SIGNAL PEPTIDASE SPECIFICITY AND SUBSTRATE SELECTION: INFLUENCE OF S1 AND S3 SUBSTRATE BINDING POCKET RESIDUES ON SPASE I CLEAVAGE SITE SELECTION

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Andrew Karla, B.S.

*****

The Ohio State University 2005

Dissertation committee: Approved by Professor Ross E. Dalbey, Advisor Professor Ming-Daw Tsai Professor Sean Taylor Professor David J. Hart Advisor Chemistry Graduate Program

ABSTRACT

Signal peptidase, which removes signal peptides from preproteins, has a substrate

specificity for small uncharged residues at -1 (P1) and small or larger aliphatic residues at

the -3 (P3) position. Structures of the catalytic domain with a 5S-penem inhibitor and a

lipopeptide inhibitor reveal candidate residues that make up the S1 and S3 pockets that

bind the P1 and P3 specificity residues of the preprotein substrate. We have used site-

directed mutagenesis, mass spectrometric analysis, in vivo and in vitro activity assays as

well as molecular modeling to examine the importance of the substrate pocket residues in their ability to promote cleavage of the pro-OmpA-nuclease A substrate with WT and mutant processing regions. Generally, we find that the S1 and S3 binding sites can

tolerate changes that are expected to increase or decrease the size of the pocket without

large effects on activity. One residue that contributes to the high fidelity of cleavage of

signal peptidase is the Ile 144 residue. Changes of the Ile 144 residue to cysteine result in

cleavage at multiple sites, as determined by mass spectrometry and Edman sequencing

analysis. In addition, we find that signal peptidase is able to cleave after phenylalanine at

the -1 residue in a double mutant where both Ile 86 and Ile 144 were changed to an alanine. Also, alteration of the Ile 144 and Ile 86 residues to the corresponding residues found in the homologous Imp1 protease changes the specificity to promote cleavage

ii following a –1 Asn residue. This work shows that Ile 144 and Ile 86 contribute to the signal peptidase substrate specificity and that Ile 144 is important for the accuracy of the cleavage reaction.

Additionally, studies with corresponding β-lactamase substrate mutants were used to further investigate the altered specificity revealed in the pro-OmpA-nuclease A studies.

The use of β-lactamase as a substrate allows for a convenient antibiotic selection method to assay for processing in vivo. In a preliminary study, a two plasmid system was developed to assay combinations of SPase and β-lactamase substrate mutants. This study assayed for growth of the temperature sensitive SPase strain, IT41, that was expressing various combinations of SPase and β-lactamase substrate mutants. Growth under conditions of high ampicillin concentration would suggest efficient processing of β- lactamase. Growth was observed in some cases when the β-lactamase substrate carried a relatively conservative -1V substitution in place of the WT -1A residue. The two-plasmid system utilizing the β-lactamase substrate developed in these studies has potential to be implemented in an unbiased genetic selection technique to further define the substrate specificity determinants of signal peptidase.

The β-lactamase substrate was also assayed in a pulse-chase study to investigate processing by select SPase mutants found to affect specificity in the pro-OmpA-nuclease

A studies. These studies largely confirm the results observed with the pro-OmpA- nuclease A substrate. One added benefit of using β-lactamase as a substrate is that the

iii signal peptide cleavage region possesses no alternate cleavage sites which greatly simplifies data analysis.

iv

This work dedicated to my parents and my family

v

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Ross Dalbey, for his advice and guidance

throughout my graduate studies. His continuing enthusiasm and support are to a great

degree responsible for my success in these investigations.

I would like to thank the other professors that have been involved in these studies

directly as collaborators and indirectly as sources of valuable advice. Their continuing

involvement has meant much to me: collaborators Dr. Mark Lively, Dr. Mark Paetzel, Dr.

Natalie Strynadka; and others Dr. Ming-Daw Tsai, Dr. Dehua Pei and Dr. Martin Caffrey.

I would also like to thank my graduate committee for their assistance in the development of this dissertation: Dr. Ming-Daw Tsai, Dr. Sean Taylor and Dr. David

Hart.

Also, I would like to express gratitude to my friends and coworkers in Johnston

Lab. It is the involvement of these people that has made this experience not only an enriching one but also an enjoyable one: Li Zhao, Minyong Chen, Joseph Carlos, Fenglei

Jiang, Özlem Doğan Ekici, Liang Yi, Hyunjin Cho, Eunjung Shim, Nil Celebi, Yuxia

Dong, Kun Xie, Jijun Yuan and members of the Pei and Tsai research groups.

vi Lastly, I would like to thank my parents Paul and Susan, my brother Stephen, my sisters Elizabeth and Margaret, and my fiancée Mary Cerny. I am forever grateful for the love and support they have given me over the years.

vii

VITA

1975……………………….. Born in Cleveland, Ohio

1998……………………….. B. S. in Biochemistry, University of Dayton, Dayton, Ohio

1998-Present……………….. Department of Chemistry, The Ohio State University,

Columbus, Ohio. Fellowship support provided by the

Chemistry Biology Interface Training Program and through

support as a Graduate Teaching and Research Associate.

PUBLICATIONS

Karla A, Lively MO, Paetzel M, Dalbey RE (2005). The identification of residues that control signal peptidase cleavage fidelity and substrate specificity. J Biol Chem 280(8): 6731-41.

Paetzel M, Karla A, Strynadka NC, Dalbey RE (2002). Signal peptidases. Chem Rev 102(12): 4549-80. Review.

Carlos JL, Paetzel M, Brubaker G, Karla A, Ashwell CM, Lively MO, Cao G, Bullinger P, Dalbey RE (2000). The role of the membrane-spanning domain of type I signal peptidases in substrate cleavage site selection. J Biol Chem 275(49): 38813-22

viii FIELDS OF STUDY

Major field: Chemistry

Specific field: Signal peptidase and membrane protein insertion

ix

TABLE OF CONTENTS

Abstract ……………………………………………………………………………. ii Dedication………………………………………………………………………….. v Acknowledgements………………………………………………………………… vi Vita………………………….……………………………………………………… viii

List of Tables……………………………………………..……………………...… xii List of Figures…………………………………………….…………………...…… xiii

Chapters:

1. Introduction………………………………….………………………...………… 1

1.1 Signal hypothesis………………………………….……………………….. 1 1.2 SPase I family of proteases………………………………….……….…..… 2 1.3 SPase I substrates: structure of signal peptides………………………...….. 6 1.4 Mechanism of type I SPases………………………………….………….… 7 1.5 Protein engineering: understanding and manipulating enzyme stability and specificity. ………………………………….………………... 12

2. The identification of residues that control signal peptidase cleavage fidelity and substrate specificity………………………………….………..…… 27

2.1 Introduction………………………………….………………………..….… 27 2.2 Results……………………………………………………………………… 29 2.2.1 In vitro analysis of SPase binding site mutants with the wild-type preprotein substrate……………………………………...… 29 2.2.2 In vitro analysis of SPase cleavage of pre-protein substrate mutants…………………………………………………………..…… 30 2.2.3 Mass Spectrometric analysis of signal peptide cleavage……….……. 31 2.2.4 In vivo analysis of signal peptidase binding site mutants………….… 36 2.2.5 Molecular modeling of binding site mutants of signal peptidase in complex with signal peptides…………………………..………..… 37 2.3 Discussion……………………………………………………………..…… 38 2.4 Materials and Methods………………………………………………..…… 45 2.4.1 Bacterials Strains and Plasmids……………………………………… 45 2.4.2 DNA methods……………………………………………………...… 45 2.4.3 Purification of signal peptidase and pro-OmpA-nuclease A…….… 46 x 2.4.4 In vitro and in vivo assay of Signal Peptidase Cleavage……….……. 46 2.4.5 Study to assay ability of mutants to complement the growth defect of the temperature-sensitive SPase strain, IT41(DE3)……..…. 47 2.4.6 Automated Edman Degradation…………………………………...…. 47 2.4.7 Matrix Assisted Laser Desorption Ionization Time-of-Flight (MALDI-TOF) Mass spectrometry………………………………...… 48 2.4.8 Prediction of theoretical mass for cleavage products…………...…… 49 2.4.9 Molecular Modeling……………………………………………….… 49

3. Analysis of cleavage fidelity and substrate specificity of signal peptidase S1/S3 mutants with β-lactamase substrate in vivo………………………….…… 67

3.1 Introduction………………………………………………………………… 67 3.2 Results: In vivo growth assays…………………………………………...… 69 3.2.1 WT β-lactamase……………………………………………………… 69 3.2.2 -1V β-lactamase……………………………………………………… 72 3.2.3 -1F β-lactamase…………………………………………………….… 73 3.2.4 -1N β-lactamase ……………………………………..……………..... 73 3.3 Discussion……………………………………………………………….…. 74 3.3.1 Genetic applications………………………………………..………… 78 3.4 Results: Pulse-Chase Studies of in vivo processing……………………...… 79 3.4.1 Processing of WT β-lactamase………………………………….…… 79 3.4.2 Processing of –1V β-lactamase………………………………….…… 80 3.4.3 Processing of –1F β-lactamase……………………………….……… 81 3.4.4 Processing of –1N β-lactamase………………………………….…… 82 3.5 Discussion…………………………………………………………….……. 83 3.6 Materials and Methods………………………………………………...…… 87 3.6.1 Bacterial strains and plasmids……………………………………...… 87 3.6.2 Construction of plasmids…………………………………………..… 87 3.6.3 Purification of β-lactamase……………………………..……………. 88 3.6.4 Pulse-chase assay of β-lactamase processing………………...……… 89 3.6.5 Growth assay for β-lactamase processing………………………….… 90 3.6.6 Conclusion…………………………………………………………… 90

List of References………………………………………………………………..… 100

xi

LIST OF TABLES

Table

2.1 Activity of mutant SPases relative to WT SPase…………………..………. 62

2.2 Mass spectral data of PONA cleavage products…………………………… 63

2.3 Predicted theoretical masses of various PONA cleavage products…….….. 65

2.4 In vivo processing of pro-OmpA in the temperature sensitive strain IT41(DE3) at the nonpermissive temperature 42ºC………….………....…. 66

xii

LIST OF FIGURES

Figure

1.1 Protease catalytic mechanisms…………………………………………….. 22

1.2 Inhibitors of SPaseI………………………………………………………… 24

1.3 A model of signal peptidase in complex with the signal peptide of the substrate pro-OmpA-nuclease A (PONA)………...………. 25

2.1 Effect of mutations that are predicted to increase or decrease the size of the binding pocket on SPase activity……………….… 50

2.2 Processing of pro-OmpA-nuclease A (PONA) with alternate -1 and -3 residues………………………………………………………….. 53

2.3 Mass analysis demonstrates altered cleavage profile for SPase mutants……………………………………………………………… 54

2.4 Activity of SPase mutants with the WT and mutant PONA substrates…………………………………………………………………… 56

2.5 Processing of WT and -1 Asn PONA substrates by WT, I144C/I86C and I144C/I86T mutants……………………………………… 57

2.6 In vivo complementation of OmpA processing defect in IT41(DE3) at 42°C………………………………………………………… 58

2.7 Molecular modeling of binding site mutants with in E. coli signal peptidase…………………………………………………….. 59

2.8 The cleavage site of signal peptide mutants by various signal peptidase mutants…………………………………………………...……… 61

3.1 Depiction of sequence surrounding PONA and β-lactamase cleavage sites………………………………………………………….…… 94

3.2 Growth assay for β-lactamase processing……………………………….…. 95 xiii

3.3 Processing of β-lactamase WT and mutants by SPase in vivo……..……… 97

3.4 Purification of WT β-lactamase from periplasmic fraction…………..……. 99

xiv

CHAPTER 1

INTRODUCTION

1.1 Signal hypothesis

The processes that allow the living cell to function as a unit and as part of an organism

require the coordination of an immense number of enzymatically-catalyzed chemical

reactions. In order to accomplish this incredible feat, higher organisms have evolved

multiple membrane systems to order and compartmentalize these enzymatic reactions in

parts of the cell that are appropriate for their function. Even the relatively simple bacterial

cell possesses one or more membrane systems that organize enzymatic processes that

occur at this interface with the external environment.

Realizing the importance of organization in the proper compartmentalization of these

reactions, Günter Blobel and David Sabatini published the first description of the signal hypothesis where they proposed that proteins carry in their primary amino acid sequence

the information necessary for targeting to the cellular membrane [1]. Subsequently,

Milstein and colleagues published a study where they examined the light chain portion of

IgG from myeloma cells and observed that this protein subunit was synthesized in a 1 higher molecular weight form than was expected and that it was then converted to the proper molecular weight upon addition of endoplasmic reticulum microsomes [2].

Several years after the first report of the signal hypothesis, Blobel added to his hypothesis suggesting that distinct signal sequences within the protein were responsible for specific targeting of proteins to the various membrane systems such as the ER, mitochondria, chloroplast and peroxisome within the cell [3]. In many studies to follow, Blobel’s hypothesis was confirmed and it was found that the distinct signal sequences are located at the amino terminus of the protein and after targeting they are often removed by various signal peptidases (SPases) that are specific to the different membrane systems where they reside.

1.2 SPase I family of proteases

SPases have been identified and characterized in a wide variety of Gram-negative and

Gram-positive bacteria, mitochondria, and the endoplasmic reticulum (ER) (for reviews see [4, 5]). There are several families of signal peptidases that have been characterized with each having distinct substrate requirements and intracellular location. However the most thoroughly studied type and indeed the subject of this work is the Type I Signal

Peptidases (SPase I)

The SPase I family includes enzymes originating in both Gram-positive and Gram- negative bacteria [6], as well as the ER, mitochondria and chloroplast in eukaryotes [7].

SPase I in its simplest form consists of a single polypeptide unit but in higher organisms

2 often consists of a multisubunit complex. In all cases these are membrane bound enzymes

that function to remove signal peptides from translocated substrate proteins. These

enzymes possess similar substrate determinants for cleavage. With the exception of the

inner membrane peptidase of the mitochondria, all favor small aliphatic residues at the –1

and –3 positions within the C-terminal region of the signal peptide relative to the

cleavage site [4]. These are usually alanine residues and result in the ubiquitous “Ala-X-

Ala” motif for signal peptide cleavage [8-10].

SPase I from bacteria were the first signal peptidases to be cloned [11], sequenced [12]

and characterized [13-17]. These enzymes consist of a single polypeptide unit with the

active site located on the trans side of the membrane. Most bacteria have a single

chromosomal copy of the SPase gene; however, some exceptions exist. In Bacillus

subtilis there are five chromosomal copies, SipS, SipT, SipU, SipV and SipW, and two

plasmid encoded signal peptidases [18, 19] while Streptomyces lividans has four [20] and

Staphylococcus aureus has two [21]. In all cases though SPase has been found to be an

essential enzyme for cell viability. Disrupting the SPase I gene in E. coli or placing it

under the control of an inducible promoter has proven the essential nature of this gene

[22, 23], while in B subtilis, both SipS and SipT must be inactivated to cause cell death

[24]. In E. coli, Nakamura et al. have developed an amber read-through mutation within

the SPase I gene to produce a temperature-sensitive strain named, IT41, that cannot

survive when this gene is deactivated [25]. Similar studies involving disruption and

selective induction have proven the essential function of Spase I in human pathogenic bacteria Streptococcus pneumonia, and S. aureus [21, 26].

3

The best-known bacterial SPase I is that found in E. coli. It consists of a single

polypeptide chain of 37 kDa. [27]. This enzyme is an integral membrane protein that

spans the membrane twice with a small N-terminal domain and a large carboxyl-terminal

domain containing the active site located in the periplasm [12, 28, 29]. Other examples of

varying topologies exist among bacterial SPases but all possess a large active domain on

the trans side of the membrane. For example, B. subtilis SPase possesses a single

membrane anchor while Haemophilus influenzae has three transmembrane regions [30].

In comparison to the bacterial SPase I, the signal peptidases of the ER membrane are relatively complex. In the ER, SPase consists of an assembly of multiple polypeptide subunits that comprise a functional SPase complex. The ER Spase complex has been characterized in a number of systems. One of the first to be characterized was described

in 1986 as isolated from canine pancreas microsomes [31]. This complex was shown to consist of five polypeptides that vary in molecular weight from 12 to 25 kDa and include

SPC12, SPC18, SPC21 SPC22/23 and SPC25. In addition, the SPase I complex has been isolated from chicken oviduct microsomes and was found to consist of two subunits, gp23 and p19 that were named according to their apparent size in kDa [32, 33]. The

SPase I complex from Saccharomyces cerevisiae has also been characterized and as in the canine complex, consists of at least four subunits of 13, 18, 20, and 25 kDa [34, 35].

The 18 kDa protein was identified as the Sec11 gene product, Sec11p from a genetic screen [36].

4 Of these polypeptides in the ER SPase I complexes discussed above, the SPC18, SPC21,

p19 and Sec11p form a family of related proteins that are believed to contain the

peptidase active site(s). The SPC18 and SPC21 subunits are homologous isoforms with

amino acid sequences that are 80% similar [37, 38] and are each approximately 47%

identical to Sec11p [38]. The avian p19 protein is also closely related to this family of

proteins (S. J. Walker & M. O. Lively, unpublished).

In addition to being found in the ER membrane, type I SPases have been isolated from

the mitochondria as well as in chloroplasts. The mitochondrial inner membrane protease

complex, as its name would suggest, is located in the mitochondrial inner membrane with its active domain within the inner membrane space [39, 40]. It exists as a complex of two

homologous polypeptide subunits, Imp1 and Imp2, of 21 and 19 kDa in mass [40-42].

These proteins are related to the bacterial and ER type I peptidases. Both Imp1 and Imp2 have proteolytic activity but are distinct within this protease family in that these subunits possess non-overlapping substrate specificities: Imp2 exhibits the typical Ala-X-Ala substrate specificity while the Imp1 protease deviates from this norm in selecting substrates with an asparagine at the –1 position.

The chloroplast type I SPase, TPP, is located in the thylakoid membrane and is also

homologous to the bacterial enzyme. Like the bacterial SPase I, it seems to function as a

single polypeptide subunit [43]. TPP has been predicted to span the membrane once with

the catalytic domain facing the thylakoid lumen.

5 1.3 SPase I substrates: structure of signal peptides

The natural substrates for type I SPases are preproteins that are destined for either export from the bacterial cell cytoplasm or import into the ER lumen, mitochondria or chloroplast. The signal peptides of these proteins generally do not have any homology but do possess the following conserved structural features: a positively charged amino-

terminal domain (n), a central hydrophobic core (h) and a hydrophilic carboxy-terminal

domain (c). It is this final carboxy-terminal domain that contains the SPase processing site which allows the signal peptide to be proteolytically removed from the mature protein.

