GROWTH RATES OF SALMONELLA COLONIES

J. L. STOKES AND H. G. BAYNE

Western Utilization Research Branch, Agricultural Research Service, United States Departmentof Agriculture, Albany 10, California

Received for publication February 4, 1957

Most species of Salmonella grow rapidly andluxuriantly on ordinary agar media. In contrast,Salmonella pullorum and a few other species ortypes such as Salmonella abortus-ovis, Salmonellasendai, and Salmonella typhi-suis develop slowlyand sparsely (Kauffmann, 1954). The reasons forthe poor growth of these strains are not known.Also, their sparse growth makes isolation moredifficult. In the present investigation, the rates ofgrowth of a large number of Salmonella on solidmedia, and to a limited extent in liquid media,have been determined under a variety of environ-mental conditions. Particular attention has beengiven to the slow growth of S. pullorum and toattempts to increase its rate of development.

MATERIALS AND METHODS

The rate and extent of growth of Salmonella onsolid media were determined by measuring thediameter of the colonies on trypticase soy agar,nutrient agar, and other media. The colonieswere obtained by streaking from 4 or 5 hourtrypticase soy broth cultures. An importantadvantage of this method is that most colonieswill arise from single cells and therefore the size ofinoculum for all of the strains in all experimentswas essentially identical and constant. Providedwell isolated colonies are measured, agreementbetween replicate colonies on a plate and betweenreplicate plates is close. Moreover, colony size isnot significantly affected by the depth of agar,total amount of growth on the plate, and age ofinoculum, in the range of 4 to 24 hr. Usually 4 to 8colonies on a plate were measured under a wide-field microscope equipped with a calibrated ocularmicrometer, at magnifications of 12, 24, or 36,and the results were averaged. Growth in liquidmedia was measured in the Klett-Summersonphotometer equipped with a red filter. All cul-tures were incubated at 35 C unless otherwiseindicated.The Salmonella species and strains investigated

included S. paratyphi B (2 strains), S. derby, S.sen*ftenberg (4 strains), S. meleagridis (2 strains),

S. anatum (2 strains), S. bareilly, S. panama, S.montevideo (2 strains), S. typhimurium, S.worthington, S. newport, S. oranienburg (3strains), S. muenchen, S. gallinarum (12 strains),and S. pullorum (10 strains). Towards the end ofour investigations several additional species andstrains were examined briefly. These includedS. typhi, S. sendai (2 strains), S. paratyphi A (3strains), S. abortus-ovis (2 strains), S. typhi-suis(2 strains), and S. fulica.

RESULTS

Growth rates. The extent to which Salmonellastrains grow in 17 hr at 35 C on trypticase soy,nutrient, and brilliant green agar is shown intable 1. The results are representative for all ofthe 44 strains originally examined. On trypticasesoy agar, all species except S. gallinarum and S.pullorum give rise to colonies which are approxi-mately 2 mm in diameter. In contrast, coloniesof all strains of S. gallinarum are about 1 mm indiameter and those of S. pullorum roughly 0.5mm. On this basis, therefore, growth of S. gal-linarum is only one-half and S. pullorum one-fourth that of the other Salmonella species. Thesedifferences are increased by a factor of two if theareas rather than the diameters of the coloniesare compared and even more if volumes are con-sidered. These striking differences in colony sizeof the 3 groups of Salmonella are shown in thephotographs in figure 1.

Similar colony diameter ratios were obtainedon nutrient and brilliant green agar. Growth,however, of all Salmonella strains was less onthese media than on trypticase soy agar. Thepoor growth of S. pullorum on brilliant green agaris especially significant because this medium isfrequently used to isolate S. pullorum from en-richment cultures. Such small colonies may bemissed if the plates are not carefully examined.Growth curves can be obtained by following

changes in colony diameter with time of incuba-tion as shown in figure 2. S. gallinarum andespecially S. pullorum exhibit a marked lag

200

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

GROWTH RATES OF SALMONELLA COLONIES

TABLE 1Differences in amount of growth of Salmonella on

solid media

Colony Diameter, mm*

Strain Number Trypti- Nutri- Brilli-case ent antsy agar green

agar agar

Salmonella typhimurium... Tm-1 2.2 1.7 1.1Salmonella oranienburg .... 200 E 2.6 1. 7 1.4Salmonella worthington..... 4661 2.1 1.6 1.7

