Campylobacter pylori Proteins - PubMed Central Canada

Vol. 57, No. 6

INFECTION AND IMMUNITY, June 1989, p. 1825-1833

0019-9567/89/061825-09$02.00/0 Copyright © 1989, American Society for Microbiology

Two-Dimensional Gel Electrophoresis and Immunoblotting of Campylobacter pylori Proteins BRUCE E. DUNN, 3* GUILLERMO I. PEREZ-PEREZ,2 4 AND MARTIN J. BLASER2'4 Laboratory' and Medical2 Services, Denver Veterans Administration Medical Center, Denver, Colorado 80220, and Department of Pathology3 and Division of Infectioius Disease, Department of Medicine ,4 University of Colorado Health Sciences Center, Denver, Colorado 80262 Received 9 November 1988/Accepted 25 February 1989

Whole-cell, outer-membrane protein, flagellum-associated antigens and partially purified urease of Campylobacter pylori were analyzed by two-dimensional gel electrophoresis. C. pylori strains were readily distinguished from strains of Campylobacterjejuni, C. coli, and C. fetus by absence of major outer membrane proteins with Mrs of 41,000 to 45,000. C. pylori strains also lacked the acidic surface-array proteins at Mr 100,000 to 149,000 identified previously in serum-resistant strains of C. fetus. Surface labeling of intact C. pyloni cells with 1251 revealed two common major proteins, which we have designated protein 2 (pl 5.6 to 5.8, Mr 66,000) and protein 3 (pI 5.2 to 5.5, Mr 63,000). Proteins 2 and 3 were also the major components (subunits) observed in partially purified urease. Partially purified preparations of flagella consistently contained proteins 2 and 3. Thus, urease appears to be associated with both outer membranes and flagella of C. pyloni. C. pyloni strains also possessed an antigen at Mr 59,000 which was cross-reactive with antiserum against flagella of C. jejuni. However, the antigen did not appear to be associated with flagella per se in C. pyloni. Protein 2 was unique to C. pylori among the Campylobacter species studied. It was not recognized by antibody against whole cells of C. jejuni or C. fetus or flagella of C. jejuni. Protein 3 was cross-reactive with antiserum against whole cells of C. jejuni and C. fetus, as were several other major protein antigens. Because protein 2 is a major outer membrane protein that is apparently unique to C. pylori, development of monospecific antibodies against this antigen may be useful for the identification of C. pylori in tissues, and purified antigen may be useful for serologic tests for specific diagnosis of C. pylori infections.

Campylobacter pylori is associated with gastritis and peptic ulcer disease (20-22, 32, 39), although it has yet to be established whether C. pylori is pathogenic per se or is merely a secondary colonizer of damaged gastric mucosa (1, 32). Despite its uncertain role in pathogenesis, the presence of C. pylori is at the least an important indicator for these inflammatory gastroduodenal conditions (1, 30). C. pyloni is characterized by highly active urease activity, which may facilitate colonization in the acidic milieu of the gastric antrum. As with other gram-negative organisms, the outer membrane of C. pylori contains lipopolysaccharide and specific proteins, which are exposed on the surface of the bacterium and are antigenic to humans (15, 16, 29, 33, 38). Outer membrane proteins (OMPs) in many enteric pathogens play important roles in adherence to and invasion of the gastrointestinal mucosa, complement fixation, and resistance to serum bactericidal activity and phagocytosis (7), and it is reasonable to assume that the OMPs of C. pylo)ii might serve one or more of these functions. Protein composition of C. pylo)i has been studied by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), a method capable of resolving proteins only on the basis of differences in molecular weight (15, 16, 18, 23, 25, 27, 29, 33, 38). Based on SDS-PAGE analysis and immunoblotting, it appears that C. pyloni shares flagellar antigens with other Campylobac ten species (18, 29). The purpose of this study was to characterize, by using two-dimensional gel electrophoresis (2DGE), the OMPs, flagellar antigens, and urease of C. pyloni and to determine

which, if any, of these antigens might be unique among Campylobacter species. MATERIALS AND METHODS Bacterial strains. The C. pylori strains used in this study were from the culture collection of the Denver Veterans Administration Medical Center Campylobacter Laboratory. All C. pylori strains used (Table 1) were identified, passaged, and stored as described previously (29). Cultures were grown on chocolate agar (Remel Laboratories, Lenexa, Kans.) at 37°C in an atmosphere containing 7.5% H2, 7.5% CO., 5% 02, and 80% N2. For comparison, we used one strain each of Campylobacter jejuni (81-94, PEN 2) and C. fetuts (82-40 LP), which we have studied previously with 2DGE (10). Preparation of Campylobacter antigens. After 72 h of incubation, cells were harvested in 10 mM Tris hydrochloride (pH 7.4) at 4°C and centrifuged twice at 5,000 x g for 10 min. Pellets were frozen in a dry ice-ethanol slurry, thawed, and then treated with 100 ,ug of DNase per ml-50 p.g of RNase A per ml for 10 min. Samples were solubilized in urea mix containing 9 M urea, 4% Nonidet P-40, 2% Servalyte 9-11 (LKB Laboratories), and 2% 2-mercaptoethanol (37). Protein concentrations were determined by the method of Bradford (6). To obtain lipopolysaccharide, whole-cell lysates were treated with proteinase K by the method of Hitchcock and Brown (14) and then solubilized in hot SDS

(31).

