MOLECULAR INTERACTIONS OF THE Streptococcus mutans SURFACE PROTEIN
PI: CONTRIBUTIONS TO PROTEIN STRUCTURE, STABILITY, AND
TRANSLOCATION














By

TREVOR BRYANT SEIFERT


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA




























Copyright 2005

by

TREVOR BRYANT SEIFERT














ACKNOWLEDGMENTS

I would like to thank all of the members of the Brady lab for sharing their

knowledge and providing technical assistance during my stay in the lab. I am especially

indebted to Dr. Brady for offering to be my mentor and providing me with a new

laboratory and research project after the failure of my initial "mentor". I would also like

to thank my committee members, Dr. Arnold Bleiweis, Dr. Paul Gulig, Dr. Daniel Purich,

and Dr. Robert Bume for their ideas and advice. Finally, I thank my parents for their

wholehearted support and encouragement during these seemingly endless years of study.














TABLE OF CONTENTS



ACKNOW LEDGMENTS..................................................... .......................

LIST OF TABLES....................... ...................... vii

LIST OF FIGURES .................... .... ................................ ii

ABSTRACT................... ... .... ....................... x

CHAPTER

1 INTRODU CTION ..................................................................

Streptococcus mutans and Dental Caries.......................... ........................
Major Surface Protein P1 ............................ ..... ....................2
Proline and Proline-Rich Regions ...........................................
Protein Translocation........................... ............................................................. 7
D naK and RopA ..................... ...... ................ ................11
Summary and Specific Aims ....................................... .. ..................... 12

2 MATERIALS AND METHODS................................. ..... ................14

Bacterial Strains, Plasmids, and Growth Conditions..................................14
Identification of an Intramolecular Interaction Involving the Proline-Rich Region
o f P l .................................................................... ............ ............ ......................14
Purification ofA-region and P-Region-MBP Fusion Proteins..........................14
Enzyme-Linked Immunosorbent Assays (ELISA) to Detect A-region and P-
Region Interaction....................................... ........................ 19
Elimination ofspaP DNA Encoding the A-Region......................................20
Evaluation of Antibody Binding to PI AA......... ....... ................. .......... 21
Assessment of Epitope Restoration by ELISA..................................... ........21
PCR and Construction ofS. mutans spaP and S. gordonii M5 sspA and sspB
Subclones.............................................................................. ...... 22
Purification and Confirmation of GST-Fusion Proteins....................................23
Competitive Inhibition ELISA to Detect A-Region and P-Region Interaction...25
Binding Stoichiometry of the A- and P-Regions by Continuous Variation........25
Analysis of Pl Translocation and the Contribution of the Alanine- and Proline-
R ich R regions .......................................................................... ..... .................26
Introduction ofspaAA into S. mutans PC3370 ................................................26








Analysis ofPIAA Cell Surface Expression in PC3370.....................................27
RNA and Dot Blotting for Confirmation of PAA Expression in PC3370.........27
Western Immunoblot Analysis of Periplasm Extracts from E. coli MC4100
and CK1953 Harboring pUC18, pDC20, pDC9, and pTS20...........................28
Construction of a Bicistronic spaP for Expression of a Discontinuous P1.........29
Evaluation of P1 Fragment Expression by Western Immunoblot.....................30
Evaluation of Surface Expression of Discontinuous P1 in S. mutans ...............31
Introduction of S. gordonii SspA and SspB A-Regions into PIAA....................31
Western Immunoblot Analysis of Chimeric P1 Containing the A-region of S.
gordonii SspA and SspB......................................... ................................... 32
Surface Expression of SspA and SspB in S. mutans PC3370 ...........................32
Involvement of RopA (Trigger Factor) and DnaK in the Maturation and
Translocation of P ......................................... ................................. ......................33
Evaluation of Pl Surface Expression by Whole Cell Dot Blot in the S. mutans
ropA- M utant, TW 90 .............................................. ..............................33
Evaluation of Pl Surface Expression by Whole Cell Dot Blot in S. mutans
SMI2, a Low-Level Expresser of DnaK.....................................................34
Analysis of dnaK Message Levels by Quantitative Real-Time PCR.................34

3 RESU LTS ...................................................................................................................36

Expression of Recombinant P1AA and Recognition by Anti-P 1 Monoclonal
A antibodies ..................................................... ....................................36
Evaluation of P1AA Expression in S. mutans ................... ......................37
Evaluation ofspaP-Specific mRNA in PC3370 Harboring the Deletion
C construct pT S2 1 ....................................................... ................................38
Evaluation of Secretion of P, PIAA, and P1AP in E. coli.................................39
Interaction of the A- and P-Regions by ELISA..............................................39
Restoration of Epitopes by the Interaction of the A- and P-Regions .................40
Inhibition ofMAb 4-10A Binding to P1 by an A- and P-Region Complex........40
Stoichiometry of the A- and P-Region Interaction............................................41
Interaction of Pl, SspA, and SspB A- and P-Regions.................................41
Introduction of the A-Regions of SspA and SspB into P1AA...........................42
Stability and Translocation of Chimeric PI Containing the A-Regions of
SspA and SspB ........................................................... ........................43
Evaluation of the Involvementof SecB in the Secretion of P1, PIAA, and
P A P in E coli....................................................................................... 44
Expression of Discontinuous P1 and Recognition by Anti-P 1 Monoclonal
A ntibodies................................................................... ............................45
Evaluation of Surface Expression of Discontinuous PI in S. mutans ...............47
Surface Expression of Pl in a RopA-Deficient S. mutans and Bacterial
Adherence to Salivary Agglutinin .......................... .............................47
Analysis of P1 Surface Expression in a Low-Level DnaK Expressing S.
m utans M utant ............. .......................................... ......... ......................49
Evaluation of dnaK mRNA Expression in S. mutans PC3370 Harboring
pDL289, pMAJJ8, pMAD, and pTS21 .........................................................49








4 DISCUSSION AND CONCLUSIONS ......................... .. .................78

Identification of an Intramolecular Interaction within P1 ........................................79
Analysis of P1 Translocation and the Contribution of the Alanine- and Proline-
R ich R egions .......................................................... ...........................................85
Involvement of RopA (Trigger Factor) and DnaK in the Maturation and
Translocation of P1 ............................ ............... 93
Conclusions ............. ...................................................... ........... 97

LIST OF REFERENCES................................ .......... ...............................99

BIOGRAPHICA L SKETCH ...................................................... ...................... ... 115







































vi















LIST OF TABLES

Table page

1. Bacterial Strains............................................................................ .......................... 15

2. Plasm ids ........................................................................................................................16

3. PCR Prim ers............................................................................... .................................. 18















LIST OF FIGURES


Figure page

1. Schem atic representation of P .................................................... ....................... 3

2. Western blot analysis of Pl and recombinant P1 lacking the A-region.......................51

3. Lack of surface expression of Pl devoid of the A-region.. .........................................52

4. RNA dot blot analysis ofspaP-specific mRNA levels in the Streptococcus mutans
spaP-negative mutant PC3370 and derivatives....................................... .....53

5. Western immunoblot of cytoplasm and periplasm fractions ofE. coli DH5a
harboring pUC18 derived plasmids expressing full-length PI, PIAP, PIAA, and
vector alone. ................................................. ......... ..... .....................................54

6. Demonstration of A-region and P-region interaction by ELISA..................................55

7. Restoration of epitopes by A- and P-region interactions as measured by ELISA........56

8. Inhibition of anti-PI MAb 4-10A binding to immobilized P1 in ELISA...................57

9. Stoichiometry of the mAb 4-10A epitope .......................... ... .............................58

10. Demonstration of similar level of mAb 3-8D reactivity to A-region-GST fusion
polypeptides of SspA, SspB, and P1 by Western immunoblot................................59

11. Demonstration of interactions between the A- and P-regions of different antigen
I/II proteins............... ..... .............................................. ......................................60

12. Evaluation of reactivity of A- and P-region dependent anti-PI mAbs with P1,
SspA and SspB .................... ............. ......................................................... 61

13. Restoration of epitopes by the interaction of various antigen 1/II A-regions with
the immobilized P-region of Streptococcus gordonii SspA...................................62

14. Restoration of epitopes by the interaction of various antigen I/II A-regions with
the im mobilized P-region of SspA. ................... ................. ........................ 63








15. Restoration of epitopes by the interaction of various antigen I/II A-regions with
the immobilized P-region of SspA. ..................... ................................64

16. Restoration of epitopes by the interaction of various antigen 1/II A-regions with
the immobilized A-region of SspA. ....................... ...............................65

17. Restoration of epitopes by the interaction of various antigen III A-regions with
the immobilized A-region of SspA. ..................... ...............................66

18. Restoration of epitopes by the interaction of various antigen I/II A-regions with
the immobilized A-region of SspA. ....................... ..............................67

20. Western immunoblot of chimeric P1 containing the A-regions of SspA and SspB...69

21. Surface expression of S. gordonii SspA and SspB in S. mutans PC3370. ................70

22. Demonstration of lack of ability of heterologous A-regions to restore surface
expression of P1AA in PC3370......................................... ......................... 70

23. Western immunoblots of cell lysates ofPC3370 harboring vector alone and
plasmids encoding PI, P1 AA P1 AA + SspA A-region, and P AA + SspB A-
region. ........................................ ..... ...... ...................... .......................71

24. Western immunoblot of cytoplasm and periplasm fractions of E. coli MC4100
(wild-type) and CK1953 (AsecB) harboring pDC20 (PI)........................................72

25. Schematic representation of discontinuous PI. ......................... ....................73

26. Western immunoblot of Pl fragments expressed from pTS30 in E. coli ...................73

27. Whole cell dot blot of S. mutans NG8 and PC3370 harboring pDL289 vector
control, pMAD encoding P1, and pTS31 encoding discontinuous P1 fragments....74

28. P1 surface expression levels ofS. mutans UA159 and TW90 (ropA- mutant) at
early log stage traced with mAb 4-10A as measured by densitometry ..................75

29. P1 surface expression levels of S. mutans UA159 and SM12 (DnaK-deficient) at
early log phase............................................. .........................76

30. Real-Time PCR quantification of dnaK mRNA from S. mutans PC3370 harboring
the pDL289 vector alone and expressing PlAP-region, full-length P1, and
P A A -region ..................... ......................................................................................77














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

MOLECULAR INTERACTIONS OF THE Streptococcus mutans SURFACE PROTEIN
PI: CONTRIBUTIONS TO SURFACE STRUCTURE, STABILITY AND
TRANSLOCATION

By

Trevor Bryant Seifert

May 2005

Chair: L. Jeannine Brady
Major Department: Oral Biology

Streptococcus mutans is considered to be the major etiologic agent of human dental

caries. Attachment of S. mutans to the tooth surface is required for the development of

caries and is mediated, in part, by the 185 kD surface protein variously known as antigen

I/II, PAc, and P1. Such proteins are expressed by nearly all species of oral streptococci

and have been identified as possible antigens for vaccine development. In addition,

researchers are utilizing P1 to study immune response and immunomodulation .

The goal of this research was to identify intramolecular interactions within P1 and

to examine their contributions to P1 structure, stability, and translocation. To that end,

this research demonstrates a) that several anti-PI monoclonal antibodies (mAbs) require

the simultaneous presence of the alanine-rich and proline-rich regions for binding, b) that

the proline-rich region of P1 interacts with the alanine-rich region, c) that like the

proline-rich region, the alanine-rich region is required for the stability and translocation








of PI, d) that both the proline-rich and alanine-rich regions are required for secretion of

P1 in E. coli, and e) that in E. coli P1 is secreted in the absence of SecB. Additionally, it

was demonstrated that the chaperone RopA (trigger factor) was not required for P1

translocation. However, its absence resulted in reduced P1 mediated adherence to

salivary agglutinin, suggesting a role in PI maturation. DnaK was also shown to be

involved in P1 translocation and dnaK mRNA levels were affected by the presence of P1

deletion constructs. Furthermore, the A- and P-regions of P1 were shown to be capable

of interacting with the A- and P-regions of the Antigen I/II proteins, SspA and SspB,

from Streptococcus gordonii. These interactions restored epitopes recognized by several

anti-Pi mAbs. Replacing the A-region of P1 with the A-regions of SspA and SspB also

restored some mAb binding, but did not restore stability and translocation of Pl to the

cell surface. The results of this research have implications for understanding surface

localization of virulence factors in pathogenic microorganisms and for understanding

how the protein structure of a vaccine antigen contributes to recognition by antibodies.













CHAPTER 1
INTRODUCTION

Streptococcus mutans and Dental Caries

Streptococcus mutans is considered to be the major etiological agent of human

dental caries [1, 2], one of the most common infectious diseases that affect humans. S.

mutans was first identified in a human carious lesion in 1924 and isolates were later

divided into eight serotypes designated a through h based on differences in cell wall

carbohydrate composition [3, 4]. Concurrent DNA hybridization studies further

categorized the serotypes as four genetic types based upon the guanine plus cytosine

(G+C) content of their genomes [5, 6]. The four genetic types were subsequently

classified as different species, Streptococcus critus serotypee a), Streptococcus rattus

serotypee b), Streptococcus sobrinus (serotypes d, g, h), and S. mutans (serotypes c, e, ).

These species are collectively known as mutans streptococci [7]. S. mutans serotype c is

the most common mutans streptococcus isolated from human dental plaque [1, 2].

S. mutans is equipped with several proteins that enable its attachment and

subsequent colonization of the tooth surface. In the presence of sucrose, extracellular

glucosyltransferases (GTF) synthesize several forms of branched extracellular glucans.

These glucans provide a matrix for the aggregation ofS. mutans and other oral

streptococci through interaction with proteins such as the bacterial surface-localized

glucan-binding proteins (GBP). S. mutans possesses four GTF genes, gtfA [8], gtfB [9],

gtfC [10], and gt/D [11], and three GBP genes, gbpA [12], gbpB [13], and gbpC [14].

Mutational inactivation of the GTF genes has shown that their products are important to








cariogenecity. However, a model for colonization of the tooth surface by S. mutans

suggests that initial attachment to the tooth pellicle is protein-mediated followed by

glucan-dependent bacterial accumulation [15]. The surface proteins that are implicated in

the initial adherence of S. mutans are members of the antigen 1/II super-family of

multifunctional adhesins and are variously known as antigen I/II [16], Ag B [17], IF [18],

P1 [19], SR [20], and PAc [21], and are encoded by the genes spaP orpac. Antigen I/1I-

like molecules are expressed in nearly all of the oral streptococci [22] and include SpaA

[23]and PAg [24] from S. sobrinus, SSP-5 from Streptococcus sanguis [25], and SspA

[26] and SspB [27] from Streptococcus gordonii.

Major Surface Protein P1

The genes spaP and pac have been cloned and sequenced [21, 28, 29]. N-terminal

amino acid sequencing of the proteins and the predicted amino acid sequences indicate

that the spaP and pac gene products differ by only 36 residues. Major characteristics of

the MR-185,000 P1 include a 38 residue amino-terminal signal sequence, a region

containing three 82-residue alanine-rich repeats, a 150 residue variable region in which

20 of the 36 aforementioned amino acid substitutions reside, a central proline-rich region

containing three 39-residue tandem repeats, carboxy-terminal wall- and membrane-

spanning regions, and an LPXTG wall anchor motif [28] (Fig. 1). Secondary structure

predictions of P1 based upon the sequence ofspaP indicate that the alanine-rich region

would form an a-helicle coiled-coil structure while the central proline-rich region would

form an extended 3-sheet structure [28]. Recently the variable region of the S.mutans

serotypefantigen I/n was subcloned, and its crystal structure determined. The crystal

structure data indicate that the variable region forms a flexible P-sandwich [30].










