Mycoplasma Genitalium, a New Species from the Human Urogenital Tract

INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, Apr. 1983, p. 387-396 Vol. 33. No. 2 0020-7713/83/020387-10$02.0010

Mycoplasma genitalium, a New Species from the Human Urogenital Tract

JOSEPH G. TULLY,'* DAVID TAYLOR-ROBINSON,3DAVID L. ROSE,' ROGER M. COLE,' AND JOSEPH M. BOVE4 Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Diseases, FCRS, Frederick, Maryland 21 701'; Laboratory of Streptococcal Diseases,' National Institute of Allergy and Infectious Diseases, Bethesda, Maryland 202052; Medical Research Council Clinical Research Centre, Northwick Park Hospital, Harrow, Middlesex, England3; and Laboratoire de Biologie Cellulaire et Moleculaire, Institut National de la Recherche Agronomique, and University of Bordeaux 11, Domaine de la Grande Ferrade, 33 140 Pont-de-la-May e, France4

Two mycoplasmas recovered from human urogenital tracts were similar in their biochemical and serological properties. These organisms possessed a unique terminal structure that appeared to be associated with attachment to tissue cells and erythrocytes. The organisms fermented glucose but did not hydrolyze urea or arginine. Growth occurred at 30 to 37°C. Cholesterol was required for growth. Unlike most other mycoplasmas, both strains were susceptible to thallium acetate. These two organisms were serologically distinct from other Mycoplasma species and from a group of unclassified serotypes of mycoplasmas. On the basis of these findings and other morphological, biological, and serological properties of the microorganisms, we propose that mycoplasmas with these characteristics belong to a new species, Mycoplasma genitalium. Strain G-37 (= ATCC 33530) is the type strain.

It is thought that the mycoplasma flora of These and other observations suggest that humans consists of 11 distinct microorganisms, other more fastidious mycoplasmas or other including nine Mycoplasma species, Achole- tetracycline-susceptible microorganisms or both plasma laidlawii, and Ureaplasma urealyticrrm might be involved in NGU and perhaps other (6). At least six of these microorganisms are clinical conditions. The availability of a new known to occur in human urogenital tracts. mycoplasma medium developed recently to cul- Mycoplasma hominis and U. urealyticum are tivate tick-derived mycoplasmas (spiroplasmas) found most frequently in urogenital tracts, and (31) was found to enhance isolation of M. each of these species has been implicated as a pneumoniae from human throat washings and to cause of some urogenital tract disease (26, 27). support growth of several other fastidious myco- Four other mycoplasmas, Mycoplasma fermen- plasmas (29). This medium (SP-4 medium) was tans, Mycoplasma primatum, Mycoplasma sali- used to examine a number of specimens from varium, and Mycoplasma pneumoniae, are patients with NGU. A mycoplasma unrelated to found only rarely in urogenital specimens, and other previously classified Mycoplasma species there is no clear evidence that these species are was isolated (27, 30). In this report, we present associated with urogenital disease. the characteristics of this microorganism. On the Although a wide variety of urogenital and basis of distinct biological and serological prop- reproductive disorders in humans have been erties, we propose that this microorganism be associated with M. hominis or U.urealyticum, it given taxonomic status as a new Mycoplasma has been difficult to establish a direct etiological species. role for these microorganisms because they often occur as frequently in patients without dis- MATERIALS AND METHODS ease as in patients with disease (26, 27). In Mycoplasmas. Two mycoplasma strains (strains G- addition, recent studies on nongonococcal ure- 37T [type strain] and M-30) were selected for charac- thritis (NGU) in humans have indicated that terization. The details of primary isolation from the although both U. urealyticum and chlamydiae urogenital tracts of two patients with NGU have been described previously (30). The other mycoplasmas are involved in this disease, in approximately used in this study, including those used as controls in 20% of the NGU cases examined no bacterial or the biochemical procedures and for serological analy- mycoplasmal agent could be recovered on the sis, were stock strains maintained in the laboratories usual laboratory media (4).Nevertheless, a proof two of us (J.G.T. and D.T.-R.). portion of patients with no apparent agent pres- Medium and cultivation. The medium used for pri- ent respond to tetracycline therapy. mary isolation and for most of the other characteriza-

