Comparison of Mycoplasma agalactiae isolates by pulsed field gel electrophoresis, SDS-PAGE and immunoblotting

ELSEVIER FEMS Microbiology Letters 143 (1996) 259-265 MICROBIOLOGY LETTERS Comparison of Mycoplasma agalactiae isolates by pulsed field gel electrophoresis, SDS-PAGE and immunoblotting Abstract Sebastiana Tola a,*, Graziano Idini a, Daniela Manunta a, Ida Casciano b, Angela M. Rocchigiani a, Antonio Angioi a, Guido Leori a a Istituto Zooprofilattico Sperimentale della Sardegna G. Pegrefi, Via Duca degli Ahruzzi 8, 07100 Sassari. Italy h Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy Received 30 July 1996; accepted 1 August 1996 We have analyzed 8 1 isolates of h4ycoplasma agahctiae from four different regions of Italy between and 1995 in order to identify antigenic differences through SDS-PAGE and Western blotting and chromosomal DNA restriction endonuclease cleavage pattern differences. Antigenic variability in M. agalactiae isolates was investigated analyzing hydrophobic membrane protein fractions by immunoblotting using pooled sheep antiserum from naturally infected sheep. Large restriction fragments obtained cleaving genomic DNAs with SmaI, NruI, MI, XhoI, BssHII and KpnI were analyzed by pulsed field gel electrophoresis. Genetic analysis indicates that isolates are all similar without intraspecific differences. This homogeneity was confirmed by immunoblotting: 80 and 50 kda antigens are present in all strains analyzed. Keywords: Mycoplasma agalactiae; Epidemiological analysis; Pulsed field gel electrophoresis; Immunoblotting 1. Introduction Mycoplasma agalactiae is the causative agent of contagious agalactia (C.A.) in sheep and goats [1,2]. C.A. is particularly widespread around the Mediterranean basin where this disease causes considerable economic damage [3]. In Sardinia, the disease was unknown until 1980 when it was introduced by sheep from the Italian mainland [4]. Control of C.A. is hindered by breeding practices (sharing of grazing grounds, uncontrollable movements of animals), by inadequate compliance with prophylactic * Corresponding author. Tel.: +39 (79) 216914; Fax: +39 (79) 229458. measures and by the lack of effective antimicrobial agents. Therefore infections caused by M agalactiae are responsible for considerable economic losses, in the order of 6 million US dollars per year. When the infection started spreading in Sardinia mainly acute and subacute clinical symptoms were observed, e.g. agalactia, mastitis, arthritis and keratoconjunctivitis. However, recently it has been possible to observe less severe clinical forms, without keratitis and arthritis, typical of an endemic situation. A reasonable strategy for control and eradication of this disease is possible only in the presence of accurate epidemiology of M. agalactiae. The possibility to identify and differentiate different strains should facilitate this aim. 0378-1097 /96/ $12.00 Copyright 0 1996 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII SO378-1097(96)00323-O

Table I M~~cophsmu uguh/iar isohes analyzed in this stud) Isolate Geographic origin Clinical condition Year of isolation PG2, type strain (bga-jena) SS 50.53,59,61,65,12,15.78 Sardmia (Sassari) mastitis. arthritis, keratitis NU 2.16.40.41 Sardinia (Nuoro) mastitis, arthritis, keratitis CA 3,8,12 Sardinia (Cagliari) mastitis, arthritis. keratitis OR 9.15,20 Sardinia (Oristano) mastitis, arthritis. keratitis SS 50,53,59,61,65,72,75.1~~~ Sardinia (Sassari) mastitis. keratitis SS 80 I 84 > 90 > 98 Sardinia (Sassari) mastitis, keratitis ss IO1. 