Surface radioiodination of intact bacteria. Cells from 72-h cultures were harvested in ice-cold Dulbecco phosphatebuffered saline (36). Cells were radioiodinated by using lodogen exactly as described previously for C. jejiuni (2).

* Corresponding author.

1825

1826

INFECT. IMMUN.

DUNN ET AL.

TABLE 1. Campylobacter sp. strains studied by 2DGE Species

Strain designation

Origin

isolationf

C. jejuni

VA 81-94 (PEN 2)

Canada

Fecal

C. fetus subsp. fetus

VA 82-40 LP

Pennsylvania

Blood

C. pylori

VA 84-182 VA 85-456 VA 86-63 VA 86-242 VA 86-260 through VA 86-265 VA 86-313 VA 86-338 VA 86-385 VA 87-1 VA 87-2 VA 87-6 VA 87-129 VA 87-130 VA 87-131 VA 87-132

Texas Australia New York Minnesota France

Gastric Gastric Gastric Gastric Gastric

Colorado Colorado Colorado Colorado Colorado Colorado England Peru England Peru

Gastric Gastric Gastric Gastric Gastric Gastric Gastric Gastric Gastric Gastric

a All strains were isolated from humans.

After radioiodination and washing, cells were solubilized in mix and subjected to 2DGE. Preparation of partially purified flagella. Flagella from 72-h cultures of C. pylori were prepared by the method of DePamphilus and Adler (9), as modified by Blaser et al. (3). Formvar-coated grids containing flagella were air dried, stained for 1 min in 1% phosphotungstic acid (pH 7.0), rinsed

urea

in distilled water, and then air dried. Grids were examined with a Philips CM 12 electron microscope. 2DGE. 2DGE with isoelectric focusing in the first dimension and SDS-PAGE in the second dimension was performed as described previously (10). Briefly, isoelectric focusing was performed in the presence of 2% ampholytes (1.6% Bio-Rad BioLyte 5/7, 0.4% Bio-Rad BioLyte 3/10). For the second-dimension separation, slab gels composed of a linear polyacrylamide gradient (8 to 20%) were run in a Protean II Multi-Cell apparatus (Bio-Rad Laboratories) at 160 V for 16 to 18 h at 14°C and then fixed and silver stained (37). Autoradiography. Silver-stained 2DGE gels containing 125I-labeled proteins were soaked in 46% methanol-8% acetic acid overnight, rehydrated for 2 to 3 h in 7% acetic acid-5% glycerol, and then dehydrated for 3 to 4 h on a Bio-Rad gel dryer at 70°C. Autoradiographs were prepared by applying the dried gels to Kodak X-Omat RP film and exposed with an intensifying screen at -70°C. Immunoblotting. After selected 2DGE runs were completed, proteins were transferred to nitrocellulose paper, and immunostaining was performed as described previously (10). Rabbit and human antisera to be tested were diluted 1:200. Urease assay. Urease was assayed by measuring the rate of release of ammonia from urea. Released ammonia was converted to indophenol, and the A625 was monitored (13). One unit of urease activity was defined as the amount capable of hydrolyzing 1 p.mol of urea min-' at 22°C in 20 mM sodium phosphate-1 mM EDTA-1 mM 2-mercaptoethanol (PEB) buffer (pH 7.0) containing 200 mM urea. The specific activity of urease was calculated as units per milligram under the conditions described above. Partial purification of urease. Cells from 72-h cultures of C. pylori 85-456 were harvested in PEB buffer. Cells were

disrupted by two passages through a French pressure cell (SLM Aminco, Urbana, Ill.). After centrifugation at 10,000 x g for 20 min at 4°C, samples were concentrated by using a concentration cell with a membrane (100,000-molecularweight cut-off; Amicon Corp., Danvers, Mass.). A sample (0.5 ml) containing 300 U of urease activity was chromatographed on a 4000SW Spherogel TSK size exclusion column at 2 ml/min in PEB buffer containing 0.2 M NaCl by using a Beckman high-performance liquid chromatography apparatus. Fractions (3 ml) with the highest specific activity were lyophilized and subjected to 2DGE. Isolation of urease for antiserum production. Urease was stained in nondenaturing polyacrylamide (6%) gels by the method of Senior et al. (35) as modified by Ferrero et al. (11). Once developed, the bands of urease activity were cut from gels and stored at -20°C. Preparation of antisera. The production of rabbit antisera against C. pylori strains and against flagella of C. jejuni was

