An improved medium for distinguishing between homofermentative and heterofermentative lactic acid bacteria

•

¡"'emmional ¡(Jumal of Food Microbiolob'Y, 18 (J 993) 37-42© 1993 Elscvicr Scicnce Publishers B.V. Al! rights rcscrvcd OI68~1605/93/$06.00

FOOD 00562

An improved medium for distinguishing betweenhomofermentative and heterofermentative lactic

acid bacteria

M. Zúñiga, 1. Pardo and S. FerrerDepartamellf de Microbiologia, Facultar de C¡imcies Biologiques, Vlliversifaf de Valencia, 811rjmwl,

Valencia, Spain

(Received 30 April 1992; accepled 3 September 1992)

37

An improved salid medium (Of differentiating between homofermentative and heterofermentative lactieacid bacteria is proposed. 11 was developed lo support ¡he growlh uf wine slrains unable lo grow inothcr media. However, il can be employed as a general medium (Of the lactic acid bacteria tha! ulilizefructase.

Key words: Lactic acid bacteria, homofcrmenlalive, hcterofermcntativc; Differential medium

Introduction

Lactic acid bacteria (LAB) are an extensive group of Gram-positive microorgan­isms that include the genera Carnobacterium, Enterococcus, Lactobaci/lus, Lacto­coccus, Leuconostoc, Pediococcus and Streptococcus. AJI are obligate fermentativebacteria with compIex nutritional requirements (Kandler, 1983; Sneath et al.,1986). They are found in a wide variety of fermented beverages and foods, theintestinal traet of animals, pIant materials, ete. (Sneath et al., 1986). Laetie aeidbacteria play an important role in winemaking: they transform L-malic acidnaturally present in wines into L( + }-Iactic acid and COz, OT, in other words,perform malolaetie fermentation.

LAB can be divided into two physiologieal groups, depending on their hexosefermentation pathway: homofermentative LAB and heterofermentative LAB. Thehomofermentative LAB degrade hexoses via glycolysis, produeing laetie aeid as themajor end-produet, whereas the heterofermentative LAB use the pentose-phos­phate pathway and yield laetie aeid, CO,' aeetie aeid and/or ethanol. LAB eanaIso produce other end-produets, depending on the speeies and growth eonditions(Condon, 1987; Kandler, 1983; Sneath et al., 1986).

The methods usually empIoyed to distinguish between homofermentative andheterofermentative LAB are based on the produetion of CO, by heterofermenta-

Correspondence address: Sergi Ferrer, Departament de Microbiologia, Universitat de Valencia, E-46100Burjassot-Valencia, Spain. Te!. +34 (6) 3864390; Fax +34 (6) 3864372.

38

tive LAR (Gibson and Abd-EI-Malek, 1945). We have routinely used this system toidcntify LAB strains isolated from Spanish musts and wines. However, non-repro­ducible results were obtained with sorne of these strains (unpublished results).

McDonald et al. (1987) designed a differential medium named HHD that madeit possible to distinguish between homofermentative and heterofermentative LARboth in solid or liquid media. This method is based on the greater acid productionfrom a fixed amount of fructose by homofermentative LAB. As the methodinvolves a pH indicator, differences in pH can be observed by means of detectablecolor changes. Homofermentative LAB produce blue colonies, whereas heterofer­mentative LAB remain white. McDonald et al. (J 987) tested HHD medium withseveral genera of LAB and obtained clear differentiation.

We evaluated HHD medium with LAR isolated from musts and wines andfound that several strains were unable 10 grow or did so very poorly. We thereforeevaluated an alternative medium (M5) able to support the growth of wine strains.Further studies showed that M5 could be used as a general medium for distin­guishing homofermentative trom heterofermentative LAB that utilizc fructase.

