Biotyping of Streptococcus thermophilus strains by DNA fingerprinting

Q 1NSTITUTPASTEUR/ELSEVIER Paris 1993

Biotypi

Res. Microbiol.

i993, i44, 38i-387

......... tt e p t o c o c c u s t h e r m o p h i l u s

strains

by DNA fingerprinting G. Salzano (2), G. Moschetti (~), F. Villani (2) and S. Coppola (~) (*; (t) istituto di Microbiologia agraria e Stazione di Microbiologia inclustriale, Universit~ degli Studi di .N~poli "'Federico H'" 1-80055 Portici, Naples (Italy), and (2) D.I.S.T.A.A.A:Y ~rniversith del Molise, 1-86100 Campobasso (Italy)

SUMMARY

DNA of 33 strains of Strep¢~coccus thermophilus (7 isolated from natural whey cultures utilized as starter, 25 from commercial yogurts and 1 reference strain) was analysed by restriction endonuclease digestion and both conventional and pulsed-field agarose gel electrophoresis. The restriction patterns after digestion of total DNA with Hindlll, EcoRI and BamHI enabled 10 different groups to be distinguished by conventional electrophoresis. Digesting genomic DNA with Smal or Notl and separating it by pulsed-field gel electrophoresis, the 33 strains could be classified into 5 groups. The same result was obtained after digestion with Haelll and conventional electrophor~sis. The 33 isotates could be divided into ! 3 groups when diversities among patterns Obtained after both the first and second method were consideredA combination of the t w o techniques turned out to be the moat reliable methodological approach for characterizing strains of S. thermophilus for scientific, industrial and legal purposes.

Key-words: DNA, Streptococcus thermophilus; Biotyping, DNA fingerprinting.

INTRODUCTION Genome analysis is considered to be an interesting approach for characterizing organisms. In microbioloby, DNA-DNA homology ~easurements have provided the basis for defirdng bacterial species. However, these are not sensitive enough to detect small variations withies, species revealed by serotyping, bacteriop!~age

Submitted November 16, 1992, accepted M~y i2, 1993. (*) Correspondingauthor.

typing, bacteriocin typing, resistotyping and by other methods of strain characterizatien. Plasmid and whole-cell protein electrophoretic profiling, 16S rRNA sequencing ar, d chromosomal restriction endonuclease digesl analysis ),~,_.esent '¢,nr new molecular methods for biotyping. Plasmid content of bacterial strains is of particular interest when referable to defined cLaracteristics. Whole-cell protein prefiling and ]6S

JO/...

G~ SA LZANO E T AL.

rRNA sequencing are widespread methods for identifying species, subspecies or other taxonomic entities such as biovar and serotype. DNArestriction-endonuclease digest analysis is very useful for obtaining information that can be used in genetic characterization (Tulloch et al,, 1991 ; Cancilla et al., 1992; Salzano et al., 1992) and enables strains of the same species to be discriminated.

We analysed the DNA fingerprinting of 33 strains of S. thermophilus after digestion with different restriction endonucleases and fragment separation by conventional and pulsed-field gel electrophoresis (PFGE) with the purpose of defining reliable procedures for typing and checking industrial strains of this species.

Owen (1989) discussed the applicability of this type of analysis to microbial pathogens in diagnostics and epidemiology. Generally speaking, a similar approach might be very useful in microbial ecology. With reference to industrial microorganisms, it has already been emphasized that sensitive, reproducible methods for strain identification could be of great use for monitoring population changes during fermentations supported by mixed or multiple strain starters (Ramos and Harlander, 1990), screening natural isolates, characterizing commercially significant strains (Salama et al., 1991) and defending patent rights (Colmin et al., 1991). In Italy, a law legal proposal nearing promulgation states that water-buffalo Mozzarella cheese from Campania, as a "controlled-production-area cheese", can be manufactured utilizing as starter both natural whey cultures and "autochthonous" strains of lactic acid bacteria, thereby necessitating reliable methods for strain characterization and "certification" (Anonimous, 1992).