1101115.117,118 Sardinia (Sassari) mastitis NU 120.125.137,140,148 Sardinia (Nuoro) mastitis CA 150,155,163 Sardinia (Cagliari) mastitis OR 171.172,173 Sardinia (Oristano) mastitis BO 3.4 Emilia Romagna mastitis TE 103,290 Abruzzo mastitis PA I 19.420.485 Sicily mastitis RM I,2,3,4 Lazio mastitis SS 183,257,298.302.304 Sardinia (Sassari) mastitis NU 318,319,322.331 Sardinia (Nuoro) mastitis CA 338.340,347.351 Sardinia (Cagliari) mastitis OR 352.361.368 Sardinia (Oristano) mastitis SS 402.410,428,437.440 Sardima (Sassari) mastitis NU 441.448.452 Sardinia (Nuoro) mastitla, Thirty-five isolates from the Sardinian Zooprophylactic Institute. Sassarl province. Sixteen isolates from the Sardinian Zooprophylactic Institute. Nuoro province. Ten isolates from the Sardinian Zooprophylactic Institute. Cagliari province. Nine isolates from the Sardinian Zooprophylactic Institute. Oristano province. Obtained from Prof. Sanguinetti, Veterinary Infectious Diseases Institute. University of Bologna Obtained from Dr. Santini. Teramo s Zooprophylactic Institute. RFrom Palermo s Zooprophylactic Institute. Obtained from Dr. Lillini, Lazio and Toscana s Zooprophylactic Institute. There exist numerous procedures for comparing animal mycoplasmal strains: they are mostly based on total DNA digestion with frequently cutting restriction enzymes [WI], and on analysis of protein profiles [9911]. A large number of recent reports have described the use of pulsed field gel electrophoresis (PFGE) for obtaining molecular fingerprints which may be useful in epidemiological studies [12-151. The restriction endonuclease analysis (REA) patterns generated by PFGE consist of a fairly small number of well-separated bands which allow the evaluation of interspecies and intraspecies genetic variations without the ambiguities and difficulties typically encountered with conventional agarose gel electrophoresis. The aim of this study was to investigate the genomic variability of 81 isolates from different Italian outbreaks using PFGE, to compare its results with the protein and antigenic profiles obtained on so- 1991 1992 I993 1995 1995 dium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and on immunoblotting. 2. Materials and methods 2.1. Bucteriul struins und identzjicution Eighty-one strains of M. uguluctiue were isolated from milk samples of different flocks with contagious agalactia between and 1995. All strains and their sources are summarized in Table 1. Strains were cloned according to the Tully technique [ 161 and identified by indirect immunofluorescence using rabbit antiserum prepared against the type strain PG2 and by polymerase chain reaction (PCR) [ 171. All strains were grown in modified Hayflick medium supplemented with 10% (v/v) equine serum until the mid-exponential phase. Viable cell number was de-

S. Tola et al. I FEMS Microbiology Letters 143 (1996) 259-265 261 termined on the last positive dilution in liquid culture according to standard methods [18]. Mycoplasmas were pelleted at 20000Xg for 30 min, washed and resuspended in phosphate saline buffer (PBS, 0.1 M phosphate, 0.33 M NaCl, ph 7.4). To avoid cross-contamination, strains were prepared one at a time. 2.2. DNA preparations for PFGE analysis Chromosomal DNA was extracted from all 81 strains and from the type strain PG2. Washed mycoplasmas and 2% low melting point agarose (Bio- Rad, Richmond, CA) in PBS ph 7.4 were mixed in equal amounts. The suspension was aliquoted in a mould and allowed to solidify at 4 C. Each block, corresponding to 100 1.11 suspension, was incubated in a lysis solution (1% Sarkosyl, 0.5 M EDTA, 10 mm Tris, ph 9.5) supplemented by 2 mg/ml of proteinase K (Promega Co., Madison, WI) for 48 h at 50 C. Blocks were then washed twice with TE buffer (10 mm Tris-HCl (ph 8.0), 1 mm EDTA (ph 8.0) for 30 min at room temperature (r.t.), twice with 40 pg/ ml phenylmethylsulfonyl fluoride (Sigma Chemical Co. ) in TE for 30 min at 50 C. After a last wash with 0.5 M EDTA ph 8.0 for 30 min at r.t., blocks were kept at 4 C in 0.5 M EDTA ph 8.0. Fragments of each block (about 50 ~1) were digested with 10 U of SalI, BarnHI, XhoI, Apa, CfaI, NotI, SmaI, NruI (Boehringer, Mannheim), BssHII, KpnI, SpeI, XbaI (Promega) for 18 h and then run on a 1% agarose in TBE buffer (O.l M Tris, 0.1 M boric acid, 2 mm EDTA, ph 8.0) at 14 C in a contour-clamped homogeneous electric field (CHEF) mapper system (Bio-Rad). Gels were then stained with ethidium bromide (5 @ml), destained with distilled water and photographed under UV light. The sizes of the restriction fragments were determined by comparison with standard size markers consisting of yeast chromosomes, Saccharomyces cerevisiae strain YNN295 (Bio-Rad), Lambda Ladder (48.5-1000 kb, Bio-Rad) and Pulse Marker (225-2200 kb, Sigma). 2.3. Triton X-114 phase partitioning Extraction of integral membrane proteins was subjected to TX-114 phase partitioning as described by Bordier [19] and adapted with some modifications [20]. Briefly, pellets of A4. agalactiae (2 mg protein/ pellet) were resuspended in 1 ml of 1% (v/v) Triton X-l 14 (Surfact Amps, Pierce, Rockford, IL, USA) in PBS containing 1 mm phenylmethylsulfonyl fluoride (Sigma) and incubated at 4 C for 30 min. Insoluble material was pelleted by two cycles of centrifugation at 1OOOOXg for 10 min at 4 C. The supernatant was transferred to a new tube, incubated at 37 C for 10 min to induce condensation of TX-l 14 and centrifuged at 1OOOOXg for 5 min at r.t. The aqueous phase was readjusted with 10% Triton X-114 to 1% (v/v) and repartitioned as described above. After five cycles of phase fractionation, the detergent phase was incubated with methanol to a final concentration of 90% (v/v) at -80 C for 18 h. Proteins were precipitated at 12 000 X g for 10 min at 4 C and subsequently processed for SDS-PAGE and immunoblotting as described below. 2.4. Antiserum For the present study, we used a pool antiserum obtained from naturally infected sheep, i.e. all sheep showing typical symptoms of contagious agalactia (mammary gland enlargement and altered milk production). 2.5. SDS-PAGE and immunoblotting Forty micrograms of total proteins and phasefractionated proteins from each strain were boiled for 5 min in sample buffer (2% SDS, 5% 2-mercaptoethanol, 10% glycerol, 62.5 mm Tris (ph 6.8)) and then electrophoresed on 1 mm SDS-12% polyacrylamide gels in a Protean II xi vertical electrophoresis cell (Bio-Rad) according to the manufacturer s instructions. After electrophoresis, separated proteins were either stained with 0.25% Coomassie blue R-250 (Bio-Rad), 25% isopropanol and 10% (v/v) acetic acid, or transferred to nitrocellulose membranes (0.45 pm pore size, Sigma) in a Trans-Blot semi-dry apparatus (Bio-Rad) as described by the manufacturer. Blots were incubated for 2 h at r.t. with PBS containing 2% skim milk (Difco, Detroit, MI, USA) and then incubated for 1 h at 37 C with pooled M. agalactiae positive antiserum diluted 1:500 in PBS-2% skim milk. After five washes in PBS-2% skim milk,

262 B 225-145.5 97-48.5-194 - 145.5 97-48.5 23.1-9.4-194 -+ 145.54 97-3 48.54 23.1 + 9.4 j 5.6-3 242.5+ 194 --_) 145.5 -b 97+ 48.6--, 23.1-+ Fig. I. PFGE of Sal1 (A). Sntul (B), BssHIl (C). and X/w1 (D) restriction endonuclease fragments of M. uguhcriae DNAs performed at 14 C in I % agarose gel in 0.5 X TBE buffer. Lane I : Yeast chromosomes. Succcrronryces cwevisiue strain YNN295 (Bio-Rad) and/or Pulse Marker 0.1-200 kb (Sigma). Sizes of marker bands are indicated in kilobases: lane 2: M. ugalactiae strain from Sardinia: lane 3: M qnhctim strain from Sicily: lane 4: M. ~qydrctiae strain from Emilia Romagna: lane 5: M qmlrrc~ficr~ strain from Abruzzo: lane 6: M. qolucticrr strain from Lario. blots were incubated for at least 1 h at 37 C in per- land). After five more washes, blots were developed oxidase-conjugated anti-sheep (heavy and light chain with 4-chloro- 1 -naphthol/hpoz. The color reaction specific, Kirkegaard & Perry, Gaithersburg, Mary- was stopped by washing blots in water.