Salmonella yallinarum ..... 271 1.4 0.8 0.5Salmonella gallinarum..... 3040 1.1 0.8 0.6Salmonella gallinarum..... 4411 1.1 0.8 0.6

Salmonella pullorum....... 3470 0.7 0.4 0.4Salmonella pullorum....... 3540 0.5 0.3 0.2Salmonella pullorum....... 3083 0.8 0.4 0.4

* After incubation for 17 hr.

period in growth compared to S. oranienburg.The latter is representative of the rapidly growingSalmonella. Moreover, the final size of the col-onies of the gallinarum-pullorum groups tends tobe less than that of the other Salmonella species.However, it is clear that even the slowly develop-ing S. pullorum colonies increase greatly in sizeif the incubation period is extended to 24 hr or

longer. Thus after a slow start, S. pullorumultimately makes considerable growth.

Similar growth curves were .obtained also inliquid medium. For these experiments, trypticasesoy broth was tubed in 7 ml amounts in Kletttubes, sterilized and then inoculated by needlefrom one-day-old broth cultures of Salmonella.The inoculated tubes were incubated at 35 C andturbidity readings were made at intervals. Typi-cal results are plotted in figure 3. S. pullorumexhibits a much longer lag period, a lower rate ofsubsequent multiplication, and also less totalgrowth than S. oranienburg.

Effect of temperature. The possibility existedthat S. pullorum grows slowly because the incuba-tion temperature of 35 C is not optimum. Accord-ingly, the effect of temperature on the growth ofSalmonella on trypticase soy agar was deter-mined. Representative data are given in figure 4.None of the Salmonella strains produce macro-

scopically visible growth at 15 C in one day. Thesame is true at 20 C except for some of the fastgrowing Salmonella species such as S. anatum

and S. senftenberg, which grow slightly and formtiny colonies. Characteristically good growth ofall strains occurs in the range of 29 C to 40 C.

Most of the rapidly growing Salmonella, however,develop best at 40 C, whereas 35 C is the opti-mum for the gallinarum-pullorum groups. At 46C, only an occasional strain of the rapidly grow-ing Salmonella develops and the9growth is slight.The Salmonella strains will grow fully at the

lower temperatures if the incubation period isprolonged beyond 1 day. At 20 C essentially fullgrowth of all strains is obtained within 3 days.The same is true at 15 C after 7 days except thatS. pullorum strains show no more than a trace ofgrowth. At 10 C, however, there is virtually nogrowth of any of the Salmonella except for slightgrowth of a few fast growing species after 2 to 4weeks of incubation. Not the slightest growth ofany of the Salmonella was obtained in 1 monthat 5 C and 1 C. The minimum temperature forthe growth of Salmonella is, therefore, approxi-mately 10 C. Likewise, the results at 46 C men-tioned above plus additional data obtained byuse of temperatures between 40 C and 46 C andby prolonged incubation at the latter tempera-ture indicate that 46 C is about the maximumtemperature for the growth of Salmonella. Similarresults were obtained when trypticase soy brothrather than agar was used.In general, however, S. pullorum and to a

lesser extent, S. gallinarum, do not tolerate theextremes of the indicated temperature range aswell as other Salmonella species. They also tendto develop best at 35 C compared to 40 C formost of the latter. It is clear that the slow growthof S. pullorum in the previous experiments cannotbe ascribed to the use of an unfavorable incuba-tion temperature.

Effect of pH. The media used in the previousexperiments were at pH 7.0 to 7.2. It was ofinterest to determine to what extent hydrogenion concentration influences the growth of Sal-monella, especially S. pullorum. Trypticase soyagar was adjusted to pH 5, 6, 7 and 8 by additionof appropriate mixtures of 1 M phosphate buffer.The buffers were sterilized separately and addedto the melted sterile trypticase soy agar inamounts sufficient to give a final concentrationof 0.05 M. In addition, a small amount of H3PO4was used to reduce the medium to pH 5. Theindicated pH values of the media are thosedetermined by direct measurement.None of the Salmonella strains grow at pH 5.

S. derby, representativ. of the rapidly growingstrains, grew well at pH 6 whereas S. gallinarumand especially S. pullorum grew poorly at this

1957] 201

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

STOKES AND BAYNE

Figure 1. Comparative size of colonies of different Salmonella species grown on trypticase soy agarfor 18 hr; natural size. (a) Salmonella worthington (b) Salmonella gallinarum (c) Salmonella pullorum.

