Vol. 45, No. 5APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 1983, p. 1670-16760099-2240/83/051670-07$02.00/0Copyright ©) 1983, American Society for Microbiology

Isolation and Partial Characterization of Two Aeromonashydrophila BacteriophagesMITCHELL S. CHOW AND M. A. ROUF*

Department of Biology and Microbiology, University of Wisconsin-Oshkosh, Oshkosh, Wisconsin 54901

Received 30 August 1982/Accepted 31 January 1983

Two Aeromonas hydrophila bacteriophages, Aehl and Aeh2, were isolatedfrom sewage. Both phages showed binal symmetry. The dimensions of A.hydrophila phages Aehl and Aeh2 differed from those of the other Aeromonasphages. Also, phage Aeh2 was the largest Aeromonas phage studied to date.Phage Aehl formed small, clear plaques, and phage Aeh2 formed turbid plaqueswith clear centers. Both phages were sensitive to chloroform treatment, beingtotally inactivated after treatment for 1 h at 60°C at pH 3 and 11. However, theinfectivity of Aehl phage stocks increased by approximately fivefold after theywere treated at pH 10 for 1 h at 22°C. Phages Aehl and Aeh2 were serologicallyunrelated and had latent periods of 39 and 52 min, respectively. The average burstsizes of phages Aehl and Aeh2 were 17 and 92 PFU per cell, respectively. PhageAehl infected 13 of 22 A. hydrophila strains tested, whereas phage Aeh2 infectedonly its original host. Phage Aehl infected some A. hydrophila strains only at orbelow 37°C. Neither phage infected the two A. (Plesiomonas) shigelloides strainsused in this study.

The bacterium Aeromonas hydrophila hasbeen known to cause diseases in frogs, alliga-tors, fishes, and even humans (10, 18, 19, 21). A.hydrophila has also been found to have a widerange of distribution in aquatic habitats (12, 16)and is considered by some investigators as apotential index organism of pollution. Becausethese organisms are also found in milk (6), meat(7), and sausage (5) and are able to grow at 0°C(15), they are potential spoilers of refrigeratedfoods.

Although A. hydrophila has been studied bymany investigators, very little is known about itsphages. Recently, Seely and Primrose (17) re-ported the isolation from fish ponds of three A.hydrophila phages used to correlate the effect oftemperature on the ecology of aquatic bacterio-phages. However, to our knowledge, there is noreported study which describes purification andcharacterization of A. hydrophila phage. There-fore, this investigation was undertaken to iso-late, purify, and characterize A. hydrophilaphages from sewage and to determine whetherthe isolated phages can be used to identify anddifferentiate between A. hydrophila strains andrelated organisms.

MATERIALS AND METHODSBacterial strains. A. hydrophila strains A3 and

ATCC 7966 were used as the original hosts for theisolation of bacteriophages. A. hydrophila A3 wasisolated from a frog with red leg disease (15).

Media and growth conditions. Cultures were incubat-ed at 30°C in Hershey (H) broth, enriched Hershey topagar, or enriched Hershey bottom agar (20).

Plaque assay: isolation and plaque morphology. Thestandard agar overlay method described by Adams (2)was used for isolation and enumeration of PFU.Plaque morphologies were observed at 12 to 16 h ofincubation. Bacterial cultures in nutrient broth wereinoculated with water samples of the final effluentfrom the city sewage treatment plant in Oshkosh, Wis.The mixtures were incubated for 36 h at room tem-perature and then centrifuged and filtered through0.45-,um (pore size) membrane filters (Gelman Sci-ences, Inc., Ann Arbor, Mich.). Pure bacteriophagestrains were obtained by six serial single-plaque isola-tions, and the two viruses so isolated were designatedAehl and Aeh2.Phage stocks. Cell lysates were prepared by the

broth and plate methods of Eisenstark (8). The lysateswere centrifuged at 5,000 x g in a Sorvall RC2-Bcentrifuge and filtered through 0.45-p.m (pore size)membrane filters (Gelman Sciences, Inc.). The filteredphage stocks were stored at 4°C.

