HAL Id: hal-00929460https://hal.archives-ouvertes.fr/hal-00929460

Submitted on 1 Jan 1995

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

The bacterial flora of surface-ripened cheeses withspecial regard to coryneforms

F Eliskases-Lechner, W Ginzinger

To cite this version:F Eliskases-Lechner, W Ginzinger. The bacterial flora of surface-ripened cheeses with special regardto coryneforms. Le Lait, INRA Editions, 1995, 75 (6), pp.571-584. �hal-00929460�

Lait (1995) 75, 571-584© Elsevier/INRA

571

Original article

The bacterial flora of surface-rlpened cheeseswith special regard to coryneforms

F Eliskases-Lechner, W Ginzinger

Bundesanstalt für Alpenlëndiscne Milchwirtschaft, A-6200 Rotholz, Austria

(Received 21 March 1995; accepted 20 June 1995)

Summary - Randomly-selected Austrian bacterial surface-ripened cheeses were examined forchanges in the microbiological composition of the smear. The bacterial counts of the Tilsit cheeses from14 cheese plants and of 3 types of soft cheeses selected varied fram 1()4 ta 109 cfu/cm2 smear 3 d aftermanufacture and fram 108 ta 109 cfu/cm2 smear after a ripening period of 3 weeks. The flora tolerateda NaCI content of at least 80 g/kg in the plate count agar. A total of 386 isolates of coryneform bacte-ria were identified. The bacterial flora proved ta be of mixed population. However, Brevibacteriumlinens accounted for a large share of the flora, comprising 30% of the total bacterial count, Besides Bre-vibacterium linens, the other main types found ta be present in the heterogeneous flora were cream-coloured and yellow-pigmented coryneforms, which were predominantly identified as Arthrobacterglobiformis and Brevibacterium ammoniagenes. The coryneforms isolated fram the cheeses 3 d aftermanufacture were more prateoly1ic than those isolated at later stages of ripening.

Tilsit cheese / smear-ripened soft cheese / smear / coryneform bacteria / cheese ripening

Résumé - Flore bactérienne de surface des fromages à pâte molle, en particulier les coryné-formes. L "évolution de la flore microbienne sur la croûte des fromages autrichiens emmorgés a été exa-minée. Des Tilsit provenant de 2 fromageries ainsi que 3 différents types de fromage à pâte mollefabriqués par la même fromagerie ont été étudiés pendant toute la période de raffinage, de même quedes Tilsit provenant de 12 fromageries à 3 différents stades d"affinage. Le nombre des germes aéro-bies mésophiles dans la morge était compris entre 1()4 et 109 ufc par cm2 de morge 3 j après la pro-duction et entre 1()B et 1()B ufc par cm2 après 3 sem d'etîineçe respectivement. La flore a toléré uneteneur en NaCI de 80 glkg dans le milieu. Au total, 386 souches de bactéries corynéformes ont été iden-tifiées à raide de 52 tests biochimiques. La flore de ces morges se trouve être une population mixte.Brevibacterium linens, considéré comme dominant dans la morge, représentait 30% de la flore bac-térienne totale. Les autres espèces principales de cette flore hétérogène étaient de couleur jaune ounon pigmentées. Elles ont été identifiées en grande partie comme Arthrobacter globiformis et Brevi-bacterium ammoniagenes. Les bactéries corynéformes isolées de la morge des fromages de 3 jétaient plus protéolytiques que celles isolées plus tardivement.

Tilsit / fromage à pâte molle / marge / bactérie corynéforme / affinage

572 F Eliskases-Lechner, W Ginzinger

INTRODUCTION

On the surfaces of red smear cheeses, bac-terial counts of between 109 and 1011 cfu/gsmear (1 OL 109 ctu/crn-' su rface) arereached within the first 2 weeks of ripeningand remain constant until the time of con-sumption (Accolas et al, 1978; Keller andPuhan, 1985; Grand et al, 1992). The vari-ations between different cheese varietiesor cheese factories are low. Only during theinitial period of ripening do differences occurwith regard to the rapidity of the changefrom yeast dominance to bacterial domi-nance.

