Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of the [alpha] 1-helical domain of the signal sequence in hemolysin recognition and transport Morden, Carla Rebecca 1998

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

Item Metadata

Download

Media
831-ubc_1998-0279.pdf [ 3.09MB ]
Metadata
JSON: 831-1.0088466.json
JSON-LD: 831-1.0088466-ld.json
RDF/XML (Pretty): 831-1.0088466-rdf.xml
RDF/JSON: 831-1.0088466-rdf.json
Turtle: 831-1.0088466-turtle.txt
N-Triples: 831-1.0088466-rdf-ntriples.txt
Original Record: 831-1.0088466-source.json
Full Text
831-1.0088466-fulltext.txt
Citation
831-1.0088466.ris

Full Text

The Role of the oti-Helical Domain of the Signal Sequence in Hemolysin Recognition and Transport by C A R L A REBECCA MORDEN B.Sc, Trinity Western University, 1995 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Biochemistry and Molecular Biology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1998 © Carla Rebecca Morden, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^\ocV>gArv^sW,_^ ^ WlrAe-ru-W^ 8> i o lo^c^ The University of British Columbia Vancouver, Canada Date ftpKi-ao-ycttg DE-6 (2788) ABSTRACT Hemolysin A (HlyA) is an RTX toxin (Repeats in Toxin) secreted by a transport complex consisting of HlyB, HlyD and TolC proteins. It is a unique transport system, in that the transport of HlyA does not require a periplasmic intermediate and its secretion is directed by an uncleaved C-terminal signal sequence. It is known that the signal sequence of leukotoxin A (LktA, a related toxin from P. hemolytica) can functionally replace the signal sequence of HlyA, even though they share little primary sequence homology. CD and NMR analysis of the C-terminal signal peptides of HlyA and LktA have demonstrated that both form similar secondary structures (helix-strand-helix) in a membrane mimetic environment (Yin et al, 1995; Zhang et ah, 1995). Thus it was postulated that a higher order structure distinguished by the amphiphiphilic helices in the signal sequence is important for recognition and transport of HlyA rather then the primary sequence of the signal sequence (Zhang et al., 1993b). In order to explore this hypothesis in greater detail, I have generated and utilized two libraries, a combinatorial library of sequences predicted to form amphiphilic a-helices and a random library of sequences, to further analyze the role of the structural domains in the HlyA signal sequence. The first library replaced the ai-helix region of the HlyA signal with a library of presumptive amphiphilic a-helices, and thus maintained the secondary structure while varying the identity of individual amino acids. The isolation of twenty-four a i-helical variants, based upon their ability to support transport, demonstrated that a conserved primary sequence was not essential for the recognition and transport of HlyA. Mutations of several amino acids, previously thought to be critical to recognition, were also shown by this approach to support HlyA transport. The second library replaced the ai-helix region of the HlyA signal with a library of random sequences, and thus allowed a variety of secondary structures to form, in order to determine if the secondary structure of the a i-helix region is required to support transport. In this case, if the isolated variants that are able to support transport form an amphiphilic oc-helical structure, it supports the hypothesis that the secondary structure of the a i-region is required for transport. The isolation of 65 variants yielded 4 variants that were able to support transport, 6 variants that were unable to support transport, as well as 55 variants with deletions, mutations, insertions or stop codons that did not conform to the design parameters. The 4 variants that were able to support transport had the expected periodicity of polar and nonpolar residues predicted by Kamtekar et al. to maintain the amphiphilic helical structure of the ai-helix (1991). These results fully support a model in which Hly A recognition by the transport complex occurs via interactions enabled by the secondary structure of the HlyA C-terminal signal sequence. T A B L E O F C O N T E N T S Page Number Abstract ii List of Tables vi List of Figures vi Acknowledgements vii CHAPTER I Introduction 1 1.1 R T X Toxin Transporters 1 1.2 The Escherichia coli Hemolysin A Transport System 2 1.3 The Hemolysin A C-terminal Signal Sequence 2 1.4 A Model for Hemolysin A Recognition and Transport 5 1.5 Thesis Research 8 CHAPTER II Methods and Materials 10 2.1 Bacterial Strains and Plasmids 10 2.2 Construction of Plasmids 10 2.3 Construction of the ai-Helical Library of HlyA Signal Sequence Variants : 11 2.4 Design of the ai-Helical Library of HlyA Signal Sequence Variants 12 2.5 Construction of the ai-Random Library of HlyA Signal Sequence Variants 16 2.6 Design of the ai-Random Library of HlyA Signal Sequence Variants 17 2.7 PCR Screening of the ai-Helical Library of HlyA Signal Sequence Variants 18 2.8 PCR Screening of the ai-Random Library of HlyA Signal Sequence Variants 18 2.9 Electroporation of Random Library of HlyA Signal Sequence Variants 19 2.1 OBlood Agar Plate Hemolysis Assay 19 2.11 Subcloning of the HlyA Signal Sequence Variants 20 2.12ELISA Assay Quantitation of HlyA Transport 20 2.13 Statistical and Sequence Analysis 21 CHAPTER III Results and Discussion 23 3.1 Screening of the ai-Helical Library of HlyA Signal Sequence Variants 23 iv Page Number 3.2 Functional Analysis of the ai-Helical Variants with the Hemolytic Zone Assay 24 3.3 Quantitation of Hemolysin A Recognition and Transport using the ELISA Assay 26 3.4 The ai-Helical Library of the HlyA Signal Sequence Variants 26 3.5 Screening of ai-Random Library of HlyA Signal Sequence Variants 29 3.6 Functional Analysis of the ai-Random Library of HlyA Signal Sequence Variants with the Hemolytic Zone Assay 31 3.7 The ai-Random Library of the HlyA Signal Sequence Variants 33 3.8 The Relationships of Helix Hydrophobic Moment, Hydrophobicity, and Beta Hydrophobic Moment with the Ability to Support Transport 33 3.9 The Amino Acid Composition of the oci-Helical Library of HlyA Signal Sequence Variants Based on Position 39 3.10 The Amino Acid Composition of the ai-Random Library of HlyA Signal Sequence Variants Based on Position 43 3.11 Does the ability to support transport correlate with secondary structure prediction? 45 CHAPTER IV Conclusion and Recommendations for Further Work 46 Bibliography 53 List of Tables Page Number Table 1. Amino Acid Sequence and Zone Assay Analysis of the ai-Helical Library of HlyA Signal Sequence Variants 25 Table2. Amino Acid Sequence and Zone Assay Analysis of the (Xi-Random Library of HlyA Signal Sequence Variants that Conform to the Design Parameters 31 Table3. Amino Acid Sequence and Zone Assay Analysis of the a i-Random Library of HlyA Signal Sequence Variants that do not Conform to the Design Parameters 33 Table4. Helix Hydrophobic Moment of HlyA Signal Sequence Variants 34 Table5. Hydrophobicity of HlyA Signal Sequence Variants 37 Table6. Beta Hydrophobic Moment of HlyA Signal Sequence Variants 40 Table7. Frequency of Amino Acids at Each Position in a i-Helical Variants 42 Table8. Frequency of Amino acids at Each Position in all (Xi-Random Variants (generated by method 2) 44 List of Figures Figure 1. Comparison of the Hemolysin A and Leukotoxin A Signal Sequences... 4 Figure2. Model of HlyA Recognition and Transport 6 Figure3. Design of the cci-Helical Library of HlyA Signal Sequence Variants.... 13 Figure4. The Genetic Code 14 Figure5. Construction of the ai-Helical Library of HlyA Signal Sequence Variants 15 Figure6. Results of the ELISA Assay of HlyA Signal Sequence Variants 27 A C K N O W L E D G E M E N T S I thank Drs. Fang Zhang, Sarah Childs and Doug Hogue for helpful discussions and suggestions. I also thank my supervisor Dr. Victor Ling for his guidance and the opportunity to work on this project. This research was supported by the Medical Research Council of Canada. I also thank Drs. M . Blight and LB. Holland for the gift of the polyclonal HlyA antibody. I would also like to acknowledge the support and encouragement of my mother and father, Karen, Keith, and my wonderful husband Darrell. Thank you. CHAPTER I: INTRODUCTION RTX Toxins Hemolysin B (HlyB) and other RTX toxin (Repeats in Toxin) transporters are part of the increasingly well studied ATP-binding cassette (ABC) transporter superfamily of ATP-dependent membrane transport systems (for reviews see Higgins, 1992; Childs and Ling, 1994; Sheps et al, 1995). These transporters consist of a conserved ATP-binding domain linked (covalently or non-covalently) to a block of 5 to 8 predicted membrane spanning helices. They may present as a functional complex of a single protein with a tandem repeat of two ATP-binding and transmembrane domains, or as a homo- or hetero-dimer of two molecules each composed on one ATP-binding and transmembrane domain. The best characterized group of prokaryotic A B C Transporters is the RTX toxin transporters (Felmlee and Welch, 1988). The RTX transport systems require an A B C protein (composed of an ATP-binding and a transmembrane domain), accessory factor and an outer membrane factor. The toxins secreted by these transport complexes are recognized via a C-terminal signal sequence and do not require a periplasmic intermediate. The RTX toxins are named for their 9 amino acid motifs rich in glycine and aspartate that repeat 9 to 40 times, L-X-G-G-X-G-N/D-D-X (where G represents glycine, L leucine, N asparagine, D aspartic acid, and X any amino acid), and importantly bind Ca which is crucial to their toxic ability. Many RTX toxins are involved in the lysis of erythrocytes and leukocytes via pore-formation, but it has been postulated that their leukocyte stimulating activities may be the more important factor in their role in pathogenesis (Bhakdi et ai, 1989; Bhakdi et al., 1990; Czuprynski and Welch, 1995). Diseases caused by microorganisms that secrete RTX toxins include, urinary cystitis, 1 whooping cough and shipping fever in cattle (Hacker et al, 1983; Glaser et al, 1988; Strathdee and Lo, 1989). The Escherichia coli Hemolysin Transport System The E. coli HlyA transport system is the best studied of the RTX toxin transport systems. Hemolysin A is a 107 kDa RTX toxin which is secreted by certain uropathogenic strains of E. coli (Gray et al, 1989). The hemolysin transport system of E. coli is distinct from the general export mechanism, the SecAY export system, as the transport of HlyA does not require a periplasmic intermediate and its secretion is directed by an uncleaved C-terminal signal sequence (Gray et al. 1989; Koronakis etal, 1989; Hess et al, 1990). The HlyA toxin is targeted to the transport complex by a 58 amino acid C-terminal signal sequence. HlyB, a member of the superfamily of A B C transporters, is involved in the energy-dependent transport of HlyA (Juranka et al, 1992; Wang etal, 1991). Hemolysin D (HlyD) and TolC (an E. coli outer membrane protein) form the transport complex along with HlyB (Wandersman and Delepelaire, 1990). The hemolysin C (HlyC) protein is involved in the activation of HlyA to its hemolytic form by transferring a fatty acyl residue to two specific internal lysine residues, K564 and K690, but this modification is not required for transport (Issartel et al, 1991; Stanley et al, 1994). The Hemolysin A C-Terminal Signal Sequence The signal sequence of HlyA is an uncleaved signal that was localized to the C-terminal 60 amino acids through deletions, gene fusions and point mutants, and does not resemble an N-terminal signal sequence (Hess, 1990; Kenny etal, 1991; Kenny etal, 1992; Koronakis etal, 1989; Stanley et al, 1991; Zhang et al. 1993a). It has been demonstrated using various chimeras that the C-terminal signal sequence can be used to secrete various heterologous 2 proteins. Further, it has been shown that these signal peptides alone, without the rest of the HlyA molecule, are sufficient for recognition and transport (Zhang et al, 1995). The HlyA signal sequence is highly tolerant to point mutations. While over 100 point mutants have been generated in the C-terminal region, only a handful have been demonstrated to decrease transport by more than 50% (Kenny et al, 1992; Zhang et al., 1993a; Sheps et al, 1995; Chervaux and Holland, 1996). Extensive deletions of the signal sequence are required before a severe reduction in transport efficiency is observed (Zhang et al, 1993a). It has been demonstrated that the HlyA signal sequence can be substituted functionally by a completely different primary sequence (Zhang et al. 1993b). The C-terminal 58 amino acids of the HlyA signal sequence can be functionally replaced by the