Within the signal peptide, the hydrophobic core is important for proper export. The importance of this region was first illustrated in a study by Beckwith et al. [44]. In this study, fusions were made between the amino-terminal region of the maltose binding protein (malE gene) and β-galactosidase (lacZ gene). Cells expressing these fusions were maltose sensitive because the high expression levels of the fusion protein upon addition of maltose resulted in jamming of the secretion machinery causing cell death. In the course of the experiment, maltose resistant revertants were isolated and upon analysis of the sequences it was discovered that nearly all possessed mutations in the signal peptide region of the malE portion. These revertant mutations consistently resulted in the incorporation of charged residues into the central hydrophobic core of the signal peptide or resulted in deletions within this region [45]. In disrupting the hydrophobicity of the h- region, the export of the fusion protein was disabled thus preventing the jamming effect.

6

The positively charged amino-terminal domain is also important for export but is of secondary importance compared to the h-region. The influence of the n-region on export is usually observed only when the hydrophobicity of the h-region is first disrupted [46].

One possible rationalization for this is that the positive charges interact electrostatically with the negative charges carried by the phospholipid head groups in the membrane [47].

This is in agreement with the observation that membrane proteins seem to have topology determinants partially dictated by the placement of positively-charged residues located on the cytoplasmic loops of the protein. This effect has come to be known as the “positive inside” rule [48].

The hydrophobic core of the signal peptide spans the membrane as an α-helix that terminates near the junction with the c-region where helix breaking residues are typically found. Following the helix breaking residue, the signal peptide cleavage region loses its helical structure and when in contact with the Spase I enzyme, assumes an extended β- sheet type structure (Figure 1.3A) [49].

1.4 Mechanism of type I SPases

Since the very early characterization of signal peptidases, it was hypothesized that these enzymes belonged to a novel peptidase class. They were thought to be of a distinct mechanistic class because they were found to be resistant to peptidase inhibitors effective against the classical serine, cysteine, aspartic acid and metallo classes of peptidases.

7

Before structural information was available, there were two residues within the E. coli

SPase I that were identified as being important for catalysis. Mutational studies showed

that when Ser 90 was changed to an alanine, this resulted in a completely inactive

enzyme [14]. This Ser 90 residue is conserved throughout this family of proteases and similar mutational studies were repeated with the conserved serine residues in the B subtilis SipS signal peptidase [50], as well as the Imp2 subunit of the mitochondrial inner membrane protease [40] which resulted in inactive enzymes. These studies strengthened support for the hypothesis that the SPase I enzymes belong to an unusual serine protease class.

Another factor suggesting a non-standard serine type protease was the fact that there are no conserved histidine residues within the prokaryotic signal peptidase family.

Additionally, when each of the three histidine residues were mutated within the E. coli

SPase, there was no observable effect on catalysis [14, 51]. These results seemed to rule out the possibility that these enzymes utilize the classical Ser-His-Asp catalytic triad mechanism (Figure 1.1A). However, it was recognized that there was a conserved lysine

(Lys 145) residue of E. coli that was present in all bacterial and mitochondrial signal peptidases that had been sequenced at that time and when Lys 145 was mutated to an alanine, there was a severe drop in catalysis observed [15, 52]. Additional residues were tried in place of Lys 145 including Met [52], His and Asn [15] and were found to abolish activity in SPase. These types of experiments were conducted on B subtilis SipS as well with similar results [50]. Though these results established that the conserved lysine is

8 critical for catalysis, they did not directly establish that this residue is involved in proton transfer in the enzyme active site. However, it was in the analysis and rationalization of this mutational information that the idea of the Ser-Lys catalytic dyad was formulated. In this model, the mechanism resembles the classic Ser-His-Asp catalytic triad mechanism with the nucleophilic serine being activated by the lysine general base instead of histidine

(Figure 1.1B).

In contrast, the conserved lysine residue is notably absent from the ER signal peptidases.

In aligned sequences at the position corresponding to the critical lysine residue, the ER and archaeal SPases have a histidine. Also present is a conserved serine corresponding to

Ser 90 and an absolutely conserved aspartic acid residue [4]. When these residues were mutated in the Sec11 protein it was discovered that the conserved serine, histidine and aspartic acid were all essential for catalytic activity but when all of the lysines in Sec11 were changed there was no effect [53]. From this information, it was concluded that the

ER type signal peptidase I likely functioned by either a Ser-His catalytic dyad mechanism where the His acts alone as general base or by the classic triad mechanism.

A significant breakthrough in the understanding of bacterial SPaseI was achieved through the crystallization and structural determination of the soluble catalytic domain in complex with a penem inhibitor [17]. This structure was important in that it was the first of the

SPase I family to be crystallized and solved in complex with an inhibitor. This structure proved that the conserved Ser 90 was the nucleophilic residue as it was shown to be

9 covalently attached to the cleaved (5S)-penem inhibitor. Also, the Ser 90 Oγ was found to be within hydrogen binding distance of the proposed general base Lys 145.

The crystal structure of the E. coli signal peptidase with covalently attached inhibitor demonstrated conclusively that SPase is a serine protease and is consistent with Lys 145 being involved in the activation of the nucleophilic Ser 90 ( Figure 1.1B, Figure 1.3) [17].

The structure provided a picture of how SPase can bind and cleave signal peptides from exported proteins. Figure 1.3 shows a structural model of the E. coli SPase (shown in a light grey ribbon diagram) with a theoretical model of a signal peptide (yellow) docked into its substrate binding sites (displayed as a molecular surface, green, red, cyan). One of the striking features of the protein structure is a large hydrophobic patch comprised of an antiparallel β-strand domain (residues 81-85, 99-105, 292-307 and 312-314) that we believe allows the catalytic domain to interact with the membrane during catalysis.

Figure 1.3A shows the proposed orientation of SPase relative to the membrane surface with the yellow hydrophobic domain of the signal peptide extending downwards into the lipid bilayer.

This type of catalytic mechanism is not without precedent and there is now a growing family of enzymes that catalyze their reactions by way of a Ser-Lys dyad mechanism.

This catalytic mechanism has been implicated in the activity of SipS of B. subtilis [50].

Also, the LexA protein, which is involved in the SOS response to DNA damage was the first protease enzyme recognized to carry out catalysis using a Ser-Lys dyad [54, 55]. In addition, UmuD is another member of the LexA family and also uses a Ser-Lys dyad

10 mechansim [56]. It is also interesting that the LexA family proteases also possess similar

–1, -3 substrate requirements as the bacterial SPase I [50]. The tail specific protease in E.

coli is another enzyme that carries out catalysis in this way. This was shown by Sauer and

colleagues in mutational studies where the Ser and Lys residues were shown to be

critical. [57]. Also likely to utilize this mechanism of catalysis are the viral Lon protease

as well as the bacterial and organellar Lon proteases [58]. Though they are not proteases,

some amidase enzymes have also been found to utilize Ser-Lys catalytic dyads in a

mechanistically-similar reaction. One example is the RTEM-1 β-lactamase from E. coli

[59].

Because of their unique mechanism, it has been a challenging problem to develop inhibitors against bacterial type I SPases. As mentioned above, none of the commonly

used inhibitors against Ser-, Cys-, Asp- or metallo- classes of proteases are effective

against SPase I. The first inhibitor that was found was discovered by Kuo et al. when they

showed that certain β-lactams could inhibit the enzyme [60]. β-lactam type molecules

have been shown to be effective inhibitors against other Ser proteases and β-lactamases

as well [61-64]. The observation that the mechanism of SPase I utilized a Ser-Lys dyad

was important information in this respect because this is similar to the mechanism for the

β-lactamase enzymes which use a lysine residue as a general base in the acylation step

[59]. With this information, work was done by Smithkline Beecham Pharmaceuticals to

develop better inhibitors based on the β-lactam architecture. Figure 1.2 shows some of

the best inhibitors that were found and include clavams, thioclavams, and penem

carboxylates. Several of these types of compounds that they developed had IC50 in the

11 0.26 to 50 µM range [65]. The most potent inhibitor that was found was the allyl (5S,6S)-

6-[(R)-acetoxyethyl]-penem-3-carboxylate inhibitor which was the one that was bound

within the active site of the SPase I crystal structure.

In addition to small molecules, SPase I has been observed to be inhibited by some signal peptides such as the signal peptide of the M13 coat protein [66]. In addition, the substitution of a proline residue at the +1 position of pre-MBP prevents processing and causes accumulation of preproteins suggesting a competitive type inhibition [67]. Finally, in paper recently published by Paetzel et al. an Arylomycins type inhibitor for SPase I

was presented [49]. Arylomycins are lipohexapeptides and inhibit SPase by non-covalent

binding in the region of normal signal peptide binding are lipohexapeptides. This

complex was stable enough that a crystal structure with this bound inhibitor could be solved to 2.5Å resolution [49].

1.5 Protein engineering: understanding and manipulating enzyme stability and

specificity

Protein engineering has been a fact of life from the very beginning. The natural process

of evolution has produced enzymes of fantastic specificity and catalytic efficiency.

However, the natural evolution of proteins is a very slow process and one that is fraught

with many dead ends and failed attempts. As proteins evolve altered specificities, they proceed by incremental random changes that have the ability to both positively and

12 negatively affect activity. Changes that result in positive change are selected by

conferring an enhanced ability of the organism to survive.

The ability to rationally engineer proteins to study the mechanisms of catalysis and

substrate specificity represents the culmination of the combined technologies of

molecular biology, chemistry and physics. Advances in the ability to crystallize proteins

and the increased accessibility to high energy x-ray sources has led to an explosion in the

number of high resolution crystal structures in recent years. These structures have

provided the previously unprecedented ability to visualize, often on the atomic scale, the

structure and arrangement of amino acids in the enzyme. Careful analysis of the

structures of enzymes as well as the structures of these enzymes in complex with

inhibitors and substrate analogs has given researchers the ability to reveal the molecular

basis of enzyme/substrate binding and catalysis. Additionally, the tools of molecular

biology have increased so much that the manipulation of the genetic sequences which

code for proteins has become routine.

Methods in protein engineering have advanced significantly over the years. One of the

first methods that allowed for the selective modification of the protein sequence was site-

directed mutagenesis. This method dates back to the 1970s with the work of Hutchison

and Smith [68] which led to the technique of oligonucleotide directed mutagenesis which was conceptually similar to the enzymatic sequencing method developed by Sanger [69].

This method has evolved slightly in its modern form but still remains one of the most useful methods of protein modification. More significantly, it and other genetic methods

13 such as random mutagenesis and gene shuffling have allowed new possibilities in the

study of protein stability and the relationships between protein structure and function

One aspect of protein engineering that has extensive applications in industry and

commercial products has been the engineering of protein stability. There are many

commercial products that are used in every day life that contain enzymes or require the

use of enzymes in their commercial manufacture. It is often the case, though, that the

properties required of an enzyme in an industrial or commercial application are very

different from the conditions experienced in the enzyme’s natural environment. These

can include extremes of pH, temperature, and ionic strength as well as the presence of

detergents and denaturants. There are many interactions with the protein structure that

contribute to enzyme stability. Such interactions include hydrophobic forces, hydrogen bonds, electrostatic forces and metal binding interactions. Modification of these factors in

some cases has led to dramatic enhancement of protein stability.

The hydrophobic effect is one of the most significant factors contributing to protein stability. Nearly all amino acid side chains buried within the protein structure are hydrophobic, or at least non-polar. The transition of these hydrophobic residues from the aqueous environment to the protein interior is thought to be a major contributing factor to protein stability. Many attempts to stabilize proteins have involved the selective modification of the hydrophobicity of these interior residues. Barnase and binase are two related members of the microbial RNase family which share 85% identity. These two enzymes differ by 17 amino acids dispersed throughout the enzyme structure. In a study

14 done by the Fersht group, each of the 17 residues was mutated stepwise and they were found to affect protein stability +1.1 to -1.1 kcal/mol each [70]. A multiple mutant form of barnase was constructed which incorporated six of the stabilizing mutations and it was discovered that they contributed in an additive manner making the mutant enzyme 3.3 kcal/mol more stable.

In another study, the Schultz group used unnatural amino acid substitutions to probe the effect of side-chain structure on T4 lysozyme stability [71]. By replacing the WT leucine

133 with norvaline, ethylglycine and alanine, they examined the effect of stepwise removal of methyl groups from the hydrophobic core. They found that removal of one methyl group destabilized the T4 lysozyme mutant by 1.1 kcal/mol and the removal of the second further destabilized the enzyme by 2.2 kcal/mol. Additional substitutions that were made in an attempt to improve the packing in the destabilized enzyme yielded predictable success. This work underscored the importance of hydrophobic core residues on protein stability in addition to the importance of compact packing of core residues.

In addition to hydrophobic effects, electrostatic forces can be influential on protein stability. Electostatic forces most significantly take the form of ionic (or salt bridge) interactions and hydrogen bonding interactions. Salt bridges have an effect on protein stability by confining the protein structure locally by way of the charged interaction.

Their influence in engineering stability has been demonstrated by disrupting these interactions and examining the destabilizing effect. In a study with chymotrypsin inhibitor 2, Leatherbarrow and co-workers demonstrated that the native protein is

15 stabilized by a factor a of 1.4 kcal/mol by examining the destabilization that resulted

from mutation of the Arg67 residue to alanine [72]. In a similar study with barnase, the

Fersht lab found that the two salt bridges formed between Arg110 and both Asp8 and

Asp12 each contribute 1 to 1.25 kcal/mol in stability [73]. Engineering stability by incorporating salt bridges is difficult to achieve due to the precise geometry necessary for a stabilizing interaction.

In cases where the native enzyme is found in a metal ion-bound form, these metal ions can contribute significantly to stability. For example, the neutral protease from Bacillus

thermoproteolyticus contains four Ca2+ binding sites. In a study by Voordouw et al., it

was found that each of these metal ions contribute between 8.1 and 9.2 kcal/mol to this

enzymes stability [74]. This effect of Ca2+ ion stabilization and the ability to utilize this

effect to stabilize proteins was also demonstrated by Kuroki et al. in studies of c-type lysozymes and α-lactalbumin [75]. These enzymes were observed to have very similar amino acid sequence and tertiary structure and while the c-type lysozymes have lost their ability to bind Ca2+, they retain a loop structure similar to the EF-hand motif that binds

Ca2+ in α-lactalbumin. Engineering aspartic acid residues into this loop restored Ca2+

binding in the lysozyme resulting in an enzyme that was significantly more thermo-stable

than the WT [75].

Although the engineering of protein stability has improved the utility of enzymes in

commercial and research applications, the true power of protein engineering resides in its

ability to reveal the nature of enzymatic catalysis and specificity. Probing catalysis and

16 specificity of enzymes has most commonly been conducted by the stepwise modification of amino acid residues within the protein structure. This can be done on the basis of three-dimensional crystal structure information but is also commonly done as a way to probe the essential nature of residues found to be conserved in primary sequence alignments. These methods have seen such ubiquitous application in the study of enzyme catalysis that the scope of the discussion presented here will be largely confined to protease and protease-related enzymes.

Understanding the basic catalytic machinery that enzymes use to accelerate chemical reactions is a very fundamental aspect where protein engineering has been particularly informative. Most early studies have employed the modification of catalytic residues to look for loss of function thereby establishing the critical importance of these residues. A classic example is the elucidation of the chemical mechanism of serine proteases which comprise almost one-third of known proteases [76]. Conserved serine, histidine and aspartic acid residues were recognized in alignments of this family of enzymes but their key importance was discovered through structural analysis and sequence modification.

The catalytic triad was first seen in the crystal structure of chymotrypsin where Blow proposed that a hydrogen binding network between Asp102, His57 and Ser195 was responsible for activating the Ser195 for nucleophilic attack [77]. The involvement of these residues in catalysis was first established using a method of chemical modification by either removing the hydroxyl of Ser195 by selective modification by PMSF followed by chemical elimination or by methylating His57 [78, 79]. These modifications individually resulted in reductions in activity of 10,000 fold [80]. Similarly in trypsin,

17 these serine and histidine residues were changed by site directed mutagenesis to alanine

and each mutant exhibited a 105-fold reduction in activity [80]. In similar mutational

experiments, the importance of Asp102 was established by substitution with Asn which

also resulted in a 10,000-fold reduction in activity [81, 82].

In several compelling experiments by Carter and Wells, it was discovered that when the

catalytic histidine of subtilisin was mutated resulting in an inactive enzyme, substrates

containing His residues near the site of cleavage could rescue the activity of these mutant

enzymes by a substrate assisted catalysis mechanism [83, 84] } [85, 86]. The elucidation

of these catalytic residues as well as the ability to engineer substrate assisted catalysis

represents the advanced state of engineering proteins and substrates together to gain key

information about the catalytic mechanism

Protein engineering has not only revealed the mechanisms of serine proteases and other

classes of proteases, but it has also allowed for the understanding of substrate selection

and subsequent modification of substrate specificity. The combination of information gained from crystal structures and mutational evidence has led to the ability to alter enzyme specificity in a relatively predictable way. This has been accomplished most

notably with subtilisin as well as with the trypsin protease.

The enzymes subtilisin and α-lytic protease (αLP) are both excellent examples of the

application of these techniques. These enzymes represent extracellular proteases with

contrasting specificities. Whereas subtilisin prefers large hydrophobic residues in the

18 substrate P1 position, αLP prefers small hydrophobic residues at this position. The x-ray structure of αLP reveals the presence of a methionine residue (methionine 192) in the S1 substrate binding pocket that occupies what would otherwise be a large hydrophobic pocket. When this methionine was replaced by an alanine, this change was accompanied by an impressive 105-106 fold increase in the activity of αLP toward substrates with a phenylalanine at the P1 position [87]. In contrast, subtilisin has been successively modified to favor small hydrophobic residues at P1 by the substitution of larger hydrophobic residues at the position of glycine-166 in the S1 pocket, thereby reducing its volume [88]. As this work demonstrates, this technique can work with simple substitutions however it does not always produce the expected result.

The proteases trypsin and chymotrypsin are homologous proteases with one significant difference located in the S1 binding pocket. Chymotrypsin has the following residues contributing to the formation of a deep, relatively hydrophobic S1 binding pocket:

Ser189, Gly216 and Gly226 [89]. In contrast, trypsin has the same glycine residues but possess an Asp189 which gives a negatively charged character to its S1 pocket [90, 91].

As a result, the enzymes have contrasting specificities: chymotrypsin prefers phenylalanine at the P1 position over alanine by a factor of 50,000 [92] while trypsin prefers substrates with arginine or lysine at P1 [90, 91]. In attempting to engineer these specificities, the obvious change was made first. However when the Asp198 of trypsin was changed to a serine, it did not result in a chymotrypsin-like specificity for P1 phenylalanine [93-95]. In order to achieve the desired change in specificity, the S1

Asp189 to Ser mutation was made in addition to altering two surfaces loops by replacing

19 them with the sequences from chymotrypsin. The total substitutions necessary were

Asp189Ser, Gln192Met, Ile138Thr, insertion of Thr219, and insertion of the two loops

185-188 and 210-224. These were sufficient to achieve chymotrypsin specificity in the

trypsin enzyme but the addition of the Tyr172Trp mutation yielded activity toward P1

phenylalanine that was 85% that of WT chymotrypsin [96]. This illustrates that success

can be achieved with single substitutions as in subtilisin, this is not typical. Often the

changes necessary to achieve an enzyme with altered specificity and respectable activity

are numerous.