The importance of the coryneforms forthe surface ripening of smear cheeses isoften stated, but only a few studies havebeen made of the identity of these bacte-ria. The following species were found onthe surfaces of different cheese varieties:Arthrobacter aurescens, Arthrobacter cit-reus, Arthrobacter globiformis, Arthrobac-ter nicotianae, Arthrobacter protophormiae,Arthrobacter uratoxydans, Arthrobacter vari-abilis, Arthrobacter spp, Brevibacteriumammoniagenes, Brevibacterium erythro-genes, Brevibacterium helvolum, Brevibac-terium linens, Brevibacterium oxydans andCorynebacterium spp (Sei 1er,1986; Sauter,1986; Mayer, 1990; Piton and Fontanier,1990; Piton-Malleret and Gorrieri, 1992).Among these species, the grey, the whiteand the cream-coloured ones formed themain part (40%), the yellow strains 36% andthe orange-pigmented strains only 22%(Sauter, 1986), in spite of using Brevibac-terium linens cultures for each smearing.This is why some commercial cultures foruse in smear-cheese ripening presently con-tain different coryneforms besides Bre-vibacterium linens such as Brevibacteriumspp and Arthrobacterspp.

ln Austria it is common practice to useBrevibacterium linens as the sole cultureorganism. Ali other microorganisms foundon the cheese surfaces, both the bacteria

and the yeasts, are caused by contaminationdu ring the manufacturing and ripening pro-cesses. In a previous paper (Eliskases-Lechner and Ginzinger, 1995), we reportedon the yeast flora of su rface-ripenedcheeses. The aim of the present study wasthe quantification and characterization ofthe bacterial flora in the smear of thesecheeses.

MATERIALS AND METHODS

Origin of cheese samples

The smears of Tilsit cheeses from 2 plants (A,B) and 3 varieties of soft cheeses from 1 plant(C) were examined during the ripening period. Inaddition, Tilsit cheeses from 13 Austrian dairyplants were examined at 3 stages of ripening, onthe 3rd, 7th and 21st d alter manufacture.

Enumeration and isolation

The smear film of 100 cm2 of the cheese surfacewas scrapedoff with a sterilespatula,dilutedseri-ally in Ringer solution and plated on plate countagar + 30 g/kg NaCI, 80 g/kg NaCI and 150 g/kgNaCI(PCA; Merck5463); it was then incubatedat30°C for 12 d (Seller, 1986) in daylight to makepigment production possible for those coryne-forms which are colourless alter groW1hin dark-ness (Mulder et al, 1966; EI-Erian, 1969). Theyeast developmenton the plateswas suppressedby Pimaricin; before use 0.1 ml of a suspension(60 mg Pimaricin/3 ml H20) was plated on thesurface of the PCA (Sauter, 1986; Engel, 1993).Three pigmented groups (orange, yellow, cream)were counted separately. Ali counts are givenper cm2 smear surface; the use of 9 smear sur-face as a unit would result in values approxi-mately 100 x higher. Five colonies of each mor-phological type were selected randomly from thePCA plates, purified and stored on plate countagar slants.

Yeast populations were detected on yeastextract-glucose-chloramphenicol-agar (YGCA;Merck 1600) with 10 /lg/kg bromophenol blue

Bacterial flora of Tilsit cheese

(Rapp, 1974) and incubated at 25°C for 5 d(IOF,1990).

Identification

Three hundred eighly-six coryneforms were iden-tified using 52 biochemical characteristicsdescribed by Seiler et al (1980), Seiler (1983,1986) and Valdés-Stauber and Sei 1er(1990, per-sonal information), including assimilation of sug-ars, ami no acids and acids, production of acidfrom carbon sources, nitrate reduction, NaCI tol-erance, KOH test (Greez and Oack, 1961), hydrol-ysis of xanthine, starch, gelatine, casein andcatabolism of tyrosin. A microtitre tray techniqueand the computer program by Seiler and Braatz(1988) were used for the identification routine.

A screening of proteolytic activities on cal-cium-caseinate agar, with a 3-mm layer (Frazierand Rupp; Merck 5409) was made. The diameterof the clear zone around the colony was mea-sured as the parameter of proteolytic activity.