[VOL. 74202

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

GROWTH RATES OF SALMONELLA COLONIES

Figure lc.

pH level. The optimum for all strains is pH 7and many but not all strains grow equally well atpH 8.

In another series of experiments in which theagar medium was adjusted with acid or alkali topH 5, 6, 7, 8, 9, and 10, results similar to thosewith phosphate buffer were obtained. Moreover,all strains grew considerably at pH 9 and 10. Atthe latter pH level, the gallinarum-pullorumgroups were more inhibited than the other Sal-monella strains. In trypticase soy broth, S.oranienburg, S. worthington, S. senftenberg and S.typhimurium, did not develop at pH 5 but madesome growth at pH 5.6 whereas four strains ofS. pullorum and two strains of S. gallinarumgrew only when the pH was raised to 6.0.

In general, therefore, S. pullorum and S.gallinarum do not grow as well as other Salmo-nella species at the extremes of the pH rangeused. In this respect, the situation parallels thatin the temperature experiments. Also, the growthrate of S. pullorum cannot be increased by shift-ing the pH level of the medium from pH 7; suchshifts usually lead to slower growth.

Stimulation of S. pullorum by yeast extract. Most

of our attempts to increase the rate of growth ofS. pullorum were unsuccessful. It was not in-creased by the addition of the following materialsto trypticase soy agar: (a) glucose or othersugars, (b) supernatant or aqueous extracts ofrapidly growing S. typhimurium, (c) a largevariety of plant and animal extracts, (d) CO2 andcompounds of the tricarboxylic acid cycle, (e)B-vitamins including pyridoxal phosphate, coen-zyme A and lipoic acid and also purines andpyrimidines. Moreover, substitution of a varietyof complete dehydrated media for trypticase soyagar did not lead to more rapid growth of S.pullorum.The only substance which markedly and con-

sistently increased the growth rate of S. pullorumon trypticase soy agar was yeast extract either inthe form of the dehydrated product (Difco) or asyeast autolysate or the Basaminobact of An-heuser-Busch. The extent of stimulation is shownin table 2. The diameter of S. pullorum coloniesincreased by approx 40 to 70 per cent and ap-proached or exceeded the diameter of S. gallin-arum colonies on trypticase soy agar alone. Therapidly growing Salmonella and S. gallinarum

19571 203

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

STO1KES AND BAYNE

HOURS

Figure B. Growth rates of Salmonella on trypti-case soy agar.

HOURS

Figure S. Growth of Salmonella in trypticasesoy broth.

also are stimulated by yeast extract so that theeffect is not a preferential one. Five tenths per

cent yeast extract is as effective as 2 per cent andgreater amounts may be inhibitory. There is verylittle stimulation with 0.1 per cent yeast extract.The responsible factors in yeast extract are notknown but they are not inorganic, since the ash isnot stimulatory. Yeast extract also increasesgrowth in trypticase soy broth. Interestingly, itdoes not appreciably reduce the lag period ofS. pullorum but rather increases its rate of growthduring the logarithmic phase and also the totalcell crop. Growth stimulation of S. pullorum by

yeast extract has been noted also by Gilfillan etal. (1955).The data in table 3 show that the addition of 2

per cent yeast extract to the selective media,brilliant green, SS, and bismuth sulfite agar

(commonly used for the isolation of Salmonella)also increases the growth of S. pullorum. Theseresults and others that we have obtained indicatethat, in general, strains of S. pullorum do notgrow as well on selective agars as on trypticasesoy agar and that some strains grow very poorlyon such agars. Bismuth sulfite medium is the

INCUBATION TEMPERATURE

Figure 4. Effect of temperature on growth ofSalmonella on trypticase soy agar.

TABLE 2Stimulation of growth of Salmonella on trypticase

soy agar by yeast extract

ColonyDiameter, mm*

PerStrain Number Wito With 2 Cent

Without Increaseyeast per cent4=tct yeast

Salmonella pullorum......... 3179 0.54 0.74 37Salmonella pullorum......... 3470 0.52 0.87 67Salmonella pullorum......... 1431 0.71 1.1 55Salmonella gallinarum....... 222 1.0 1.2 20Salmonella gallinarum ....... 4411 1.0 1.3 30Salmonella senftenberg ....... 3252 2.0 2.3 15Salmonella oranienburg ...... 6266 2.1 2.8 33

* After incubation for 16 hr.