Wall M.&Nrasu Cytoplatm
sp.anni Spain Tla
Rgin. Regin (a.- 1557-1561)
Sinpal Se&q A-REGION V REGION P-REGION (a.. 1486-1535) (a. 1536-1556)
(U 1-3) (Lt 16-464) (..a679-I23) (a... 40 -93)



Figure 1. Schematic representation of P1



Antigen I/II polypeptides are structurally complex and exhibit diverse binding

properties, which mediate interactions with a variety of substrates including host salivary

agglutinin, fibronectin, fibrinogen, and collagen [31, 32]. Several regions have been

implicated in the binding activities of antigen I/II polypeptides. Brady et al [33] provided

evidence that P1 possessed multiple sites contributing to salivary agglutinin binding and

that different regions might be involved in binding to soluble or immobilized salivary

agglutinin. Later, Scatchard analysis of antigen I/II binding to saliva-coated

hydroxyapetite showed the binding to be mediated by two sites [34]. Investigators have

showed that recombinant peptide fragments derived from the A-region bound salivary

agglutinin [35] or salivary glycoproteins [36] and Senpuku et al. (1995) demonstrated

that antibodies specific to a peptide fragment derived from PAc aa residues 200-481

inhibited the binding of fluid-phase salivary components to immobilized PAc.

Furthermore, an antigen I/II peptide fragment consisting of a.a. residues 816-1213

blocked S. mutans cell adhesion to saliva-coated hydroxyapatite [37] and Kelly et al. [38]

found antigen I/II derived peptides consisting of residues 1005-1044 and 1085-1114 to be

inhibitory to S. mutans adhesion to salivary glycoproteins.








Proline and Proline-Rich Regions

Proline is unique among the amino acids in that its side chain is covalently bound

to the backbone amide. As a result of this unusual bond, the proline residue has a

restricted backbone conformation [39], the bulkiness of its side chain restricts the

conformation of the preceding residue [40], and the proline is unable to act as a hydrogen

bond donor. Proline is recognized as an a-helix and P-sheet breaker and is often located

one or two residues C-terminal of an a-helix [41]. A sequence of four or more prolines

in a row adopts the conformation of an extended structure with three residues per turn,

known as a polyproline II helix [42]. The polyproline II helix is a major structural

element in collagen, pancreatic polypeptides, and neuropeptides [43]. In a survey of

surface proteins, there were 90 occurrences of polyproline II helixes in 80 non-

homologous proteins [44].

Proline-rich regions are biologically important in numerous unrelated proteins in a

variety of organisms, both eukaryotic and prokaryotic. Although the functions of

proteins containing proline-rich regions can be quite diverse, the roles of the proline-rich

regions appear to be fairly conserved: protein-protein interaction, folding, and structure.

In eukarotes, anandamide amidase is responsible for the hydrolysis of neuromodulatory

fatty acid amides and esters [45]. The enzyme contains a nine residue proline-rich

region, which upon removal resulted in loss of enzymatic activity and a change in

subcellular localization of the enzyme. It is suggested that the proline-rich region may be

essential for the correct folding of the amidase, although similar proline-rich regions are

not found in non-mammalian amidases [46]. Involvement in cellular localization has also

been demonstrated for the proline-rich region in the eukaryotic enzyme dynamin.








Okamoto et al. has shown that a proline-rich domain is involved in the enzyme's

localization to coated pits and in its GTP-binding activity [47].

The relevance of proline-rich domains to the field of medicine can be seen in the

human disease Liddle syndrome. Liddle syndrome is a disease of the amiloride-sensitive

epithelial sodium channel [48]. The sodium channel is composed of three subunits, two

of which contain highly conserved C-terminal proline-rich domains. Frameshift

mutations resulting in the deletion of at least 45 residues from the C-terminal segments of

the subunits have been the identified causes of this disease. More recently, a missense

mutation that results in the substitution of a leucine for a proline (P616) in the conserved

proline-rich domain of subunit 0 has been identified and correlated with the disease. The

substitution has greater effect on channel activity than the deletion of the complete C-

terminal segment of both the 0 and y subunits. It is suggested that this proline residue is

involved in an essential interaction with another protein and possibly another subunit

[49].

Proline-rich regions are also involved in protein-protein interactions between

organisms. In the world of virology, the transformation of primary B lymphocytes with

Epstein-Barr virus (EBV) is known to be dependent upon the expression of the EBV

nuclear protein 2 (EBNA2) [50]. The 483-residue EBNA2 contains a 36 residue proline-

rich region in the N-terminal third of the protein. Of the 230 residues of the N-terminal

half of EBNA2, 222 were not essential for transformation ofB lymphocytes. The eight

essential residues are seven prolines and a glutamine, and it is suggested that they may

constitute a critical domain for structure or intramolecular interaction [51]. Lastly, an

interaction between a proline-rich insect peptide and a molecular chaperone in bacteria








has been identified. Pyrrhocoricin, an antibacterial peptide originally isolated from the

European sap-sucking bug Pyrrhocoris apterus [52], kills sensitive species by binding to

the bacterial DnaK [53].

Previously, in an attempt to define a role for the P-region in the adhesive function

of P1, an internal deletion, P1AP (A826-996), was constructed [54]. The proline-rich

region (P-region) is highly conserved among the antigen 1/II family of oral streptococcal

proteins, and similar highly repetitive proline-rich sequences have been identified in a

wide variety of bacterial proteins [55-64]. Homology to the P-region of Pl is found in

numerous surface proteins in both prokaryotes and eukaryotes. Among these are the

fibronectin binding proteins of Streptococcus pyogenes and Staphylococcus aureus [65,

66], an immunogenic secreted protein (isp) ofS. pyogenes [67], and the virulence

associated surface protein, PspA, of S. pneumonia [68]. The internally deleted

polypeptide P1AP was expressed in both E. coli and in S.mutans PC3370, an isogenic

spaP-negative mutant. Western blots of PIAP expressed in E. coli revealed a loss in

reactivity for fiveofeleven Pi-specific MAbs. These five mAbs also did not react to a

subclone of the P-region (826-996), suggesting that they recognize a complex P1 epitope

that is dependent upon the presence of the P-region. Although P1AP contains the signal

sequence, it was not translocated to the surface ofS. mutans PC3370 (spaP). Also, in

comparison to full-length P1 expressed from pDL289, only low levels of PIAP were

detected in the cytoplasm of PC3370, while mRNA levels were equivalent. These data

suggest that the P-region may be required for P stability and subsequent translocation to

the cell surface.








Protein Translocation

Since P1AP possessed the N-terminal signal sequence and C-terminal cell wall

anchor motif, the lack of P1AP expression on the cell surface was unexpected. Proline-

rich regions are known to be involved in a variety ofintra- and intermolecular protein-

protein interactions [69-78], including chaperone-like activities. Wang et al. [79]

identified a centrally located proline-rich region in the serine protease, Factor C.

The Limulus polyphemus (horseshoe crab) Factor C is a 132-kDa secreted serine

protease and contains a centrally located proline-rich region. The role of the proline-rich

region in the secretion and function of Factor C was investigated through the construction

and expression of homologues with and without the proline-rich region. The proline-rich

region is flanked by an amino-terminal lectin binding domain and a carboxy-terminal

protease domain. Factor C is 1019 amino acids in length, and the proline-rich region

spans residues 630 690. Deletion of either the lectin binding or protease domains

resulted in peptides that were stably expressed and secreted by the cell. In contrast,

deleting the proline-rich region resulted in a protein that was found in the cytoplasm but

was no longer secreted. In addition, expression of a truncated peptide consisting of the

amino-terminal 329 residues of Factor C was stable and secreted while a peptide

consisting of the secretion signal fused to the carboxy-terminal protease domain was not

secreted. Interestingly, the addition of the proline-rich region amino-terminal to the

protease domain restored secretion of the peptide, but fusing the proline-rich region

carboxy-terminal to the protease domain did not restore secretion.

To establish a role for the proline-rich region in the folding of Factor C, a partial

trypsin digestion analysis was performed. Trypsin cleaves preferentially at unfolded

regions in proteins, and although there are nearly 100 potential trypsin cleavage sites in








Factor C, four bands were detected after partial digestion. The visible fragments were

attributed to compactly folded domains. However, Factor C without the proline-rich

region was not detectable after 10 minutes of trypsin digestion indicating that the protein

was in an unfolded trypsin-susceptible conformation. This study revealed that the

proline-rich region is essential for the stability and secretion of Factor C. Because this

effect was similar to that previously described for the P-region of P1 [54] these authors

suggested that internal proline-rich regions may act as intramolecular chaperones for

correct folding and secretion of proteins that contain them. The homology of the P-

region of P1 to the surface proteins of numerous organisms and its conservation within

oral streptococci suggest that it plays an important role in PI, and considering the

prevalence of proline-rich regions in protein-protein interactions, it is likely that the P-

region is involved in such an interaction.

Recently, Van Dollarweed et al. (2003) demonstrated that the P-region of Pl binds

to a polypeptide fragment of P1 that contains the alanine-rich region (A-region). This

interaction restored the reactivity of a P specific Mab that was not reactive to either of

the fragments individually and suggests that these regions interact in mature, surface

expressed, P1. X-ray crystallography has revealed that the variable region of PI forms a

flexible beta-sandwich that would place the P-region and A-region into close proximity

[30]. Given that the P-region is required for the native structure, stability, and

translocation of Pl and that the P-region interacts with a fragment of P1 containing the

A-region, it is likely that the A-region may also play a role in the structure, stability, and

translocation of Pl.








To fully elucidate the role of the P-region in P translocation, a better

understanding of the molecule's route of translocation represents an important goal.

There is no experimental data that identify the secretion pathway P1 or antigen I/II-like

proteins use. However, based upon the method of Pl cell wall anchoring, a route of

translocation has been predicted. Gram-positive surface proteins containing the

conserved C-terminal LPXTG motif, such as PI, are anchored to the cell wall by the

membrane anchored transpeptidase, sortase. During cell wall anchoring, sortase cleaves

surface proteins between the threonine and the glycine of the LPXTG motif [80].

Following cleavage in S. aureus and Listeria monocytogenes, the proteins are linked to

cell wall peptides via an amide bond [81, 82]. Although several aspects of peptidoglycan

structure in Gram-positive bacteria are variable [83], the principles of surface protein

anchoring appear to be conserved [84]. Lee and Boran[85] identified and insertionally

inactivated the gene encoding sortase, srtA, in S. mutans. As predicted, SrtA mutants

secreted P1 into the supernatant demonstrating that P1 is indeed a sortase anchored

protein. Current evidence suggests that sortase anchored proteins are translocated via the

Sec translocase [86].

The Sec-dependent secretion pathway has been well characterized and studied in E.

coli and to a lesser extent in B. subtilis. In E. coli, the Sec-translocase consists of SecA,

SecY, SecE, SecG, SecD, SecF, and YajC [87]. Two major targeting pathways converge

on the Sec-translocase, the signal recognition particle (SRP) pathway and the SecB

pathway. The E. coli SRP consists of a 4.5s RNA and the GTPase Ffh, both of which are

required for cell viability [88]. Signal peptides of nascent polypeptides are recognized by

the SRP as they emerge from the ribosome [89]. SRP binding stalls translation and








targets the SRP-ribosome complex to the SRP receptor, FtsY [90, 91]. The complex is

then targeted to the Sec-translocon where the ribosome docks and translation is restored.

The preprotein is cotranslationally translocated across the membrane via an integral

membrane complex consisting of SecY, SecE, and SecG. The ATPase SecA provides

energy for the translocation [87]. In the case ofposttranslational secretion, the

cytoplasmic chaperone SecB targets preproteins to the Sec-translocon. SecB binds to

nascent and full-length preproteins as they emerge from the ribosome [92]. SecB

interaction prevents premature folding of the preprotein and delivers it to the Sec-

translocon in a secretion-competent state. Binding of the SecB-preprotein complex with

SecA results in the transfer of the preprotein to SecA and the release of SecB [93]. The

preprotein is subsequently translocated across the membrane through the Sec-translocon

[94].

The SRP pathway exists in both gram-negative and gram-positive bacteria.

Identified homologs of the Sec-dependent pathway components in B. subtilis include

SecA, SecYEG, SecDF, YrbF, Ffh, and scRNA. As the genome sequences of gram-

positive bacteria have become available, investigators have searched for homologs of

SecB to no avail. However, a B. subtilis complementation study of an E. coli SecB null

mutant revealed a functional ortholog, CsaA, with partially overlapping binding

characteristics [95-97]. As previously stated, the SRP is essential for viability in E. coli,

and this was assumed to be the case in all organisms. However, an Ffh null mutant in S.

mutans is viable, and P1 is translocated and expressed on the cell surface [98]. This

suggests that ifPl secretion is Sec-dependent, the targeting pathway is likely to be SecB-








like and may require a SecB ortholog or possibly an unrelated chaperone with similar

functions.

DnaK and RopA

The translocation of P1 to the cell surface in an S. mutans mutant devoid of the

SRP pathway [98] would suggest that P1 is post-translationally secreted, and a protein of

1561 residues would most certainly require interactions with chaperones to prevent

misfolding and aggregation while transiting the cytosol. The 70-kD heat shock proteins

(Hsp70s) are ubiquitous proteins found in the bacterial cytosol and several compartments

of eukaryotic cells including the endoplasmic reticulum, the mitochondria, and the

cytosol [99]. The E. coli Hsp70, DnaK, has been extensively studied and is involved in a

variety of cellular processes, including both protein folding and degradation. In studies

of substrate specificity it has been shown that DnaK preferentially binds to peptides that

contain hydrophobic residues [100]. In proteins these hydrophobic residues are typically

found in the core of the folded structure, or in subunit interfaces [101]. Nascent

polypeptides emerging from the ribosomes, as well as malfolded proteins, display short

hydrophobic regions that are not exposed in the protein's native conformation. DnaK

binds to these exposed hydrophobic segments, thereby preventing aggregation and further

misfolding.

Another chaperone that interacts with nascent polypeptides is trigger factor, a

ribosome-associated peptidyl-prolyl cis-trans isomerase (PPlase). In S. mutans trigger

factor is known as RopA. Trigger factor associates with the large ribosomal subunit at

the peptide exit channel and binds to nearly all nascent polypeptides [102]. There is

evidence that trigger factor cooperates with DnaK to promote the folding of a variety of

cytosolic E. coli proteins [103] and that they share substrates and binding specificities








[104]. In fact, DnaK is not recruited to translating ribosomes that lack trigger factor

[105]. Besides its chaperone activities, trigger factor can catalyze cis-trans isomerizaion

ofpeptidyl-prolyl peptide bonds. The PPIase activity of trigger factor is not required by

all of the proteins that require trigger factor for proper folding, however the PPIase

activity is essential for some [106]. In S pyogenes, the PPIase activity of trigger factor

influences the conformation of the nascent cysteine protease, SpeB, which in turn directs

the protease into one of several alternative folding pathways[107]. The malfolded

proteases are subsequently not targeted to the secretion pathway.

Summary and Specific Aims

In an effort to characterize the role of the proline-rich region of P1 in the adherence

properties of the molecule, Brady et al. (1998) deleted the region from P1 (P1AP). While

PIAP retained the sequences believed sufficient for expression and translocation,

unexpectedly, it was unstable and not translocated to the cell surface. As there is a lack

of research regarding protein translocation in gram-positive organisms and Streptococcus

in particular, it was of interest to identify the role of the proline-rich region in P1 stability

and translocation. Since proline-rich regions are known to be involved in intra- and

intermolecular protein-protein interactions, the first specific aim of this work was to

identify regions within P1 that interact with the proline-rich region. Once a proline-rich

region interaction was discovered, the second specific aim of this study was to analyze

the role of the interacting region in the structure, stability and translocation of P1.