387 388 TULLY ET AL. INT. J. SYST.BACTERIOL. tion procedures was SP-4 medium (31). This medium (1:2,000 to 15,000, equivalent to 0.05 to 0.02% was prepared as follows: 3.5 g of mycoplasma broth [wt/vol]) of thallium acetate and polymyxin B (500 base (BBL Microbiology Systems, Cockeysville, U/ml) on the growth of strain G-37T in SP-4 broth Md.), 10 g of tryptone (Difco Laboratories, Detroit, medium. The compounds were incorporated into the Mich.), 5.3 g of peptone (Difco), and 5 g of glucose SP-4 medium, which was distributed to a series of were placed in a 2-liter flask, and 615 ml of deionized screw-capped vials (1.8 ml in each). A 0.2-ml volume water was added. As soon as the base medium was in of a suspension of strain G-37T grown in a T-25 flask solution, it was adjusted (with 1 N NaOH) to pH 7.6 to was added to the first vial in the thallium series, the 7.8, and the preparation was then sterilized at 121°C polymyxin series, and a control SP-4 medium series. for 15 to 20 min. After cooling, the base medium was Tenfold dilutions of the mycoplasma suspension were supplemented with the following sterile components: then made, up to a lo-' dilution. The growth of strain 50 ml of tissue culture supplement CMRL 1066 (lox, G-37T in each series of vials was assessed by noting with glutamine; GIBCO Laboratories, Grand Island, pH changes (red to yellow) in the SP-4 broth medium; N.Y.), 35 ml of a fresh 25% yeast extract solution the vials were incubated at 37°C for 28 days. (Flow Laboratories, Inc., Rockville, Md.), 100 ml of a Morphology. The morphologies of the two myco- 2% TC Yeastolate solution (Difco), 170 ml of fetal plasma strains were examined by conventional dark- bovine serum (FBS) heated at 56°C for 1 h (Sterile field techniques and by electron microscopy, using Systems, Logan, Utah), 10 ml of penicillin (100,000 U/ both negative staining procedures and sectioning of ml), and 20 ml of a 0.1% phenol red solution. The final organism pellets. For negative staining, organisms pH was 7.4 to 7.5. Solid medium was prepared by were removed from the surface of a T-25 flask with a adding 0.8% Noble agar (Difco) or 0.6% agarose small amount of SP-4 broth medium and centrifuged, (Seakem ME; RMC Corp., Rockland, Maine) to the and the resulting pellet was suspended in 1% ammoni- base medium before sterilization. um acetate. The organisms were then stained with 2% The components of tissue culture supplement ammonium molybdate, as described previously (9). CMRL 1066 sometimes interfered with the biochemi- Some suspensions were initially fixed briefly in 1% cal tests. For this reason, a culture medium containing glutaraldehyde or by using glutaraldehyde vapor. For 70% mycoplasma broth base (BBL), 10% fresh yeast sectioning, organism pellets prepared from growth in a extract (25% solution; Flow), and either 20% FBS T-25 flask or strain G-37-infected Vero cell cultures (Sterile Systems) or 20% horse serum (Flow) was were fixed for 5 h at room temperature with 3% used. Medium containing FBS was supplemented with glutaraldehyde in 0.1 M phosphate buffer (pH 7.0). 0.5% glucose, 500 U of penicillin per ml, and 0.002% After two washes in buffer, secondary fixation was phenol red and was solidified by adding 0.6% agarose carried out in 1% osmium tetroxide for 5 h at room (Seakem ME). temperature or overnight at 4°C. After one or two Strains G-37T and M-30 were usually grown in 10 ml washes in Verona1 acetate buffer (18) without CaCI,, of SP-4 medium in T-25 tissue culture flasks that were the material was exposed overnight at room tempera- incubated horizontally at 37°C. As the pH moved to ture to 0.5% uranyl acetate in the same buffer and the acid range (yellow) during growth, a layer of cells subsequently enrobed in 3% Noble agar. Cubes of agar adherent to the surface of each plastic flask became were dehydrated and embedded in Spurr medium by apparent. The medium was pipetted off, and 2 ml of standard methods (9). Silver sections were cut with an fresh SP-4 medium was added to the flask. The cells LKB Ultratome 111 and stained with Reynold lead were then scraped off the surface with a sterile, bent citrate, as previously described (9). All electron mi- glass rod which had a rubber tip. The resulting cell croscopy and photography were performed by using a suspension was used to inoculate agar media or to Hitachi model HU 11-C electron microscope operated carry out other procedures in the characterization at 75 kV. process. Agar cultures were usually incubated anaero- Filtration studies. A suspension of strain G-37T cells bically in a GasPak container (BBL) at 37°C. obtained from growth in SP-4 medium in a T-25 flask Strains G-37T and M-30 were purified by a threefold for 5 days was examined for passage through a series filtration-cloning procedure (23). This involved pass- of membrane filters with graded pore diameters (aver- ing a cell suspension scraped from a T-25 flask through age pore diameters, 450, 300, 220, and 100 nm). An a membrane filter (Millex; pore size, 220 nm; Millipore unfiltered suspension (0.2 ml) was used as a control, Corp., Bedford, Mass.), inoculating agarose plates and this suspension and each filtrate were diluted in with undiluted filtrate or appropriate 10-fold dilutions 10-fold steps in 1-dram (3.75-ml) vials containing 1.8 (usually lo-' and lop2)in SP-4 broth medium, select- ml of SP-4 broth medium. The vials were incubated at ing single colonies from the resulting growth, and 37"C, and the endpoint of each titration was consid- inoculating each of these colonies into 5 ml of SP-4 ered to be the highest dilution at which the medium broth medium; this procedure was repeated two more changed color to yellow and there was granular turbid- times. ity (typical of the growth of strain G-37T). The highest SP-4 broth medium prepared as described above but dilution showing a color change contained 1 color- devoid of penicillin or any other antibiotic was used to changing unit (CCU) of mycoplasmas. detect a possible change of strain G-37T to a bacterial Tests for biological and biochemical properties. The form. At least five passages in antibiotic-free medium procedures used to determine glucose fermentation, were performed in T-25 flasks, and the cell suspen- arginine hydrolysis, urea hydrolysis, phosphatase ac- sions were inoculated onto blood agar plates and Gram tivity, tetrazolium reduction, serum liquefaction, and stained at each passage. The agar plates were incubat- film and spot reactions have been described previously ed aerobically at 37°C and examined for bacterial (2, 5). Most of these procedures and the digitonin test colonies. (13) were performed with broth or agarose media We examined the effects of different concentrations containing FBS. Hemadsorption of guinea pig erythro- VOL. 33, 1983 MYCOPLASMA GENITALIUM SP. NOV. 389 cytes to colonies of strains G-37T and M-30 on agarose Serological tests. Hyperimmune antisera to the my- medium or to growth adherent to T-25 plastic flasks coplasmas listed in Table 1 were prepared. Some of was assessed by techniques described previously (19). these antisera were antisera prepared as part of the We also examined the ability of human type 0 erythro- National Institutes of Health mycoplasma reference cytes before and after treatment with purified neur- reagent program (11). The methods used to prepare aminidase (20) to absorb to strain G-37T colonies on antisera to the remaining mycoplasmas have been agar. described elsewhere (28). Disk growth inhibition tests Sterol requirement. The sterol requirement of strain (8) were performed with suspensions of strain G-37T G-37T was determined by a technique that was modi- cells that had been grown at 37°C on the surfaces of T- fied slightly from the recommended direct broth cul- 25 plastic flasks containing SP-4 broth medium. The ture method (21), as described recently (32). Briefly, suspensions were usually diluted 150 or 1:lOO and the modified method involved growing strain G-37T in then plated onto SP-4 medium or FBS medium con- serum-free base medium to which different amounts of taining agarose. When the media appeared to be dry, cholesterol or FBS were added and comparing the paper disks saturated with 0.25 ml of undiluted myco- growth (recorded as number of CCU per milliliter) plasma antiserum were placed onto the surfaces, and with the growth in serum-free medium. the plates were incubated anaerobically at 37°C in the