described previously (3, 30). Polyclonal antisera against urease and against proteins 2 and 3 of C. pylori were produced by solubilizing these proteins from multiple enzyme-stained bands and multiple 2DGE gels, respectively, and then injecting the solubilized protein into rabbits. Rabbits were injected subcutaneously with a total of approximately 75 Fg of the purified proteins in acrylamide in divided dosed on days 1, 21, 27, and 28. An additional intraperitoneal injection was made on day 29. One week after the final boost, rabbits were bled, and the serum was separated and stored at -20°C. Antibody production was screened as described previously (4). In addition, we used an enzymelinked immunosorbent assay (ELISA) with pooled C. pylori antigen (30) to measure serum titers in a group of 24 patients from whom histologic and culture studies were also performed to assess the possible presence of C. pylori. From these patients, we selected for immunoblotting studies four sera from persons who were infected with C. pylori and had high ELISA values and two sera from persons who were not infected and had low values. RESULTS 2DGE of whole cell preparations. The 2DGE profile of a whole-cell preparation of C. pylori 86-260 is shown in Fig. 1. Isoelectric points (pIs) ranged from 4.5 to 6.8, whereas MrS ranged from 10,000 to 120,000. Three characteristic proteins of variable intensity were observed in all 20 strains of C. pylori studied. The proteins are labeled 2, 3, and 5 in Fig. 1. All C. pyloni strains differed from C. jejuni and C. fetls strains in that they lacked major OMPs at Mr 41,000 to 45,000 (10). All C. pyloni strains lacked the high-molecularweight surface-array proteins characteristically present (28) in serum-resistant strains of C. fetus (10). The most consistent differences observed between C. pylori strains involved proteins in the regions of the brackets in Fig. 1. Proteins in this region were variable among strains of C. pylori, independent of the amount of protein added to gels. In contrast, proteins migrating at positions 7, 8, and 9 were present in all strains of C. pylori studied. Autoradiography of extrinsically labeled C. pylori strains. Autoradiographs of three strains of C. pylori (86-265, 84-182, and 85-456) are shown in Fig. 2b, c, and d, respectively. For comparison, a silver-stained whole-cell preparation of C. pylori strain 86-265 before autoradiography is shown in Fig. 2a. Proteins 2 and 3 were radiolabeled in all three strains (Fig. 2). All three strains showed radiolabeling of proteins labeled 6 and 10 (Fig. 2), which varied in relative intensity

VOL. 57, 1989

IEF

~ ~ ~ .1

ANALYSIS OF CAMPYLOBACTER PYLORI PROTEINS

---

ACIDIC SDS

I

116k1 27k66k_

- _,

45Sk36k2 9k-

20k

;.,

*

&t

"--

;-

-_

,v

-

?~JL;O

-41

-G~

14k-

FIG. 1. Silver-stained 2DGE of whole-cell preparation of C. pylori 86-260. All silver-stained gels in this study contained 60 ,ug of protein, and immunoblots contained 90 p.g, unless stated otherwise. All gels and blots are oriented with the acidic side of the gel to the left and high-molecular-weight proteins at the top. The pH gradient extended from 4.5 (left side) to 6.8 (right side) under the conditions used. The molecular weights of marker proteins in thousands are shown at the left. Numbers 1 through 9 indicate individual proteins selected for characterization as to molecular weight and pl (Table 2). Brackets show the region of greatest variability in 2DGE pattern among the 20 C. pylori strains studied.

among strains. Proteins 7, 9, and 11 were labeled variably (Fig. 2b through d). Isolated flagellar preparations. Proteins 2 and 3 and a protein below Mr 29,000 were present in preparations of partially purified flagella (Fig. 3). A more basic protein spot observed in the flagellar preparation (arrow, Fig. 3) may represent a modified form of protein 3, may have been a protein present in whole cells in a low concentration, or may have been an artifact of preparation, as no such spot was observed in the whole-cell preparation (Fig. 3). By electron microscopy, numerous flagellar filaments were evident (data not shown). 2DGE of partially purified urease. Purification by size exclusion chromatography resulted in an approximately 10fold increase in the specific activity of urease. The 2DGE profile of partially purified urease is shown in Fig. 4a. Charge trains at Mr 66,000 and 63,000 were the only significant proteins observed in silver-stained gels, corresponding in pl, Mr, and shape to proteins 2 and 3, respectively. Immunoblotting with antisera against urease. Urease isolated from nondenaturing gels showed three major bands at Mr 66,000, 62,000, and less than 30,000 by SDS-PAGE in 10% polyacrylamide gels (data not shown). Polyclonal antiserum against urease isolated from gels recognized both protein 2 and protein 3 in urease partially purified by size exclusion chromatography (Fig. 4b). No subunit with lower molecular weight was observed in either silver-stained or Western-blotted 2DGE gels of partially purified urease, however (Fig. 4). Immunoblotting of whole-cell preparations of C. pylori isolates with antiserum against urease demon-