10

9

8'"Eo 7--::::>u..Ü 60>0

5

4

30

573

RESULTS

Changes in bacterial countsduring ripening

Figures 1 and 2 show the increase in bac-terial counts on the surface of Tilsit cheesesfrom 2 plants (codes A, B) during ripening.The bacterial counts of the Tilsit cheesesafter removal from the brine were104 ctu/cm'' in plant A and 106 ctu/crn'' inplant B. lt must be considered that at thisstage of ripening no "smear" had developed.Within 20 d, the maximum count, 109

ctu/crnè, was reached and the countremained at this level until the time of con-sumption. The counts of the 80 g/kg-NaCI-resistant bacteria were equivalent ta the totalnumber of cheese surface bacteria; accord-ingly, the bacterial flora can be character-

1---,/-/------------------1-- total count-4- 80 g/kg NaCI-<>- 150 g/kg NaCI

10 30 40 50 8C60 70ripening time (d)

Fig 1. Oevelopment of the 80 g/kg-NaCI- and 150 g/kg-NaCI-tolerant bacteria compared with the totalbacterial count of Tilsit cheese during the who le ripening period (plant A).Évolution de la flore bactérienne tolérant des teneurs de 80 g/kg et 150 g/kg NaCI comparée à la floretotale au cours de la maturation de Tilsit (fromagerie A).

20

F Eliskases-Lechner, W Ginzinger574

10

9

8'"Eo 7--:::>LLo 60>.2

5

4

3

0

1----1/----------------1---- total cou nt--<>- 80 g/kg NaCI

'------------------1-.--150 g/kg NaCI

20 30 40 6C10 50

ripening time (d)

Fig 2. Development of the 80 g/kg-NaCI- and 150 g/kg-NaCI-tolerant bacteria compared with the totalbacterial count of Tilsit cheese du ring the whole ripening period (plant B).Évolution de la flore bactérienne tolérant des teneurs de 80 g/kg et 150 g/kg NaCI comparée à la floretotale au cours de la maturation de Tilsit (fromagerie B).

ized as 80 g/kg-NaCI tolerant. The 150 g/kg-NaCI-tolerant bacteria counts were approx-imately 1 log unit lower than the total cou nt.

The development of the total bacterialcount of 3 different soft cheese varietiesis characterized by a maximum of> 108 ctu/cm'' (fig 3) after 10 d. In the fol-lowing ripening periods, a slight decreasein the counts of the smears of 2 varietiesoccurred, but considerable differencesbetween the 3 varieties were not found.

The percentages of orange- and yellow-pigmented colonies in the smears of Tilsitcheeses from 2 cheese plants during ripen-ing are shown in figure 4. The proportion ofpigmented colonies reached a maximum of43% of the total bacterial count in cheesesfrom plant A, and of 35% in those from plantB. Furthermore, a difference was foundbetween the smears from the 2 plants in therelation of orange- to yellow-pigmentedcolonies.

By means of the results just given, 3 dif-ferent times of ripening (3, 7 and 21 d aftermanufacture) were chosen to characterizethe smears of Tilsit cheeses from 14 cheeseplants. From each plant, the cultures used,the kind and the frequency of smearing wererecorded. Ali plants other than E and F useda Brevibacterium linens culture; plant L addi-tionally used a yeast culture. Moreover, in ailplants except E and F, the repeated smear-ing began with the older cheeses andmoved onto the younger on es so that theyounger cheeses were contaminated by thesmear of the older ones.

The variation in the bacterial counts washigh in the initial period of ripening, withcounts between 104 and 109 cfu/crn'' smearbeing found (table 1). After 3 weeks, thecounts in the smears of ail plants were> 108 cfu/cm2. In the smears of the 3- and 7-d-old cheeses, pigmented colonies occurredonly sporadically, and were found in levels

Bacterial flora of Tilsit cheese

9

8

7N

Eo--::> 6u,0Cl.Q 5

4

30

575

4C

f-------jr-----------------j-- variety 11--_---1

......variety 2____variety 3

15 20 25

ripening time (d)Fig 3. Development of the total bacterial count in the smears of 3 different soft cheese varieties dur-ing the whole ripening period (plant Cl.Évolution de la flore totale dans la morge de 3 différents types de fromages à pâte molle pendanttoute la période de raffinage (fromagerie G).