C-terminal 70 amino acids of LktA (from Pastuerella hemolytica). The fact that these signal sequences have very little primary sequence homology, but share a similar predicted secondary structure, suggested that a higher ordered structure within the signal sequence of HlyA may be important for its recognition and transport (figure 1) (Zhang et al., 1993a). This finding also provided an explanation for the resistance of the HlyA signal sequence to inactivation by point mutations (Chervaux and Holland, 1996; Kenny etal, 1992; Sheps etal, 1995; Zhang etal, 1993a). CD and NMR using purified signal peptides of HlyA and LktA have confirmed the similarity of the predicted secondary structure, an amphiphilic helix-strand-helix motif (oci-strand-a2) (Yin et al, 1995; Zhang et al, 1995). Both these peptides exhibit similar biophysical properties. By circular dichroism (CD) analyses it was observed that both peptides are unstructured in an aqueous environment, but in certain membrane mimetic environments they assume a helical secondary structure (Zhang et al, 1995). A 2D NMR spectroscopy 3 3- § 9- 2 cv 3' 03 o CD cu 3- 3] o-^3 i » O CD oj O 3 CD o CD <9 § ^ § 0 3 ^ 3 -03 QJ Qj 0 0 ~ - CO ~ -^ 3' 3> 0 *8 3' o O , Q CD O O- Cl ^ Q.' V ; Co s . Co < -s -< -h ^ ~* 5 Q3 9J. 03 •= 3. £ o Cr Ti A. co CD C Q " CD CO Q, c •Pi ~ - S _ 3 DO 03 O- ~~ C D • Q3 . O CD CQ' a-S. 2 Q3 - CD O ~A 5. S: CO Q co DO 03 2 o 03 03 Q. co - - 3 ' CQ % 0 ~ CD CD 2 3 Q3_ CO C Q ' 03_ CO CD -Q c CD o CD -< CD CO D a CD r-7; 2 D H D 2 m — m CO — C O 7s —7s «55 r— I — 7s CO 7; CO D 7; CO CO — C O > D C O — c o ^ CD D > J C O — c o n lyA co " c ktA 0 CD 0 r—K CD —1 D" CD —1 3 ' D ' Q3_ Q3_ CO CO CQ CQ' 1 3 Q3_ Q3_ CO CO CD CD -Q - Q C C CD CD 1 3 O O CD CD CD < m m T J co $ T J H CO T l CO -< CD T J 2 CO 2 -co > CD 2 CD 7; 'H D O m r -CO 7s I T l -< m 1— |— 7; CO 7; — -< CD CO D O 5 I CO m O CO 7s — 7s | — CO — CO CO < > CD CO > 2 T l n -—1 D CO < CO 7s 2 m O m CO T J T J CO 2 > < > l ~ CO < 1— > 1— T J O —1 r— CO — CO ^ CD 1— . 2 0 > J 0 CO CO D 1— T l co -— CO CO -< 1— CD O 73 FAF NS 3 CO > o I r—t-CD 3 03_ CO C Q ' 0 3 _ CO CD - Q C CD O CD CQ C 3 A 3 study localized comparable regions of helices in both peptides (Yin et al, 1995). A similar finding of a helix-strand-helix motif has also been reported for the C-terminal 56 residues of Erwinia chrysanthemi protease G (Wolff et al, 1994) which is secreted by a transport system analogous to the hemolysin system. In these studies it was noted also that the helices form in an apolar environment. Such findings suggest that the ability to form an appropriate secondary structure in an membrane environment may be an important in the recognition and transport of HlyA (Yin et al, 1995; Zhang et al, 1995). The deletion of the a-helices within HlyA C-terminal signal region severely decreased transport, again indicating the possible importance of the signal sequence secondary structure in recognition and transport. A Model for HlyA Recognition and Transport These findings have led to the proposed model of HlyA recognition and transport as a working hypothesis in the Ling Lab (figure 2) (J. Sheps and F. Zhang personal communication). This model takes into account extensive mutational and biophysical analyses of the HlyA signal sequence undertaken by various laboratories as discussed above. The main features of the proposed model are: 1) The signal sequence of HlyA is targeted to the inner membrane lipids which allowed the helix-strand -helix motif to form prior to interacting with HlyB (Zhang et al, 1995). To initiate transport, the C-terminal signal sequence of HlyA must first fit into the binding pocket formed by the HlyB/D transport complex. This step would allow the HlyA molecule to search for the transport complex in two dimensional space along the plane of the membrane which is more efficient than in three dimensional space i.e. cytoplasm (Zhang etal, 1995). 5 HlyB Outer membrane Periplasmic space inner membrane ATPase domain Figure 2. This model for the recognition and transport of HlyA involves the following proposed steps: (1) interaction of HlyA with the inner membrane lipids via the C-terminal signal sequence, (2) induction of secondary structure formation of the signal sequence to fit into the binding pocket of HlyB, (3) contact of the ccj and a.2 helices in the signal sequence with specific residues in the binding pocket of HlyB to trigger the ATPase activity, (4) translocation of the HlyA molecule into the growth medium. 2) This binding pocket contains amino acid residues which function as "triggers" of transport through activating HlyB ATPase activity. Such triggers are stimulated by specific interactions with residues from the HlyA signal sequence. The number of potential triggers residues may be large and more than one trigger may need to be "pulled" before the HlyB ATPase is turned on to initiate the translocation process. This could account for the ability of peptides with completely different primary sequences serving as efficient transport signals (Zhang et al, 1993a). Such peptides presumably possess the necessary 3D shape to occupy the binding pocket but interaction with the triggers will necessarily be mediated by residues different from those of the HlyA signal sequence (Zhang et al, 1995). 3) The ATP binding domain of HlyB has intrinsic ATPase activity which, when activated, transduces energy for HlyA transport through a channel formed by the HlyB/D and TolC complex. Based on this model, it would be predicted that the required common higher order structure of the signal sequence reflects a 3D shape that it is able to fit into the binding pocket of the transport complex. As the LktA and HlyA signal sequences are functionally equivalent, they likely form a similar secondary structure in order to fit into the binding pocket of the transporter but since they have different primary sequences their interaction with specific residues of HlyB for triggering the intrinsic ATPase will likely be different, as perhaps a certain number of interactions is required. Thus, the interaction of the signal with of HlyB may require a specified secondary structure and may be multivalent, and that there may be many residues within the binding pocket which may function as ATPase triggers for transport of the toxin. 7 THESIS RESEARCH Based on the above observations, it is predicted that the maintenance of the secondary structure, but not necessarily the amino acid sequence, of the C-terminal signal of HlyA would be sufficient to support its recognition and transport. In order to test this hypothesis more definitively, I have taken the approach of generating variants in the cti-helical domain of the HlyA signal sequence. In this study, two combinatorial libraries of sequences were generated and used to test this hypothesis. The first library has the potential of yielding 1.5 x 108 predicted amphiphilic a-helical variants. It was substituted into the oti-helical region within the C-terminal signal of HlyA to determine the relative importance of specific amino acid residues within the context of an amphiphilic helix. If the hemolysin transport system is not sensitive to the amino acid sequence, most sequences generated will be able to support transport. On the other hand, if the signal sequence is dependent on a few critical amino acids, then few of the combinatorially generated sequences will be able to support transport. As will be detailed, results from this line of investigation provide evidence that the amphiphilic helical structure of the oci-helix is sufficient for recognition and transport, as all helical variants isolated were able to support transport. The second library of random sequences has the potential of producing 7.4 x 1015 variants. It was substituted into the cci-region within the C-terminal signal of HlyA to determine if the amphiphilic helical structure of this region is required for transport. If the hemolysin transport system is not sensitive to secondary structure, then variants may form an amphiphilic a-helical structure in the ai-region but yet may be unable to support transport. Additionally, the system may not be sensitive to secondary structure if variants that do not form an amphiphilic a-helical structure in the ai-region are able to support transport. The 8 results described herein provide evidence that the amphiphilic helical structure may be required for recognition and transport, as all variants able to support transport are predicted to form an amphiphilic helical structure in the oti-region, while those variants unable to support transport, are not predicted to form an amphiphilic helical structure in this region. In a broader context, this study HlyA recognition and transport allows further insight into one of the most captivating problems in the A B C transporter family, that of substrate specificity. A B C transporters such as the multidrug transporter P-glycoprotein, have the capacity to transport a wide range of drugs and peptides, and may also function as a chloride channel or as an ATP channel. In contrast, the HlyB/D transporter has one native substrate, HlyA. Suprisingly, upon further investigation, this transporter has also been shown to have a capacity to transport a wide variety of proteins when attached to the HlyA C-terminal signal sequence, LktA C-terminal signal sequence, and a variety of signal sequence variants. The challenge of the A B C transporter family, is to understand how they can accommodate such a wide variety of substrates, within what is likely a conserved transport mechanism. 9 C H A P T E R II: M E T H O D S AND M A T E R I A L S Bacterial Strains and Plasmids Escherichia coli strains JM83( X' ara A(pro-lac)rpsLthi §-%0dlacZ A Ml5 X~), DH5cc (F- supE44 hsdRll' recA, gyrA96 endA, thi-1 relAi deoR X-), and TOP10 (F- mcrA A (mrr-hsdRMS-mcrBC) <|> -80 lacZ-A-M15 A-lacX74 deoR recAi araD139 A-(ara-leu)7679 galU galK rpsL(StrR) endAj nupG) were used for all experiments. The hemolysin system was expressed in JM83 using pUCAC494, which encodes HlyA and pLGBCD which encodes HlyB, D, and C. HlyC is required for the addition of a fatty acyl moiety to HlyA, a covalent modification required for hemolytic activity, which does not effect transport (Issartel et al., 1991). Where appropriate, cells with plasmids were grown in the presence of 34 pg/ml chloramphenicol and 50 pg/ml ampicillin. Routine DNA purification and manipulations were as described (Sambrook et al., 1989). Construction of Plasmids The pUCAC494 plasmid containing the hlyA gene was constructed as follows. pUC19 was digested with Sail and Hindlll and blunt end ligated, this plasmid was then digested with BamHI and Kpnl, and the 3.06Kb plasmid was ligated to a BamHI-Kpnl 2.9Kb fragment (hlyA gene) from pLG583SK previously described (Holland et al., 1990; Zhang et al., 1995). pUCAC494 was digested with BgUI and Sail, sites located in the hemolysin A C-terminal region. The large fragment of pUCAC494 was isolated by the method of Favre and ligated to a double stranded adapter encoding 3 stop codons (forward adapter: 5'-T C G A C G C G G C C G C T T T A A T A G T G A T C G A T G A - 3 ' , reverse adapter: 5'-G A T C T C A T C G A T C A C T A T T A A A G C G G C C G C G - 3 ' ) (Favre, 1992). The adapter insertion 10 was confirmed by DNA sequencing using Sequenase (U.S. Biochemical Corp) as per the manufacturers instructions, and the resulting plasmid, designated pUCAC494-adapter, contains a hlyA gene with a 58 C-terminal amino acid truncation. Construction of the cti-Helical Library of HlyA Signal Sequence Variants To construct the library, a series of 4 overlapping synthetic oligonucleotides (Gibco-BRL), one being degenerate for the ai-helix, were used. Oligonucleotide A 5'-A T G A T G C G T C G A C T T A T G G G A G C C A G G A C A A T C T T A A T C C A - 3 ' , oligonucleotide B 5'-A A C A T C G A A G T T A C C B B C A G Z r N B r T N A N N A N N T B r N B r T N A N N T B r N T B r N A N N A N T G G A T T A A G A T T G T C - 3 o l i g o n u c l e o t i d e C 5'-G G T A A C T T C G A T G T T A A G G A G G A A A G A T C T G C C - 3 ' . and oligonucleotide D 5 ' - G G C A G A T C T T T C C T C C T T - 3 ' (the underlined portions indicate overlapping regions). For the degenerate oligonucleotide B, N represents a, g, t or c; B represents a, g, or c; B r represents t, c, or g; and Z r represents a, t, or c. Equimolar concentrations of the four oligonucleotides in buffer (10 mM Tris, pH 7.5, 50 mM NaCI, 10 mM MgCl2, and 5mM dithiothreitol) were mixed and heated (95°C, 3 min), and left to cool to room temperature, and then placed at 4°C. The annealed oligonucleotides were incubated (4 hrs, 37°C) with large fragment of DNA polymerase I and