1827

strated that proteins 2 and 3 were the major proteins recognized (data not shown). Mixing of partially purified urease with antiserum against urease (1:10 dilution of antiserum) at 4°C for 1 h had no measurable effect upon enzyme activity under the assay conditions used. Immunoblotting of whole-cell preparations. Immunoblot analysis of whole-cell preparations of C. pylori 85-456 with homologous rabbit antiserum showed that a large number of proteins are immunogenic (Fig. Sa). Immunodominant proteins included charge trains at positions 2, 3, and 5, another charge train denoted by the asterisk, and several proteins below Mr 29,000 shown in brackets (Fig. Sa). The immunogenic charge train at the asterisk probably corresponds to the silver-stained protein at position 4 in Fig. 1. A poorly focused band migrating near the bottom of the gel representing lipopolysaccharide was also immunogenic (Fig. Sa). Immunoblots of this same C. pylori strain with rabbit antisera against either C. jejuni 81-94 (Fig. Sb) or C. fetrns 82-40 LP (Fig. Sc) showed patterns that were similar to one another. Neither of the latter two antisera recognized the protein at position 2; however, both antisera recognized the charge trains at positions 3 and S and at the asterisk and recognized multiple proteins shown with the braces (Fig. Sb and c). Western blotting was more sensitive than silver staining in the detection of a variety of minor proteins of C. pylori. Antiserum against C. fetus 82-40 LP recognized a protein at p1 5.2, Mr 25,000 (arrow, Fig. Sc), which was recognized faintly by antiserum to C. jejiuni 81-94. Control immunoblots with normal rabbit serum showed no staining (Fig. Sd). Immunoblotting with rabbit antiserum against C. jejuni flagella. Figure 6 shows the results of immunoblot analyses of whole-cell preparations of C. pylori 85-456 (Fig. 6a), C. jejuni 81-94 (Fig. 6c), and C. fetis 82-40 LP (Fig. 6d) and of isolated flagella from C. pylori 85-456 (Fig. 6b) with monospecific antiserum against isolated flagella from a wild-type flagellate and motile (F+M+) C. jejuni strain (3). Against whole cells of C. pylori 85-456, the antiserum recognized several acidic charge trains at Mr 50,000 to 63,000 (Fig. 6a). The central charge train recognized (arrowhead) was the same as that denoted by the asterisks in Fig. 5a through c. There was no significant recognition of proteins at positions 2 and 3. Additional low-molecular-weight proteins were also recognized (brackets, Fig. 6a). Repeated Western blot analysis both by SDS-PAGE (data not shown) and by 2DGE (Fig. 6b) failed to demonstrate recognition of proteins in partially purified flagella of C. pylori with antisera against flagella of C jejuni. Immunoblots of whole cells of C. jejuni 81-94 (Fig. 6c) and of C. fetus 82-40 LP (Fig. 6d) showed significant staining of a broad flagellar band at Mr 61,000 to 63,000, although additional components were also recognized in C. fetuis 82-40 LP cells (Fig. 6d). Immunoblot analysis with human serum. Immunoblot analysis of C. pylori strain 85-456 was performed with reactive (Fig. 7) and nonreactive (data not shown) human sera as determined in the C. pylori ELISA (25). Sera from persons not infected with C. pylori and nonreactive in the ELISA showed no recognition of C. pylori antigens (data not shown). Positive sera recognized proteins at positions 2, 3, and S (Fig. 7). Immunostaining of many other proteins, including an acidic, low-molecular-weight spot (arrow, Fig. 7b), was variable, suggesting that the infecting strains possessed different immunogens than did the tests strains. Similar results were obtained with these four sera against 2DGE blots of C. pylori 84-182 (data not shown).

1828

INFECT. IMMUN.

DUNN ET AL. IF..I A

')

-

.. D lC IC)

1.5 A IIC.. --

k_

I&

:

9 *

I 5 k-

4;

-

v

104-

10

50-. s11

X :5k -

.-

-

_4

-

1

-

At' *_o *_ ft

6

w

.^

41

I|

1_

I

b

a

10

_

7v

2_484 6

1t-

c

d

FIG. 2. Comparison of silver-stained 2DGE of a whole-cell preparation of C. pylori 86-265 (a) with 2DGE autoradiographs of C. pylori 86-265 (b), 84-182 (c), and 85-456 (d) that were surface radioiodinated as indicated in the text. Proteins are labeled as in Fig. 1. Arrowheads in panel a show lightly silver-stained proteins that were intensely radiolabeled (see corresponding arrowheads in panels b through d).

DISCUSSION In general, the results of this 2DGE analysis of C. pyloni proteins are in agreement with those obtained with SDSPAGE of whole cells (15, 16, 18, 23, 25, 27, 29, 38). However, studies of OMPs and flagellar proteins by SDSPAGE have been limited. To date, analyses of partially purified urease by SDS-PAGE have appeared in abstract form only. Separation in two dimensions facilitated identification of the multiple proteins between Mr 45,000 and 66,000, which by SDS-PAGE alone are difficult to distinguish from one another. It is notable that the C. pylori strains studied showed remarkable similarity by 2DGE, despite their diverse geographic origins. The major conserved surface-exposed proteins observed in C. pyloni strains, which we have designated 2 and 3, also appear to be major subunits of the enzyme urease. Evidence

supporting this hypothesis is as follows: (i) 2DGE of partially purified urease showed major proteins at M, 66,000 and 63,000 with pIs and shapes corresponding to proteins 2 and 3, respectively, observed in whole-cell preparations; and (ii) polyclonal antisera against urease extracted from nondenaturing gels recognized proteins 2 and 3 as major antigens in Western blots of chromatographically enriched urease preparations and of whole-cell preparations. Surface exposure of urease in C. pylori helps to explain the following observations: (i) the kinetics of inhibition of urease activity by a variety of agents are similar in cell lysates compared with whole cells (24); and (ii) high-molecular-weight antigens enriched for urease activity serve as excellent antigens in recently described ELISAs (8; D. J. Evans, D. G. Evans, D. Y. Graham, and P. D. Klein, Proceedings of the First Meeting of the European Caimpylohbater pyloni Study Group, abstr. P51, 1988).

