon A K T activation is not due to decreased expression of any of these Treg-associated proteins. 78 PIA.L F0XP3 CTLA4 CD25 o O Granzyme A-101 102 1 0 3 1 0* 10' 10z 103 10« Granzyme B - R e s t i n g | | Resting — aCD3/28 — aCD3/28*4HT CD4+CD25 C04+CD25+ C04+CD25+ CD4+CD25+ Figure 3.5. Enhanced AKT activity in CD4+CD25+ Tregs does not reduce expression of FOXP3, CTLA-4, CD25, or Granzymes A or B. AKT-ER- (or control pCCL-) transduced CD4+CD25+Treg or CD4+CD25"T cells were activated with ccCD3/28 mAbs in the presence of either vehicle alone (EtOH) or 4HT (150 nM). 79 After 48 hours, cells were stained for CD25, CTLA-4, FOXP3, and Granzyme A and B. Unstimulated CD4+CD25+and CD4+CD25" T cell lines are included for comparison. A single representative experiment of 4 is depicted. Enhanced AKT activity in CD4+CD25+ Tregs restores their capacity to produce IFN-y, TNF-a, IL-4, and IL-10, but not IL-2. One o f the def in ing characteristics o f C D 4 + C D 2 5 + Tregs is their inab i l i ty to produce significant amounts o f cytokines, par t icular ly I L - 2 and I F N - y 1 0 . G i v e n the capacity o f exogenous I L - 2 to reverse suppression, and evidence that constitutive A K T act ivat ion i n T cel ls can induce product ion o f I L - 2 and I F N - y 3 1 , it was important to determine whether the lack o f suppression i n the A K T - E R C D 4 + C D 2 5 + Tregs i n the presence o f 4 H T was s imply due to increased I L - 2 product ion. C o n t r o l p C C L - and A K T - E R - e x p r e s s i n g C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T ce l l l ines were st imulated w i t h c tCD3/28 m A b s , i n the absence or presence o f 4 H T , and supernatants were col lec ted after 24hr (Figure 3.6A). A s expected, control p C C L and, i n the absence o f 4 H T , A K T - E R + C D 4 + C D 2 5 + Tregs produced s ignif icant ly lower levels o f I L - 2 than the C D 4 + C D 2 5 " T ce l l l ines. Surpr i s ing ly , the capacity o f the A K T - E R C D 4 + C D 2 5 + T reg c e l l l ines to produce I L - 2 was not altered upon addi t ion o f 4 H T . In contrast, addi t ion o f 4 H T to the A K T - E R + C D 4 + C D 2 5 " T c e l l l ines resulted i n a 3 . 4 6 ± 0 . 5 9 fo ld increase i n I L - 2 product ion (n=3, p=0.009.4). 80 A . IL-2 120 p C C L A K T - E R p C C L A K T - E R CD4-CD25* C D 4 - C D 2 S -8 . 3.0-1 1.6, p C C L A K T - E R p C C L A K T - E R p C C L A K T - E R p C C L A K T - E R CD4*CD2S* C D 4 - C D 2 5 " CD4*CD2S- C D 4 - C D 2 5 " Figure 3.6. Enhanced AKT activity in CD4+CD25+ Tregs restores their capacity to produce IFN-y, TNF-a, IL-4, and IL-10, but not IL-2. AKT-ER (or control pCCL)-transduced CD4+CD25+Treg or CD4+CD25" T cell lines were activated with aCD3/28 mAbs in the presence of either vehicle alone or 4HT (150 nM). Supernatants were collected after 24 hours (A) or 48 hours (B) and assayed for amounts of cytokines by CBA. A representative experiment of 3 performed is depicted. W e next examined whether act ivat ion o f A K T i n the C D 4 + C D 2 5 + T r eg ce l l l ines affected product ion o f other T h l - or Th2-associated cytokines. In contrast to I L - 2 , act ivat ion o f A K T by addi t ion o f 4 H T i n A K T - E R + C D 4 + C D 2 5 + Tregs restored their capacity to produce I F N - y (7.8-58 fo ld compared to without 4 H T control , n=4, p=0.0026), T N F - a (2.5-21.3 fo ld , n=4, p=0.0096), I L - 4 (3.2-7.9 f o l d n=4, p=0.0045) and I L - 1 0 (3.1-12.1 fo ld , n=4, 81 p=0.0035) (Figure 3 . 6 B ) to levels equivalent to, or i n the case o f I L - 4 and I L - 1 0 , higher than those produced by C D 4 + C D 2 5 " T cel ls . 3.4 D I S C U S S I O N T h i s report represents the first characterization o f intracellular s ignal ing cascades i n ex v i v o human C D 4 + C D 2 5 + Tregs. U s i n g f l o w cytometry based assays, we found that C D 4 + C D 2 5 + Tregs have a capacity equivalent to that o f non-Tregs to activate E R K and p38 M A P K s , but have a significant defect i n the phosphoryla t ion o f A K T upon T C R - m e d i a t e d act ivat ion. Th i s abnormal A K T act ivat ion resulted i n decreased act ivi ty o f downstream effectors. Moreove r , TCR- independen t condi t ional act ivat ion o f exogenous A K T i n Tregs reversed their suppressive capacity, indicat ing that the defect i n the capacity o f C D 4 + C D 2 5 + Tregs to fu l ly activate A K T contributes to their unique suppressive function. Interestingly, enforced act ivat ion o f A K T i n C D 4 + C D 2 5 + Tregs neither reversed their inabi l i ty to produce I L - 2 f o l l o w i n g T C R st imulat ion, nor decreased expression o f F O X P 3 , C T L A 4 or C D 2 5 , indica t ing that addi t ional factors are i n v o l v e d i n the reversal o f suppression. Together, these data represent the first causal associat ion between an altered molecu la r s ignal and the b i o l o g i c a l funct ion o f C D 4 + C D 2 5 + Tregs. Phosphory la t ion o f A K T at both Thr308 and Ser473 is required for m a x i m a l kinase act ivi ty 3 3 . A c t i v a t e d C D 4 + C D 2 5 + Tregs showed decreased phosphoryla t ion o f A K T at Ser473, but no decrease i n phosphoryla t ion at Thr308 . A K T act ivi ty i n C D 4 + C D 2 5 + Tregs, however , was c lear ly impai red since the phosphoryla t ion o f both direct ( F O X O l and F O X 0 3 A ) and indirect (S6) downstream targets was also reduced. T h i s observat ion is consistent w i t h the f inding that murine C D 4 + C D 2 5 + Tregs had lower A K T act ivat ion upon I L - 2 receptor s t imulat ion 1 5 , and suggests these cel ls may have a g loba l defect in their 82 capacity to activate this kinase. In contrast, L i et a l . recently reported that i n v i t ro expanded C D 4 + C D 2 5 + Tregs der ived f rom cord b l o o d were found to have normal ac t ivat ion o f A K T at longer (15 min -16 hr) t ime points after act ivat ion l 7 . O u r f ind ing that C D 4 + C D 2 5 + T r eg ce l l l ines f rom adult peripheral b lood have reduced phosphoryla t ion o f A K T and S6 indicates that this difference is not s imply related to i n v i t ro expansion. Rather, the apparent discrepancy between our observations and those o f L i et a l . cou ld be due to contaminat ion by effector T cel ls , and/or the t ime points chosen by these authors for examinat ion o f A K T act ivat ion. O u r data indicate that the reduced phosphoryla t ion o f A K T i n C D 4 + C D 2 5 + Tregs does not result f rom changes i n the ac t iv i ty o f P O ' K itself, or o f the phosphatases S H I P or P T E N . Thus , the upstream defect must l ie either i n their capacity to activate or recruit P D K 2 , the putative Ser473 kinase, or i n increased funct ion o f a Ser473-specif ic phosphatase. The identity o f P D K 2 has long been the subject o f intensive research, and kinases such as In tegr in- l inked kinase ( I L K ) 4 3 _ 4 5 , D N A - P K 4 6 , and even A K T i t s e l f 4 7 can phosphorylate this residue. M o r e recently, it has been suggested that w h e n complexed w i t h r ictor, m T O R is the dominant Ser473 kinase 4 8 . In terms o f Ser473-specif ic phosphatases, the recently identif ied P H doma in leuc ine- r ich repeat protein phosphatase ( P H L P P ) 4 9 or the more broadly acting protein phosphatase 2 A ( P P 2 A ) 5 0 may be invo lved . Current ly , the relative contr ibut ion o f these molecules to the state o f Ser473 phosphoryla t ion i n pr imary human T cel ls has not been investigated, and further studies are required before the upstream defect i n C D 4 + C D 2 5 + Tregs can be explored. A c t i v a t i o n o f the P I 3 ' K - A K T pathway is a fundamental requirement for T ce l l su rv iva l and c e l l cyc l e progression 3 9 , and transgenic m i c e expressing a const i tut ively active fo rm o f A K T displayed enhanced T ce l l ac t ivat ion and cytokine product ion , lost the 83 requirement for C D 2 8 co-s t imula t ion to fu l ly activate T-ce l l s , and developed autoimmune-l ike syndromes 5 1 ' 5 2 . W e therefore hypothesized that restoration o f A K T s ignal ing i n C D 4 + C D 2 5 + Tregs w o u l d reverse their hyporesponsive state. U p o n len t i -v i ra l mediated expression o f an induc ib ly active fo rm o f A K T , however , their capacity to proliferate was on ly s l ight ly (~5 fold) enhanced, and remained s ignif icant ly lower than that o f C D 4 + C D 2 5 " T cel ls . C D 4 + C D 2 5 + Tregs may therefore have irreversible epigenetic changes and/or unident i f ied b locks i n other s ignal ing pathways, w h i c h underl ie this phenotype. T h i s f inding may also be related to the cont inued h i g h express ion o f F O X P 3 i n the A K T - E R C D 4 + C D 2 5 + Tregs, since ectopic expression o f F O X P 3 alone is sufficient to induce anergy i n human C D 4 + T ce l l s 2 4 . Unexpec ted ly , act ivat ion o f A K T d i d not restore the capacity o f C D 4 + C D 2 5 + Tregs to produce I L - 2 , al though it s ignif icant ly enhanced product ion o f a l l other cy tokines tested. In contrast, ac t ivat ion o f A K T i n C D 4 + C D 2 5 " T cel ls resulted i n s ignif icant ly enhanced I L - 2 product ion. These latter data are consistent w i t h findings f rom T cel ls f rom A K T transgenic mice , w h i c h produced tenfold higher levels o f both I L - 2 and I F N - y (but not I L - 4 ) 3 1 than wi ld - type T cel ls , and w i t h the k n o w n role o f A K T i n act ivat ion o f N F A T . The lack o f I L -2 product ion by C D 4 + C D 2 5 + Tregs w i t h restored A K T act ivi ty suggests that these cel ls have a permanent b lock i n product ion o f this cy tokine , w h i c h may be related to their inabi l i ty to undergo chromat in remodel ing at the I L - 2 locus after act ivat ion 1 6 : The mechan i sm o f C D 4 + C D 2 5 + T r eg ce l l suppression has yet to be fu l ly defined. W e investigated whether restoration o f A K T act ivi ty might alter expression o f molecules p rev ious ly associated w i t h suppressive act ivi ty . O u r data, however , indicate that the loss o f suppressive capacity is not due to decreased expression o f F O X P 3 , G D 2 5 , C T L A 4 , and/or 84 granzymes A and B . W e also considered the poss ib i l i ty that expression o f A K T - E R may enhance the capacity o f C D 4 + C D 2 5 + T r eg cel ls to respond to I L - 2 . Restorat ion o f A K T act ivi ty , however , neither caused the cel ls to respond to lower amounts o f I L - 2 nor s ignif icant ly increased the magnitude o f their response (data not shown). Thus , the basis for the reversal o f suppression by expression o f active A K T i n C D 4 + C D 2 5 + Tregs remains unclear. It is possible that enhanced product ion o f cytokines other than I L - 2 may be i nvo lved . It seems u n l i k e l y that I L - 1 0 , I F N - y or T N F - a w o u l d be impl ica ted , since these cytokines generally have anti-proliferative, pro-suppressive effects 5 3 - 5 5 . H i g h levels o f I L - 4 might be i n v o l v e d , since exogenous I L - 4 has been shown to reverse suppression i n mur ine cel ls 5 6 , a l though w e have found this not to be the case for human cel ls ( M K L , unpubl ished data). Future experiments i n v o l v i n g neutral izat ion o f these cytokines w i l l be required to investigate their possible role. In conc lus ion , we have demonstrated for the first t ime a causal connec t ion between the b i o l o g i c a l characteristics o f C D 4 + C D 2 5 + Tregs and an altered s ignal ing pathway. Our data provide further evidence that the i n v i t ro hyporesponsive state o f C D 4 + C D 2 5 + Tregs does not correlate w i t h suppressive capacity. In addi t ion, the observat ion that C D 4 + C D 2 5 + Tregs have a profound inabi l i ty to produce I L - 2 , even when their capacity to secrete other cytokines is restored, suggests that these cel ls have irreversible epigenetic changes. F i n a l l y , the condi t iona l ly suppressive A K T - E R C D 4 + C D 2 5 + Tregs represent the first system a l l o w i n g induc ib le abrogation o f suppression at the level o f the Treg ce l l , rather than by changing the susceptibi l i ty to suppression o f the target T cel ls . Induct ion o f A K T act ivi ty i n C D 4 + C D 2 5 + Tregs represents a powerfu l tool for the study o f the mechanism(s) o f suppression, and a major advance towards therapeutic modula t ion o f peripheral tolerance. 