T4 DNA ligase and the enzymes heat inactivated (65°C, 5 min). The ligation mixture is then digested with BgUI and Sail and double-stranded oligomers were purified using the Qiagen QIAquick Nucleotide Removal Kit. The mixture of double stranded oligonucleotides was cloned directionally in the pUCAC494-Adapter plasmid that was digested with Bgllll and Sail. The Pst I site 11 (nucleotide position 541) had been destroyed in the variants, this provided an additional measure with which to screen the library. Design of the aj-Helical Library of HlyA Signal Sequence Variants To investigate the role of amino acids within the first helix (cti) in the helix-strand-helix motif of the HlyA C-terminal signal sequence, a combinatorial library of a i-helices that maintained the amphiphilic helical structure of the helix was designed. The combinatorial approach applied is based on a general strategy described by Kamtekar which periodicity of the polar and nonpolar residues is conserved to maintain amphiphilic helical structure while the identity of the individual residues is varied (Kamtekar et ai, 1993). The principle is to arrange the periodicity of the nonpolar and polar residues to approximate the 3.6 residue repeat in such a manner that polar residues and nonpolar residues reside on opposite sides of the helix (figures 3 and 4). A series of 4 overlapping oligonucleotides, one of which was degenerate, encoding for the aphelix was used to construct the library (figure 5). The degenerate oligomer was designed in the following manner. Nonpolar amino acids utilized the degenerate codon NtN (where N represents a mixture of a, g, t, and c). With the NtN codon, positions requiring a nonpolar amino acid are then filled by Phe, Leu, lie, Met, or Val. The degenerate codon BaN (where B represents a, g, and c) was used for the polar amino acids, excepting Ser and Thr. These positions are filled by, Glu, Asn, Asp, Lys, Gin, and His. The degenerate codon aBN was used for Ser and Thr. These positions are filled by Ser, Thr, Asn, Lys, Arg. The first nucleotide in the 10th codon of the ai-helix was changed from t to a in order to circumvent the possible generation of three stop codons. This variation maintains charged polar residues at this position. As the 11th and 12th amino acids in Figure 3 Design of the ai-Helical Library of HlyA Signal SequenceVariants Amino Acid Sequence of the cti-Helix HlyA Sequence L I N E I S K I I S A A Nucleotide Sequence NtN NtN BaN BaN NtN aBN BaN NtN NtN aBN Zct gBrBr L L E E L S E L L S S G Variety of I I D D I T D I I T T V Amino Acids M M K K M N K M M N A A Possible F F N N F K N F F K V V Q H Q H V R Q H V V R Figure 3. Nonpolar amino acids utilize the degenerate codon NTN (where N represents a mixture of a, g, t, and c). With the NTN codon, positions requiring a nonpolar amino acid are then filled by Phe, Leu, He, Met, or Vol. The degenerate codon BAN (where B represents a, g, and c) represents polar amino acids, excepting Ser and Thr. These positions are filled by, Glu, Asn, Asp, Lys, Gin, and His. The degenerate codon ABN replaced the Ser in the aj-helix. These positions are filled by Ser, Thr, Asn, Lys, Arg. The 11th amino acid position in the ai-helix is replaced by the degenerated codon Zct (where Z represents a mixture of t, a, and g) which generates the polar (neutral) amino acids Ser and Thr as well as Ala. The 12th amino acid position is replaced by the degenerate codon gBrBr (where Br represents t, c, and g), which generates the nonpolar amino acids Gly and Vol, as well as Ala. 13 Figure 4 The Genetic Code First Second Position Third Position Position (3' (5' end) end) U C A G Phe (F) Ser (S) Tyr(Y) Cys (C) U U • Phe (F) Ser (S) Tyr(Y) Cys (C) c Leu (I.) Ser (S) Stop Stop A Leu(L) Ser (S) Stop Trp (W) G Leu.(L) , Pro (P) His (H) Arg(R) U c Leu.(L)" Pro (P) His (H) Arg(R) c Loud.) Pro (P) Gin (Q) Arg (R) A Leu (I.) Pro (P) Gin (Q) Arg (R) G He (I) Thr (T) Asn (N) Ser (S) U A He (0 Thr (T) Asn (N) Ser (S) C He (I). '• • ' Thr(T) Lys (K) Arg(R) A Met (M) Thr(T) Lys (K) Arg(R) G Val (V) Ala (A) Asp (D) Gly(G) U G Val (V) Ala (A) Asp (D) Gly(G) c Val(V) Ala (A) Glu (E) Gly (G) A Val (V) Ala (A) Glu (E) Gly (G) G Figure 4. If one position of the bases in a codon is held constant, while the other two are varied, it is possible to generate nonpolar codons or polar codons as is described in figure 3. The gray boxes indicate nonpolar amino acids generated by the codon NtN, where N represents any one of the four nucleotides. (Stryer, 1988). 14 Figure 5 Construction of the ai-Helical Library of HlyA Signal Sequence Variants Sail Site 5' at gat peg tcq act tat ggg age cag gac aat ctt aat cca ggt aac ttc gat gtt aag gag gaa aga tct gec 3' Figure 5. Four oligonucleotides, one being degenerate for the aj-helix region, are used in the construction of the helical library of HlyA signal sequence variants. For the degenerate oligonucleotide B, N represents a, g, t or c; B represents a, g, or c; Br represents t, c, or g; and Zr represents a, t, or c. 15 the ai- helix are alanine, they must be replaced by polar and nonpolar amino acids, respectively, in order to maintain the amphiphilicity of the helix (figure 3). Therefore a total 5 3 2 2 8 of 5J x 6J x 5Z x r = 1.5 x 10° different amino acid sequences are theoretically possible in this library. The 4 oligonucleotides were ligated and complementary strands were synthesized to the non-overlapping regions and directionally cloned into pUCAC494-adapter to generate a plasmid library representing a full length HlyA variant. Construction of the ai-Random Library of HlyA Signal Sequence Variants To construct the random library two different methods were used. The first method involved two overlapping oligonucleotides, one being degenerate for the ai-helix. Oligonucleotide Random 100 5' A T G A T G C G T C G A C T T A T G G G A G C C A G G A C A A T C T T A A T C C A N N N N N N N N N N N N N NNNNNNNNNNN^ G A T G T T A A G G A G G A 3' and oligonucleotide DI 5' G G C A G A T C T T T C C T C T T C A A C G T C G A A G T T A C C 3' (the underlined portions indicate overlapping regions). Equimolar concentrations of the two oligonucleotides in buffer (10 mM Tris, pH 7.5, 50 mM NaCl, 10 mM MgCl 2 , and 5mM dithiothreitol) were mixed and heated (95°C, 3 min), and left to cool to room temperature, and then placed at 4°C. The annealed oligonucleotides were incubated (1 hr, 70°C) with Taq DNA polymerase, purified using a Qiaex II Gel Isolation Kit according to manufacturer's instructions, and then digested with BgUI and Sail. The double-stranded, digested oligomers were purified using the Qiagen QIAquick Nucleotide Removal Kit (according to manufacturer's instructions). The mixture of double stranded oligonucleotides was cloned directionally in the pUCAC494-Adapter plasmid that was digested with Bgllll and Sail. 16 The second method involved a series of 4 overlapping oligonucleotides one being degenerate for the a i-helix was used. Oligonucleotides A, C and D were used as described in the construction of the combinatorial library with the replacement of the degenerate oligonucleotide B with oligonucleotide HABrandom 5'-A A C A T C G A A G T T A C C N N N N N N N ^ GGATTAAGATTGTC-3 ' (the underlined portions indicate overlapping regions). The library was then created in the same manner as the combinatorial library. Design of the ai-Random Library of HlyA Signal Sequence Variants To further investigate the role of the a i-region of the hemolysin A C-terminal signal sequence, a random library of sequences to replace the amphiphathic ai-helix was designed. Two approaches to creating the random library were employed. The first method involved two oligonucleotides, one of which was degenerate, encoding for the ai-region (figure 3). The degenerate oligonucleotide was designed with the codon NNN replacing each of the 12 codons in the ai-region (where N represents a,g,t or c). The second method involved a series of four oligonucleotides, one of which was degenerate, encoding for the oci-region. The degenerate oligonucleotide was also designed with the codon NNN replacing each of the 12 codons in the ai-region (where N represents a,g,t or c). In both methods, the oligonucleotides were ligated and complementary strands were synthesized to the non-overlapping regions, and directionally cloned into pUCAC494-adapter to generate a plasmid library representing a 12 full length HlyA variant. Therefore from the 20 amino acids and 3 stop codons, a total of 21 = 7.4 x 1015 different sequences are theoretically possible in this library. 17 PCR Screening of the aj-Helical Library of HlyA Signal Sequence Variants The pUCAC494 library of helical variants was transformed into JM83 cells that contained pLGBCD, and plated out (1 .Ox 102 cfu/plate) on 20 mL LB agar plates containing ampicillin and chloramphenicol, with the top 10 mL supplemented with 5% sheep red blood cells. All colony forming units on a given blood agar plate were analyzed by PCR, regardless of the size of hemolytic zone in order to generate, an unbiased library of helices. Direct colony PCR amplification was performed to screen the potential helical variants using an internal primer HA23 (5 ' -GACGGCAGGGTAATCACA-3'; nucleotide position 380-397 of HlyA) and M13R1 (-40) (Lee and Cooper, 1995). The HlyA mutants isolated that did not contain the double stranded oligonucleotide insert yielded PCR products that were approximately 63 to 93 bp smaller than those with inserts. Mutant plasmids, containing the double stranded oligonucleotide insert were subjected to DNA sequencing using Sequenase and Thermosequenase (Amersham Life Science) as per manufacturers instructions, and PCR primers. PCR Screening of the a,] -Random Library of HlyA Signal Sequence Variants The pUCAC494 library of random variants was electroporated into JM83 or DH5a cells that contained pLGBCD, and plated out (1.0 x 10 cfu/plate) on 20 mL LB agar plates containing ampicillin and chloramphenicol, with the top 10 mL supplemented with 5% sheep red blood cells. Direct colony PCR amplification was performed to screen the potential cti-variants by the same method as used in the helical library (Lee and Cooper, 1995). The HlyA mutants isolated that did not contain the double stranded oligonucleotide insert yielded PCR products that were approximately 63 to 93 bp smaller than those with inserts. The PCR products were then isolated using the Qiagen PCR Purification Kit. The PCR products 18 containing the double stranded oligonucleotide insert were subjected to DNA sequencing using a Perkin Elmer Automated Sequencer and both the dRhodamine and Big Dye Kits (Perkin Elmer) as per manufacturer's instructions. An internal primer aSEQ (5-GACGGCAGGGTAATCACACC-3*; nucleotide position 380-399 of HlyA) and M13R1 (-40) were used in sequencing the PCR products. Electroporation of Random Library of HlyA Signal Sequence Variants DH5a and JM83 cells containing pLGBCD (hlyB, C, and D) were transformed using electroporation (BIORAD Gene Pulser, and Pulse Controller) with the pUCAC494-random library plasmids as per the manufacturer's instructions. Blood Agar Plate Hemolysis Assay JM83 cells containing pLGBCD (hlyB, C, and D) were transformed with pUCAC494 (wild type hlyA) (positive control), a 58 amino acid truncation of the HlyA C-terminal signal sequence which deletes both the cti and 0 C 2 helices (negative control), as well as the library of HlyA helical variants. The transformants were plated at a density of 1.0 x 10 cfu on 20 mL LB agar plates containing the antibiotics ampicillin (50"ug/mL) and chloramphenicol (34 pg/ml) and the top 10 mL supplemented with 5% defibrinated sheep blood and were grown at 37°C for 10-12 hours. The colonies were assessed for hemolytic activity based on the size and brightness of the zone of lytic clearing as determined by visual inspection. The zones were assessed relative to wild type hemolysin A (positive control), a 58 amino acid truncation of the HlyA C-terminal signal sequence which deletes both the cti and C C 2 helices (negative control). Comparisons are made between colonies of similar size. Isolated helical variants were plated on blood agar medium in triplicate, and the results confirmed by an independent observer. 