TABLE 1. Mycoplasma strains and antisera used in serological tests Strains Source" St rains Source" Mycoplasma agalactiae PG2' ...... NIH Mycoplasma iouwe 69ST...... Mycoplasma alkalescens DBS803 ...... MS-LMM Mycoplasma lipophilum MaByT ...... NIH Mycoplasma alvi IlsleyT ...... MS-LMM Mycoplasma maculosum PGIST...... NIH Mycoplasma anatis 1340T ...... NIH Mycoplasmu meleagridis 17529T ...... NIH Mycoplasma arginini G230T ...... NIH Mycoplasma moatsii MK405T...... MS-LMM Mycoplasma arthritidis PG6T ...... NIH Mycoplasma molare H542T...... MS-LMM Mycoplasma bovigenitalium PG1lT...... NIH Mycoplasma mycoides subsp. Mycoplasma bovirhinis PG43T ...... NIH mycoides B3 ...... MS-LMM Mycoplasma bovis DonettaT...... MS-LMM Mycoplasma mycoides subsp. Capri PG3T . . MS-LMM Mycoplasma bovoculi M165/69T...... MS-LMM Mycoplasma neurolyticum type AT...... NIH Mycoplasma buccale CH20247T...... NIH Mycoplasma opalescens MH5408T ...... MS-LMM Mycoplasma californicum ST-6T ...... MS-LMM Mycoplasmu orale CH19299T ...... NIH Mycoplasma canadense 275CT...... MS-LMM Mycoplasma ovipneumoniae Y9tIT...... MS-LMM Mycoplasma canis PG14T ...... MS-LMM Mycoplasma pneumoniae FHT ...... NIH Mycoplasma capricolum California kidT . .MS-LMM Mycoplasma primatum HRC292T...... MS-LMM Mycoplasma caviae G122T ...... MS-LMM Mycoplasma pullorum CKKT ...... MS-LMM Mycoplasma citelli RG-2CT...... MS-LMM Mycoplasma pulmonis PG34T...... NIH Mycoplasma colurnbinasale 694T...... MS-LMM Mycoplasma putrefaciens KS-lT ...... MS-LMM Mycoplasma columbinum MMP-lT...... MS-LMM Mycoplasma salivarium PG20T...... NIH Mycoplasma columborale MMP-4T, ...... MS-LMM Mycoplasmu spumans PG13T ...... NIH Mycoplasma conjunctivae HRC581T...... MS-LMM Mycoplasma sualvi Mayfield BT...... MS-LMM Mycoplasma cynos H831T...... MS-LMM Mycoplasmu subdolum TBT ...... MS-LMM Mycoplasma dispar 462/2T ...... MS-LMM Mycoplasma synoviae WVU 1853T ...... MS-LMM Mycoplasma edwardii PG24T ...... MS-LMM Mycoplasma ve recundum 107T ...... M S- LMM Mycoplasma equigenitalium T37T ...... MS-LMM Mycoplasma sp. primate strain B-291 .....MS-LMM Mycoplasma equirhinis M432/72T ...... MS-LMM Mycoplasma sp. bovine strain California Mycoplasma fastidiosum 4822T ...... MS-LMM calf ...... M S-LMM Mycoplasma faucium DC333T ...... NIH Mycoplasma sp. bovine strain B5P Mycoplasma feliminutum BenT ...... MS-LMM (group 7) ...... MS-LMM Mycoplasma felis Cat 27 ...... MS-LMM Mycoplasma sp. strain 70-159...... MS-LMM Mycoplasma fermentans PG18T...... NIH Mycoplasma sp. caprine strain G-145 .....MS-LMM Mycoplasma Jlocculare M~42~...... MS-LMM Mycoplasma sp. plant strain Melaleuca ...MS-LMM Mycoplasma gallinaceum DDT...... MS-LMM Mycoplasma sp. plant strain 831-C4 ...... MS-LMM Mycoplasma gallinarum PG16T ...... NIH Acholeplasma axanthum H86N ...... MS-LMM Mycoplasma gallisepticum PG31T...... NIH Acholeplasrna equifetale N93 ...... MS-LMM Mycoplasma gallopavonis WRIT ...... MS-LMM Acholeplasma granularum BTS39T ...... MS-LMM Mycoplasma gateae Mart ...... MS-LMM Acholeplasma hippikon CIT ...... MS-LMM Mycoplasma hominis PG21T, ...... NIH Acholeplasma laidlawii PGST ...... MS-LMM My coplasma hy opneumonia e JT...... M S-LM M Acholeplasma modicum PG49T...... MS-LMM Mycoplasma hyorhinis BTS7T ...... NIH Acholeplasma morum 72-043T...... MS-LMM Mycoplasma hyosynoviae S16T ...... MS-LMM Acholeplasma oculi 19LT...... MS-LMM Mycoplasma hers PG30T ...... NIH Acholeplasma sp. plant strain L1, ...... MS-LMM