85 3.5 S U P P L E M E N T A R Y F I G U R E S Figure S3.1. C D 4 + C D 2 5 + Tregs have defective A K T phosphorylation in response to stimulation via the T C R in the absence or presence of costimulation via CD28. Ex vivo CD4 + T cells were (A) stimulated with aCD3 or aCD3/CD28 mAbs and stained for CD4, CD25 and phospho-AKT (Ser473) or (B) unstimulated cells were stained for CD4, CD25 and total A K T . A is an average of 3 independent experiments (p=0.028 aCD3 alone), B is representative of 3 experiments. 8 6 A. PIP3 16, 14. 12-U -S 10 . u ' C 8 . IE £ o BC a> m CD 4 -2 • 0--0 • CD4+C025+ DC04+CD25-20 40 Time (min) 80 B. Basal SHIP 300, z 2004 1004 CD4* CD25 • CD4+CD25+ • CD4+CD25" D. Basal PTEN 250 n r z 2 0 0 2 o '= 150 a> e S100 o 30 40 Time (Min) 50 0 • • 004X025* CD4*CD25" Figure S3.2. CD4+CD2S+ Tregs and CD4+CD25 T cells have equivalent PI3K function and levels of the phosphatases PTEN and SHIP. (A and C) Ex vivo CD4+ T cells were stimulated with aCD3/28 mAbs for the indicated times and MFIs following staining with anti-PIP3 Abs (A) or anti-phosphoSHIP (C) were determined in CD4+CD25hi8h or CD4+CD25" T cells. For A and C, a single experiment representative of at least 3 is depicted (p=NS), and the inset (A only) is the fold change in MFI from resting to activation (10 minutes) with each point representing a separate experiment. (B and D) Ex vivo, unstimulated CD4+ T cells were stained for CD4, CD25, and MFIs following staining with anti-SHIP (B) and anti-PTEN (D) were determined in CD4+CD25high or CD4+CD25_ T cells. B is the average of 3 experiments (p=NS) and (D) each point represents a separate experiment (p=NS). 87 A . B. • Resting — «CD3/CD28 CD4+CD25 CD4+CD25+ tr 10 to pAKT 10* 10 •o 10' 105 103 10' • Resting — otCD3/CD28 A CD4+CD25 l CD4+CD25+ 1 0 10 10 JO 10 0 10' 102 103 10* pS6 • Figure S3.3. Defective activation of phospho-AKT and S6 in CD4+CD25+ Treg cell lines. FACS-sorted CD4+CD25+ Tregs and CD4+CD25- T cells were stimulated with <xCD3/CD28 mAbs and stained for (A) phospho-AKT (Ser473) or (B) phospho-S6. A single representative experiment of 3 is depicted (A and B). 8 8 A. Pre-purification 10' 10= 10 a 10* 10 ' 1CP 10 3 10 ' pS6 • Figure S3.4. Transduction efficiency and relative purity of transduced T cell lines. (A) FACS-sorted CD4 + CD25 + Tregs and CD4 + CD25"T cells were transduced with control (pCLL) or AKT-ER-encoding lenti virus, and the percent of A N G F R + cell was determined after 6 days. Transduced cells were purified and stained intracellularly for HA expression to monitor expression of A K T - E R . (B) Transduced cells were stimulated with 4HT (luM) for 1 hr at 37°C, and stained for phosphor-S6. A single representative experiment of 4 is depicted (A and B). 89 3.6 R E F E R E N C E S 1. L e v i n g s M K , Bacchet ta R , S c h u l z U , Ronca ro lo M G . The role o f I L - 1 0 and T G F - b e t a i n the differentiation and effector funct ion o f T regulatory cells . Int A r c h A l l e r g y I m m u n o l . 2002;129:263-276. 2. Sakaguchi S. Na tu ra l ly ar is ing Foxp3-express ing C D 2 5 + C D 4 + regulatory T cel ls i n i m m u n o l o g i c a l tolerance to se l f and non-self. Na t I m m u n o l . 2005;6:345-352. 3. C o o m b e s J L , R o b i n s o n N J , M a l o y K J , U h l i g H H , P o w r i e F . Regula tory T cel ls and intestinal homeostasis. I m m u n o l R e v . 2005;204:184-194. 4. Shevach E M . Regulatory/suppressor T cel ls i n health and disease. Ar th r i t i s R h e u m . 2004;50:2721-2724. .5. B e l k a i d Y , S u n C M , B o u l a d o u x N . Parasites and immunoregula tory T cel ls . Cur r O p i n I m m u n o l . 2006. 6. Fontenot J D , Rudensky A Y . A w e l l adapted regulatory contr ivance: regulatory T ce l l development and the forkhead fami ly t ranscript ion factor F o x p 3 . N a t I m m u n o l . 2005;6:331-337. 7. Z i e g l e r S F . F O X P 3 : o f m ice and men . A n n u R e v I m m u n o l . 2006;24:209-226. 8. G a v i n M A , Torgerson T R , Hous ton E , et a l . S ing le -ce l l analysis o f no rma l and F O X P 3 - m u t a n t human T cel ls : F O X P 3 expression without regulatory T ce l l development. P roc N a t l A c a d S c i U S A . 2006;103:6659-6664. 9. B a e c h e r - A l l a n C , V i g l i e t t a V , Haf le r D A . H u m a n C D 4 + C D 2 5 + regulatory T cells . S e m i n I m m u n o l . 2004;16:89-98. 10. L e v i n g s M K , Sangregorio R , Sartirana C , et al . H u m a n C D 2 5 + C D 4 + T suppressor c e l l c lones produce t ransforming g rowth factor beta, but not in ter leukin 10, and are dist inct f rom type 1 T regulatory cel ls . J E x p M e d . 2002;196:1335-1346. 11. L e v i n g s M K , Sangregorio R , R o n c a r o l o M G . H u m a n C D 2 5 + C D 4 + T regulatory cel ls suppress naive and memory T - c e l l prol i fera t ion and can be expanded in vitro wi thout loss o f function. J E x p M e d . 2001;193:1295-1302. 12. L e v i n g s M K , Ronca ro lo M G . Phenotyp ic and funct ional differences between human C D 4 + C D 2 5 + and type 1 regulatory T cel ls . C u r r T o p M i c r o b i o l I m m u n o l . 20'05;293:303-326. 13. Can t re l l D A . T - c e l l antigen receptor s ignal transduction. Immuno logy . 2002; 105:369-374. 90 14. Schwar tz R H . T ce l l anergy. A n n u R e v Immuno l . 2003;21:305-334. 15. Bens inger SJ , W a l s h P T , Z h a n g J , et a l . Dis t inc t I L - 2 receptor s ignal ing pattern i n C D 4 + C D 2 5 + regulatory T cel ls . J I m m u n o l . 2004;172:5287-5296. 16. S u L , Creusot R J , G a l l o E M , et a l ! M u r i n e C D 4 + C D 2 5 + regulatory T cel ls fa i l to undergo chromat in remodel ing across the p r o x i m a l promoter region o f the I L - 2 gene. J I m m u n o l . 2004;173:4994-5001. 17. L i L , Godf rey W R , Porter S B , et a l . C D 4 + C D 2 5 + regulatory T - c e l l l ines f rom human cord b l o o d have functional and molecular properties o f T - c e l l anergy. B l o o d . 2005;106:3068-3073. . 18. I r ish J M , H o v l a n d R , K r u t z i k P O , et a l . S ing le c e l l p ro f i l i ng o f potentiated phospho-protein networks i n cancer cel ls . C e l l . 2004;118:217-228. 19. K r u t z i k P O , Ha le M B , N o l a n G P . Character izat ion o f the murine i m m u n o l o g i c a l s ignal ing ne twork w i t h phosphospecif ic f l o w cytometry. J Immuno l . 2005;175:2366-2373. 20. Sachs K , Perez O , Pe'er D , Lauffenburger D A , N o l a n G P . Causa l prote in-s ignal ing networks der ived from multiparameter s ingle-ce l l data. Science. 2005;308:523-529. 21 . L e v i n g s M K , Sangregorio R , Ga lb i a t i F , Squadrone S, de W a a l M a l e f y t R , Ronca ro lo M G . I F N - a l p h a and I L - 1 0 induce the differentiation o f human type 1 T regulatory cel ls . J I m m u n o l . 2001;166:5530-5539. 22. A m e n d o l a M , V e n n e r i M A , B i f f i A , V i g n a E , N a l d i n i L . Coordina te dual-gene transgenesis by lent iv i ra l vectors ca r ry ing synthetic b idi rect ional promoters. N a t B io t echno l . 2005;23:108-116. 23. K o h n A D , Bar the l A , K o v a c i n a K S , et a l . Cons t ruc t ion and characterizat ion o f a condi t iona l ly active vers ion o f the serine/threonine kinase A k t . J B i o l C h e m . 1998;273:11937-11943. 24. A l l a n S E , Passerini L , Bacchet ta R , et a l . The role o f 2 F O X P 3 isoforms i n the generation o f human C D 4 + Tregs. J C l i n Invest. 2005; 115:3276-3284. 25. K r u t z i k P O , N o l a n G P . Intracellular phospho-protein staining techniques for f l ow cytometry: moni to r ing single ce l l s ignal ing events. Cytomet ry A . 2003;55:61-70. 26. Perez O D . Apprec i a t i ng the heterogeneity i n autoimmune disease: mult iparameter assessment o f intracel lular s ignal ing mechanisms. A n n N Y A c a d S c i . 2005;1062:155-164. 27. Perez O D , K r u t z i k P O , N o l a n G P . F l o w cytometric analysis o f kinase s ignal ing cascades. Me thods M o l B i o l . 2004;263:67-94. 91 28. Perez O D , N o l a n G P . Simultaneous measurement o f mul t ip le active kinase states us ing po lychromat ic f l o w cytometry. N a t B io t echno l . 2002;20:155-162. 29. Perez O D , N o l a n G P . Phospho-proteomic immune analysis by f l o w cytometry: f rom mechan i sm to translational medic ine at the s ing le -ce l l l eve l . I m m u n o l R e v . 2006;210:208-228. 30. Okkenhaug K , B i l a n c i o A , E m e r y J L , Vanhaesebroeck B . Phosphoinosi t ide 3-kinase i n T ce l l ac t ivat ion and surv iva l . B i o c h e m Soc Trans. 2004;32:332-335. 31 . K a n e L P , Andres P G , H o w l a n d K C , A b b a s A K , Wei s s A . A k t provides the C D 2 8 cost imulatory signal for up-regulat ion o f I L - 2 and I F N - g a m m a but not T H 2 cytokines . N a t I m m u n o l . 2001;2:37-44. 32. A r i m u r a Y , S h i r o k i F , K u w a h a r a S, et a l . A k t is a neutral ampl i f ie r for T h ce l l differentiation. J B i o l C h e m . 2004;279:11408-11416. 33. Woodget t JR . Recent advances i n the protein kinase B s ignal ing pathway. C u r r O p i n C e l l B i o l . 2005;17:150-157. 34. Seminar io M C , Wange R L . L i p i d phosphatases i n the regulat ion o f T c e l l act ivat ion: l i v i n g up to their P T E N - t i a l . I m m u n o l R e v . 2003;192:80-97. 35. Fontenot J D , G a v i n M A , Rudensky A Y . Foxp3 programs the development and funct ion o f C D 4 + C D 2 5 + regulatory T cel ls . N a t Immuno l . 2003;4:330-336. 36. Roncador G , B r o w n P J , Maestre L , et a l . A n a l y s i s o f F O X P 3 prote in expression i n human C D 4 ( + ) C D 2 5 ( + ) regulatory T cel ls at the s ingle-ce l l l eve l . E u r J Immuno l . 2005;35:1681-1691. 37. K r y s t a l G . L i p i d phosphatases i n the immune system. S e m i n I m m u n o l . 2000;12:397-403 . 38. Stahl M , Di jkers P F , K o p s G J , et a l . The forkhead t ranscript ion factor F o x O regulates t ranscript ion o f p 2 7 K i p l and B i m i n response to I L - 2 . J Immuno l . 2002;168:5024-5031. 39. Can t re l l D . Pro te in kinase B ( A k t ) regulat ion and funct ion i n T lymphocytes . S e m i n I m m u n o l . 2002;14:19-26. 40 . G o n d e k D C , L u L F , Quezada S A , Sakaguchi S, N o e l l e R J . Cu t t ing Edge : Contact-M e d i a t e d Suppression by C D 4 + C D 2 5 + Regula tory C e l l s Involves a G r a n z y m e B-Dependent , Perforin-Independent M e c h a n i s m . J I m m u n o l . 2005; 174:1783-1786. 4 1 . Gros sman W J , V e r b s k y J W , To l l e f sen B L , K e m p e r C , A t k i n s o n JP , L e y T J . Dif fe rent ia l expression o f granzymes A and B i n human cyto toxic lymphocy te subsets and T regulatory cel ls . B l o o d . 2004;104:2840-2848. 92 42. de l a R o s a M , R u t z S, Dorn inger H , Scheffold A . Interleukin-2 is essential for C D 4 + C D 2 5 + regulatory T ce l l function. E u r J I m m u n o l . 2004;34:2480-2488. 43. L i u E , S inha S, W i l l i a m s C , et a l . Targeted deletion o f in tegr in- l inked kinase reveals a role i n T - c e l l chemotaxis and surv iva l . M o l C e l l B i o l . 2005;25:11145-11155. 44. B r a v o u V , K l i r o n o m o s G , Papadaki E , Taraviras S, V a r a k i s J. I L K over-expression i n human c o l o n cancer progression correlates w i t h act ivat ion o f beta-catenin, down-regula t ion o f E-cadher in and act ivat ion o f the A k t - F K H R pathway. J Pathol . 