Due to the numerous mutational changes necessary to generate enzymes with altered

specificities or novel activities, there has been recent work to develop genetic methods to make many changes at once to generate vast libraries of possible combinations. One such method involves the exchange of genetic modules between related enzymes to produce hybrid enzymes with novel properties. This has been done extensively with sections of genes that possessed sequence homology [97-100]. Recently, though, a PCR-based technique was used in the Tsai group that focused on exchange of portions of genes based of structurally-conserved regions, which was named SCOPE for structure-based combinatorial protein engineering [101]. By exchanging these elements based on their structural folds, they were able to isolate several novel DNA polymerases with enhanced

phenotypes.

In addition, in the Benkovic laboratory, they have pioneered a method of Incremental

Truncation for the Creation of HYbrid enzymes or ITCHY [102, 103]. This method

20 utilizes enzymatic methods to shorten stretches of DNA incrementally with exonuclease

III. This enables the subsequent recombination of these pieces of DNA with others to produce hybrid enzymes of nearly limitless combinations. Coupled with an effective selection technique for the desired activity, these methods hold promise to be extremely efficient and effective in generating enzymes with novel properties.

21

A

O O P1' P1' Ser NH H O NH N N NH H O O H O N His P1 H P1 N NH N Ser HN His HN H N O O O P Asp P2 O 2 Asp O O Michaelis Complex Tetrahedral Intermediate I

Ser H Ser O N O N H H O O OH O H H H P1 H P1 N N N HN N HN His O O N His P2 N P2 H O O Asp Asp O O Acyl-enzyme Intermediate Tetrahedral Intermediate II

Figure 1.1 Protease catalytic mechanisms Above are depicted the proposed catalytic mechanisms of A. the classical serine protease catalytic triad mechanism involving Ser, His and Asp catalytic residues and B. the novel Ser, Lys catalytic dyad mechanism of Spase I.

(Continued)

22

Figure 1.1 (Continued)

B

23

H H (O)n O S R1 R N N 1 O R O 2 R2 Clavams Thioclavams

H H S O S H3C N N O O O O O O O R1

Penem Carboxylates Ally(5S,6S)-6-[(R)-acetoxyethyl]-penem-3-carboxylate

Figure 1.2 Inhibitors of SPaseI Above are depicted some of the most potent inhibitors of SPaseI and include clavams, thioclavams, and penem carboxylates. These inhibitors have IC50s in the 0.26 to 50 µM range [65]. The most potent inhibitor that was found was the allyl (5S,6S)-6-[(R)- acetoxyethyl]-penem-3-carboxylate inhibitor which is the one that is bound within the active site of the SPase I crystal structure (Figure 1.3).

24

Figure 1.3 A model of signal peptidase in complex with the signal peptide of the substrate pro- OmpA-nuclease A (PONA). A. A ribbon diagram of signal peptidase is rendered in white. The molecular surface is shown for those residues of signal peptidase within 5 Å of the PONA signal peptide (yellow). Two phospholipids molecules are shown in van der Waals spheres (green and red) in order to give perspective and orientation relative to the membrane. B. A close up view of the signal peptidase substrate binding pockets. The molecular surface is colored green for those signal peptidase residues contributing side chain atoms to the S1 pocket. The molecular surface is colored cyan for those residues contributing side chain atoms to the S3 pocket. The molecular surface is colored red for the residues isoleucine 86 and isoleucine 144. The surface main chain atoms and surface atoms not part of the S1 or S3 binding pockets are colored white. The signal peptidase residues are labeled in black and the P1 alanine and P3 alanine of the PONA signal peptide are labeled in yellow.

(Continued)

25

Figure 1.3 (Continued)

26

CHAPTER 2

THE IDENTIFICATION OF RESIDUES THAT CONTROL SIGNAL PEPTIDASE

CLEAVAGE FIDELITY AND SUBSTRATE SPECIFICITY

2.1 Introduction

In both prokaryotic and eukaryotic organisms, SPase exhibits similar substrate requirements. With the exception of mitochondrial Imp1, type 1 SPase specifically recognizes substrates with small amino acid residues at the –1 (P1) position or small uncharged residues at the –3 (P3) position of the signal peptide relative to the cleavage site[104-107]. Alternatively, the mitochondrial Imp1 cleaves substrates following a -1 asparagine residue [40, 108, 109]. The identity of determinants within SPase that control cleavage of the substrate at the correct site and not at other nearby potential sites remains unknown. The substrate specificity determinants within SPase that allow it to cleave

signal peptides following the -3, -1 rule are also undefined. One clue comes from the structure. The structure of E. coli SPase with a bound inhibitor indicates that non-polar atoms from the residues Ile 86, Pro 87, Ser 88, Gly 89, Ser 90, Met 91, Leu 95, Tyr 143,

Ile 144, and Lys 145 form the S1 pocket. In addition, modeling studies suggest that non-

27 polar atoms from the residues Phe 84, Ile 86, Ile 101, Val 132, Asp 142, and Ile 144 comprise the S3 pocket[17]. The residues Ile 144 and Ile 86 (colored red in Figure 1.3) bridge the S1 and S3 binding pockets.

As a first step to identify residues that control cleavage fidelity and substrate specificity, we mutated residues in the E. coli SPase substrate binding site and examined the effects of those changes on the cleavage of various substrates. We found that substitution of

SPase Ile 144 by a cysteine residue results in cleavage of the pro-OmpA-nuclease A substrate at multiple sites. Cleavage at multiple sites is in contrast to the action of wild- type SPase which cleaves exclusively at the correct site. The cleavage at multiple sites suggests that Ile 144 contributes to the high fidelity of cleavage by SPase enzymes. In addition, we found that mutation of Ile 86 and Ile 144 to alanine residues changes the specificity of the enzyme, allowing it to cleave after a (-1) Phe residue. Lastly, substitution of Ile 144 and Ile 86 residues by the homologous residues found in the Imp1 protease changes the specificity by promoting cleavage following a (–1) Asn residue.

Taken together, the results demonstrate that Ile 144 and Ile 86 play important roles in substrate specificity and cleavage fidelity.

28 2.2 Results

2.2.1 In vitro analysis of SPase binding site mutants with the wild-type preprotein substrate

To test the importance of amino acids that comprise the S1 and S3 pockets for SPase catalysis, site-directed mutagenesis was used to alter substrate-binding residues predicted to increase or decrease the size of the S1 and S3 pockets. The pocket residues were changed to alanines to increase the pocket size (see Figure 2.1A). Each of the mutant

SPases was purified to homogeneity and then assayed using pro-OmpA-nuclease A substrate (PONA) with different dilutions of the SPase. The dilution factor required to achieve 50% processing can be used to estimate the relative activity of the mutant enzyme relative to the wild-type (WT) (see Table 2.1). The mutant with the lowest enzymatic activity was the I86A variant with a 200-fold lower activity relative to the wild-type enzyme. The L95A and M91A mutants were impaired roughly 10- to 20-fold compared to the wild-type enzyme while the V132A mutant exhibited a 100-fold decrease in activity. In contrast, the other mutants had near WT activity. To decrease the size of the S1 and S3 pockets, residues were changed to those indicated in Figure 2.1B.

As can be seen, these mutations that resulted in a smaller predicted pocket size had only a modest effect on the activity. Of these mutants, the F84W SPase was the most affected with an activity 1/10 of that of the WT enzyme. The L95R SPase, which has an arginine introduced into the S1 substrate binding pocket, served as a negative control. Placement

29 of Arg in this location was expected to interfere with the binding and structure of the

enzyme. The L95R SPase had no measurable activity in vitro.

2.2.2 In vitro analysis of SPase cleavage of pre-protein substrate mutants

As a first step to examine the substrate specificity of the different SPase mutants we examined the ability of the WT and selected SPase mutants to process PONA substrates with different -1 or -3 residues (see Figure 2.2A). The WT SPase cleaved the WT PONA

(with a -1 and -3 alanine) to generate the mature nuclease A domain. In contrast, the WT enzyme cleaved PONA with phenylalanine at the signal peptide -1 position to produce a product running slower on SDS-PAGE than the mature nuclease A domain.

Alternatively, a PONA substrate with a -3 phenylalanine is cleaved by the WT SPase to produce a cleavage product of a similar size as that observed with the WT substrate. A similar cleavage pattern is observed for the -3 Phe mutant with the other SPase mutants assayed. In contrast, the SPase I144A, I144C and I144S mutants can cleave the -1 Val

PONA to a rapidly migrating OmpA-nuclease A like that observed when the WT SPase processes WT PONA. Only a small amount of cleavage of the -1Val PONA mutant is observed with the double mutant I144A/I86A. With the -3 phenylalanine pro-OmpA- nuclease A variant, all the I144 mutants produce a processed band similar in size to the wild-type. However, almost no processing of the -1 phenylalanine mutant is observed for any of these mutants.

30 We followed the time course of -1 Phe PONA processing using higher enzyme concentrations since very little processing of the PONA was observed for the Ile 144 mutants in lower enzyme concentrations. Indeed, cleavage is observed at these higher

SPase levels for all of the mutants (see Figure 2.2B). Interestingly, a lower molecular weight processed band is observed when cleavage is performed with the WT, I144A and

I144S SPase enzymes. These studies showed that the WT and SPase mutants cleaved the

PONA mutant substrates, but it was not clear where cleavage was occurring.

2.2.3 Mass Spectrometric analysis of signal peptide cleavage

In order to identify the sites of cleavage of the various substrates by the SPase enzymes, mass spectrometry was utilized to measure the masses of the major (major peak (*) in

Table 2.2) and minor products. The mass data are shown in Table 2.2 and Table 2.3 shows the predicted theoretical masses corresponding to cleavage of the PONA substrates at various positions from -11 to +6. By comparing the mass data (Table 2.2) to the theoretical masses (Table 2.3) for processing at various sites, we determined where processing had occurred. For selected samples, Edman degradation was performed on the cleavage products to confirm the processing site. Edman degradation showed that the

WT enzyme cleaved the WT PONA substrate exclusively at the predicted site. In addition, mass spectrometry showed a peak at 16,812 +/- 11 Da (16,812 Da theoretical) was produced by SPase cleavage (Figure 2.3A) indicating cleavage at the predicted site.

In contrast, WT SPase processed the -1 Phe mutant eight residues toward the N-terminus from the wild type cleavage site following the Ala-Leu-Ala sequence. Edman degradation

31 yielded the sequence, GFATVAQ. MALDI TOF analysis determined a mass of 17,637

Da (17,634 Da theoretical) (Figure 2.3B). Processing of the –1 Val substrate by WT

Spase shows the majority of cleavage occurring at the same Ala-Leu-Ala sequence, in

this case producing a peak of mass 17,601 Da (17,586 Da theoretical). For this digest,

other minor cleavage products are observed as well with masses: 17,052, 17,810 and

16,651 corresponding to cleavage following –3 Ala (17,040 Da theoretical), –11 Ala

(17,770 Da theoretical) and +2 Thr (16,640 Da theoretical).

In contrast to the fidelity seen when the WT SPase cleaves the wild-type PONA, the

I144C mutant cleaved the same substrate at a number of sites, as determined by mass

spectrometry. Although the major peak is at the normal cleavage site, significant

cleavage was also observed at other sites (Figure 2.3A).Mass analysis revealed a major

product of 16,819 Da as well as alternate cleavage products of mass 17,028, 16,751,

16,646 and 16,559 Da (compare this with the WT where only a 16,812 Da cleavage

product is observed). Analysis of these data with the theoretical masses of various PONA

cleavage products (Table 2.3) revealed that cleavage occurred after the -3 Ala (17,011 Da

theoretical), +1 Ala (16,741 Da), +2 Thr (16,640 Da), and +3 Ser (16,553 Da). A second

trial digestion confirmed the results from the first trail producing the same cleavage

products (Table 2.2). Multiple cleavage sites occurred as well with the -1 Phe PONA

substrate by the I144C mutant (Table 2.2). The first trial shows the following peaks:

17,075, 16,808, 16,547, and 17,628 Da. Analysis of these masses showed corresponding

cleavage sites after -3 Ala (17,088 Da theoretical), -1 Phe (16,812 Da), +3 Ser (16,553

Da), as well as the upstream -9 Ala (17,634 Da) site seen with the WT enzyme. The

32 second trial showed a different pattern: 16,550, 16,637, and 16,738 Da with major cleavage occurring after +3 Ser (16,553 Da theoretical), +2 Thr (16,640 Da), and +1 Ala

(16,741 Da). Taken together, the results demonstrate that the Ile 144 residue is important for determining the accuracy of SPase cleavage.

Although the I144C mutant looses its fidelity of cleavage, the I144A/I86A mutant

cleaved the WT PONA substrate only at the correct site producing a mass of 16,812 Da

(theoretical 16,812 Da) (Table 2.2). Strikingly, this double mutant also cleaves the -1

Phe PONA at the correct cleavage site position (following the -1 Phe) generating a

16,808 Da product in our first trial and a 16,817 Da product in the second trial, both closely matching the predicted mass of 16,812 Da within the error range of +/- 11 Da.

Due to the fact that the –1 Phe PONA is a poor substrate, some of the uncleaved

preprotein of theoretical mass 19,740 Da is observed in both spectra (Figure 2.3B). This

demonstrates that the specificity of the SPase mutant has been changed.

A different cleavage pattern was observed for the Ile 144 mutants processing the -1 Val

PONA. None of the mutants were capable of efficient cleavage following the -1 Val,

with only I144A, I144C and I144S showing cleavage at this site to a very limited extent.

Cleavage occurred almost exclusively after the +2 Thr, according to mass spectrometry.

Edman degradation was performed and the results showed that the substrate was cleaved

by I144C after the sequence Val-Ala-Thr (after the +2 site); a very weak sequence was

observed: Ser-Thr-X-X-Leu-X-X-Glu-Ala.

33 Similar results were observed for the processing the –3 Phe substrate. Mass analysis confirmed cleavage after the +2 Thr residue for the WT SPase in addition to the I144A,

I144C, I144S and I144A/I86A mutants (see Table 2.2). Limited processing is also observed at the –9 Ala (17,643 Da) site when WT SPase is incubated with this substrate.

Figure 2.4 shows processing of different substrates using serial dilutions of the WT and mutant SPases to determine the efficiency of cleavage of SPase and substrate pairs. As can be seen, the I144A, I144C, and I144S SPase mutants have roughly 1/10th the activity compared to the WT SPase in cleaving the WT substrate (Figure 2.4A). In contrast, the

I144A/I86A mutant has roughly 1/1000 the activity for cleaving the WT substrate relative to the WT SPase. Even greater reduction in activity is seen toward the mutant substrates

(Figure 2.4B). Processing is highly impaired for the WT and mutant SPases in cleaving the -1 Phe substrate. Only a small amount of processing is observed at the highest enzyme level. This indicates that for the WT enzyme, the WT substrate is preferred more than 10,000-fold over the -1 Phe substrate. Also, when cleavage occurs, it is at the alternative upstream site. Likewise, the single Ile 144 mutants have roughly 1000-fold preference for cleaving the WT PONA over the -1 Phe PONA. Although not as pronounced for the other tested SPases, there is a preference for cleavage of the WT substrate over the -1 Phe substrate even for the I144A/I86A SPase. Similarly, cleavage of PONA with a -1 Val or -3 Phe is strongly reduced relative to the wild-type.

Since the I144A/I86A mutant has a change of specificity, it is evident that the Ile 144 and

Ile 86 residues are important for substrate specificity in SPase. It is remarkable that Imp1,

34 which is the only member of the type I SPase family to possess unusual substrate specificity, cleaving after a -1 asparagine, has different residues at the homologous 144 and 86 positions than other members of the family. Instead of an Ile at the homologous

144 position, it has a cysteine. At the homologous 86 position it has either a cysteine or a threonine, not an Ile. Therefore, we constructed SPase mutants incorporating cysteine or threonine at the homologous positions to generate the two double mutants: I144C/I86C and I144C/I86T. These mutants along with WT SPase were assayed for their ability to cleave the WT and –1 Asn PONA in a dilution study as described above (Figure 2.5). In processing the WT PONA, both mutants are clearly impaired exhibiting a 1000-fold reduction with the I144C/I86T mutant being slightly more active (Figure 2.5A). When the activities are compared for the processing of -1 Asn PONA, all three enzymes exhibit very low activity showing processing at only the most concentrated enzyme condition.

However, the activities are now quite similar for the WT and these two mutants. In switching substrates from WT to –1 Asn, the WT SPase activity has been reduced

10,000-fold while the I144C/I86C and I144C/I86T mutants have suffered only around an order of magnitude loss in activity.

Processing of the -1 Asn PONA by the WT and mutant SPase was observed at two discrete sites (Table 2.2). The digest of –1 Asn PONA with WT enzyme yielded a mixture of products: the major peak was 17,067 Da which corresponds to processing upstream after the –3 Ala (17,055 Da theoretical). Surprisingly, the minor peak occurred at 16,825 Da which is in close agreement to the theoretical mass of 16,812 Da for processing following the –1 Asn residue. This is an unanticipated result as processing

35 following –1 Asn has not been previously observed for WT SPase [105, 106]. In the

I144C/I86C and I144C/I86T digests, these same peaks were observed. However, in the case of these mutants, the major peak is the one representing cleavage following –1 Asn

(Figure 2.3C). In addition, cleavage following the –1 Asn as the major product was

observed with all the Ile 144 mutants and the I144A/I86A mutant (Table 2.2).

2.2.4 In vivo analysis of signal peptidase binding site mutants

To investigate how detrimental the SPase mutations would be with the proteins in the

natural membrane environment, we examined processing of pro-OmpA in vivo using the temperature-sensitive IT41 strain [25], modified by the incorporation of the T7 RNA polymerase as described above. IT41(DE3) was grown in M9+glucose media at 30°C to a density of 0.3 OD600 and then shifted to 42°C for 1.5h. The cells were pulsed with

[35S]-methionine for 10 sec and then chased with non-radioactive methionine for the

indicated times. Figure 2.6 shows that the processing of pro-OmpA is dramatically

delayed. However, IT41(DE3) bearing a plasmid encoding WT SPase shows rapid

processing. I86A SPase, which showed 1/1000 the activity as WT enzyme in a detergent

extract with pro-OmpA-nuclease A, was active in vivo. Processing was completely normal at the non-permissive temperature when IT41(DE3) was transformed with a plasmid expressing I86A SPase. No in vivo activity was observed for L95R and active site residue mutant, K145A. Except for the L95R SPase mutant, all of the SPase mutants

with substrate binding pocket mutations showed activity in vivo (see Table 2.4).

36 Complementation of the growth defect of IT41(DE3) at 42°C was assayed for all the

SPase mutants. Most mutants clearly showed complementation in this experiment with equivalent colonies on both 30 and 42°C plates. Other mutants showed a reduced number of colonies on the 42°C plate while still other exhibited poorly reproducible growth.