MATERIALS AND METHODS

Streptococcus thermophilus, one of the most widely used lactic acid bacteria in the dairy industry, has already been studied from this point of view. Ramos and Harlander (1990) analysed DNA fingerprinting of strain ATCC 19987 isolated from a cheese starter and of strain ATCC 19258 isolated from pasteurized milk and found them to be identical. Colmin et aL (1991) examined 25 strains of this species. After digestion with Haelll and N d e l l , they obtained 17 different patterns and judged the method inconvenient for use, preferring to develop a species-specific D N A probe by which diversities were reduced to 10.

CHEF = contour-clampedhomogeneouselectric field. PFGE =- pulsed-fieldgel electrophoresis.

Bacterial strains and growth media

A total of 33 strain¢ of S. thermophilus were studied, 7 isolated during previous studies on the microftora of natural whey cultures utilized as starters for the traditional manufacture cf water-buffalo Mozzarella cheese in the provinces of Naples mad Caserta (Coppola et al., 1988), 25 isolated from commercial samples of yogurt and strain 822 belonging to the National Collection of Food Bacteria, Reading, UK. The origins of the strains are reported in table 1. Their identification was ascertained according to Diebel and Seeley (1974) by techniques described by Harrigan and McCance (1976). MI7 broth (Oxoid) was used as standard growth medium. Isolation of total DNA, restriction digestion and conventional electrophoresis

Total DNA was prepared by the method of Marmur (1961) from cells harvested by centrifugation (8,000 rpm) of overnight cultures at 42°C in Mi7 broth of each of the bacterial strains. The purity of the DNA samples was confirmed spectroscopically by determining the ratio of the absorbance at 260 to that at 280 nm. DNA concentrations were determined by measuring absorbance at 260 nm. DNA (5 ~g) was digested with Hindlll, BamHl or EcoRI according to the supplier's (Promega) instructions. Digests were separated by electrophoresis through 0.6 e/0 (w/v) agarose gel in TBE Luffer at 2.5 V/cm for 12h. Genomic DNA preparation, restriction digestion and

PFGE Chromosomal DNA of high molecular weight were prepared by the McClelland method (1987). Samples were dialysed in 20 ml TE (Tris 10 'alb¢1,

PMSI7 = phenylmelhylsul.~honylfluoride. [E ~= "lris+EDTA.

B ! O T Y P ! N G Off S T R E P T O C O C C U S T H E R M O P H i L U S

ABC

DE F GH

B Y DNA H N G E R P R i N T I N G

383

| LM

Fig. I. Type pattgr~s resulting from Hindlll digestion and e!ectrophoresis in 0.6 % agarose gel of total DNA of S, thermophilu:¢ strains. Lane A = ~. DIqA H/t/dill restriction fragments used as ~. size marker ; lane B = group 1 ; lane C = group II; lane D = group Ill ; lane E = group IV; lane F ~ group V; lane G = group V1 ; lane H = group VII ; lafl~ I ~ group VIii; idqe L = grot~p IX ; lane M = group X.

T a b l e I. S. ther~tlo~vtzilus strains g r o u p e d on the basis o f restriction pv,ttern s i m i l a r i t y

306W 317W 321Y 322Y 324Y" 327Y 340Y 342Y 344Y 337Y

It

Ill

IV

V

V',

VII

VII1

IX

X

314W 318W 320Y 323Y 345Y 348Y 352Y 353Y 354Y

330Y 338Y 33eJY 353Y

302W 334Y 343Y

326Y 329Y

347Y

332Y

308W

NCFB822

3 t 0W

The ten different restriction Patterns we~-eobtained by digestion of ~oial DNA, The sanle len groups ~,ere delec'~ed u=ing d~ree different restriction enzymes. W = nalural whey culture utililed iq tradilional K~nufac~ure ~i ~ ater-bt;ffalo Mozzarella cheese ; Y = commercial sample of yoghurt; NCFB= National Collection of kood Bac~eria~ Reading (UK).