P1 is a large and structurally complex molecule as is evident by the change in

antibody reactivity seen against P1AP that suggests complex and possibly conformational

epitopes. Further evidence of the structural complexity of PI was revealed in the solved





13


crystal structure of the variable region [30]. Based upon the surface expression of PI in

S. mutans lacking the SRP pathway [98] and the presumption that sortase-anchored

proteins are secreted via the Sec translocase, the successful post-translational

translocation of a large and complex molecule, such as P must be dependent upon

chaperones. The final specific aim of this work was to examine whether the chaperones

DnaK and RopA contributed to P1 translocation or function.













CHAPTER 2
MATERIALS AND METHODS

Bacterial Strains, Plasmids, and Growth Conditions

Bacterial strains used in this study are listed in Table 1, and all plasmids used are

listed in Table 2. Unless otherwise noted, all S. mutans strains were grown under

anaerobic conditions at 370C in Todd-Hewitt broth (BBL, Cockeysville, Md.)

supplemented with 0.3% yeast extract (THBYE) and kanamyacin (500 pg/ml) as needed.

E. coli strains were grown aerobically at 370C with vigorous shaking in Luria-Bertani

broth (LB) (1% [wt/vol] tryptone, 0.5% [wt/vol] yeast extract, 1% [wt/vol] NaCI, pH 7.0)

supplemented with ampicillin (100 gg/ml) or kanamycin (50 gg/ml) as appropriate. E.

coli strains MC4100 and CK1953 were grown aerobically at 370C with vigorous shaking

in M9 medium (0.625% [wt/vol] Na2HPO4, 0.075% [wt/vol] KH2PO4, 0.2% [wt/vol]

NaCI, 0.028% [wt/vol] MgSO4, 0.1% [wt/vol] (N H. i: S:,, 1% glucose) supplemented

with kanamycin (50 pg/ml) and ampicillin (100 ig/ml) as appropriate.

Identification of an Intramolecular Interaction Involving the Proline-Rich Region of
P1

Purification of A-region and P-Region-MBP Fusion Proteins

Overnight cultures of E. coli harboring pMA3 [54] or pMA41 [35] (Table 1) were

diluted 1:100 into fresh Luria-Bertani (LB) broth containing 100 g/ml of ampicillin and

grown to an OD6o of 0.5. The medium was supplemented with 0.3 mM isopropyl-b-D-

thiogalactopyranoside (IPTG), and the culture was incubated for an additional 2 hours at

370C Periplasmic contents were extracted by osmotic shock [108]. Affinity

















0



*0
C



80)
2

N C.- 4
i. 'i

^ C
~se





2 I
e' 0
EgM

st s
G st
F T<


I






0 ,








00
cr,







oo
0

5

K

a. IP. C. d )





SS> Smmo
a $ agg
slll l,^ l ^"~
u*|| 55o


a <5ooo

- m2m zma
W)
















00
(_)



a2

I)
8 ^



o1 E0








a -0- n. L- *
I 2'a 'I1 s o" a o*


E u
"E 7F:, a
S 0 0 I 0" o










1>1
I~ ~ ~ ~~' II 9 T> S .IsI



-0 .0C,
00u E 00
< F.

4S < 0-3 S i < 3
.gia 4)a ,, "S t












4))
tn.. m m CS' 0. I -a. m 4) -.
a " sR a 11,X 3 g
| i E' 2
-QL) Oo.o o. r B r t


Igsg"' g~~g,-. "" ^ g- s-~r
s 1 . 1 1 1 1.
SU 5" 0 0 20
1 30 8





w 0-0
.o S2




04zo- 5 < &<0< 0
0-j ^ s" a c3 l. 9 9 xx x x c x~ .
0 d 'o o -. El~! caq~'~E




^a i g|^ O a I P@ e B
= E>0. 0. a0. 0. 0. >0 0. 0 0. 0. 0.


. s



(0. C"
o8~rL O: 8ja ~ 00 ~,
m~ Ul ~ a
0 a 0.0 0. 0.. 0. 0. 00 0. 0. 0

H

















0







z z
0
Q 0









0
0 E
ti '0 4) '0 '0 03
o a a 3 3 8
O 2 2 .2 .2 .2 5







< 1



zo
c c _@.
021



oo o
0 0 < 0 <
< <










0 c
.o -a M S 0 510
e~ .E '3 "



I U U Q U



00 0
F- M F H

1=: 1 *ill.a c-l l l
BO M 0O " n O


c c T3 3 t 'S '6a '8 *C C qg-

4'' c B cC

.& '" ," o M o 1 u .' u0 ~Di : o \o
Q


Q~ B. Q.l O. D. Q OW C W QU o.





c 0

*s)
"O. Q. .~ "' O 0C. 0.0
'a S
0 -L
o0.o .'0000'
d ~,3~''










-~ c1 c ;5 : S.

C2, CI :I 4 P
:Z)



k a.






gS 0
~u


0s



u


H '-, I? <; u .
o<< u

"< u F.
"~~~ ~~~ 5I-

-


uU U
^IblB^^lEl^ .




uuuFu UU : I
L Cd) Cd)









u <0
^i S~~3l^|^ =98^
< oo
u EO .o~
E-I-u -- < I u--- _-
S < < < u u u
e u<:uuuuouou-d-c--.j -..- j guUU U CD<<<<*<*- -''i :*i : '0( u
~HH

0 H



u

0 E
jH g- i k0 E- os F-4 F. F- rN F - -' co F-Is ' ciiE--
U "S S SS S S SgN
~~ ~r~~~~~~~a








purification of the fusion proteins was performed by passage of the periplasmic fractions

through a column of amylose resin (Bio-Rad) and elution with 10 mM maltose by a

standard protocol [108]. Purified fusion proteins were quantified using the bicinchoninic

acid (BCA) protein assay kit (Sigma) with bovine serum albumin as the standard.

Enzyme-Linked Immunosorbent Assays (ELISA) to Detect A-region and P-Region
Interaction

Binding of the A-region to the P-region was measured by ELISA. Sample wells of

Costar High Binding plates (Coming Incorporated, Coming, N.Y.) were coated overnight

at 40C, in triplicate, with 100 pl of 0.1 M carbonate-bicarbonate buffer (pH 9.6)

containing 0.02% sodium azide and 100 ng of purified maltose binding protein (MBP),

A-region-MBP, or P-region-MBP. Coating buffer and unbound antigens were removed

from the ELISA plate wells, and unreacted sites were blocked with PBS-Tw and

overnight incubation at 40C. Plates were washed four times with PBS-Tw. Purified A-

region-MBP, P-region-MBP, and MBP were two-fold serially diluted in PBS-Tw and

added to the wells, beginning at 1000 ng/well. The plates were incubated overnight at

40C and washed four times with PBS-Tw. A-region-specific MAb 3-8D [35] or rabbit

anti-MBP Ab (NEB, Beverley, MA) was added to the wells at a 1:1000 dilution. Plates

were washed with PBS-Tw and peroxidase-labeled goat anti-mouse IgG or goat anti-

rabbit Ig (Cappell) were added to the wells at a 1:1000 dilution. After washing, 100 pl of

0.01 M phosphate citrate buffer (pH 5.0) containing 0.1 M o-phenylenediamine

dihydrochloride and 0.012% (vol/vol) hydrogen peroxide were added to each well. Plates

were incubated for 30 min at room temperature, and the absorbance at 450 nm was

recorded by using an MPM Titertek model 550 ELISA plate reader (Bio-Rad).








Elimination of spaP DNA Encoding the A-Region

Fragments of spaP both upstream and downstream of the A-region were amplified

by polymerase chain reaction (PCR) and subsequently ligated together to create spaPAA.

Fidelity of the reactions was confirmed by restriction and sequence analysis. Forward

primer TS9k and reverse primer TS8 were used to amplify spaP DNA upstream of the A-

region, including the spaP promoter. Forward primer TS7 and reverse primer TSl0k

were used to amplify spaP downstream of the A-region. Primers TS9 and TS10 contain

engineered KpnI restriction sites. Primers TS7 and TS8 were engineered with single base

changes that introduce silent mutations, which upon ligation of the PCR products produce

a unique SfoI restriction site. Reactions were carried out in a UNO thermoblock

thermocycler (Biometra, Tampa, FL) with plasmid-encoded spaP, pDC20 [54] as the

template and VENT polymerase (NEB) under the following conditions fo 30 cycles:

denaturation at 940C for 30 seconds, primer annealing at 530C for 30 seconds, primer

extension at 720C for 1 minute or 3 minutes and 30 seconds; and final extension at 720C

for 7 min. The resulting 727 and 3,568 bp gene fragments were ligated together and

cloned into the SmaI site ofpUC18, creating pTS20, which was introduced into E. coli

DH5ac by electroporation. Clones were screened on LB agar supplemented with 100

ig/mL ampicillin and 0.75 gg/mL X-gal (5-bromo-4 chloro-3 indolyl-P-D-

galactopyranoside). White colonies were picked and tested for the presence ofspaPAA

insert DNA after alkaline lysis. Sequences of all recombinant constructs were confirmed

by the DNA sequencing core (University of Florida).








Evaluation of Antibody Binding to P1AA

E. coli DH5a harboring pTS20 or pDC20 were grown for 16 hours at 370C,

harvested by centrifugation, and lysed by boiling for 5 minutes in SDS-sample buffer

(4% [wt/vol] sodium dodecyl sulfate [SDS], 2% [vol/vol] 2-mercaptoethanol, 20%

[vol/vol] glycerol, 125 mM Tris-HC1 [pH 6.8], 0.1 mg ofbromophenol blue per ml).

Proteins were separated by SDS-polyacrylamide gel electrophoresis on 7.5% (vol/vol)

acrylamide preparatory gels by the method of Laemmli [117]. Proteins were

electroblotted onto nitrocellulose membrane (Schleichter and Schuell, Keene, N.H.) for 1

h at 100 V by the method of Towbin et al. [118]. Immunoblots were blocked with PBS-

Tw and cut into 0.5 cm strips. Strips were incubated with anti-Pt MAbs [119, 120] at

dilutions of 1:1000 in individual troughs of an Incutray (Schleichter and Schuell). After

washing, strips were incubated in peroxidase-labeled goat anti-mouse IgG (Cappel) and

developed with 4-chloro-l-naphthol solution (7 ml of PBS, I ml of 4-chloro-1- naphthol

[Sigma; 3 mg/ml in ice-cold methanol], and 8 pl of 30% [vol/vol] hydrogen peroxide).

Assessment of Epitope Restoration by ELISA

Sample wells of Costar High Binding plates were coated as before with 500 ng of

purified maltose binding protein (MBP), A-region-MBP, or P-region-MBP. Following

blocking and washes as previously, purified A-region-MBP, P-region-MBP, and MBP

were serially diluted two-fold in PBS-Tw and added to the wells, beginning at 500

ng/well. The plates were incubated overnight at 40C and washed four times with PBS-

Tw. MAbs 4-10A, 5-5D, and 6-11A were added to the wells at a 1:1000 dilution.

Binding of the MAbs was traced with peroxidase-labeled goat anti-mouse IgG at a








1:2000 dilution, the plates were developed, and the absorbance was measured as

previously described.

PCR and Construction of S. mutans spaP and S. gordonii M5 sspA and sspB
Subclones

The P-regions ofS. gordonii M5 sspA and sspB and several regions ofspaP were

amplified by PCR and cloned into the pGEX-4T-2 vector (Amersham Biosciences) for

expression as fusion polypeptides with glutathione S-transferase (GST). Forward and

reverse primers were designed based upon the published sequences ofsspA and sspB

(accession numbers U40025 and U40026 respectively) and the unpublished sequence of

NG8 spaP. The primer sequences and engineered restriction sites are shown in Table 3.

The following primers were used in these amplifications: sspA P-region (a.a. 808-1008) -

TS41 and TS42, sspB P-region (a.a. 749-942) TS42 and TS43, spaP A-region (a.a. 179-

466) TS24 and TS25, and spaP P-region (a.a. 816-1016) TS28 and TS29. These

primers were engineered with BamHI and Sall restriction sites to enable subsequent

cloning into the pGEX-4T-2 expression vector.

PCR was performed for 30 cycles under the following conditions: denaturation at

950C for 3 min; primer annealing at 51 C for 30 sec; and primer extension at 720C for 40

sec. Final primer extension was carried out for an additional 7 min after the last cycle.

The amplified PCR products of correct predicted size were cloned into the T/A cloning

vector pCR2.1-TOPO (Invitrogen). This vector is supplied linearized with overlapping

thymidine residues that can be ligated to the overhanging adenosine residues generated in

the PCR-amplified products. Insertion of foreign DNA into this region prevents the

expression of lacZa allowing for blue-white screening of E. coli transformants. Ligated

pCR2.1 and PCR amplified products were used to transform E. coli DH5a by calcium








chloride [121]. Clones were screened on LB agar supplemented with 50 gg/mL

kanamycin and 0.75 |g/mL X-gal. White colonies were picked and tested for the

presence ofsspA, sspB, and spaP insert DNA after alkaline lysis. Plasmid DNA from

each recombinant was restricted with BamHI (Promega) and Sall (Promega), and

electrophoresed on 0.7% (wt/vol) agarose. The appropriate sized DNA fragments were

S excised from the gel and purified using a Qiagen gel extraction kit. The sspA, sspB, and

spaP fragments were ligated into BamHI (Promega) and Sall (Promega) linearized

pGEX-4T-2 resulting in directional cloning downstream of the gst gene, which encodes

the glutathione S-transferase (GST) of Schistosomajaponicumi. Ligated DNA was used

to transform E. coli BL21 and transformants were selected for on LB agar supplemented

with 100 |ig/mL of ampicillin and 75 ig/mL X-gal. White colonies were screened for the

presence of spaP-containing insert DNA as described previously.

The pGEX-4T-2-derived plasmids encoding the PI A- and P-regions were

designated pGEX-A and pGEX-P, respectively. The plasmids encoding the SspA and

SspB P-region-GST fusions were designated pGEX-AP and pGEX-BP.

Purification and Confirmation of GST-Fusion Proteins

Recombinant E. coli harboring plasmids encoding GST-fusion proteins and pGEX-

4T-2 as a control were grown for 16 hours in LB broth supplemented with 100 gg/ml of

ampicillin (LB/A100) and passage 1:100 into LB/A100. Following shaking at 250C

until an optical density at 600 nm of 0.5 was achieved, cultures were supplemented with

0.1 mM IPTG and grown for an additional 4 hours at 250C.

To confirm expression of each fusion protein, lysates from each recombinant E.

coli, as well as uninduced controls, were analyzed by SDS-PAGE and Western








immunoblot on 7.5% (wt/vol) polyacrylamide gels. Proteins in the gels were

electroblotted onto nitrocellulose membranes for 1 hour at 100 volts. Replicate filters

were stained with colloidal gold (Diversified Biotech, Boston, MA) or blocked for 1 hour

at room temperature with PBS-Tw. Membranes were incubated overnight at room

temperature with anti-GST rabbit polyclonal antisera (Amersham Biosciences) diluted

1:1000 in PBS-Tw. Membranes were washed four times with PBS-Tw prior to

incubation for 2 hours at room temperature with affinity-purified peroxidase-labeled goat

anti-rabbit conjugate (ICN/Cappell ICN Biomedicals, Aurora, OH) diluted 1:1000 in

PBS-Tw. Membranes were washed twice with PBS-Tw and twice with PBS prior to

development with 4-chloro-l-napthol solution for 30 min.

The Pl-GST fusion proteins were purified by affinity chromatography using

glutathione sepharose 4B (Amersham Biosciences). IPTG induced recombinant E. coli

were resuspended in PBS containing 1 mM PMSF (phenylmethylsulfonyl fluoride) and

lysed by sonication on ice 5 times for 15 seconds at power setting 3 using a Sonic 300

Dismembrator (ARTEK Systems Corporation, Farmingdale, NY). Triton X-100 was

added to the sonicate to a final volume of 1% (vol/vol). Following a 30 minute

incubation at 250C, the sonicate was centrifuged for 10 minutes at 12,000 xg, and the

supernatant was applied to the glutathione sepharose 4B. After repeated washes with

PBS, the bound fusion proteins were eluted with 10 mM reduced glutathione in 50 mM

Tris-HC1, pH 8. The recovery of purified Pl-GST fusion proteins was confirmed by

Western immunoblot using anti-GST rabbit polyclonal antisera (Amersham Biosciences).