a NIH, Reference reagent collection of the National Institute of Allergy and Infectious Diseases (11); MS- LMM, Mycoplasma Section, Laboratory of Molecular Microbiology, National Institute of Allergy and Infectious Diseases. 390 TULLY ET AL. INT. J. SYST.BACTERIOL.

GasPak system. Zones of colony inhibition around the late or fibrillar layer, 8 to 12 nm thick, which disks were recorded after an incubation period of 14 extended outward from the characteristic single days. The plate immunofluorescence antibody test limiting membrane of the cell (Fig. 1B and C). In was performed by a direct method (12), using myco- negatively stained preparations of naturally or plasma colonies which had developed on SP-4 or FBS sonically disrupted organisms (Fig. 1D through agarose media. The colonies were stained with appro- F) priate dilutions (usually 1:lO to 1:60) of fluorescein- or in sections (Fig. 1C) of some organisms conjugated antisera to the mycoplasmas listed in Table (most were uniformly densely stained), we ob- 1. served a fibrillar rodlike structure or terminal Deoxyribonucleic acid. The guanine-plus-cytosine organelle which was approximately 210 to 250 content of purified deoxyribonucleic acid of strain G- nm long and from 35 to 60 nm wide. The 37T was determined by buoyant density, which was proximal end of this structure was surmounted measured by equilibration centrifugation in a CsCl by a slightly wider cap (67 to 80 nm) (Fig. 1B gradient, using techniques described previously (17). through F). The ultrastructure of the mycoplas- Purified deoxyribonucleic acid from Spiroplasmu citri ma, whether unfixed, glutaraldehyde fixed, neg- (guanine-plus-cytosine content, 26 mol%) was used as a reference in this procedure (17). atively stained, or fixed and sectioned or wheth- Tissue culture studies. Attachment of strains G-37T er from broth cultures or from infected cell and M-30 to tissue cells was examined by using Vero cultures was the same. monkey kidney cells (received from M. F. Barile, Cultural properties. Cloned and culture-adapt- Food and Drug Administration, Bethesda, Md.). Por- ed strains G-37T and M-30 grown in SP-4 broth tions (0.1 ml) of a 5-day culture of each mycoplasma medium in glass tubes or vials produced a slight- strain in SP-4 medium were inoculated into 35-mm ly granular turbidity and a lowering of the pH of plastic dishes containing 1-day-old monolayers of the medium after about 4 to 5 days at 37°C. Vero cells on glass cover slips covered by Eagle When the broth was removed from the tube or minimum essential medium. After incubation at 37°C for 5 to 7 days, the monolayers were examined by vial, a ring of material adherent to the glass was optical microscopy for cytopathic changes. In addi- apparent, suggesting that the cells could attach tion, infected tissue cells on cover slips were fixed in to this surface. Inoculation of the mycoplasmas 3% glutaraldehyde for 5 to 18 h at room temperature. into T-25 plastic flasks containing 10 ml of SP-4 The cells were then scraped off the cover slips, centri- broth medium, followed by incubation at 37°C in fuged at 3,000 X g for 5 min, washed in phosphate a horizontal position, resulted in a similar buffer, and secondarily fixed in 1% osmium tetroxide; change in the pH of the medium and a sheet of the cell pellets were then washed, stained, embedded, attached mycoplasma cells on the surface of the sectioned, and examined by electron microscopy, us- plastic after 5 days (Fig. 2A). Similar but less ing the techniques described above. marked growth occurred at 30°C (30). The two mycoplasma strains also grew well, at least after RESULTS adaptation to the SP-4 medium, in conventional The following description covers the proper- mycoplasma broth medium containing 20% ties of both strain G-37= and strain M-30, except FBS. However, attachment to plastic or glass where noted otherwise. surfaces occurred only with strain G-37T when Morphology and ultrastructure. Dark-field mi- the organisms were grown in FBS medium. croscopy of cultures showed small, discrete, When conventional mycoplasma medium sup- apparently nonmotile, coccoid bodies; no details plemented with 20% horse serum was used, of these bodies could be resolved. Gram staining strain G-37T grew only in suspension and did not revealed small coccoid to irregular gram-nega- attach to the surfaces of the T-25 plastic flasks. tive units. Electron microscopy of either nega- Strain M-30 did not grow or attach to surfaces tively stained (Fig. 1A) or sectioned (Fig. 1B and when it was cultured in the medium containing C) material showed that the mycoplasma cells horse serum. Because few colonies of strain G- were predominantly flask shaped, although 37T or M-30 were apparent when those orga- stages of rounding, blebbing, vacuolization, and nisms were grown on SP-4 solid medium supple- disruption were observed as cultures aged. The mented with 0.8% Noble agar, a more highly dimensions of the cells were as follows: length, purified agar preparation (0.6% agarose) was 0.6 to 0.7 pm; width, 0.3 to 0.4 pm at the used. On this medium, both strains produced broadest part and 0.06 to 0.08 Fm at the tip. The colonies, which varied in diameter from 20 to tip was often slightly broadened or truncate. A 200 pm. Although a few colonies had a “fried- characteristic feature was a prominent outer egg” appearance, most colonies appeared to layer or nap extending distally from the tip for grow into the agarose and exhibited little surface only approximately 40 to 60% of the cell length. growth (Fig. 2B). More successful colony devel- In negatively stained preparations, the nap ap- opment occurred on SP-4 agarose medium incu- peared to be discrete 78-nm irregular particles or bated at 37°C in the GasPak system for 14 days. aggregates (Fig. 1A and D through F), but in The inhibitory effects of different concentra- sections we observed a continuous finer particu- tions of thallium acetate are shown in Table 2. VOL. 33, 1983 MYCOPLASMA GENITALIUM SP. NOV. 391