30 355 10

Table 1. Yeast counts and total bacterial counts in the smears of Tilsit cheeses fram 14 cheese plants,3, 7 and 21 d after manufacture, and the percentages of pigmented bacterial colonies after 21 d.Nombre de levures et flore totale bactérienne dans la morge de Tilsit de 14 fromageries, 3, 7 et 21 j aprèsla fabrication, et pourcentages des colonies pigmentées après 21 j.

Log yeast count/cm2 Log bacterial count/cm2 % Pigmentedcolonies

Plant 3d 7d 21d 3d 7d 21d (21 d)

A 3.5 6.5 7.7 6.3 6.8 9.3 24B < 2.0 < 2.0 4.8 4.5 7.0 9.1 13D 5.9 4.1 5.9 6.2 6.3 8.4 30E 6.3 5.6 5.4 8.0 8.9 9.0 0F > 6.5 6.8 6.6 9.5 9.3 9.5 10G 2.8 6.4 4.4 5.3 9.4 9.2 0H < 2.0 6.7 5.9 5.2 8.8 9.0 01 5.2 7.1 6.8 5.4 8.4 8.6 5J 4.1 7.1 5.8 7.0 8.8 9.7 0K 2.2 7.1 6.9 5.4 8.4 8.8 0L 5.5 6.7 6.3 5.9 8.1 9.2 20M 3.1 6.0 6.5 5.0 8.5 9.6 20N 5.3 6.4 2.6 7.1 8.6 9.0 00 5.3 7.0 5.8 5.9 8.1 9.0 5

F Eliskases-Lechner, W Ginzinger576

50

40

30~0

20

10

00 30 40 50

ripening time (d)Fig 4. Percentages of orange- and yellow-pigmented colonies of the total bacterial count in the smearsof Tilsit cheeses from 2 cheese plants during ripening.Pourcentages des colonies de couleur jaune et orange présentes dans la flore totale de la marge deTilsit provenant de 2 fromageries pendant J'affinage.

10 20

of between 0 and 30% of the total bacterialcount after 3 weeks of ripening.

1n the smears of the 3 different softcheese varieties, a change from yeast dom-inance to bacterial dominance occurredwithin the first week. This change was nottypical for the Tilsit cheeses: ail smear sam-pies 3, 7 and 21 d after manufacture showedlower yeast counts than bacterial counts.The percentages of yeasts were lower inthe smear samples of the older cheesesthan in the younger ones.

Identification results

Three hundred eighty-six isolates of coryne-form bacteria were isolated from the smearsof the previously mentioned cheese sam-pies and were identified. Table Il outlines

60 70 sc

the occurrence of the different species inthe different cheese plants. Thirteen differ-ent species were found. Besides Brevibac-terium linens, the other main types foundpresent in the heterogeneous flora werecream-coloured and yellow-pigmentedcoryneforms, which were identified asArthrobacter citreus, Arthrobacter globi-formis,Arthrobacternicotianae,Arthrobactervariabilis, Brevibacterium ammoniagenes,Brevibacterium fuscum, Brevibacteriumhelvolum, Brevibacterium imperiale, Bre-vibacterium oxydans, Corynebacteriumbetae, Corynebacterium insidiosum andCorynebacterium poinsettiae. In the bacte-rial flora in plants G and N, Brevibacteriumammoniagenes was dominant, while inplants A, Band L Brevibacterium ammoni-agenes and Brevibacterium linens dorni-nated. In the smears from plants C, D and l,a predominance of Arthrobacter globiformis

Bacterial flora of Tilsit cheese 577

Table Il. Subdivision according to species of coryneforms isolated from the surfaces of cheeses from15 plants.Fréquence des espèces corynéformes isolées de la morge des fromages fabriqués dans les 15fromageries.