Purified fusion proteins were quantified using the bicinchoninic acid (BCA) protein assay

kit (Sigma) with bovine serum albumin as the standard.








Competitive Inhibition ELISA to Detect A-Region and P-Region Interaction

Individual wells of Costar High Binding plates (Coming Incorporated, Coming,

N.Y.) were coated overnight at 40C, in triplicate, with 100 pl of 0.1 M carbonate-

bicarbonate buffer (pH 9.6) containing 0.02% (wt/vol) sodium azide and 100 ng of Pl.

Coating buffer and unbound antigens were removed from the ELISA plate wells and

unreacted sites were blocked with PBS-Tw and overnight incubation at 40C. Plates were

washed four times with PBS-Tw. Purified A-region-GST, P-region-GST, a 1:1 molar

ratio of A-region-GST and P-region-GST, and GST were added to MAb 4-10A, diluted

1:8000, to a final concentration of 1 nM. Controls included MAb 4-10A alone and no

primary antibody. The mixtures of Pl-GST fusions and MAb 4-10A were incubated at

4C for 30 minutes and then applied to the P1 coated ELISA plate at 100 tl per well.

The plates were incubated for 2 hours at 370C and washed four times with PBS-Tw. The

binding of MAb 4-10A to the immobilized P1 was traced with peroxidase-labeled goat

anti-mouse IgG (Cappell) at a 1:2000 dilution. The plates were developed with 4-CN,

and absorbance was measured as previously described. The percent inhibition of MAb 4-

10A binding to captured PI was calculated as 100 [(mean OD of Mab 4-10A + P1-

GST fusion /mean OD of MAb 4-10A alone) x 100].

Binding Stoichiometry of the A- and P-Regions by Continuous Variation

A variation on the Job Plot [122] was used to measure the binding stoichiometry of

the A- and P-regions required for the formation of the MAb 4-10A epitope. Purified A-

region-GST and P-region-GST fusion proteins were diluted in 0.1 M carbonate-

bicarbonate buffer (pH 9.6) containing 0.02% (wt/vol) sodium azide and then mixed in 7

different molar ratios (0:6, 1:5, 2:4, 3:3, 4:2, 5:1, and 6:0) while maintaining a constant








total concentration of 0.67 .M. The mixtures were incubated at 40C for 1 hour, and then

100 l per well was applied to a Costar High Binding plate (Coming Incorporated,

Coming, N.Y.) in triplicate. The plate was incubated overnight at 40C. Coating buffer

and unbound antigens were removed from the ELISA plate wells, and unreacted sites

were blocked with an overnight incubation in PBS-Tw at 40C. After washing 4 times

with PBS-Tw, MAb 4-10A was added to the wells at a 1:1000 dilution. Wells were

washed with PBS-Tw and peroxidase-labeled goat anti-mouse IgG was added to the wells

at a 1:2000 dilution. After washing, the plate was developed with OPD, and absorbance

was measured as above.

Analysis of P1 Translocation and the Contribution of the Alanine- and Proline-Rich
Regions

Introduction of spaAA into S. mutans PC3370

The spaPAA DNA, including the promoter, was restricted by Kpn I from pTS20

and isolated by gel electrophoresis and purification with a Qiagen gel extraction kit. The

purified spaPAA was ligated into the Kpn I site of the E. coli S. mutans shuttle vector

pDL289, creating pTS21, and introduced into E. coli DH5a by electroporation. Clones

were screened on LB agar supplemented with 50 .ig/mL ampicillin. Colonies were

picked and tested for the presence ofspaPAA insert DNA after alkaline lysis. pTS21 was

subsequently introduced to the S. mutans spaP-negative mutant PC3370 by natural

transformation. An overnight culture of PC3370 grown in THYE media was diluted 1:20

into THYE media containing 5% (vol/vol) sterile horse serum (THYHS). The culture

was grown to Klett 100 at 370C at which time pTS21 (pg/ml) was added. After an

additional 30 minutes at 370C, an equal volume of THYHS was added to the cultures.








Following 90 minutes at 370C, transformants were screened on THYE agar supplemented

with 500 pg/mL kanamycin. The sequences of all recombinant constructs were

confirmed by the DNA sequencing core (University of Florida).

Analysis of P1AA Cell Surface Expression in PC3370

The spaP isogenic mutant PC3370 harboring plasmids encoding P1 (pMAD),

P1AA (pTS21), PIAP (pMAJJ8) and vector only (pDL289) were grown for 16 hours at

37C, the cells were harvested by centrifugation and washed twice with PBS. Cells were

resuspended in PBS, and the densities of the suspensions were equalized at Klett 160.

Twofold serial dilutions of the cell suspensions were made in PBS, and 100 Ptl of each

dilution was applied in duplicate to two nitrocellulose membranes (Schleicher and

Schuell) by using a 96-well dot blot manifold (Schleicher and Schuell). Wells were

washed twice with 200 ml of PBS, and the filters were removed from the apparatus and

blocked with PBS containing 0.25% (wt/vol) gelatin and 0.25% (vol/vol) Tween 20. Cell

surface PI was detected with rabbit antiserum 230 [37] or Mab 3-10E [119] as the

primary antibodies diluted 1:500, peroxidase-conjugated goat anti-rabbit IgG and goat

anti-mouse IgG as the secondary antibodies diluted 1:1000, and development with 4-

chloro-l-naphthol solution.

RNA and Dot Blotting for Confirmation of P1AA Expression in PC3370

Following the manufacturer's protocol, RNA was isolated from stationary phase

cultures of PC3370 harboring pDL289, pTS21 (P1AA), and pMAD (PI) using the Qiagen

Rneasy kit (Qiagen, Valencia, CA). Total RNA concentration was measured by OD260/280

nm and standardized to -92 ig/ml by the addition of RNA dilution buffer (6x SSC, 20%

formaldehyde). Samples were serially diluted two-fold, and 50 il of each were applied








to a nylon membrane using a 96-well dot blot manifold (Schleicher & Schuell). The

membrane was baked for 30 minutes at 1200C and incubated in DIG Easy Hyb (Roche,

Indianapolis, IN) for 2 hours at 37C. The membrane was probed overnight at 370C with

digoxigenin-labeled, PCR amplified, DNA complementary to the 3' end ofspaP,

nucleotides 3985-4125. The membrane was washed, blocked for 1 hour at 250C in

Roche blocking buffer, and incubated in alkaline phosphatase-labeled anti-digoxigenin

antibodies. After washing in detection buffer, chemiluminence substrate, CSPD, was

added and the membrane was exposed to Super Rx x-ray film (Fuji, Tokyo).

Western Immunoblot Analysis of Periplasm Extracts from E. coli MC4100 and
CK1953 Harboring pUC18, pDC20, pDC9, and pTS20

Periplasm contents of E. coli DH5a harboring pUC18, pDC20, pDC9, or pTS20

and E. coli MC4100 and CK1953, asecB mutant [109], harboring pUC18 or pDC20 were

extracted by osmotic shock [108]. Briefly, cells were grown for 16 hours at 37C. The

medium was supplemented with 0.3 mM IPTG to induce p-galactasidase expression, and

the culture was incubated for an additional 2 hours at 37C harvested by centrifugation

at 7000 x g for 10 minutes, and resuspended in 30 mM Tric-HCl/20% (wt/vol) sucrose,

pH8.0, and EDTA to a final concentration of ImM. The cells were incubated at 250C for

10 minutes while shaking, harvested by centrifugation for 10 minutes at 10,000 x g, and

resuspended in ice-cold 5 mM MgSO4. Cells were next incubated in an ice bath for 10

minutes, centrifuged for 10 minutes at 10,000 x g, and again decanted. One molar Tris-

HC1, pH 7.4, was added to the supernatant to a final concentration of 20 mM. The

supernatant containing the periplasm contents was diluted 5:1 with SDS sample buffer

and incubated for 5 minutes at 1000C. The cell pellets were resuspended in SDS-sample

buffer and also heated for 5 minutes at 1000C. Proteins were separated on 7.5% (wt/vol)








SDS-polyacrylamide gels and transferred to nitrocellulose for 1 hour at 100V.

Immunoblots were blocked and developed as described above for the dot blot assay.

Construction of a Bicistronic spaP for Expression of a Discontinuous P1.

The following engineering produced a genetic construct encoding spaP that

expressed P1 as two independent fragments, the N-terminal 465 residues and the C-

terminal 1095 residues. Fragments ofspaP both upstream and downstream of the 3' end

of the A-region were amplified by PCR and subsequently ligated together to create a

"split" spaP (Fig. 3-5). Forward primer TS9k and reverse primer TS17 were used to

amplify spaP DNA upstream of the 3' end of the A-region, including the spaP promoter.

Forward primer TS18 and reverse primer TS10k were used to amplify spaP downstream

of the A-region. Primers TS9 and TS10 were engineered with KpnI restriction sites and

primers TS17 and TS18 contain engineered XhoI restriction sites (Table 3). Primer TS17

also encodes multiple stop codons for the termination of the N-terminal P1 fragment

translation. Primer TS18 contains the spaP ribosome binding site and encodes a start

codon for translation of the C-terminal P1 fragment. Reactions were carried out in a

UNO thermoblock thermocycler (Biometra, Tampa, FL) with plasmid-encoded spaP,

pDC20 [54] as the template and HiFi DNA polymerase (Invitrogen) for 30 cycles under

the following conditions: denaturation at 940C for 30 seconds, primer annealing at 51C

for 1 minute, primer extension at 680C for 2 minutes and 30 seconds or 720C for 1

minute and 30 seconds; and a final extension at 720C or 680C for 7 min. The resulting

1,653- and 3,536-bp gene fragments were restricted withAXhoI before being ligated

together. The ligated fragments were gel purified and amplified by PCR as before using

primers TS9 and TS10 under the following conditions for 30 cycles: denaturation at 94C

for 30 seconds, primer annealing at 580C for 1 minute, primer extension at 680C for 3








minutes 45 seconds; and 680C for an additional 7 min. The PCR product was cloned into

the pCR 2.1-TOPO vector, creating pTS30, which was introduced into E. coli Top 10

cells according to manufacturer's instructions. Clones were screened on LB agar

supplemented with 50 ug/mL kanamycin and 0.75 tg/mL X-gal. White colonies were

picked and tested for the presence ofspaP insert DNA after alkaline lysis. Plasmid

pTS30 from the recombinant was restricted with KpnI and electrophoresed on 0.7%

(wt/vol) agarose. The appropriate sized split spaP DNA fragment was excised from the

gel and purified with the Qiagen gel extraction kit. The split spaP sequence was ligated

into the KpnI site of the shuttle vector pDL289, creating pTS31, and used to transform E.

coli DH5a by electroporation. Transformants were selected for on LB agar

supplemented with 50 jtg/mL of kanamycin and 0.75 pg/mL X-gal. White colonies were

screened for the presence of spaP-containing insert DNA as before. Sequences of all

recombinant constructs were confirmed by the DNA sequencing core (University of

Florida).

Evaluation of P1 Fragment Expression by Western Immunoblot.

E. coli DH5a harboring pTS30 was grown for 16 hours at 370C, harvested by

centrifugation, and lysed by boiling for 5 minutes in SDS-sample buffer (4% [wt/vol]

sodium dodecyl sulfate [SDS], 2% [vol/vol] 2-mercaptoethanol, 20% [vol/vol] glycerol,

125 mM Tris-HCl [pH 6.8], 0.1 mg ofbromophenol blue per ml). Proteins were

separated by SDS-polyacrylamide gel electrophoresis on 7.5% acrylamide preparatory

gels by the method of Laemmli [117]. Proteins were electroblotted onto nitrocellulose

membrane (Schleichter and Schuell, Keene, N.H.) for 1 h at 100 V by the method of

Towbin et al. [118]. Immunoblots were blocked with PBS-Tw. Membranes were








incubated with A-region specific mAb 3-8D, A- and P-region dependent mAb 4-10A, and

C-terminal specific mAb 5-3E [119, 120] at dilutions of 1:000. After washing,

membranes were incubated in peroxidase-labeled goat anti-mouse IgG (Cappel) and

developed with 4-chloro-l-naphthol solution.

Evaluation of Surface Expression of Discontinuous P1 in S. mutans

Plasmid pTS31 was introduced into S. mutans PC3370 by natural transformation as

before. S. mutans NG8 and PC3370 harboring pDL289 and derivatives expressing P1

(pMAD) and discontinuous P1 fragments (pTS31) were grown for 16 hours at 370C .

Cells were harvested, applied to a nitrocellulose membrane, and surface expression of P

was traced as before with MAbs, 3-8D, 4-9D, 4-10A, 5-5D, 6-11A, and 3-10E.

Introduction of S. gordonii SspA and SspB A-Regions into P1AA

DNA encoding the A-regions ofsspA and sspB were amplified by PCR and ligated

into the plasmid encoding P1AA, pTS21. pTS21 was constructed with two silent

mutations that created a unique Sfo I recognition sequence at the site of the deleted A-

region DNA [115]. PCR was used to amplify DNA fragments of sspA and sspB that

encode 287 residues, which are homologous to the deleted A-region in the spaP construct

PlAA. Primers TS19 and TS20 were used to amplify the sspA DNA fragment and

primers TS21 and TS22 were used to amplify the sspB fragment (Table 3). Reactions

were carried out in a UNO thermoblock thermocycler (Biometra, Tampa, FL) with

chromosomal sspA and sspB as the templates and VENT polymerase (NEB) under the

following conditions: (i) 94C for 2 minutes; (ii) 30 cycles of 940C for 30 seconds, 50C

for 30 seconds, 720C for 30 and (iii) 720C for an additional 7 min. The resulting 861 bp

gene fragments were cloned into the Sfo I site ofpTS21 with E. coli DH5a as the host

strain. Plasmid DNA was isolated from clones and insert orientation was confirmed by








restriction digest and sequencing. The resulting plasmids, designated pTS22 (sspA A-

region) and pTS23 (sspB A-region), were introduced into the S. mutans spaP-negative

mutant strain PC3370 by natural transformation as previously described. Transformants

were selected for their ability to grow on THYE containing 500tgg/ml of kanamycin.

Western Immunoblot Analysis of Chimeric P1 Containing the A-region of S.
gordonii SspA and SspB

Whole cell lysates ofE. coliDH5a harboring pTS22 and pTS23 and mechanically

lysed S. mutans PC3370 harboring the same plasmids were electrophoresed on 7.5%

(wt/vol) SDS-PA gels, transferred to nitrocellulose, and traced with a panel of eleven

anti-P 1 mAbs as previously described.

Surface Expression of SspA and SspB in S. mutans PC3370

S. gordonii M5 SspA and SspB were expressed in the spaP-negative mutant

PC3370, and translocation to the cell surface was determined by whole cell dot blot.

Plasmids containing sspA driven by the sspB promoter, pGEM-sspA (unpublished), and

sspB, pEB-5 [25] were kindly donated by D. Demuth (University of Pennsylvania,

Philadelphia, PA). pGEM-sspA was linearized with SacI and blunted with Klenow

fragment, followed by a second digestion with SphI. The sspA fragment was gel purified,

ligated into the SphI-SmaI site of the streptococcal shuttle vector pDL289 and introduced

into PC3370 by natural transformation. pEB-5 [25] was digested with BamHI and EcorI

and the appropriate-sized sspB band was gel purified. The sspB fragment was ligated

into pDL289 and introduced into PC3370 as previously described. Transformants were

selected for their ability to grow on THYE containing 500tg/ml ofkanamycin.