Throughout this study, the binding pocket mutant, L95R, and active site mutant, K145A, negative controls exhibited no growth at 42 °C in any of the trials conducted. These results indicate that the mutants, though impaired to varying degrees in their in vitro activities, can still complement in vivo in most cases, whereas the negative controls clearly cannot.

2.2.5 Molecular modeling of binding site mutants of signal peptidase binding in complex with signal peptides

To help in the interpretation of the mutations made both within the binding pockets of

SPase and within the recognition residues of the PONA substrate we have modeled the

SPase/PONA complex. The models use the recently solved crystal structure of the complex between signal peptidase and the lipopeptide Arylomycin A2 as the template

(PDB code 1T7D) [49]. As described in the methods section, the model for PONA was built into the active site of SPase using the acyl-enzyme penem inhibitor crystal structure and non-covalently bound lipopeptide inhibitor complex crystal structure as a guide to position the P1 and P3 residues into the S1 and S3 binding pockets. The path taken by the substrate along the surface of SPase and the non-specific binding interactions with SPase are basically the same as those observed in the lipopeptide inhibitor complex structure. 37 The ribbon diagram of WT SPase in complex with the WT PONA signal peptide is shown Figure 1.3. PONA makes parallel β-sheet interactions with the strands that line the binding sites (residues 83-90 and 142-145) (Figure 1.3B). As can be seen from Figure

1.3B and Figure 2.7A, the P1 Ala residue points into the S1 pocket and the P3 Ala points into the shallow S3 pocket. The residues Ile 86 and Ile144 sit just at the boarder line between the S1 and S3 pockets and contribute atoms to both pockets. The change in the volume of the binding pockets can be seen when the residues contributing to the binding site are depicted in van der Waals spheres (Figure 2.7). The ability of the double mutant

I86A/I144A to cleave a PONA substrate with a -1 Phe can be rationalized from the modeling which shows the increased volume within the S1 binding pocket provided by the isoleucine to alanine substitutions allows the phenyl side chain to fit into the pocket without any van der Waals clashes (Figure 2.7C). A similar reasoning based on steric effects can be used to explain the ability of the I86C/I144C to bind and cleave PONA substrates with a -1 Asn (Figure 2.7D). It is not clear from the molecular modeling why the mutant I144C (Figure 2.7B) has the ability to cleave at multiple sites within PONA substrate.

2.3 Discussion

Here we investigated which residues within the substrate binding pocket of SPase are responsible for SPase’s high fidelity and contribute to the substrate specificity of SPase.

Toward this objective, we changed all the candidate S1 and S3 residues within SPase that are believed to bind to the -1 and -3 residues of the precursor substrate. Changes were

38 made that were designed to either increase or decrease the size of the binding pocket.

With the exception of the I86A SPase mutant, the results of increasing the size of the

pocket showed that most of the mutants were still quite capable of cleaving the substrate

with only a few mutants demonstrating a 10 to 20-fold reduction in activity. These

results suggest that there is plasticity in the substrate binding pocket. The mutants that

were designed to decrease the size of the substrate binding pocket again only had modest

effects. This is probably due to the fact that the pockets are already quite small within the

WT SPase and that the enzyme is most likely able to make the necessary structural

adjustments required to bind the mutant substrates. A comparison of the three crystal structures so far available for the E. coli signal peptidase (the apo-enzyme[110] the

covalently bound acyl-enzyme inhibitor complex [111], and the non-covalently bound

peptide inhibitor form [49] show that adjustments in both the mainchains and side chains

surrounding the binding site are observed [49].

Our results showed that mutation of isoleucine 144 to cysteine alters the fidelity of the

enzyme. The I144C SP cleaved not only at the normal processing site, but also at other

sites; after the -3 Ala, -1 Ala, +1 Ala, +2 Thr, and +3 Ser with respect to the normal

cleavage position (Figure 2.8 shows a cleavage site summary). The proximal sequences

including the cleavage sites are: TVA-Q for the -3 Ala; AQA-A for the -1 Ala; QAA-T

for the +1 Ala; AAT-S for the +2 Thr; and ATS-T for +3 Ser where the – indicates

position of cleavage. One possible explanation for this lack of fidelity may be that the cysteine mutation in the binding pocket may lead to sliding of the prebound substrate within the active site region. Sliding of the substrate once pre-bound would lead to a

39 cleavage profile with sites clustered around the canonical site as observed in our experiments. The WT leucine in the substrate binding pocket leads to optimal binding, which prevents sliding, and thereby cleavage occurs at only the normal site. These results support the idea that the isoleucine 144 residue is important for SPase to cleave its substrate accurately at the proper site.

In addition to identifying a residue contributing to the accuracy of SPase cleavage, we identified residues that contribute to the substrate specificity of SPase. As mentioned in the introduction, the substrate specificity determinants of this enzyme are at the -1 and -3 position of the substrate protein. Signal sequences within preproteins contain a -1 Ala,

Gly, Ser, and Thr and a -3 Ala, Val, Ser, Thr, Gly, and Leu [112]. The E. coli SPase has been shown to cleave precursor substrates with a -1 Ala, Gly, Ser, Cys, Pro and, in some cases, Thr [105-107]. In addition, it can cleave substrates with a -3 Gly, Leu, Val, Thr,

Cys, Ser, Ile, Pro, and Asn. In bacteria, Phe is never found at the -1 or -3 position [113].

Out of 334 Gram negative signal peptide sequences in the Swiss-Prot data set, there are 2

Val and one Asn at the -1 position [113]. However, they are believed to be incorrectly assigned signal peptidase cleavage site residues [113]. Strikingly, the substrate specificity of SPase is changed with mutation of the Ile 144 and Ile 86 residues to alanine. Whereas the WT enzyme cannot cleave the pro-OmpA-nuclease A substrate after the -1 Phe, the I86A/I144A mutant cleaved the substrate after the -1 Phe with no other cleavage products observed by mass spectrometry (Figures 2.3B and 2.8). On the other hand, the WT SPase cleaved the -1 Phe substrate at an alternative site eight residues upstream (following the -9 Ala) from the normal cleavage site, following the standard

40 processing rules. This suggests that residues well into the proposed hydrophobic helical

(H) region of these mutant substrates are able to melt into the appropriate β-strand conformation needed to make the intermolecular contacts with the SPase binding site.

Previously, processing at an alternative site has been observed when processing at the normal site was prevented by mutating the -1 residues [105]. The I144C mutant cleaves the -1 Phe PONA substrate at multiple sites, again showing inaccuracy in cleavage of the substrates.

While I144A/I86A SPase can cleave after -1 Phe, it is a very poor enzyme. Its activity is around 1/10,000 when compared to the WT enzyme processing the WT pro-OmpA- nuclease A substrate. A similar large reduction in activity was found with trypsin when the active site Gly 226 was changed to an Ala resulting in increased preference for cleaving substrates with lysines relative to arginines [114]. In addition, the substrate specificity of Granzyme B can be extended at the P4 position (but the activity is lowered dramatically from 10 to 100 fold) by mutating the active site residues I99 and N218 to

Ala residues [115]. Further changes in the SPase protein both in the active site and out of the active site region may be necessary to increase the catalytic potency toward the -1

Phe substrate.

Our results indicate that the –1 and -3 residues within the PONA substrate strongly influence where the WT SPase cleaves the substrate. Whereas the WT SPase cleaves the

–1 Val and –3 Phe substrates downstream within the mature domain at the cleavage sites,

VAT-S and AAT-S, respectively (Figure 2.2A; Table 2.2), it cleaves the –1 Phe PONA

41 substrate eight residues upstream (from the normal cleavage site) within the leader

sequence at ALA-G. For this substrate, WT SPase cannot cleave after the +2 Thr because

the F is an inappropriate -3 residue (FAT-S) thereby preventing cleavage at this site

[112].

A similarly strong influence of the -1/-3 residue is seen with the I144C and I144A/I86A

SPases. The -1 Val PONA is processed by the SPase mutants mainly after the +2 Thr

residue (following the cleavage sequence VAT-S), although I144C processing yielded at

least one other minor peak indicating cleavage after the -1 Val. Cleavage of the –1 Val

PONA by the I144A/I86A mutant occurs at the VAT-S site exclusively over the normal

cleavage position at -1. This is in contrast to the AQF-A cleavage site observed with the –

1 Phe PONA where SPase I144A/I86A cleavage does occur after the –1 position. It

appears that in this case as well, the binding of valine by the S1 pocket is highly

unfavorable causing the shift to the VAT-S site. In addition, the -3 Phe PONA is

processed by the WT, I144C, and I144A/I86A almost exclusively after the +2 Thr with

limited cleavage following -2 Gln. Cleavage does not occur at the WT processing site

because the -3 Phe is inappropriate for the signal peptidase specificity.

The cleavage of all the mutant substrates by WT and the tested SPase mutants is very

poor (Figure 2.4B), even though cleavage is occurring at alternative sites (with the

exception of -1F mutant and I86A/I144A which cleaves at the normal position after -1

Phe) following the SPase -1, -3 processing rule of von Heijne [112]. This shows clearly

there is a positional dependence for the cleavage efficiency by SPase (normal position

42 versus alternate position). However, it is perplexing that the I144C SPase mutant can cleave within the WT substrate at multiple sites quite efficiently. A possible explanation for this result is that the WT substrate first binds to SPase in a normal fashion via the -1 and -3 residue of the substrate and then slides in the active site area allowing cleavage at the alternative sites. The cleavage products observed suggest that the active site of the

enzyme preferentially slides into the mature domain to the next most preferred binding

site, which in the case of I144C SPase results in cleavage following the +2 Thr. This is observed for the other PONA mutants as well except in the case of the –1F substrate

where cleavage following the +2 Thr places the engineered –1F interacting unfavorably

with the P3 SPase binding pocket. In this case, the enzyme preferentially cleaves

following the next residue, +3 Ser.

The results with the I86A/I144A SPase and the -1F PONA substrate show that the Ile 144

and Ile 86 residues of SPase play a central role in the process of accurately cleaving the

preprotein substrate: change in the Ile 144 and Ile 86 residues effects a change in –1,-3 substrate recognition. In light of this result, we observe an interesting correlation with the

only SPase that exhibits a significantly altered substrate recognition sequence, Imp1

mitochondrial protease. Imp1 protease lacks the conserved 144 and 86 residues. Ile 144

is a cysteine and Ile 86 is either a cysteine or threonine residue. The result is an enzyme

which cleaves in vivo after a -1 Asn cleavage site which is unique among SPases [108,

116]. The SPase mutants constructed to resemble the Imp1 enzyme: I144C/I86C and

I144C/I86T, yielded interesting results in the digestion of the -1 Asn PONA substrate.

This analysis shows that both enzymes in addition to WT SPase produce the same two

43 cleavage products. Processing was observed following the -3 Ala and -1 Asn residues.

The proportions, however, are different with the WT enzyme selecting the -3 Ala site

preferentially while both Imp1-type double mutants and other Ile 144 mutants (data not

shown) demonstrated a preference for cleavage following -1 Asn. WT SPase cleaves WT

PONA 10,000 times better than the -1 Asn PONA. In contrast, the SPase Ile 144/Ile 86

mutants mimicking the Imp1 binding residues cleave WT PONA only 100 times better.

Taken together, these observations strengthen the body of evidence that these residues, in

particular, hold a special importance in determining the –1,-3 substrate recognition

sequence.

Future studies will have to examine the substrate specificity of the SPase with the Imp1

Ile 144/Ile 86 residues. Chen et al. [109] has shown that, in addition to cleaving its endogenous substrate, i-cyt b2, with a -1 Asn, Imp1 can process substrates with a -1 Ser,

Cys, Met, Ala, and Leu. In addition, we will analyze the SPase S1 and S3 mutants utilizing a library of peptide substrates with random sequence surrounding the OmpA cleavage site. These experiments hold the promise of identifying differences in the preferred cleavage sequence for each of the mutants not possible using designed substrate sequences alone. These future data could help in clarifying the effects on substrate

selection and fidelity observed in the current studies. In addition, crystallographic and

spectroscopic analysis of these binding site mutants and substrate mutants in complex will help in the understanding of this interesting molecular recognition event that most likely occurs at or within the membrane surface.

44 2.4 Materials and Methods

2.4.1 Bacterials Strains and Plasmids

E. coli strains BL21(DE3), BLR(DE3), and DH5α were from our collection. E. coli

IT41, a temperature-sensitive SPase I strain, was obtained from Dr. Yoshikazu Nakamura

[25]. The IT41 cell line was modified by incorporation of the T7 RNA polymerase gene using λDE3 prophage supplied in the λDE3 Lysogenization Kit (Novagen) to produce

IT41(DE3). Lysogens successfully expressing the T7 RNA polymerase were verified using the included T7 tester phage in addition to verifying TS growth. The pET 23b vector, which contains IPTG inducible SPase with a 6-His tag, was described previously

[117]. The pONF1 vector overexpressing pro-OmpA-nuclease A was described in [118].

This nuclease A substrate has a six-histidine sequence at codons 2 through 7 [119].

2.4.2 DNA methods

The Quikchange mutagenesis method (Stratagene Inc) was used to make oligonucleotide-

directed mutations within the SPase and pro-OmpA-nuclease A proteins. Mutations were

verified by gene sequencing. The calcium chloride and electroporation methods were

used for DNA transformation into E. coli strain, DH5α.

45 2.4.3 Purification of signal peptidase and pro-OmpA-nuclease A

BL21(DE3) cells containing 6-His tagged SPase encoded by the pET23b vector were used for overexpression of SPase. SPase was expressed and purified using ion exchange and nickel affinity chromatography, as described by [117]. Pro-OmpA-nuclease A containing a 6-His tag was purified over a Ni2+-nitrilotriacetic acid-agarose (NTA)

(Qiagen, Inc) column, as described in [120] with modifications to utilize denaturing

conditions. Following ammonium sulfate precipitation, proteins were dissolved in 6M

guanidine buffer and loaded on 1ml bed volume of Ni-NTA resin. Resin was washed with 8M urea buffer (pH 8.0) to remove guanidine and elution of bound protein was accomplished using 8M urea buffers of reducing pH (pH 6.3, 5.9, 4.5). Following elution, proteins were assayed for purity using 17% SDS-PAGE and selected fractions were refolded slowly by dialyzing against increasing volumes of 25mM HEPES (pH 7.5) with 1% Triton X-100 utilizing a total of 4 liters buffer.

2.4.4 In vitro and in vivo assay of Signal Peptidase Cleavage

The in vitro activity of wild-type and mutant SPases was determined using various pro-

OmpA-nuclease A substrates. Cleavage of the substrate was performed at different dilutions of SPase, as described [117]. To determine the activity of SPase in vivo, we

used the temperature-sensitive IT41(DE3) strain. The IT41(DE3) strain was modified

from the original strain by introducing the T7 RNA polymerase gene allowing expression

of the pET 23b encoded SPases. The procedure for measuring in vivo activity of the

46 SPase mutants is the same as that described by [117]: Single colony transformants of

IT41(DE3) were transferred into M9+glucose media and grown to a density of

approximately 0.3 OD600. The cells were then shifted to 42°C and incubated for an additional 1.5h. The cells were pulsed with [35S]-methionine for 10 sec and then chased

with non-radioactive methionine for the indicated times.

2.4.5 Study to assay ability of mutants to complement the growth defect of the

temperature-sensitive SPase strain, IT41(DE3)

The processing of pro-OmpA was assayed in the temperature-sensitive SPase strain,

IT41(DE3). All media used to culture IT41(DE3) was prepared with a low salt

concentration of 2.5g NaCl/liter (LS2.5 media) and 30µg/ml tetracycline. Competent cells of IT41(DE3) were prepared using the CaCl2 method and transformed with a plasmid

expressing the SPase mutant to be studied. The transformed IT41(DE3) cells were plated

on solid LS2.5 media with 100µg/ml Amp and grown at 30°C until colonies reached 1-

2mm in size. Single transformants were selected and resuspended in fresh media and streaked in duplicate on fresh plates for incubation at 30 and 42°C to assay for complementation of growth at the non-permissive temperature.

2.4.6 Automated Edman Degradation

For N-terminal protein sequence analysis, enzyme reaction mixtures were separated by

SDS-PAGE then transferred electrophoretically to PVDF membranes. Transferred

47 proteins were visualized on the PVDF blots by staining with Coomassie Blue then

excised for sequence analysis. N-terminal Edman degradations were performed using an

Applied Biosystems Model 492 automated protein sequencer.

2.4.7 Matrix Assisted Laser Desorption Ionization Time-of-Flight (MALDI-TOF)

Mass spectrometry

SPase cleavage of pro-OmpA-nuclease A was also analyzed using mass spectrometry.

Small aliquots of reaction mixtures (10-20µL) were precipitated by addition of 10

volumes of ice-cold acetone. After standing on ice for 10 minutes, the acetone

precipitated proteins were sedimented by centrifugation in an Eppendorf centrifuge at

13,000 rpm for 10 minutes. The supernatant was carefully removed by aspiration and the

protein pellets (usually not visible) were redissolved by addition of 5µL 10% formic acid.

The dissolved proteins were mixed with the matrix solution: 5µL saturated solution of α-

cyano-4-hydroxycinnamic acid (10mg dissolved in 500µL 0.1% trifluoroacetic acid and

500µL CH3CN). The stainless steel MALDI-TOF target plate was spotted with 1 µL of

each reaction sample and analyzed using a Bruker Daltonics Autoflex mass spectrometer

in the linear mode. The instrument was calibrated with a mixture of protein standards

including insulin (5,734.6 Da); cytochrome C (12,361.1 Da), and myoglobin (16,952.6

Da). The mass accuracy in the 17 kDa mass range was +/- 11 Da.

48 2.4.8 Prediction of theoretical mass for cleavage products

To predict the size of postulated cleavage products to match with the MS data, various portions of the sequence of the mature domain of pro-OmpA-nuclease A were entered into the PeptideMass tool on the ExPASy Proteomics Server

(http://us.expasy.org/tools/peptide-mass.html). Singly charged average masses were calculated without enzymatic cutting to obtain the total mass of the predicted protein products of cleavage by signal peptidase.

2.4.9 Molecular Modeling

Models of each of the SPase mutants were made using the atomic coordinates from the recent crystal structure of signal peptidase in complex with a lipopeptide inhibitor (PDB

code 1T7D) as the template [49]. The mutations were made with the program XtalView

[121]. The model for the pro-OmpA-nuclease A (PONA) substrate was built into the

active site of SPase using the acyl-enzyme penem inhibitor [111] and non-covalently

bound lipopeptide inhibitor complexes [49] as a guide to position the P1 and P3 residues

into the S1 and S3 binding pockets. The path taken by the signal peptide along the

surface of SPase and the parallel β-sheet type hydrogen bonding interactions with SPase

are basically the same as those observed in the lipopeptide inhibitor complex structure.

The models of the SPase/PONA complexes were energy minimized using the program

CNS [122]. Figures were prepared using the program PyMol.