19 Subcloning of the HlyA Signal Sequence Variants The BgllII-a\-fos\\x-SalI inserts of selected mutant plasmids were isolated and re-introduced into Bglll/Sall digested pUCAC494-adapter and then re-assessed for hemolytic ability as previously described. ELISA Assay Quantitation of HlyA Transport An ELISA assay of HlyA was performed using a polyclonal antibody against HlyA (a gift from Drs. M . Blight and LB. Holland, Institut de Genetique et Microbiologie). Cultures were grown to O D 6 0 0 0.800, and the supernatant collected after centrifugation (3 000 x g, 5 min). The supernatant was then concentrated using Centricon-30 filtration columns (Amicon). Supernatants were applied to immulon-2 microtitre plates (Dynatech labs) and incubated at 37°C for 2 hrs; the wells then were blocked with 1% BSA, Tween-20 in phosphate-buffered saline; the wells were probed with anti-HlyA diluted 1:5 000, followed by horse radish peroxidase (HRP) conjugated anti-rabbit IgG diluted 1:10 000. Washes with 3 x 300 mL washing buffer were performed after each step to remove unabsorbed protein/antibody. Absorbed HRP conjugated anti-rabbit IgG was detected using ABTS (Sigma) and A430 readings taken. The following controls were used: JM83 with pLGBCD, JM83 with pLGBCD and pUCAC494, no anti-HlyA antibody, no HRP-conjugated Anti-Rabbit IgG, neither anti-HlyA or HRP-conjugated Anti-Rabbit IgG. Note that all ELISA and Blood Agar Plate Hemolysis Assay analysis was performed using the strain JM83. Also, in both assays, wild-type HlyA and HlyA variants are expressed from a high expression promoter and, as has been previously established by Zhang et al., slight variations in transcription do not account for variation in the amount of HlyA transported (1993b). Rather, 20 it is the amount of HlyB which is the rate limiting step in recognition and transport (Zhang et al, 1993b). Statistical and Sequence Analysis The hydrophobic moment analysis was performed with Peptool (vs 1.0) using the Eisenberg consensus scale (Eisenberg et al, 1984). Calculation of the hydrophobic moment (m) was performed using the following equation: m 2 = Z Hn(sin On) + Hn(cos 0 „ ) 2 where m is the hydrophobic moment, Hn is the hydrophobicicty of residue n, and O is the angle between successive side chains viewed along the central axis of the assumed structure (characteristic hydrophobic periodicity). Helices typically have one hydrophobic residue for every 3.6 residues and beta-strands have a repeating pattern of hydrophobic and hydrophilic residues. The hydrophobicities (Hn) are normalized in order that the mean value of the 20 amino acids is 0.00 and the standard deviation is 1.00. The hydrophobic moment is calculated by the Fourier transform of the hydrophobicity of an amino acid sequence (Evans, 1997). The Eisenberg consensus scale of 1984 (is based on 5 other scales) is the table used by the hydrophobic moment algorithm to determine the hydrophobic degree of similarity between amino acid residues (Eisenberg et al, 1982; Eisenberg et al, 1984). A 7 residue window is used for the helix and beta hydrophobic moment. The hydrophobicity analysis was performed with Peptool vs 1.0 also using the Eisenberg consensus scale (Eisenberg et al, 1984) and a 7 residue window. The hydrophobicity algorithm of Peptool was calculated using the Eisenberg consensus scale of 1984 to determine the hydrophobic degree of similarity between amino acid residues (Eisenberg et al, 1982; Eisenberg et al, 1984). Mean values were compared using the Student t-test. Amino amino acid distribution has been analyzed with the cumulative binomial test, with the expected probabilities of comparison calculated from the number of possible codons that may generate a given amino acid in the two libraries of variants. The position-specific amino acid compositions were analyzed as a probability that the calculated deviation would occur in a binomial distribution (Samuels, 1989). Secondary structure predictions were performed using AGADIR and SSCP. AGADIR predicts the helical content of peptides. It is an algorithm based on the helix/coil transition theory (standard conditions of ionic strength, pH, and temperature are used in the calculation) (Muiioz and Serrano, 1994a; Munoz and Serrano, 1994b; Munoz and Serrano, 1997). SSCP (Secondary Structural Content Prediction) predicts the content of helix, strand, and coil using amino acid composition, and involves analytic vector decomposition methods applied on the composition vector of the query protein (Eisenhaber et al, 1996a; Eisenhaber etal, 1996b; Eisenhaber etal, 1995). 22 CHAPTER III: RESULTS AND DISCUSSION Screening of the cti-Helical Library HlyA Signal Sequence Variants To investigate the role of amino acids within the first helix (cti) in the helix-strand-helix motif of the HlyA C-terminal signal sequence, a combinatorial library of ai-helices that maintained the amphiphilic helical structure of the helix was designed. The combinatorial approach applied is based on a general strategy described by Kamtekar which periodicity of the polar and nonpolar residues is conserved to maintain a predicted amphiphilic helical structure while the identity of the individual residues is varied (Kamtekar et al., 1991) (figures 3 and 4). The HlyA variant pUCAC494 library was co-transformed with pLGBCD, which encodes hlyB, C, and D, into JM83 cells. Transformed JM83 cells were screened for hemolytic activity by the blood agar plate assay which indirectly examines the ability of the transport complex to recognize and secrete HlyA. Colonies that secrete HlyA have a clear zone of lysis (hemolytic zone) around the colony. All colonies on a given blood agar plate, regardless of the size of hemolytic zone were subjected to direct colony PCR to confirm the presence of the double oligonucleotide insert. Those containing inserts produced PCR fragments approximately 63 to 93 base pairs larger than those lacking inserts, a size differential is accounted for by both blunt ended ligations and incompletely digested plasmid. Of the 1400 colonies screened by PCR, only 54 isolates were found to have inserts of the correct size, and these isolates fell into three categories; those with nonsense insertions, those with inserts that were inconsistent with the design parameters, and those with the correct insertion. A total of 24 of these 54 isolates, upon DNA sequencing of the region encoding 23 the ai-helix, had the expected periodicity of polar and nonpolar residues predicted to maintain the amphiphilic helical structure of the ai-helix. In an attempt to explain the large proportion of colonies screened by PCR that were not found to have inserts, digested pUCAC494-adapter and digested pUCAC494-adapter that had undergone the identical ligation conditions in the absence of the insert were both separately transformed into JM83 in the presence of and absence of unannealled single stranded oligonucleotides. A greater number of background transformants were observed when the plasmid was incubated with unannealled oligonucleotides in comparison to those incubated in the absence of unannealled oligonucleotides. This may help to explain the high background of isolates without inserts. In addition the low efficiency of insertion may be accounted for by the technical difficulty of annealing four oligonucleotides. Functional Analysis of the aj-Helical Variants with the Hemolytic Zone Assay The relative degree of hemolytic activity exhibited by colonies on blood agar plates provides a convenient assay for testing the efficiency of HlyA secretion. The relative size of hemolytic zones around colonies plated on blood agar plates was graded from the largest zone (++++), which corresponds to the wild-type HlyA phenotype (positive control), to the smallest (+), which corresponds the phenotype of a 58 amino acid truncation of the C-terminal signal sequence which deletes both the cti and 0,2 helices (negative control), and also no detectable zone (-) (see table 1). Hemolytic zones of the isolated helical variants were then assessed relative to these controls. Isolated helical variants were plated on blood agar medium in triplicate, and the results confirmed by an independent observer. The summary of Table 1 Amino Acid Sequence and Zone Assay Analysis of the ai-Helical Library of HlyA Signal Sequence Variants Description Phenotype Amino Acid Sequence of the ai-Region HlyA +++++ 1. 1 N i- I S K I 1 S A A P63N2 ++++ V V D D L T N F V S T A P48C3 ++++ V L E Q F T E F L T S V P48C4 ++++ V 1 E Q F N D V L T S V P32C1 ++++ V F E D V T E F V K S A P48C1 ++++ V I E Q F T D V L T S V P48C2 ++++ V 1 F D F T D V F T A V P6301 ++++ 1 I; N . H F T N M L T S G P3C13 +++ V 1 E Q F T N V L T S V P35C1 +++ F F E Q F R D F F N A G P63Q2 +++ mi L D N F S D L L S S V P63C3 +++ i. F D Q L T N F V S S V P5C2 +++ V L H D V S D F F K S V P63C49 +++ F F H F. L N N M L R S V P4C1 +++ V V H Ii I S H V F S A V P30C1 +++ 1 F H N F S E F L T s V P6C2 +++ 1. L H H L S Q F F T s V P63,22C5 +++ I F N N F S H M L T A V P35C5 ++ V L E F. F s D V F T s G P63W2 ++ F F H D V R D V L S A A P3C11 ++ F V D D V T N F F T s V P5C1 ++ V I E K I T Q 1 V R T A ' P5C8 ++ 1. L Q Q L R E F 1 T s G P63C15 ++ F F K H L R H L L S A V P32C4 ++ F L Q E L R D F L T s V Table 1. Amino acid sequences of the aj-helical library of HlyA signal sequence variants. The gray boxes represent amino acids identical to the native oci-helix sequence of hemolysin A, which is listed in the first row of the table. In the zone assay analysis, the relative size of the hemolytic zones around colonies plated on blood agar plates is graded from the largest zone (+ + + + +) as seen in wild type hemolysin A to the smallest (+) which is a truncation of the C-terminal 58 amino acids of the HlyA signal sequence. 25 the results is shown in table 1. All a i-helical variants were able to support transport to a greater degree than that of a 58 amino acid truncation of the C-terminal signal sequence. Quantitation of Hemolysin A Recognition and Transport using the ELISA Assay An ELISA assay was developed to quantitatively measure the level of HlyA in the culture supernatant. The liquid hemolysis assay was not used since it has previously been found to be subject to variation. Cells containing pLGBCD (having hlyB,D and Q alone were used as a base line while cells containing pLGBCD and pUCAC494 (hlyA) were used to establish a standard of 100 percent secretion. The ELISA was performed on representative helical variants to determine the amount of secreted HlyA, which directly corresponds to the ability of the a i-helix within the C-terminal signal to function in the recognition and i subsequent transport of HlyA, by the transport complex (figure 6). The amount of HlyA protein in the culture supernatant representing each helical variant corresponded to the relative ranking of its hemolytic zone size and brightness as observed in the blood agar assay (table 1). The ai-Helical Library of HlyA Signal Sequence Variants It was postulated that if the recognition and transport of HlyA is not sensitive to amino acid sequence but rather the secondary structure of its C-terminal signal sequence, most oti-helical variants generated would be able to support transport. Alternatively, if primary sequence is important, then few a i-helical variants generated would be able to support transport and the work of Zhang et al. may have fortuitously chosen a different primary sequence (i.e. LktA) with critical amino acids or characteristics (1993b). The results described herein demonstrated that all helical variants isolated were found to 26 f CP 0\ Percentage of Absorbance in Comparison to Wild Type HlyA O m r > > V) W Bi < o tn CD ro o p+ (D Q. X 2. n SL < 0) CD 3 I I I 27 Figure 6. In the blood agar plate assay analysis, the relative size of the hemolytic zones around colonies plated on blood agar plates is graded from the largest zone (5) as seen in wild type hemolysin A to the smallest (1) which is a truncation of the C-terminal 58 amino acids of the signal sequence. (Note that zero corresponds to the zone observed in the presence of HlyB/D but the absence of HlyA). The ELISA assay results are represented as a percentage of transport based on wild type HlyA in the presence of HlyB/D to establish a standard of 100 percent secretion. Variants used in the analysis are listed in order of their appearance on the X-axis starting at the first phenotype category 4; P63N2, P48C3, P3C13, P63C49, P3C11, and P5C1. For the phenotypic category 1, a C-terminal 58 amino acid truncation of the signal sequence was used, and for category 0, HlyB/D were expressed in the absence of HlyA. 28 support transport. Seventeen of the twenty-four variants isolated had a phenotype of (+++) or higher corresponding to a degree of transport approximately 67% of that of wild type HlyA, while the remaining seven variants had a phenotype of (++) corresponding to a degree of transport approximately 50% of that of wild type HlyA as calculated from the ELISA results (figure 6). Among these 24 helical variants, the number of unchanged amino acids in the ai-helix did not correlate with transport efficiency established by the hemolytic zone analysis and ELISA assays. For example, variant P48C3 (with no conserved residues) results in a higher degree of transport than variant P4C1 in which six residues were conserved (those identical to the original sequence) (table 1). Two variants were initially found to have hemolytic zones smaller than that of a 58 amino acid truncation of the C-terminal tail. However, upon subcloning of these variants back into the pUCAC494-adapter plasmid, the variants were found to support transport. Therefore it is likely that there had been mutations in the calcium binding motif or sites of modification by HlyC that would have affected the toxin's hemolytic activity. Screening of ai-Random Library of HlyA Signal Sequence Variants To further investigate the role of the a i-region of the HlyA C-terminal signal sequence, a random library of sequences to replace the amphiphathic ai-helix was designed. Two approaches to creating the random library were employed as described in methods and materials. Using the first method, the plasmid library was electroporated into JM83 with pLGBCD, and colonies were screened for their ability to support transport using blood LB plates as previously described. Three variants, RCA7, RCH, and RCN, were selected and sequenced (table 2). In the second method described for creating a plasmid library of 29 Table 2 Amino Acid Sequences and Zone Assay Analysis of the ai-Random Library of HlyA Signal Sequence Variants Variant Zone Assay Analysis Amino Acids Sequence of the on-Region HlyA +++++ L I N E I S K I I S A, A RCA7 +++++ 1. 