Involvement of RopA (Trigger Factor) and DnaK in the Maturation and
Translocation of P1

Evaluation of P1 Surface Expression by Whole Cell Dot Blot in the S. mutans ropA
Mutant, TW90

Whole cell dot blots of TW90 [112], courtesy of Tom Wen (University of Florida,

Gainesville, FL), were used to determine whether the reduction in adherence was due to a

reduced level of surface localized P1. UA159 and TW90 were grown for 16 hours at

370C in THYE broth. The cells were passage into triplicate cultures at 1:50 in THYE

broth, grown at 370C to a Klett reading of 50, and passage again in THYE broth at 1:50.

Cells were grown to Klett readings of 20 and 150, harvested by centrifugation, and

washed twice with PBS. Cells were resuspended in 50% of the original culture volume

of PBS. Two-fold serial dilutions of the cell suspensions were made in PBS, and 100 tl

of each dilution was applied to replicate nitrocellulose membranes (Schleicher and

Schuell) by using a 96-well dot blot manifold (Schleicher and Schuell). Wells were

washed twice with 200 gl of PBS, and the filters were removed from the apparatus and

blocked with PBS containing 0.25% (wt/vol) gelatin and 0.25% (vol/vol) Tween 20. PI

was detected with five anti-Pi monoclonal antibodies [33] that recognize cell surface PI

as well as with rabbit polyclonal serum as the primary antibodies, each diluted 1:500.

Secondary antibodies were goat anti-mouse IgG or goat anti-rabbit Ig (MP Biomedicals,

Irvine, Ca) diluted 1:1000. The membranes were developed with 4-chloro-l-naphthol

solution. Quantification of Pl surface expression was performed by densitometry using a

Fluorchem imager and software (Alpha Innotech, San Leandro, Ca).








Evaluation of P1 Surface Expression by Whole Cell Dot Blot in S. mutans SM12, a
Low-Level Expresser of DnaK

To examine the contribution of DnaK to P1 surface expression, a whole cell dot

blot experiment as above was performed using S. mutans SM12, which was engineered to

express approximately 5% of the level of DnaK as the parent strain, UA159 (Lemos and

Bume, in preparation, University of Florida, Gainesville, FL)). S. mutans UA159 and

SM12 were grown and harvested at early-log and stationary phases, Klett readings of 20

and 150, and surface expression was determined as previously detailed.

Analysis of dnaK Message Levels by Quantitative Real-Time PCR

Real-Time PCR was utilized to evaluate the effects of expression of the A- and P-

region deletion constructs on dnaK mRNA level. Cultures of PC3370 harboring pDL289

(shuttle vector), pMAJJ8 (P1AP-region), and pMAD (P1), and pTS21 (PlAA-region)

were grown in triplicate to Klett 100 after multiple passages, as above. RNA from each

culture was isolated according to supplier's instructions using the Qiagen RNeasy kit

(Qiagen, Valencia, CA). The total RNA concentration was measured by OD 260/280 nm.

cDNA of dnaK and 16S RNA was synthesized from 0.5 utg of RNA using primer

dnaKAS and 16sRVS respectively, and SuperScript II reverse transcriptase (Invitrogen,

Carlsbad, CA) for 10 minutes at 250C, 50 minutes 420C, and 15 minutes at 700C.

Transcript levels were determined by using iQ SYBR Green Supermix (Bio-Rad,

Hercules, CA). Reactions were performed in a 25-l volume using the manufacturer's

protocols. The forward primers for dnaK and 16s RNA were DnaKS and 16SFWD.

Amplification was performed under the following conditions: 30 seconds at 950C,

followed by 40 cycles of 10 seconds at 95C and 45 seconds at 600C. Melt curve data

was collected with an additional 100 cycles of 10 seconds starting at 600C and increasing





35


by 0.40C after cycle 2 and 15 seconds at 720C. DNA amplification and fluorescence

detection was performed with the iCycler IQ real-time PCR detection system and

accompanying software (Bio-Rad). A standard curve was plotted for the reaction with

values obtained from the amplification of known quantities of DNA from dnaK. 16s

RNA was used to normalize RNA abundance for all reactions. For each experiment,

cDNA was amplified from RNA which was freshly isolated from each of three parallel

cultures of each transformant. Real-Time PCT was conducted on each cDNA sample

triplicate resulting in 36 data points for each transformant













CHAPTER 3
RESULTS

Expression of Recombinant P1AA and Recognition by Anti-Pi Monoclonal
Antibodies

Regions of Pl that contributed to the epitopes for eleven anti-Pi mAbs were

identified based on immunoblot analysis of full-length and truncated P1 polypeptides.

Several of these mAbs, 6-1 IA, 5-5D, 3-1 OE, and 1-6F, were initially mapped to the

central region of the protein, which contained the P-region [ 120]. Deletion of the P-

region (a.a. 826-996) from PI eliminated the binding of mAbs 6-11A, 5-5D, 3-10E, and

4-10A. A region contributing to the binding of mAb 4-10A had also been mapped to the

region just amino-terminal to the central region. Surprisingly, none of the P-region-

dependent mAbs bound to a P-region subclone (a.a. 819-1017) [54] suggesting that the

epitopes were complex and possibly conformational. Characterization of the epitope for

mAb 6-11A was conducted by Rhodin et al [123]. Construction of several P1 subclones

and analysis by Western immunoblot revealed that residues 465-1561 were not sufficient

for the binding of mAb 6-11A. These data suggested that the P-region and residues

amino-terminal of a.a. 465 were necessary for mAb 6-11A binding. Taken together with

crystal structure data indicating that the A-region and P-region may be in close proximity

[30], we elected to examine whether the A-region contributed to a complex structure also

involving the P-region. A spaP gene lacking DNA encoding the A-region (a.a. 179-466)

was constructed by PCR and cloned into pUC 18, creating pTS20, as detailed in Chapter

2. P1 lacking the A-region (P1AA) was detectable by Western immunoblot in whole cell








lysates of recombinant E. coli DH5a using anti-P polyclonal antibodies (data not

shown). While full-length P1 migrates approximately 20-kD larger than its predicted

-165 kD on SDS-polyacrylamide gels, PIAA, like P1AP, migrates at its predicted

molecular weight. The effect of deleting the A-region on the antigenicity of P was

examined by Western blotting utilizing the eleven anti-P 1 monoclonal antibodies [33].

Deletion of the A-region from P1 eliminated the reactivity of five of the eleven mAbs

(Figure 2). Three of the non-reactive mAbs, 4-10A, 5-5D, and 6-11A, are also not

reactive with P1AP [54]. Reactivity of mAbs 5-3E, 2-8G, 3-3B, and 6-8C, which are

specific to the C-terminus of P1 [120], confirmed that the deletion of the DNA encoding

the A-region did not disrupt the reading frame. The Western blot also shows that like

P1AP, P1AA is stably expressed and easily detectable in E. coli.

Evaluation of P1AA Expression in S. mutans

When expressed in S. mutans, PIAP was unstable and not detected on the cell

surface [54]. To determine whether the deletion of the A-region resulted in similar P1

characteristics, P1AA was expressed in the spaP mutant PC3370. Whole cell dot blot

analysis of PC3370 harboring pDL289 (vector), pMAD (P1), or pTS21 (PIAA) was used

to examine whether the A-region, like the P-region, is necessary for P1 surface

expression in S. mutans. These results are shown in Figure 3. Two-fold serial dilutions

of the cells were applied to the nitrocellulose membrane in duplicate. The positive

control PC3370C expressing full-length P1 (row 4) demonstrated the reactivity of the

antiserum with surface expressed Pl. Negative controls, PC3370 and PC3370A, vector

only, (rows 1 and 2) showed lack of reactivity of the antiserum with cells lacking P1.

PC3370 harboring pTS21 encoding P1AA (row 3) was not reactive with the polyclonal








antiserum indicating a lack of surface expression of Pl. These results indicated that

P1AA was not translocated to the surface of PC3370. No P1AA was detected in spent

culture liquor, although P1 was found in the spent culture liquor of PC3370C

(complemented spaP mutant) and NG8 (wild-type) (data not shown).

To determine if the lack of detectable PIAA on the surface was due to a problem

with translocation out of the cytoplasm, cell lysates were examined for detectable P1AA.

NG8, PC3370A, PC3370C, PC3370, and PC3370 harboring pTS21 were subjected to

mechanical lysis in a Mini Beadbeater apparatus, and samples were analyzed by Western

blotting (data not shown). Full-length P1 was present in both cell extracts and cell debris

of NG8 (wild-type) and PC3370C (complemented spaP mutant). P1AA was not detected

in either the cell extract or the cell debris of PC3370 harboring pTS21, and no P1 was

observed in the negative controls, PC3370A (vector only) or PC3370.

Evaluation of spaP-Specific mRNA in PC3370 Harboring the Deletion Construct
pTS21

With a lack of detectable P1AA in PC3370, an RNA dot blot was used to confirm

that spaPAA was transcribed from the pDL289 shuttle vector in PC3370 (Figure 4).

Dilutions of total cellular RNA were probed with a digoxinin-labeled probe

corresponding to the 3' end ofspaP. The top two rows correspond to the negative

controls, PC3370 and PC3370A harboring the vector only. The third row contains RNA

from PC3370 harboring pTS21 and the bottom row contains RNA from the positive

control, PC3370 harboring pMAD. The dot blot shows that spaPAA message is

expressed at levels equivalent to the full-length spaP expressed from pMAD.








Evaluation of Secretion of P1, P1AA, and P1AP in E. coli

Since P1AA and P1AP were undetectable and possibly unstable in S. mutans while

being clearly detectable in E. coli, the use ofE. coli as a model to determine the secretion

competency of these proteins was examined. To this end, periplasmic extracts ofE. coli

DH5a harboring pUC18, pDC20, pDC9, or pTS20 (vector only, and expressing P1,

PIAP, and PIAA respectively) were prepared by osmotic shock, and the presence of Pl

and derivatives was detected by electrophoresis on SDS-polyacrylamide gels followed by

Western immunoblotting. These results are shown in Figure 5. Lanes marked "C"

contain cellular lysates and lanes marked "P" contain periplasmic extracts. Lanes marked

pDC20 contain cellular extracts from E. coli DH5a harboring pDC20 (full-length P1) and

clearly show that P1 is present in both the cytoplasm and the periplasm. Lanes marked

pDC9 show cellular fractions from DH5a harboring pDC9 and show that P1AP is present

in the cytoplasm, but absent from the periplasm. Lanes pTS20 correspond to cell

fractions from DH5a harboring pTS20 and show that, like P1AP, P1AA is present in the

cytoplasm, but not translocated to the periplasm. Lanes marked pUC 18 are cellular

lysates and periplasm extracts from E. coli harboring pUC18 (vector only). These results

suggest that E. coli may be a viable model for the study of the intramolecular

requirements for P1 translocation.

Interaction of the A- and P-Regions Detected by ELISA

The demonstration that the binding of mAbs 4-10A, 5-5D, and 6-11A to Pl were

dependent upon the simultaneous presence of the A- and P-regions and work by van

Dolleweerd et al [124] characterizing a complex epitope comprised of the P-region and a

fragment of P1 containing the A-region suggested a possible interaction between these








domains. To determine whether the isolated A- (a.a. 186-469) and P-regions (a.a. 819-

1017) were capable of such an interaction, ELISA was used to evaluate binding (Figure

6). To facilitate protein purification, the A-region and P-region of Pl were expressed as

fusions with maltose-binding protein (MBP), pMA41 and pMA3 respectively (Table 1).

Purified P-region-MBP or MBP alone as a negative control was immobilized in ELISA

plate wells. After washing and blocking the plates, two-fold serial dilutions of A-region-

MBP were added to the wells. Binding of A-region-MBP to the immobilized proteins

was detected by the A-region-specific MAb 3-8D. As shown, the A-region-MBP bound

to P-region-MBP in a dose-dependent manner, but not to MBP alone. MBP alone did not

bind to MBP or to P-region-MBP (data not shown).

Restoration of Epitopes by the Interaction of the A- and P-Regions

The requirement for the simultaneous presence of both the A- and P-regions for

MAb 4-10A, 5-5D, and 6-11 A binding to P suggested that both of these regions

contribute to the epitopes for these mAbs. Reactivity of these mAbs against ELISA

plates coated with the A- and P-regions was tested as above (Figure 7). The ELISA

revealed that while the mAbs did not react to the A- or P-regions alone, they did react to

the wells containing both regions. Additionally, MAb 4-10A reacted equally well

regardless of which region is immobilized, while mAbs 5-5D and 6-11A clearly reacted

better when the P-region is immobilized to the plate.

Inhibition of MAb 4-10A Binding to P1 by an A- and P-Region Complex

Competitive inhibition ELISA was used to assay the ability of the A- and P-regions

to interact in solution phase. It was previously shown that the binding of MAb 4-10A to

P1 was dependent upon the presence of both the A- and the P-regions. In this assay, the

ability of A-region and P-region-GST fusions alone and together to inhibit MAb 4-10A








binding to immobilized P1 was analyzed. As shown in Figure 8 as a percentage of

inhibition, MAb 4-10A binding to immobilized Plwas not inhibited by A-region-GST, P-

region-GST, or GST alone. However, MAb 4-10A binding was inhibited by a mixture of

A-and P-region-GST as well as by P1 in solution.

Stoichiometry of the A- and P-Region Interaction

A quasi-continuous variation assay was performed to determine the stoichiometry

of the A- and P-region interaction required for the reconstitution of the epitope required

by MAb 4-10A. Varying molar ratios of A- and P-region-GST fusion proteins were

mixed while maintaining a constant total molar concentration. The A-region/P-region

mixtures were immobilized on a 96-well ELISA plate, and binding of MAb 4-10A to the

mixtures was traced with peroxidase-labeled goat anti-mouse antibody. The maximum

binding of MAb 4-10A to the A-region/P-region complex was clearly seen at a 1:1 molar

ratio (Figure 9).

Interaction of P1, SspA, and SspB A- and P-Regions

The A- and P-regions of Pl are approximately 70% identical to the same regions in

the S. gordonii M5 SspA and SspB, also members of the antigen I/I family. To

determine whether the SspA and SspB A- and P-regions exhibited the same binding

characteristics as the P1 regions, ELISA was used to analyze their ability to interact with

one another and with the A- and P-regions of P1. A- and P-region-GST fusion proteins

were purified by affinity column chromatography and immobilized to ELISA plates as

previously described. Two-fold serial dilutions of A-region-GST fusion proteins were

incubated with the immobilized P-region-GST proteins, and binding was traced with the

A-region specific mAb 3-8D as before. Figure 11 shows that the A-regions of P1, SspA,

and SspB are capable of binding to the P-regions of all three of the antigen I/II family








proteins. As measured by ELISA and Western blot (Figure 10) mAb 3-8D reacts equally

to each of the A-region-GST fusions.

Anti-Pi mAb Epitope Restoration by the Interaction of A- and P-Regions of SspA
and SspB

Based on the demonstrated restoration of epitopes by the interaction of the A- and

P-regions of Pl, the interactions of the A- and P-regions of SspA and SspB, and the

reactivity ofmAbs 4-10A and 5-5D to full-length SspA and SspB (Figure 12), restoration

of epitopes for the mAbs by the interaction of the A- and P-regions of SspA and SspB

was examined. Reactivity of mAbs 4-10A, 5-5D, and 6-11 A with a combination of the

A- and P-regions was tested by ELISA as previously described. As shown in Figures 13

through 18, none of the mAbs reacted to the A- or P-regions alone; however, the results

showed that all three mAbs reacted to all A-region/P-region interactions in which a P1

fragment was the overlaid moiety. The binding of all three mAbs was also restored upon

the interaction of any P-region to the immobilized A-region of P1. The interaction of

SspA A-region with immobilized P1 P-region was also able to restore binding of mAb 5-

5D. MAb 5-5D binding was also restored to a more limited extent when the P-region of

PI was overlaid on the immobilized A-region of SspA. In summary, all heterologous A-

and P-regions interacted, yet not all of the interactions restored anti-PI mAb epitopes.