2006;208:91-99. 45 . N h o R S , X i a H , K a h m J , K l e i d o n J , D i e b o l d D , H e n k e C A . R o l e o f in tegr in- l inked kinase i n regulat ing phosphoryla t ion o f A k t and fibroblast surv iva l i n type I co l lagen matrices through a be ta l integrin v i ab i l i t y s ignal ing pathway. J B i o l C h e m . 2005;280:26630-26639. 46. F e n g J , Park J , C r o n P , Hess D , H e m m i n g s B A . Identif icat ion o f a P K B / A k t hydrophob ic m o t i f Ser-473 kinase as D N A - d e p e n d e n t . protein kinase. J B i o l C h e m . 2004;279:41189-41196. E p u b 42004 Ju l 41115. 47. T o k e r A , N e w t o n A C . Akt /p ro te in kinase B is regulated by autophosphorylat ion at the hypothet ical P D K - 2 site. J B i o l C h e m . 2000;275:8271-8274. 48. Sarbassov D D , Guer t in D A , A l i S M , Sabatini D M . Phosphoryla t ion and regulat ion o f A k t / P K B by the r i c t o r - m T O R complex . Science. 2005;307:1098-1101. 49. G a o T , Furnar i F , N e w t o n A C . P H L P P : a phosphatase that direct ly dephosphorylates A k t , promotes apoptosis, and suppresses tumor growth. M o l C e l l . 2005;18:13-24. 50. Janssens V , G o r i s J . Pro te in phosphatase 2 A : a h igh ly regulated fami ly o f serine/threonine phosphatases impl ica ted i n ce l l g rowth and s igna l l ing . B i o c h e m J. 2001;353:417-439. 51. Parsons M J , Jones R G , Tsao M S , Odermatt B , Ohash i P S , Woodget t JR . Express ion o f active prote in kinase B i n T cel ls perturbs both T and B ce l l homeostasis and promotes in f lammat ion . J Immuno l . 2001;167:42-48. 52. R a t h m e l l J C , E l s t r o m R L , C i n a l l i R M , T h o m p s o n C B . A c t i v a t e d A k t promotes increased resting T ce l l size, CD28- independent T ce l l growth, and development o f au to immuni ty and lymphoma . Et i r J I m m u n o l . 2003;33:2223-2232. 53. S a w i t z k i B , K i n g s l e y C I , O l i v e i r a V , K a r i m M , Herber M , W o o d K J . I F N - g a m m a product ion by alloantigen-reactive regulatory T cel ls is important for their regulatory funct ion i n v i v o . J E x p M e d . 2005;201:1925-1935. 54. M o o r e K W , de W a a l M a l e f y t R , C o f f m a n R L , O 'Ga r r a A . Inter leukin-10 and the in ter leukin-10 receptor. A n n u R e v I m m u n o l . 2001;19:683-765. 93 55. Fas S C , F r i t z sch ing B , Sur i -Payer E , K j a m m e r P H . Death receptor s ignal ing and its funct ion i n the i m m u n e system. C u r r D i r A u t o i m m u n . 2006;9:1-17. 56. Pace L , R i z z o S, P a l o m b i C , Brombacher F , D o r i a G . Cut t ing Edge : IL-4- Induced Protec t ion o f C D 4 + C D 2 5 - T h C e l l s f r o m . C D 4 + C D 2 5 + Regula tory T C e l l - M e d i a t e d Suppression. J I m m u n o l . 2006; 176:3900-3904. 94 4. C D 4 + C D 2 5 + Tregs express A q u a p o r i n 9 3 4.1 I N T R O D U C T I O N Unders tanding the peripheral regulat ion o f tolerance has been greatly advanced by the d iscovery and characterization o f C D 4 + C D 2 5 + T regulatory cel ls (Tregs), a t h y m i c a l l y -der ived subset o f C D 4 + T cel ls . C D 4 + C D 2 5 + Tregs are cr i t ica l regulators o f immune homeostasis and peripheral tolerance, and their absence results i n severe auto immune disease i n both human and mouse M . C D 4 + C D 2 5 + Tregs are hypo-prol i ferat ive to p o l y c l o n a l s t imul i i n v i t ro , do not produce any pro- inf lammatory cytokines such as I L - 2 , and suppress the ac t iva t ion and prol i fera t ion o f T effector cel ls i n a cell-contact dependant manner 4 . C D 4 + C D 2 5 + Tregs are currently best ident i f ied by their expression o f F O X P 3 , a t ranscript ion l 2 5 factor necessary for their development ' ' . C D 4 + C D 2 5 + Tregs const i tut ively express h i g h levels o f the I L - 2 R a , C D 2 5 , an act ivat ion marker that is also transiently up-regulated on activated T effector cel ls . T reg select ion methods based on the i so la t ion o f C D 2 5 + cel ls therefore result i n contaminat ion w i t h activated T effector cel ls , as has been previous ly described 4 ' 6 . Considerable efforts have been made to find a more specif ic marker o f Tregs, i nc lud ing microarray analysis o f both human and mouse Tregs. T o date, microarray analysis has ident i f ied a number o f 7 9 molecules expressed at higher levels on Tregs, such as G I T R , C T L A - 4 , and O X - 4 0 " . H o w e v e r , a l l o f the molecules are also act ivat ion markers for T effectors, and thus are not specif ic . Recent ly , neuropi l in-1 has been suggested to be speci f ica l ly expressed on murine C D 4 + C D 2 5 + Tregs 1 0 , however it has been also observed i n F O X P 3 - anergic T cells suggesting that it is not specific A d d i t i o n a l l y , the expression o f the I L - 7 R a ( C D 1 2 7 ) has 3 Manusc r ip t i n preparation 95 be proposed for T reg selection, as C D 4 + C D 2 5 + Tregs have lower levels o f C D 127 than do C D 4 + C D 2 5 " T effector cel ls 6 . H o w e v e r , the expression o f C D 1 2 7 on T effector cells decreases after act ivat ion, suggesting that select ion o n the basis o f C D 127 select ion is s t i l l i m p e r f e c t 1 2 . M i c r o a r r a y analyses performed thus far used resting, p o l y c l o n a l populat ions o f C D 4 + C D 2 5 + Tregs, w h i c h were l i k e l y contaminated w i t h activated T effector ce l ls . W e felt that the on ly way to re l iably el iminate the p rob lem o f effector ce l l contaminat ion was to perform an analysis o f T reg s ingle-ce l l clones, each tested funct ional ly to ensure a suppressive phenotype. Thus , we performed microarray analysis on resting or activated Treg clones compared to pooled T effector clones f rom the same donor. U s i n g this nove l approach, we observed that C D 4 + C D 2 5 + T r eg clones expressed h i g h levels o f aquaporin-9 ( A Q P 9 ) . A Q P 9 is an aquaglyceropor in that permits the passage o f water and uncharged solutes such as water and g lyce ro l 1 3 ' 1 4 . It has been reported that A Q P 9 is expressed i n l iver , brain , testis, placenta, peripheral leukocytes and neutrophils, osteoclasts and their myelogenous precursors, but has not been reported on T cel ls 1 3 ' 1 5 " 2 0 . Here w e report the prev ious ly undiscovered selective expression o f A Q P 9 on human C D 4 + C D 2 5 + Tregs. 4.2 M A T E R I A L S A N D M E T H O D S C e l l purification. Per ipheral b lood was obtained f rom healthy volunteers fo l lowing informed consent and approva l o f the p ro toco l by the U B C C l i n i c a l Research Eth ics Boa rd . A s p rev ious ly described 2 1 ' 2 2 , P B M C s were isolated b y densi ty centrifugation over F i c o l l (Stemcell Technologies) , and C D 4 + T cells were subsequently pur i f ied b y negative selection w i t h magnetic beads ( M i l t e n y i B i o t e c h or Stemcell Technologies) . C D 2 5 + ce l ls were either 96 pur i f i ed by posi t ive select ion ( M i l t e n y i B io t ech ) and passed over 2 M S co lumns to ensure 9 0 - 9 5 % puri ty, or F A C S sorted. The f low-through f rom the C D 2 5 + posi t ive select ion was then passed over an L D deplet ion co lumn to remove any C D 2 5 1 0 cells, and used as the C D 2 5 " fraction, as p rev ious ly d e s c r i b e d 2 I ' 2 2 . T cel l c lon ing . Pur i f ied C D 4 + T cells were F A C S sorted into C D 4 + C D 2 5 " or C D 4 + C D 2 5 h i g h (gating o n 1-2% brightest C D 2 5 + ) . Ce l l s were then cloned b y l imi t ing d i lu t ion , and plated at a concentration o f 0.3 ce l ls /wel l i n a 96 w e l l round bo t tom plate i n T cel l med ium ( X v i v o - 1 5 , 5% human serum [Nor th B i o ] , l x pen ic i l l in / s t rep tomycin [Invitrogen], l x Glutamax [Invitrogen]), i n the presence o f rh IL-2 ( l O O U / m l , Ch i ron) , as p r ev ious ly described 2 3 . C l o n i n g efficiency was determined by addi t ion o f [ 3 H] thymid ine (1 u C i per w e l l ; A m e r s h a m Biosc iences) to one 96 w e l l plate. Clones were re-stimulated every 14 days as p r ev ious ly described 4 and selected o n the basis o f C D 2 5 expression ( in the resting phase), i n vi t ro anergy and suppress ion 4 ' 2 2 . Microarrays . R N A was extracted f rom C D 4 + C D 2 5 " T effector or C D 4 + C D 2 5 + T reg clones either at rest or fo l lowing act ivation for 24 hours (plate-bound a C D 3 and soluble a C D 2 8 ( l u g / m l each) i n assay med ium ( X v i v o - 1 5 , 10% F C S (Gibco) , 1% human serum [Nor th B i o ] , l x pen ic i l l in / s t rep tomycin [Invitrogen], l x Glu tamax [Invitrogen]). T r i z o l (Invitrogen) was used for R N A extraction, and further pur i f ica t ion was performed using R N e a s y k i t (Qiagen) clean-up p ro toco l . R N A (0.5-1 u.g) was then ampli f ied us ing Message A m p antisense ( a R N A ) k i t ( A m b i o n ) . a R N A (1 \xg) was then labeled w i t h C y 3 or C y 5 using C y S c r i b e Di rec t m R N A labell ing k i t (Amersham) , and pur i f ied using C y S c r i b e G F X puri f icat ion k i t 97 (Amersham) . Label l ing w i t h C y 3 or C y 5 alternated between biological replicates to account for potent ial dye effects. Probes were then concentrated using M i l l i p o r e M i c r o c o n Y M - 3 0 filters, and hyb r id i z e d fo l lowing manufacturer 's recommended p ro toco l (Amersham) . In short, C y 3 and C y 5 labelled samples were h y b r i d i z e d for 14-16 hr at 4 2 ° C i n h u m i d condi t ions i n hybr id i za t ion buffer (Amer s h am C y S c r i b e Direc t k i t ) and formamide. Pre-warmed buffers ( 4 2 ° C ) were used to wash slides l x ( l x S S C , 0 .2% S D S ) and 2x wi th ( O . l x S S C , 0 .2% S D S ) for 10 m i n at r o o m temperature o n a rotary shaker, p r io r to a final (10 seconds) wa sh i n water. Slides used were H u m a n Operon v.2.1 (21, 3 2 9 K ) glass arrays produced (based o n human 70mers f rom Operon , Hun t sv i l l e , A L ) by the M i c r o a r r a y Fac i l i t y o f T h e Prostate Centre at V a n c o u v e r Genera l Hosp i t a l , Vancouve r , Canada . F o l l o w i n g overnight hybr id iza t ion and washing , arrays were imaged using a S c a n A r r a y Express scanner ( P e r k i n E l m e r , Bos ton , M A ) . M i c r o a r r a y analysis. Scanned images were processed us ing Imagene software. B i o l o g i c a l replicates were paired, and GeneSpr ing software used to per form L o w e s s normal iza t ion . Gene lists were created us ing a filter set on expression level (>200) and no rma l i zed values i n both resting and activated condit ions. B i o m e t r i c rank expression analysis was performed on the mean f o l d expression ratio o f Treg c lone :T effector poo led clones. Quantitative R T - P C R . F o r quantitative R T - P C R , amounts o f target m R N A were determined us ing Sybr green on an A B I 5700 R e a l T i m e P C R machine. A l l samples were run i n t r ipl icate, and relative expression o f target gene was determined by n o r m a l i z i n g to G A P D H i n order to calculate a fold-change i n value. O l i g o s were as described i n Table 4 .1 : 98 Target Sense Anti-sense LEFl 5' CGA AGA GGA AGG CGA TTT AG 3' 5' TCC TGA GAG GTT TGT GCT TG 3' LILRA2 5' GTG ACA GGA GCC TAC AGC A A 3' 5' CGT TGT GGG TGT TCA TCT TC 3' ITGAM 5' TCA GAG TCT GCC TCC ATG TC 3' 5' GCG TGT GCT GTT CTT TGT CT 3' MA4A 5' TGG TGC TCC TCC TAA GTG TG 3' 5' TGG CAG AAT TAA CAC AAC CC 3' S100A9 5' GGG AAT TCA AAG AGC TGG TG 3' 5' GAA GCT CAG CTG CTT GTC TG 3' TGFpT 5' ATC TCC ACC ATC ACC AAC A A 3' 5' TGT ACT GGC CGT TAC CTT CA 3' CSF1R 5' GAG CAC AAC CAA ACC TAC GA 3' 5' TGT GAA GAG GAA CTC ATC CG 3' AQP9 5' CTC