ANALYSIS OF CAMPYLOBACTER PYLORI PROTEINS

VOL. 57, 1989

IEFF ACIDIC SDS

I 97k -

66k -

3... 2

14.111

45k -

29k -

-e

4

FIG. 3. Silver-stained gel from 2DGE of partially purified flagella of C. pylori 85-456. Proteins are labeled as in Fig. 1. The solid arrow denotes a protein observed in the flagellar preparation which was not observed in the corresponding whole-cell preparation (not shown). The open arrow denotes a protein below M, 29,000 which was also present.

Newell (25) demonstrated that Sarkosyl-insoluble membrane proteins, acid-extracted proteins, and partially purified flagella of C. pylori contained major bands at Mr 61,000, 54,000, and 31,000 by SDS-PAGE. The band at 61,000 did not react with antiserum against C. jejuni, whereas the band at 54,000 was cross-reactive. Hawtin and Newell (P. R. Hawtin and D. G. Newell, Proceedings of the First Meeting of the European Campylobacter Pylori Study Group, abstr. 02, 1988) purified urease from C. pylori by gel filtration. The molecular weight of the intact urease molecule was approximately 512,000. SDS-PAGE analysis of purified urease revealed polypeptide bands at Mr 61,000, 54,000, and 31,000.

1829

Based on our observations, it seems likely that the proteins at Mr 61,000 and 54,000 recognized by Hawtin and Newell represent subunits of urease. The protein at Mr 61,000 described by Hawtin and Newell most likely corresponds to protein 2 in our analysis, since both are OMPs, are not recognized by antiserum against C. jejuni, and appear to be subunits of urease. The protein at Mr 54,000 described by Hawtin and Newell most likely corresponds to protein 3 in our analysis, since both are recognized by antiserum against C. jejuni and also appear to be subunits of urease. The differences in molecular weight reported by Hawtin and Newell and in the present study are probably technical in nature. No protein was recognized consistently by 2DGE corresponding to the urease component at Mr 31,000 described by Hawtin and Newell. However, we have confirmed that a protein at approximately Mr 30,000 is present in SDS-PAGE profiles of partially purified urease (G. I. PerezPerez, B. E. Dunn, and M. J. Blaser, unpublished data). The reason for this discrepancy is not known. The urease component at Mr 31,000 may be either too acidic or too basic to be resolved by 2DGE under the conditions used. Polyclonal antiserum made against proteins 2 and 3 did not recognize the protein at Mr 31,000 by either SDS-PAGE or 2DGE Western blot analysis. When proteins 2 and 3 were eluted from 2DGE gels, heated for 5 min at 100°C with dithiothreitol and SDS, and then subjected to SDS-PAGE, no significant low-molecular-weight band was observed (Perez-Perez et al., unpublished data). Thus, there is no evidence that the urease-associated protein at Mr 31,000 is a degradation product of protein 2 or 3. Analysis by 2DGE demonstrated that C. pylori strains possess an antigen at Mr 59,000 that is cross-reactive with antisera against flagella of C. jejini. However, the antigen was not present in preparations of partially purified flagella of C. pylori, but was observed in whole-cell preparations only. Using 2DGE, we were able to demonstrate that this cross-reactive antigen is distinct from protein 3 in C. pylori (Mr 63,000), since the isoelectric point of the cross-reactive protein (pl approximately 4.7; Table 2) is different from that of protein 3 (pl 5.2 to 5.5). Newell (25) performed Western blot analysis of whole cells of C. pylori with antisera against flagella of C. jejiuni. A reactive band was present between Mr

IEF F---

BASIC

ACIDIC SDS

66k-

2

-4

-

3

_ummIma

*:eb

45k-

a

b

FIG. 4. Silver-stained gel from 2DGE of 5 ,ug of partially purified urease (a) and immunoblot antiserum against urease (diluted 1:200) (b). Proteins are labeled as in Fig. 1.

analysis of partially purified

urease

using

1830

INFECT. IMMUN.

DUNN ET AL.

B At S I C'

A C I11) I(C 1) s

#6:k

-

Jl

_~~

1~ * #P# ~

.wi

r

2isk _

A7.*S * b~-6 * "1'. * *

^'! t) k L.

a

I Ps

-4

p 9

-,

...4

*40pdolol

0

Am U.

10

ow4

-0 0

. -0

- :

C

d

6

FIG. 5. Immunoblot analysis of C. pylori strain 85-456 with homologous rabbit serum (a), rabbit antiserum directed against C. jejuni 81-94 (b), rabbit antiserum directed against C. fetius 82-40 LP (c), and normal rabbit serum (d). Proteins 2, 3, and 5 are labeled as in Fig. 1. Brackets in panel a denote strongly immunogenic low-molecular-weight proteins recognized only by homologous antiserum. Braces in panels b and c show very similar patterns of recognition of proteins by the two heterologous antisera, which were also recognized by the homologous antiserum (panel a). The arrow in panel c denotes the major protein detected by C. fietts 82-40 LP antiserum, which was not recognized significantly by C. jejiuni antiserum. There was no staining reaction with normal rabbit serum (panel d).