Materials and Methods

The components of M5 medium are given in Table 1. It is based on the MLümedium (Claus et al., 1983), with the following modifications: omission of glucose,

TABLE J

Composition of differential media M5, HHD and MLO for homo- and helerofermenlative laclie acidbacteria

Componenl (g/l) M5 HHD MLO

Tryptone JO lO JOVeas! extrae! 5 I 5Casaminoacids 3Baclosoytone 1.5Glucose JOFructose 2.5 2.5 5KH 2 P04 2.5 2.5Tween 80 J mi I g I miL-Cysteine Hel 0.5 0.5MgSO,'7H,O 0.2 0.2MnS04 "lH 2 O 0.05 0.05(NH 4 )z citrate 3.5Calcium pantothenate 0.01Tomato juice )00 miBromocresol greeo " 20 mi 20 miAgar 20 20 20pH 6.5 7.0 4.8

" Stock solution, 0.1 g of bromocresol greeo in 30 mI of 0.01 N NaOH.

39

tomata juice and diarnmonium citrate, and the addition of potassium phosphatc,calcium pantothenate, and also of fructase as the sale carbon saUTce, andbromocresol green as a pH indicator. The pH was adjusted to 6.5. Strains used inthis study are listed in Table 11, and were oblained from the Colección Españolade Cultivos Tipo (CECT)· or directly isolated from wines. SlreplOcoeeus andEnterococcus strains were roulinely grown in Braio Heart Infusion (Oxoid) al 3rC,Laeloeoccus strains in M17 brolh (Terzaghi and Sandine, 1975), with glueoseadded instead of lactase, al 32°C. Carnobacterium strains (synonyrnolls Lactobacil­lus) were grown in YGPB (Garvie, 1978) at 28°C, and Laelobacillus, Pediocoecus,and Leuconostoc strains except Leucol1ostoc oenos in MRS broth (de Mao el al.,1960) at 28°C. L. oenos strains were grown in MLO medium (Claus et al., 1983) at28°C. AII the media were sterilized by autociaving for 30 min at IJ5°C.

Inocula for the assays with HHD or M5 were prepared as follows: eells weregrown unlil the early slationary phase in lhe appropriate medium, harvested bycenlrifugalion, and washed twice wilh sterile dislilled waler. The pellet wasresuspended in distillcd water and appropriatc dilutions were sprcad onto plales ofHHD or M5 media. Unless otherwise indicated plates of HHD were incubatedunder aerobie conditions (ambient atmosphere), as McDonald et al. (1987) de­scribed, and plates of M5 in anaerobic jars (Oxoid) under an atmosphere of IUU%CO, at 25"C to improve the use of fructose as an electron acceptor.

The differentiation bctwecn homofermentative and heterofermentativc LAB inmixed populations was investigatcd, and variations in viable counts in M5 andstandard media for isolation of LAB (de Man et al., 1960) were also tested.Lae/obacillas planlarum CECT 220 (homofermentative) and LaelObacillus eellobio­sas CECT 562 (hetcrofermentative) were used beeause they produce colonies withdifferent morphologies. Cell suspensions of both strains were spread on M5 piatesat ralios of I : 1, I : 2 and 1: 10.

Results and Discussion

The results of the assays of the strains of LAB tested in HHD and M5 areshown in Table 11. As in HHD medium, homofermentative LAB colonies wereblue in M5, while heterofermentative colonies remained white. None of the L.oenos strains nor Lactobacillus fructivorans UR 3839 could grow in HHD cvenafter 20 days of incubation. Minor modificalions in HHD composition or culturecondilions (variations in tempcraturc or presence of CO,) did not allow the growthof thcsc strains. Furthcrmorc, Carnobacterium diuergens CECT 4016, Carnobac­leriam piscieola CECT 4020, LaelObacillas acidophilas CECT 289, Laelobacillasalimenlarias CECT 570, Laetobacillas easei ssp. casei CECT 475 and Laelobacil/as[ermenlam CECT 285 showed an inappropriate response in HHD. Most of thestrains tested showed a betler growth in M5 medium than in HHD. L. acidophilasCECT 289, SlreplOcoeeus milis CECT 804, SlreplOcoeeas mutans CECT 479 andStreptococcus saliuarius ssp. saliuarius CEeT 805 could not grow on M5 plates at25"C, but they did at 37°C. For the rest of LAB strains the incubation temperature