Electron microscopy. Phage lysates were preparedby the plate method (8) and washed twice by low-speed (6,800 x g, 15 min) and high-speed (41,300 x g,2 h) centrifugation. The phage pellets were suspendedin 0.26 M ammonium acetate (pH 7.0) and negativelystained with 1% potassium phosphotungstate (pH 7.0)or 1% uranyl acetate (pH 4.0) by the method ofHaschemeyer and Myers (11). Stained phage sampleson carbon-coated, 400-mesh copper grids were ob-served with an EMU-3G transmission electron micro-scope (RCA Electro-Optics & Devices, Lancaster,Pa.) operated at 50 kV.

1670

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

ISOLATION OF A. HYDROPHILA BACTERIOPHAGES 1671

The dimensions of the phages were determined bycomparing the negatively stained phage samples online diffraction grids (catalog no. 0734; Polaron Equip-ment Ltd., Watford, England) with the line spacings(426.9 nm) of the grids. Tail lengths of the phagesincluded the necks and baseplates.

Sensitivity to chloroform. The experimental proce-dure used for determining chloroform sensitivity wasmodified from that of Feldman and Wang (9). A 1-mlphage suspension with an infectivity of approximately2 x 103 PFU was mixed with 0.05 ml of chloroform in atest tube (13 by 100 mm). To another tube of phagesuspension, 0.05 ml of physiological saline was addedinstead of chloroform as a control. The mixtures wereshaken by hand for 10 min at room temperature. Thetubes were then centrifuged in a bench-top centrifugefor 5 min, and the uppermost layer of the fluid in thetube was assayed for enumeration of PFU per millili-ter.

Stability to temperature, pH, and storage time. Sta-bility to temperature was tested in 5-ml volumes ofphage suspension by placing them in a water bath atappropriate temperatures for 1 h and then plaqueassaying. The pH stability was tested in 10-ml volumescontaining 9.9 ml ofH broth at the appropriate pH and0.1 ml of phage suspension. Stability in time of thephage stocks in 10-ml samples was determined at 4 and300C.

Adsorption of phage to host cell. Appropriate cellsuspensions were repeatedly washed in H broth,counted microscopically in a Petroff-Hausser countingchamber, and resuspended at 108 cells per ml in Hbroth containing 107 PFU/ml, 1 mM CaCl2, and 10-3 MMgCl2 or NaCl.At 5-min intervals, 0.1-ml samples of the phage-cell

mixtures were withdrawn from the tubes and dilutedinto 9.9 ml of ice-cold H broth. The diluted mixtureswere immediately centrifuged in a bench-top centri-fuge for 10 min at room temperature and assayed forPFU. The rates of absorption of the phages to the hostcells were calculated by the method of Adams (2).

Preparation of antisera. Phage lysates prepared bythe plate method (8) were centrifuged at 5,000 x g for15 min. The supernatants were heated in a water bathat 4°C for 1 h and then partially purified by two cycleseach of low-speed (6,000 x g, 15 min) and high-speed(41,300 x g, 2 h) centrifugation. The phage pelletswere resuspended overnight in phosphate-buffered sa-line at 4°C. The infectivities of phage Aehl and Aeh2vaccines were 3.0 x 1010 and 4.0 x 1010 PFU/ml,respectively. Rabbits were immunized by intravenousinjections of the viral vaccines into the marginal earveins by the immunization schedule of Burrell (4).Sera were stored at -20°C until use.

Serological properties. The antisera were diluted to1:2,000, and neutralizations of 10 original PFU/mlwere conducted at 37°C. The serum values (K values)for the rates of serum neutralization of the phageinfectivity and the serological relationships of thephages were determined by the method of Adams (2).Growth cycles. One-step growth curves were investi-

gated as described by Adams (2).Host range. The host ranges of the phages were

determined by spot tests (8) at incubation tempera-tures of 20, 30, and 37°C. The inocula of phages Aehland Aeh2 were 5.0 x 108 and 3.0 x 109 PFU/ml,respectively. The plates were examined at 18 and 24 hof incubation.

RESULTSPhage isolation. The bacteriophages isolated

on A. hydrophila strains A3 and ATCC 7966were designated Aehl and Aeh2, respectively.Designation of the phages followed the sugges-tion of Ackermann et al. (1).

Plaque morphology. The plaque morphologiesof phages Aehl and Aeh2 are shown in Fig. 1.Phage Aehl produced small, clear plaques of 0.5to 1.0 mm in diameter which remained approxi-mately the same size at 12 to 36 h of incubation.