204 [VOL. 74

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

GROWTH RATES OF SALMONELLA COLONIES

TABLE 3Increased growth of Salmonella pullorum onaddition of yeast extract to selective agar media

Colony Diameter, mm*Per Cent

Strain Number extrasct Trypti- Brimi- Bis-Etatcase ant SS muth

s-Isoy green fite

Salmonella 3230 0 0.80 0.50 0.300.17pullorum 2 1.37 0.66 0.690.35

Salmonella 3470 0 0.67 0.72 0.810.16pullorum 2 1.00 1.12 1.010.26

* After incubation for 18 hr.

most inhibitory and brilliant green the least.Inhibition of Salmonella by bismuth sulfite agar

when streaked from selenite enrichment cultureshas been reported by Williams Smith (1952).The addition of yeast extract to selective agar

media would be helpful in the isolation of S.pullorum provided the selective properties of suchmedia are retained in the presence of yeastextract. Experiments with Proteus vulgaris,Escherichia coli, Aerobacter cloacae and Alcal-igenes faecalis indicate that the selective activityof brilliant green is largely nullified by yeastextract as might be expected from previouslypublished data (Stokes and Osborne, 1955).However, the inhibition of the above mentionedbacteria was equally as effective on SS and bis-muth sulfite agar with yeast extract as without it.

Additional slow growing Salmonella. After com-

pletion of most of the experimental work, addi-tional cultures of Salmonella which have beenreported to grow slowly on ordinary agar mediawere obtained from various sources and examined.These included S. typhi, S. paratyphi A, S. sendai,S. fulica, S. abortus-ovis and S. typhi-suis. Onboth trypticase soy and nutrient agar, S. typhi, 2strains of S. paratyphi A and 1 strain of S. sendaigrew as well as other rapidly developing Sal-monella. One strain each of S. paratyphi A, S.sendai and S. abortus-ovis and also S. fulicaexhibited the rate characteristic of S. gallinarum.The remaining S. abortus-ovis strain grew as

slowly as S. pullorum. The latter results are inaccord with those of Lovell (1931) who reportedan average diameter of 0.7 mm for 6 strains ofS. abortus-ovis when grown on agar for 18 hr at37 C. The 2 cultures of S. typhi-suis developedextremely slow on both media; readily visible

colonies were present only after 2 days of incuba-tion. This organism, therefore, grows even moreslowly than S. pullorum. As in the case of otherSalmonella strains, the 6 species were stimulatedby the the addition of 2 per cent yeast extractto trypticase soy agar.The extremely slow growth of S. fulica, S.

abortus-ovis and S. typhi-suis on brilliant greenagar is particularly striking and significant. Thecolonies were 0.1 mm or less in diameter after 17hr incubation. These organisms could easily bemissed during isolation from natural sourcesunless extended incubation periods were used andthe plates carefully examined. Such small colo-nies do not redden the medium surrounding them.S. fulica frequently failed to grow on brilliantgreen agar and only occasionally formed a fewsmall colonies.

DISCUSSION

The results indicate that most species of Sal-monella grow rapidly on solid culture media butthat strains of S. pullorum develop much moreslowly while those of S. gallinarum occupy anintermediate position. The differences in growthrates between the 3 groups are large and constantand may offer still another diagnostic means foridentifying and differentiating S. pullorum and S.gallinarum (Hinshaw, 1941).The slow development of S. pullorum compared

to other Salmonella is due to a combination of alonger lag period and a lower rate of multiplica-tion during the exponential phase. The morelimited pH and temperature tolerance of S.pullorum may be due to this lower growth rateand may disappear when rapid growth isachieved. The slow growth of S. pullorum differsfrom that of the dwarf colonies of S. eastbourneand S. typhi (Kauffmann, 1954) in that the latterdo not increase in size on continued incubation.Also, the dwarf colony variants give rise tonormal sized colonies when grown in the presenceof small amounts of thiosulfate or sulfite, whereasthese compounds have no effect on the growth ofS. pullorum.The difficulty encountered in preferentially

increasing the growth rate of S. pullorum bymodification of pH, temperature, nutrients andgrowth factors suggests that inadequate perme-ability of the cells to nutrients may be a con-controlling factor in its slow growth. In thisconnection it may be significant that S. pullorumstrains require exogenous supplies of amino