49

Figure 2.1 Effect of mutations that are predicted to increase or decrease the size of the binding pocket on SPase activity. A. Mutations that are predicted to increase the size of the binding pocket. B. Mutations that are predicted to decrease the size of the binding pocket. Wild-type pro-OmpA-nuclease A (PONA) was digested by serial dilutions of stock wild-type and mutant forms of signal peptidase in order to study the effect of changing the size of the substrate binding pockets. The 0 dilution lane in the upper gel is a control in which no enzyme has been added and indicates the position of the unprocessed form of PONA. Processing of PONA was initiated by the addition of 1µl of signal peptidase stock (0.1mg/ml) or corresponding serial dilution of the stock to 10µl PONA reaction mixture (0.04mg/ml in 50 mM Tris, 10mM CaCl2, 1% Triton X-100, pH 8.0). Reactions were incubated for 1 hour at 37°C and then run on 17% SDS-PAGE. “P” indicates the position of the uncleaved preprotein and “M” indicates the position of the mature protein.

(Continued)

50

Figure 2.1 (Continued)

A. WT P M

Dilution 0 1 101 102 103 104

P V132A M

P I86A M

P M91A M

P L95A M

P I144A M

P F84A M

Dilution 1 101 102 103 104

(Continued)

51 Figure 2.1 (Continued)

B.

P WT M

Dilution 0 1 101 102 103 104

P D142I M

P D142E M

F84W P M

P G89A M

P L95R M

P V132I M

P I144C M

Dilution 1 101 102 103 104

52

hours of incubation WT –1V –1F –3F 0 1 6 12 24 Std

P P WT M M

P P I144A M M

P P I144C M M

P P I144S M M

P I144A P M I86A M

Figure 2.2 Processing of pro-OmpA-nuclease A (PONA) with alternate -1 and -3 residues. WT and three PONA mutants, -1 Val, -1 Phe and -3 Phe, were assayed for processing by the indicated SPase mutants. A. To demonstrate processing of PONA mutants, an overnight incubation was conducted. SPase (WT or mutant) 0.5µg was added to the PONA reaction mixture (0.04mg/ml in 50 mM Tris, 10mM CaCl2, 1% Triton X-100, pH 8.0) and incubated at 37°C for 18hrs. Samples were run on 17% SDS-PAGE and detected with silver staining. B. In order to demonstrate the development of processed bands over extended incubations, reactions were set up as in Figure 2.1 and incubated at 37°C. At the designated time point, a portion was removed and the reaction stopped by the addition of denaturing loading buffer. All time points were then analyzed by 17% SDS-PAGE followed by silver staining, “P” indicates the position of the uncleaved preprotein and “M” indicates the position of the cleaved protein band.

53

Figure 2.3 Mass analysis demonstrates altered cleavage profile for SPase mutants. Selected mass spectra are paired to illustrate alteration in PONA processing in the various SPase mutants. A. Processing of WT PONA by WT SPase yields a single peak corresponding to the WT cleavage product (17,023 is the peak due to the matrix adduct). In the lower panel, digestion by I144C SPase shows multiple peaks demonstrating a loss of fidelity. B. In the upper panel, processing of the -1 Phe PONA by the WT enzyme yields peaks which are heavier than the product corresponding to cleavage at the -1 site indicating selection of alternate processing sites. In contrast, the I144A/I86A mutant shows cleavage following the -1F residue (lower panel). C. Mass spectra data from digestion of -1 Asn PONA by the WT and I144C/I86T SPase are shown. Processing is observed following the -1 Asn and -3 Ala residues in both cases. In the lower panel, the shift in relative intensity of the peaks indicates a change in preference for the -1 Asn cleavage site for the I144C/I86T mutant.

(Continued)

54

Figure 2.3 (Continued)

A B 25000 2500 16,812 17,637 20000 2000

1500 15000 1000 10000 Intensity Intensity matrix adduct 500 17,081 5000 19,744 17,023 (matrix adduct) 0 0

16,819 12000 16,808 3000 10000 8000 16,646 2000 16,751 6000 19,753

Intensity

Intensity 4000 1000 16,559 matrix adduct 2000 matrix adduct 0 0 -2000 15000 16000 17000 18000 19000 20000 21000 15000 16000 17000 18000 19000 20000 21000 m/z m/z C 8000 17,067 6000 16,825

4000

Intensity 2000 matrix adduct

0

16,827

4000

3000

2000 Intensity 17,045 1000 0

15000 16000 17000 18000 19000 20000 21000 m/z

55

A.

0 1 10 100 1000 10000 Dilution Factor SP P WT M

P I144A M

P I144C M

P I144S M

I144A P I86A M

B. 6A 8 /I 4A A 44C none WT I14 I1 I144S I144

-1F PONA P M

-1V PONA P M

-3F PONA P M

Figure 2.4 Activity of SPase mutants with the WT and mutant PONA substrates. The dilution series experiment was repeated as described in Figure 2.1 utilizing the - 1Val, -1 Phe and -3 Phe PONA mutants in addition to WT. A. For the processing of WT PONA, all dilution lanes are shown while in B. only the “1” or undiluted lane is shown for the mutants due to their low observed activity. “P” indicates the position of the uncleaved preprotein and “M” indicates the position of the mature protein. 56

A.

1 101 102 103 104 Dilution Factor

P WT PONA M WT

P M I144C/I86C

P M I144C/I86T

B. 1 101 102 103 104 Dilution Factor

P WT -1N PONA M

P M I144C/I86C

P M I144C/I86T

Figure 2.5 Processing of WT and -1 Asn PONA substrates by WT, I144C/I86C and I144C/I86T mutants. The dilution series experiment was conducted as in Figure 2.1 using the WT (A.) and -1 Asn (B.) PONA substrates. “P” indicates the position of the uncleaved preprotein and “M” indicates the position of the mature protein.

57

0 10 30 50 70 90 120s -WT Lep P M

P +WT Lep M

P I86A M

L95R P M

P K145A M

Figure 2.6 In vivo complementation of OmpA processing defect in IT41(DE3) at 42°C. SPase mutants were studied for their ability to complement for the temperature sensitive SPase defect of IT41(DE3) at 42°C. Only fresh transformants of SPase and SPase mutants into IT41(DE3) were used. Colonies incubated at 30°C (See Methods) were transferred to 2ml of M9 +glucose media and incubated at 30°C until a density of 0.3 OD600 was reached. The cultures were then shifted to a 42°C incubator and incubated for an additional 1.5hrs. Cell cultures were labeled with 35S-methionine for 10 seconds at which time the “0” time point was taken. After the “0” time point was taken, the chase was begun with the addition of excess non-radioactive methionine with additional time points taken at indicated times. At each time point, 250µL of culture was quenched by addition to 250µL of ice cold 20% TCA. These were then subjected to immunoprecipitation with OmpA antiserum (See Methods) and run on 15% SDS-PAGE. “P” indicates the position of the uncleaved pro-OmpA protein and “M” indicates the position of mature OmpA.

58

Figure 2.7 Molecular modeling of binding site mutants with in E. coli signal peptidase. The amino acid residues of signal peptidase that contribute to the binding pocket surface are shown in van der Waals spheres. The side chain atoms from those residues that contribute to the S1 binding pocket (Pro 87, Ser 88, Gly 89, Ser 90, Met 91, Leu 95, Tyr 143, Lys 145) are colored green. The side chain atoms from those residues that contribute to the S3 binding pocket (Phe 84, Ile 101, Val 132, Asp 142) are colored cyan. The side chain atoms from those residues that contribute to both the S1 and the S3 binding pockets (Ile 86 and Ile 144) are colored hot pink. The main chain atoms are colored white. The P3’ to P5 residues of the pro-OmpA-nuclease A (PONA) substrate are shown in yellow (carbon) blue (nitrogen) and red (oxygen). A. The wild-type signal peptidase with wild-type pro-OmpA-nuclease A. B. The I144C mutant of signal peptidase. C. The I86A/I144A mutant of signal peptidase with the bound P1 Ala to P1 Phe mutant of pro-OmpA-nuclease A. D. The I86C, I144C mutant of signal peptidase.

(Continued)

59

Figure 2.7 (Continued)

60

WT PONA ALAGFATVAQA-ATSTKK WT SP

ALAGFATVAQA-ATSTKK I144C SP

ALAGFATVAQA-ATSTKK I144A/I86A SP

-1F PONA ALAGFATVAQF-ATSTKK WT SP

ALAGFATVAQF-ATSTKK I144C SP

ALAGFATVAQF-ATSTKK I144A/I86A SP

-1V PONA ALAGFATVAQV-ATSTKK WT SP

ALAGFATVAQV-ATSTKK I144C SP

ALAGFATVAQV-ATSTKK I144A/I86A SP

-3F PONA ALAGFATVFQA-ATSTKK WT SP

ALAGFATVFQA-ATSTKK I144C SP

ALAGFATVFQA-ATSTKK I144A/I86A SP

Figure 2.8 The cleavage site of signal peptide mutants by various signal peptidase mutants. The peptide sequence of the WT PONA and each PONA mutant construct is listed in addition to the processing sites observed for each indicated SPase enzyme. Arrows indicate the site of processing and the relative size of the arrows indicate major and minor peaks.

61

Mutant enzyme Proposed pocket Activity relative to 100% WT

F84A S3 50 F84W S3 10 I86A S1/S3 0.5 G89A S1 100 M91A S1 5 L95A S1 10 L95R S1 0 I130A S1 100 V132A S3 1 V132I S3 100 D142E S3 10 D142I S3 100 Y143A S1/S3 10 Y143W S1/S3 100 I144A S1/S3 50 I144C S1/S3 50 I144S S1/S3 10 I144A/I86A S1/S3 0.1

Table 2.1 Activity of mutant SPases relative to WT SPase Activities of mutant SPases were determined from dilution studies shown in Figure 2.1. Dilution of WT and mutant necessary to achieve 50% processing was determined and comparison yielded the relative activities.

62

Table 2.2 Mass spectral data of PONA cleavage products. Mass data for the major and minor peaks determined by MS is tabulated above. The first entry in the column indicates the major peak (*), followed by any other minor peaks observed. Predicted masses for theoretical cleavage are indicated in parentheses.

(Continued)

63

Table 2.2 (Continued)

WT PONA WT SP 1 *16812 (16812) 2 *16815 (16812) I144C SP 1 *16819 (16812), 16646 (16640), 16751 (16741), 16559 (16553), 17028 (17012) 2 *16823 (16812), 16651 (16640), 16754 (16741) 16565 (16553), 17031 (17012) I144A/I86A SP *16812 (16812) I144C/I86C SP *16819 (16812) I144C/I86T SP *16822 (16812), 17033 (17012) -1F PONA WT SP 1 *17637 (17634), 17081 (17088) 2 *17639 (17634), 17085 (17088) I144C SP 1 *17075 (17088), 16808 (16812), 16547 (16553), 17628 (17634) 2 *16550 (16553), 16637 (16640), 16738 (16741) I144A/I86A SP 1 *16808 (16812) 2 *16817 (16812) -1V PONA WT SP *17601 (17586), 17052 (17040), 17810 (17770), 16651 (16640) I144A SP *16648 (16640), 16852 (16812) I144C SP *16645 (16640), 16818 (16812) I144S SP *16651 (16640), 16837 (16812) I144A/I86A SP *16646 (16640) -1N PONA WT SP *17067 (17055), 16825 (16812) I144A SP *16828 (16812), 17049 (17055) I144C SP *16823 (16812) I144S SP *16826 (16812) I144A/I86A SP *16825 (16812) I144C/I86C SP *16825 (16812), 17067 (17055) I144C/I86T SP *16827 (16812), 17045 (17055) -3F PONA

WT SP *16652 (16640), 17643 (17634)

I144A SP *16647 (16640)

I144C SP *16653 (16640)

*16645 (16640) I144S SP I144A/I86A SP *16658 (16640), 16864 (16883)

64

AA Sequence Name WT -1V -1F -1N -3F M (H)6 K K 19,663.6 19,691.7 19739.7 19706.6 19739.7 T A I A I A V A -11 17,813.417,841.5 17889.5 17856.4 17,889.5 L -10 17,742.417,770.4 17818.4 17785.4 17,818.5 A -9 17,629.217,657.2 17705.3 17672.2 17,705.3 G -8 17,558.117,586.2 17634.2 17601.1 17,634.2 F -7 17,501.117,529.1 17577.2 17544.1 17,577.2 A -6 17,353.917,381.9 17430.0 17396.9 17,430.0 T -5 17,282.817,310.9 17358.9 17325.8 17,358.9 V -4 17,181.717,209.8 17257.8 17224.7 17,257.8 A (F) -3 17,082.6 17,110.6 17158.7 17125.6 17,158.7 Q -2 17,011.517,039.5 17087.6 17054.5 17,011.5 A (V, F, or N) -1 16,883.4 16,911.4 16959.5 16926.4 16,883.4 A +1 16,812.3 16,812.3 16,812.3 16,812.3 16,812.3 T +2 16,741.2 16,741.2 16,741.2 16,741.2 16,741.2 S +3 16,640.1 16,640.1 16,640.1 16,640.1 16,640.1 T +4 16,553.0 16,553.0 16,553.0 16,553.0 16,553.0 K +5 16,451.9 16,451.9 16,451.9 16,451.9 16,451.9 K +6 16,323.7 16,323.7 16,323.7 16,323.7 16,323.7 -PONA 16,195.6 16,195.6 16,195.6 16,195.6 16,195.6

Table 2.3 Predicted theoretical masses of various PONA cleavage products. The theoretical masses of possible PONA cleavage products were predicted using the PeptideMass tool on the ExPASy Proteomics Server. Average masses are reported as [M + H]+.

65

Lep Mutant Time point of 50% processing No plasmid control 120s WT (+ Control) <10s I144A/I86A 10s L95R 50s K145A (- Control) 70s

The following mutants exhibited 50% processing in <10s like WT SP: F84A, F84W, G89A, I86A, M91A, M91I, L95A, V103I, I130A, V132A, V132I, D142E, D142I, Y143A and I144A.

Table 2.4 In vivo processing of pro-OmpA in the temperature sensitive stain IT41(DE3) at the nonpermissive temperature 42°C. This table indicates the amount chase time required to observe 50% processing for the WT and mutant SPases. Times are indicated in seconds of chase.

66

CHAPTER 3

ANALYSIS OF CLEAVAGE FIDELITY AND SUBSTRATE SPECIFICITY OF

SIGNAL PEPTIDASE S1/S3 MUTANTS WITH Β-LACTAMASE SUBSTRATE IN

VIVO

3.1 Introduction

In the previous studies, processing of the pro-OmpA-nuclease A preprotein substrate revealed intriguing results confirming a change in specificity of select SPase I mutants.

These studies, however, required the labor-intensive purification of multiple mutant substrates followed by SPase I cleavage in vitro and final characterization of cleavage

products by mass spectrometry. Utilizing this process, I was able to identify new

cleavage sequences that could bind and be processed by the mutant SPases to a greater

degree than the WT enzyme.

In the effort to expand these results to other substrates, I chose to work with the TEM-1

β-lactamase enzyme. TEM-1 is one of an extended family of β-lactamase enzymes that

are responsible for bacterial resistance to β-lactam antibiotics. TEM-1 was originally 67 isolated from E. coli [123] and inactivates β-lactam antibiotics such as ampicillin by

hydrolysis of the β-lactam ring.

β-lactamase is a protein that is exported to the periplasm in E. coli and, as such, possesses

a cleavable signal peptide. Under normal conditions, export of the enzyme begins with

the insertion of the signal peptide into the bacterial inner membrane. After translocation,

the signal peptide is cleaved by SPase I and the active mature domain of β-lactamase is

released into the periplasm and then ultimately into the extracellular medium.

As a substrate, β-lactamase offers several advantages over pro-OmpA-nuclease A.

Sequence analysis of the cleavage region reveals a single SPase I processing site (Figure

3.1). Due to the fact that the SPase I cleavage region within the β-lactamase signal peptide has no alternative cleavage sites, when the canonical site is altered by mutagenesis, SPase I can no longer cleave off the signal peptide. This can be seen clearly in my experiments when the acceptable processing site is changed such that the WT

SPase cannot cleave, no processing is seen at the canonical or any other alternate site even after extended chase times (Figure 3.2C, D). In the absence of signal peptide cleavage, the catalytically-active domain of β-lactamase remains in the periplasm tethered to the bacterial inner membrane.

Another advantage of β-lactamase as a substrate involves its ability to inactivate β-lactam antibiotics. In fact, it is possible to use the processing of β-lactamase by SPase as a selection technique based on the dramatic difference in ampicillin tolerance that is

68 observed when the mature domain is located in the periplasm verses the extracellular environment.

While it is common in selective media to use concentrations of ampicillin in the range of

50-100µg/ml, cells possessing WT β-lactamase can survive ampicillin levels as high as

1mg/ml [124]. In contrast, when processing is prevented and β-lactamase remains in the periplasm, the amount of ampicillin that can be tolerated by these cells is dramatically reduced, with cells unable to tolerate much more than 100µg/ml (data not shown). My experiments show that this can be used as an effective selection technique for SPase cleavage using an ampicillin concentration of 500µg/ml. Under conditions where β- lactamase cannot be cleaved from the membrane, the cells can survive at low concentrations of ampicillin such as 50µg/ml; however at high ampicillin concentrations, they cannot. Only when the signal peptide of β-lactamase is cleaved can the cells survive concentrations at or above 500µg/ml.

3.2 Results: In vivo growth assays

3.2.1 WT β-lactamase

As an initial method to assess the activity of SPase mutants toward various β-lactamase mutants in vivo, a plate-based assay was developed which employed two different ampicillin concentrations to differentiate unprocessed β-lactamase from that which was fully processed by the SPase in question. Growth experiments that I have done indicate

69 that cells producing fully processed β-lactamase can survive at concentrations of

ampicillin as high as 1mg/ml (data not shown). Other experiments demonstrated that

when β-lactamase is incapable of being processed (Figure 3.2B, C) the cells can survive

only low concentrations of 50-100µg/ml but cannot grow at 500µg/ml.

Based on this observation, a system was developed in which two plasmids were

employed to express separately β-lactamase and SPase. Since each protein is expressed from a different plasmid, various combinations of mutants of β-lactamase and SPase can be easily assayed. For this study, β-lactamase was expressed from the pING plasmid which was generated by removing the SPase gene from the pRD8 plasmid which has been used in previous studies to express SPase [23]. With the SPase gene removed, the pING plasmid was subjected to site-directed mutagenesis to generate different mutations at the -1 position within the signal peptide of β-lactamase. WT SPase and selected mutants were sub-cloned from the pET23b plasmid using the SalI and SmaI sites into the pGZ119HE [125]. The pGZ119HE plasmid possesses the ColD origin of replication and as such can be cotransformed with pING that has a compatible ColE1 origin.

In the studies that follow, a temperature sensitive SPase strain, IT41, was used. The IT41 strain does not grow at 42ºC but does grow, albeit slowly, at 30ºC. Even at the permissive temperature of 30ºC the activity of signal peptidase is strongly reduced, however this

level of activity is sufficient to allow the cell to grow.