1 Y E I S K I .1 S A A. R C H ++++ F V D V V T N F F T S V B92 +++ F V H S H V F .s" A ; V R C N ++ F V D D V T N F F T S V 86 + Q G S H N s. K P K A H Q B l + s K R E. Q T v Q G E K K B29 + R H R P R Y E P Y T T I D121 + V R P P N T K Q K T I K D125 + E R D L s S Q T E K H T 47 - H T A P H T N K P Q K K Table 2. Amino acid sequences of HlyA signal sequence variants from theai-random library that conform to the design parameters outlined in materials and methods. In the zone assay analysis, the relative size of the hemolytic zones around colonies plated on blood agar plates is graded from the largest zone (+ + + + +) to (+) which represents a 58 amino acid truncation, and also (-), which represents no detectable zone. Variant 47 had a zone smaller than that of a 58 amino acid truncation. The gray boxes represent amino acids identical to the native aj-helix sequence of hemolysin A which is listed in the first row of the table. 30 random ay-region sequences, the strain DH5a rather than JM83 (which is recA+) was used to reduce the chances of mutation in the gene of interest. Using this method, colonies were randomly selected in order to obtain variants that were and were not able to support transport. This method generated 62 variants, seven of which did not contain stop codons, insertions or deletions in the ay-region, nor in the last 100 amino acids at the C-terminus. Twenty-five of the variants were found to have a Leu-Ile (ctt-att) mutation three amino acids upstream from the cci-region (one of these variants had a 2 nucleotide deletion in the ai-region, and thirteen of these variants had at least one stop codon in the oci -region). An additional twenty-six variants were found to have stop codons, deletions or insertions in the a i-region (eleven had at least one stop codon). Two variants created a Bglll site in the 12th position of the cti-region, thus deleting eight amino acids and two nucleotides immediately downstream from the ai-region. Additionally, one variant has a deletion and another variant had an insertion outside ct/-region. Overall, the two methods yielded 10 variants that corresponded to the design parameters previously outlined (tables 2 and 3). Functional Analysis of the ai-Random Library of HlyA Signal Sequence Variants with the Hemolytic Zone Assay Hemolytic zone assays were performed on the a i-region random variants as per the method described for the helical variants isolated from the helical library. The summary of the results is shown in table 1. Four of the ten variants were shown to be able to support transport to a greater degree, than that of a 58 amino acid truncation of the C-terminal signal sequence. While six variants were not able to support transport, with a 31 Amino Acid Sequences of Hly A Signal Sequence Variants in the arRandom Library that do not Conform to the Design Parameters Table 3 a Variants 42 44 87 90 108 201 207 A4 A5 A18 B21 B22 B26 B80 D123 D124 D128 D141 D155 D165 D169 D175 D176 D83 B5 B57 B81 D94 B93 P7AC8 A W W I S * Y C T E K N Y E * L G V P * P E G D W N G Y S S A A A C L D K A K L P P F L * T Q Q P * p P Q D S M T R Q Amino Acid Sequence of the ai-Region L C V R K H A S C K Q A E T S K E K Q A E T S K E N R L L Y C M L S - -L Q V T S M L R R K Q A Q Y I E V T S M L R R K P G I S N R F Q - -R N K * Y S G E T L L L T * G K H I * F T T R Q T L L L -Y A G S K C - - - -Q Q I W N T L K - -V G L R R N I T R T L S L R G * T R S E Y Q G W * T N V G K N R Q P A R I N V E T * S A T * Q E P A R D K L - - -W I N E K R - -Y Q G W * T N V G N E P I A F Y Q K * K E L Q R I Q * S T I T H K Q K T l D T T R Q D L T T l G T T Q R Q L Q I K Table 3b Leu-Ile Variants Amino Acid Sequence of the ai-Region 101 L R A * R R K H K K T K 106 I A G S F M H M S S N A 109 S * R Y K V K S N S A Q 210 N A R H * S R 1 Q K Y Q 213 L R G * R R K H K K T K 215 D A Q Q S I E * * K N T 219 I A S G I * N R F Q N K 50 C A A T V * I N R T A L 81 T K I R Y S P Y L N N D 82 Q Q S H N S K P N A H Q 99 S L R I I T Q K F G V Q A17 • p A D K L L A R C Q S T A4 D A Q Q S I E * * K N T A6 D A Q Q S I E * * K N T A8 P A D T L L A R c Q S T B85 R I R P R K I K p G K -B87 P G S F S N N M S T T Y B52 G * S * T R D T K T S D DI L L T Q D * K N R * T D116 F G N T Y W * P R S K H D120 P T K S A K T S N K K K D126 G P Q S V D K H L R K K D133 Q S A H F V N H A R E K D153 V R P * Y M R L K R G N D161 R * * A C R G N L P V P Table 3. Amino acid sequences of the «/-random library of HlyA signal sequence variants that do not conform to the design parameters. Stop codons are designated by a '*', and deletions by dashes. Table 3a summarizes variants with mutations, insertions or stop codons in the ai-helix region. As well, variant B93 listed in table 3a has a codon deleted outside the ai-region, while variant P7AC8 listed has a codon insertion outside the arregion. Table 3b summarizes variants with a Leu-Ile mutation 3 amino acids upstream from the arregion. 32 degree of transport the same (five variants) as or less (one variant) than that of a 58 amino acid truncation of the HlyA C-terminal signal sequence. The ai-Random Library of HlyA Signal Sequence Variants Based on the results from the a i-helical library, it was postulated that if the recognition and transport of HlyA required an amphiphilic helical structure of the oci-region of the C-terminal signal sequence, variants able to support transport would be predicted to have an amphiphilic a-helical structure, and variants unable to support transport would not be predicted to form an amphiphilic a-helical structure. The results support this hypothesis, as the 6 variants unable to support transport are not predicted by the method of Kamtekar et al. to form an amphiphilic a-helical structure, while the 4 variants able to support transport are predicted to form an amphiphilic a-helical structure and have the expected periodicity of polar and nonpolar amino acids seen in a helical wheel format (table 2, figure 3) (1991). It is also interesting to note that 4 of the 6 variants unable to support transport have prolines present in the ai-region, which also suggests that the a\-region does not form a-helices in these variants. The Relationships of Helix Hydrophobic Moment, Hydrophobicity, and Beta Hydrophobic Moment with the Ability to Support Transport The average mean helix hydrophobic moment correlates with the ability to support transport (table 4). In the random library, there is no average helix hydrophobic moment below 0.304 among variants able to support transport, while there is no average helix hydrophobic moment above 0.240 among variants unable to support transport. And the variant 47, with an ability to support transport lower than that of a 58 amino acid C-terminal 33 Table 4a Helix Hydrophobic Moment of Randomly Generated Variants Able to Support Transport Mean Helix Variant Position Hydrophobic Moment 1 2 3 4 5 6 7 8 9 10 11 12 R C N 0.27 0.3 0.4 0.4 0.39 0.37 0.33 0.33 0.32 0.31 0.28 0.19 0.324 R C H 0.28 0.28 0.38 0.39 0.37 0.33 0.35 0.32 0.32 0.31 0.28 0.19 0.317 R C A 7 0.24 0.24 0.4 0.43 0.41 0.43 0.36 0.38 0.4 0.35 0.3 0.21 0.346 B92 0.26 0.28 0.36 0.37 0.34 0.34 0.29 0.32 0.34 0.3 0.24 0.21 0.304 Avg Mean Helix Hydrophobic Moment 0.323 Table 4b Helix Hydrophobic Moment of Randomly Generated Variants Unable to Support Transport Helix Variant Position Hydrophobic Moment 1 2 3 4 5 6 7 8 9 10 11 12 B29 0.19 0.15 0.06 0.13 0.13 0.13 0.11 0.26 0.28 0.25 0.21 0.15 0.171 D121 0.27 0.29 0.25 0.25 0.27 0.23 0.15 0.22 0.27 0.17 0.21 0.3 0.240 D125 0.25 0.29 0.24 0.19 0.26 0.39 0.22 0.17 0.17 0.22 0.16 0.12 0.223 47 0.11 0.09 0.06 0.19 0.19 0.17 0.25 0.21 0.16 0.16 0.23 0.2 0.168 86 0.15 0.06 0.16 0.2 0.16 0.1 0.05 0.13 0.14 0.08 0.13 0.19 0.129 B1 0.15 0.18 0.24 0.24 0.2 0.17 0.2 0.25 0.18 0.17 0.19 0.19 0.197 Avg Mean Helix Hydrophobic Moment 0.188 Table 4c Helix Hydrophobic Moment of Helical Variant Position 1 2 3 4 5 6 7 HlyA 0.29 0.31 0.46 0.48 0.47 0.47 0. .39 P63N2 0.26 0.3 0.39 0.4 0.37 0.37 0. .33 P48C3 0.26 0.29 0.39 0.4 0.37 0.36 0 ,32 P48C4 0.28 0.28 0.41 0.44 0.39 0.39 0. .33 P32C1 0.26 0.29 0.39 0.4 0.37 0.46 0 .39 P48C1 0.28 0.3 0.41 0.42 0.39 0.39 0 .33 P48C2 0.28 0.3 0.41 0.42 0.4 0.39 0 .32 P6301 0.27 0.29 0.39 0.38 0.36 0.34 0 .31 Avg Mean Helix Hydrophobic Moment by Functional Analysis Mean Helix Hydrophobic Phenotype Moment 8 9 10 11 12 0.38 0.4 0.35 0.3 0.21 0.376 +++++ 0.3 0.3 0.29 0.25 0.17 0.311 ++++ 0.33 0.32 0.31 0.27 0.18 0.317 ++++ 0.33 0.33 0.34 0.28 0.18 0.332 ++++ 0.39 0.39 0.37 0.33 0.26 0.358 ++++ 0.33 0.33 0.31 0.28 0.18 0.329 ++++ 0.34 0.35 0.31 0.26 0.2 0.332 ++++ 0.27 0.26 0.24 0.22 0.15 0.290 ++++ 0.324 P3C13 0. 28 0.3 0.4 0. .41 0, .38 0.38 0.32 P35C1 0. .27 0.35 0.45 0. .51 0. .44 0.47 0.43 P63Q2 0. .27 0.3 0.4 0. .41 0. .38 0.37 0.34 P63C3 0. .27 0.3 0.39 0, .41 0. .38 0.38 0.33 P5C2 0. .24 0.26 0.37 0, ,38 0. .36 0.46 0.39 P63C49 0. .24 0.25 0.37 0. .36 0. .33 0.49 0.42 P4C1 0. .26 0.28 0.35 0. .37 0 .34 0.34 0.29 P30C1 0. .23 0.24 0.35 0. .37 0 .33 0.34 0.28 P6C2 0. 22 0.25 0.35 0. ,37 0. .34 0.33 0.3 P63.22C5 0, .27 0.31 0.38 0, ,36 0 .35 0.32 0.26 Avg Mean Helix Hydrophobic Moment 0.32 0.32 0.3 0.27 0.18 0.322 +++ 0.38 0.45 0.45 0.27 0.23 0.392 +++ 0.33 0.34 0.32 0.29 0.19 0.328 +++ 0.33 0.33 0.32 0.28 0.19 0.326 +++ 0.43 0.43 0.42 0.38 0.29 0.368 +++ 0.45 0.47 0.48 0.4 0.34 0.383 +++ 0.32 0.34 0.3 0.24 0.21 0.303 +++ 0.32 0.32 0.31 0.27 0.18 0.295 +++ 0.31 0.33 0.32 0.29 0.19 0.300 +++ 0.27 0.29 0.26 0.21 0.19 0.289 0.331 +++ P35C5 0.26 0.29 0.39 0.4 P63W2 0.25 0.32 0.42 0.47 P3C11 0.27 0.3 0.4 0.04 P5C1 0.32 0.34 0.44 0.46 P5C8 0.26 0.34 0.43 0.49 P63C15 0.3 0.4 0.48 0.53 P32C4 0.26 0.34 0.45 0.5 Avg Mean Helix Hydrophobic Moment 0.38 0.36 0.33 0.29 0.3 0.28 0.39 0.39 0.36 0.33 0.4 0.41 0.39 0.37 0.33 0.33 0.32 0.31 0.42 0.61 0.53 0.52 0.5 0.46 0.42 0.4 0.38 0.32 0.37 0.39 0.45 0.43 0.41 0.35 0.41 0.43 0.43 0.41 0.38 0.36 0.41 0.43 0.25 0.16 0.308 ++ 0.23 0.19 0.347 ++ 0.28 0.19 0.294 ++ 0.41 0.34 0.446 ++ 0.25 0.17 0.352 ++ 0.25 0.21 0.388 ++ 0.29 0.18 0.370 ++ 0.358 35 truncation of the signal sequence, has a helix hydrophobic moment of 0.099. These results also correlate with the average helix hydrophobic moment of helical variants,(all were able to support transport), among which there was no average helix hydrophobic moment below 0.289. However, the mean average helix hydrophobic moments of each phenotypical category of helical variants does not appear to correlate a progressively greater helix hydrophobic moment with an increase in ability to support transport. Interestingly, the overall trend does remain that there is a higher mean helix hydrophobic moment for variants able to support transport. It is important to note that the algorithm employed by Peptool vs 1.0 for calculating beta and helix hydrophobic moments does not take into account that amino acids with long side chains that are flexible may have unexpected contributions to the hydrophobic moment of a helical structure (Lemire et al., 1989). The correlation of helix hydrophobic moment with transport supports the proposed model for HlyA recognition and transport in which the amphiphilic nature of the ai-helix is postulated to be involved in partitioning into the