JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 1985, p. 1007-10100095-1137/85/091007-04$02.00/0Copyright © 1985, American Society for Microbiology

Characterization of "Campylobacter pyloridis" by Culture,Enzymatic Profile, and Protein Content

FRANCIS MEGRAUD,î* FRANCOISE BONNET,2 MONIQUE GARNIER,2 AND HERVÉ LAMOULIATTE3Laboratoire de Bactériologie, Hôpital des Enfants, Bordeaux, 33077 Bordeaux Cédex'; Laboratoire de Biologie Cellulaire

et Moléculaire, Institut National de la Recherche Agronomique, Pont de la Maye2; and Service de Gastroentérologie,Hôpital Saint-André, 33 Bordeaux,3 France

Received 16 May 1985/Accepted 20 August 1985

"Campylobacter pyloridis" has been recently described as a gastritis-associated bacterium. We studied 20strains. The bacteria had most of the characteristics of Campylobacter spp. strains. They were hippuratenegative and tolerant to triphenyl tetrazolium chloride (at 0.4 and 1 mg/ml). They grew on all the mediacommonly used in laboratories, although chocolate agar was the most effective for isolation. They grew in amicroaerophilic atmosphere as well as in an atmosphere enriched in C02 and when incubated at 37°C but notat 30 or 42°C. A total of 31 enzymes were present among the 78 studied. y-Glutamyl-transpeptidase activitywas, in addition to urea hydrolysis, an interesting feature for the identification of these bacteria. The proteinprofiles of the 20 strains were similar.

Campylobacters have emerged these last few years as

bacteria of great concern in the gastrointestinal tract ofhumans. Campylobacter jejuni was recognized first as amajor cause of intestinal infections worldwide, and, more

recently, campylobacterlike organisms (CLO) have beenfound: "Campylobacter fenneliae" and "Campylobactercineadi" are associated with homosexual proctitis (25);another campylobacter designated as gastric CLO (18)pyloric campylobacter (12), "Campylobacter pyloridis" (13)or "Campylobacter pylori" (6), has been associated withgastritis and peptic ulcers. Few data are presently availableconcerning this bacterium. The aim of this study was tocharacterize the organism, in particular by using growthcharacteristics, enzymatic profile, and protein content.

MATERIALS AND METHODS

Strains. We tested 20 strains. They were all isolated fromgastric antrum biopsies performed with an Olympusfibroscope on adult patients with upper gastrointestinal tractproblems. The fibroscope and forceps were disinfected aftereach use and were treated with 2% glutaraldehyde daily. Thebiopsy fragment was conserved in saline solution in a sterilecontainer at 4°C and was inoculated within 4 h.The fragment was dissociated in 0.5 ml of brucella broth

and inoculated on chocolate agar (BioMerieux, Marcyl'Etoile, France), brucella agar (Difco Laboratories, Detroit,Mich.) enriched with 10% sheep blood, and on the samemedium containing Skirrow's selective supplement (Oxoid,Basingstoke, United Kingdom). All the plates were incu-bated in a microaerophilic atmosphere (Gas Generating Kit,Anaerobic System Oxoid; in an anaerobic jar without cata-lyst) for 6 to 7 days.The strains were maintained thereafter by passage every 3

to 4 days. We were able to passage the strains more than 30times. As they adapted to culture conditions, their growthcycle became shorter. We could not study the proposed typestrain NCTC 11637 which was not available.Morphology. The strains were observed with Gram stain at

different incubation periods and by electron microscopy forselected strains by a technique previously described (3).

* Corresponding author.

Briefly, the organisms were fixed with glutaraldehyde in a

collidin buffer at a final concentration of 1% and depositedon a 2% agar. The organisms were then transferred onto a0.25% procelloidin film in amylacetate and observed directlywith a Siemens Elmiskop 101.