G. SALZANO E T AL.

384

EDTA 1 raM; pH 8.00) containing 0.1 mM phenylmethylsulphonyl fluoride (PMSF) for 1 h at room temperature. After dialysis repeated twice without PMSF, the inserts were maintained in TE at 4°C. For digestion, each sample was incubated in 200 ml of appropriate buffer supplemented with 15 units of restriction enzyme (Sinai, Notl) and incubated for 8 h at the appropriate temperature.

A

B

C

D

E

F

Electrophoresis of the restriction digests was performed on the CHEF (contour-clamped homogeneous electric field) system "Chef-Mapper" (BioRad) for 14 h and 6 V/cm with a 25-s pulse time at 14°C.

RESULTS Digestion of total DNA of the 33 strains of

S. thermophilus with HindIII and separation by conventional agarose electrophoresis enabled the distinction of 10 different patterns (fig. 1). Strains could be grouped as reported in table I. Strain NCFB822 as well as other four strains showed a pattern that could not be compared with those of the other strains analysed. The seven strains isolated from natural whey cultures showed patterns referable to 5 different types. Profiles obtained from these strains were also encountered in isolates from commercial yogurts. Four patterns (III, V, VI and VII) could be considered as typically belonging to yogurt strains. The same 10 groups were confirmed after digestion of total DNA of the 33 strains with EcoRI and BamHI (data not shown). Results obtained after digestion of genomic DNA with Smal and separation by pulsed field gel electrophoresis are reported in figure 2 and in table !l. In this instance, the 33 strains could be referred to five groups, wholly confirmed after digestion with Notl and separation by the same electrophoresis method. ~Strain 310W (isolated from a natural whey culture), strain 348Y (isolated from a sample of yogurt) and strain NCFB822 showed singular restriction patterns; all other straJ,~ were referable to two groups. This same arrangement of tl~e 33 strains was obtained after digestion of total DNA with Haelll and conventional electrophoresis. The five restriction patterns obtained by this method are shown in figure 3.

Fig. 2. Type patterns resulting from Sinai digestion and PFGE analysisof genomicDNAof S. thermophilusstrains. Lane A = ~.concatamers (monomer43.3 kb); lane B = groupl; laneC = group IV; laneD = grouplll; lane E = group !1; lane F = group V.

The analysed isolates could be divided into 13 groups when diversities among patterns obtained both after digestion with HindllI (or EcoRI or BamHI) and after digestion with Sinai (or NotI or HaellI) were considered. Table III reports, for each group, the strains presenting similar patterns using the first and second method.

BIOTYPING OF STREPTOCOCCUS THERMOPHILUS B Y D/~JA FINGERPRiNTllVG Table I!. S. thermophilus strains grouped on the

basis of restriction pattern similarity, after pulsed field gel electrophore~is. I

II

Ill

302W 308W 317W 321Y 322Y 323Y 324Y 327Y 329Y 332Y 337Y 340Y 342Y 344Y 326Y

306W 314W 3t8W 320Y 330Y 334Y 338Y 343Y 345Y 348Y 352Y 353Y 354Y 355Y 339Y

1V

310W NCFB822

385

ABCDE

V

347Y

?

The five different restriction patterns were obtained by digestion of chromosomal DNA with S i n a i and N o t l . The same five groups were detected u~ing H a e l l l and conventional electrophoresis.

DISCUSSION Up-to-date DNA restriction endonuclease analyses have provided results of differing significance and utility. Our results highlight a particular restriction polymorphism among 33 strains of S. thermophilus. In fact, strains isolated from the same habitat showed different restriction patterns; moreover, the same pattern was encountered in strains of different origins. This represents an intermediate situation between that reported, tbr instance, in species belonging to other genera such as Brucetla (AllardetServent et al., 1988) or Staphylococcus (Etienne et aL, 1990), where no difference was evidenced for strains of the same species, and in results concerning other species, such as Pseudomonas aeruginosa (Grothues and Tummler, 1987), where strictly strain-dependent restriction patterns were found.

considered. We distinguished 10 biotypes after digestion of DNA of 33 strains with HindlII, EcoRl and BamHI, but 5 of them were represented by a single strain; the remaining 28 (about 80 % of the isolates analysed) could be gathered into just 5 groups.