Introduction of the A-Regions of SspA and SspB into P1AA

The identification of the interaction between the A-regions of SspA and SspB and

the P-region of Pl including the restoration of the mAb 5-5D epitope suggested that some

degree of Pt structure was attained but native structure was not fully restored at the

polypeptide level. To determine whether introduction of the A-regions of SspA and SspB








into P1AA could restore native structure and translocation of the deletion construct, each

A-region was ligated in-frame into the site of the deletion in P1AA. The homology

between each of the A-regions is illustrated in Figure 19. The resulting chimeric P1

proteins were examined for restoration of mAb binding by Western immunoblot (Figure

20). The binding of mAb 3-8D demonstrated that the S. gordonii A-regions had been

inserted into PIAA and were in-frame (panel A). Restoration of binding was only seen

for mAb 5-5D with the chimeric PI protein containing the A-region of SspA (panel B).

Carboxy-terminal specific mAb 6-8C reacted to both chimeric proteins indicating that the

insertion was in-frame and that the proteins were not truncated (data not shown). The

bands that mAb 6-8C reacted to were of the same molecular weight as the band that 5-5D

reacted to in panel B. Full-length P1 migrates on SDS-polyacrylamide gels near 185 kD.

Oddly, neither chimeric protein appeared to migrate slower than P1AA. It is apparent

that the introduction of the A-regions did not restore native P1 migration characteristics

to the proteins.

Stability and Translocation of Chimeric P1 Containing the A-Regions of SspA and
SspB.

Although full native structure based upon recognition by all A- and P-dependent

mAbs was not achieved, the binding of 5-5D to the SspA chimeric construct indicated

that some degree of A- and P-region interaction had been restored, therefore it was still of

interest to examine the possible restoration of translocation of the chimeric molecules to

the cell surface. To insure that there was not an intrinsic problem with the translocation

of the S. gordonii proteins to the surface of S. mutans, SspA and SspB were expressed in

PC3370. Whole cell dot blots were performed and expression was traced with mAb 5-5D

as it demonstrated the strongest cross-reactivity with SspA and SspB in Western blots.








Figure 21 demonstrates that both SspA and SspB were translocated to the surface in S.

mutans.

The surface expression of chimeric P1 containing the A-regions of SspA and SspB

was also examined by whole cell dot blot. PC3370 harboring pDL289, pMAD (P1),

pTS21 (P1AA), pTS22 (P1AA+SspA A-region), and pTS23 (P1AA+SspA A-region)

were bound to nitrocellulose membrane using a 96-well dot blot manifold. Surface

expression of P1 was traced with mAbs 4-10A and 5-5D as they are reactive to P1 on the

cell surface and were reactive to SspA and SspB by Western immunoblot (Figure 22).

No surface expression of either chimeric P1 was detected on the surface of PC3370. No

full-length chimeric P1 proteins were detected in cell lysates of transformed PC3370 by

Western blot with C-terminal specific anti-P mAbs (Figure 23, upper panel).

Breakdown products of the proteins were, however, detected with the A-regions specific

mAb 3-8D (Figure 23, lower panel).

Evaluation of the Involvement of SecB in the Secretion of P1, P1AA, and P1AP in
E. coli

It is presumed that P1 is translocated to the cell surface via the general secretary

pathway and the Sec translocase [86]. The two major routes to the Sec translocase are

via the chaperones, signal recognition particle (SRP) or SecB. In S. mutans, however, Pl

is secreted in the absence of the SRP pathway and S. mutans does not possess SecB or a

known ortholog. To determine whether the translocation of Pl is dependent upon SecB

in E. coli and possibly a SecB ortholog in S. mutans, secretion of PI to the periplasm was

examined in the E. coli SecB-negative mutant CK1953 [109]. Periplasmic extracts of

CK1953 and the wild-type MC4100 expressing Plwere prepared by osmotic shock, and

the presence of P1 was detected by Western immunoblotting using mAbs 5-3E, 2-8G,








and 6-8C. These mAbs are specific to the carboxy-terminus of PI and were used to

insure that only full-length molecules were traced. These results are shown in Figure

24A. Lanes marked "C" contain cellular lysates and lanes marked "P" contain periplasm

extracts. The host and plasmid expressed is indicated above each pair of lanes. The first

pair of lanes contains cellular extracts from MC4100 harboring pDC20 (full-length P1)

and show that P1 is detected in both the cytoplasm and the periplasm. The cellular

fractions of the second pair of lanes contains cellular extracts from the SecB' mutant

CK1953 harboring pDC20 and show that, like in MC4100, P1 is detected in both the

cytoplasm and the periplasm. The final pair of lanes corresponds to cell fractions from

MC4100 harboring pUC18 (vector only). The above cell extracts were also analyzed by

Western immunoblotting for the cytoplasmic protein P-galactosidase to confirm the

integrity of the periplasm extractions (Figure 24B). No 1-galactosidase was detected in

the periplasmic extracts. These results show that P1 translocation in E. coli is not

dependent on SecB, the chaperone that is central to the general secretary pathway of E.

coli.

Expression of Discontinuous P1 and Recognition by Anti-P1 Monoclonal Antibodies

It has been proposed that proteins frequently contain "uncleaved intramolecular

chaperone-like fragments". These fragments are believed to assist in protein stabilization

and folding by binding to adjacent regions [125]. Intramolecular chaperones (IMC) have

been identified in a number ofproteases, including a proline-rich IMC in the Limulus

Factor C [79] in which deletion of the IMC resulted in a malfolded and non-secreted

protein. Complementation of a secretion defect resulting from an IMC deletion has been

demonstrated in the Pseudomonas aeruginosa elastase, LasB. Mclver et al [126]








successfully rescued secretion and activity by expressing the IMC in trans. Additionally,

not all secreted proteins are translocated in an unfolded conformation. The TAT system

is capable of secreting proteins that are first folded in the cytoplasm [127]. Although the

TAT system has not been found in S. mutans, it was of interest to determine whether the

A- or P- regions possessed IMC activity and whether such an interaction would result in

the translocation of a non-linear or folded P1. To examine this, aspaP gene engineered

to express P1 as two peptides (a.a. 1-480 and a.a. 481-1561, see Figure 25) was

constructed by PCR and cloned into pCR2.1, creating pTS30, as detailed in Methods and

Materials.

The predicted molecular weights of the PI fragments are 51 kD for the 480 residue

N-terminal fragment and 119 kD for the 1081 residue C-terminal fragment. Cell lysates

of E. coli harboring pTS30 were examined by Western blotting utilizing anti-Pl mAbs, 3-

8D, 4-10A, and 5-3E (Figure 26). MAb 3-8D is specific to the A-region of P1 [35],

reactivity of MAb 4-10A has been shown to be dependent upon the simultaneous

presence of both the A- and P-regions, and MAb 5-3E is specific to the C-terminal end of

P1 [120]. In lane A, MAb 3-8D is shown to be reactive with bands migrating between

approximately 65 and 80 kD. In lane B, MAb 4-10A reacts with a band that corresponds

to the molecular weight of the C-terminal fragment at 119 kD. The binding of mAb 4-

10A suggests that enough N-terminal fragments co-migrate with the 119 kD C-terminal

fragment to restore epitope recognition by this antibody. In lane 3, mAb 5-3E also reacts

with a 119 kD band. The reactivity of all three mAbs indicates that both the N-terminal

and C-terminal P1 fragments are expressed and detectable in E. coli.








Evaluation of Surface Expression of Discontinuous Pl in S. mutans

The spaP mutant PC3370 was used as the host for plasmids pMAD and pTS31,

encoding full-length P1 and discontinuous P1 respectively. Whole cell dot blot analysis

was used to examine whether an interaction of the A- and P-regions, when expressed in

trans, could result in translocation of the fragments to the cell surface. These results are

shown in Figure 27. The positive controls S. mutans NG8 (wild-type) and PC3370C

expressing full-length PI (columns A and C) demonstrate the reactivity of the mAbs with

surface expressed Pl. MAb 3-8D has previously been shown to be unreactive with full-

length P1 on the cell surface. The negative control, PC3370A, harboring the vector only,

(column B) showed lack of reactivity of the mAbs with cells lacking Pl. PC3370

harboring pTS31 encoding the PI fragments (column D) was not reactive with the mAbs

indicating a lack of surface expression of the P1 fragments. These results indicated that

P1 fragments were not anchored to the surface of PC3370. The N-terminal fragment was,

however, detected in spent culture liquor with MAb 3-8D (data not shown).

To determine if the P1 fragments were detectable in S. mutans cell lysates, PC3370

harboring pTS31 was subjected to mechanical lysis in a Mini Beadbeater apparatus and

samples were analyzed by Western immunoblotting (data not shown). While both the N-

and C-terminal fragments were detected in E. coli, only the N-terminal fragment was

present in the S. mutans cell extracts, indicating that the C-terminal fragment was

unstable in S. mutans.

Surface Expression of P1 in a RopA-Deficient S. mutans and Bacterial Adherence to
Salivary Agglutinin

The first chaperone encountered by nascent polypeptides is believed to be the

polyprolyl isomerase (PPIase), RopA or trigger factor. Trigger factor is involved in








protein secretion and maturation. The involvement of RopA in the expression of

functional P1 was analyzed using an adherence assay. PI mediates binding to salivary

agglutinin, and the binding can be inhibited by the Pi-specific monoclonal antibody

(mAb) 4-10A, but not by the Pi-specific mAb 6-11A [33]. Adherence ofS. mutans

TW90, a RopA-deficient mutant [112], to human salivary agglutinin immobilized on an

F1 sensor chip was assayed using the BlAcore 3000 (BIAcore AB, Uppsala, Sweden) by

Monika Oli by the method described in [128]. Briefly, agglutinin was immobilized on

the BIAcore F1 sensor chip surface, and suspensions of S. mutans UA159 (wild-type) and

TW90 (AropA) in adherence buffer were injected onto the sensor chip. A substantial

reduction in adherence (> 50%) in three separate assays was observed for the RopA-

deficient cells (Brady laboratory, unpublished). The complete inhibition of adherence of

both UA159 and TW90 by the addition of anti-P1 mAb 4-10A indicated that the residual

adherence was Pl mediated (Brady laboratory, unpublished).

In light of the laboratory's findings that the function of P1 appeared to be altered in

a ropA-negative strain, whole cell dot blots of TW90 were used to determine whether the

reduction in adherence was due to a reduced level of surface localized P1 in the ropA-

negative background. Cells were grown to early-log and stationary phases, and samples

were standardized for cell number by absorbance measurements. Replicate blots were

reacted with mAbs 3-8D, 4-9D, 4-10A, 5-5D, 6-11A, 3-10E, 1-6F, 5-3E, 2-8G, 3-3B, or

6-8C. Quantification of P1 surface expression was performed by densitometry using a

Fluorchem imager and software (Alpha Innotech, San Leandro, Ca). The mAb 4-10A

results shown in Figure 28 are representative of all data. There were no differences

detected in the surface expression of Pl between wild-type UA159 and TW90.








Analysis of P1 Surface Expression in an S. mutans Mutant Expressing Low-Levels of
DnaK

The route of P1 translocation to the cell surface and the chaperones involved are

unknown. P1 is secreted in the absence of the chaperones SRP, SecB, and RopA,

although RopA appears to affect PI function. DnaK binds to proline-rich proteins [129]

and is involved in chaperoning a wide variety of proteins. DnaK also has a pool of

substrates that overlaps with RopA [104]. The contribution of DnaK to P1 surface

expression was examined by whole cell dot blot as was performed with the RopA mutant.

The experiment was performed using S. mutans SM12, which was engineered to express

approximately 5% of the level of DnaK as the parent strain, UA159 (Lemos and Bure,

in preparation, University of Florida, Gainesville, FL). S. mutans UA159 and SM12 were

grown and harvested at early-log and stationary phases, samples were standardized for

cell number by absorbance measurements, and surface expression was determined as

previously detailed. Figure 29 shows that there is a significant reduction in the amount of

surface expressed P1 at early-log growth in SM12 (P<0.0001), but P1 levels were equal

in both strains at stationary phase (data not shown).

Evaluation of dnaK mRNA Expression in S. mutans PC3370 Harboring pDL289,
pMAJJ8, pMAD, and pTS21

The reduction of Pl surface expression seen in the early-log phase of SMI2

suggested that DnaK might have a role in P1 translocation. Changes in dnaK mRNA

levels in response to the expression of P1 and P1 deletion constructs in S. mutans could

indicate an interaction between the chaperone and the PI proteins. To this end,

quantitative Real-Time PCR was utilized to measure levels of dnaKmRNA expression.

DnaK message was quantified from total RNA isolated from early-log phase cultures of

PC3370 harboring pDL289 (shuttle vector), pMAJJ8 (PIAP-region), and pMAD (PI),





50

and pTS21 (P1AA-region). Compared to the vector only control or PC3370

complemented with full-length P1, the level of dnaK message was significantly decreased

(P<0.005 and P<0.05, respectively) in the presence of P1AP and increased (P<0.005 and

P<0.05, respectively)in the presence of P1AA (Figure. 30) The mRNA levels of 16S

RNA were used as an internal control, and no significant difference was found between

samples (P<0.38).












SQ < a '" & in w a u


P1 185 kD


P1AA I 135 kD



Figure 2. Western blot analysis of P1 and recombinant P1 lacking the A-region (PIAA).
The reactivity of eleven anti-PI mAbs against whole cell lysates of E. coli
harboring pDC20 (P1) or pTS20 (P1AA) were analyzed by Western blot to
determine the effect of the A-region deletion on antigenicity. The mAbs used
are listed above each strip. The epitope ofmAb 3-8D is within the A-region,
however mAb 3-8D does not bind to full-length Pl. The reactivity of mAbs
4-10A, 5-5D, 6-11A, and 3-10E are dependent upon the P-region. The
reactivity ofmAbs 5-3E, 2-8G, 3-3B, and 6-8C are dependent upon the C-
terminal terminal third of P1.














PC3370

PC3370 + vector

PC3370 + P1AA


PC3370 + P1
0 0 4



CFU/well

Figure 3. Lack of surface expression of Pl devoid of the A-region. Whole cell dot blots
of S. mutans spaP isogenic mutant PC3370, PC3370 harboring shuttle vector
pDL289 alone, and PC3370 harboring the pDL289 construct encoding PIAA
or full-length P1. Blots were reacted with anti-Pi mAbs 1-6F and 3-10E.
These antibodies had been shown in previous experiments to react with
recombinant P1AA. Identical results were obtained using a polyclonal anti-PI
rabbit antiserum (data not shown).





53






PC3370
PC3370 + vector

PC3370+PI s l
PC3370 +P1AA :i 4
10 5 2.5 1.2 0.62 0.31 0.15
pg/well

Figure 4. RNA dot blot analysis ofspaP-specific mRNA levels in the S. mutans spaP-
negative mutant PC3370 and derivatives. Twofold serial dilutions of total
cellular RNA, beginning with 10 mg, were probed with DNA encoding the C-
terminus ofspaP. From top to bottom, the rows contain mRNA from
PC3370, PC3370A (vector only), PC3370C (full-length spaP), and PC3370
harboring pTS21 (spaP with A-region encoding DNA deleted), respectively.













PI PIAP PIAA Vector
C PC P C P C P


Figure 5. Western immunoblot of cytoplasm (C) and periplasm (P) fractions of E. coli
DH5a harboring pUC18 derived plasmids expressing full-length PI (pDC20),
PIAP (pDC9), P1AA (pTS20), and vector alone detected with C-terminus
specific mAbs 5-3E, 6-8C, and 2-8G. Migration of molecular weight
standards are indicated in kilodaltons.






