40

TABLE 11

Differential growth of laclic acid bacteria on HHD and M5 media

Species

CarllObllcleriwlIC. di¡;erget/sC. piscico/a

EllterococC/H

E. faccalisE. mwultü

LllctobacillllsL. llcidoplúlllSL. alimentariusL. brel'isL. casé ssp. caseiL. cdlobios/lsL. CllnJatll5

L. fermellfllm

L. frueriroran.\'L. /¡ilgardiiL. plalltllrlllll

L. plallfaml1l

L. sakcL. ¡'irideseens

Lactococcl/SL. [actis ssp. cremorisL. l(letís ssp. laclis

L. ml/illolac'isLellcoflo.\"fUc

L. mesertteroides ssp. dexrrafliclIt1IL. mesellteroides ssp. mesemeroidesL. (JalOs

L. oerlOS

L. OC/lOS

L. UC1l05

L. OC1l0S

L. OC/lO.\'

L. oe'IO~'

L. oeflOS

L. pllramesellferoides

PetliococCllsP. acidilacricil'. dam"o~ilu

P. parv/I/usP. pelltosllCCIIS

StreplOCocClIS

S. mifis

S. mlllilflS

S. salirariw; ssp. salil'arillsS. sal/guis

* Aberran! resull.Blue colonies.

b Whitc colonies.Assay 1101 carricd out.

d No growth.

L

Slrain

CEcr 4016CEcr402D

CEcr 187CEcr972

CEcr 289CEcr 570CEcr 216CEcr 475CEcr 562CEcr 904CECT 285UR 3839UR 3817CEcr 220UR 3809CEcr 906CEcr283

CEcr697CECT916CEcr988

CEcr 912CEcr 215UR 3840UR 3841UR 3843UR 3866UR 3872UR 3873UR 3874ML34 (CEcr 218)UR 3816

CEcr98CEcr 793CEcr 813CECT 923

CEcr804CEcr 479CEcr 805CEcr 480

Medium

HHD

B a

B *

B

w*w*ww*WB

B *N°WBB

B

NNNNNNNNW

B

B

M5

BW b

BB

BBWBWBWWWBBBB

BBB

WWWWWWWWWWW

BBBB

BBBB

41

did not modify the response in M5, and the time required to observe the test resultwas strain-dependent. Most of tbe strains grew in 7 days or less, except lhe L.oenos and Lactobacillus fructivorans UK 3839 slrains, which needed more tban 10days.

In the development of Ihe M5 medium, different amounts of nutrients weretested, and their concentration optimized: e.g., too much yeast extract (lO gil) orcalcium pantothenate (O. ¡ gil) could yield anomalous results. Factors affecting thefinal pH of the M5 medium were also analysed, because an ¡ncrease in the acidproduction can result in heterofermentative LAB being assessed as homofermenta­tive. We optimized the phosphate concentration, the ¡ni tia! pH value, the type andamounl of sugar, and the innuence of oxygen. As LAS can use oxygen as anelectron acceptor, this can lead lo a change of lhe end-producls (Candon, 1987;van Seelen et al., 1986). When M5 was used, incubation under aerobic conditionsled to inappropriate response in sorne cases due to a higher acid production by theheterofermentative strains. Heterofermentative LAS can increase the acid produc­tiao under aerobic conditions for two rcasons: first, more fructase can be COI1­

verted into acids because it is nol reduced to mannilol; secand, more acetic acidcan be produced (Condon, 1987). Reduction of fructose to mannitol is essential toobtain the appropriate results from all strains. If the reduction of fructose tomannitol is not required for the reliability of the method, the replacement of thissugar by glucose will not modify the results in either aerobic ar anaerobicconditions. When fructase was replaced by glucose in the M5 medium, weobserved lhat for sorne strains the growth was very poor, and many heterofermen­tative strains behaved as homofermentatives both under aerobic and anaerobicconditions. This is also suggested by the facl that the heterofermentalive Lacto­bacillus viridescens behaves as homofermentative even on M5 piates incubatedunder CO, atmosphere (Table 11), because it does not produce mannitol fromfructose (Holzapfel and Gerber, 1983). Although Carnobacterium divergells (syn.Lactobacillus divergens) was initially described as an atypical heterofermenter(Holzapfel and Gerber, 1983), furlher slUdies reclassified it as a homofermentativeorganism (De Sruyn et al., ¡987; 1988). So the resulls provided by this organismboth in HHD and M5 media were correcl.