FIG. 1. Electron micrographs of M. genitaliurn. (A) Negative staining of an intact mycoplasma cell from a 5- day culture in SP-4 medium. (B and C) Sections from infeded Vero cell cultures after 5 days. (D through F) Negative staining of disrupting mycoplasmas from a culture. Note the flask-shaped cells, the truncated tip, the nap (N) on the terminus extending distally to the area marked by unlabelled arrowheads, the blebs and vacuoles of unfixed negatively stained organisms (A and E), the internal terminal organelle consisting of a rod (R) and a cap (C), the sites of cytadsorption (AD) by the nap, and the characteristic single membrane (M).Bar = 0.1 pm. 392 TULLY ET AL. INT. J. SYST.BACTERIOL.

FIG. 2. (A) Growth in flasks. The top flask was a Corning T-25 plastic flask containing M. genitafiurn strain G- 37T cultured for 5 days in SP-4 medium, showing adherent growth stained with Crowle triple stain (10). The bottom flask was a control containing only medium and stained with the same stain. (B) Colonies of M. genitalium grown for 14 days anaerobically on SP-4 medium containing 0.6% agarose. Bar = 200 pm.

The concentrations of thallium used in most bacteria in clinical specimens cultured in SP-4 mycoplasma media to suppress bacterial growth broth medium. (1:2,000 to 1:3,000) inhibited the growth of strain Reversion studies. Cultures of strain G-37T G-37T in SP-4 broth medium. A concentration of maintained in antibiotic-free SP-4 broth medium thallium acetate of 1 :4,000 was partially inhibi- for five to eight consecutive passages exhibited tory, and only when a dilution of 1:5,000 was growth characteristics similar to those observed used was the growth of strain G-37'r comparable when penicillin was present. No evidence of a to the growth in SP-4 medium without thallium. change to bacterial forms was apparent when the The effects of polymyxin B on the growth of antibiotic-free broth cultures were inoculated strain G-37Tin SP-4 broth medium indicated that onto blood agar or when Gram-stained cells a concentration of 500 U/ml was not inhibitory were examined microscopically. (Table 2), suggesting that this compound might Filtration studies. A cloned broth culture of be a suitable replacement for thallium acetate in strain G-37= contained lo8 CCU/ml before filtra- suppressing contamination by gram-negative tion through a series of membrane filters having