54,000 and 61,000, but this band was not observed in silver-stained preparations of purified flagella (25), similar to results in the present study. Lee et al. (18) observed a cross-reactive protein at Mr 57,000 to 59,000 in C. pylori strains with antiserum against flagella of C. jejiunli. Neither Newell (25) nor Lee et al. (18) analyzed preparations of purified flagella by Western blots with antisera against flagella of C. jejuni, however. Taken together, the available data suggest that the antigen that is cross-reactive with flagella of C. jejuni is not associated with flagella, per se, in C. pylori. Additional methods, such as immunoelectron microscopy will be necessary to localize this cross-reactive antigen. Of interest, the two major antigens that copurified with

flagella, proteins 2 and 3, also appear to be subunits of urease. The specific activity of urease in preparations of partially purified flagella was measured to determine whether urease was present as a contaminant. The specific activity was not significantly different from that observed in French press lysates of whole cells, suggesting that urease was not selectively concentrated in preparations of partially purified flagella. Thus, either flagellar preparations were consistently contaminated with bacterial membrane fragments with associated urease activity, or, more likely, urease is associated with flagella as well as with the outer membrane of C. pylori strains. Similarly, Newell (25) demonstrated that proteins at Mr 61,000 and 54,000, shown by Hawtin and Newell to be components of urease (see above), were prominent in prep-

VOL. 57, 1989

ANALYSIS OF CAMPYLOBACTER PYLORI PROTEINS

ACXIDIC

BASIC

A C 11)1 C

I

1831

97k --i

66k

I'_-

-

45k -_

44

29k-

_iH

I

L

20k --

a 97k_ 66kMML-

p4b

TT-r-

WI

45k-

-

29k 20k-

d

c

FIG. 6. Immunoblot analysis with monospecific rabbit antiserum to C.jejuni F+M+ flagella. Antigens studied were C. pylori 85-456 (whole cell) (a), C. pylori 85-456 (isolated flagella, 5 ,ug) (b), C. jejiini 81-94 (whole cell) (c), and C. fetuis 82-40 LP (whole cell) (d). The arrowhead in panel a denotes an immunogenic protein, which corresponds to the protein shown by asterisks in Fig. Sa through c. Brackets in panel a denote proteins of low molecular weight recognized only in C. pylori strains.

arations of partially purified flagella. Goodwin et al. (12) demonstrated previously that flagella of C. pylori possess a sheath that is continuous with the bacterial outer membrane. We were not able to determine which protein(s) comprise the sheath of C. pylori. However, since urease appears to be a TABLE 2. Characterization of selected Campylob(cter proteins by 2DGE indicating pi and Mr Label"

1 2 3

Identification

pI

M,

6.1 5.6-5.8 5.2-5.5 4.7

79.000 66,000 63.000

4

Urease subunit Urease subunit Protein cross-reactive with C. jejini flagella

5 6 7 8 9

Surface-exposed protein Surface-exposed protein

5.2 5.2

54,000 44,000

5.7-6.0

24.000

Surface-exposed protein

6.1 5.8

19,000 16,000

"

Numbers correspond to those in Fig. 1.

59,000

major conserved surface-exposed OMP of C. pylori, we speculate that the enzyme may contribute to the structure of the flagellar sheath. This hypothesis is supported by the presence of urease subunits in preparations of partially purified flagella (25) (present study). Immunoelectron microscopic analyses are in progress to determine whether urease is directly associated with the flagella of C. pylori. Western blot analysis of C. pylori proteins with four ELISA-positive sera demonstrated that a variety of C. pylori proteins were immunogenic. Since protein 2 is a surfaceexposed OMP that appears to be unique to C. pylori among the Campylobacter species analyzed, this antigen may prove useful in an ELISA system to enhance the sensitivity and specificity for detection of antibodies against C. pylori. Studies to test this hypothesis are in progress. Recent studies of 16S rRNA sequence homology among Campylobacter species have demonstrated that C. pylori demonstrates greater homology with the obligate anaerobe Wolinella succinogenes than it does with other campylobacters (17, 34). In general, the present 2DGE studies are consistent with such genotypic differences. The whole-cell

1832

DUNN ET AL.

INFECT. IMMUN.

I1 F -

'-

ACIDIC

BASIC

97k66k# -*.

so*

5,C -P

2

t I

0

9 .

-

45k-

-

O

.-O

--O

29k20k-

-w

a

4

b

FIG. 7. Immunoblot analysis of C. pylori 85-456 with two ELISA-reactive human sera. Immunoreactive charge trains 2, 3, and 5 are labeled as in Fig. 1. The arrow in panel b denotes immunoreactive protein recognized faintly by the serum in panel a.