Colmin et aL (1991) recently showed the validity of a species-specific DNA probe to distinguish 10 biotypes of the bacterial species here

The 33 strains studied were referable to 5 biotypes after DNA digestion with Sinai or NotI and PFGE, as well as after digestion with

.....

:

:

2

:i:

.

,

Fig. 3. Type p a t t e r n s resulting f r o m H a e l l l digestion and e]ectrophoresis in 0.6 % a g a r o s e gel o f t o t a l D N A o f S. thermophilus strains. Lane A = g r o u p l ; lane B = g r o u p IV; t a n e C g r o u p 11I; lane D = g r o u p I I ; lane E = g r o u p V

=

386

G. SALZANO ET AL.

Table III. S. thermophilus strains grouped cumulating the diversities among restriction patterns obtained after

digestion with Hindill (or EcoRl or BamHl) and with Smal (or Notl or Haelll). I

il

II1

lV

V

VI

VII

VIII

IX

X

Xi

317W 321Y 322Y 324Y 327Y 337Y 340Y 342Y 344Y

314W 318W 320Y 345Y 348Y 352Y 353Y 354Y

330Y 338Y 339Y 355Y

334Y 343Y

326Y 329Y

302w

323Y

332Y

347Y

308W

3i0W

Haelll and conventional electrophoresis. However, analysis o f the 28 strains mentioned a b o v e shows that they are referable to only 2 groups. It is interesting that the c o m p o s i t i o n o f the five groups does not vary after D N A digestion with restriction endonucleases recognizing sequences o f eight (NotI), six (Smal) and four (Haelll) base pairs (all c o m p o s e d by G C pairs). Unfortunately, consistent criteria for establishing the exact limits o f an authentic " b i o t y p e " within a bacterial species are not available. It is therefore not possible to point out the most suitable and correct a p p r o a c h for achieving this objective. Nevertheless, when characterizing strains by simple and reproducible methods, restriction analysis carried out with H i n d l l l and conventional electrophoresis and digestion with Smal and pulsed electrophoresis ( P F G E ) can be considered fairly satisfactory. This combination o f techniques proved particularly reliable, mainly due to an appropriate number o f restriction fragments resulting from the Smal d i g e s t i o n / P F G E system, which permitted straightforward interpretation.

Acknowledgements

This research was supported by the National Research Council of Italy, Special Project RAISA, Sub-project N. 4, Paper N. 923.

XII

XIII

306W NCFB822

Caract6risation de Streptococcus thermophilus par profils de restriction de I'ADN

L'ADN de 33 souches de Streptococcus thermophilus (7 isoldes h partir de c,altures naturelles utilisdes comme levain, 25 isoldes de yogourts du commerce et I souche type de collection) a dt6 analysd apr~s digestion par des endonucldases de restriction et dectrophor~se en agarose, conventionnelie ou en champ pulsd. Les profils de restriction obtenus apr6s digestion de I'ADN total avec ltindI!l, EcoRI et BamHI, suivie d'une dlectrophor~se conventionnelle, ont permis de distinguer 10 groupes de souches. La digestion de I'ADN gdnomique par Smal ou Notl suivie d'une dlectrophor6se en champ pulsd a mis en dvidence 5 groupes de souches, les mfimes rdsultats dtant obtenus apr~s digestion par Haelll et dlectrophor~se conventionnelle. Si on rdunit les donndes fournies par la premiere et la deuxi~me mdthode, les 33 souches peuvent ~tre rdparties en 13 groupes de biotypes. L'analyse de I'ensemble de ces rdsultats devrait faciliter la caractdrisation des souches de S. thermophilus. Mots-cl~s : ADN, Streptococcus thermophilus ; Caractdrisation, Profils de restriction.