TGG TGG TCA CAT CAA CC 3' 5' ACA AAG GCT CCC AAG AAC TG 3' CD 166 5' AGC ATG ATT GCT TCA ACA GC 3' 5' CCA TAC CAC AGT TGC ATT CC 3' Goliath 5' GGA TCC CTG GCT TAG TGA AC 3' 5' GTT CTG GTG AGC CTT TCC AT 3' PU.l 5' AAG ACC TGG TGC CCT ATG AC 3' 5' AAG TCC CAG TAA TGG TCG CT 3' PTDGR 5' TGA TGA CCG TGC TCT TCA CT 3' 5' CGG AGG TCT TCT GCT TCT TC 3' KSP37 5' AGT CCT TTC CCA GCT GTG TT 3' 5' CCA TAT GCA GCA GGT GAC TT 3' PTGS2 5' CCA CTC AAG TGT TGC ACA TAA TC 3' 5' ACA GGA GCA TCC TGA ATG G 3' EOMES 5' TGG GAT TGA GTC CGT TTA TG 3' 5' CTC TGT GGC TCA AAT TCC AC 3' ENTPD1 5' GTG GAG TGG GAG AGA GGT GT 3' 5' GGA GCA CAT CCA TTT CAT TG 3' CD6 5' GAT GGG TAC CAC TCA CTG TCA 3' 5' CTA CTG CGG CCA CAA AGA G 3' HRH1 5' GGA TGT TCA TAG GCA TGA CG 3' 5' ACA GTA GGG CTC AAC CTG CT 3' S100A8 5' GGA CAC TCG GTC TCT AGC A A 3' 5' GCT GGA GAA AGC CTT GAA CT 3' FCER1G 5' TGT GGT GGT TTC TCA TGC TT 3' 5' CTG AAG ATC CAA GTG CGA A A 3' VLA1 5' ACA TCT GTC AGA CCG TCA CC 3' 5' GCA CAA CGT ATT CCA TCA GG 3' ENV127 5' TCC ATT TAC CAC CAG AGG GT 3' 5' GAT TTG TGT GAC GGG CAA 3' GAPDH 5' TTG CCA TCA ATG ACC CCT TC 3' 5' GTT CTC AGC CTT GAC GGT GC 3' 99 Table 4 .1. Primer sequences used for quantitative RT-PCR. Primers were designed using Primer Express software. Western blotting. H i g h l y pur i f ied C D 4 + C D 2 5 + T reg or C D 4 + C D 2 5 " T cel ls were lysed by sonicat ion i n lys i s buffer containing 1% S D S , l O m M H E P E S , and 2 m M E D T A ( p H 7.4). Lysates f rom ~3 m i l l i o n cel ls were quantitated by B C A (Pierce), separated by S D S - P A G E gel electrophoresis, transferred to ni t rocel lulose membranes, and immunoblo t ted w i t h anti-A Q P 9 A b s ( A l p h a Diagnost ics) . Membranes were re-probed w i t h anti-p38 A b s (Santa C r u z Bio tech) to determine equivalency o f load ing . Lenti-viral vectors. The bidi rec t ional l en t i -v i ra l • vector ( p C C L . s i n . c P P T . S V 4 0 p o l y A . C T E . m C M V A N G F R . P G K ) 2 4 was mod i f i ed by replac ing the P G K promoter w i t h the human E F l c t promoter to improve transgene expression i n human T cel ls . A c D N A cassette containing human aquaporin 9 a long w i t h an hemagglu t in in ( H A ) epitope tag at its c a rboxy l terminus 2 5 , was inserted downstream o f the E F l c t promoter . The resul t ing cont ro l ( p C C L ) and p C C L . A Q P 9 ( A Q P 9 ) vectors both dr ive const i tut ive expression o f A N G F R as a marker gene. V S V - p s e u d o t y p e d third-generation lenti-viruses were produced by transient four-p lasmid co-transfection into H E K 2 9 3 T cel ls and concentrated by ultracentrifugation 2 4 . Expres s ion titers were determined on H E K 2 9 3 T cel ls by l i m i t i n g ft 7 d i lu t ion and ranged f rom 7x10 to 1.5x10 transducing uni t s /ml . Generation of T cell lines and lenti-viral transduction. C D 4 + T cells were stained for C D 4 and C D 2 5 and F A C S sorted into C D 2 5 h i and C D 2 5 " fractions o n a B D F A C S A r i a as described 4 . T cells (150,000/well) were activated b y s t imulat ion w i t h a C D 3 (1 pg/ml) and A P C s (500,000/wel l irradiated 50 G y ) for 24 hours i n T cel l med ium ( X v i v o - 1 5 , 5% human serum [Nor th B i o ] , l x pen ic i l l in / s t rep tomycin [Invitrogen], l x Glu tamax [Invitrogen]), i n the presence o f r h I L - 2 ( l O O U / m l , Chi ron) . C o n t r o l p C C L or p C C L A Q P 9 encoding lenti-virus 100 was added to the activated cel ls at a mu l t i p l i c i t y o f infect ion o f 5. The % o f A N G F R + -transduced T cel ls was moni tored after 6 days and was routinely 20-40%. A N G F R + cel ls were pur i f i ed w i t h a n t i - N G F R magnetic beads ( M i l t e n y i B io tech) f o l l o w i n g the manufacturer 's instructions and a l l o wed to expand. Equiva len t expression o f the A Q P 9 prote in i n the var ious c e l l l ines was conf i rmed by f l o w cytometr ic analysis us ing an a n t i - H A -F I T C m A b (Roche) as described 2 6 . Pu r i f i ed C D 4 + C D 2 5 + Treg and C D 4 + C D 2 5 " T cel l lines were re-stimulated every 14 days as previous described 4 and moni tored at the end o f every cyc le to ensure preservation o f their suppress ive capaci ty. Proliferation and suppression of T cells. T ce l l clones or transduced T cel ls were plated at 50,000 ce l l s /we l l i n 9 6 - w e l l plates, and st imulated w i t h soluble c t C D 3 m A b s (1 pg /ml ) i n the presence o f A P C s (CD3-deple ted P B M C s , irradiated at 50 G y , 50,000 ce l l s /wel l ) . Pro l i fe ra t ion was assessed after 72 hours, by addi t ion o f [ 3 H]thymid ine (1 u C i per w e l l ; A m e r s h a m Biosc iences) for the f inal 16 hours o f the assay. T o test for suppressive capacity C D 4 + T cel ls (50,000 ce l l s /wel l ) or C D 4 + T ce l l clones (25,000 ce l l s /wel l ) were stimulated w i t h a C D 3 / A P C s , i n absence or presence o f transduced T cel ls (50,000 ce l l s /wel l ) . Suppress ion was assessed by determining [ 3 H]thymid ine incorporat ion after 72 hours. In a l l cases untransduced C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T ce l l l ines were ana lyzed i n paral lel and found not to be different f rom control vector-transduced T ce l l l ines (data not shown). Cytokine production. Transduced T ce l l l ines were plated i n a 48 w e l l plate ( l x l 0 6 / m l ) i n assay m e d i u m and stimulated w i t h a C D 3 / 2 8 beads (Dyna l ) at a ratio o f 1 bead: 16 cells . Supernatants were col lected at 24 or 48 hours and the T h l / T h 2 cytometr ic bead assay ( B D Biosc iences ) was used to determine amounts o f cytokine. 101 Statistical analyses. A l l analysis for statistically significant differences was performed with the student's paired t test, p values less than 0.05 were considered significant. A l l cultures were performed in triplicate and error bars represent the SD. 4.3 R E S U L T S Generation and screening of human CD4+CD25+ Treg and CD4+CD25" T effector single cell clones Selection on the basis of CD25 expression greatly increases the frequency of Tregs in the resultant population, but a significant number of contaminating Te will remain 4 . In order to ensure pure, homogeneous populations of Tregs and T effector cells, we FACS sorted CD4 + T cells into CD25 h i g h (1-2% brightest) and CD25" populations, then used limiting dilution to achieve clonal populations, as previously described (Figure 4.1)4. m <N Q O 2.9% . WHS 63.G%; Single cell cloning CD4 R2 0 © 0 0 0 © CD4 +Tcell clones CD4 Figure 4.1. Single cell cloning of C D 4 + T cells. Ex vivo CD4+ T cells were stained with CD4 and CD25 and FACs sorted into populations of CD4+CD25high or CD4+CD25" T cells, from which limiting dilution produced single cell-derived CD4+ T cell clones. As the expression of CD62L had been suggested to define a more potent population of Tregs 2 7 ' 2 8 , we also sorted and cloned CD4 + CD25 h i g h CD62L h i g h or CD4 + CD25 h i g h CD62L" T cells. 102 Sing le -ce l l clones were expanded us ing a feeder ce l l mixture as prev ious ly reported 4 , and then screened o n the basis o f C D 2 5 expression (Table 4.2). CD25 CD25 + CD25 + CD62L CD25 + CD62L + Clone # CD25 M F I Clone # CD25 M F I Clone # CD25 M F I Clone # CD25 MFI 1 432 3 1874 5 1005 2 186 2 625 4 3039 10 2299 7 3113 3 790 5 2100 18 162 9 1511 4 686 •7 3141 26 411 11 2161 5 175 9 3329 34 3580 15 2482 6 742 15 2745 42 2750 16 2187 7 817 16 2821 44 5056 19 2833 8 514 18 2259 47 2546 19b 1480 9 1288 20 4601 49 3672 22 1644 10 129 34 1850 50 1779 24 2820 11 455 38 274 56 1848 27 1471 12 118 40a 2543 66 1177 30 2973 13 322 40b 1120 70 1220 31 2199 14 106 62 2852 78 1799 32 1367 15 194 69 1740 79 3449 36 3083 16 634 71 3209 88 4953 40 2438 17 300 72 2302 42 2015 18 343 • 7 ? . 3335 50 2022 19 828 79 1345 51 2410 20 435 80 2308 53 1421 21 344 84 133 54 1614 22 702 92 3855 56 1341 23 295 102 1027 24 444 Table 4.2. CD25 expression on C D 4 + T cell clones. Resting CD4+ T cell clones were stained for CD25 expression, and mean fluorescence intensity ( M F I ) is entered beside clone numbers. Clones selected for microarray analysis are indicated in bold type. Clones w i t h a h igh C D 2 5 M F I were then assayed for anergy (Figure 4.2a) and suppressive capacity (Figure 4.2b) i n vi t ro. Unfortunately, many p romis ing clones expressing h i g h levels o f C D 2 5 either d id not survive i n vi t ro culture, or d i d not proliferate i n numbers sufficient to assay. O f more than 300 clones in i t i a l ly screened, we ult imately 103 selected 2 T reg clones that expressed h igh levels o f C D 2 5 (indicated i n b o l d type in Table 4.2), and were consistently anergic and suppressive in vitro (Figure 4.2). The 3 T effector clones selected were not anergic or suppressive i n v i t ro (Figure 4.2), and inc luded one clone that was C D 2 5 h l g h at the t ime o f in i t ia l sorting and was thus a contaminat ing activated T effector c e l l . W e d i d not find sorting on the basis o f C D 2 5 h i g h C D 6 2 L + expression to improve the percentage o f Treg clones generated, as compared to C D 2 5 h l g h or C D 2 5 h l g h C D 6 2 L + , a l though we d i d observe a higher number o f T ce l l clones emerge f rom the C D 2 5 h l g h C D 6 2 L + popula t ion . 104 60 40 J x 0) c -a E 20 * * j _ ft i 1 1 i 1 r—— i CD25* CD25+ CD62L" CD62L+ CD25+ CD62L* CD25" CD25" CD25* CD62L" #34 #40a #42 #86 #7 #94 #5 #10 #102 #18 CD25+ CD25+ CD62L" CD62L+ CD25+ CD62L+ CD25" CD25" CD25+ CD62L" #34 #40a #42 #86 #7 #94 #5 #10 #102 #18 100 • * ft ft •400 Figure 4.2. Screening of human C D 4 + C D 2 5 + Treg and C D 4 + C D 2 5 T effector single cell clones. (A) CD4+ T cell clones were stimulated with aCD3 (lug/ml) and irradiated APCs and proliferation was assessed. (B) Autologus CD4+ T cells were stimulated with aCD3 (lug/ml) and irradiated APCS in the absence or presence of CD4+ T cells clones (1:1 ratio). Data are depicted as percent suppression ([l-(Te+Treg/Te alone]*100). (A&B) Proliferation and suppression were assessed by 3H-thymidine incorporation, and a single representative experiment of 3 is depicted. Clones marked with a star were used for microarray experiments. 