I

.r._

..Ali. -~

FIG. 1. Plaque (dark areas) morphologies of phages Aehl (A) and Aeh2 (B). The incubation times for phagesAehl and Aeh2 were approximately 16 and 24 h, respectively. Bar, 5 mm.

VOL. 45, 1983

a

t

416 41. * 41*

'O. AO'.

4-If

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

1672 CHOW AND ROUF

FIG. 2. Electron micrographs of A. hydrophila phages Aehl (A) and Aeh2 (B). Phage Aehl was negativelystained by 1% potassium phosphotungstate (pH 7.0); phage Aeh2 was negatively stained by 1% uranyl acetate(pH 4.0). Bar, 100 nm.

Phage Aeh2 produced turbid plaques with clearcenters. The turbid plaques of Aeh2 were 0.8 to1.8 mm in diameter at 12 h of incubation andincreased to 1.0 to 4.0 mm in diameter at 36 h ofincubation. The clear centers remained approxi-mately the same size at 12 to 36 h of incubation.

Electron microscopy. The virion of phageAehl was composed of: (i) a head 134.4 nm longand 89.0 nm wide, (ii) a tail 122.8 nm long and21.7 nm wide, and (iii) a baseplate 30.2 nm wideto which several spikes were attached.The virion of phage Aeh2 consisted of: (i) an

isometric head 133.9 nm in diameter betweenopposite apexes, (ii) a tail 212.4 nm long and 28.1nm wide, and (iii) terminal appendages attachedto the tail (Fig. 2). The tails of both phagespossessed contractile sheaths which were ableto contract along the long axes of the tails (notshown).

Sensitivity to chloroform. After chloroformtreatment, the infectivities of the phage Aehland phage Aeh2 suspensions were reduced to72.0 and 0.74% of their original infectivities,respectively.

Stability to temperature, pH, and time. Thetemperature stabilities were tested at 20 to 60°C,at 10-degree intervals. Phage Aehl was stablefor 1 h at 50°C. Phage Aeh2 was stable for 1 h at40°C, and only about 75% of the phage survivedfor 1 h at 50°C. Both phages were inactivated toless than 1% survival after 1 h at 60°C. PhageAehl was stable at pH 5 to 10, and phage Aeh2was stable at pH 5 to 9. The activities of both

phages were totally lost at pH 3 or 11 for 1 h at22°C (Table 1). However, the infectivity of thephage Aehl suspension increased by approxi-mately fivefold after it was suspended at pH 10for 1 h at 22°C. When stored at 4°C, both phagesshowed initial losses of infectivity over the first10 days. Thereafter, the phage stocks werestable (Fig. 3). On day 60 of storage at 4°C,phages Aehl and Aeh2 were decreased to 60 and65% of their original infectivities, respectively.At 30°C, the viral infectivities of phages Aehland Aeh2 were reduced to 1% of the original ondays 37 and 78, respectively.

TABLE 1. Stabilities of phages Aehl and Aeh2 topHa

% Survival of phage:

Aehl Aeh2

3 0.00 0.004 0.74 0.115 100 766 99 1017 100 1008 99 1009 106 8710 522 5711 0.00 0.00

a Phage particles were suspended at the indicatedpH for 1 h at 22°C and then pipetted for the phageassay.

APPL. ENVIRON. MICROBIOL.

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

ISOLATION OF A. HYDROPHILA BACTERIOPHAGES 1673

Adsorption of phages to host cells. The adsorp-tion rates of phage Aehl to A. hydrophila A3 andof phage Aeh2 to A. hydrophila ATCC 7966 areshown in Fig. 4. The adsorption constants forphage Aehl in H broth alone, H broth plus 10-3M MgCl2, and H broth plus 10-3 M CaCl2 were5.3 x 10-"°, 7.1 x 10-10, and 8.2 x 10-10,respectively. Similarly, the adsorption constantsfor Aeh2 in these media were 5.7 x 10-10, 4.3 x10-10, and 7.7 x 10-1o, respectively. Bothphages adsorbed more rapidly to their host cellswhen 10-3 M CaC12 was added to the H broth.