S. Tola et al. I FEMS Microbiology Letters 143 (1996) 259-265 263 Fig. 2. Coomassie R-250 SDS-PAGE analysis of PG2 and 13 M. agafactiae proteins segregating into the Triton X-l 14 phase. Lane I : kaleidoscope prestained standard, Bio-Rad; lane 2: PG2, reference strain; lanes 3-15: M. agalactiae strains (~~59, ss72, ss80, ~~410, nu40, nu318, ca3, ca340, or368, bo4, te103, pa485 and rm2). 3. Results 3.1. Restriction endonuclease analysis We tested 12 restriction endonucleases on M. agalactiae DNA: NotI, ApaI, ClaI, BarnHI, MI, XhoI, ClaI, SmaI, NruI, BssHII, SpeI and KpnI. We failed to observe DNA cutting by Not1 and ApaI of M agalactiae chromosomal DNA. ClaI, BarnHI, spe1 and Xbal produced numerous fragments smaller than 48.5 kb. Ultimately, we selected six restriction enzymes: SalI, SmaI, NruI, XhoI, BssHI and KpnI, to digest all 81 isolates. PFGE profiles obtained for each enzyme are absolutely indistinguishable and we could not find any differences among the PG2 reference strain, Sardinian, Sicilian, Emilian and Abruzzese strains. DNA cleavage patterns resulting from digestion with San, SmaI, BssHII and X/z01 of five different strains are shown in Fig. la-d. SalI produced 9 DNA fragments between 7 and 194 kb after 20.18 h migration at pulse times of 0.47-12.91 s. Under the same conditions, BssHII cut DNA into 10 fragments of 15-280 kb and XhoI into 5 fragments of 25400 kb. SmaI cleaved AL agalactiae DNA into 7 fragments of 4465 kb after 15.16 h of migration at pulse times of 0.22-30.82 s. 3.2. Integral membrane protein analysis by SDS-PAGE In this study we used SDS-PAGE and immuno- blotting techniques to compare the 81 strains and the reference strain PG2. The Coomassie R-250 SDS- PAGE surface protein profiles of 13 M. agalactiae isolates and reference strain PG2 are shown in Fig. 2. Apart from minor differences in intensity and positions of some bands, all profiles were similar. In particular, bands corresponding to 104, 80, 70, 50, 45, 41 and 26 kda were always present. 3.3. Antigenic analysis by immunoblotting In order to evaluate antigenic differences among various strains, Western immunoblot analysis of gels loaded with the same amount of proteins was performed with pooled antiserum from naturally infected sheep. The result is shown in Fig. 3. Pooled sheep antiserum reacted with several protein bands both of PG2 and of the isolates. Although differences in the molecular size of recognized protein bands among strains were evident, immunodominant proteins of 80 and 50 kda were detected in all field strains and in the reference strain. 4. Discussion The choice of methods to employ in epidemiological investigations depends on their reproducibility and on the microorganism under investigation. Various genetic techniques are currently used to compare strains. Digestion of total mycoplasmal DNA is also often used [5,7,8]; however under conventional electrophoretic conditions, the results are of- Fig. 3. Immunoblot analysis of PG2 reference strain and 13 M. agalactiae strains. The samples were subjected to SDS-PAGE, as described in Fig. 2, transferred to nitrocellulose and incubated with pooled sheep antiserum. All lane indications as in legend to Fig. 2.