70 IT41 cells were cotransformed with the plasmids expressing the SPase and β-lactamase mutants in question and cultured on media selective for both plasmids. Single colonies were isolated and streaked out on plates containing either 50 or 500µg/ml ampicillin in duplicate for incubation a both 30 and 42ºC. The controls for these studies involved cotransforming the WT β-lactamase along with either the WT SPase or the K145A mutant. As cited previously in this work, K145 is an active site residue of SPase and when mutated to an alanine, abolishes all activity [15, 52].

Upon incubation at 42ºC (data not shown), it was observed that none of the cells could grow, except for the cells transformed with WT SPase and WT β-lactamase, even at the lower ampicillin concentration. Cells transformed with the K145A SPase and WT β- lactamase showed no growth at 42ºC although in some rare cases minor growth was observed, most likely due to reversions.

We performed our genetic selection at 30ºC (permissive) temperature where the chromosomally-encoded SPase has strongly reduced activity. At this temperature, we observed the desired pattern of selection (Figure 3.2). When grown under these conditions, the WT SPase control can grow normally under either ampicillin condition

(Figure 3.2A, C). The K145A SPase yielded slightly different results. Instead of a total lack of growth, the K145A SPase with the WT β-lactamase consistently showed a low background level of slowly growing cells that behaved the same at either 50 or 500µg/ml ampicillin (Figure 3.2A, B). The slow growth at both ampicillin levels is due to the low

71 levels of signal peptidase encoded by the chromosome of IT41. Sufficient amounts of

mature β-lactamase are generated to allow growth of the strain at 500µg/ml ampicillin.

3.2.2 -1V β-lactamase

The results for the -1V β-lactamase appear in Figure 3.2A. The –1V substrate was the

most easily tolerated substrate by SPase and the mutants (Figure 3.2A). This was observed by the presence of growth in the cases where the 500µg/ml concentration was used. In these plates all SPase mutants were able to grow to some degree in the presence of 500µg/ml ampicillin. Upon comparison of the 50µg/ml and 500µg/ml concentrations, it can be seen that some of the mutants show less reduction of growth at the higher concentration. While the WT, I144A and I144C/I86T SPases all appear to show a dramatic reduction in growth at 500µg/ml ampicillin, the I144C and I144A/I86A mutants exhibit only minor reduction in growth under the more selective conditions. This result agrees with the pro-OmpA-nuclease A studies where the I144C SPase was shown to have reduced fidelity in substrate selection in addition to the ability to cleave after a -1V

residue in the pro-OmpA-nuclease A substrate. The ability of the I144A/I86A SPase to

support such growth at high ampicillin concentrations agrees the PONA results which

demonstrated the ability to accommodate larger -1 residues in the S1 pocket. Finally,

when the K145A mutant was assayed with the -1V β-lactamase, there was minor growth

at 50µg/ml but no growth at all at the higher ampicillin concentration.

72 3.2.3 -1F β-lactamase

As predicted by the in vitro PONA experiments, the growth of IT41 containing the the signal peptidase mutants with the –1F β-lactamase substrate was greatly impaired at high ampicillin concentrations (Figure 3.2B). While IT41 containing all the SPase mutants assayed could grow normally at the lower ampicillin concentration, none of the mutants could grow at the higher concentration. The control of K145A SPase with the WT β- lactamase exhibited low-level growth under both ampicillin concentrations. As mentioned in Section 3.2.1, this result for the strain containing the K145A mutant most likely is due to the background, although poor, processing by the chromosomally- encoded signal peptidase in IT41. The slow growth at 30ºC in IT41 is due to the fact that

SPase processing is less than optimal for cell growth. This is reasonable since it is also observed that when SPase processing at this site is prevented by the incorporation of the

–1F residue, no colonies are observed from SPase processing or from processing by any other protease.

3.2.4 -1N β-lactamase

Except for the I144A/I86A mutant, cells bearing the –1N β-lactamase substrate yielded similar results as the -1F substrate when assayed in the plate-based experiment (Figure

3.2C). Only the IT41 expressing the I144C and I144A/I86A SPase showed growth of any kind at 500µg/ml ampicillin. A few colonies were observed for the I144C SPase mutant that most likely represents reversion of the strain.

73 3.3 Discussion

The plate assay for β-lactamase processing yielded some interesting results. Primarily,

there was the fact that only the WT β-lactamase/WT SPase combination could grow at the non-permissive temperature of 42ºC and that the desired selection was observed at the permissive temperature of 30ºC. This effect could be attributed to the idiosyncrasies of the IT41 strain. This strain is very challenged in its growth and even at the permissive

temperature, requires extended incubation to form sizable colonies. IT41 relies on a read-

through effect of an amber mutation inserted into the SPase I gene for its ability to grow

at 30ºC. At the permissive temperature, read-through can occur but at the non-permissive

temperature, read-through is largely inhibited resulting in the temperature-sensitive

phenotype. In practice, this method of control lacks optimum stringency and often results

in reversion to a non-temperature sensitive phenotype. In addition, the slow growth

phenotype that I observe suggest that even at the permissive temperature, the SPase

produced is not fully functional or is not produced at WT levels. It is likely that at the

highly restrictive non-permissive temperature, only the cells with the greatest β-lactamase

processing ability can survive long enough to produce visible growth.

As controls the WT and K145A SPase were effective in demonstrating that the system

was working properly and that the growth observed represented selection for β-lactamase

processing. Under all conditions of temperature and ampicillin concentrations assayed,

the WT SPase could process WT β-lactamase adequately to allow abundant growth. In

contrast, the K145A mutant with the WT β-lactamase exhibited very limited growth. The

74 slow growth at the permissive temperature of IT41 containing the K145A SPase mutant is occurring by the residual low activity of the signal peptidase encoded by the chromosome. This is supported by the observation that this growth is not observed when the IT41 strain expressing K145A contains β-lactamase with a non-canonical –1 residue such as in the -1V β-lactamase, or when WT or mutant SPases were matched with the -1F

β-lactamase.

The –1V substrate is chemically more similar to the WT substrate containing an alanine residue than the other two mutant substrates assayed. The valine residue substitution was chosen to represent an incremental change in size and chemical nature from the alanine residue in the WT sequence yet remaining more conservative than the other two phenylalanine and asparagine substitutions chosen. As expected, valine substitution was tolerated more easily by both the WT and mutant SPases assayed. In the in vitro studies with PONA as a substrate, both the I144C and I144A mutants were able to process the -

1V PONA to some extent and here as well, these mutants do permit survival at 500µg/ml.

Also able to survive at 500µg/ml were the double mutants I144A/I86A and I144C/I86T.

The ability of the I144C/I86T SPase to cleave after -1V in β-lactamase is not too surprising as it was observed to be relatively active toward the WT substrate in the

PONA study in addition to being able to cleave after another larger residue, -1N PONA.

The real surprise here was the abundance of growth of IT41 with the I144A/I86A SPase mutant. With all substrates studied, this enzyme has consistently had poor activity. In the study with –1V PONA, this double mutant could not produce any cleavage product after

75 one hour of incubation (Figure 2.4B). In this same data set, the WT SPase also produced no cleavage product. In this study, however, the I144A/I86A mutant allowed abundant growth of IT41 containing the -1V β-lactamase while the WT SPase allowed few colonies. This result is somewhat puzzling: The I144A/I86A substrate clearly had the worst activity in the in vitro studies and was thus expected to perform similar to the

K145A negative control in this study. One explanation for this result is that the IT41 cells

used for this transformation have reverted to phenotype resembling the WT SPase strain.

This explanation is unlikely because the cells used for this study were from the same

stock as the other transformations that appear on this plate which did behave as expected.

Also, if the reversion had occurred, the results would more closely resemble those of the

WT SPase processing the -1V β-lactamase.

The -1F substitution is a large deviation from the WT alanine and as shown in my previous experiments cannot be recognized and cleaved by the WT SPase. The resulting data for the studies with -1F β-lactamase are easy to explain. There were no SPase mutants studied that were able to allow IT41 containing the β-lactamase mutant to grow at the 500µg/ml ampicillin concentration suggesting that the SPase mutants were not able to process this substrate adequately. It is possible that there is some low-level processing that is occurring in these cases but the mature β-lactamase produced is insufficient to allow significant growth.

Finally, the -1N substrate, as explained previously in Chapter 2, is analogous to the substrate recognition sequence of the mitochondrial Imp1 protease. As demonstrated with

76 the -1N PONA substrate, the I144C/I86T substrate was capable of cleaving this sequence with enhanced specificity compared to the WT. In light of this result, it was expected that there would be some ability to grow at 500µg/ml ampicillin in this experiment. However, this was not observed for the I144C/I86T SPase mutant. The WT and I144A mutant exhibited no growth under the highly selective conditions. In the case of the I144C mutant, it was expected that there would be some growth since this particular SPase mutant demonstrated relaxed specificity in the PONA studies. However, growth in this case was not observed. The few colonies that grew were most likely the result of suppressor mutations.

As in the –1V study, the I144A/I86A mutant produced more growth with the –1N β- lactamase than any of the other mutants assayed. When examining these results in light of the pulse-chase results, it is likely that the growth seen here is the result of some unanticipated factor. The pulse-chase data for the I144A/I86A mutant SPase with the -1N

β-lactamase show very slight processing and with this result in mind, it was expected that very little growth would be observed. On the contrary, a relatively large amount of cells were seen on this plate (Figure 3.2C). It is possible that the high concentrations of these cells applied in this case are partially responsible for this anomalous growth. More work will have to be done to understand this result.

From these studies, it is clear that this plate-based assay system can be effective for the rapid determination of acceptable substrates for SPase and SPase mutants. β-lactamase mutants that were adequate SPase substrates supported growth at 500µg/ml ampicillin

77 while those that were not could only grow at 50µg/ml. One weakness of the assay is the growth observed at 500µg/ml ampicillin of IT41 containing K145A SPase and wild-type

β-lactamase. Apparently, there still remains some SPase activity in IT41 to promote growth (and presumably cleavage of β-lactamase) at 500µg/ml ampicillin. Future studies will optimize the genetic screen by using a modified β-lactamase protein such that growth

of IT41 does not occur at 500µg/ml ampicillin unless a SPase protein that cleaves β-

lactamase is introduced.

3.3.1 Genetic applications

With such a genetic screen, one application would be to select from a large set of

potential mutants for ones that exhibit efficient cleavage. For example, a library of

random mutations within the cleavage region of β-lactamase could be used to select for

the preferred cleavage sequence(s) of candidate SPase mutants. Additionally, the reverse

could be done where randomly mutated SPase could be screened to search for mutations

that allow the efficient processing of a non-canonical cleavage sequence in β-lactamase.

In these cases, the colonies that grow well represent substrates that are well tolerated by the SPase mutant in question. In this way, a substrate selection profile could be formed

for each mutant SPase to contrast with the known preference of the WT. The –1F

substrate would be appropriate for this type of study due to the lack of growth observed

for the assayed SPases at the high ampicillin concentration. Conversely, the I144A/I86A

mutant would be an interesting mutant to assay with the randomized β-lactamase screen to develop a specificity profile to compare to the WT.

78 3.4 Results: Pulse-Chase Studies of in vivo processing

3.4.1 Processing of WT β-lactamase

The plate assay worked well for identifying potential substrate/enzyme combinations that

presumably reflect cleavage of pre-β-lactamase by signal peptidase. Therefore, the

processing of the β-lactamase protein by WT SPase and other select SPase mutants was

examined directly. This was done by examining the amounts of pulse-labeled β-lactamase

pre- and mature forms present at increasing chase times (Figure 3.3A). As seen in this

figure, the results agree with the relative activities seen with the same mutants processing

the pro-OmpA-nuclease A substrate (Figures 2.4A, 2.5A and Table 2.1). The WT SPase

exhibited essentially complete processing in the earliest time point as well as did I144A and I144C SPase. Processing by the double mutant I144C/I86T was more impaired but still resembled the WT. Finally, the I144A/I86A mutant was significantly impaired with incomplete processing even after 5 minutes of chase. As a negative control, the catalytic residue mutant K145A showed only minor processing at the longer chase points, presumably due to the action of low levels of the chromosomally-encoded signal peptidase produced in IT41 cleaving at the SPase site. This is supported by previous

mutational and crystallographic studies that have confirmed that K145 is a catalytic

residue that is essential for activity. Also consistent with this notion is the fact that when

β-lactamase with a valine substitution at the -1 position is studied, no cleavage at all is observed with the IT41 strain containing the K145A mutant (Figure 3.3B) as any WT

79 SPase originating from the chromosome would cleave the -1V β-lactamase poorly. These

results mirror the results obtained in the plate-based assay (see above).

Another observation that was made throughout these studies was that in all experiments

where pre-β-lactamase was present at longer chase times, a reduction in band intensity starting around the 120 second chase point was uniformly observed suggesting a slight instability in the premature form of this enzyme. Peak processing of β-lactamase is observed at this time as well with little increase in the intensity of the processed bands in cases where there still remains unprocessed material to be cleaved.

3.4.2 Processing of –1V β-lactamase

Of the three mutant substrates of β-lactamase studied in this section, the -1V substrate represents the substrate with a -1 residue with the most minor deviation from the chemical nature of the alanine residue found at this position in the WT β-lactamase. With

this relatively minor deviation in the size of the -1 residue, it is possible that the WT

SPase could tolerate this cleavage sequence without the option to cleave a nearby alternative sequence. My results with the WT enzyme confirm this with the generation of a processed band in the first chase point which peaks in intensity by the 120 second time point (all pulse-chase data for this substrate can be seen in Figure 3.3B). As expected, the

K145A negative control demonstrates no processing, even after 15 minutes of chase.

80 Since the amount of substrate processed by the WT SPase peaked after 120 seconds and because it is approximately 50% processed (slightly less, Figure 3.3B), comparisons of the WT and mutant SPases will be made up to this point. The I144C mutant was capable of processing this substrate slightly more efficiently than the WT SPase: at 120 seconds

I144C processed 50% of the -1V β-lactamase while the WT SPase had processed noticeably less than 50%. The I144A and I144C/ I86T mutants had similar activity toward this substrate as the WT though somewhat reduced. As observed previously, the

I144A/I86A mutant was the most impaired of all enzymes studied with little cleavage product seen, even at the 120 second time point.

3.4.3 Processing of –1F β-lactamase

The –1F β-lactamase is a poor substrate much like the –1F pro-OmpA-nuclease A substrate analyzed in Chapter 2. As observed previously, the WT SPase cannot cleave this substrate with a phenylalanine at the –1 position. Since there are no other cleavage sites present, the β-lactamase remains unprocessed in the pulse-chase study, even at the longest chase time of 300 seconds. All results are shown in Figure 3.3C.

While the results for the WT SPase showed no cleavage, some of the mutants studied fared better. The mutants I144A and I144C exhibited almost no cleavage of the –1F residue substrate with only a very faint mature band developing at the longest chase point. The double mutant I144A/I86A produced a slightly increased amount of cleavage, with a processed band of slightly greater intensity developing. Finally, the I144C/I86T mutant exhibited the greatest activity of all with this substrate, however the processing is

81 very poor, nevertheless. In Figure 3.3C, a processed band can be seen in the early chase

points that is of greater intensity than that observed with the other mutants, even at the

300 second chase times.

3.4.4 Processing of –1N β-lactamase

Among signal peptidase enzymes, the ability to cleave substrates with an asparagine

residue at the –1 position is the unique ability of the mitochondrial inner membrane

protease, Imp1. As demonstrated in Chapter 2, modifying the I144 and I86 residues of

SPase I to the cysteine and/or threonine residues present in the Imp1 enzymes had the

result of changing the specificity of the mutant enzyme to enhance its activity toward the

-1N mutant (relative to the WT substrate with -1A) of pro-OmpA-nuclease A.

In these experiments the same mutant SPases assayed above with the WT β-lactamase were tested against the –1N substrate. As Figure 3.3D clearly shows, the WT SPase is

unable to cleave the –1N sequence, even after extended incubation times. In contrast,

limited cleavage was observed with all the mutant SPases assayed: both double mutants

produced a faint processed band at the longer chase times while the single mutants

exhibited more efficient processing of this substrate with I144A exhibiting the most

favorable reactivity. Admittedly, the activity of the SPase enzymes is very poor for the -

1N substrate compared to the WT substrate.

82 3.5 Discussion

The results of these in vivo experiments parallel the in vitro data with the PONA substrate. The SPase mutants studied here exhibit similar activity levels toward the β- lactamase substrate as was observed with the PONA substrate. In the case of the WT cleavage sequence, the pulse-chase results agree with the activity estimates made in

Chapter 2 based on the dilution series experiments with PONA (Table 2.1). WT SPase could cleave efficiently along with the I144A and I144C mutants. These mutants were found to possess 50% of the WT activity with the PONA substrate which is consistent with the efficient cleavage seen here with pre- β-lactamase in vivo. In fact, an examination of the data reveals the mutants appear to cleave better than the WT with complete processing in the first chase while some unprocessed material was still present in the first lane of the WT. This is unlikely to represent a real advantage to the mutants here and is most likely due to differences in the health and condition of the cells used in this study. The IT41 SPase temperature sensitive strain used in this study is challenging to cultivate and use effectively. Each transformed strain grows at a different rate and fitness level. Though great care was taken to control cell density and labeling conditions, some small level of variation is inevitable. The double mutants I144C/I86T and

I144A/I86A also behaved as expected with increasing processing defects evident.

An interesting observation using the WT β-lactamase substrate was that the processing defect predicted from the studies with pro-OmpA-nuclease A was not as extreme as expected. In the in vitro studies with PONA, the I144 single mutants had 50% of WT

83 activity (Figure 2.4A, Table 2.1) while the I144C/I86T mutant had approximately 1%

WT activity (Figure 2.5A). In spite of these figures, relatively large amounts of processed material are generated in these in vivo studies. With the single mutants appearing slightly better than WT while the I144C/I86T mutant appearing to be only slightly more impaired than WT. It is possible that the in vivo conditions are more conducive to promoting the full potential activity in these mutants that the detergent micelle is not able to do in vitro.

The natural membrane environment could promote a more cleavage friendly context that is not capable of being simulated with detergent.

The processing of the K145A negative control was minimal and consistent with some residual signal peptidase activity in the conditional-lethal signal peptidase strain. When an unconventional -1V residue was substituted, there was no processing observed in the

IT41 strain containing the K145A SPase; the residual signal peptidase activity is inadequate to process the -1V substrate. Since this background processing was slight and only occurred with the WT β-lactamase, it is unlikely to significantly affect the processing observed with the other SPase mutants.

In the studies involving the -1V β-lactamase, the results are more interesting. As expected, the WT SPase processing is impaired. Maximum processing occurred at 120 seconds chase with approximately 30% processing. Comparing favorably, the I144A mutant showed a similar amount of processing after 120 seconds while the I144C mutant processed around 50%. These figures demonstrate little difference in the activities of these mutants and WT for the -1V substrate. There may be differences in the abilities of

84 the mutants and WT to accommodate this substrate but they are not great enough to be

detected by this assay. In the experiments with the -1F and -1N β-lactamase, there are

clear differences. For both of these substrates, the WT produces no processed material.