lipid bilayer prior to interacting the transport complex. Hydrophobicity also appears to correlate with the ability to support transport (table 5). The mean hydrophobicity of variants decreases with a decreased ability to support transport. All variants unable to support transport have mean hydrophobicities below -4.717, while the mean hydrophobicity of randomly generated variants able to support transport was above -0.450, and the mean hydrophobicity of helical variants (able to support transport) was above -2.008. It is interesting to note that the average mean hydrophobicity of helical variants (able to support transport) in each category of phenotype decreases as the degree of ability to support transport decreases. It has previously been suggested that average hydrophobicities 36 Tab le 5 a Hydrophobicity of Randomly Generated Variants Able to Support Transport Variant Position Mean Hydrophobic i ty 1 2 3 4 5 6 7 8 9 10 11 12 R C N -1.4 -0.7 -1.2 -1.7 -1.3 -0.8 0.0 1.1 1.4 0.8 -0.1 -1.5 -0.450 R C H -0.6 0.8 1.1 1.4 1.0 0.7 0.8 1.1 1.4 0.8 -0.1 -1.5 0.575 R C A 7 -0.2 1.2 1.2 0.0 -0.2 -0.7 -0.5 1.0 1.6 1.2 0.6 -1.3 0.325 B92 -0.9 0.1 0.0 -0.4 -0.1 0.0 0.6 1.7 2.0 1.6 0.9 -0.7 0.400 A v g Mean Hydrophobic i ty 0.213 Table 5b Hydrophobicity of Randomly Generated Variants Unable to Support Transport Variant Position Mean 1 2 Hydrophobic i ty 3 4 5 6 7 8 9 10 11 12 D121 -4.6 -5.3 -4.2 -4.2 -5.0 -5.3 -6.7 -6.2 -5.6 -4.5 -3.6 -4.8 -5.000 D125 -8.0 -7.7 -5.8 -3.8 -3.1 -2.8 -4.0 -5.0 -5.5 -5.9 -5.5 -5.5 -5.217 47 -3.6 -2.3 -1.3 -1.5 -2.7 -3.7 -5.0 -6.1 -6.5 -7.5 -8.1 -8.3 -4.717 B29 -9.9 -10.0 -10.1 -8.7 -7.7 -5.6 -3.9 -2.7 -0.9 -0.7 -0.7 -1.3 -5.183 86 -4.5 -3.5 -3.3 -3.9 -4.4 -5.3 -5.6 -4.9 -4.8 -4.2 -4.5 -5.4 -4.525 B1 -7.8 -9.0 -9.3 -7.5 -5.8 -3.3 -1.7 -2.5 -3.5 -5.6 -7.5 -8.0 -5.958 A v g Mean Hydrophobic i ty -5.100 37 Table 5c Hydrophobicity of Helical Variants by Functional Analysis V a r i a n t P o s i t i o n Hydrophobicity Phenotype 1 2 3 4 5 6 7 8 9 10 11 12 HlyA -1.0 0.0 -0.4 -1.1 -1.0 -1.1 -0.5 1.0 1.6 1.2 0.6 -1.3 -0.167 +++++ P63N2 -1.5 -0.8 -1.3 -1.7 P48C3 -1.4 -0.6 -0.9 -1.3 P48C4 -1.0 -0.1 -0.8 -1.7 P32C1 -1.3 -0.4 -0.9 -1.4 P48C1 -1.0 -0.1 -0.5 -1.1 P48C2 -1.0 -0.1 -0.6 -1.2 P6301 -1.3 -0.3 -0.7 -1.1 Avg Mean Hydrophobicity -1.3 -0.9 0.0 1.0 1.0 0.2 -0.9 -0.6 0.1 1.1 1.2 0.6 -1.8 -2.0 -1.1 0.1 0.7 0.5 -1.1 -0.7 -0.5 0.0 -0.6 -1.9 -0.9 -0.9 -0.2 0.7 1.0 0.5 -1.0 -0.8 -0.1 1.2 1.8 1.6 -1.0 -1.0 -0.5 0.2 0.3 -0.2 -0.7 -2.2 -0.683 ++++ -0.2 -1.6 -0.375 ++++ -0.2 -1.6 -0.750 ++++ -2.4 -3.4 -1.217 ++++ -0.2 -1.6 -0.358 ++++ 1.1 -0.6 0.025 ++++ -1.1 -2.5 -0.767 ++++ -0.589 P3C13 -1.0 -0.1 -0.5 -1.1 P35C1 -1.1 -0.2 -1.7 -3.2 P63Q2 -1.5 -0.7 -1.1 -1.5 P63C3 -1.4 -0.6 -1.1 -1.6 P5C2 -1.1 -0.2 -0.6 -1.3 P63C49 -0.8 0.1 -0.4 -1.3 P4C1 -1.1 0.0 -0.1 -0.4 P30C1 -0.7 0.4 0.3 -0.4 P6C2 -1.0 0.1 0.0 -0.5 P63.22C5 -0.8 0.0 -0.5 -1.0 Avg Mean Hydrophobicity -0.8 -0.7 0.0 0.9 1.1 0.5 -3.9 -4.7 -3.2 -1.1 -0.1 0.0 -1.3 -1.1 -0.4 0.5 0.7 0.3 -1.2 -0.8 0.0 0.9 1.1 0.4 -1.1 -0.9 -0.8 -0.2 -0.4 -1.5 -1.7 -2.1 -2.4 -2.3 -2.6 -3.7 -0.1 0.0 0.6 1.7 2.0 1.6 -0.4 -0.4 0.0 1.0 1.2 0.6 -0.5 -0.5 0.0 0.9 1.3 0.7 -0.9 -0.8 0.0 0.8 1.4 1.3 -0.2 -1.6 -0.292 +++ -0.4 -2.1 -1.808 +++ -0.4 -1.7 -0.683 +++ -0.3 -1.7 -0.525 +++ -1.8 -2.7 -1.050 +++ -3.4 -3.5 -2.008 +++ 0.9 -0.7 0.367 +++ -0.2 -1.6 -0.017 +++ -0.1 -1.5 -0.092 +++ 0.8 -0.6 -0.025 +++ -0.613 P35C5 -1.4 -0.5 -0.8 -1.3 P63W2 -0.8 0.0 -1.3 -3.0 P3C11 -1.4 -0.7 -1.2 -1.7 P5C1 -1.3 -0.5 -1.2 -1.9 P5C8 -1.5 -0.8 -2.2 -3.6 P63C15 -1.5 -0.8 -2.4 -3.3 P32C4 -1.3 -0.5 -2.0 -3.4 Avg Mean Hydrophobicity -1.1 -0.9 -0.2 0.8 0.9 0.2 -3.9 -4.8 -3.3 -1.0 0.3 0.9 -1.3 -0.8 0.0 1.1 1.4 0.8 -1.3 -0.8 -0.8 -0.5 -1.6 -3.3 -4.1 -4.6 -2.6 -0.5 0.5 0.5 -3.8 -4.3 -2.3 -0.4 0.8 1.4 -4.1 -4.8 -3.1 -1.0 0.1 0.6 -0.8 -2.5 -0.633 ++ 0.3 -1.4 -1.500 ++ -0.1 -1.5 -0.450 ++ -3.4 -4.1 -1.725 ++ -0.6 -2.4 -1.825 ++ 0.8 -0.7 -1.375 ++ -0.2 -1.6 -1.775 ++ -1.326 38 strongly deviating from 0 are rare among mitochondrial presequences and may interfere with function (von Heijne, 1986; Lemire et al, 1989), this seems to be in agreement with the observation of C-terminal signal sequence variants unable to support transport which have very low average hydrophobicities. It also may be that the correlation with transport and hydrophobicity may be a reflection of a requirement for an amphiphilic helical structure. For completion, beta hydrophobic moment was also calculated to evaluate the possibility that the ability to support transport might correlate with an amphiphilic B-sheet. As expected, no clear correlation was observed to a 99.9999% confidence level (a=0.05) (table 6). Therefore the ability to form an amphiphilic B-sheet does not appear to correlate with the ability to support transport. However, this feature may be important for individual sequences. Lower averages may be explained in part by a strong bias toward F (Lemire et al, 1991). The Amino Acid Composition of the ai-Helical Library of HlyA Signal Sequence Variants Based on Position The helical library appears to have a strong bias for F in position 5 (probability of this finding happening by chance, (9.1 x 10"6) and 8 (9.1 x 10"6) (table 7). There also appears to be a bias for F in position 2 (probability of this finding happening by chance is 3.4 x 10"4), E in position 3 (3.2 x 10"3), S in position 11 (1.9 x 10"4), and V in position 12 (1.9 x 10"4). A bias against T in position 11 is observed with a probability of this finding happening by chance to be 1.9 x 10"3. Yet there does not appear to be any correlation between the number of amino acids conserved in relation to wild type function. For example, there are 39 Table 6a Beta Hydrophobic Moment of Randomly Generated Variants Able to Support Transport Mean Beta Variant Position Hydrophobic Moment 1 2 3 4 5 6 7 8 9 10 11 12 RCN 0.13 0.15 0.12 0.18 0.07 0.25 0.18 0.11 0.14 0.26 0.25 0.15 0.166 RCH 0.13 0.27 0.3 0.26 0.23 0.21 0.18 0.11 0.14 0.26 0.25 0.15 0.208 RCA7 0.03 0.22 0.27 0.26 0.22 0.3 0.16 0.18 0.25 0.34 0.22 0.05 0.208 B92 0.06 0.19 0.2 0.23 0.19 0.25 0.24 0.06 0.14 0.33 0.22 0.04 0.179 Avg Mean Beta Hydrophobic Moment 0.190 Table 6b Beta Hydrophobic Moment of Randomly Generated Variants Unable to Support Transport Mean Beta Variant Position Hydrophobic Moment 1 2 3 4 5 6 7 8 9 10 11 12 B29 0.62 0.47 0.96 0.61 0.81 0.45 0.46 0.1 0.1 0.15 0.28 0.17 0.432 D121 0.47 0.59 0.41 0.3 0.31 0.18 0.39 0.23 0.23 0.32 0.16 0.4 0.333 D125 0.13 0.2 0.12 0.16 0.38 0.2 0.22 0.11 0.11 0.08 0.08 0.17 0.163 47 0.08 0.15 0.06 0.06 0.14 0.11 0.11 0.25 0.1 0.34 0.05 0.13 0.132 86 0.22 0.04 0.27 0.08 0.26 0.21 0.5 0.42 0.55 0.25 0.35 0.1 0.271 B1 0.18 0.11 0.19 0.14 0.31 0.26 0.23 0.43 0.28 0.29 0.07 0.12 0.218 Avg Mean Beta Hydrophobic Moment 0.258 Table 6c Beta Hydrophobic Moment of Helical Variants by Functional Analysis Mean Beta Variant Posi t ion Hydrophobic Phenotype Moment 1 2 3 4 5 6 7 8 9 10 11 12 HlyA 0.16 0.09 0.14 0.23 0.09 0.3 0.16 0.18 0.25 0.34 0.22 0.05 0.184 +++++ P63N2 0.14 0.13 0.1 0.18 0.07 0.25 0.17 P48C3 0.12 0.15 0.13 0.19 0.1 0.25 0.19 P48C4 0.16 0.11 0.11 0.28 0.16 0.28 0.1 P32C1 0.13 0.14 0.11 0.19 0.09 0.25 0.17 P48C1 0.16 0.11 0.11 0.2 0.09 0.25 0.15 P48C2 0.16 0.12 0.11 0.2 0.09 0.25 0.17 P6301 0.14 0.11 0.1 0.19 0.08 0.21 0.11 Avg Mean Beta Hydrophobic Moment P3C13 0.16 0.11 0.11 0.2 0.09 0.24 0.17 P35C1 0.12 0.15 0.13 0.49 0.38 0.42 0.33 P63Q2 0.14 0.12 0.1 0.2 0.07 0.24 0.13 P63C3 0.16 0.11 0.09 0.19 0.06 0.24 0.17 P5C2 0.07 0.2 0.17 0.21 0.14 0.25 0.15 P63C49 0.08 0.18 0.15 0.27 0.18 0.25 0.16 P4C1 0.08 0.18 0.19 0.23 0.19 0.25 0.24 P30C1 0.05 0.21 0.19 0.22 0.18 0.25 0.17 P6C2 0.08 0.13 0.15 0.17 0.07 0.21 0.16 P63.22C5 0.1 0.16 0.14 0.2 0.13 0.23 0.14 Avg Mean Beta Hydrophobic Moment 0.11 0.14 0.19 0.2 0.07 0.146 ++++ 0.11 0.15 0.24 0.25 0.15 0.169 ++++ 0.03 0.16 0.23 0.25 0.15 0.168 ++++ 0.2 0.17 0.14 0.26 0.11 0.163 ++++ 0.12 0.16 0.23 0.25 0.15 0.165 ++++ 0.11 0.17 0.35 0.21 0.05 0.166 ++++ 0.08 0.09 0.1 0.19 0.08 0.123 ++++ 0.157 0.1 0.14 0.23 0.25 0.15 0.163 +++ 0.33 0.21 0.19 0.25 0.14 0.262 +++ 0.1 0.14 0.21 0.24 0.14 0.153 +++ 0.11 0.14 0.23 0.24 0.14 0.157 +++ 0.2 0.18 0.12 0.28 0.08 0.171 +++ 0.4 0.31 0.29 0.34 0.19 0.233 +++ 0.06 0.14 0.33 0.22 0.04 0.179 +++ 0.09 0.15 0.24 0.25 0.15 0.179 +++ 0.1 0.15 0.26 0.25 0.15 0.157 +++ 0.06 0.13 0.28 0.21 0.05 0.153 +++ 0.181 P35C5 0.12 0.14 0.12 0.2 0.09 0.24 0.15 0.1 0.14 0.17 0.19 0.08 0.145 ++ P63VV2 0.08 0.2 0.17 0.5 0.4 0.42 0.31 0.25 0.18 0.26 0.19 0.05 0.251 ++ P3C11 0.13 0.15 0.12 0.18 0.07 0.25 0.18 0.11 0.14 0.26 0.25 0.15 0.166 ++ P5C1 0.16 0.19 0.2 0.26 0.19 0.34 0.22 0.31 0.28 0.26 0.37 0.26 0.253 ++ P5C8 0.13 0.13 0.1 0.48 0.37 0.42 0.35 0.28 0.12 0.21 0.21 0.08 0.240 ++ P63C15 0.22 0.16 0.1 0.44 0.33 0.4 0.37 0.32 0.14 0.31 0.21 0.04 0.253 ++ P32C4 0.12 0.14 0.1 0.47 0.34 0.4 0.29 0.22 0.17 0.24 0.25 0.15 0.241 ++ Avg Mean Beta Hydrophobic Moment 0.221 41 — - i j ^ - i ft ~ (Ti cz CD CO *<L c • TJ D ID fl) C 7 > O O O O C 0 - O < D - * ^ O O O O O O O O 0 0 0 0 0 ) 0 0 0 - ^ 0 0 0 0 0 0 0 0 W O O O O C O O O - J O l O O O O O O O O O O O O O O O - ^ O O C O O C O r O J ^ h J O O o o o o o o o - * o o r o o t n - N j O ) w o o - t > - o o o o ^ o o - ^ r v ) O o o o o o o o O O O Q - V | O O O O O O O O O O N 3 0 1 0 O O O O O O O O O O C 0 O - U M C D C J 3 O O - N i o o o o l t c o o r o - » - o o o o o o o o 4 ^ 0 0 0 0 - N J O O J ^ - ^ o o o o o o o o O O O ^ O O O K J O O O O O O O - ^ r o O o o o r o ^ o o o o o o o o o o o o o ^ O O O O O O O O O O - N O O O O O A O O O O O O O A O O J > O J > J > 4 ^ - N O O O O O O D - N O O ^ O O O O O O O - f v ^ O O O O O O O O O O O O O O O O O O O O C O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 =• £3 3 5" SP 3 f 5" = s » o O £ £ > > O = . ^ ^ = ^ i ! ! a ) i ! i . c n c < t | « - < C 3 - O 3 c 3 o ) > 3 3' o > "1 ^ - > « ! > . 0 3. .M =• I fs m| « a i § 3 00 TJ > 2 2 CO 3 3 o cr (ft CD CD Q . 3 cr CD 5 > i S3? m 0) o 3T TJ o ( A 8 > - n O 3 5- =• I 3 * a § § • ai TJ > 2 3 O O = ^ ? o 3 3 ^ (ft rn o > _ O. ^ V It ID ^ 8 " 2 "\ O » o r S a -cn 3 > T l T j 3 CD m w 2 c - a — O <B CD 1 * 1 Ii 3 > T l T J 3 CD m O q - - a X ; » O CD CD g M I s f f a a > 3 3 o > —i 0) CT CD > 3 5' o > o ai V) a> m u o 3" TJ o (fl 3>' o 3 5' R i 2. B) 42 no amino acids conserved in P48C3 yet it maintains a phenotype of (++++) in relation to wild type (table 1). One possible explanation for the biases for F, E, S, and V, and against T, may lie in the hydrophobic nature of these amino acids. While F and V have a strong hydrophobic nature, E and S are hydrophilic, while T is weakly hydrophilic, falling closest to the mean on the hydrophobicity scale (Eisenberg et al., 1984). Thus it is possible that the hydrophobic nature of these amino acids contributes to a degree required for recognition of transport of the signal sequence. However, in order to determine the nature of the biases observed in the helical library, it will be important to further investigate natural biases in amino acid composition among C-terminal signal sequences, as there may be an explanation of the biases that lies within the nature of the C-terminal signal sequences as have been observed in N-terminal signal sequence (von Heijne, 1986; Lemire et al., 1989). Additionally, it is unlikely that that there is a bias in the construction of the library, as the pUCAC494-cti -helix library ligation was sequenced, and in each degenerate position in the oti-region, the expected nucleotides were observed. The Amino Acid Composition of the ai-Random Library (generated by method 2) of HlyA Signal Sequence Variants Based on Position In the random library generated by method 2, there appears to be a strong bias for K in positions 7, 10, and 12, with a probability of these findings happening by chance to be 1.8 x 10"8, 3.2 x 10"9, and 1.7 x 10"12 respectively (table 8). Note that this analysis excluded the three variants selected by method 1 in the random library, as they were selected on their ability to support transport, rather than by a random, independent selection. There also appears to be a strong bias against A in position 2 (probability of 2.3 x 10~5). Biases 43 Table 8 Frequency of Amino Acids at Each Position in all a r Random Variants (generated by method 2) Amino Acids Frequency of 0 b s e r v e d N u m b e r o f A m i n o A * d s in Each Position of 1 2 3 4 5 6 7 8 9 10 11 12 Ala (A) 6.25 1 13 4 8 1 1 2 1 1 2 3 1 Arg(R) 9.375 3 6 9 2 6 9 4 5 2 7 3 0 Asn(N) 3.125 3 0 2 1 4 1 7 3 6 2 6 2 Asp(D) 3.125 4 2 4 0 0 3 1 1 0 0 0 2 Cys (C ) 3.125 2 1 0 1 1 1 1 1 2 0 0 0 Gin (Q) 3.125 3 5 8 7 6 1 4 3 3 6 1 5 Glu(E) 3.125 4 0 2 4 3 1 4 0 3 1 1 2 Gly (G) 6.25 5 3 3 4 3 0 2 1 1 2 3 0 His (H) 3.125 1 1 2 4 1 0 3 4 0 1 3 1 lie (I) 4.6875 3 1 2 3 7 4 2 