Biochemical and tolerance tests. The following characterswere tçsted: oxidase (Pathotec; General Diagnostics, Div.Warner-Lambert Co., Moriis Plains, N.J.), catalase produc-tion (3% hydrogen peroxide solution), and hippurate hydro-lysis with a 1% sodium hippurate solution (Sigma ChemicalCo., St. Louis, Mo.) in phosphate-buffered saline (pH 7.2;0.4-mnl volume) inoculated with a suspension equivalent to a

McFarland opacity standard of 5 and incubated 2 h in a waterbath, followed by the addition of 0.2 ml of ninhydrin.Reading was performed after 10 min. A deep blue similar tothat of crystal violet solution was considered a positivereaction (4).The production of H2S was tested on iron-bisulfite-

pyruvate (FBP) medium (24) incubated 2 h at 37°C in a waterbtth and also on blood agar over which a lead acetate paperwas suspended. Tolerance tests were performed on Mueller-Hinton agar (BioMerieux) containing 0.4 or 1 mg of triphenyltetrazolium chloride (Sigma) per ml, 0.1% sodium selenite(Sigma), 1% glycine (Sigma), 3.5% sodium chloride(Prolabo, Paris, France), or 8% glucose (Prolabo). Thesensitivity to the vibriostatic compound 0129 (InstitutPasteur Production, Marne la Coquette, France) was studiedon Mpeller-Hinton agar. AUl Mueller-HIinton agar contained5% sheep blood. All the plates were incubated at 37°C undermicroaerophilic conditions. Urea hydrolysis was performedin Christensen medium without agar (BioMerieux) and inurea-indole medium (22) (BioMerieux) incubated 2 h in a

37°C water bath. All these tests were performed with 3- to4-day-old cultures, but the results did not change when oldercultures were used.The following Campylobacter spp. strains were used as

controls: C. jejuni CIP 702, C. jejuni biotype 2, and Cam-pylobacterfetus subsp. fetus CIP 5396.Growth characteristics were tested as follows: (i) on

brucella agar with 10% sheep blood under microaerophilicconditions at 30 and 42°C, (ii) on brucella agar with 10%sheep blood at 37°C in air, in an atmosphere enriched in C02

1007

Vol. 22, No. 6

on March 28, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

1008 MEGRAUD ET AL.

i_

FIG. 1. Electron microscope photograph of "C. pyloridis," oxbow shaped (x 5,000, transfer technique).

(Gas Generating Kit, Carbon Dioxide System Oxoid), andunder anaerobic conditions, and (iii) at 37°C undermicroaerophilic conditions on tryptic soy agar with 5%sheep blood, Mueller-Hinton agar with 5% sheep blood,Columbia agar with 5% sheep blood (all these media were

from BioMerieux), Wilkins-Chalgren medium (Oxoid) with10% agar and 5% sheep blood, brain heart infusion (InstitutPasteur Production) with 10% agar and 5% sheep blood,charcoal agar (Oxoid) with 10% sheep blood, and brucellabroth (Difco).

Five strains were also passed through a membrane filter(0.65-,um pore size; Millipore Corp., Bedford, Mass.) andplated on brucella agar to see whether this technique couldbe used to select strains.Growth was recorded after 4 days (except for growth at

30°C [7 days]).Enzymatic profile. A suspension equivalent to a

McFarland standard of 7 to 8 was prepared in saline solution,and 0.1 ml was used to inoculate each of the microtubes ofstrips AP 1, 2, 3, 4, 5, and 6; AP esterase; and API Zym (APISystem, La Balme les Grottes, France). Strips AP 1 to 6contain substrates for 57 arylamidases; the AP esterase stripcontains substrates for 10 esterases. These strips are notcommercially available. For API Zym, we recorded only theresults for alkaline phosphatase, acid phosphatase,phosphoamidase, and osidases.The strips were incubated at 37°C in air for 18 h, and

reagents A (2.5 g of Tris, 1.1 ml of 37% HCI, 1 g of laurylsulfate, 10 ml of distilled water) and B (35 mg of fast blue BB,10 ml of 2 methoxyethanol) were added. Reading was

performed after 10 min. The results were recorded as eithermild or strong enzymatic activity.Every strain was tested twice on different days, and

discordant results were verified.Electrophoresis: preparation of protein samples. A washed