On the other hand, the I144A and I144C mutants both produce processed material in the longer chase lanes. For the -1F β-lactamase, the bands are very faint and difficult to see, but are present at the final chase time of 300 seconds. For the -1N β-lactamase, the processed bands for these mutants are unambiguous and appear early on. This is consistent with the relaxed specificity seen for these mutants in the in vitro PONA studies.

The double mutants I144C/I86T and I144A/I86A also demonstrate altered specificity in comparison to the WT enzyme in these studies. Both mutants can process the -1V substrate but the significant differences are seen with the -1F and -1N β-lactamases. Both

mutants can handle the -1N cleavage site better than the WT SPase with the I144C/I86T

mutant resembling the single mutants and the I144A/I86A mutant showing significant

amounts of processed material only at the longest chase time. The results for the -1F

substrate are rather interesting in that the I144C/I86T mutant generates a processed band

that is much greater what the WT or any other mutant could produce. While the double

alanine mutant was shown to process following a phenylalanine residue in PONA and in

these studies, as well, it appears that the I144C/I86T Spase is quite tolerant of this large

aromatic residue at -1.

85 As stated previously the I144C/I86T mutation was designed based on the mitochondrial

Imp1 protease. When aligned with SPaseI, there are little differences in the conserved regions associated with the substrate binding pockets with the notable exception of the positions homologous to 144 and 86 in SPaseI. It is likely that the differences observed here represent an evolved change in the substrate binding pockets to allow the enzyme to recognize and bind larger residues at the -1 position.

For the assay of a large variety of mutant substrates and SPase enzymes, the method of studying β-lactamase processing in vivo is superior to the study of the PONA substrate.

As demonstrated here, the pulse-chase study can yield clean and unambiguous immunoprecipitations to demonstrate a time course of processing. And since β-lactamase has only a single potential processing site, manipulation of this site can be done with little worry of alternate processing. Only in the case of a SPase mutant with a significantly altered specificity concomitant with high activity could this become an issue. In such as case, the β-lactamase cleavage product can be purified by phenyl-boronate affinity chromatography for mass analysis to determine the site of cleavage. I have purified β- lactamase in this way and the purification procedure produced reasonably pure enzyme suitable for mass spectrometry (Figure 3.4).

86 3.6 Materials and Methods

3.6.1 Bacterial strains and plasmids

The E. coli strain DH5α was obtained from our laboratory collection while the E. coli temperature-sensitive SPase I strain, IT41, was obtained from Dr. Yoshikazu Nakamura

[25]. The plasmid pRD8 which contains the SPase I gene in the pING vector was obtained from our collection. The plasmid pGZ119HE was generously provided by Dr.

Andreas Kuhn.

3.6.2 Construction of plasmids

To examine the ability of β-lactamase mutants to be processed by the various SPase binding pocket mutants, a two plasmid system was employed requiring the preparation of two constructs. The construction of the two plasmids was conducted as follows:

First, the SPase mutants needed to be subcloned from the pET 23b to avoid the concomitant expression of WT β-lactamase from this vector in these studies. To this end,

SPase was subcloned into the SmaI/SalI site of the pGZ119HE vector [125]. The pGZ119HE is suitable for this study because it possesses the ColD origin of replication and confers chloramphenicol resistance. The pET 23b plasmids bearing the various SPase mutants were digested with SalI and SmaI in the same reaction vessel using 1.5X

Universal Buffer (Stratagene) and the DNA fragment containing SPase was purified by

87 excision from an agarose gel. The pGZ119HE vector was prepared in the same way and the two were ligated to produce the pGZ119HE-SPase expression vector. The resulting

DNA was sequenced for confirmation.

Second, a plasmid capable of expressing β-lactamase was needed that could be cotransformed with the SPase mutants. For this vector, I choose to modify pRD8 by removing the SPase gene [23]. The pRD8 plasmid contains the ColE1 replication origin and the bla gene for ampicillin resistance and is thus compatible with the pGZ119HE-

SPase expression vector. Additionally, β-lactamase can be highly expressed from this plasmid by the addition of 0.2% arabinose. The pRD8 plasmid was digested with the SalI and SmaI enzymes as described above and run on an agarose gel. The DNA fragment corresponding to the double cut vector was excised and purified from the gel. The resultant DNA was then ligated to produce what is essentially the original pING vector.

For information on pRD8, see reference [23].

The pING vector was then modified using the Quikchange (Stratagene Inc) site specific mutagenesis method to incorporate different amino acid residues at the position -1 to the cleavage site.

3.6.3 Purification of β-lactamase

β-lactamase was purified using the PheBo system from MoBiTec (Goettingen, Germany) that utilizes phenyl-boronate agarose resin for the specific purification of β-lactamase.

88 The purification was conducted as described in the manual. To summarize, DH5α cells, which were first transformed with the pET23b vector, were grown to saturation at 37°C.

Cells were isolated by low speed centrifugation (4000 rpm, JA-10) and the periplasmic

fraction was isolated in the following manner: Cells were resuspended in ice cold STE

buffer (20% sucrose, 200mM Tris/HCl, 100mM EDTA, pH 9.0) with gentle shaking for

20 minutes and then pelleted (10,000 rpm, JA-10). Pellet was then resuspended in ice

cold 10mM Tris/HCl, pH 9.0 with gentle shaking for 20 minutes and then pelleted

(10,000 rpm, JA-10). At this step, the supernatant was isolated and contains the

periplasmic cell fraction. β-lactamase was precipitated from the periplasmic fraction with

ammonium sulfate (130g into 200ml periplasmic fraction) and the precipitated proteins were redissolved and applied to the phenyl-boronate column. Elution was performed with borate buffer elute purified β-lactamase.

3.6.4 Pulse-chase assay of β-lactamase processing

As described in section 2.4.5, all media intended for the culturing of IT41 was prepared with a reduced salt concentration of 2.5g NaCl/liter (LS2.5 media). Competent cells of the IT41 strain were prepared using the CaCl2 method and cotransformed with the various mutants of the pGZ119HE-SPase and pING vectors. These were then grown at

30°C on solid LS2.5 media until colonies were 1-2mm in size. These were then transferred to liquid LS2.5 media and grown to a cell density of 0.3-0.5 OD600. At this point, the cells were transferred to M9+fructose media and incubated at 30°C. After 30 minutes, the temperature was shifted to 42°C and incubation was continued for one hour. 89 At 5 minutes prior to pulse, arabinose was added to a final concentration of 0.2% to increase the expression of β-lactamase. The cells were pulsed with [35S]-methionine for 1 minute and then chased with non-radioactive methionine for the indicated times. At each time point, samples were quenched with a 10% final concentration of TCA in preparation for immunoprecipitation with rabbit anti-β-lactamase polyclonal antibody (Chemicon

Intl.).

3.6.5 Growth assay for β-lactamase processing

The same cotransformants in the pulse-chase study were also assayed for their ability to grow under different ampicillin concentrations. Agarose plates were prepared with 2.5g

NaCl/liter, 15µg/ml tetracycline, 30µg/ml chloramphenicol, and either 50 or 500 µg/ml ampicillin. The cells were streaked out media of both ampicillin concentrations and grown at 30 and 42 °C. For the extended incubation necessary for these studies, it was found that efforts must be taken to prevent the solid media from drying out at 42 °C.

Dehydration of the media results in a higher effective salt concentration and increased likelihood of reversion. This was accomplished by pouring plates that were extra thick and also by adding an open dish of water to the incubator to maintain high humidity.

3.6.6 Conclusion

These studies have provided insight into the nature of SPase substrate selection and binding by utilizing rational enzyme modification based on structural data. They also

90 illustrate that structural data of an enzyme not only represent the structural features of the

WT enzyme but can also be used to model hypothetical changes in the enzyme structure before actually making these changes in the form of a purified mutant enzyme. Computer

modeling was then used to dock the substrate and enzyme together thus allowing rational planning for SPase binding pocket mutations that may affect substrate binding. These

methods allowed for the selection of S1 and S3 pocket residue changes that were most

likely to effect the desired changes in pocket size. Finally the best candidate changes

were made and the effects studied using WT and mutant signal peptide cleavage

sequences.

Significantly, the enzyme substrate specificity profile was altered in several mutants by

the alteration of the I144 residue in addition to the I86 residue of SPase. It was our

objective to enlarge these pockets in order to accommodate non-canonical residues within

the substrate peptide. Mass data presented in Chapter 2 clearly illustrate an alteration in

the ability of SPase to bind and process substrates with larger residues at the –1 position.

In these cases the substrate specificity was extended to accommodate non-canonical -1

residues in addition to the normally preferred -1 alanine. Though enzyme specificity was not altered while maintaining good catalytic efficiency, in fact some mutants such as the

I144A/I86A mutant had very poor activity. This mutant was measured to have activity

only 0.1% that of the WT SPase when assayed with the PONA substrate. Though the

I144C/I86T mutant had better activity with approximately 1% WT activity, it was still a very inefficient enzyme.

91 These results are significant in that they deepen our understanding of the E. coli SPase I

substrate specificity and how alteration in the substrate binding region can result in reciprocal changes in substrate binding ability.

Some of the alterations based on computer modeling resulted in the desired alteration in substrate binding. For example, when the I144A/I86A and I144C/I86T double mutants were assayed with the –1F substrates, it was observed that cleavage could occur at this site which is not processed by the WT. In addition when signal peptidase was designed to mimic the Imp1 specificity, it was observed that this alteration did enhance this mutant’s ability to process this substrate over the WT SPase.

There were surprises, though, and changes did not always result in the desired effect. For instance, the I144C mutant exhibited a dramatic reduction in the fidelity of cleavage site selection with the WT and –1F PONA substrate (Table 2.2, Figure 2.8). In addition to processing following the –1 site, it also processed at many other sites surrounding –1. It is not clear why such a minor change in the enzyme sequence should have such a dramatic effect on the fidelity of substrate selection.

Another unexpected result occurred with the I144C/I86T mutant that did cleave following –1N residue but could also cleave following the –1F residue. In fact, it seemed to prefer the –1F cleavage site (compare Figure 3.3C and D). This should not be entirely surprising in that there are many other differences between SPase I of E. coli and the

92 Imp1 protease and that the process of evolution has had much more time than I have had

to make all the possible changes.

This leads to the next logical step in these studies that I have begun in the development of

a plate based assay for IT41 growth based on in vivo β-lactamase processing. It should be

noted that these are preliminary results and that more work needs to be done to

characterize these results. Though the results of this study establish the ability of this

assay to discriminate between situations where there was efficient processing and where

processing was strongly inhibited. In the cases where processing is revealed to be more

intermediate, the plate-based assay can sometimes yield results that cannot be directly

correlated to the pulse-chase processing data.

In developing the in vivo system based on β-lactamase processing, I have established the

groundwork and developed the system that will allow future studies with in vitro

evolution. Just as in the natural process of evolution these studies will employ random

mutations in both enzyme and substrate coupled with survivability in high concentrations of ampicillin to provide selection of successful processing. It is in this way that altered specificities could be developed while maintaining catalytic efficiency.

93

PONA Cleavage region:

TM -IAIAVALAGFATVAQAATSTKKLHKEPA- Mature

β-lactamase Cleavage region

TM -VALIPFFAAFCLPVFAHPETLVKVKDAE- Mature

Figure 3.1 Depiction of sequence surrounding PONA and β-lactamase cleavage sites The amino acid sequence surrounding the site of cleavage is depicted in single letter code for the PONA and β-lactamase substrates. WT cleavage site is indicated with a vertical arrow and -1 and -3 residues are shown in boldface. Note lack of alternative cleavage sites within the β-lactamase protein.

94

50µg/ml 500µg/ml Ampicillin

A -1V β-lactamase

A

AA K145A

C WT

CT K145A/ WT Blac WT/WT Blac

Figure 3.2 Growth assay for β-lactamase processing. Plates shown were incubated at 30ºC. In each photograph, the plate containing 50µg/ml is shown at left and the one with 500µg/ml is shown at right. Plates are divided into sections with each section labeled for the SPase mutant assayed (order of mutants is the same for each set of 50 and 500µg/ml plates): I144A “A”, I144C “C”, I144A/I86A “AA” and I144C/I86T “CT”. Plates in (A) assayed primarily the -1V substrate with the WT/WT control appearing at bottom and the K145A/WT negative control adjacent in the counterclockwise direction. In (B) the -1F substrate is studied along with the K145A/WT control at bottom. And finally, in (C) the -1N substrate is assayed with the WT/WT control repeated at bottom.

(Continued)

95

Figure 3.2 (Continued)

50µg/ml 500µg/ml Ampicillin

B -1F β-lactamase

AA

C CT

WT A

K145A/WT Blac

C -1N β-lactamase

AA

C CT

A WT

WT/WT Blac

96 A (WT) 10 30 60 120 300 seconds of chase

P WT M

P K145A M

P I144A M

P I144C M

P I144C/I86T M

P I144A/I86A M

10 30 60 120 300 seconds of chase B (-1V) WT P M

K145A P M

I144A P M

P I144C M

P I144C/I86T M

I144A/I86A P M

Figure 3.3 Processing of β-lactamase WT and mutants by SPase in vivo. Pulse-chase studies depicting WT (A), -1V (B), -1F (C) and -1N (D) β-lactamase processing by SPase WT and mutants in vivo. Cells were pulse-labeled with 35S- methionine and chased with cold methionine for the indicated times. Labeled β-lactamase was immunoprecipitated and run on SDS-PAGE to be detected finally by phosphorimaging. (Continued) 97 Figure 3.3 (Continued)

C (-1F) 10 30 60 120 300 seconds of chase

P WT M

P I144A M

P I144C M

P I144C/I86T M

P I144A/I86A M

D (-1N) 10 30 60 120 300 seconds of chase

P WT M

P I144A M

P I144C M

P I144C/I86T M

P I144A/I86A M

98

Marker Fraction number

Size (kDa)

75 - 50 - 37 - β-lactamase 25 -

20 -

Total

lysate

Figure 3.4 Purification of WT β-lactamase from periplasmic fraction. β-lactamase was purified from DH5α cells harboring the pET23b vector. As described in Methods section 3.6.3, the periplasmic fraction was first isolated and β-lactamase was precipitated with the addition of ammonium sulfate. Precipitated proteins were redissolved and applied to a phenyl-boronate column (MoBiTec). Elution was achieved by the application of borate buffer. The position of β-lactamase is indicated.

99

REFERENCES

1. Blobel, G. and D.D. Sabatini, Ribosome-membrane interaction in eukaryotic cells, in Biomembranes, L.A. Manson, Editor. 1971, Plenum: New York. p. 193- 195.

2. Milstein, C., et al., A possible precursor of immunoglobulin light chains. Nat New Biol, 1972. 239(91): p. 117-20.

3. Blobel, G. and B. Dobberstein, Transfer of proteins across membranes. I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. J Cell Biol, 1975. 67(3): p. 835-51.

4. Paetzel, M., et al., Signal peptidases. Chem Rev, 2002. 102(12): p. 4549-80.

5. Tjalsma, H., et al., Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome. Microbiol Mol Biol Rev, 2000. 64(3): p. 515-47.

6. Paetzel, M., R.E. Dalbey, and N.C. Strynadka, The structure and mechanism of bacterial type I signal peptidases. A novel antibiotic target. Pharmacol Ther, 2000. 87(1): p. 27-49.

7. Dalbey, R.E., et al., The chemistry and enzymology of the type I signal peptidases. Protein Sci., 1997. 6(6): p. 1129-38.

8. von Heijne, G., Patterns of amino acids near signal-sequence cleavage sites. Eur J Biochem, 1983. 133(1): p. 17-21.

9. Perlman, D. and H.O. Halvorson, A putative signal peptidase recognition site and sequence in eukaryotic and prokaryotic signal peptides. J Mol Biol, 1983. 167(2): p. 391-409.

10. von Heijne, G., Signal sequences. The limits of variation. J Mol Biol, 1985. 184(1): p. 99-105.

100 11. Date, T. and W. Wickner, Isolation of the Escherichia coli leader peptidase gene and effects of leader peptidase overproduction in vivo. Proc Natl Acad Sci U S A, 1981. 78(10): p. 6106-10.

12. Wolfe, P.B., W. Wickner, and J.M. Goodman, Sequence of the leader peptidase gene of Escherichia coli and the orientation of leader peptidase in the bacterial envelope. J Biol Chem, 1983. 258(19): p. 12073-80.

13. Chatterjee, S., et al., Determination of Km and kcat for signal peptidase I using a full length secretory precursor, pro-OmpA-nuclease A. J. Mol. Biol., 1995. 245(4): p. 311-4.

14. Sung, M. and R.E. Dalbey, Identification of potential active-site residues in the Escherichia coli leader peptidase. J. Biol. Chem., 1992. 267(19): p. 13154-9.

15. Tschantz, W.R., et al., A serine and a lysine residue implicated in the catalytic mechanism of the Escherichia coli leader peptidase. J Biol Chem, 1993. 268(36): p. 27349-54.

16. Paetzel, M., et al., Use of site-directed chemical modification to study an essential lysine in Escherichia coli leader peptidase. J Biol Chem, 1997. 272(15): p. 9994- 10003.

17. Paetzel, M., R.E. Dalbey, and N.C. Strynadka, Crystal structure of a bacterial signal peptidase in complex with a beta-lactam inhibitor [see comments] [published erratum appears in Nature 1998 Dec 17;396(6712):707]. Nature, 1998. 396(6707): p. 186-90.

18. Tjalsma, H., et al., Bacillus subtilis contains four closely related type I signal peptidases with overlapping substrate specificities. Constitutive and temporally controlled expression of different sip genes. J Biol Chem, 1997. 272(41): p. 25983-92.

19. Meijer, W.J., et al., The endogenous Bacillus subtilis (natto) plasmids pTA1015 and pTA1040 contain signal peptidase-encoding genes: identification of a new structural module on cryptic plasmids. Mol Microbiol, 1995. 17(4): p. 621-31.

20. Parro, V., et al., Four genes encoding different type I signal peptidases are organized in a cluster in Streptomyces lividans TK21. Microbiology, 1999. 145 ( Pt 9): p. 2255-63.

21. Cregg, K.M., I. Wilding, and M.T. Black, Molecular cloning and expression of the spsB gene encoding an essential type I signal peptidase from Staphylococcus aureus. J Bacteriol, 1996. 178(19): p. 5712-8.

101 22. Date, T., Demonstration by a novel genetic technique that leader peptidase is an essential enzyme of Escherichia coli. J Bacteriol., 1983. 154(1): p. 76-83.

23. Dalbey, R.E. and W. Wickner, Leader peptidase catalyzes the release of exported proteins from the outer surface of the Escherichia coli plasma membrane. J. Biol. Chem., 1985. 260(29): p. 15925-31.

24. Bolhuis, A., et al., Different mechanisms for thermal inactivation of Bacillus subtilis signal peptidase mutants. J Biol Chem, 1999. 274(22): p. 15865-8.

25. Inada, T., et al., Conditionally lethal amber mutations in the leader peptidase gene of Escherichia coli. J Bacteriol, 1989. 171(1): p. 585-7.