10 -e- P-regon ImmobiBed
MBP Immobilized

08

06

04

02 -

0.0" ~ I------- --- i-i-i ---- ----
0.0
1000 500 250 125 62.50 31,25 1562 781 3.90

A-region-MBP (ng/well)





Figure 6. Demonstration of A-region and P-region interaction by ELISA. 100 ng of P-
region-maltose binding protein (MBP) fusion polypeptide (.) or MBP alone
(A) were used to coat ELISA plate wells. Two-fold serial dilutions of purified
A-region-MBP fusion polypeptide starting at 1000 ng/well were added to the
coated wells, and binding of the A-region to the P-region or to the MBP
negative control was detected with the A-region specific MAb 3-8D.










4-10A








060
08 -- -A1 3 ----









5-50
030








010




B 25 15 65 315 15S 7B 3 1 0


A- 6-11A o
030 ,

025 -.3P 0 ,ni




8 015












Figure 7. Restoration of epitopes by A- and P-region interactions as measured by ELISA.
500 ng of P-region-MBP fusion polypeptide, A-region-MBP, or MBP were
used to coat ELISA plate wells. Two-fold serial dilutions of purified A-
region-MBP starting at 500 ng/well were added to the P-region and MBP
coated wells and vice versa. MAbs 4-10A, 5-5D, and 6-1 IA were tested for
reactivity. Panel titles indicate the mAb tested and the legends indicate the PI
fragment that was immobilized. A- and P-regions did not interact with MBP
alone and no mAb binding was detected with the controls (data not shown).














100 -- -





820


-U-
20


-40





Figure 8. Inhibition of anti-Pl MAb 4-10A binding to immobilized P1 in ELISA. To
determine whether the A- and P-region polypeptides could interact in solution
and produce an epitope recognized by mAb 4-10A, the antibody was mixed
with soluble PI, glutathione S-transferase (GST), A-region-GST, P-region-
GST, or a 1:1 molar mixture of A-region-GST and P-region-GST
polypeptides. The mAb 4-10A mixtures were applied to P1 immobilized to an
ELISA plate, and inhibition of binding to the immobilized P1 was measured.
Bars indicate percent inhibition.















0.25

0.2

0.15

O 0.1

0.05

0
A: 0 0.16 0.33 0.5 0.66 0.83 1
R 1 0.83 0.66 0.5 0.33 0.16 0
Molar Fraction





Figure 9. Stoichiometry of the mAb 4-10A epitope. Varying molar ratios of P1 A- and
P-region polypeptides with a constant total concentration of 3.3 pmoles were
immobilized in ELISA plate wells and epitope restoration was detected with
mAb 4-10A. The experiment was performed in triplicate and standard
deviation is represented by the error bars.






59







SspA SspB PI


100-
75-


50-

37-


Figure 10. Demonstration of similar level of mAb 3-8D reactivity to A-region-GST
fusion polypeptides of SspA, SspB, and PI by Western immunoblot.







60


P1 P-region
3-8D




12


8 2,1









SspA P-region
3-80
0,-










04





02
50 0 125 2 312 s 156 7S















SspA P-region
3-8D
-I S-pA A-.gn0
10
IPs A-.eBl.n

: tPIA
5 2 125 S 31.2 IS6 78 0















Figure 11. Demonstration of interactions between the A- and P-regions of different
antigen I/11 proteins. Panel titles indicate the source of the immobilized P-
regions. Binding of the different A-regions identified in the legends were
traced with the cross-reactive A-region specific mAb 3-8D. Legends identify
the overlaid polypeptides.
0. ------------o---------
















the overlaid polypeptides.













0<
o 0 -
c o < V
7 7


Figure 12. Evaluation of reactivity of A- and P-region dependent anti-P1 mAbs with P1,
SspA, and SspB. Whole cell lysates ofE. coli DH5a harboring pDC20 (P1),
pDDA (SspA), and pEB-5 (SspB) were electrophoresed on 7.5% SDS
polyacrylamide gels, transferred to nitrocellulose and probed with the anti-P
mAbs shown above. The indicated molecular weights are in kilodaltons.











SspA P-region
4-10A





















SspA P-region
5-50











S A P-reg spon A-ro
1A- $00 00


12

















S6-11A0
02
5 20 15 M 5 312 156 7,o o





















































immobilized P-region and the SspAb tested. Legend indicates the source of the
overlaid A-regions.
overl-a- sApegA-regs














































overlaid A-regions.











SspB P-region
4-1OA
SspAA-rBglon
::, PI A I
1,0 -0- sS 0 A- n












0A0 -

500 250 125 625 312 15-e 78 0



SspB P-region
5-5D
1a
--- spA A-rgo



12















SspB P-region
QA-





P1B ARon
14



8 o.







0-
O 4













M 250 M 1 V 5 12 s.6 7.8l 0












n ell


Figure 14. Restoration of epitopes by the interaction of various antigen I/II A-regions
with the immobilized P-region of SspA. Panel titles indicate the source of
immobilized P-region and the mAb tested. Legend indicates the source of the
overlaid A-regions.
overlaid A-regions.











P1 P-region
4-10A
18










P1 P-reg io







5-5 250 25 7 0

P1 P-region
5-5D


500 20 125 S25 312 15 78 0



P1 P-region
6-11A
14-



1,2- V 1


08 .l







500 250 125 2.5 312 15 18 o0
ngwell


Figure 15. Restoration of epitopes by the interaction of various antigen I/II A-regions
with the immobilized P-region of SspA. Panel titles indicate the source of
immobilized P-region and the mAb tested. Legend indicates the source of the
overlaid A-regions











SspA A-region
4-10A


--C SSpA Pp-egn
-- Sps p-reion
T- PI P-rwlon


500 250 125 825 312 156 78 0


SspA A-region
5-50

-0- sp P-r.en
-.- Pt P1R5221


5 25 S 22 5 MS 312 15.6 78 0



SspA A-region
6-11A

SSp.sregon


500 2S0 125 625 31,2 155 78 0
ngtwell


Figure 16. Restoration of epitopes by the interaction of various antigen I/II A-regions
with the immobilized A-region of SspA. Panel titles indicate the source of
immobilized P-region and the mAb tested. Legend indicates the source of the
overlaid P-regions








66



SspB A-region
4-10A


Sopu P-r05n
-0- sB P-rob~ln
T- PI P-reion


00-
50O 250 125 a5 312 158 78 0




SspB A-region
5-5D

-*- SpAP-W.lon
16- -0- 5-08a7p-f
'we PI-f
1.4







12



1,

0
o --i------,-,-,-i-i-,-
600 250 12S 25 31.2 156 78 o




SspB A-region
6-11A
SSl P-gon



124


10



0.6

04

02-


50 250 125 B2 312 156 7.8 0
ng/wel!


Figure 17. Restoration ofepitopes by the interaction of various antigen I/II A-regions

with the immobilized A-region of SspA. Panel titles indicate the source of

immobilized P-region and the mAb tested. Legend indicates the source of the

overlaid P-regions.











P1 A-region
4-10A
18






Is

8 P
oI



w 4

5o0 2 12 65 312 156 78 0



P1 A-region
5-5D
spA P-reln
-I PI P-b,





S040





so

m r 18 t 5 312 15f 7v 8 0a



P1 A-region
6-11A
-- SspA P-raegi







18



0.4-


5 25 125 65 312 15. 7s 0
ngwell


Figure 18. Restoration of epitopes by the interaction of various antigen I/II A-regions
with the immobilized A-region of SspA. Panel titles indicate the source of
immobilized P-region and the mAb tested. Legend indicates the source of the
overlaid P-regions.
















10 20 30 40

SpB vk T.# 1 V .1U AR .Y ipJ KL L V Q AV ED S I

50 60 70 80 W0
S'pA
sPB |T... E. ,IL& 4 j " . L l 1 q IA t I

100 /10 120 130




140 150 160 170 IO




190 200 210 220




230 240 250 260 270

S.pA .A N A.


2900 J00 310
PI
S.OA 2Io ?y +






Figure 19. CLUSTAL W alignment of the A-regions of PI, SspA, and SspB. Dark


grey shading indicates identity. Light grey shading indicates similarity.













1 2 3 4
A.
250-






I00 -







Panel A was reacted with the A-region specific mAb 3-8. Panel B was3-8D
Sw .







Figure 20. Western immunoblot of chimeric P1 containing the A-regions of SspA and
SspB. Whole cell lysates of E. coli DHSa harboring plasmids encoding P1
containing with the A-regions of SspA and SspB. Lanes contain P1 (1),
P1AA (2), P1AA + A-region of SspA (3), P1AA + A-region of SspB (4).
Panel A was reacted with the A-region specific mAb 3-8D. Panel B was
reacted with the A- and P-region dependent mAb 5-5D.












P1 SsoA


Figure 21. Surface expression ofS. gordonii SspA and SspB in S. mutans PC3370.
Whole cell dot blot ofPl-deficient S. mutans PC3370 complemented with
plasmid-encoded P1, SspA, and SspB. Surface expression was traced with
mAb 5-5D.


1 2 3 4 5

So


Figure 22. Demonstration of lack of ability ofheterologous A-regions to restore surface
expression of PAA in PC3370. Whole cell dot blot ofPC3370 harboring
vector alone (1) and plasmids expressing Pl (2), P1AA (3), and P1 containing
the A-regions of SspA (4) and SspB (5). Surface expression of P1 was
detected with mAb 5-5D.


Vector


SsoB








A B C DE


5-3E
2-8G
3-3B
6-8C





3-8D


Figure 23. Western immunoblots of cell lysates of PC3370 harboring vector alone (A),
and plasmids encoding PI (B), PIAA (C), P1AA + SspA A-region (D), and
P1AA + SspB A-region. P1 was detected with C-terminals specific mAbs
(upper panel) and A-region specific mAb 3-8D (lower panel).















pDC20 pDC20 pUCS8
MC4100 CK1953 CK1953


pDC20 pDC20
MC4100 CK193
C P C P


Figure 24. Western immunoblot of cytoplasm (C) and periplasm (P) fractions of E. coli
MC4100 (wild-type) and CK1953 (AsecB) harboring pDC20 (P1). PI was
traced with mAbs 5-3E, 2-8G, and 6-8C in panel A. In panel B, 3-
galactosidase was traced with a rabbit polyclonal antibody. Migration of
molecular weight standards are indicated in kilodaltons.












A-Region
( a 16AA


P-Region
a( a 4dn-9l63


a.a. 1-465 a.a. 466-1561



Figure 25. Schematic representation of discontinuous PI. Black arrow represents the
spaP promoter. N-terminal open reading frame (ORF) expresses residues 1-
465 including the A-region (shaded). C-terminal ORF expresses residues
466-1561, which includes the P-region (shaded).








A B C


100-
75- ii


Figure 26. Western immunoblot of PI fragments expressed from pTS30 in E. coli and
traced with mAbs 3-8D (A), 4-10A (B), and 5-3E (C). Migration of molecular
weight standards are indicated in kilodaltons.














A B C D
3-8D

4-9D

4-10D

5-5D



3-10E 7


Figure 27. Whole cell dot blot of S. mutans NG8 (A) and PC3370 harboring pDL289
vector control (B) pMAD encoding Pl (C), and pTS31 encoding
discontinuous P1 fragments (D). Surface expression of P1 polypeptides was
traced with the indicated anti-P mAbs.















5000

S4000

& 3000 T

S2000

1000

0
UA159 TW90


Figure 28. P1 surface expression levels ofS. mutans UA159 and TW90 (ropA- mutant)
at early log stage traced with mAb 4-10A as measured by densitometry.














5000

4000

._ 3000 *

g 2000

1000

0
UA159 SM12




Figure 29. P1 surface expression levels ofS. mutans UA159 and SMI2 (DnaK-deficient)
at early log phase traced with mAb 4-10A as measured by densitometry. No
difference was detected at stationary phase ([n = 12]* statistically significant,
P<0.0001. Significance was determined by student's t-test.)














6e+6

] 5e+6
E
Z 4e+6

S3e+6 T
**
E 2e+6 +1

Sle+6

0
Vector PIAP P1 PIAA




Figure 30. Real-Time PCR quantification ofdnaK mRNA from S. mutans PC3370
harboring the pDL289 vector alone and expressing PlAP-region, full-length
P1, and PlAA-region. ( [n = 36]* statistically significant compared to vector,
P<0.005. ** statistically significant compared to P1, P<0.05.
Significance was determined by student's t-test.)














CHAPTER 4
DISCUSSION AND CONCLUSIONS

Dental caries is one of the most prevalent oral diseases worldwide, affecting 60-

90% of schoolchildren and the majority of adults. In the United States, dental caries is

the most common chronic childhood disease with 78% of 17 year olds having at least one

cavity or filling [130] and, according to the U.S. Department of Health and Human

Services, it is estimated that over $84 billion dollars is spent annually on dental treatment

and caries prevention in the United States alone. While advances in dental care and caries

prevention has reduced the incidence of caries in developed countries, the incidence of

caries worldwide has remained unchanged for the past 20 years [131].

A major contributing factor to the decline of caries in developed countries was the

introduction of fluoride to water and toothpaste. Unfortunately, in developing countries,

where the incidence of caries is on the rise, fluoridated community water is commonly

not a viable option. Although fluoridated water reaches 60% of the U.S. population,

more than 90% of toothpastes contain fluoride, and processed food and beverages often

contain fluoride, the reduction in caries incidence has been uneven across the general

population in the United States. The majority of the disease is now being borne by a

disproportionately small segment of the population; those of low socioeconomic status,

low education, and lack of access to dental care [132-134], in essence a mirror of the

populations in developing countries. The inability to manage caries in this subpopulation

of the United States, where the dentist to population ratio is better than 1:2000 people,








illustrates the improbable task of preventing caries in developing countries, such as in

Africa where the ratio is :150,000. A better understanding of the virulence factors and/or

targets of protective immunity in S. mutans could lead to preventative measures that

could help overcome the lack of resources, education, and infrastructure that is currently

required for caries prevention.

The major surface protein P1 of the cariogenic organism, S. mutans, is a

multifunctional adhesin and plays a role in the attachment of the bacterium to the tooth

surface. P1 shares similarities to virulence factors of several other bacterial species,

including the fibronectin binding proteins ofS. aureus and S. pyogenes [65, 66], and the

pneumococcal surface protein (PspA) of S. pneumonia [68]. P1 has been identified as a

target for protective immunity and has been studied as a potential antigen candidate for

an anti-caries vaccine [135]. It has also been used for the production of mAbs that are

currently being investigated for their ability to modulate the immune response in mice

that are challenged with mAb-S. mutans or mAb-Pl complexes [136]. Also, with little

known about the maturation and translocation of Streptococcal surface proteins, P1 is a

potential model for studies in these areas.

The goal of this research was to further our understanding of the structure and

antigenic properties of this large and complex molecule with an emphasizes on

identifying intramolecular interactions, the contribution of intramolecular interactions to

structure, stability, and translocation of P1, and to begin to identify chaperones that

contribute to P1 maturation and translocation.

Identification of an Intramolecular Interaction within P1

Previously, by process of elimination using truncated PI polypeptides, the central

region of Pl was determined to contribute to the epitopes of six of eleven anti-P mAbs








(4-9D, 4-10A, 5-5D, 6-11A, 3-10E, and 1-6F) [120]. It was additionally shown that

deletion of the P-region of P1 (PIAP) abrogated the binding of four of the eleven mAbs

(4-10A, 5-5D, 6-11 A, and 3-10E) and that none of these antibodies recognized a

subcloned P-region peptide suggesting that their epitopes were complex and

conformational. Surprisingly, although P1AP retained its N-terminal signal sequence and

C-terminal cell wall anchoring motif it was unstable in S. mutans and not translocated to

the cell surface [54]. Proline-rich regions have been shown to be involved in both

protein-protein interactions and intramolecular chaperone-like interactions. The initial

objective of these studies was to identify interactions between the P-region and other

regions of P .