Species Cheese plants

A B C 0 E F G H J K L M N 0

A citreus 12 3 2A globiformis 72 19 8 1A nicotianae 1 1 2 1 4Aspp 4 5 4 8 1 1 6 3 9A variabilis 1 4 3 2 2 2B ammoniagenes 8 6 11 4 1 2 7 12B fuscum 1 1B helvolum 1B imperiale 1 3 1 2B linens 20 5 8 13 5 5 13 5Boxydans 3Bspp 2 1C betae 4C insidiosum 3 1 1C poinsettiae 1 1 1 2 5Cspp 2 4 4 2 6 2 3

Total no of isolates 42 19 90 21 22 25 22 19 10 9 15 24 29 21 18

A: Arlhrobacter; B: Brevibacterium; C: Corynebacterium

was characteristic. Brevibacterium linens incombination with Afthrobacterspp was typ-ical for the smears from plants E, F and M.ln the remaining 4 plants, no single specieswas found to dominate the flora.

Biochemical propertiesof the bacterial flora

ln tables III and IV the biochemical proper-ties of the species are outlined as the per-centage of positive results. The isolatestested showed a wide range of combina-tions of characteristics.

Proteolytic activity

Changes in the frequency of proteolytic bac-teria du ring the ripening were found (tableV). The percentage of strains showingmedium and high levels of proteolytic activ-ity was higher at the beginning of ripeningthan after 21 d.

DISCUSSION

At the beginning of ripening, the bacterialcounts in the smears of the cheese surfacesvaried from 104-109 cfu/cmê. In the follow-

578 F Eliskases-Lechner, W Ginzinger

ing period, the counts increased to 108-109

ctu/cm'' and the differences between thesmears from the different plants were equal-ized. In analysing Limburger (EI-Erian,1969), Tilsit (Keller and Puhan, 1985),Weinkase (Sauter, 1986), Gruyère andBeaufort (Accolas et al, 1978; Piton-Malleretand Gorrieri, 1992), counts of this level wereregularly found on the surfaces of smear-ripened cheeses. Pigmented coryneformsshowed a similar development as the totalbacterial count but in lower numbers, occur-ring at levels of between a and 43% of thetotal counts. Grand et al (1992) found counts

of pigmented strains more than 1 log unitlower.

The bacterial flora can be characterizedby means of a marked resistance to NaCI.Thus, no difference between the total num-ber of bacteria and the 80 g/kg resistantbacteria was found. Only with 150 g/kg NaCIin the medium were the counts 1 log unitlower than the total number. The cheesesurface can therefore be described as anecosystem in which NaCI-tolerant bacteriapredominate. The salt tolerance of the sur-face flora of smear-ripened cheeses wasalso noted by EI-Erian (1969), Accolas et

Table III. Percentages of positive tests of the Arthrobacter spp mast commonly isolated from thecheese surfaces.Pourcentage d'essais positifs concernant Arthrobacter spp les plus fréquemment isolés sur la croûtedes fromages.

Character Arthrobacter Arthrobacter Arthrobactercitreus globiformis variabilis(n = 19) (n = 102) (n = 14)

100 99 100100 99 10094 91 83

0 3 081 89 8356 55 8388 100 75

100 86 7556 90 17

100 100 92100 99 8331 66 075 93 83

100 100 10094 100 10075 66 5831 46 025 79 0

100 98 8344 95 088 92 1781 99 5094 99 92

Assimilation of:AcetatePropionateValerateCapronateHeptanoateCaprylate4-Amiobutyrate4-AminovalerateMalonateSuccinateDL-malateAdipinateCitrateDL-lactateFumarateLevulinate4-HydroybenzoateGlyoxylateD-gluconate2-Keto-gluconateGlycineL-IeucineL-proline

Bacterial flora of Tilsit cheese

al (1978) and Grand et al (1992), who alsoequated the halotolerant flora with the totalbacterial count. lt is possible to select thebacterial flora by varying the salt concen-tration on the cheese surface. In practice,the salt content on the cheese surface canbe varied using different concentrations inthe smear water. At the beginning of ripen-ing this concentration should not be too highbecause of the yeasts. A NaCI concentrationof 50 g/kg in the smear water is ideal(Sauter, 1986). The NaCI concentration inthe smear water can be increased paraI/el to

Table III. (continued)

579

the increase in pH to 120 g/kg (Busse,1989).