26. Zhang, Y.B., B. Greenberg, and S.A. Lacks, Analysis of a Streptococcus pneumoniae gene encoding signal peptidase I and overproduction of the enzyme. Gene, 1997. 194(2): p. 249-55.

27. Wolfe, P.B., P. Silver, and W. Wickner, The isolation of homogeneous leader peptidase from a strain of Escherichia coli which overproduces the enzyme. J Biol Chem, 1982. 257(13): p. 7898-902.

28. Moore, K.E. and S. Miura, A small hydrophobic domain anchors leader peptidase to the cytoplasmic membrane of Escherichia coli. J Biol Chem, 1987. 262(18): p. 8806-13.

29. Whitley, P., L. Nilsson, and G. von Heijne, Three-dimensional model for the membrane domain of Escherichia coli leader peptidase based on disulfide mapping. Biochemistry, 1993. 32(33): p. 8534-9.

30. Dalbey, R.E., et al., The chemistry and enzymology of the type I signal peptidases. Protein Sci, 1997. 6(6): p. 1129-38.

31. Evans, E.A., R. Gilmore, and G. Blobel, Purification of microsomal signal peptidase as a complex. Proc Natl Acad Sci U S A, 1986. 83(3): p. 581-5.

32. Baker, R.K. and M.O. Lively, Purification and characterization of hen oviduct microsomal signal peptidase. Biochemistry, 1987. 26(26): p. 8561-7.

33. Lively, M.O., A.L. Newsome, and M. Nusier, Eukaryote microsomal signal peptidases. Methods Enzymol, 1994. 244: p. 301-14.

34. YaDeau, J.T. and G. Blobel, Solubilization and characterization of yeast signal peptidase. J Biol Chem, 1989. 264(5): p. 2928-34.

102 35. YaDeau, J.T., C. Klein, and G. Blobel, Yeast signal peptidase contains a glycoprotein and the Sec11 gene product. Proc Natl Acad Sci U S A, 1991. 88(2): p. 517-21.

36. Bohni, P.C., R.J. Deshaies, and R.W. Schekman, SEC11 is required for signal peptide processing and yeast cell growth. J Cell Biol, 1988. 106(4): p. 1035-42.

37. Shelness, G.S. and G. Blobel, Two subunits of the canine signal peptidase complex are homologous to yeast SEC11 protein. J Biol Chem, 1990. 265(16): p. 9512-9.

38. Greenburg, G., G.S. Shelness, and G. Blobel, A subunit of mammalian signal peptidase is homologous to yeast SEC11 protein. J Biol Chem, 1989. 264(27): p. 15762-5.

39. Schneider, A., et al., Inner membrane protease I, an enzyme mediating intramitochondrial protein sorting in yeast. Embo J, 1991. 10(2): p. 247-54.

40. Nunnari, J., T.D. Fox, and P. Walter, A mitochondrial protease with two catalytic subunits of nonoverlapping specificities. Science, 1993. 262(5142): p. 1997-2004.

41. Behrens, M., G. Michaelis, and E. Pratje, Mitochondrial inner membrane protease 1 of Saccharomyces cerevisiae shows sequence similarity to the Escherichia coli leader peptidase. Mol Gen Genet, 1991. 228(1-2): p. 167-76.

42. Schneider, A., W. Oppliger, and P. Jeno, Purified inner membrane protease I of yeast mitochondria is a heterodimer. J Biol Chem, 1994. 269(12): p. 8635-8.

43. Chaal, B.K., et al., Characterization of a cDNA encoding the thylakoidal processing peptidase from Arabidopsis thaliana. Implications for the origin and catalytic mechanism of the enzyme. J Biol Chem, 1998. 273(2): p. 689-92.

44. Schatz, P.J. and J. Beckwith, Genetic analysis of protein export in Escherichia coli. Annu. Rev. Genet., 1990. 24: p. 215-48.

45. Bedouelle, H., et al., Mutations which alter the function of the signal sequence of the maltose binding protein of Escherichia coli. Nature, 1980. 285(5760): p. 78- 81.

46. Puziss, J.W., J.D. Fikes, and P.J. Bassford, Jr., Analysis of mutational alterations in the hydrophilic segment of the maltose-binding protein signal peptide. J Bacteriol, 1989. 171(5): p. 2303-11.

47. Nesmeyanova, M.A., et al., Positively charged lysine at the N-terminus of the signal peptide of the Escherichia coli alkaline phosphatase provides the secretion

103 efficiency and is involved in the interaction with anionic phospholipids. FEBS Lett, 1997. 403(2): p. 203-7.

48. von Heijne, G., Net N-C charge imbalance may be important for signal sequence function in bacteria. J Mol Biol, 1986. 192(2): p. 287-90.

49. Paetzel, M., et al., Crystallographic and biophysical analysis of a bacterial signal peptidase in complex with a lipopeptide-based inhibitor. J Biol Chem, 2004. 279(29): p. 30781-90.

50. van Dijl, J.M., et al., Identification of the potential active site of the signal peptidase SipS of Bacillus subtilis. Structural and functional similarities with LexA-like proteases. J Biol Chem, 1995. 270(8): p. 3611-8.

51. Black, M.T., J.G. Munn, and A.E. Allsop, On the catalytic mechanism of prokaryotic leader peptidase 1. Biochem J, 1992. 282 ( Pt 2): p. 539-43.

52. Black, M.T., Evidence that the catalytic activity of prokaryote leader peptidase depends upon the operation of a serine-lysine catalytic dyad. J Bacteriol, 1993. 175(16): p. 4957-61.

53. VanValkenburgh, C., et al., The catalytic mechanism of endoplasmic reticulum signal peptidase appears to be distinct from most eubacterial signal peptidases. J Biol Chem, 1999. 274(17): p. 11519-25.

54. Shinagawa, H., SOS response as an adaptive response to DNA damage in prokaryotes. Exs, 1996. 77: p. 221-35.

55. Little, J.W., LexA cleavage and other self-processing reactions. J Bacteriol, 1993. 175(16): p. 4943-50.

56. Peat, T.S., et al., Structure of the UmuD' protein and its regulation in response to DNA damage. Nature, 1996. 380(6576): p. 727-30.

57. Keiler, K.C. and R.T. Sauer, Identification of active site residues of the Tsp protease. J Biol Chem, 1995. 270(48): p. 28864-8.

58. Birghan, C., E. Mundt, and A.E. Gorbalenya, A non-canonical lon proteinase lacking the ATPase domain employs the ser-Lys catalytic dyad to exercise broad control over the life cycle of a double-stranded RNA virus. Embo J, 2000. 19(1): p. 114-23.

59. Strynadka, N.C., et al., Molecular structure of the acyl-enzyme intermediate in beta-lactam hydrolysis at 1.7 A resolution. Nature, 1992. 359(6397): p. 700-5.

104 60. Kuo, D., et al., Determination of the kinetic parameters of Escherichia coli leader peptidase activity using a continuous assay: the pH dependence and time- dependent inhibition by beta-lactams are consistent with a novel serine protease mechanism. Biochemistry, 1994. 33(27): p. 8347-54.

61. Chabin, R., et al., Mechanism of inhibition of human leucocyte elastase by monocyclic beta-lactams. Biochemistry, 1993. 32(34): p. 8970-80.

62. Faraci, W.S. and R.F. Pratt, Mechanism of inhibition of RTEM-2 beta-lactamase by cephamycins: relative importance of the 7 alpha-methoxy group and the 3' leaving group. Biochemistry, 1986. 25(10): p. 2934-41.

63. Knight, W.B., et al., Mechanism of inhibition of human leukocyte elastase by two cephalosporin derivatives. Biochemistry, 1992. 31(21): p. 4980-6.

64. Wilmouth, R.C., et al., Mechanistic insights into the inhibition of serine proteases by monocyclic lactams. Biochemistry, 1999. 38(25): p. 7989-98.

65. Black, M.T. and G. Bruton, Inhibitors of bacterial signal peptidases. Curr Pharm Des, 1998. 4(2): p. 133-54.

66. Wickner, W., et al., Inhibition of purified Escherichia coli leader peptidase by the leader (signal) peptide of bacteriophage M13 procoat. J Bacteriol, 1987. 169(8): p. 3821-2.

67. Barkocy-Gallagher, G.A. and P.J. Bassford, Jr., Synthesis of precursor maltose- binding protein with proline in the +1 position of the cleavage site interferes with the activity of Escherichia coli signal peptidase I in vivo. J Biol Chem, 1992. 267(2): p. 1231-8.

68. Hutchison, C.A., 3rd, et al., Mutagenesis at a specific position in a DNA sequence. J Biol Chem, 1978. 253(18): p. 6551-60.

69. Sanger, F., S. Nicklen, and A.R. Coulson, DNA sequencing with chain- terminating inhibitors. Proc Natl Acad Sci U S A, 1977. 74(12): p. 5463-7.

70. Serrano, L., A.G. Day, and A.R. Fersht, Step-wise mutation of barnase to binase. A procedure for engineering increased stability of proteins and an experimental analysis of the evolution of protein stability. J Mol Biol, 1993. 233(2): p. 305-12.

71. Mendel, D., et al., Probing protein stability with unnatural amino acids. Science, 1992. 256(5065): p. 1798-802.

72. Jandu, S.K., et al., Role of arginine 67 in the stabilization of chymotrypsin inhibitor 2: examination of amide proton exchange rates and denaturation thermodynamics of an engineered protein. Biochemistry, 1990. 29(26): p. 6264-9. 105

73. Horovitz, A., et al., Strength and co-operativity of contributions of surface salt bridges to protein stability. J Mol Biol, 1990. 216(4): p. 1031-44.

74. Voordouw, G., C. Milo, and R.S. Roche, Role of bound calcium ions in thermostable, proteolytic enzymes. Separation of intrinsic and calcium ion contributions to the kinetic thermal stability. Biochemistry, 1976. 15(17): p. 3716- 24.

75. Kuroki, R., et al., Design and creation of a Ca2+ binding site in human lysozyme to enhance structural stability. Proc Natl Acad Sci U S A, 1989. 86(18): p. 6903- 7.

76. Hedstrom, L., Serine protease mechanism and specificity. Chem Rev, 2002. 102(12): p. 4501-24.

77. Blow, D.M., J.J. Birktoft, and B.S. Hartley, Role of a buried acid group in the mechanism of action of chymotrypsin. Nature, 1969. 221(178): p. 337-40.

78. Weiner, H., et al., The formation of anhydrochymotrypsin by removing the elements of water from the serine at the active site. J Am Chem Soc, 1966. 88(16): p. 3851-9.

79. Nakagawa, Y. and M.L. Bender, Methylation of histidine-57 in alpha- chymotrypsin by methyl p-nitrobenzenesulfonate. A new approach to enzyme modification. Biochemistry, 1970. 9(2): p. 259-67.

80. Henderson, R., Catalytic activity of -chymotrypsin in which histidine-57 has been methylated. Biochem J, 1971. 124(1): p. 13-8.

81. Craik, C.S., et al., The catalytic role of the active site aspartic acid in serine proteases. Science, 1987. 237(4817): p. 909-13.

82. Sprang, S., et al., The three-dimensional structure of Asn102 mutant of trypsin: role of Asp102 in serine protease catalysis. Science, 1987. 237(4817): p. 905-9.

83. Carter, P. and J.A. Wells, Engineering enzyme specificity by "substrate-assisted catalysis". Science, 1987. 237(4813): p. 394-9.

84. Carter, P., L. Abrahmsen, and J.A. Wells, Probing the mechanism and improving the rate of substrate-assisted catalysis in subtilisin BPN'. Biochemistry, 1991. 30(25): p. 6142-8.

85. Dall'Acqua, W., et al., Elastase substrate specificity tailored through substrate- assisted catalysis and phage display. Protein Eng, 1999. 12(11): p. 981-7.

106 86. Corey, D.R., et al., Trypsin specificity increased through substrate-assisted catalysis. Biochemistry, 1995. 34(36): p. 11521-7.

87. Bone, R., J.L. Silen, and D.A. Agard, Structural plasticity broadens the specificity of an engineered protease. Nature, 1989. 339(6221): p. 191-5.

88. Estell, D.A., T. P. Graycar, J. V. Miller, D. B. Powers, J. P. Burnier, P. G. Ng, and J. A. Wells, Science, 1986. 233: p. 659.

89. Blow, D.M., in The Enzymes, P.D. Boyer, Editor. 1971, Academic Press: Boca Raton.

90. Huber, R., et al., Structure of the complex formed by bovine trypsin and bovine pancreatic trypsin inhibitor. II. Crystallographic refinement at 1.9 A resolution. J Mol Biol, 1974. 89(1): p. 73-101.

91. Krieger, M., L.M. Kay, and R.M. Stroud, Structure and specific binding of trypsin: comparison of inhibited derivatives and a model for substrate binding. J Mol Biol, 1974. 83(2): p. 209-30.

92. Knowles, J.R., Enzyme specificity: alpha-chymotrypsin. J Theor Biol, 1965. 9(2): p. 213-28.

93. Hedstrom, L., L. Szilagyi, and W.J. Rutter, Converting trypsin to chymotrypsin: the role of surface loops. Science, 1992. 255(5049): p. 1249-53.

94. Graf, L., et al., Electrostatic complementarity within the substrate-binding pocket of trypsin. Proc Natl Acad Sci U S A, 1988. 85(14): p. 4961-5.

95. Hedstrom, L., Trypsin: a case study in the structural determinants of enzyme specificity. Biol Chem, 1996. 377(7-8): p. 465-70.

96. Hedstrom, L., J.J. Perona, and W.J. Rutter, Converting trypsin to chymotrypsin: residue 172 is a substrate specificity determinant. Biochemistry, 1994. 33(29): p. 8757-63.

97. Stemmer, W.P., DNA shuffling by random fragmentation and reassembly: in vitro recombination for molecular evolution. Proc Natl Acad Sci U S A, 1994. 91(22): p. 10747-51.

98. Stemmer, W.P., Rapid evolution of a protein in vitro by DNA shuffling. Nature, 1994. 370(6488): p. 389-91.

99. Coco, W.M., et al., DNA shuffling method for generating highly recombined genes and evolved enzymes. Nat Biotechnol, 2001. 19(4): p. 354-9.

107 100. Crameri, A., et al., DNA shuffling of a family of genes from diverse species accelerates directed evolution. Nature, 1998. 391(6664): p. 288-91.

101. O'Maille, P.E., M. Bakhtina, and M.D. Tsai, Structure-based combinatorial protein engineering (SCOPE). J Mol Biol, 2002. 321(4): p. 677-91.

102. Ostermeier, M., A.E. Nixon, and S.J. Benkovic, Incremental truncation as a strategy in the engineering of novel biocatalysts. Bioorg Med Chem, 1999. 7(10): p. 2139-44.

103. Ostermeier, M., J.H. Shim, and S.J. Benkovic, A combinatorial approach to hybrid enzymes independent of DNA homology. Nat Biotechnol, 1999. 17(12): p. 1205-9.

104. Dierstein, R. and W. Wickner, Requirements for substrate recognition by bacterial leader peptidase. Embo J, 1986. 5(2): p. 427-31.

105. Fikes, J.D., et al., Maturation of Escherichia coli maltose-binding protein by signal peptidase I in vivo. Sequence requirements for efficient processing and demonstration of an alternate cleavage site. J Biol Chem, 1990. 265(6): p. 3417- 23.

106. Shen, L.M., et al., Use of site-directed mutagenesis to define the limits of sequence variation tolerated for processing of the M13 procoat protein by the Escherichia coli leader peptidase. Biochemistry, 1991. 30(51): p. 11775-81.

107. Karamyshev, A.L., et al., Processing of Escherichia coli alkaline phosphatase: role of the primary structure of the signal peptide cleavage region. J Mol Biol, 1998. 277(4): p. 859-70.

108. Pratje, E. and B. Guiard, One nuclear gene controls the removal of transient pre- sequences from two yeast proteins: one encoded by the nuclear the other by the mitochondrial genome. Embo J, 1986. 5(6): p. 1313-7.

109. Chen, X., et al., Signal peptides having standard and nonstandard cleavage sites can be processed by Imp1p of the mitochondrial inner membrane protease. J Biol Chem, 1999. 274(53): p. 37750-4.

110. Paetzel, M., R.E. Dalbey, and N.C. Strynadka, Crystal structure of a bacterial signal peptidase apoenzyme: implications for signal peptide binding and the Ser- Lys dyad mechanism. J Biol Chem, 2002. 277(11): p. 9512-9.

111. Paetzel, M., R.E. Dalbey, and N.C. Strynadka, Crystal structure of a bacterial signal peptidase in complex with a beta-lactam inhibitor. Nature, 1998. 396(6707): p. 186-90.

108 112. Nielsen, H., et al., Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng, 1997. 10(1): p. 1-6.

113. Bendtsen, J.D., et al., Improved prediction of signal peptides: SignalP 3.0. J Mol Biol, 2004. 340(4): p. 783-95.

114. Craik, C.S., et al., Redesigning trypsin: alteration of substrate specificity. Science, 1985. 228(4697): p. 291-7.

115. Ruggles, S.W., R.J. Fletterick, and C.S. Craik, Characterization of structural determinants of granzyme B reveals potent mediators of extended substrate specificity. J Biol Chem, 2004. 279(29): p. 30751-9.

116. Hahne, K., et al., Incomplete arrest in the outer membrane sorts NADH- cytochrome b5 reductase to two different submitochondrial compartments. Cell, 1994. 79(5): p. 829-39.

117. Klenotic, P.A., et al., The role of the conserved box E residues in the active site of the Escherichia coli type I signal peptidase. J Biol Chem, 2000. 275(9): p. 6490- 8.

118. Takahara, M., et al., The ompA signal peptide directed secretion of Staphylococcal nuclease A by Escherichia coli. J Biol Chem, 1985. 260(5): p. 2670-4.

119. Carlos, J.L., et al., The role of the membrane-spanning domain of type I signal peptidases in substrate cleavage site selection [In Process Citation]. J Biol Chem, 2000. 275(49): p. 38813-22.

120. Carlos, J.L., et al., The role of the membrane-spanning domain of type I signal peptidases in substrate cleavage site selection. J Biol Chem, 2000. 275(49): p. 38813-22.

121. McRee, D.E., XtalView/Xfit--A versatile program for manipulating atomic coordinates and electron density. J Struct Biol, 1999. 125(2-3): p. 156-65.

122. Brunger, A.T., et al., Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr, 1998. 54 ( Pt 5): p. 905-21.

123. Medeiros, A.A., Beta-lactamases. Br Med Bull, 1984. 40(1): p. 18-27.

124. Palzkill, T., et al., Selection of functional signal peptide cleavage sites from a library of random sequences. J Bacteriol, 1994. 176(3): p. 563-8.

109 125. Lessl, M., et al., Dissection of IncP conjugative plasmid transfer: definition of the transfer region Tra2 by mobilization of the Tra1 region in trans. J Bacteriol, 1992. 174(8): p. 2493-500.

110