105 Microarray analysis of human CD4+CD25+ Treg and CD4+CD25 T effector clones R N A was extracted f rom 2 T r e g clones and 3 T effector clones, either i n the resting phase or 24 hours after act ivat ion w i t h a C D 3 / C D 2 8 (each 1 ug/ml) . R N A f rom the 3 selected T effector clones was pooled pr ior to ampl i f i ca t ion and subsequent hybr id iza t ion . The R N A f rom each i nd iv idua l T reg clone was hyb r id i zed w i t h pooled T effector clones, either i n the resting phase or f o l l o w i n g act ivat ion. A s the purpose o f these microarrays was as an in i t ia l screen for potential unique ce l l surface markers o f C D 4 + C D 2 5 + Tregs, 2 b i o l o g i c a l replicates were performed for each clone, and rank expression analysis was performed. A subset o f the genes observed to be either up- or down-regulated i n Tregs, as compared to T effector cel ls , i n both resting and activated condit ions is shown i n Table 4.3. In general, T r eg clones expressed lower amounts o f cytokine m R N A s , such as I L - 4 and I F N -Y, and had decreased expression o f S T A T 4 , suggesting a non T h l / T h 2 phenotype, consistent w i t h previous reports (Table 4.3)29. A d d i t i o n a l l y , T reg clones had altered expression o f a variety o f adhesion molecules and increased expression o f H L A - D R A and H L A - D R M (Table 4.3). In agreement w i t h previous reports, we observed that T reg clones expressed increased levels o f galect in 3 ( L G A L S 3 ) , a molecu le i n v o lv e d i n the control o f apoptosis that has been proposed as a mediator i n the induc t ion o f T r l cel ls by C D 4 + C D 2 5 + Tregs 2 9 ' 3 0 . Unfor tunate ly , F O X P 3 expression was b e l o w the leve l o f detection, and so c o u l d not be compared. Other T reg associated molecules , such as C D 2 5 and G I T R , were not expressed at higher levels i n T reg clones. Th i s cou ld be a result o f the increased expression o f act ivat ion markers i n T effector ce l l l ines associated w i t h repeated i n v i t ro e x p a n s i o n 2 6 , or reflect a lack o f t ranscript ional act ivi ty as previous ly described 8 . Considerable var ia t ion i n expression 106 existed between T reg clones for some genes, though others were consistent, ref lect ing the c lona l nature o f the cel ls . Fold expression in Tregs as compared to T effectors Resting Activated Clone #40a Clone #42 Clone #40a Clone #42 LEF1 AF288571 5.33 (4.07-6.98) 5.78 (2.61-12.8) 6.54 (4.27-10) 3.24 (2.78-3.8) LILRA3 NM 006865 6.28 (6.28-*) 3.16 (1.52-6.35) 5.87 (4.9-7.04) 9.03 (7.87-10.4) LILRA2/ILT-1 NM 006866 77.19 (71.8-83) 3.34 (1.92-5.82) 5.67 (5.23-6.14) 7.93 (3.05-20.64) MS4A4A NM 016650 15.05 (10.06-22.51) 13.54 (4.51-40.57) 10.07 (6.83-14.84) 10.80 (8.6-13.56) MRC1 NM 002438 134.17 (71.7-250.9) 3.38 (1.56-7.32) 10.02 (3.06-32.85) 12.53 (9.25-16.98) TIMP2 AL110197 12.66 (8.45-19) 4.15 (3.68-4.67) 8.91 (4.45-17.82) 3.63 (1.63-8.1) AVPI1 NM 021732 17.68 (14.76-21.2) 8.39 (5.34-13.2) 4.34 (3-6.26) 3.45 ' (2.69-4.42) ITGAM/MAC 1/CDllb NM 000632 13.86 (3.9-49) 3.74 (3.31-4.2) 4.91 (3.65-6.61) 4.34 (4.1-4.6) S100A8/Calgr anulin A NM 002964 78.65 (54.05-114.4) 6.76 (4.16-10.97) 14.10 (5.66-35.1) 11.09 (8.48-14.5) RAS ALI NM 004658 4.07 (3.6-4.6) 3.08 (2.32-4.08) 10.58 (2.48-45.15) 3.12 (1.94-5) S100A9/ CalgranulinB NM 002965 8.53 (7.43-9.77) 3.83 (2.38-6.15) 7.96 (3.36-18.9) 5.04 (4.67-5.43) TGFBI NM 000358 8.43 (4.89-14.54) 13.91 (5.06-38.25) 10.42 (5.02-21.63) 10.94 (10.35-11.6) LGALS3/ Galectin 3 NM 002306 7.13 (4.66-10.91) 3.72 (2.59-5.35) 7.86 (6.25-9.88) 8.62 (4.25-17.5) CSF1R NM 005211 55.66 (39.42-78.6) 3.35 (1.79-6.28) 9.43 (3.52-25.2) 18.92 (13.25-27) PU.l NM 003120 34.47 (11.06-107) 6.48 (3.6-11.65) 5.27 (4.9-5.67) 3.17 (2.2-4.53) FCER1G NM 004106 12.58 (7.3-21.66) 3.74 (2.67-5.23) 5.71 (2.56-12.76) 3.33 (3.07-3.6) PTGDR U31099 22.60 (22.6-22.6) 6.82 (1.24-37.7) 3.50 (2.45-5.01) 18.8 (18.8-18.8) FABP4 NM 001442 80.25 (43.59-148) 8.75 (7.03-10.89) 19.66 (8.04-48) 18.59 (13.73-25.2) A2M NM 000014 18.85 (6.61-9.17) 7.60 (6.29-9.17) 15.66 (9.19-26.71) 6.20 (3.25-1 1.84) AQP9 NM 020980 24.44 (8.88-67.3) 4.23 (3.47-5.15) 5.56 (4.17-7.39) 28.43 (27.97-28.9) memD/ CD166 Y10183 5.86 . (4.13-8.31) 3.52 (3.49-3.55) 4.57 (3.45-6.05) 7.06 (5.85-8.51) 107 Fold expression in Tregs as compared to T effectors Resting Activated Clone #40a .Clone #42 Clone #40a Clone #42 LEF1 AF288571 5.33 (4.07-6.98) 5.78 (2.61-12.8) 6.54 (4.27-10) 3.24 (2.78-3.8) LILRA3 NM 006865 6.28 (6.28-*) 3.16 (1.52-6.35) 5.87 (4.9-7.04) 9.03 (7.87-10.4) HCK NM 002110 6.71 (5.44-8.26) 7.12 (4.81-10.54) 5.39 (5.39-5.39) 10.11 (7.66-13.34) RNF130/ Goliath NM 018434 39.95 (11.77-135) 12.73 (10.9-14.79) 16.16 (5.15-50.6) . 9.41 (7.72-11.46) IFNG NM 000619 0.26 (0.26-0.26) 0.14 (0.057-0.32) 0.03 (0.021-0.029) 0.07 (0.03-0.16) EVI27 AF208111 . 0.24 (0.16-0.36) 0.08 (0.05-0.12) 0.17 (0.12-0.25) 0.16 (0.14-0.18) 1L4 NM 000589 0.11 (0.032-0.35) 0.18 (0.12-0.27) 0.05 (0.05-0.05) 0.05 (0.01-0.27) VLA1 X68742 0.11 (0.08-0.16) 0.06 (0.06-0.07) 0.26 (0.07-0.94) 0.11 (0.11-0.11) CCL5/ RANTES NM 002985 0.23 (0.12-0.42) 0.14 (0.12-0.16) 0.16 (0.14-0.19) 0.08 (0.05-0.13) pcta-1 L78132 0.16 (0.09-0.3) 0.23 (0.22-0.24) 0.32 (0.19-0.56) 0.31 (0.31-0.31) KSP37 NM 031950 0.05 (0.01-0.18) 0.08 (0.04-0.18) 0.06 (0.02-0.15) 0.09 (0.06-0.13) PTGS2 NM 000963 0.08 (0.06-0.1) 0.09 (0.07-0.1) 0.19 (0.8-0.48) 0.20 (0.11-0.37) EOMES NM 005442 0.25 (0.14-0.42) 0.05 (0.02-0.11) 0.05 (0.44-0.06) 0.22 (0.09-0.5) ENTPD1 NM 001776 0.10 (0.04-0.28) 0.25 (0.15-0.41) 0.23 (0.08-0.63) 0.22 • (0.17-0.27) CD6 X60992 0.17 (0.09-0.31) 0.22 (0.19-0.25) 0.27 (0.27-0.27) 0.28 (0.14-0.57) Table 4.3. Microarray analysis of human CD4+CD25+ Treg and CD4+CD25" T effector clones. Microarray analysis was performed on resting and activated CD4+CD25+ Treg and CD4+CD25" T effector clones. An abbreviated list of differentially expressed genes is depicted, with normalized fold expression values for each clone. Activated clones were stimulated for 24 hours with plate-bound aCD3 (lug/ml) and soluble aCD28 (lug/ml) prior to RNA extraction. Normalized mean fold expression values incorporate 2 biological replicates in which labelling with Cy3 and Cy5 was reversed, and values in brackets represent the range between the replicates. Gene names that are in bold were chosen for qPCR validation. Quantitative PCR Validation of Microarray Results In order to be considered as a potential c e l l surface marker for C D 4 + C D 2 5 + Tregs, genes were required to be expressed i n human T r e g clones more than 3 fo ld higher or lower as compared to T effector clones. O f the 122 genes that met this expression l eve l cr i ter ion, 21 108 were selected for va l ida t ion us ing quantitative P C R ( q P C R ) on p o l y c l o n a l populat ions o f ex v i v o C D 4 + C D 2 5 + Tregs. Selec t ion for further va l ida t ion was made o n the basis o f ce l l surface expression, consistency between c lones , and relevance to T ce l l b i o l o g y , and inc luded molecules that were both up- and down-regulated on C D 4 + C D 2 5 + T r eg clones by microar ray . Spec i f i c pr imers were designed for each va l ida t ion target, and q P C R analyses were carr ied out o n R N A der ived f rom p o l y c l o n a l populat ions o f ex v i v o C D 4 + C D 2 5 + Tregs and C D 4 + C D 2 5 " T effector cel ls (Figure 4.3). T h e most consistently up-regulated genes i n C D 4 + C D 2 5 + Tregs inc luded the calgranul ins S 1 0 0 A 8 and S 1 0 0 A 9 , the adhesion molecule I T G A M / C D 1 l b , and the channel protein aquaporin-9 ( A Q P 9 ) (Figure 4.3). Genes that were consistently expressed at lower levels i n C D 4 + C D 2 5 + Tregs than i n C D 4 + C D 2 5 " T effector ce l l s inc luded the adhesion molecu le V L A - 1 , the T - b o x gene E O M E S w h i c h is associated w i t h C D 8 d e v e l o p m e n t 3 1 , and k i l l e r specif ic secretory protein ( K S P 3 7 ) , a protein associated w i t h T h i cel ls and cyto toxic cel ls 3 2 (Figure 4.3). Considerable var ia t ion was found i n the q P C R results between donors, w h i c h was l i k e l y reflective o f the po lyc lona l nature o f the pur i f i ed cel ls , and potential contaminat ion w i t h activated effector T cel ls . 109 c g m CL X LU 0> 5 cu K 0) O) c x: O o LL. 100 -, 80 60 10 9 8 -7 -6 5 4 A 3 • Donor 1 • Donor 2 • Donor 3 1 4 4 M Up-regulated on microarray Down-regulated on microarray Figure 4 .3. Quantitative PCR validation of microarray results. The fold change in expression of genes indicated was determined by quantitative RT-PCR. Values plotted are the fold change of genes in CD4+CD25+ Treg/CD4+CD25" Te, where CD4+CD25" Te equal 1 . Relative expression is normalized to GAPDH expression. Standard deviation for each bar is < 1 0 % . Results from 3 different donors are shown. H u m a n C D 4 + C D 2 5 + Tregs express more A Q P 9 m R N A as compared to C D 4 + C D 2 5 T effector cells O f the genes chosen for va l ida t ion by q P C R analysis, A Q P 9 appeared to be the most p romis ing candidate for a c e l l surface marker o f human C D 4 + C D 2 5 + Tregs . A Q P 9 is a c e l l surface membrane channel protein that permits the passage o f water, g l y c e r o l , and urea 1 3 . W e compared the expression o f A Q P 9 i n human ex v i v o C D 4 + C D 2 5 + Tregs to C D 4 + C D 2 5 " Te i n a number o f donors, and found that A Q P 9 message was expressed ~ 75 fo ld (range 35-100, p= 0.001, n=7) higher i n Tregs (Figure 4.4a). The differential expression o f A Q P 9 110 message was also observed i n in vitro expanded c e l l l ines i n the resting phase (Figure 4.4b). Pre l imina ry evidence suggests that A Q P 9 is also expressed at higher levels i n murine C D 4 + C D 2 5 + Tregs than C D 4 + C D 2 5 " T effector cel ls (data not shown), suggesting that increased expression o f A Q P 9 is not unique to human C D 4 C D 2 5 Tregs. A. CD Q_ O < c o to CL X 111 100 m 75 50 •c 25 JS CC 4JUUULJ-CO D_ O < c o '</> 10 2 a i3 800 600 400 to 200 ID a. CD25- CD25+ CD25-cell line CD25+ cell line Figure 4.4. Human CD4+CD25+ Tregs express more AQP9 mRNA as compared to CD4+CD25" T effectors. (A) RNA was isolated from ex vivo purified CD4+CD25+ Treg and CD4+CD25"T cells or (B) CD4+CD25+ Treg and CD4+CD25" T effector cell lines and expression of AQP9 was determined by quantitative RT-PCR. (A) Each point represents a specific experiment, and (B) one representative experiment of three is shown. (A&B) Values plotted are normalized by GAPDH expression. AQP9 is not an activation marker for human CD4+CD25" T effector cells A s ment ioned above, the major f l aw w i t h the use o f C D 2 5 as a c e l l surface marker for Tregs is that it is also transiently up-regulated o n activated effector T cel ls . Therefore, it was cr i t i ca l to assess the expression o f A Q P 9 f o l l o w i n g act ivat ion, to ensure that this marker was also not correlated w i t h act ivat ion i n T effector cel ls . Pur i f i ed C D 4 + C D 2 5 + Tregs and C D 4 + C D 2 5 " T effector cel ls were activated w i t h plate-bound a C D 3 (1 ug/ml) and soluble a C D 2 8 ( l u g / m l ) for 1, 2, or 5 days i n the presence o f m i n i m a l I L - 2 ( l O U / m l ) (Figure 4.5). A Q P 9 m R N A levels were not found to increase i n C D 4 + C D 2 5 " T effector cel ls f o l l o w i n g T C R s t imula t ion (p=NS, n=3) (Figure 4.5), despite significant upregulat ion o f C D 2 5 , 111 ind ica t ing that T effector cel ls were strongly activated (data not shown). Interestingly, levels o f A Q P 9 m R N A i n C D 4 + C D 2 5 + Tregs were decreased f o l l o w i n g T C R act ivat ion (Figure 4.5). T h i s is consistent w i t h previous reports i n w h i c h A Q P 9 m R N A levels decreased after protein kinase C ( P K C ) s t imulat ion by T P A i n astrocytes 3 3 . W h i l e the b io log i ca l s ignif icance o f this act ivat ion- induced down-regula t ion o f A Q P 9 i n C D 4 + C D 2 5 + Tregs is unclear, it is clear that A Q P 9 is not an act ivat ion marker o f C D 4 + C D 2 5 " T effector cel ls . 