TABLE 2. Effects of thallium acetate and polymyxin B on the growth of strain G-37T No. of CCU/ml Length of No. of CCU/ml in 10-fold dilutions of a strain G-37T culture grown with different in 10-fold incubation concn of thallium acetate and polymyxin B: dilutions at 37°C in SP-4 Thallium acetate concn Polymyxin B (days) medium alone 1:2,000 1 :3,000 1:4,000 1: 5,000 (500 U/ml) 7 103 Neg" Neg 10' 1o2 10' 14 105 Neg N eg 1 o2 103 1o5 21 107 Neg 1o2 1o4 1Oh 1Oh 28 107 Neg 1o2 1o5 1Oh 1oh Neg, Negative. VOL. 33, 1983 MYCOPLASMA GENITALIUM SP. NOV. 393 graded pore diameters. After passage through No significant growth was apparent in the base membrane filters having average pore diameters medium alone or when cholesterol supplements of 450, 300, 220, and 100 nm, the filtrates con- (1 to 20 kg/ml) were added to the base medium. tained lo5,lo4, lo2,and 0 CCU/ml, respectively. However, there was increased multiplication Biochemical and biological properties. The re- when 15 to 20% FBS was incorproated into the sults of various biochemical tests performed on base medium. the two mycoplasma strains and other biological Serological studies. Growth inhibition and epi- properties determined in this study were as immunofluorescence tests with antisera or con- follows: fermentation of glucose, positive; hy- jugates prepared to the mycoplasmas listed in drolysis of arginine, negative; hydrolysis of Table 1 indicated that strains G-37T and M-30 urea, negative; phosphatase production, nega- were serologically unrelated to mycoplasmas tive; tetrazolium reduction (aerobic), weak; tet- described previously, including both fermenta- razolium reduction (anaerobic), positive; lique- tive and nonfermentative Mycoplasma and faction of coagulated serum, negative; film and Acholeplasma species. spot reaction, negative; susceptible to 1.5% digi- Deoxyribonucleic acid. The guanine-plus-cyto- tonin (diameter of inhibition zone, 7 mm); hem- sine content of deoxyribonucleic acid obtained adsorption (guinea pig or human type 0 erythro- from strain G-37T was 32.4 5 1.0 mol%. cytes), positive; sterol required for growth; Tissue culture studies. Infection of Vero mon- culture turbidity, granular; adherent to glass and key kidney cell monolayers with the mycoplas- plastic surfaces; cytopathic to tissue culture mas resulted in the appearance of foci of vacuoli- cells (Vero monkey); colonies on agarose vari- zation and cell lysis after 5 days at 37"C, as able in size, most without a central zone; pre- observed by optical microscopy. An examina- ferred atmosphere in broth, aerobic; preferred tion of sections by electron microscopy verified atmosphere on agar, anaerobic; preferred tem- the cytopathogenic changes and showed that perature, 37°C. The most notable characteristics there were numerous mycoplasmas closely asso- were fermentation of glucose and the strong ciated with both intact and disrupting cells (Fig. adsorption of erythrocytes (guinea pig or human 3). Adherence of the mycoplasmas to the host type 0) to colonies of strains G-37T and M-30. cell membranes appeared to be mediated only by Treatment of erythrocytes (human type 0) with the nap on the tip and anterior (proximal) two- purified neuraminidase completely abolished ad- thirds of the cells (Fig. 1B and C). A similar role sorption of these cells to colonies of strain G- of the nap was observed in sections of the 37T. mycoplasmas adhering to erythrocytes (data not Sterol requirement. The growth responses of shown). strain G-37* to cholesterol and to FBS added to serum-free base medium are shown in Table 3. DISCUSSION A preliminary analysis of some of the morphological, biological , and serological characteristics of the two new genital tract mycoplasma TABLE 3. Effects of cholesterol and serum on the growth of strain G-37T strains has been presented previously (27, 30). The most unusual characteristics of this orga- No. Of (CCU'rnl) Addition of cho- after incubation of nism are its differentiated terminal structure and lesterol and FBS Concn strain G-37~I at 37°C for: its ability to attach to tissue cells and to erythro- to base medium cytes, apparently by the restricted nap on this 14 Days 21 Days structure. None" NG~ 10' The morphology and ultrastructure of the new None'' NG 10' mycoplasma resemble most closely the mor- Cholesterol" 1 pg/ml NG 10' phology and ultrastructure of M. pneumoniae (3, 5 pglml 10' 10' 32), although Mycoplasma gallisepticum (l), 10 pg/ml 10' 10' Mycoplusmu alvi (14), and Mycoplasma sualvi 20 pg/ml 10' 10' FBS" 0.5% NG NG (15) are also flask shaped and possess terminal 5% 10' 10' organelles of some sort. Two of these species 10% lo1 1o3 (M. pneumoniae and M. gallisepticum) are 15% 104 1o7 pathogenic and are known to attach to glass (22, 20% 1o8 1ox 25), to hemadsorb (19,20), and to cytadsorb (24, ~~ ~ 32-34); however, unlike the apparently non- ' Serum-free base medium alone. M. M. sualvi, M. NG, No growth. pathogenic species alvi and ' pneumoniae M.gallisepticum c' Base medium supplemented with 0.5% albumin, and have not been 0.01% Tween 80, and 10 kg of palmitic acid per ml. found in the urogenital tracts of their hosts. A All medium preparations contained 0.5% albumin, surface nap has been observed on M. gallisepti- 0.01% Tween 80, and 10 pg of palmitic acid per ml. cum (7), but apparently this nap is not restricted 394 TULLY ET AL. INT. J. SYST. BACTERIOL.