Work by Rhodin et al. [123] on the characterization of the mAb 6-11 A epitope

further defined regions of PI that were required for reactivity of the 6-11A. Analysis of

several P-region spanning P1 subclones revealed that in addition to the P-region (a.a.

819-1017), residues N-terminal of D465 also contributed to the reactivity of mAb 6-11A.

In addition, the crystal structure of the P1 variable region suggested that the A- and P-

regions may be in close proximity [30], and it was reported that a polypeptide containing

the P-region bound to the N-terminal third of P1 [124]. Based upon these reports, the A-

region was examined to determine whether it contributed to a complex structure by

association with the P-region.

The initial experiment was to examine the effects of the removal of the A-region

from P1. Therefore, a spaP gene lacking the A-region (a.a. 179-466) was constructed by

PCR and cloned into pUC 18, creating pTS20. The construct was engineered with a silent

mutation that produced a unique SfoI restriction site that would later be used to insert








heterologous A-regions. The insertion of the SfoI site dictated the exact residues that

were deleted. Deletion of the A-region resulted in a loss of reactivity of five of eleven of

the anti-P1 MAbs (3-8D, 4-9D, 4-10A, 5-5D, and 6-11A), 3 of which are also dependent

upon the presence of the P-region (4-10A, 5-5D, and 6-11A) (see Figure 2). This

suggests that the epitopes of these three antibodies are complex and composed of portions

of discontinuous segments or that an interaction between the regions results in

conformational epitopes being produced within one or both of the regions. Reactivity of

Mabs 5-3E, 2-8G, 3-3B, and 6-8C, which are specific to the C-terminal of P1, confirmed

that the deletion of the DNA encoding the A-region did not disrupt the reading frame.

The Western immunoblot also shows that like PIAP, P1AA is stably expressed and

detectable in E. coli.

The presence of internal proline-rich regions has been associated with aberrant

migration of streptococcal and staphylococcal proteins on SDS-polyacrylamide gels. The

molecular mass of Pl has been predicted to be -166 kDa, although the protein migrates

with an apparent molecular mass of-185 kDa by SDS-PAGE. Interestingly, P1AP and

P1AA run at their predicted sizes of 152 kDa and 135 kDa, respectively, suggesting that

an interaction between the A- and the P-regions may contribute to anomalous migration

of P1 by SDS-PAGE. On a sided note, the abberant migration of P1 was observed even

after denaturingg" in 8 M urea and SDS-PAGE at both 40C and 600C.

With the data from the A-region deletion indicating that the A- and P-regions

contribute to the same epitopes, it was of interest to determine if these regions were, in

fact, capable of interacting. An interaction between recombinant polypeptides

corresponding to the isolated A-region (a.a. 186-469) and P-region (a.a.819-1017) was








examined by ELISA. A-region polypeptide was incubated with immobilized P-region

and A-region binding was detected with anti-P1 MAb 3-8D, which recognizes an epitope

contained entirely within the alanine-rich repeats. The ELISA revealed a direct, dose-

dependent, interaction (see Figure 6).

The required simultaneous presence of both the A- and P-regions for the binding of

mAbs 4-10A, 5-5D, and 6-11A to PI and the ability of recombinant A- and P-region

polypeptides to interact suggested that the interaction of these regions could contribute to

the epitopes that are recognized by these mAbs. To further analyze the characteristics of

the epitopes recognized by these mAbs, ELISA were performed with A- and P-region

polypeptides to determine whether binding of mAbs 4-10A, 5-5D, or 6-11A was restored

upon interaction of these two discontinuous domains. Interestingly, mAbs 5-5D and 6-

11A reacted considerably better when the A-region was applied to immobilized P-region

rather than visa versa, while MAb 4-10A displayed no apparent preference (see Figure 7).

This would suggest that the contact residues for MAbs 5-5D and 6-11A are largely

contained within the A-region while both regions may contain residues required for the

MAb 4-10A epitope. Since, all three of these mAbs bind to PI on the surface ofS.

mutans, these results indicate that the A- and P-regions interact in the context of the

whole molecule in its native conformation.

Due to the possibility that the interaction of the A- and P-regions may be an artifact

of being immobilized to the ELISA plate, a competitive inhibition ELISA was performed

to assess the interaction of the A- and P-regions in solution. Since it was shown that an

interaction of the A- and P-region is required for the binding of MAb 4-10A, recombinant

A- and P-region polypeptides were used alone and in combination to inhibit the binding








ofMAb 4-10A to immobilized P1. As can be seen in Figure 8, neither the A-region nor

the P-region individually inhibits MAb 4-10A binding to PI, however the combination of

both polypeptides does. This indicates that the A- and P-regions are capable of

interacting in solution.

In an effort to establish the stoichiometry of the A- and P-region interaction, a

variation of the Job Plot was performed. The Job Plot, or continuous variation, consists

of mixing two binding partners, or an enzyme and substrate, at various molar ratios while

holding the total concentration constant and then recording a measurable change. While

continuous variation is normally performed in solution, due to the use of antibodies to

measure the A- and P-region interaction, the A- and P-region polypeptides had to be

immobilized to an ELISA plate to afford the removal of excess unbound antibody. Since

MAb 4-10A was the tool used to measure the A- and P-region interaction, the

stoichiometry that was determined would actually be that required for the formation of

the MAb 4-10A epitope. According to the assay, the epitope of mAb 4-10A consists of a

1:1 ration of A-region to P-region (see Figure 9).

Although antigen I/II proteins are highly conserved, the functional properties of

individual members of this family of proteins differ. S. gordonii possesses two antigen

I/II proteins, SspA and SspB, which have been well characterized. Several functional

differences between these two proteins have been identified, including coaggregation

with other oral flora and interaction with type I collagen [137, 138]. PI is closer to SspA

in homology, 67% identity versus 57% with SspB. Specific amino acid residues that are

not in PI have been identified to be important for SspB binding to Porhromonas

gingivalis [139] and interactions of SspB and P1 with salivary agglutinin also differ








[140]. Recent studies focusing on the A-regions of SspA, SspB, and P1 have also

identified structural and functional variation. It was reported that the A-regions of P1 and

SspA bound to salivary agglutinin but that the A-region of SspB did not. In addition,

structural analysis suggested that the A-region of SspB is less stable than that of SspA

and Pl, both at high temperature and low pH. It should be noted that the A-regions of

SspA and SspB exhibit approximately 87% primary sequence identity with one another

while the A-regions of SspA and P1 only share 70% (see Figure 19) [35, 36, 116, 141].

In light of the similarities and differences reported between the A-regions of SspA,

SspB, and P1, it was of interest to see if the A- and P-regions of these S. gordonii

proteins interacted with one another as well as with the A-and P-regions of P1. Again,

interaction between these regions was examined by ELISA and a dose-dependent

interaction was observed with mAb 3-8D reacting to A-region polypeptides binding to

immobilized P-region polypeptides (see Figure 11). The results indicate that the A-

region of Pl interacts more strongly with all of the P-regions than either of the S.

gordonii polypetides and that the SspB A-region is the weakest binder, which follows the

trend of P1 being more like SspA.

The contribution of the P1 A-region and P-region interactions to native structure as

determined by epitope restoration and the ability of the heterologous interaction of A-

and P-regions led to the examination of epitope restoration by the interaction of the

heterologous A- and P-regions. As before, ELISA was used to detect the restoration of

epitopes for mAbs 4-10A, 5-5D, and 6-11A by every combination of PI, SspA, and SspB

A- and P-region interactions in which eith the A- or P-region was immobilized. The

results demonstrated that the epitopes recognized by these three A- and P-region








dependent mAbs were restored regardless of which P-region interacted with the

immobilized A-region from P1 (see Figure 18). While when P-regions were interacted

with either of the S. gordonii A-regions, only the P-region of Pl restored any mAb

reactivity, which was for mAb 5-5D and at a low level (see Figuresl6 and 17).

Additionally, the A-region of Pl was able to restore all of the epitopes when interacted

with any of the immobilized P-regions and the A-region of SspA was able to restore the

mAb 5-5D epitope when interacted with immobilized PI P-region (see Figure 15). And

lastly, the interaction of the A-region of SspB with any immobilized P-region failed to

restore epitopes.

Comparing the results of the A- and P-region interactions to the reactivity of mAbs

4-10A, 5-5D, and 6-11A against full-length SspA and SspB as examined by Western

immunoblot was interesting (see Figure 12). While mAb 5-5D reacted to both full-length

SspA and SspB, its epitope was not restored by the interaction of the SspA and SspB A-

and P-regions. The same held true for mAb 4-10A which bound to full-length SspB and

weakly reacted to SspA. To summarize, although mAbs 4-10A and 5-5D bound to full-

length SspA and SspB, their epitopes were not restored by interactions of the A- and P-

regions of these proteins unless one of the interacting regions was from P1.



Analysis of P1 Translocation and the Contribution of the Alanine- and Proline-Rich
Regions

All life depends upon the targeting of newly synthesized proteins to their site of

action. During transit to its destination, a protein must avoid a variety of hazards such as

malfolding, aggregation, and degradation and may be required to pass through one or

more membranes, known as translocation. Protein translocation has been extensively








studied in E. coli and the models established through this research are believed to be

representative of all bacterial cells [142-144] but more recently, Bacillus subtilis has

become the model for Gram-positive bacteria. Proteins that are targeted for translocation

across the cytoplasmic membrane contain an N-terminal signal sequence [145] that

generally contains positively charged residues followed by 15 to 20 hydrophobic

residues[146, 147], which are usually removed during or shortly after translocation.

Signal peptides can be classified by the type of signal peptidase that is responsible for

their proteolytic processing and these classifications can be used to predict the

translocation pathways [148]. Based upon surveys of signal peptides in the genomes of

B. subtilis and several other Gram-positive bacterium, it is predicted that most

extracellular proteins in these organisms are secreted via the Sec-translocase [149-151] In

Bacillus subtilis, there are four predicted protein transport pathways; (i) the Sec-

dependent pathway, (ii) Twin-arginine translocation (Tat), (iii) ABC transporter-

dependent secretion pathways, (iv) and a pseudopilin-specific export pathway. A survey

of the S. mutans UA159 genome failed to reveal any homologues of the Tat machinery or

Tat signal peptides.

In addition to the requirement for MAb reactivity, the simultaneous presence of

both the A- and the P-regions appear to be required for P1 stability in S. mutans.

Analysis of mRNA encoding P AA, like P AP [54], demonstrated that the internally

deleted spaP gene was transcribed at levels equivalent to the wild-type spaP gene (see

Figure 4). Differences in dnaK mRNA levels in S. mutans harboring plasmids encoding

P1, P1AA, and P1AP also suggested that the PIAA was being translated. However, no

PlAA was detected in the cytoplasm, on the cell surface (see Figure 3), nor in the culture








liquor. While P1AP contains a deletion of 170 residues and P1AA lacks 287 residues,

there are examples of stable antigen I/II polypeptides that, when compared to P1, are

lacking large segments of the molecules. The antigen I/II protein expressed by S.

intermedius, Pas, lacks -270 residues from the A-region and -80 residues from the P-

region [152]; Paa from S. cricetus possesses an additional -139 residues in the A-region

and -39 residues less in the P-region [152]; and S. mutans GS-5 expresses a PAc

molecule lacking the C-terminal -400 residues [153]. The A-region of Pl consists of

three-82 residue repeats and the P-region consists of three 39-residue repeats and both

Paa and Pas retain repeats in both regions. Not all internal deletions in P1 result in the

apparent level of instability seen in P1AA and P1AP. Rhodin et al. constructed a PI

construct lacking residues 84-190 which was detectable in S. mutans PC3370, but was

not translocated to the surface (unpublished). This suggests that the A- and P-regions

may contain inherent structural information, possible chaperone binding sites, or perhaps

possess chaperone-like activities that are critical to PI stability. A proline-rich region has

been implicated as an intramolecular chaperone by Wang et al. [79]. The central proline-

rich region of the Limulus secreted serine protease, Factor C, was shown to be required

for secretion of the molecule. Their data suggested that the correct folding of the

molecule C-terminal of the proline-rich region was dependent upon the presence of the

proline-rich region and that the lack of secretion was due to malfolding.

To fully understand the role of the A- and P-regions in P1 translocation, identifying

the molecule's route of translocation is necessary. There is no experimental data that

identifies the secretion pathway employed by PI or antigen I/II-like proteins. Cell wall

anchoring of Pl and PAc is mediated by the transpeptidase sortase [85, 154] and sortase








anchored proteins are presumed to be translocated via the sec translocase [86]. As

detailed in the introduction, the Sec-dependent secretion pathway has been thoroughly

studied in E. coli and the characteristics of the pathway are presumed to be conserved for

all bacteria. In E. coli, the Sec-translocase consists of a structure composed of several

proteins including the ATPase SecA, which provides the energy for translocation [87].

Current literature identifies two major pathways that a nascent protein destined for the

Sec-translocase would be transported upon, the signal recognition particle (SRP) pathway

and the SecB pathway.

The SRP pathway is involved in co-translational protein secretion. The SRP

recognizes and binds to the signal peptides of nascent polypeptides as they emerge from

the ribosome [89]. Binding of the SRP stalls translation and targets the SRP-ribosome

complex to the SRP receptor, FtsY [90, 91]. The SRP-ribosome-FtsY complex is then

targeted to the Sec-translocon where the ribosome docks and the protein is co-

translationally translocated across the membrane [87].

The cytoplasmic chaperone SecB targets preproteins to the Sec-translocon for post-

translational translocation. SecB binds to nascent and full-length proteins as they emerge

from the ribosome [92]. SecB interaction prevents premature folding of the preprotein

and delivers it to the Sec-translocon in a secretion-competent state. Binding of the SecB-

protein complex with SecA results in the transfer of the preprotein to SecA and the

release of SecB [93]. The protein is subsequently translocated across the membrane

through the Sec-translocon [94].

Due to the faint expression of PIAP and the undetectable expression of PIAA in S.

mutans, E. coli was used to begin to examine P1 secretion. PlAP and PIAA are stable








and detectable, albeit at reduced levels, in whole cell lysates ofE. coli by Western

immunoblot. Analysis of periplasmic extracts by Western immunoblot revealed that P1

was secreted into the periplasm, but P1AP and P1AA were not. This suggests that while

the A- and P-regions are apparently not required for stability in E. coli, the regions are

required for secretion. If a lack of chaperone interaction with the deletion recombinant

proteins results in the lack of secretion, perhaps a similar lack of interaction also favors

degradation of the molecules in S. mutans. It is known that secretion incompetent

proteins are subject to rapid turnover, which is a likely scenario when these constructs are

expressed in S. mutans. With such a rapid turnover in S. mutans and considering the time

required for lysis of the bacterium, pulse-chase experiments to determine the half-life of

these deletion proteins in Streptococci are currently technically improbable.

The SRP pathway has been identified in both gram-negative and gram-positive

bacteria. In B. subtilis, numerous homologs of the general secretary pathway

components have been identified. However, as is the case with S. mutans, no SecB

homolog has been identified, but B. subtilis has been found to possess a functional

ortholog, CsaA, which has been shown to have partially overlapping binding

characteristics [95-97]. As previously stated, the SRP is essential for viability in E. coli

and it was assumed to be the case in all organisms. However, it has been demonstrated

that S. mutans is viable without SRP [98] and PI is translocated and expressed on the cell

surface in its absence (unpublished). This would suggest that if P1 secretion is Sec-

dependent, the targeting pathway should likely be SecB-like and require a functional

SecB ortholog. To examine the possibility of a role for SecB in P1 secretion, P1 was

expressed in a SecB-negative E. coli mutant, CK1953. PI was shown to be stable and