264 S Tola c, cd / FEMS.Microhiolog>~ Letters 143 f 1996 1 259-265 ten difficult to analyze. PFGE makes it possible to separate large DNA fragments generated by rare cutters and provides a new approach to molecular typing of bacterial pathogens [21]. In the choice of restriction enzymes we selected those that have a recognition sequence rich in G+C in order to limit the number of fragments generated by mycoplasma digestion. Mycoplasma agalactiar is reported to have a mol% of G+C in the range of only 33.5-34.2 [22]. Among the 12 enzymes tested we found six that generate easily analyzable patterns. Using these six enzymes on all samples and on reference strain PG2 we noticed that the patterns obtained are 100% similar independently of the geographical origin of the samples tested. This surprising homogeneity for samples coming from regions as far apart as Sicily, Sardinia, Emilia Romagna, Abruzzo, Lazio and Tuscany suggests that only one strain of M. agalactiac is present in Italy. This fact is very important for the production of an effective vaccine. The homogeneity revealed by REA on PFGE is in accordance with Triton X-l 14 extracted membrane protein profiles on SDS-PAGE. Fig. 2 show the numerous common bands between samples and reference strain. These common proteins are all in the size range between 40 and 90 kda; in particular, 50 and 80 kda bands are quantitatively dominant. Immunoblotting analysis using pooled sheep antiserum shows more heterogeneity in comparison with SDS-PAGE analysis but the 50 and 80 kda proteins are the only immunogenic proteins in all strains analyzed. This observation is in contrast with data published by Solsona et al. [8] since their protein analysis by immunoblotting on French M. agabctiac strains reveals immunodominant antigens of sizes between 25 and 36 kda. Further studies should verify if these two bands are always present during all phases of infection so that they could be utilized to set up a serological diagnostic system specific for IV. agalactiar. Acknowledgments This work was supported by grants of Minister0 della Sanitl (Health Ministry), Ricerca Finalizzata (Aim specific research). We thank P. Melis and G. Delogu for support and helpful discussion and F. Masia, M.L. Lanza and T. Sechi for technical assistance. References [I] Cottew, G.S. (1979) Caprine-Ovine Mycoplasmas. The Mycoplasmas, Vol. II, pp. 103-132. Academic Press, New York. [2] DaMassa, A.J. (1983) Recovery of Mycoplasmcr ugulrrctiur from mastitic goat milk. J. Am. Vet. Med. Assoc. 183. 548% 54. [3] Lambert. M. (1987) Agalaxie contagieuse des brebis et des ch&vres. Rev. Sci. Tech. Off. Int. Epizoot. 6, 681-697. [4] Sanguinetti, V. and Chicco. D. (1987) Contagious agalactiae and M_vcoplasmu mycoides: the situation in Italy. In: Agriculture Contagious Agalactia and Other Mycoplasmal Diseases of Small Ruminants. pp. l-21. Off. Offic. Publ. Eur. Comm. Luxembourg. [S] lonas. G.. Norman. G.H.. Clarke, J.K. and Marshall. R.B. (1991) A study of the heterogeneity of isolates of Mycoplasmcr or~ipneumoniue from sheep in New Zealand. Vet. Microbial. 29. 339-347. [6] Lcy. D.H. and Avakian. A.L. (1992) An outbreak of M~ L.oplusmu.s.vnovrue infection in North Carolina turkeys: compar- ison of isolates by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and restriction endonuclease analysis. Avian Dia. 36, 672-678. [7] Poumarat, F., Solsona, M. and Boldini, M. () Genomic. protein and antigenic variability of Mycoplasma hovi.