FIG. 3. Electron micrograph of a section of a Vero cell culture infected with M. genitalium for 5 days. Note the small dark mycoplasmas and the stages of cell destruction. Bar = 5 pm. to the terminal bleb, which is said to mediate define the true habitat. However, the mycoplas- cytadsorption (24, 33). No surface layers have mas recovered from two patients with NGU are been defined on M. alvi and M. sualvi, and the important additions to the mycoplasma flora of status of hemadsorption and cytadsorption for the human urogenital tract, Although the data these two species is unknown. When originally presented here and elsewhere (30) suggest that reported (27, 30), the new mycoplasma was the SP-4 medium is a crucial factor in recovery of first example of a structurally differentiated my- the organisms, the clear demonstration of the coplasma that adheres to cells by a clearly inhibitory activity of thallium acetate (a compo- defined nap restricted to its terminal region. nent frequently utilized in mycoplasma media to After this original report, a similar but more suppress growth of urogenital tract and oral restricted terminal nap was observed on M. cavity bacteria) indicates that the absence of pneumoniae, and this nap has been shown by thallium in the SP-4 medium used for primary labeling with monoclonal antibody (16) to in- isolation was an important factor. There are not clude or consist of P1, a trypsin-sensitive sur- enough data to relate this microorganism to face protein responsible for the previously ob- NGU or to other urogenital infections, but its served (32) terminus-mediated cytabsorption of presence in urethras, its isolation on special this mycoplasma. The nature of the nap of the media, its specialized structure, and its capacity new genital mycoplasma has not been deter- to adhere to cells and produce cytopathic effects mined yet. suggest that it may have a role in human disease. The recovery of the organisms from human Efforts to define the distribution of this myco- genital tracts is not proof that this site represents plasma and to clarify its role in human disease the primary habitat of this mycoplasma. Further are in progress. attempts to isolate the organism from other Our evaluation of the properties of the micro- anatomical sites in humans by using the methods organisms described above fulfills the essential described previously (30) will be necessary to criteria for mycoplasmas, including absence of VOL. 33, 1983 MYCOPLASMA GENITALIUM SP. NOV. 395