pellet of the strains grown 4 to 6 days on brucella agar with10% sheep blood under microaerophilic conditions at 37°Cwas suspended in 0.5 ml of sodium dodecyl sulfate buffer,consisting of 50 mM Tris hydrochloride (pH 6.8), 5% -mercaptoethanol (vol/vol), 1% sodium dodecyl sulfate(wt/vol), 15% glycerol (vol/vol), and 0.01% bromophenolblue. The homogenate was heated for 5 min in a boilingwater bath. Insoluble debris was removed by centrifugationat 10,000 x g for 5 min at 20°C. Supernatants were stored at-20°C before testing.The sodium dodecyl sulfate-polyacrylamide gel electro-

phoresis system consisted of a separating gel containing anexponential 15 to 20% acrylamide gradient with an upperstacking 5% acrylamide gel as described by Laemmli (9).Electrophoresis was performed at room temperature for 24 hat 90 V. A silver stain was used (17).

RESULTSFor isolation, chocolate agar was more effective than

brucella agar. All the strains grew on this medium, and therapidity of growth was better than on brucella agar.

Classical tests. All the strains were gram-negative rods,sometimes spiral or "ox bow" shaped (Fig. 1). The haloaround the organisms is an artifact. The organisms werelarge compared with the other Campylobacter species (size,3 to 5 by 0.5 Jxm), and their motility was weak or absentwhen they were grown on agar. The cells were eitherlophotrichate or monotrichate, although the two forms wereoften found in the same strain. Successive examinations ofthe same culture showed coccoidal transformation morethan 7 days after isolation and 2 to 3 days after a number ofpassages.The strains were oxidase and catalase positive and sensi-

tive to the vibriostatic compound 0129, and they did nothydrolyze hippurate. They did not produce H2S in FBPmedium, although we observed a blackening of the leadacetate paper. They were tolerant to triphenyl tetrazoliumchloride at 0.4 and 1 mg/ml; they did not reduce triphenyltetrazolium chloride but gave a typical metallic aspect after4 days. They were also tolerant to 0.1% sodium selenite and,to a lesser extent, to 1% glycine but not to 8% glucose and3.5% NaCl. All but one strain hydrolyzed urea in bothmedia. In Christensen medium, the color change was usuallyobserved after less than 10 min.Growth characteristics. The strains grew in an atmosphere

enriched in C02 and in a microaerophilic atmosphere butnever in air or under anaerobic conditions. The colonieswere 0.5 to 1 mm in diameter and were weakly hemolytic on10% blÔQd agar. They grew poorly at 30°C (8/20) and 42°C(6/20) as compared with 37°C. Mueller-Hinton agar, trypticsoy agar, Wilkins-Chalgren medium, and brain heart infusionmedium yielded growth comparable to that on chocolate andbrucella agar. On Columbia and charcoal agar, only a fewcolonies appeared, and they did so at a slower rate than onthe other media. There was no growth in brucella broth evenwhen it was supplemented with 10% horse serum. Thebacteria did not pass through the 0.65-,um-pore-size filter.Enzymatic profile. A total of 31 enzymes were present at

least once: 20 of 56 arylamidases, 7 of 10 esterases, 1transpeptidase, 1 alkaline phosphatase, 1 acid phosphatase,and 1 phosphoamidase (Table 1).

J. CLIN. MICROBIOL.

on March 28, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

CHARACTERIZATION OF "CAMP YLOBACTER PYLORIDIS"

TABLE 1. Enzymes found in 20 "C. pyloridis" strains

Enzyme No. of positivestrains

Arylamidase of:L-Tyrosine ......................................... 5L-Pyrrodidonyl ..................................... 6L-Lysyl ......................................... 14L-Arginyl......................................... 14L-Alanyl......................................... 2N-Benzoyl-leucinea ................................ 20S-Benzyl-cysteinea ................................. 20DL-Methionine ..................................... 17Glycyl-phenylalanine ............................... 2Glycyl-proline...................................... 14L-Glutamine ........................................ 13a-L-Glutamate...................................... 2L-Tryptophane ..................................... 10L-Alanyl-L-arginine................................. 17L-Alanyl-L-phenylalanyl-L-prolinea................. . ...............19L-Argininyl-L-arginine .............................. 3L-Phenylalanyl-L-arginine .......................... 4L-Phenylalanine-L-prolinea ......................... 20L-Prolyl-L-arginine ................................. 3L-Seryl-L-methionine ............................... 8

Esterase of:Butyrate (C4) .......... ............................. 20Valerate (C5)a ...................................... 20Caproate (C6)a ..................................... 20Caprylate (C8)a ..................................... 20Nonanoate (C9)a................................... 20Caprate (C10)a ...................................... 20Laurate (C12)a ...................................... 19

Transpeptidase ...................................... 20

Alkaline phosphatasea ................ ................ 20

Acid phosphatase"a.................................... 20

Phosphoamidase" ..................................... 20

a Strong activity.