Serological properties. The rate constants (Kvalues) of anti-Aehl serum neutralization were353 and 0 for phages Aehl and Aeh2, respective-ly, and those of anti-Aeh2 serum neutralizationwere 0 and 339 for phages Aehl and Aeh2,respectively (Table 2).Growth cycles. The one-step growth curves of

1 010

109

E

108

1 '7

A. hydrophila phages Aehl and Aeh2 understandard growth conditions indicated that thelatent periods of phages Aehl and Aeh2 were 39and 52 min, respectively. The average burstsizes of phages Aehl and Aeh2 were 17 and 92PFU per cell, respectively. The rise periodswere 34 and 50 min for phages Aehl and Aeh2,respectively.Host range. A. hydrophila phages Aehl and

Aeh2 infected some A. hydrophila strains, butnone of the other genera or species tested (Table3). Phage Aehl infected 13 of 22 A. hydrophilastrains. Phage Aehl formed plaques on plates ofA. hydrophila Cl at 20 and 30°C, but not at 37°C.Also, phage Aehl formed plaques on plates ofA.hydrophila S1 at 30°C but not at 20 or 37°C.Phage Aeh2 infected its original host (A. hydro-phila ATCC 7966), but none of the other bacteriain this study.

20 40 60 80Time (days)

FIG. 3. Stabilities of phages Aehl (closed symbols) and Aeh2 (open symbols) during storage at 4°C (O, *) or300C (0, 0).

VOL. 45, 1983

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

167 CHO AN ROU APP. ENIO.MIRBo.

100 A

80

60(D

40

(0~~~~~~~~~~~~~~~~~~~~~~~~~~~~(

20

5 10 15 20 5 10 1 5 20Time (min)

FIG. 4. Adsorption of phage Aehl to A. hydrophila A3 (A) and of phage Aeh2 to A. hydrophila ATCC 7966(B). The phages were adsorbed to their host cells in H broth (A, A), in H-broth plus 0.001 M MgCl2 (U, O), and inH broth plus 0.001 M CaCl2 (0, 0). At the indicated time intervals, unadsorbed phages were measured by thephage assay. The adsorptions were conducted at 30°C.

DISCUSSION

TABLE 2. Neutralization of phages Aehl and Aeh2by antiseraTime (min)

required for 80% K valueAntiseruma neutralization

Aehl Aeh2 Aehl Aeh2

AehlUnadsorbed 9.1 b 353 0Adsorbed 10.2 - 315 0

Control(unadsorbed)c NDd 0

Aeh2Unadsorbed 9.5 0 339Adsorbed 10.9 0 295

Control(unadsorbed)c 0a Sera were adsorbed with the host cells. Anti-Aehl

serum adsorbed to A. hydrophila A3 and anti-Aeh2serum adsorbed to A. hydrophila ATCC 7966.

b , No decrease of phage titer was observedduring the experimental period.

c Nonimmune serum collected from rabbits beforeimmunization.

d ND, Not done.

Two A. hydrophila bacteriophages, Aehl andAeh2, were isolated from sewage. Because eachvirion of the two phages is composed of a headand a contractile tail, phages Aehl and Aeh2belong to the morphological group A of bacterio-phages described by Bradley (3). The dimen-sions of phages Aehl and Aeh2 are differentfrom those of the other Aeromonas phages de-scribed by Paterson et al. (13) and Popoff (14).Phage Aeh2 is the largest Aeromonas phagestudied to date.

Seeley and Primrose (17) reported that onlylow-temperature phages (phages able to formplaques at or below 30°C) were found for A.hydrophila in the water samples they studied. Inour study, phage Aehl formed plaques on platesof many A. hydrophila strains at 20, 30, and37°C, and yet the infection of the same phagewas found to be temperature dependent forstrains Cl and Si. Phage Aehl infected A.hydrophila Cl at 20 and 30°C and strain Si at30°C. Phage Aehl was unable to infect A. hydro-phila Cl at 37°C or strain Si at 20 or 37°C (Table3). These results indicate that temperature mayplay a role in phage infections of certain A.hydrophila strains, but not in others. Generaliza-tion on the role of temperature in classifyingphages may be valid for only a specific host.