2DGE pattern of C. pylori is dissimilar to that of either C. jejuni or C. fetus; C. pylori strains lack the major OMPs at Mr 41,000 to 45,000 that are characteristic of other Campylobacter sp. (4, 19, 26) but have large amounts of surfaceexposed urease. Nevertheless, there is significant antigenic cross-reactivity among C. pylori, C. jejluni, and C. fetns. Further analysis is necessary to determine which antigens, if any, may be shared between C. pylori and W. succinogenes. Since C. pylori is able to persist in hosts who have high levels of specific serum antibodies (30), one hypothesis we considered is that these strains possess acidic high-molecular-weight surface-array proteins that could confer serum resistance, as has been described for C. fetus (5, 10, 28). However, since we could not identify such surface-array proteins, C. pylori must possess other survival mechanisms. In summary, at least one of the major proteins identified in this study (protein 2) appears to be unique to C. pylofi. The characteristics of protein 2 were as follows: (i) it copurified with urease activity and appeared to be a subunit of the urease enzyme; (ii) it was one of only two common major proteins labeled with apparent high specific activity by 1251 surface labeling and thus was surface exposed; (iii) it was immunogenic both io rabbits immunized experimentally with whole cells of C. pyl'oi and in naturally infected humans; (iv) antiserum against this protein recognized both cells and flagella of C. pylori; and (v) the protein was not crossreactive with antiserum directed against whole cells of C. jejini or C. fetius. Use of this antigen in purified form might enhance the specificity and sensitivity of ELISA systems for detection of antibody to C. pyloni in patients. Similarly, antibodies directed against this specific antigen could prove useful for the detection of C. pyloni in gastric biopsies. ACKNOWLEDGMENTS We thank Magda Altmann and Gail Campbell for excellent technical assistance and George Buck, Joel Chodos. Francis Megraud, and Stewart Goodwin for providing C. pylori strains.

This work was supported in part by Public Health Service grant BRSG-05357 awarded by the Biomedical Research Grant Program, Division of Research Resources, National Institutes of Health, by the Medical Research Service of the Veterans Administration, by an Interagency Agreement with the U.S. Army Medical Research and Development Command, and by Norwich-Eaton Pharmaceuticals, Inc. LITERATURE CITED

1. Blaser, M. J. 1987. Gastric Campylobacter-like organisms, gastritis and peptic ulcer disease. Gastroenterology 93:371-383. 2. Blaser, M. J., J. A. Hopkins, R. M. Berka, M. L. Vasil, and W.-L. L. Wang. 1983. Identification and characterization of 3. 4.

5.

6.

7.

8. 9.

Canmpylobacterjejuni outer membrane proteins. Infect. Immun. 42:276-284. Blaser, M. J., J. A. Hopkins, G. I. Perez-Perez, H. J. Cody, and D. G. Newell. 1986. Antigenicity of Campylobacterjejiuni flagellae. Infect. Immun. 53:47-52. Blaser, M. J., J. A. Hopkins, and M. L. Vasil. 1984. Campylob(ac terjejuni outer membrane proteins are antigenic for humans. Infect. Immun. 43:986-993. Blaser, M. J., P. F. Smith, J. A. Hopkins, I. Heinzer, J. H. Bryner, and W.-L. L. Wang. 1987. Pathogenesis of Campyloboctej frtus infections. Serum-resistance associated with high molecular weight surface proteins. J. Infect. Dis. 155:696-706. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. Buchanan, T. M., and W. A. Pearce. 1979. Pathogenic aspects of outer membrane components of gram-negative bacteria, p. 475-494. In M. Inouye (ed.), Bacterial outer membranes. John Wiley & Sons, Inc., New York. Dent, J. C., C. A. M. McNulty, J. S. Uff, M. W. L. Gear, and S. P. Wilkinson. 1988. Campylobacter pylori urease: a new serological test. Lancet i:1002. DePamphilus, M. L., and J. Adler. 1971. Purification of intact flagella from Esclhrichia coli and Baccillius siubtilis. J. Bacteriol.