1 1 0 3 3 Leu (L) 9.375 4 5 4 4 6 2 0 2 10 1 1 1 Lys (K) 3.125 1 4 1 1 2 2 12 7 7 12 6 14 Met(M) 1.5625 0 1 0 0 0 2 0 5 0 0 0 0 Phe(F) 3.125 2 1 1 1 2 0 1 0 4 0 0 0 Pro(P) 6.25 6 6 5 5 0 0 1 4 2 1 0 1 Ser (S) 6.25 5 3 7 4 6 8 5 2 2 5 4 0 Stop 4.6875 4 5 1 5 1 5 5 4 3 1 2 1 Thr(T) 6.25 2 4 1 7 5 1 1 1 6 2 7 6 8 Trp(W) 1.5625 3 0 1 0 0 4 0 0 0 0 0 0 Tyr(Y) 3.125 3 0 3 1 3 3 1 2 1 0 1 1 Val (V) 625 3 1 2 0 5 2 1 1 0 2 2 1 Total 62 62 62 62 62 60 57 53 50 50 45 43 for Q in positions 3 (1.3 x IO"4), 4 (6.8 x IO"4), 5 (3.2 x 10"3), 10 (8.9 x 10"4), K in positions 8 (2.3 x IO"4), 9 (1.5 x IO"4), 11 (4.6 x IO"4), M in position 8 (1.8 x 10"4), T in positions 6 (3.0 x 10"4), 3 (1.2 x 10"3), and L in position 9 (5.8 x 10"3). A bias against L in position 7 was observed with a probability of this occurring by chance to be 3.7 x 10"3). Overall, there was a strong bias for K (1.1 x 10"15) and Q (1.3 x 10"9), a weaker bias for T (2.2 x 10"3), and a bias 5 3 against V (9.9 x 10") and C (6.1 x 10") occurring in any one of the 12 positions of the helix. As there were only 6 variants isolated that were unable to support transport, it is not possible to determine that the presence of any single amino acid in a given position may be the key factor in rendering the variant unable to support transport. Additionally, due to the frequency of stop codons, insertions or deletions, it is difficult to correlate these biases for or against with biological function, thus explanations for the bias are speculative. The observed biases in the random library were unexpected. It is unlikely that these biases arose in the construction of the library but rather after transformation into E. coli. This explanation is further supported by the observation that no bias was observed against stop codons, and therefore selective pressure against truncated signal sequences without an ct2-helix were not observed. Thus one may postulate that there is a selection against proteins with functional oci and 0:2 regions that are in some manner lethal to E. coli. 45 Does the ability to support transport correlate with secondary structure prediction? The predicted secondary structure of all variants able to support transport follows the pattern predicted by Kamtekar et al. (1991) to form amphiphilic a-helices in both the helical and random libraries, while variants unable to support transport do not follow this pattern. However, secondary structure prediction results using the AGADIR and SSCP programs do not correlate with the ability to support transport. This may reflect the finding that both the HlyA and LktA signal sequences were unstructured in an aqueous environment, while in certain membrane mimetic environments they assumed a helical secondary structure (Yin et al., 1995). As the AGADIR program predicts the helical content of peptides, but is limited in it ability to predict helical content for peptides with Pro, His, Cys, Trp and other aromatic residues, variants with these residues may contribute to the lack of correlation (tables 1 and 2). The AGADIR program also takes into account pH, temperature and ionic strength which may not reflect the in vivo conditions under which the a i-region forms a helical secondary structure. 46 C H A P T E R IV: CONCLUSIONS AND R E C O M M E N D A T I O N S F O R F U T U R E W O R K In the two libraries generated in this study, random and helical, the predicted secondary structure of all variants able to support transport follows the pattern predicted by Kamtekar et al. (1991) to form amphiphilic a-helices, while variants unable to support transport do not follow this pattern. The helix hydrophobic moment calculation, which gave a measure of the degree of amphiphilicity, calculated a greater degree of amphiphilicity for variants able to support transport over that of variants unable to support transport. However, secondary structure prediction results using the AGADIR and SSCP programs do not correlate with the ability to support transport. This may reflect the CD and N M R analyses which found that both the HlyA and LktA signal sequences were unstructured in an aqueous environment, while assuming a helical secondary structure in certain membrane mimetic environments. Although these results support a hypothesis that an amphiphilic a-helical structure is required for transport, it is apparent from this study that considerable selective pressures exist. This finding was not initially anticipated. For example, in both libraries only a small proportion of the variants were obtained that yielded the intended construction. In the random library of variants generated using the strain DH5a a high frequency of variants (48.4%) with insertions of codons, deletions of codons and deletions resulting in frame shifts was observed both within the ai-helical region as well as three amino acids upstream from the ai-region. (Note that the percentage 48.4 is calculated by excluding variants with insertions and 47 deletions that also contained stop codons.) One possible explanation for this observation is that there may be strong selective pressure for the a\ region to form an amphiphilic helix, due to toxicity of non-helical structures. However, I do not exclude the possibility that there may be amphiphilic a-helices that are unable to support transport, or that there may be non-amphiphilic a-helices that may support transport but due to the high frequency of mutations, insertions and deletions (strong selective pressure) it becomes technically difficult to isolate these variants due to possible toxicity problems within the current experimental design. Although selection pressure against certain sequences by the host may be multifactorial and complex, analysis of variant 47 may give insight into a possible mechanism of toxicity. Preliminary observations of variant 47, (which does not support transport), have demonstrated that it begins to grow, then lyses in DH5a with pLGBCD, while other ai-helical variants and wild type HlyA grow at identical rates in DH5a with pLGBCD. Variant 47, along with the other ai-helical variants and wild type HlyA grow at identical rates in DH5a, JM83, and JM83 with pLGBCD. One model to explain these observations in DH5a, a recA- strain, is a situation in which the second helix (a2) is recognized by the binding pocket of the transport complex, and locks the transporter in an open position (resulting in conformational change of the transport complex to an open position), but cannot trigger transport as the ai-helix is not in position to allow the necessary amino acid interactions. The open position of the transport complex may allow other molecules to pass through the transporter destroying the proton gradient, and resulting in lysis. However, in JM83 (recA+) strains carrying the transport complex on pLGBCD, mutations that would circumvent cell lysis, may take place, and in JM83 and DH5a strains 48 without the transport complex, cell lysis would not occur. There is precedent for the passage of molecules through the hemolysin transport complex in the reverse direction, as it has been observed that vancomycin resistance could be reversed in the presence of the hemolysin transport system (Blight et al, 1994). This observation also brings to light that there still may be helices that do not support transport. However of those variants analyzed in this sample, an ai-amphipathic helix was sufficient to support transport. Thus in employing two libraries to investigate the role of secondary structure in the recognition and transport of HlyA the following conclusions were drawn: (1) all variants isolated that were predicted to form an amphiphilic helical structure for the a i-region by the method of Kamtekar et al (1991) were able to support transport, (2) the six variants isolated that were unable to support transport, were not predicted to form an amphiphilic helical structure for the cti-region by the method of Kamtekar et al (1991), (3) the helix hydrophobic moment and hydrophobicity as calculated by Peptool (vs. 1.0) using the algorithm of Eisenberg et al. (1994) correlate with the ability of variants to support transport, (4) the method of Kamtekar et al. (1991), predicts with greater accuracy biological activity (ability to support transport) than do the secondary structure prediction programs AGADIR and SSCP, (5) selective pressures were observed in the random library which may indicate that enormous biological selectivity in the evolution of the hemolysin transport system occurred through the use of the biological alphabet (genetic code) much in the same way as is employed in the helical library whose design for amphiphilic a-helices was based on the method of Kamtekar et al. (1991). These conclusions support the model of HlyA recognition and transport as described in the introduction: (1) interaction of HlyA with the inner membrane lipids via the C-49 terminal signal sequence, (2) induction of secondary structure formation of the signal sequence to fit into the binding pocket of HlyB, (3) contact of the cti and oc2 helices in the signal sequence with specific residues in the binding pocket of HlyB to trigger the ATPase activity, (4) translocation of the HlyA molecule into the growth medium. Additionally greater appreciation of the role, the nature of the helix plays in transport has been gained. The amphiphilic nature of the oci-helix appears to be important as variants able to support transport are predicted to be amphiphilic helices and the helix hydrophobic moment correlates with ability to support transport. Thus it may be postulated that the amphiphilicity of ai-region is important in order for the helix to partition into the lipid bilayer, and may be a rate limiting step in allowing the toxin to search for the transporter in a 2D rather than 3D manner. To further investigate the validity and detail of the steps outlined in this model, the following recommendations for future work are made. (1) As the ability to support transport correlates with the predicted amphiphilic structure of Kamtekar et al. (1991), helix hydrophobic moment, and hydrophobicity, but not with other secondary structure prediction programs employed, what structure is being predicted and generated in those variants able to support transport, and conversely, in those unable to support transport? As N M R analysis was previously performed on HlyA and LktA signal peptides and demonstrated that they assume similar secondary structures, a similar approach could be applied to study the structures of the selected variants isolated from both libraries (Yin et al, 1995). This line of study will investigate the predicted structure of the selected signal variants and analyze their structural characteristics in relation to their ability to support transport. Results from this line 50 of investigation will enable further interpretation of the role of the C-terminal signal sequence secondary structure. (2) To address the question of what is the structure of variants in the random library that were not selected due to biological pressure - is there a lethal effect of a non-helical structure in the a i-helix? It is proposed that electroporation of the pUCAC494-random library of variants into DH5cc cells in the absence of HlyB/D may circumvent lethality problems i f they exist as there has not been any adverse side effects reported for the expression of HlyA in the absence of a transport complex (rather it has been shown to be degraded quickly in the absence of the transport complex) (Koronakis et al., 1991). Additionally the cloning of the variant 47, observed to be lethal in DH5ct with BCD, while nonlethal in DH5oc, JM83, JM83 with pLGBCD, into a vector with a tightly controlled promoter will further dilineate at which step in recognition and transport or by another unexpected means, this variant is lethal. These experiments will investigate the hypothesis of biological selection pressure in the libraries HlyA variants. (3) To further examine the model proposed for the toxicity of variant 47 in the E. coli DH5a containing pLGBCD, two sets of experiments are proposed. Firstly, i f the model is correct, then a non-helical structure in both the a i and ct2 regions would not be able to support transport and yet would not be toxic in DH5a containing pLGBCD (hlyB, C, and D), as the variant would not be recognized by the transport complex even partially. Additionally, one would also predict that if a random library of variants was substituted into the ct2 region, while maintaining the wild type sequence a i-region, and variants were selected that were unable to support transport, non-helical ct2 variants would also be toxic in the DH5cc strain 5 1 with pLGBCD. A second line of investigation