No significant difference was foundbetween the bacterial status on Tilsit cheeseand that on the 3 soft cheese varieties, withthe exception of the relation between yeastsand coryneforms. Whereas the character-istic change from yeast dominance to bac-terial dominance was found in the smears ofthe 3 different soft cheese varieties, in thesmears of Tilsit cheeses from ail 14 plants,the bacteria were dominant al ail stages ofripening (Eliskases-Lechner and Ginzinger,

L-threonine 94 98 25L-aspargine 100 100 58L-aspartate 88 71 17

. L-arginine 94 99 8t-hlstldine 100 100 17D-xylose.,. 50 100 50D-ribose 81 97 33L-arabinose 13 94 8D-galactose 100 98 67L-rhamnose 67 96 8D-xylitol 91 8M-inositol 86 02,3-Butyleneglycol ~;-88 95 25Glycerol 81 100 25Acetamide 13 87 0

Acidfrom:D-xylose 38 16 0D-glucose 75 92 75Sucrose 63 61 8Lactose 50 59 0Starch 0 1 0Dextrin 0 16 0

Hydrolysis of:Xanthine 19 39 0Starch 19 46 9Gelatin 69 88 9Casein 19 72 0Catabolism of tyrosine 60 84 0Resistance to 70 g/kg NaCI 100 99 100Nitrate reduction 94 82 25Pigment with KOH 0 0 0

580 F Eliskases-Lechner, W Ginzinger

Table IV. Percentages of positive tests of the Corynebacterium spp and Brevibacterium spp mostcommonly isolated trom the cheese surfaces.Pourcentage d'essais positifs concernant Corynebacterium spp et Brevibacterium spp les plus fré-quemment isolés sur la croûte des fromages.

Character Corynebacterium Brevibacterium Brevibacteriumpoinsettiae ammoniagenes Iinens

(n = 12) (n = 53) (n = 77)

Assimilation of:Acetate 0 35 99Propionate 15 89 99Valerate 0 9 99Capronate 38 0 97Heptanoate 0 5 94Caprylate 0 4 764-Amiobutyrate 8 94 974-Aminovalerate 0 7 94Malonate 0 17 92Succinate 8 98 99DL-malate 0 83 95Adipinate 30 0 83Citrate 0 13 80DL-lactate 73 98 99Fumarate 0 98 j~Levulinate 0 44-Hydroybenzoate 0 0 L4Glyoxylate 0 2 .r- 74'"D-gluconate 92 (li 912-Keto-gluconate 31 38Glycine 0 53L-Ieucine 85 ~~ 80L-proline 23 62 73

1995). It can therefore be concluded thatthe knowledge concerning smear-ripenedsoft cheeses is not in every case applica-ble to semi-hard cheeses. Keller and Puhan(1985) also found that the yeast counts onTilsit cheese were never greater th an thebacterial counts. For soft cheese varieties,this change is of great importance, andshould occur at a certain stage of ripening toavoid the growth of undesired mou Ids(Sauter, 1986; Mayer, 1990).

ln 12 of the 14 cheese plants, Brevibac-terium linens cultures were used. Never-theless, Brevibacterium linens was not adominant species within the bacterial flora of

the cheese smear. It reached a maximumlevel of approximately 30% of the total floraon the 21 st d of ripening. The speciesArthrobacter globiformis and Brevibacteriumammoniagenes took a dominant position.Ali others of the 13 different species foundoccurred only sporadically, or were typicalfor the smear from a single plant. Althoughthe morphology of the colonies was the maincriterion for selection, the exact percent-ages in the smear are difficult to specify.Additionally, the change in composition ofthe bacterial flora during the ripening periodis also difficult to specify, although tenden-cies are distinguishable.