The following enzynsi were always negative: aryla-midases of L-phenylalaninei L-hydroxyproline, L-histidine,glycine, L-aspartate, glycyl-glycine, leucyl-glycine, L-seryl-tyrosine, N-carbobenzoxy-arginine-4-methoxy, L-iSO-leucine, L-ornithine, L-proline, L-serine, L-threonine, N-carbobenzoxy-2-glycyl-glycyl-L-arginine, a-alanine, aE-L-alanyl-L-phenylalanyl-L-prolyl-L-alanine, a-L-aspartyl-alanine, a-L-aspartyl-L-arginine, a-L-glutamyl-a-L-glutamic,a-L-glutamic-L-histidine, glycyl-L-alanine, glycyl-L-arginine,glycyl-L-tryptophane, L-histidyl-L-leucyl-L-histidine, L-

histidyl-L-serine, L-leucyl-L-alanine, L-leucyl-L-leucyl, L-leucyl-L-valyl-L-tyrosyl-L-serine, L-Iysyl-L-alanine, L-lySyl-L-lysine, L-phenylalanyl-L-prolyl-L-alanine, L-valyl-L-tyrosyl-L-serine, N-benzoyl-L-alanine-4-methoxy, N-acetyl-glycyl-L-lysine, L-histidyl-L-phenylalanine, andL-lysyl-L-serine-4-methoxy; esterases of myristate (C14),palmitate (C16), and stearate (C18); and osidases of a-D-galactopyranoside, 3-D-galactopyranoside, P-D-glucuronate,a-D-glucopyranoside, ,B-D-glucopyranoside, P-D-glUCoS-aminide, a-D-mannopyranoside, and a-L-fucopyranoside.

Protein profiles. The protein profiles were similar from onestrain to another (Fig. 2) and different from the profilesobtained with reference strains of Campylobacter spp. andrepresentatives of the CLOs (data not shown). A total of 30to 35 bands were observed. There were seven major protein

bands (molecular weights: 74,000, 64,000, 58,000, 43,500,21,500, 17,000, and 12,000) always present and little varia-tion among the minor bands.

DISCUSSIONMuch interest has resulted from the discovery by Warren

and Marsiiall of a CLO on the surface of gastric epithelium(11, 26). This finding not only changed our knowledge of themicroflora of the stomach, which was thought to be sterile,but also, because of the association of this bacterium withgastritis and possibly peptic ulceration, opened a tremen-dous field for investigation. In the past, these organisms hadbeen observed, but culturing techniques were lacking (1, 19).We have been concerned with studying the characteristics of"C. pyloridis." Morphologically, this bacterium has threeparticularities when compared with the other campylobac-ters: size, after having grown on agar, which is much larger(13); flagellar arrangement, which can be lophotrichate (13,23) or monotrichate (21); and frequent presentation as an oxbow-shapeç bacterium. Similar to the other campylobacters,it was mainly spiral shaped and became coccoid in' oldcultures (1(, 13, 14).These bacteria grew on most of the blood media com-

motdly used in a microbiology laboratory, such as trypticsoy, Mueller-Hinton, brucella, Wilkins-Chalgren, brainheart infusion, and chocolate agar. However, the last one isthe most recommended for isolation if only one medium isused. Coluinbia agar and charcoal medium did not permitgood growth of the organism; Kasper and Dickgiesser (7)noted no growth on Columbia agar. We were not able togrow these bacteria in a broth medium, as noted by others (5,7, 16). Our organisms grew poorly at 30 and 40°C. Kasperand Dickgiesser (7) found no growth under these conditions,whereas Marshall et al. noted growth at 30°C (13), andLangenberg et al. noted growth at 42°C (10). These discrep-ancies are difficult to explain; differences between isolates ofdifferent geographical areas possibly exist. Concerning theincubation atmosphere, there is agreement for 5% 02 or anatmosphere enriched in C02, aerobic and anaerobic condi-tions being unsuitable.These bacteria are oxidase and catalase positive (5, 7, 10,