References

Allardet-Servent, A., Bourg, G., Ramez, M., Papes~ M., Bellis, M. &Raizes, G. (1988), DNA polymorphism in strains of the genus Brucella. J. BacterioL, 170, 4603-4607.

BIOTYPING OF STREPTOCOCCUS THERMOPHILUS BY DNA FINGERPRINTING Anonimous (1992), Ministero dell'Agricoltura e delle Foreste. Parere del comitato nazionale per la tutela delle denominazioni di origine e tipiche dei formaggi. Gazzetta Ufficiale della Repubblica ltaliana, Serie Generale, 10-4-1992, 85, 36. Cancilla, M.R., Powell, I.B., Hillier, A.J. & Davidson, B.E. (1992), Rapid genomic fingerprinting of Lactococcus lactis strains by arbitrarily primed polymerase chain reaction with 3zp and fluorescent labels. AppL Environ. )dicrobiol., 58, 1772-1775. Colmin, C., Pebay, M., Simonet, J.M. & Decaris, B. (1991), A species-specific DNA probe obtained from Streptococcus salivarius subsp, thermophilus detect strain restriction polymorphism. FEMS MicrobioL Letters, 81, 123-128. Coppola, S., Parente, E., Dumontet, S. & La Peccerella, A. (1988), The microflora of natural whey cultures utilized as starters in the manufacture of Mozzarella cheese from water-buffalo milk. Lait, 68, 295-310. Diebel, R.H. & Seeley, H.W. (1974), Streptoeoccaceae fam. nov., in "Bergey's manual of determinative bacteriology" (Buchanan R.E. and Gibson N.E., Gibbson R.E., Baltimore N.E.). Williams and Wilkins, London. Etienne, J., Poitevin-Later, F., Renaud, F. & Fleurette, J. (1990), Plasmid profiles and genomic DNA restriction endonuclease pattern of 30 independent Staphylococcus lugdunensis strains. FEMS MicrobioL Letters, 67, 93-98. Grothues, D. & Tummler, B. (1987), Genome analysis of

387

Pseudomonas aeruginosa by field inversion electrophoresis. FEMS MicrobioL Letters, 48, 419-422. Harrigan, W.F. & McCance, M.E. (1976), Laboratory methods in food and dairy microbiology. Academic Press, London, New YorkMarmur, J. (1961), A procedure for lhe isolation of deoxyribonucleic acid from microorganisms. J. MoL BioL, 3, 208-218. McClelland, M., Jones, R., Patel, Y. & Nelson, M. (1987), Restriction endonuclea~es for pulsed field mapping of bacterial genomes. Nucl. Acids Res., 15, 5985-6005. Owen, R.J. (1989), Chromosomal DNA fingerprinting - a new method of species and strain identification applicable to microbial pathogens. J. Med. MicrobioL, 30, 89-99. Ramos, M.S. & Harlander, S.K. (1990), DNA fingerprinting of lactococci and streptococci used in dairy fermentations. Appl. MicrobioL BiotechnoL, 34, 368-374. Salama, M., Sandine, W. & Giovannoni, S. (1991), Development and application of oligonucleotide probes for identification of Lactococcus lactis subsp. cremoris. Appl. Enviro~m. MicrobioL, 57, 1313-1318. Salzano, G., Vilani, F., Pepe, O., Sorrentino, E., Moschetti, G. & Coppola, S. (1992), Conjugal transfer of plasmid-born bacteriocin production in Enterococcus faeealis 226 NWC. FEMS MicrobioL Letters, 99, 1-6. Tultoch, D.L., Finch, L.R., Hillier, A.J. & Davi6son, B.E. (1991), Physical map of the chromosome of Lactococcus lactis subsp, lactis DLt t and localization of six putative rRNA operons. J. Bacteriol., 173, 2768-2765.