105:376-383. 10. Dunn, B. E., M. J. Blaser, and E. L. Snyder. 1987. Two-

VOL. 57, 1989

ANALYSIS OF CAMPYLOBACTER PYLORI PROTEINS

dimensional gel electrophoresis and immunoblotting of Cainipvlobaciter outer membrane proteins. Infect. Immun. 55:15641572. 11. Ferrero, R. L., S. L. Hazell, and A. Lee. 1988. The urease enzymes of Canmpylobactle pylori and a related bacterium. J. Med. Microbiol. 27:33-40. 12. Goodwin, C. S., R. K. McCulloch, J. A. Armstrong, and S. H. Wee. 1985. Unusual cellular fatty acids and distinctive ultrastructure in a new spiral bacterium (Caimplohbacter pyloridis) from the human gastric mucosa. J. Med. Microbiol. 19:257-267. 13. Hausinger, R. P. 1986. Purification of a nickel-containing urease from the rumen anaerobe Selenionionas ruminantiuin. J. Biol. Chem. 261:7866-7870. 14. Hitchcock, P. J., and T. M. Brown. 1983. Morphological heterogenicity among Stlmoiell(a lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J. Bacteriol. 154:269-277. 15. Jones, D. M., J. Eldridge, A. J. Fox, P. Sethi, and P. J. Whorwell. 1986. Antibody to the gastric campylobacter-like organism (Campylobacter pyloridis)-clinical correlations and distribution in the normal population. J. Med. Microbiol. 22: 57-62. 16. Kaldor, J., W. Tee, C. Nicolacopolous, K. Demirtzoglon, D. Noonan, and B. Dwyer. 1986. Immunoblot confirmation of immune response to Campylobacter pvloridis in patients with duodenal ulcers. Med. J. Aust. 145:133-135. 17. Lau, P. P., B. Debrunner-Vossbrink, B. E. Dunn, K. Miotto, M. T. McDofinell, D. M. Rollins, C. J. Pillidge, Z. B. Hespell, R. R. Colwell, M. L. Sogin, C. R. Woesse, and G. E. Fox. 1987. Phylogenetic diversity and position of the genus Camnpylobacter. Syst. Appl. Microbiol. 9:231-238. 18. Lee, A., S. M. Logan, and T. J. Trust. 1987. Demonstration of a flagellar antigen shared by a diverse group of spiral-shaped bacteria that colonize intestinal mucus. Infect. Immun. 55: 828-831. 19. Logan, S. M., and T. J. Trust. 1983. Molecular identification of surface protein antigens of Catmpylobaccter jejiuni. Infect. Immun. 42:675-682. 20. Marshall, B. J., J. A. Armstrong, D. B. McGechie, and R. J. Glancy. 1985. Attempt to fullfill Koch's postulates for pyloric Cainpylobacter. Med. J. Aust. 142:436-439. 21. Marshall, B. J., D. B. McGechie, P. A. Rogers, and R. J. Glancy. 1985. Pyloric Campylobacter infection and gastroduodenal disease. Med. J. Aust. 142:439-444. 22. McNulty, C. A. M., and D. M. Watson. 1984. Spiral bacteria of the gastric antrum. Lancet i:1068-1069. 23. Megraud, F., F. Bonnet, M. Garnier, and H. Lamouliatte. 1985. Characterization of Caimpylobacter pyloridis by culture, enzymatic profile and protein content. J. Clin. Microbiol. 22:10071010. 24. Mobley, H. L. T., M. J. Cortesia, L. E. Rosenthal, and B. D. Jones. 1988. Characterization of urease from Campylobacter

1833

pylori. J. Clin. Microbiol. 26:831-836. 25. Newell, D. G. 1987. Identification of the outer membrane proteins of Campvlohbactr pyloridis and antigenic cross-reactivity between C. pyloridis and C. Jjuiuni. J. Gen. Microbiol. 133: 163-170. 26. Newell, D. G., H. McBride, and A. D. Pearson. 1984. The identification of outer membrane proteins and flagella of Caimpyobhacterjejuni. J. Gen. Microbiol. 120:1201-1208. 27. Pearson, A. D., J. Bamforth, L. Booth, G. Holdstock, A. Ireland, C. Walker, P. Hawtin, and H. Millward-Sadler. 1984. Polyacrylamide gel electrophoresis of spiral bacteria from the gastric antrum. Lancet i:1349-1350. 28. Pei, A. H., R. T. Ellison, R. V. Lewis, and M. J. Blaser. 1988. Purification and characterization of a family of high molecular weight surface array proteins from Campylobacter fetus. J. Biol. Chem. 263:6416-6420. 29. Perez-Perez, G. I., and M. J. Blaser. 1987. Conservation and diversity of Caimpylobatcer pylori major antigens. Infect. Immun. 55:1256-1263. 30. Perez-Perez, G. I., B. M. Dworkin, J. A. Chodos, and M. J. Blaser. 1988. Campvlobacter py'lori antibodies in humans. Ann. Intern. Med. 109:11-17. 31. Perez-Perez, G. I., J. A. Hopkins, and M. J. Blaser. 1985. Antigenic heterogeneity of lipopolysaccharides from Campylobacter jejiini and Campylobacter fetus. Infect. Immun. 48: 528-533. 32. Rathbone, B. J., J. I. Wyatt, and R. V. Heatley. 1986. Cainpylohblter pylori-a new factor in peptic ulcer disease. Gut 27: 635-641. 33. Rathbone, B. J., J. I. Wyatt, B. W. Worsley, L. K. Trejdosiewicz, R. V. Heatley, and M. S. Losowsky. 1985. Immune response to C. p'lori. Lancet i:1217. 34. Romaniuk, P. J., B. Zoltowska, T. J. Trust, D. J. Lane, G. J. Olsen, N. R. Pace, and D. A. Stahl. 1987. Catmpylobacter pylori, the spiral bacterium associated with human gastritis, is not a true Campylobacter sp. J. Bacteriol. 169:2137-2142. 35. Senior, B. W., N. C. Bradford, and D. S. Simpson. 1980. The urease of Protease strains in relation to virulence for the urinary tract. J. Med. Microbiol. 13:507-512. 36. Swanson, J. 1981. Surface-exposed proteins of the gonococcal outer membrane. Infect. Immun. 34:804-816. 37. Tollaksen, S. L., N. L. Anderson, and N. G. Anderson. 1984. Operation of the ISO-DALT system, 7th ed. U.S. Department of Energy Publication ANL-BlM-84-1. Argonne National Laboratory, Argonne, 111. 38. Von Wulffen, H., J. Heisemann, G. H. Butzow, T. Loning, and R. Laufs. 1986. Detection of Camplohbacter pyloni in patients with antrum gastritis and peptic ulcers by culture, complement fixation test, and immunoblot. J. Clin. Microbiol. 24:716-720. 39. Warren, J. R., and B. J. Marshall. 1983. Unidentified curved bacilli on the gastric epithelium in active chronic gastritis. Lancet i:1273-1275.