cell walls, filterability, lack of reversion to bac- National Institutes of Health, Bethesda, Md. teria with cell walls when grown in antibiotic- 12. Del Gludlce, R. A., N. F. Robward, and T. R. Cprski. 1%7. Immunofluorescence identification of Mycoplasma free medium, penicillin resistance, and produc- on agar by use of incident illumination. J. Bacteriol. tion of typical colonies on agarose. The two 93: 1205-1 209. strains which we studied are serologically dis- 13. Freundt, E. A., B. E. Andrews, H. Erne, M. Kunze, and tinct from other members of species, subspe- F. T. Black. 1973. The sensitivity of Mycoplasmatales to sodium-polyanethol-sulfonate and digitonin. Zentralbl. cies, or unclassified serotypes of the order My- Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1 Orig. coplasmatales. The growth requirement of these 255: 104-1 12. strains for sterol and their morphology place 14. Gourlay, R. N., S. G. Wyld, and R. H. Leach. 1977. these microorganisms in the family Mycoplas- Mycoplasmu alui, a new species from bovine intestinal and urogenital tracts. Int. J. Syst. Bacteriol. 27:89-%. mataceae. An inability to hydrolyze urea indi- 15. Gourlay, R. N., S. G. Wyld, and R. H. Leach. 1978. cates that the organism should be assigned to the Mycoplasma sualvi, a new species from the intestinal and genus Mycoplasma. We believe that this micro- urogenital tracts of pigs. Int. J. Syst. Bacteriol. 28:289- organism represents a new species within this 292. 16. Hu, P. C., R. M. Cole, Y. S. Huaq, J. A. Graham, D. E. genus and, therefore, propose the name Myco- Gardner, A. M. Collier, and W. A. Clyde, Jr. 1982. Myco- plasma genitalium (ge. ni. ta’ lium. L. n. genita- plasma pneumoniae infection: role of a surface protein in lis genitals; genitalium pertaining to the geni- the attachment organelle. Science 216313-315. tals). Strain is the type strain. A cloned line 17. Junca, P., C. Saillard, J. Tully, 0. Garcla-Jurado, J.-R. G-37 Degorce-Dumas, C. Mouches, J.-C. Viiult, R. Vogel, R. of this strain has been deposited in the American McCoy, R. Whitcomb, D. Williamson, J. htrllle, and Type Culture Collection as strain ATCC 3353OT. J. M. BovC. 1980. Caracterization de spiroplasmes isoles d’hsectes et de fleurs de France continentale, de Corse et ACKNOWLEDGMENTS du Maroc. Proposition pour une classification des spiro- R. 290:1209-1212. We thank Patricia Carle (University of Bordeaux, Pont-de- plasmes. C. Acad. Sci. la-Maye, France) for assistance in the deoxyribonucleic acid 18. Kellenberger, E., A. Ryter, and J. Sechaud. 1958. Electron studies. We also thank Colis Blood (Laboratory of Molecular microscopy study of DNA-containing plasms. 11. Vegeta- Biology, National Institute of Allergy and Infectious Diseases, tive and mature phage DNA as compared with normal Frederick, Md.) and Pat Furr (Medical Research Council bacterial nucleoids in different physiological states. J. 4:671-678. Clinical Research Centre, Northwick Park Hospital, Harrow, Biophys. Biochem. Cytol. 19. Manchee, R. J., and D. Taylor-Roblnson. 1968. England) for technical assistance. We also thank Terry Popkin Haemad- (Laboratory of Streptococcal Diseases, National Institute of sorption and haemagglutination by mycoplasmas. J. Gen. Allergy and Infectious Diseases, Bethesda, Md.) for assist- Microbiol. 50:465-478. 20. Manchee, R. J., and D. Taylor-Robinson. 1%9. ance in electron microscopy. Utilization of neuraminic acid receptors by mycoplasmas. J. Bacteri- LITERATURE CITED 01. 98:914-919. 21. Razin, S., and J. G. Tully. 1970. Cholesterol requirement 1. Allen, T. C., J. 0. Stevens, E. R. Fhrance, and R. 0. of mycoplasmas. J. Bacteriol. 102306-310. Hampton. 1970. Ultrastructure of Mycoplasma gallisepfi- 22. Somerson, N. L., W. D. James, B. E. Walls, and R. M. cum isolate 1056. J. Ultrastruct. Res. 33318-331. Chanock. 1%7. Growth of Mycoplasma pneumoniae on a 2. Aluotto, B. B., R. G. Wlffler, C. 0. Williams, and J. E. glass surface. Ann. N. Y. Acad. Sci. 143384-389. Fa&. 1970. Standardized bacteriologic techniques for the 23. Subcommittee on the Taxonomy of Mollicutes. 1979. Pro- characterization of Mycoplasma species. Int. J. Syst. posal of minimal standards for description of new species Bacteriol. m35-58. of the class Mollicutes. Int. J. Syst. Bacteriol. 29172-180. 3. Biberfeld, G., and P. Biberfeld. 1970. Ultrastructural 24. Tajima, M., T. Nunoya, and T. Yagihashi. 1979. An features of Mycoplasma pncumoniae. J. Bacterioi. ultrastructural study of the interaction of Mycoplasma 102~855-861. gafliseppticum with the chicken tracheal epithelium. J. Vet. 4. Bowie, W. R., R. Alexander, J. B. Stimson, J. F. Floyd, Res. 40:1009-1014. and K. K. Holmes. 1981. Therapy for nongonococcal 25. Taylor-Robinson, D., and R. J. Manchee. 1%7. Adherence urethritisdouble blind randomized comparison of 2 doses of mycoplasmas to glass and plastic. J. Bacteriol. W1781- and 2 durations of minocycline. Ann. Intern. Med. 1782. 95:306--311. 26. Taylor-Robinson, D., and W. M. McCormack. 1979. My- 5. Bradbury, J. M. 1977. Rapid biochemical tests for charac- coplasmas in human genito-urinary infections, p. 307-366. terization of Mycoplasmatales. J. Clin. Microbiol. 5531- In J. G. Tully and R. F. Whitcomb (ed.), The mycoplas- 534. mas, vol. 2. Human and animal mycoplasmas. Academic 6. Cassell, G. H., and B. C. Cole. 1981. Mycoplasmas as Press, Inc., New York. agents of human disease. N. Engl. J. Med. 304:80-89. 27. Taylor-Robinson, D., J. G. Tully, P. M. Furr, R. M. Cole, 7. Chu, H. P., and R. W. Horne. 1%7. Electron microscopy D. L. Rose, and N. F. Hanna. 1981. Urogenital mycoplas- of Mycoplasma gallisepticum and Mycoplasma mycoides ma infections of man: a review with observations on a using the negative staining technique and their compari- recently discovered mycoplasma. Isr. J. Med. Sci. son with Myxovirus. Ann. N.Y. Acad. Sci. 143:190-203. 17524-530. 8. Clyde, W. A., Jr. 1964. Mycoplasma species identifica- 28. Tully, J. G., M. F. Barile, D. G. if. Edward, T. S. Theo- tion based upon growth inhibition by specific antisera. J. dore, and H. Erne. 1974. Characterization of some caprine lmmunol . 92:958-%5. mycoplasmas, with proposals for new species, Mycoplas- 9. Cole, R. M., and T. J. Popkin. 1981. Electron microscopy ma capricolum and Mycoplasma putrefaciens. J. Gen. and ultrastructure of bacteria, p. 34-51. In P. Gerhardt Microbiol. 85:102-120. (ed.), Manual of methods for general bacteriology. Ameri- 29. Tully, J. G., D. L. Rose, R. F. Whitcomb, and R. P. can Society for Microbiology, Washington, D.C. Wenzel. 1979. Enhanced isolation of Mycoplasma pneu- 10. Crowle, A. J. 1961. Immunodiffusion. Academic Press, moniae from throat washings with a newly modified Inc., New York. culture medium. J. Infect. Dis. 139:478-482. 11. Cunningham, S., and J. E. Nutter (ed.). 1975. NIAID 30. Tully, J. G., D. Taylor-Robinson, R. M. Cole, and D. L. catalog of research reagents. Publication (NIH) 75-899. Rose. 1981. A newly discovered mycoplasma in the human 3% TULLY ET AL. INT. J. SYST.BACTERIOL.

urogenital tract. Lancet i:1288-1291. rial. 32:92-100. 31. Tully, J. G., R. F. Whitcomb, H. F. Clark, and D. L. 33. Wilson, M. H., and A. M. Collier. 1976. Ultrastructural Williamson. 1977. Pathogenic mycoplasmas: cultivation study of Mycoplasma pneumoniae in organ culture. J. and vertebrate pathogenicity of a new spiroplasma. Sci- Bacteriol. 125332-339, ence 195892-894. 34. Zucker-Franklin, D., M. Davidson, and L. Thomas. 1966. 32. Tully, J. G., R. F. Whitcomb, D. L. Rose, and J. M. Bod. The interaction of mycoplasmas with mammalian cells. I. 1982. Spiroplasma mirum, a new species from the rabbit HeLa cells, neutrophils, and eosinophils. J. Exp. Med. tick (Haemaphysalis leporispalusrris). Int. J. Syst. Bacte- 124:521-532.