Bacteriophages are usually stable over a pHrange of 5 to 8 (2). Phage Aehl was stable at a

APPL. ENVIRON. MICROBIOL.1674 CHOW AND ROUF

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

ISOLATION OF A. HYDROPHILA BACTERIOPHAGES 1675

TABLE 3. Host range of phages Aehl and Aeh2 by spot test

Plaque formation of following phage at indicated temp

Bacterium and strain no. or (SC)cAorigin Source Aehl Aeh2

20 30 37 20 30 37

A. hydrophilaA3C1

C2

C3

S1

ATCC 7966CDC 3067-68CDC 3067-68 SCICDC 3325-68CDC 3326-68 SCICDC 3890-68CDC 3910-68CDC 5244-68BBWc

Bear River, IdahocBLMB23C

CDSc

GLlcHuman bloodcJackson Hole, Wyo.cLN1c

Madison River, Mont.c

Frog (Wisconsin)Raw chicken

(Wisconsin)Raw chicken

(Wisconsin)Raw chicken

(Wisconsin)Raw sausage

(Wisconsin)NAbNANANANANANANAWater (South

Carolina)

Lake (NorthCarolina)

Sediments (SouthCarolina)

AlligatorNorth Carolina

Lake (NorthCarolina)

+ + + _ _ _

+ +±

_- + + +

+ + +

+ + + _ _ _

+ + +_ _ _

+ + + _ _ _

+ + + _ _ _

+ + + _ _ _

+ + +

+ + + _ _ _

+ + + _ _ _

A. (Plesiomonas)shigelloidesCDC 2418-69CDC 4337-69

Unrelated organismsda Infectivities of phage Aehl and Aeh2 suspensions were 5 xb NA, Not available.

108 and 3 x 109 PFU/ml, respectively.

c Kindly provided by T. C. Hazen, University of Puerto Rico, Rio Piedras.d Enterobacter aerogenes, Escherichia coli, Proteus vulgaris, Pseudomonas aeruginosa, Salmonella paraty-

phi A, Salmonella paratyphi G, Salmonella typhi, Shigella boydii, Shigella dysenteriae, Shigella sonnei, Vibrioparahaemolyticus ATCC 17802, and Vibrio cholerae ATCC e14035.

pH range of 5 to 10, and phage Aeh2 was stableat a pH range of 5 to 9 (Table 1). Popoff (14)found that Aeromonas salmonicida phages are

quite resistant to alkaline conditions, some beingresistant to pH 12 for 30 min at 26°C. In our

study, both phages showed excellent stability toalkaline pH. This stability may be due to the factthat both phages were isolated from water sam-

ples collected from the alkaline final effluent of a

sewage treatment plant.It was interesting to find that the infectivity of

phage Aehl increased by approximately fivefoldafter it was suspended at pH 10 for 1 h at 22°C(Table 1). This experiment was repeated withdifferent Aehl stocks, and similar results wereobtained. Further experimental studies are beingplanned to explore this phenomenon.For the serological studies, the K values for

the homologous system of neutralization were

339 and 353, but the K values for the heterolo-gous systems of neutralization (anti-Aehl serum

reaction with phage Aeh2 and anti-Aeh2 serum

NANA

NA

VOL. 45, 1983

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from

1676 CHOW AND ROUF

reaction with phage Aehl) were 0. It is thereforeconcluded that A. hydrophila phages Aehl andAeh2 are serologically unrelated.The two phages showed different degrees of

stability to physical and chemical agents. PhageAeh2 was more sensitive to chloroform and lessstable at 50°C than was phage Aehl. For stor-age, phages Aehl and Aeh2 were much morestable at 4 than at 30°C.

ACKNOWLEDGMENTS

We thank T. C. Hazen, University of Puerto Rico, RioPiedras, for supplying us with many Aeromonas cultures. Wealso thank Dorothy Parker, Department of Biology. Universi-ty of Wisconsin-Oshkosh, for helpful suggestions and readingthe manuscript.

LITERATURE CITED

1. Ackermann, H. W., A. Audurier, L. Berthiaume, L. A.Jones, J. A. Mayo, and A. K. Vidaver. 1978. Guideline forbacteriophage characterization, p. 1-24. In M. A. Lauffer,F. B. Bang, K. Maramorosch, and K. M. Smith (ed.),Advances in virus research, vol. 23. Academic Press,Inc., New York.