20519571

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

STOKES AND BAYNE

acids for growth (Johnson and Rettger, 1943). Ifpermeability to essential amino acids were indeedlimiting, this could account for the slow growth ofS. pullorum. Interestingly, many strains of S.gallinarum, S. sendai, S. paratyphi A and all ofthe other slow growing Salmonella also requireeither growth factors or amino acids or both,since they will not grow in a medium of glucose,ammonium nitrogen, and inorganic salts. Incontrast virtually all other Salmonella strains-and these grow rapidly-can develop in this orsimilar simple media (Kauffmann, 1954; Leder-berg, 1947). Nutritional data on our strains sup-port these conclusions. The rapidly growingSalmonella developed well in a medium composedof glucose, citrate, ammonium nitrogen andmineral salts, whereas the slowly growingstrains required amino acids for growth and insome instances also B-vitamins. RecentlyClowes and Rowley (1955) suggested thatpermeability may be the limiting factor in thegrowth of small colony variants of Escherichiacoli. We have experiments now under way todetermine to what extent the metabolism ofamino acids and permeability influence thegrowth rates of S. pullorum and other Salmonella.

ACKNOWLEDGMENTS

We are greatly indebted to Dr. Phillip Ed-wards of the Communicable Disease Center,Chamblee, Georgia, Dr. Ivan Saphra of theNational Salmonella Center, Beth Israel Hos-pital, New York City, and Dr. F. Kauffmann ofthe State Serum Institute, Copenhagen, Den-mark, for supplying us with many of the Salmo-nella cultures.

SUMMARY

The growth rates of a large number of Sal-monella strains on agar media, and to a morelimited extent in liquid media, have been deter-mined under a variety of cultural conditions.Most species grow rapidly on nutrient andtrypticase soy agar and form colonies that are1.5 to 2.5 mm or more in diameter in 18 hr at35 C. In contrast, all strains of Salmonella pul-lorum develop slowly and the colonies are there-fore small, 0.3 to 0.8 mm, or approximately one-fourth the diameter of the rapidly growingSalmonella. Strains of Salmonella gallinarumexhibit growth patterns that are intermediate

between the above two groups. The slowerdevelopment of S. pullorum is due to a combina-tion of a longer lag period and a lower rate ofmultiplication during the exponential phase.The minimum, optimum and maximum growth

temperatures for most Salmonella species are10 C, 40 C, and 46 C, respectively. The pullorum-gallinarum groups grow best at 35 C and do nottolerate the extremes of the temperature rangeas well as other Salmonella. They also do notdevelop as readily as the rapidly growing Sal-monella at pH 5 to 6 and above pH 8.The rate of growth of S. pullorum on trypticase

soy agar could not be increased by a variety ofchanges in physiological and nutritional condi-tions. Only the addition of yeast extract markedlyincreased the growth rate. This occurred alsowith the other Salmonella species. The dataindicate that the addition of yeast extract toselective agars such as SS and bismuth sulfitemay be helpful in the isolation of slow growingSalmonella.

REFERENCESCLOWES, R. C. AND ROWLEY, D. 1955 Genetic

studies on small-colony variants of Esche-richia coli K-12. J. Gen. Microbiol., 13,461-473.

GILFILLAN, R. F., HOLTMAN, D. F., AND Ross, R.T. 1955 A synthetic medium for propaga-tion and maintenance of virulent strains ofSalmonella pullorum. Poultry Sci., 34, 1283-1288.

HINSHAW, W. R. 1941 Cysteine and relatedcompounds for differentiating members of thegenus Salmonella. Hilgardia, 13, 583-621.

JOHNSON, E. A. AND RETTGER, L. F. 1943 Acomparative study of the nutritional require-ments of Salmonella typhosa, Salmonellapullorum, and Salmonella gallinarum. J.Bacteriol., 45, 127-135.

KAUFFMANN, F. 1954 Enterobacteriaceae, 2nded., pp. 23-45. E. Munksgaard, Copenhagen.

LEDERBERG, J. 1947 The nutrition of Sal-monella. Arch. Biochem., 13, 287-290.

LOVELL, R. 1931 A member of the Salmonellagroup causing abortion in sheep. J. Pathol.,34, 13-22.

STOKES, J. L. AND OSBORNE, W. W. 1955 Aselenite brilliant green medium for the isola-tion of Salmonella. Applied Microbiol., 3,217-220.

WILLIAMS SMITH, H. 1952 The evaluation ofculture media for the isolation of Salmonellafrom faeces. J. Hyg., 50, 21-36.

206 [VOL. 74

on January 2, 2020 by guesthttp://jb.asm

.org/D

ownloaded from