The possibility of using M5 medium for direcl isolation and differentiation ofLAB from natural environments was also tested by inoculating in the same platesboth L. plantarum and L. brevis at different ratios. Alter incubation it wasobserved that colonies of both strains yielded the appropriate results, and therelative ralios were maintained. Similar results had been reported by McDonald etal. (1987) with HHD for combinations of several strains of LAS.

Therefore, M5 can be employed as a general medium for differentiatingbetween homofermentative and heterofermentative lactic acid bacteria that utilizefructose. Best results are achieved when plates are incubated under an atmosphercof 100% CO, at 25°C.

42

Acknowledgements

This research was supported by the Comisión Interministerial de Ciencia yTecnología (BI089-0430) and by a gran! from the Generalitat Valenciana 10 M.Z.

References

Claus, D., Lack, P. and Neu, P. (1983) O.S.M. Catalogue of strains. Dcutsche Sarnmlung vonMikroorganismcn.

Condon, S. (1987) Responses uf lactic acid bacteria lo oxygen. FEMS Microbio!. Rev. 46, 269-280.Dc Bruyn, I.N., Louw, A. l., Visscr, L. and Holzapfcl, W.H. (1987) LaC!obllcillllS divergclls is a

homofermenlative organismo Syst. Appl. Microbio!. 9,173-175.Oc Bfuyn, I.N., Holzapfel, W.I-I., Visscr, L. and Louw, A.1. (988) Glucose metabolism by Lac/obacilllls

divcrgem. J. Gen. Microbio!. 134,2103-2109.de Man. J.c.. Rogosa, M. and Sharpe, M.E. (1960) A medillm for ¡he cultivation of lactobacilli. J. Appl.

Bacterio!. 23, 130~ 135.Garvic, E.I. (1978) Streptococcu.\· mffillofaclis (Orla-Jcnsen and Hansen): a group N streptoeoccus found

in raw milk. Inl. J. Syst. Bacteriol. 28,190-193.G¡bson. T. and Abd-EI-Malek, Y. (1945) The formation of CO 2 by laelle acid bacteria and Bacifflls

ficheni[ormis and a cultural method of detecting Ihe proccss. J. Dairy Res. 14, 35-44.Holzapfel, W.H. and Gerber, E.S. (1983) Laclobaciffll.~ divergefls sp. nov., a new heterofermentative

Lactubacill/ls spccics producing L( + )-Iactate. Syst. Appl. Microbio!. 4, 522-534.Kandler, O. (1983) Carbohydrale melaholism in laetle acid baeleria. Antonie van Leeuwenhoek. 49,

209-224.McDonald, L.e., McFeelers, R.F., Daeschel, M.A. and Flemlng, H.P. (1987) A differenlial medium for

Ihe enumeration of homofermentative and heterofermentative lactic acid bacleria. Appl. Environ.Microbiol. 53, 1382-1384.

Sneath, P.H.A., Mair, N.S., Sharpe, M.E. and Holt, J.G. (1986) Bergey's Manual of SystemalicBacleriology, Vol. 2, W¡lliams & Wilk¡ns, Baltimore, MD.

Terzaghi, B. and Sandine, W.E. (1975) Improved medium for laclie slreplococcí and their bacterio­phages. Appl. Microbio!. 29,807-813.

van Beelen, P., van der Hoeven, J.S., de Jong, M.H. and Hoogendoorn, H. (1986) The effect of oxygenon the growth and acid production uf Srreprococcus mlUallS and Streptococclls sanguis. FEMSMicrobio!. Eco!. 38, 25-30.

,