~. Vet. Microbial. 40. 305-321. [S] Solsona, M.. Lambert. M. and Poumarat, F. (1996) Genomic. protein homogeneity and antigenic variability of Mycoplasmo ugalucticrr. Vet. Microbial. 50. 45-58. [9] Thirkell, D., Spooner, R.K., Jones, G.E. and Russell, W.C. (I 990) Polypeptide and antigenic variability among strains of MJX-oplasmu ovipneumoniue demonstrated by SDS-PAGE and immunoblotting. Vet. Microbial. 21, 241-254. [IO] Avakian, A.P., Kleven, S.H. and Ley, D.H. (1991) Comparison of Mycoplusma gullisepticum strains and identification of immunogenic integral membrane proteins with Triton X-l 14 by immunoblotting. Vet. Microbial. 29, 319-328. [I I] Sachse, K:, Krajetzki, C., Pfutzner. H. and Hass, R. (1992) Comparison of Mycoplasmu hovi.y strains based on SDS- PAGE and immunoblot protein patterns. J. Vet. Med. B. 39, 24&252. [I21 Pyle, L.E., Lynn, N.C., Cocks, B.G., Bergemann, A.D., Whitley, J.C. and Finch. L.R. (1988) Pulsed-field electrophoresis indicates larger-than-expected sizes for mycoplasma genomes. Nucleic Acids Res. 16, 6015-6025. 1131 Smith, C.L., Kleco. S.R. and Cantor, R. (1988) Pulsed field gel electrophoresis and the technology of large DNA molecules. In: Genome Analysis: A Practical Approach (Davies. K.. Ed.). pp. 41 ~72. IRL Press. Oxford. 1141 Renaud. F.. Freney, J.. Etienne, J.. Bes. M., Brun. Y.. Barsotti. O., Andre. S. and Fleurette, J. (1988) Restriction endonuclease analysis of Staphylococcus epidermidis DNA may be

S. Tola et al. I FEMS Microbiology Letters 143 (1996) 259-265 265 a useful epidemiological marker. J. Clin. Microbial. 26, 1729- ization (Razin, S. and Tully, J.B., Eds.), pp. 185-196. Aca- 1134. demic Press, New York. [15] Allardet-Servent, A., Bouziges, N., Carles-Nurit, M.J., Bourg, G., Gouby. A. and Ramuz, M. (1989) Use of low-frequencycleavage restriction endonucleases for DNA analysis in epidemiological investigations of nosocomial bacterial infections. J. Clin. Microbial. 27, 2057-2061. [16] Tully, J.G. (1983) Cloning and filtration techniques for myco- plasmas. In: Methods in Mycoplasmology, Vol. I, Mycoplasma Characterization (Razin, S. and Tully, J., Eds.), pp. 173-177. Academic Press, New York. [17] Tola, S., Idini, G., Manunta, D., Galleri, G., Angioi, A., Rocchigiani, A. and Leori, G. (1996) Rapid and specific detection of Mycoplasma agaluctiae by polymerase chain reaction (in press). [18] Rodwell, A.W. and Whitcomb, R.F. (1983) Methods for di- rect and indirect measurement of mycoplasma growth. In: Methods in Mycoplasmology, Vol. I, Mycoplasma Character- [19] Bordier, C. (1981) Phase separation of integral membrane proteins in Triton X-114 solution. J. Biol. Chem. 256, 1604 1607. [20] Reithman, H.C., Boyer, M.J. and Wise, K.S. (1987) Triton X- 114 phase fractionation of an integral membrane surface pro- tein mediating monoclonal antibody killing in Mycoplasma hyorhinis. Infect. Immun. 55, 1094-l 100. [21] Maslow, J.N., Slutsky, A.M. and Arbeit, R.D. () Application of pulsed-field gel electrophoresis to molecular epidemiology. In: Diagnostic Molecular Microbiology: Principles and Applications (Persing, D.H., Smith, F.T., Tenover, F.C. and White, T.J., Eds.), pp. 563-572. American Society for Microbiology, Washington, DC. [22] Freundt, E.A. and Edward, D.G. (1979) Classification and taxonomy. In: The Mycoplasmas (Barile, M.F. and Razin, S., Eds.), pp. 141, Vol. I. Academic Press, New York.