40 n co 0-O < c o "55 <D 1 30 20 10H Resting 24 hr 48 hr 5 day Resting 48 hr 5 day CD4+CD25- CD4+CD25+ Figure 4.5. AQP9 is not an activation marker for human CD4+CD25" T effector cells. Ex vivo purified CD4+CD25+ Tregs and CD4+CD25" Te were activated with plate-bound aCD3 (lug/ml) and soluble ctCD28 (lug/ml), in the presence of minimal IL-2 (lOU/ml) for times indicated prior to RNA extraction, and the expression of AQP9 was determined by quantitative PCR. Values plotted are normalized by GAPDH expression. A single representative experiment of 3 is shown. Human CD4+CD25+ Tregs express more AQP9 protein as compared to CD4+CD25" T effector cells. W e next sought to conf i rm that the increased levels o f A Q P 9 gene expression i n C D 4 + C D 2 5 + Tregs translated to an increase i n protein expression, as compared to C D 4 + C D 2 5 " T effector cel ls . W e performed western blot analysis o n who le ce l l lysates f rom pur i f i ed C D 4 + C D 2 5 + Tregs and C D 4 + C D 2 5 " T effector cel ls , and on ly i n C D 4 + C D 2 5 + T reg lysates were w e able to detect l o w levels o f A Q P 9 protein (Figure 4.6). T h i s suggests that 112 the h i g h levels o f A Q P 9 message i n C D 4 + C D 2 5 + Tregs correspond w i t h a b io log ica l ly significant increase i n A Q P 9 protein. In order to assess the potential u t i l i ty o f A Q P 9 as a ce l l surface marker for C D 4 + C D 2 5 + Tregs, it w o u l d be necessary that A Q P 9 protein was detectable o n the surface o f viable Tregs. Unfortunately , the on ly c o m m e r c i a l l y available reagent for detection o f A Q P 9 is a p o l y c l o n a l antibody raised against an intracel lular epitope o f the protein, and is thus not appropriate for sort ing o f l ive cel ls . In addi t ion, w e were unable to detect A Q P 9 i n ex v i v o Tregs by immunocytochemis t ry w i t h this po lyc lona l antibody, l i k e l y due to a lack o f sensit ivity (data not shown). Thus , i n future, more sensitive antibodies raised against extracellular epitopes o f A Q P 9 w i l l be necessary to conc lus ive ly determine whether this protein can be used to sort pure populat ions o f l ive Tregs. AQP9+ control CD25- CD25+ -32KD Figure 4.6. Human CD4+CD25+ Tregs express more AQP9 protein as compared to CD4+CD25' T effectors. Whole cell lysates were made from ex vivo purified CD4+CD25+ Tregs and CD4*CD25" Te and AQP9-transfected HEK 293T cells were resolved by SDS-PAGE and immunoblotted with ct-AQP9 pAbs. The membrane was re-probed with ct-p38 Abs to assess equivalency of loading. A single representative experiment of 3 is shown. Lent i -v i ra l transduction of human C D 4 + C D 2 5 " T effectors wi th A Q P 9 It has been prev ious ly demonstrated that a transcription factor required for C D 4 + C D 2 5 + T reg development, F O X P 3 , can induce a T reg phenotype when transduced into C D 4 + C D 2 5 " T effector cells A s the h igh expression o f A Q P 9 was found exc lus ive ly i n C D 4 + C D 2 5 + Tregs, we therefore investigated whether a s imi l a r change i n phenotype w o u l d occur upon 113 A Q P 9 over-expression i n C D 4 + C D 2 5 " T effector cel ls . In order to over-express A Q P 9 i n p r imary T cel ls , we constructed a len t iv i ra l vector encoding A Q P 9 c D N A and a truncated vers ion o f the human nerve g rowth factor receptor ( A N G F R ) as a cell-surface marker for accurate sorting o f transduced cel ls . F l o w cytometry pur i f ied C D 4 + C D 2 5 " T effector ce l l l ines were either left untransduced ( U T ) , or transduced w i t h a control l en t i -v i ra l vector conta ining the A N G F R marker gene alone ( p C C L ) or the A Q P 9 - e x p r e s s i n g len t iv i ra l vector ( p C C L - A Q P 9 ) . Transduced T ce l l l ines were then expanded in vitro, and pur i f ied o n the basis o f A N G F R expression (Figure 4 .7a) . In order to investigate b io log i ca l consequences o f A Q P 9 expression, w e assayed transduced T cel ls for phenotype and function. Firs t , we examined the resting expression o f k n o w n Treg-associated molecules F O X P 3 , C D 2 5 , and C T L A - 4 , and d i d not observe any difference between control and A Q P 9 - t r a n s d u c e d C D 4 + C D 2 5 " T effector c e l l l ines (p=NS, n=3) (Figure 4 . 7b) . Further experiments d i d not reveal any change i n prol i fera t ion o f A Q P 9 expressing T e (Figure 4 .7c) , or acquis i t ion o f suppressive capacity (Figure 4 . 7 d ) (p=NS, n=3). In contrast, a C D 4 + C D 2 5 + T reg ce l l l ine from the same donor showed h i g h expression o f Treg-associated molecules as w e l l as an anergic and suppressive phenotype (Figure 4 . 7 b , c, d). Therefore, lent i -vira l mediated over-expression o f A Q P 9 d i d not induce a Treg phenotype i n C D 4 + C D 2 5 " T effector ce l l l ines. Howeve r , p re l iminary data suggests that A Q P 9 expression is increased i n F O X P 3 - t r a n s d u c e d T effector cel ls (data not shown), suggesting that expression o f A Q P 9 may be correlated w i t h F O X P 3 expression. 114 CD25-Figure 4.7. Lenti-viral transduction of human CD4 +CD25" T effectors with AQP9. CD4 +CD25" Te were transduced with lenti virus encoding AQP9 (pCCL-AQP9) or control (pCCL) vector. (A) A N G F R expression of transduced cells following purification. (B) Resting transduced cells were stained for FOXP3, CD25, and C T L A - 4 expression. (C) Cells were stimulated with aCD3 (lug/ml) and irradiated APCs and proliferation was assessed. (D) Allogenic CD4 + T cells were stimulated with ccCD3 (lug/ml) and irradiated APCS in the absence or presence of CD4 + T cell lines (1:1 ratio). Data are depicted as percent suppression ([l-(Te+Treg/Te alone)]* 100). (C&D) Proliferation and suppression were assessed by 3H-thymidine incorporation. (A, B, C&D) A single representative experiment of 3 is depicted. 115 4.4 DISCUSSION In this study, we performed comparat ive microarray analysis o n funct ional ly defined human C D 4 + C D 2 5 + T r eg and C D 4 + C D 2 5 " T effector clones i n order to identify new specific markers o f human Tregs. W e f ind that human C D 4 + C D 2 5 + Tregs express h igh levels o f A Q P 9 as compared to C D 4 + C D 2 5 " T effector cells , as observed by microarray analysis o f C D 4 + C D 2 5 + T r eg clones and q P C R va l ida t ion o f ex v i v o C D 4 + C D 2 5 + Tregs. T o date, it has s not yet been previous ly reported that A Q P 9 is expressed o n T cel ls , a l though it is k n o w n to be expressed on m y e l o i d cel ls and neutrophils 2 0 ' 3 4 . A Q P 9 does not appear to be an act ivat ion marker o f C D 4 + C D 2 5 " T effector cel ls , as expression levels were not increased f o l l o w i n g T C R activat ion. Expres s ion o f A Q P 9 m R N A i n ex v i v o C D 4 + C D 2 5 + Tregs decreased after act ivat ion, i n l ine w i t h a previous report i n w h i c h act ivat ion o f P K C decreased levels o f A Q P 9 m R N A and protein i n astrocytes 3 3 . A s h i g h levels o f A Q P 9 m R N A have been observed i n both C D 4 + C D 2 5 + T r e g c lones and c e l l l ines i n the resting phase, f o l l o w i n g repeated i n v i t ro expansion, A Q P 9 expression l i ke ly fluctuates through the ce l l cyc le . A l t h o u g h A Q P 9 m R N A levels are s ignif icant ly upregulated i n the Treg cel ls o f a l l donors tested, w e have not been able to address the potential ut i l i ty o f A Q P 9 as a c e l l surface marker for sorting o f C D 4 + C D 2 5 + Tregs due to the lack o f appropriate detection reagents. Func t iona l ly , w e have demonstrated that lent i -v i ra l mediated over-expression o f A Q P 9 i n C D 4 + C D 2 5 " T effector cel ls does not promote the acquis i t ion o f a T reg phenotype, suggesting that this protein is not sufficient to induce regulatory function. Future s i R N A -based experiments to inhib i t A Q P 9 expression i n human Tregs , or the study o f A Q P 9 deficient mice , w i l l be necessary to determine whether this prote in is essential for regulatory 116 funct ion i n these cel ls . In addit ion, s tudying the consequences o f A Q P 9 over-expression i n C D 4 + C D 2 5 + Tregs w i l l certainly be o f interest. A Q P 9 is a ce l l surface membrane channel protein that permits the passage o f water, g lyce ro l , urea, and other smal l solutes 1 3 ' 1 4 . M i c r o a r r a y analysis on T reg clones d i d not reveal any significant expression o f any other aquaporins, suggesting that A Q P 9 may have a non-redundant funct ion on C D 4 + C D 2 5 + Tregs (data not shown). A Q P 9 is required for neutrophi l mot i l i t y 1 6 , and the increase i n size required for osteoclast differentiation 2 0 , thus one function on C D 4 + C D 2 5 + Tregs may relate to water transport. It has been recently shown that A Q P 9 is up-regulated o n rat hepatocytes dur ing i n response to starvation, and return to baseline i n response to in su l in release f o l l o w i n g feeding 1 3 . G l y c e r o l is released into the b l o o d as a product o f the fatty ac id metabol i sm result ing f rom starvation, where it can then be used for gluconeogenesis 3 5 . Thus , the increased express ion o f A Q P 9 on the c e l l surface w o u l d facilitate the maintenance o f the energy balance o f the ce l l 1 3 ' 1 4 . C D 4 * C D 2 5 + Tregs are k n o w n to have altered intracellular s ignal ing downstream o f the T C R , and have defective act ivat ion o f A K T and its targets 2 2 . A K T act ivat ion is associated w i t h decreased expression o f pro-apoptotic proteins, and increased protein and g lycogen synthesis leading to ce l l cycle entry and s u r v i v a l 3 6 ' 3 7 . It is therefore interesting to speculate that C D 4 + C D 2 5 + Tregs express A Q P 9 to compensate for an alteration i n energy balance result ing f r o m a decrease i n A K T funct ion. O u r strategy for Treg c lon ing based on differential expression o f C D 2 5 and C D 6 2 L expression revealed that T reg c lon ing o n the basis o f C D 6 2 L expression does not increase the frequency o f T reg clones. The use o f C D 4 + C D 2 5 + T reg clones for microarray analysis ensured a homogeneous regulatory popula t ion that was not contaminated w i t h effector T 117 cel ls , a c o m m o n l y encountered p rob lem w h e n w o r k i n g w i t h bu lk populat ions o f T r e g cells . H o w e v e r , the l imi ta t ion o f ana lyz ing on ly 2 T reg clones der ived f rom a single donor may also bias results. F o r this reason we proceeded to va l ida t ion o f ident i f ied genes us ing quantitative P C R on po lyc lona l T c e l l populat ions f rom mul t ip le donors. Interestingly, some o f genes suggested by the microar ray to be expressed at higher levels i n C D 4 + C D 2 5 + Tregs were m y e l o i d lineage associated genes. The differential expression o f some o f these genes, for example C S F 1 R , was not val idated by q P C R . It is possible that the o r ig ina l microarray observat ion was a result o f contaminat ion w i t h remnants o f feeder cel ls that were not di luted out and/or phagocytosed by the s low g r o w i n g T reg ce l l c lones. H o w e v e r , other myeloid-associated genes such as I T G A M / C D 1 l b were val idated by q P C R and expressed at higher levels i n T r e g clones. In particular, calgranul ins A and B ( S 1 0 0 A 8 and S 1 0 0 A 9 ) were consistently expressed at higher levels i n human C D 4 + C D 2 5 + Tregs compared to C D 4 + C D 2 5 " T effectors. Together w i t h S 1 0 0 A 1 2 , ca lg ranu l in A and B are referred to as m y e l o i d related proteins ( M R P ) , and are considered damage-related proteins because they are found at h igh levels at sites o f inf lammat ion , consistent w i t h the presence o f neutrophils and monocytes 3 8 . W h i l e it is unclear what funct ion ca lgranul in A and B w o u l d serve i n C D 4 + C D 2 5 + Tregs, future experiments to explore this quest ion w o u l d certainly be o f interest. O v e r a l l , this report has contributed informat ion on the gene expression prof i le o f C D 4 + C D 2 5 + T r eg single ce l l clones, and has identif ied the channel prote in A Q P 9 as a potential Treg-speci f ic ce l l surface marker , and a p romis ing target o f functional invest igat ion i n human C D 4 + C D 2 5 + Tregs. 118 4.5 R E F E R E N C E S 1. H o r i S, N o m u r a T , Sakaguchi S. C o n t r o l o f regulatory T c e l l development by the t ranscript ion factor F o x p 3 . Science. 2003;299:1057-1061. 2. Kha t t r i R , C o x T, Y a s a y k o S A , R a m s d e l l F . A n essential role for Scur f in i n C D 4 + C D 2 5 + T regulatory cel ls . Na t I m m u n o l . 2003;4:337-342. 3. Sakaguchi S, Sakaguchi N , A s a n o M , Itoh M , T o d a M . Immuno log i c self-tolerance mainta ined by activated T cel ls expressing I L - 2 receptor alpha-chains ( C D 2 5 ) . B r e a k d o w n o f a s ingle mechan i sm o f self-tolerance causes var ious autoimmune diseases. J I m m u n o l . 1995;155:1151-1164. 4. L e v i n g s M K , Sangregorio R , Sartirana C , et a l . H u m a n C D 2 5 + C D 4 + T suppressor c e l l clones produce transforming growth factor beta, but not in ter leukin 10, and are dist inct f rom type 1 T regulatory cells . J E x p M e d . 