would involve examining whether or not there is direct interaction between the signal sequence of variant 47 and HlyB/D. As GST fusion proteins with the HlyA C-terminal signal sequence have previously been shown to compete with a full length wild-type HlyA for transport (note that the GST fusion protein cannot be transported), competition assays can be used to further determine the nature of the interaction between variant 47 and the hemolysin transport complex. If the proposed model of toxicity is correct, then one would expect there to competition between a full-length wild-type HlyA molecule and a GST fusion to the C-terminus of variant 47 when coexpressed, resulting in cell lysis, and confirming that the variant 47 toxicity effect is limited to the C-terminus of the molecule. However, i f there is an unexpected effect due to an interaction with the rest of the HlyA molecule with the C-terminus, one would not observe cell lysis in a competition experiment of this nature. (4) To examine the role of the Leu-Ile mutations 3 amino acids upstream from the a\-helix region observed at a high frequency (25) in the random library. This position will be mutated back to Leu, and HlyA recognition and transport analyzed by the ELISA assay for HlyA and hemolytic zone assay. In a wider context, this study of the hemolysin transport system have provided insights into fundamental aspects and evolution of A B C transporters. One of the most captivating problems in the A B C transporter family is that of substrate specificity. The A B C transporters are distributed widely throughout eukaryotic and prokaryotic organisms, and are responsible for transporting a diverse array of substrates which include ions, hydrophobic drugs, short peptides, sugar polymers, and large proteins. In eukaryotes, these include the multidrug resistance P-glycoprotein (mdr genes), the cystic fibrosis transmembrane conductance regulator (CFTR), 52 transporters of antigenic peptides (TAPi and TAP 2 ) , the exporter of a-type mating factor in S. cerevisiae (STE6), and transporters of peroxisomal entities (Pmp70, and ALD-P) (Childs and Ling, 1994). In prokaryotes, these proteins include the periplasmic permeases, the hemolysin A transporter (HlyB), the leukotoxin A transporter (LktB), and the nisin A transporter (NisT) (Fath and Kolter, 1993). The challenge is to understand how a single family of proteins can accommodate such a wide range of substrates within what is a likely a conserved transport mechanism. The manner in which the hemolysin system has evolved to selectively recognize a broad variety of primary sequences, and is predicted to recognize a given secondary structure motif, may parallell curent thought that there may be many different modes of binding a given substrate by the P-glycoprotein. Additionally the fact that the HlyA signal sequence may need to be targetted to the inner membrane lipids in order to form a secondary structure recognized by the transport complex further parallels the current concept that anticancer drugs are recognized by P-glycoprotein via the lipid bilayer (Shapiro et al, 1997). Thus, the studies in this thesis have allowed us to further refine the understanding of the recognition and transport of HlyA by its dedicated A B C transport system. 53 B I B L I O G R A P H Y Bradley, G., P.F. Juranka, and V. Ling. 1988. Mechanism of multidrug resistance. Biochim Biophy Acta. 948:87-128. Bhakdi, S., S. Greulich, M . Muhly, B. Eberspacher, H. Becker, A . Thiele, and F. Hugo. 1989. Potent leukocidal action of Escherichia coli hemolysin mediated by permeabilization of target cell membranes. J Exp Med. 169:737-754. Bhakdi, S., M . Muhly, S. Korom, and G. Schmidt. 1990. Effects of Escherichia coli hemolysin on human monocytes. Cytocidal action and stimulation of interleukin 1 release. J. Clin Invest. 85:1746-1753. Blight, M.A . , A . L . Pimenta, J.-C. Lazzaroni, C. Dando, L. Kotelevets, S.J. Seror, and LB. Holland. 1994. Identification and preliminary characterization of temperature-sensitive mutations affecting HlyB, the translocator required for the secretion of haemolysin (HlyA) from Escherichia coli. Mol Gen Genet. 245:431-440. Chervaux, C , and L B . Holland. 1996. Random and directed mutagenesis to elucidate the functional importance of helix II and F-989 in the C-terminal secretion signal of Escherichia coli hemolysin. J Bacteriol. 178:1232-1236. Childs, S., and V . Ling. 1994. The M D R superfamily of genes and its biological implications. Important Advances in Oncology. DeVita Jr., V.T., S. Hellman., and S.A. Rosenberg, Eds. JB Lippincott Co.: Philadelphia. Czuprynski, C.J. and R.A. Welch. 1995. Biological effects of R T X toxins: the possible role of lipopolysaccharide. 3:480-483. Eisenberg, D., E. Schwarz, M . Komaromy, and R. Wall. 1984. Analysis of membrane and surface protein sequences with the hydrophoic moment plot. J Mol Biol. 179:125-42. Eisenberg, D., R . M . Weiss, and T.C. Terwilliger. 1982. The helical hydrophobic moment: a measure of the amphiphilicity of a helix. Nature. 299:371-374. Eisenberg, D., R . M . Weiss, and T.C. Terwilliger. 1984. The hydrophobic moment detects periodicity in protein hydrophobicity. Proc Nat Acad Sci USA. 81:140-144. Eisenhaber, F., F. Imperiale, P. Argos, and C. Froemmel. 1996. Prediction of Secondary Structural Content of Proteins from Their Amino Acid Composition Alone. I. New Analytic Vector Decomposition Methods. Proteins: Struct Funct Design. 25:157-168. 54 Eisenhaber, F., C. Froemmel, and P. Argos. 1996. Prediction of Secondary Structural Content of Proteins from Their Amino Acid Composition Alone. II. The Paradox with Secondary Structural Class. Proteins: Struct Funct Design. 25:169-179. Eisenhaber, F., B. Persson, and P. Argos. 1995. Prediction of Protein Structure. Recognition of Primary, Secondary, and Tertiary Structural Features from Amino Acid Sequence. Critical Reviews in Biochemistry and Molecular Biology. 30:1-94. Endicott, J.A., and V. Ling. 1989. The biochemistry of P-glycoprotein-mediated multidrug resistance. Annu Rev Biochem. 58:137-151. Evans, S. 1997. Peptool. Vers. 1.0. CD-ROM. Edmonton: U of Alberta. Favre, D. 1992. Improved phenol-based method for the isolation of DNA fragments from low melting temperature agarose gels. Biotechniques. 13:22, 25-26. Felmlee, T., and R.A. Welch. 1988. Alterations of amino acid repeats in the Escherichia coli haemolysin affect cytolytic activity and secretion. Proc Natl Acad Sci USA., 85:5269-5273. Glaser, P., H. Sakamoto, J. Bellalou, A. Ullman, and A. Danchin. 1988. Secretion of cyclolysin, the calmoldulin-sensitive adenylate cyclase-haemolysin bifunctional protein of Bordetellapertussis. EMBO J. 7:3997-4004. Gray, L., K. Baker, B. Kenny, N. Mackman, R. Haigh, and LB. Holland. 1989. A novel C-terminal signal sequence targets Escherichia coli haemolysin directly to the medium. JCell Sci Suppl. 11:45-57. Hacker, J., C. Hughes, H. Hof, and W. Goebel. 1983. Cloned haemolysin genes from Escherichia coli that cause urinary tract infection determine different levels of toxicity in mice. Infect Immun. 42:57-63. Hess, J., I. Gentschev, W. Goebel, and T. Jarchau. 1990. Analysis of the haemolysin secretion system by PhoA-HlyA fusion proteins. Mol Gen Genet. 224:201-208. Higgins, C F . 1992. A B C Transporters: from microorganisms to man. Annu Rev Cell Biol. 8:67-113. Holland, LB., B. Kenny, and M . Blight. 1990. Haemolysin secretion from E. coli. Biochimie. 72:131-141. Issartel, J.P., V. Koronakis, and C. Hughes. 1991. Activation of Escherichia coli prohaemolysin to the mature toxin by acyl carrier protein-dependent fatty acylation. 351:759-761. 55 Juranka, J., F. Zhang, J. Kulpa, J. Endicott, M . Blight, LB. Holland, and V. Ling. 1992. Characterization of the haemolysin transporter, HlyB, using an epitope insertion. Biol Chem. 267:3764-3770. Kamtekar, S., J.M. Schiffer, H. Xiong, J.M. Babik, and M.H. Hecht. 1991. Protein Design by Binary Patterning of Polar and Nonpolar Amino Acids. Science. 262: 1680-1685. Kenny, B., R. Haigh, and LB. Holland. 1991. Analysis of the haemolysin transport process through the secretion from Escherichia coli of PCM, C A T or beta-galactosidase fused to the HlyA C-terminal signal domain. Mol. Microbiol. 5:2557-2568. Kenny, B., S. Taylor, and I. B. Holland. 1992. Identification of individual amino acids required for secretion within the haemolysin (HlyA) C-terminal targeting region. Mol. Micro. 6:1477-1489. Koronakis, V., E. Kornonakis, and C. Hughes. 1989. Isolation of analysis of the C-terminal signal directing export of Escherichia coli haemolysin protein across both bacterial membranes. EMBO J. 8:595-605. Koronakis, V. , Hughes, C , and E. Kornonakis. 1991. Energetically distinct early and late stages of HlyB/HlyD-dependent secretion across both Escherichia coli membranes. EMBO J. 10:3263-3272. Lee, A.B., and T.A. Cooper. 1995. Improved Direct PCR Screen for Bacterial Colonies: Wooden Toothpicks Inhibit PCR Amplification. BioTechniques. 18:225-226. Lemire, B.D., c. Fankhauser, A. Baker, and G. Schatz. 1989. The Mitochondrial Targeting Function of Randomly Generated Peptide Sequences Correlates with Predicted Helical Amphiphilicity. J Biol Chem. 264:20206-20215. Munoz, V., and L. Serrano. 1994. Elucidating the folding problem of a -helical peptides using empirical parameters, II. Helix macrodipole effects and rational modification of the helical content of natural peptides. J Mol Biol. 245:275-296. Munoz, V., and L. Serrano. 1994b. Elucidating the folding problem of a -helical peptides using empirical parameters III: Temperature and pH Dependence. J Mol Biol. 245:297-308. Munoz, V. and L. Serrano. 1997. Development of the Multiple Sequence Approximation within the Agadir model of a-Helix formation. Comparison with Zimm-Bragg and Lifson-Roig Formalisms. Biopolymers. 41:495-509. Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press: Cold Spring Harbor. 56 Samuels, M.L. 1988. Statistics for the life Sciences. Dellen Publishing Company: San Francisco. 74-96, 190-258. Shapiro, A.B. , A.B. Corden, and V. Ling. 1997. P-glycoprotein Mediated Hoechst 33342 Transport out of the Lipid Bilayer. Eur J Biochem. 250:115-121. Sheps, J. A., I. Cheung, and V. Ling. 1995. Hemolysin Transport in Escherichia coli. J Biol Chem. 270:14829-14834. Sheps, J.A., and F. Zhang. 1998. Personal Communication. Stanley, P., V. Koronakis, and C. Hughes. 1991. Mutational analysis supports a role for multiple structural features in the C-terminal secretion signal of Escherichia coli haemolysin. Mol Microbiol. 5:2391-2403. Stanley, P., L. Packman, V. Koronakis, and C. Hughes. 1994. Fatty acylation of two internal lysine residues required for the toxic activity of E. coli hemolysin. Science. 266:1992-1996. Strathdee, C.A., and R.Y.C. Lo. 1989. Cloning, nucleotide sequence, and characterization of genes encoding the secretion function of the Pasteurella haemolytica leukotoxin determinant. J Bacteriol. 171:916-928. Styer, L., 1988. Biochemistry. 3rd ed. W.H. Freeman and Co: New York. 107. von Heijne, G. 1986. Mitochondrial targeting sequences may form amphiphilic helices. EMBO J. 5:1335-1342. Wandersman, C , and P. Delepelaire. 1990. TolC, an Escherichia coli outer membrane protein required for haemolysin secretion. Proc Natl Acad Sci USA. 87:4776-4780. Wang, R., S.J. Seror, M . Blight, J.M. Pratt, J.K. Broome-Smith, and LB. Holland. 1991. Analysis of the membrane organization of an Escherichia coli protein translocator, HlyB, a member of a large family of prokaryote and eukaryote surface transport proteins. J Mol Biol. 217:441-454. Wolff, N., J.M. Ghigo, P. Delepelaire, C. Wandersman, and M . Delepelaire. 1994. C-terminal secretion signal of an Erwinia chrysanthemi proteinase secreted by a signal peptide-independent pathway: proton NMR and CD conformational studies in membrane-mimetic environments. Biochem. 33:6792-6801. Yin, Y., F. Zhang, V. Ling, and C. H. Arrowsmith. 1995. Structural analysis and comparison of the C-terminal transport signal domains of hemolysin A and leukotoxin A. FEBS Letters. 366:1-5. 57 Zhang, F., J.A. Sheps, and V. Ling. 1993a. Complementation of transport-deficient mutants of Escherichia coli a-hemolysin by second-site mutations in the transporter hemolysin B. J Biol Chem. 268:19889-19895. Zhang, F., D.I. Greig, and V. Ling. 1993b. Functional replacement of the hemolysin A transport signal by a different primary sequence. Proc Natl Acad Sci USA. 90: 4211-4215. Zhang, F., Y. Yin, C. H. Arrowsmith, and V. Ling. 1995. Secretion and circular dichroism analysis of the C-terminal signal peptides of HlyA and LktA. Biochem. 34:4193-4201. 58 

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.831.1-0088466/manifest

Comment

Related Items