13, 16), hippurate negative (8, 10), and H2S positive in thelead acetate paper test (10, 13). We found them to be tolerantof triphenyl tetrazolium chloride (0.4 and 1 mg/ml), sodiumselenite (0.1%), and glycine (1%), for which Kasper andDickgiesser did not obtain uniform results (7). All the strainspossessed the three enzymes that we found in all 270 strainsof the campylobacters and CLOs tested (Z. El Harrif and F.Megraud, submitted for publication): esterases of butyricand valeric acid and L-phenylalanine-L-proline arylamidase.Alkaline phosphatase was always present in "C. pyloridis"as described by Langenberg et al. (10), but another salientfeature was the presence of -y-glutamyl-transpeptidase in allthe strains. We also found this enzyme in some strains of C.jejuni but in no other Campylobacter species or CLOs. Thischaracteristic could be helpful in the identification of "C.pyloridis." The other interesting property of the bacteriawas the presence of strong urease activity. We detected thisenzyme in Christensen medium without agar, as well as withour traditional medium derived from Ferguson medium (22).However, this characteristic is not exclusive for "C.pyloridis." Campylobacter nitrofigilis can also hydrolyzeurea (2, 15), and we also found a "C. pyloridis" strain whichdid not have this characteristic.Our protein profiles were different from those of Pearson

et al. (20), who used sonication instead of sodium dodecyl

1009VOL. 22, 1985

on March 28, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

1010 MEGRAUD ET AL.

*..~~~~~~~~~~~î.2:.~~~ ~ ~ ~ ~ ~ s,-s.e:_n-................ ___3Wp

_ i{LS ? é |Éj;j; i>xMM qw* M-e ,r .

1t,~~~ ~ ~ ~ ~ ~ ~~~~.. = = !! r V . . -;Fi w _ :mt_ 6-; -

FIG. 2. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of 20 "C. pyloridis" strains, silver stained. Numbered major proteinbands and their molecular weights: i 74,000; 2, 64,000; 3, 58,000; 4, 43,500; 5, 21,500; 6, 17,000; and 7, 12,000.

sulfate, but gave similar results such as seven major bandsand a close profile from one strain to another.

"C. pyloridis" constitutes a relatively homogeneous spe-cies, as is noted by protein profile, enzymatic studies, andclassical tolerance and biochemical tests. Its position in thegenus Campylobacter is probable. One strain has beendeposited at the Pasteur Institute Collection under the num-ber CIP 101260.

ACKNOWLEDGMENTS

We thank P. Lestage and C. Camou for excellent technicalassistance and J. P. Gayral (API System) for the strips.

ADDENDUM IN PROOF

After this paper was submitted, the name Campylobacterpyloridis was validated in the International Journal ofSystematic Bacteriology, list No. 17 (35:223-225, 1985).

LITERATURE CITED1. Freedburg, A. S., and L. E. Barron. 1940. The presence of

spirochaetes in human gastric mucosa. Am. J. Dig. Dis.7:443-445.

2. Fricker, C. R., and R. W. A. Park. 1985. Urease-positivecampylobacters. Lancet i:394.

3. Garnier, M., M. Clerc, and J. M. Bove. 1981. Growth anddivision of spiroplasmas: morphology of Spiroplasma citri dur-ing growth in liquid medium. J. Bacteriol. 147:642-652.

4. Harvey, S. M. 1980. Hippurate hydrolysis by Campylobacterfetus. J. Clin. Microbiol. 11:435-437.

5. Jones, D. M., A. M. Lessels, and J. Eldridge. 1984. Campylo-bacter-like organisms on the gastric mucosa: culture, histologi-cal, and serological studies. J. Clin. Pathol. 37:1002-1006.