2002;196:1335-1346. 5. B r u n k o w M E , Jeffery E W , Hje r r i l d K A , et al . D i s rup t ion o f a n e w forkhead/winged-he l i x protein, scurfin, results i n the fatal lymphoprol i fera t ive disorder o f the scurfy mouse. N a t Genet. 2001;27:68-73. 6. L i u W , Pu tnam A L , X u - Y u Z , et a l . C D 1 2 7 expression inversely correlates w i t h F o x P 3 and suppressive funct ion o f human C D 4 + T reg cel ls . J E x p M e d . 2006;203:1701-1711. 7. G a v i n M A , C la rke S R , N e g r o u E , Ga l l egos A , Rudensky A . Homeostas is and anergy o f C D 4 ( + ) C D 2 5 ( + ) suppressor T ce l l s i n v i v o . N a t I m m u n o l . 2002;3 ;33-41. 8. M c H u g h R S , Whit ters M J , P i c c i r i l l o C A , et a l . C D 4 ( + ) C D 2 5 ( + ) immunoregula tory T cel ls : gene expression analysis reveals a functional role for the g lucocor t ico id- induced T N F receptor. Immuni ty . 2002;16:311-323. 9. S h i m i z u J , Y a m a z a k i S, Takahashi T , Ishida Y , Sakaguchi S. S t imula t ion o f C D 2 5 ( + ) C D 4 ( + ) regulatory T cel ls through G I T R breaks i m m u n o l o g i c a l self-tolerance. Na t I m m u n o l . 2002;3:135-142. 10. Brude r D , Probs t -Kepper M , Wes tendor f A M , et a l . N e u r o p i l i n - 1 : a surface marker o f regulatory T cel ls . E u r J Immuno l . 2004;34:623-630. 11. K n o e c h e l B , L o h r J , Z h u S, et a l . Func t iona l and molecular compar i son o f anergic and regulatory T lymphocytes . J I m m u n o l . 2006;176:6473-6483. 12. Shevach E M . F r o m v a n i l l a to 28 f lavors : mul t ip le varieties o f T regulatory cel ls . Immuni ty . 2006;25:195-201. 13. Carbrey J M , G o r e l i c k - F e l d m a n D A , K o z o n o D , Praetorius J , N i e l s e n S, A g r e P . A q u a g l y c e r o p o r i n A Q P 9 : solute permeat ion and metabol ic control o f expression i n l iver . P roc N a t l A c a d S c i U S A . 2003;100:2945-2950. 119 14. Badaut J , R e g l i L . D i s t r ibu t ion and possible roles o f aquaporin 9 i n the brain. Neurosc ience . 2004;129:971-981. 15. N i h e i K , K o y a m a Y , T a n i T , et a l . Immunoloca l i za t ion o f aquaporin-9 i n rat hepatocytes and L e y d i g cells . A r c h H i s t o l C y t o l . 2001;64:81-88. 16. L o i t t o V M , Fors lund T , Sundqvis t T , M a g n u s s o n K E , Gustafsson M . Neu t roph i l leukocyte mot i l i t y requires directed water inf lux . J L e u k o c B i o l . 2002;71:212-222. 17. W a n g S, C h e n J , B e a l l M , Z h o u W , Ross M G . Express ion o f aquaporin 9 i n human chor ioamnio t i c membranes and placenta. A m J Obstet G y n e c o l . 2004;191:2160-2167. 18. Badaut J , Petit J M , Brunet J F , Magis t re t t i P J , Char r iau t -Mar langue C , R e g l i L . Di s t r ibu t ion o f A q u a p o r i n 9 i n the adult rat brain: preferential express ion i n catecholaminergic neurons and i n g l i a l cel ls . Neurosc ience . 2004;128:27-38. 19. Ishibashi K , K u w a h a r a M , G u Y , Tanaka Y , M a r u m o F , Sasaki S. C l o n i n g and funct ional expression o f a new aquaporin ( A Q P 9 ) abundantly expressed i n the peripheral leukocytes permeable to water and urea, but not to g lycero l . B i o c h e m B i o p h y s Res C o m m u n . 1998;244:268-274. 20. A h a r o n R , Bar -Shav i t Z . Involvement o f aquaporin 9 i n osteoclast differentiation. J B i o l C h e m . 2006;281:19305-19309. 2 1 . C r e l l i n N K , G a r c i a R V , Hadisfar O , A l l a n S E , Steiner T S , L e v i n g s M K . H u m a n C D 4 + T cel ls express T L R 5 and its l igand f lage l l in enhances the suppressive capacity and expression o f F O X P 3 i n C D 4 + C D 2 5 + T regulatory cel ls . J Immuno l . 2005;175:8051-8059. 22 . C r e l l i n N K , G a r c i a R V , L e v i n g s M K . A l t e r e d act ivat ion o f A K T is required for the suppressive funct ion o f human C D 4 + C D 2 5 + T regulatory cel ls . B l o o d . 2006. 23. L e v i n g s M K , Sangregorio R , Ronca ro lo M G . H u m a n C D 2 5 + C D 4 + T regulatory cel ls suppress naive and memory T - c e l l prol i fera t ion and can be expanded in vitro wi thout loss o f funct ion. J E x p M e d . 2001;193:1295-1302. 24. A m e n d o l a M , V e n n e r i M A , B i f f i . A , V i g n a E , N a l d i n i L . Coordinate dual-gene transgenesis by len t iv i ra l vectors ca r ry ing synthetic b id i rec t ional promoters. N a t B io t echno l . 2005;23:108-116. 25 . K o h n A D , Bar the l A , K o v a c i n a K S , et a l . Cons t ruc t ion and characterizat ion o f a condi t iona l ly active vers ion o f the serine/threonine kinase A k t . J B i o l C h e m . 1998;273:11937-11943. 26. A l l a n S E , Passerini L , Bacchet ta R , et a l . The role o f 2 F O X P 3 isoforms i n the generation o f human C D 4 + Tregs. J C l i n Invest. 2005;115:3276-3284. 120 27. E r m a n n J , Hof fmann P , Ed inge r M , et a l . O n l y the C D 6 2 L + subpopulat ion o f C D 4 + C D 2 5 + regulatory T cel ls protects f rom lethal acute G V H D . B l o o d . 2005; 105:2220-2226. 28. Y o u S, Stehoffer G , Bar r io t S, B a c h J F , Chatenoud L . U n i q u e role o f C D 4 + C D 6 2 L + regulatory T cel ls i n the control o f autoimmune diabetes i n T ce l l receptor transgenic mice . P roc N a t l A c a d S c i U S A . 2004;101 Supp l 2:14580-14585. 29. Pfoertner S, Jeron A , Probs t -Kepper M , et a l . Signatures o f human regulatory T cel ls : an encounter w i t h o l d friends and new players. Genome B i o l . 2006 ;7 :R54 . 30. H u a n g C T , W o r k m a n C J , F l i e s D , et a l . R o l e o f L A G - 3 i n regulatory T cel ls . Immuni ty . 2004;21:503-513. 31 . Pearce E L , M u l l e n A C , Mar t i n s G A , et a l . Con t ro l o f effector C D 8 + T ce l l funct ion by the t ranscript ion factor Eomesodermin . Science. 2003;302:1041-1043. 32. O g a w a K , Tanaka K , Ish i i A , et a l . A nove l serum protein that is select ively produced by cy to toxic lymphocytes . J Immuno l . 2001;166:6404-6412. 33. Y a m a m o t o N , Sobue K , M i y a c h i T , et a l . Different ia l regulat ion o f aquaporin expression i n astrocytes by protein kinase C . B r a i n Res M o l B r a i n Res . 2001 ;95:110-116. 34. Bhattacharjee H , Carbrey J , R o s e n B P , M u k h o p a d h y a y R . D r u g uptake and pharmaco log ica l modula t ion o f drug sensit ivi ty i n l eukemia by A Q P 9 . B i o c h e m B i o p h y s Res C o m m u n . 2004;322:836-841. 35. F i n n P F , D i c e JF . Proteolyt ic and l i po ly t i c responses to starvation. Nu t r i t i on . 2006;22:830-844. 36. Seminar io M C , Wange R L . L i p i d phosphatases i n the regulat ion o f T ce l l act ivation: l i v i n g up to their P T E N - t i a l . I m m u n o l Rev . 2003;192:80-97. 37. Woodget t JR . Recent advances i n the protein kinase B s igna l ing pathway. C u r r O p i n C e l l B i o l . 2005;17:150-157. 38. F o e l l D , W i t t k o w s k i H , V o g l T , R o t h J . S100 proteins expressed i n phagocytes: a nove l group o f damage-associated molecula r pattern molecules . J L e u k o c B i o l . 2006. 121 5. D i s c u s s i o n and C o n c l u s i o n s The in i t i a l a i m o f this research was to investigate and further elucidate the molecular phenotype o f human C D 4 + C D 2 5 + Tregs. The role o f T L R 5 on human C D 4 + C D 2 5 + Tregs was studied, and it was reported that a l l C D 4 + T cel ls express T L R 5 . Further, it appears that the T L R 5 l igand f lage l l in is a co-st imulatory molecu le for C D 4 + C D 2 5 " T effector cel ls , and that f l age l l in can increase the suppressive capacity o f C D 4 + C D 2 5 + Tregs i n the absence o f A P C s . In an effort to understand the mechan i sm under ly ing the in vitro anergy o f C D 4 + C D 2 5 + T regulatory ce l ls , their intracellular s ignal ing downstream o f the T C R has been investigated. T h i s research has led to the ident if icat ion o f an alteration i n C D 4 + C D 2 5 + T regulatory ce l l s ignal ing at the leve l o f A K T act ivat ion, w h i c h is required for their suppressive capacity. In addi t ion, microarray analysis o n c lona l c e l l l ines was performed i n an attempt to identify a more specif ic ce l l surface marker for human C D 4 + C D 2 5 + T regulatory cel ls , and it was observed that human C D 4 + C D 2 5 + Tregs speci f ica l ly express the membrane channel protein A Q P 9 . The research on the role o f T L R 5 on C D 4 + T cel ls has p rov ided fundamental evidence to an emerging f ie ld s tudying the direct interaction o f bacterial products and T L R l igands w i t h cel ls o f the adaptive immune system. It n o w appears that the conceptual d i v i s i o n and l inks between the innate and adaptive immune systems should be reconsidered. Indeed, it appears that C D 4 + T cel ls speak the language o f bacteria, and are not solely reliant on dendri t ic cel ls as translators. Recent publ icat ions have described a role for T L R 2 i n the expans ion and funct ion o f both C D 4 + C D 2 5 + Tregs and C D 4 + C D 2 5 " T effectors u , and suggested that H S P 6 0 can enhance C D 4 + C D 2 5 + T reg f u n c t i o n 3 . These recent publ icat ions are consistent w i t h the mode l proposed i n Chapter 2, w h i c h suggests that T L R l igands can be 122 sensed direct ly by C D 4 + T cells and promote both C D 4 + T effector ce l l ac t ivat ion and C D 4 + C D 2 5 + T r eg function. In v i v o , the most relevant site to study the interaction o f bacterial products and C D 4 + C D 2 5 + Tregs is the gastrointestinal tract, where there is a c lear ly defined requirement for C D 4 + C D 2 5 + Tregs to control aberrant inf lammatory responses to commensa l bacteria 4 . Therefore, a log ica l continuat ion o f this research w o u l d be to study the role o f f l age l l in and T L R 5 o n the suppression o f col i t is by C D 4 + C D 2 5 + Tregs i n a murine an imal m o d e l . A d d i t i o n a l l y , as it has been observed that f l age l l in is the dominant antigen i n C r o h n ' s disease 5 , it w o u l d be interesting to examine the numbers and functional capacity o f C D 4 + C D 2 5 + Tregs i n the gut o f C r o h n ' s disease patients. A s discussed i n Chapter 1, a process o f infectious tolerance has been suggested whereby C D 4 + C D 2 5 + t r e g s can convert C D 4 + T effector cel ls into cy tokine secreting T r l cel ls . Cy tok ines are k n o w n to p lay a par t icular ly important role i n regulat ing intestinal homeostasis 4 . Therefore, a l og ica l cont inuat ion o f this w o r k w o u l d be to investigate the potential o f T L R l igands, such as f lage l l in , to modulate the process o f infectious tolerance. Investigations o f intracellular s igna l l ing downstream o f the T C R i n C D 4 + C D 2 5 + Tregs has greatly advanced scientific understanding o f the alterations that contribute to C D 4 + C D 2 5 + T r e g anergy. A l t h o u g h the induc t ion o f A K T act ivi ty increased C D 4 + C D 2 5 + T r e g prol i ferat ion, it was not sufficient to fu l ly break anergy. Recent publ icat ions f rom the T u r k a laboratory have suggested that the phosphatase P T E N is responsible for the anergic response o f C D 4 + C D 2 5 + Tregs to I L - 2 alone 6 . The hypo-prol i ferat ive profi le o f C D 4 + C D 2 5 + Tregs presents a challenge to any appl ica t ion requir ing i n v i t ro expansion, whether for experimental manipulat ions or potential therapeutic uses. A better understanding 123 o f the intracel lular determinants o f C D 4 + C D 2 5 + T reg anergy w i l l i n fo rm and direct ex v i v o expans ion protocols . B y seeking to reverse the defect i n A K T act ivat ion observed i n C D 4 + C D 