2. Adams, M. H. 1959. Bacteriophages. Interscience Pub-lishers, Inc., New York.

3. Bradley, D. E. 1967. Ultrastructure of bacteriophages andbacteriocins. Bacteriol. Rev. 31:230-314.

4. Burrell, R. 1974. Experimental immunology, 4th ed. Bur-gess Publishing Co., Minneapolis, Minn.

5. Buttiaux, R. 1959. The value of the association Escheri-chieae-group D streptococci in the diagnosis of contami-nation in foods. J. Appl. Bacteriol. 22:153-158.

6. Driessen, F. M., and J. Stadhouders. 1972. Suitability offour different media for the enumeration of pseudomonadsin milk. Neth. Milk Dairy J. 2:91-99.

7. Eddy, B. P., and A. J. Kitchell. 1959. Cold-tolerant fer-mentative gram negative organisms from meat and othersources. J. AppI. Bacteriol. 22:57-63.

8. Eisenstark, A. 1967. Bacteriophage techniques, p. 449-524. In K. Maramorsch and H. Koprowski (ed.), Methodsin virology, vol. 1. Academic Press, Inc., New York.

9. Feldman, H. A., and S. S. Wang. 1961. Sensitivity of

various viruses to chloroform. Proc. Soc. Exp. Biol. Med.106:736-738.

10. Gibbs, E. L., T. J. Gibbs, and P. C. Van Dyck. 1966.Rana pipiens health and disease. Lab. Anim. Care16:142-146.

11. Haschemeyer, R. H., and R. J. Myers. 1972. Negativestaining, p. 99-147. In M. A. Hayat (ed.), Principles andtechniques of electron microscopy: biological applica-tions, vol. 2. Van Nostrand Reinhold Co., New York.

12. Hazen, T. C., C. B. Fliermans, R. P. Hirsch, and G. W.Esch. 1978. Prevalence and distribution of Aeromonashydrophila in the United States. Appl. Environ. Microbi-ol. 36:731-738.

13. Paterson, W. D., R. J. Douglas, I. Grinyer, and L. A.McDermott. 1969. Isolation and preliminary characteriza-tion of some Aeromonas salmonicida bacteriophages. J.Fish. Res. Board Can. 26:629-632.

14. Popoff, M. 1971. Etude sur les Aeromonas salmonicida.II. Caracterisation des bacteriophages actifs sur les"aeromonas salmonicida" et lysotypie. Ann. Rech. Vet.2:33-45.

15. Rouf, M. A., and M. M. Rigney. 1971. Growth tempera-tures and temperature characteristics of Aeromonas.Appl. Microbiol. 22:503-506.

16. Schubert, R. 1976. Der Nachweis von Aeromonaden der"hydrophila-punctata-gruppe' im rahmen der hydienis-chen trinkwasserbeurteilung. Zentralbl. Bakteriol. Parasi-tenkd. Infektionskr. Hyg. Abt. 1 Orig. Reihe B 161:482-497.

17. Seeley, N. D., and S. B. Primrose. 1980. The effect oftemperature on the ecology of aquatic bacteriophages. J.Gen. Virol. 46:87-95.

18. Shotts, E. B., Jr., J. L. Gaines, Jr., L. Martin, and A. K.Prestwood. 1972. Aeromonas-induced deaths among fishand reptiles in an eutrophic inland lake. J. Am. Vet. Med.Assoc. 161:603-607.

19. Snieszko, S. F., and G. L. Bullock. 1962. Freshwater fishdiseases caused by bacteria belonging to the generaAeromonas and Pseudomonas. Fishery leaflet 459. Bu-reau of Sport Fisheries and Wildlife, U.S. Department ofthe Interior, Washington, D.C.

20. Snustad, D. P., and D. S. Dean. 1971. Genetics experi-ments with bacterial viruses. W. H. Freeman & Co., SanFrancisco.

21. Von Graevenitz, A., and A. H. Mensch. 1968. The genusAeromonas in human bacteriology. Report of 30 cases andreview of the literature. N. Engl. J. Med. 278:245-249.

APPL. ENVIRON. MICROBIOL.

on April 15, 2021 by guest

http://aem.asm

.org/D

ownloaded from