6. Kasper, G., and N. Dickgiesser. 1984. Antibiotic sensitivity of"Campylobacter pylori." Eur. J. Clin. Microbiol. 3:444.

7. Kasper, G., and N. Dickgiesser. 1984. Isolation of campylobac-ter-like bacteria from gastric epithelium. Infection 12:179-180.

8. Kasper, G., and N. Dickgiesser. 1985. Isolation from gastricepithelium of campylobacter-like bateria that are distinct from"Campylobacter pyloridis." Lancet i:111-112.

9. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

10. Langenberg, M. L., G. N. J. Tytgat, M. E. I. Schipper, P. J. G.

Rietra, and H. C. Zanen. 1984. Campylobacter-like organisms inthe stomach of patients and healthy individuals. Lancet i:1348.

11. Marshall, B. J. 1983. Unidentified curved bacilli on gastricepithelium in active chronic gastritis. Lancet i:1273-1275.

12. Marshall, B. J., D. B. McGechie, G. J. Francis, and P. J. Utley.1984. Pyloric campylobacter serology. Lancet ii:281.

13. Marshall, B. J., H. Royce, D. I. Annear, C. S. Goodwin, J. W.Pearman, J. R. Warren, and J. A. Armstrong. 1984. Originalisolation of Campylobacter pyloridis from human gastricmucosa. Microbios 25:83-88.

14. Marshall, B. J., and J. R. Warren. 1984. Unidentified curvedbacilli in the stomach of patients with gastritis and pepticulceration. Lancet i:1311-1314.

15. McClung, C. R., D. G. Patriquin, and R. E. Davis. 1983.Campylobacter nitrofigilis sp. nov., a nitrogen-fixing bacteriumassociated with roots of Spartina alterniflora Loisel. Int. J.Syst. Bacteriol. 33:605-612.

16. McNulty, C. A. M., and D. M. Watson. 1984. Spiral bacteria ofthe gastric antrum. Lancet i:1068-1069.

17. Oakley, B. R., D. R. Kirsch, and N. R. Morris. 1980. Asimplified ultrasensitive silver stain for detecting proteins inpolyacrylamide gels. Anal. Biochem. 105:361-363.

18. Owen, R. J., S. R. Martin, and P. Borman. 1985. Rapid ureahydrolysis by gastric campylobacters. Lancet i:111.

19. Palmer, E. D. 1954. Investigation of the gastric spirochaetes ofthe human. Gastroenterology 27:218-220.

20. Pearson, A. D., J. Bamforth, L. Booth, G. Holdstock, A. Ireland,C. Walker, P. Hawtin, and H. Millward-Sadler. 1984. Polyacryl-amide gel electrophoresis of spiral bacteria from the gastricantrum. Lancet i:1349-1350.

21. Philipps, A. D., K. R. Hine, G. K. T. Holmes, and D. F.Woodings. 1984. Gastric spiral bacteria. Lancet ii:100-101.

22. Roland, F., D. Bourbon, and S. Sztrum. 1947. Différenciationrapide des entérobactériacées sans action sur le lactose. Ann.Inst. Pasteur (Paris) 73:914-916.

23. Skirrow, M. B. 1984. Taxonomy and biotyping, p. 33-38. InA. D. Pearson, M. B. Skirrow, B. Rowe, J. R. Davies, and D.M. Jones (ed.), Campylobacter Il. Public Health LaboratoryService, London.

24. Skirrow, M. B., and J. Benjamin. 1980. Differentiation ofenteropathogenic campylobacter, J. Clin. Pathol. 33:1122.

25. Totten, P. A., C. L. Fennell, F. C. Tenover, J. M. Wezenberg,P. L. Perine, W. E. Stamm, and K. Holmes. 1985. Campylobac-ter cinaedi (sp. nov.) and Campylobacterfenneliae (sp. nov.):two new campylobacter species associated with enteric diseasein homosexual men. J. Infect. Dis. 151:131-139.

26. Warren, J. 1983. Unidentified curved bacilli on gastric epithe-lium in active chronic gastritis. Lancet i:1273.

J. CLIN. MICROBIOL.

iwi"., f M-.4- .41,

on March 28, 2021 by guest

http://jcm.asm

.org/D

ownloaded from