2 5 + Tregs, it was observed that induced A K T act ivi ty reversed the i n vi tro suppressive capacity o f C D 4 + C D 2 5 + Tregs. T h i s represents a major advance i n understanding the molecu la r requirements o f C D 4 + C D 2 5 + T r eg suppression, and is the first system reported i n w h i c h suppression can be ' turned o f f at the leve l o f the T reg ce l l , rather than the T effector. T h i s system c o u l d be o f great benefit i n research to elucidate the poor ly defined mechan i sm o f suppression i n C D 4 + C D 2 5 + Tregs. A d d i t i o n a l l y , this observation may fuel further research into the therapeutic modula t ion o f C D 4 + C D 2 5 + Tregs. In order to perform intracellular s ignal ing studies w i t h l imi ted numbers o f human ex v i v o C D 4 + C D 2 5 + Tregs, previous ly developed f l o w cytometr ic techniques were adapted 7 " 1 0 . Me thodo log ies were developed for the detection o f C D 2 5 and/or F O X P 3 i n combina t ion w i t h phospho-specif ic f l o w cytometry antibodies, thus a l l o w i n g the specific analysis o f intracel lular s ignal ing i n C D 4 + C D 2 5 + Tregs f rom w i t h i n a m i x e d popula t ion o f C D 4 + T cel ls . A detailed report o f this method has been prepared for submiss ion to the Journal o f I m m u n o l o g i c a l Methods . A l t h o u g h a defect i n A K T phosphoryla t ion i n C D 4 + C D 2 5 + Tregs was identif ied, the cause o f that defect was not isolated. One potential candidate molecu le c o u l d be the kinase responsible for phosphoryla t ing S473 , P D K 2 , n o w thought to be the m T O R c o m p l e x m T O R C 2 1 1 1 3 . It w o u l d be o f great interest to study the association o f m T O R w i t h the m T O R C 2 complex i n C D 4 + C D 2 5 + Tregs. Al te rna t ive ly , the over expression o f a 124 phosphatase, such as the recently discovered P H L P P 1 4 , cou ld produce the same defect i n A K T act ivat ion. T h e forced act ivat ion o f A K T i n the presence o f T C R s t imulat ion restored C D 4 + C D 2 5 + T r e g cytokine product ion o f I L - 4 , I L - 1 0 , I F N - y , and T N F - a , but not I L - 2 . It is interesting that induc t ion o f A K T act ivi ty should stimulate both T h i and T h 2 cytokine release f rom C D 4 + C D 2 5 + Tregs. A good fo l l ow-up experiment w o u l d be to perform intracel lular cy tokine staining o n these cel ls to determine i f there was un i fo rm expression o f a l l cytokines produced, or i f there was a split into T h i and T h 2 subtypes. S i m i l a r l y , it w o u l d be o f interest to study the i n v i v o suppressive capacity o f murine C D 4 + C D 2 5 + Tregs hav ing consti tutive A K T act ivi ty. A d d i t i o n a l l y , the observation o f defective A K T phosphoryla t ion i n C D 4 + C 2 5 + Tregs is consistent w i t h reports o f preferential expansion o f C D 4 + C D 2 5 + Tregs when cul tured i n the presence o f rapamycin , an inhibi tor o f m T O R l 5 1 6 . One explanat ion is that because C D 4 + C D 2 5 + Tregs have decreased A K T act ivat ion, they are resistant to the suppression o f m T O R by rapamycin , and thus their prol i ferat ion is unchanged. Converse ly , C D 4 + T effector cel ls are inh ib i ted by rapamycin , and thus expansion o f a m i x e d culture o f C D 4 + T cel ls i n the presence o f r apamyc in skews towards C D 4 + C D 2 5 + T r eg growth. H o w e v e r , an alternative hypothesis w o u l d suggest that suppression o f m T O R cou ld facilitate the convers ion o f C D 4 + C D 2 5 ~ T effector cells into C D 4 + C D 2 5 + Tregs. C lea r ly further experiments invest igat ing these hypotheses w o u l d be o f interest. T h e microarray analysis described i n Chapter 4 represents an attempt to use the gene expression prof i le o f funct ional ly tested, human C D 4 + C D 2 5 + T r eg s ingle-ce l l der ived clones as a tool for ident if icat ion o f unique C D 4 + C D 2 5 + T reg ce l l surface markers. T h i s approach 125 may be considered both a strength and a l imi ta t ion . The potential o f contaminat ing T effector cel ls has been removed, and thus presumably w i l l have removed some 'no ise ' f rom the system. H o w e v e r , i n order to achieve sufficient ce l l numbers the clones went through mul t ip l e rounds o f i n v i t ro expansion. U l t ima te ly , the microarray analysis was performed on on ly 2 T r e g clones der ived f rom a single donor, and this may bias the results. Nonetheless , this research led to the nove l observation that human C D 4 + C D 2 5 + Tregs express the membrane channel protein aquaporin 9. I f appropriate detection reagents are developed, there is potential for A Q P 9 to be used as a ce l l surface marker for the detection and i so la t ion o f C D 4 + C D 2 5 + Tregs, w h i c h w o u l d have great commerc i a l and scient if ic impact . A s it i s , these observations w i l l open a new avenue o f research into the characterizat ion and function o f A Q P 9 on C D 4 + C D 2 5 + Tregs. A l t h o u g h it has been established that the over-expression o f A Q P 9 i n C D 4 + C D 2 5 " T effector cel ls does not induce suppressive T r e g phenotype, the funct ion o f A Q P 9 i n C D 4 + C D 2 5 + T r eg cel ls awaits invest igat ion. It may be that C D 4 + C D 2 5 + Tregs express A Q P 9 i n order to facilitate the impor t o f g lyce ro l , w h i c h can then be used for energy product ion w i t h i n the c e l l . A n interesting hypothesis is that C D 4 + C D 2 5 + Tregs express A Q P 9 i n order to compensate for the decreased act ivi ty o f A K T , w h i c h w o u l d result i n an alteration o f the metabol ic balance o f the c e l l . H o w e v e r , this remains to be tested. H u m a n C D 4 + C D 2 5 + Tregs are k n o w n to differ s ignif icant ly f rom their mur ine counterparts. F o r example, i n human cel ls the expression o f F O X P 3 is not exc lus ive to Tregs, and i n fact is an act ivat ion marker for C D 4 + C D 2 5 " T effector cel ls 1 7 ' 1 8 . A d d i t i o n a l l y , a spl ice varient o f F O X P 3 exists i n human Tregs but not mouse 1 9 . A n important molecu le i n the funct ion o f murine Tregs is C D 103 2 0 ' 2 1 , however this molecule is not detectable o n 126 human Tregs 2 2 . One o f the strengths o f the research presented here is it was a l l performed using human cel ls , and thus has more therapeutic relevance. In summary, the research presented i n this thesis represents a significant contr ibut ion to the f i e ld o f immuno logy . The interaction o f C D 4 + C D 2 5 + Tregs w i t h bacterial products v i a T L R s has been explored, the unique alterations i n the intracellular s ignal ing pathways o f C D 4 + C D 2 5 + Tregs, and their contr ibut ion to suppression have been elucidated, and a study o f the gene expression profi le o f Treg clones has led to the discovery o f a new molecu le associated w i t h the T reg phenotype. Thus , the molecular phenotype o f C D 4 + C D 2 5 + Tregs has been explored v i a mul t ip le approaches, y i e ld ing nove l observations o f considerable impor t to the study o f human C D 4 + C D 2 5 + Tregs and their control o f immune homeostasis. 127 5.1. REFERENCES 1. L i u H , K o m a i - K o m a M , X u D , L i e w F Y . T o l l - l i k e receptor 2 s ignal ing modulates the functions o f C D 4 + C D 2 5 + regulatory T cel ls . P roc N a t l A c a d S c i U S A . 2006;103:7048-7053. 2. Su tmul le r R P , den B r o k M H , K r a m e r M , et a l . T o l l - l i k e receptor 2 controls expansion and funct ion o f regulatory T cells . J C l i n Invest. 2006; 116:485-494. 3. Z a n i n - Z h o r o v A , Caha lon L , T a i G , M a r g a l i t R , L i d e r O , C o h e n IR . Heat shock protein 60 enhances C D 4 + C D 2 5 + regulatory T ce l l function v i a innate T L R 2 s ignal ing. J C l i n Invest. 2006; 116:2022-2032. 4. P o w r i e F . Immune Regula t ion i n the Intestine: A Ba lanc ing A c t between Effector and Regula tory T C e l l Responses. A n n N Y A c a d S c i . 2004;1029:132-141. 5. L o d e s M J , C o n g Y , E l s o n C O , et a l . Bac te r ia l f lage l l in is a dominant antigen i n C r o h n disease. J C l i n Invest. 2004;113:1296-1306. 6. W a l s h P T , B u c k l e r J L , Z h a n g J , et a l . P T E N inhibi ts I L - 2 receptor-mediated expansion o f C D 4 + C D 2 5 + Tregs. J C l i n Invest. 2006;116:2521-2531. 7. K r u t z i k P O , Ha le M B , N o l a n G P . Character izat ion o f the murine i m m u n o l o g i c a l s ignal ing ne twork w i t h phosphospecif ic f l o w cytometry. J Immuno l . 2005;175:2366-2373. 8. K r u t z i k P O , Ir ish J M , N o l a n G P , Perez O D . A n a l y s i s o f protein phosphoryla t ion and ce l lu la r s igna l ing events by f l o w cytometry: techniques and c l i n i c a l appl icat ions. C l i n I m m u n o l . 2004;110:206-221. 9. Perez O D , K r u t z i k P O , N o l a n G P . F l o w cytometric analysis o f kinase s ignal ing cascades. Me thods M o l B i o l . 2004;263:67-94. 10. Perez O D , N o l a n G P . Simultaneous measurement o f mul t ip le active kinase states us ing po lychromat ic f l o w cytometry. N a t B io t echno l . 2002;20:155-162. 11. Jacinto E , L o e w i t h R , Schmid t A , et a l . M a m m a l i a n T O R complex 2 controls the actin cytoskeleton and is rapamycin insensit ive. Na t C e l l B i o l . 2004;6:1122-1128. 12. Sarbassov D D , Guer t in D A , A l i S M , Sabatini D M . Phosphory la t ion and regulat ion o f A k t / P K B by the r i c t o r - m T O R complex . Science. 2005;307:1098-1101. 13. Y a n g Q , Inok i K , Ikenoue T , G u a n K L . Identif icat ion o f S i n l as an essential T O R C 2 component required for complex format ion and kinase act ivi ty . Genes D e v . 2006;20:2820-2832. 14. G a o T , Furnar i F , N e w t o n A C . P H L P P : a phosphatase that d i rect ly dephosphorylates A k t , promotes apoptosis, and suppresses tumor growth. M o l C e l l . 2005;18:13-24. 128 15. Bat tag l ia M , S tab i l in i A , M i g l i a v a c c a B , Hore j s -Hoeck J , Kauppe r T , Ronca ro lo M G . R a p a m y c i n promotes expansion o f functional C D 4 + C D 2 5 + F O X P 3 + regulatory T cel ls o f both healthy subjects and type 1 diabetic patients. J Immuno l . 2006;177:8338-8347. 16. Bat tag l ia M , S tab i l in i A , Ronca ro lo M G . R a p a m y c i n selectively expands C D 4 + C D 2 5 + F o x P 3 + regulatory T cel ls . B l o o d . 2005;105:4743-4748. 17. G a v i n M A , Torgerson T R , Hous ton E , et a l . S ing le -ce l l analysis o f no rma l and F O X P 3 - m u t a n t human T cells: F O X P 3 expression without regulatory T c e l l development. P roc N a t l A c a d S c i U S A . 2006;103:6659-6664. 18. W a n g J , Ioan-Facsinay A , v a n der V o o r t E I , H u i z i n g a T W , Toes R E . Transient expression o f F O X P 3 i n human activated nonregulatory CD4(+) T cel ls . E u r J I m m u n o l . 2006. 19. A l l a n S E , Passerini L , Bacchet ta R , et a l . The role o f 2 F O X P 3 isoforms i n the generation o f human C D 4 + Tregs. J C l i n Invest. 2005;115:3276-3284. 20. Suf f ia I, R e c k l i n g S K , Salay G , B e l k a i d Y . A role for C D 103 i n the retention o f C D 4 + C D 2 5 + T r e g and control o f L e i s h m a n i a major infection. J I m m u n o l . 2005;174:5444-5455. 21 . S h i m i z u J , M o r i i z u m i E . Aging-dependent generation o f suppressive C D 4 + C D 2 5 -R 1 2 3 1 o C D 1 0 3 + T cel ls i n mice . E u r J I m m u n o l . 2003;33:2449-2458. 22. L e v i n g s M K , Roncaro lo M G . Phenotypic and functional differences between human C D 4 + C D 2 5 + and type 1 regulatory T cel ls . C u r r T o p M i c r o b i o l Immuno l . 2005;293:303-326. 129 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100385/manifest

Comment

Related Items