Bacterial flara of Tilsit cheese 581

L-threonine 0 48 86L-aspargine 0 96 100L-aspartate 0 59 73L-arginine 7 13 99L-histidine 7 89 96D-xylose 92 46 49D-ribose 54 83 78L-arabinose 54 28 8D-galactose 54 34 96L-rhamnose 0 9 61D-xylitol 31 5 10M-inositol 0 11 352,3-Butyleneglycol 0 5 77Glycerol 85 41 99Acetamide 38 13 43

Acid from:D-xylose 38 11 5D-glucose 85 89 68Sucrose 62 20 19Lactose 77 26 18Starch 0 0 0Dextrin 0 4 4

Hydrolysis of:Xanthine 15 6 66Starch 0 24 3Gelatin 64 22 97Casein 83 33 71Catabolism of tyrosine 15 41 64Resistance to 70 g/kg NaCI 100 91 100Nitrate reduction 38 93 55Pigment with KOH 8 0 98

Table V. Proteolytic activity of coryneforms isolated fram Tilsit cheese at 3 stages of ripening.Activité protéolytique des bactéries corynéformes isolées de fromage Tilsit à 3 stades d'affinage.

Time of isolation(days)

No ofisolates

Activity %

Law Medium High

3 87 33.3 41.4 25.3

7 87 42.5 34.5 23.0

21 18.089 55.0 27.0

582 F Eliskases-Lechner, W Ginzinger

The comparison with Iiterature data isdifficult, because often no identification wasmade, a differentiation by means of mor-phological characteristics being carried outinstead. For example, Mulder et al (1966)and EI-Erian (1969) found 65-80% grey-white strains on the surface of Limburgercheese, whereas isolates of the type Bre-vibacterium Iinens made up between 9 and24%. Sauter (1986) found, on the surfaceof Weinkase, 42% white to cream-colouredcoryneforms, 36% yellow-pigmented coryne-forms and only 22% showing the typical pig-mentation of Brevibacterium linens. Seiler(1986) and Mayer (1990) defined the iso-lated strains as Arthrobacter spp. On thesurfaces of different cheese varieties, egRomadur, Limburger cheese, Weinkase,Harzer Handkâse and Camembert,Arthrobacter nicotianae, Arthrobacter vari-abilis and Brevibacterium ammoniageneswere found. The species Arthrobacter pro-tophormiae and Arthrobacter uratoxydansoccurred in the smear of Vacherin Mont-d 'Or (Anonymous, 1988; Hug-Michel et al,1989). Sauter (1986) identified the coryne-forms of Weinkase as Arthrobacteraurescens, Arthrobacter citreus, Arthrobac-ter globiformis, Brevibacterium Iinens, Bre-vibacterium helvolum and Brevibacteriumoxydans.

The results concerning the proteolyticactivity of the coryneforms correspond tothe descriptions given by Mulder et al (1966)and EI-Erian (1969), who classified themajority of the grey-white coryneforms ofthe cheese surface as nonproteolytic, andfound the flora at earlier stages of ripeningmore proteolytic than at later stages.

ln conclusion, it is remarkable that Bre-vibacterium linens is never the dominantbacterium despite its regular use as a cul-ture. It is to be assumed that the majorityof those coryneforms, which were not addedas culture organisms, are just as importantas Brevibacterium linens for the ripeningprocess, eg by their proteolytic activity, for

the production of aroma and the appearenceof the cheese surface.

REFERENCES

Accolas JP, Meldon D, Vassal L (1978) Étude de laflore superficielle des fromages de Gruyère et deBeaufort. 20e Congr Int Lait, Paris, France, 773-774

Anonymous (1988) Untersuchung über die Dynamik derObertlâchenflora beim Vacherin Mont-d'Or.Tatigkeitsbericht FAM 205-207

Busse M (1989) Die Obertlâchentlora von geschmiertemKase. Mi/chwirtschBer99, 137-141

EI-Erian AFM (1969) Bacteriological studies on Lim-burger cheese. Thesis, Agricultural University,Wageningen, Netherlands

Eliskases-Lechner F, Ginzinger W (1995) The yeastflora of surface-ripened cheeses. Milchwissenschaft50,458-462

Engel G (1993) Hemmung von Hefe- und Schim-melpilzwachstum beim quantitativen Nachweis vonBakterien. Milchwissenschaft48, 325-327

Grand M, Weber A, Perret J, Zehntner U, Glatlli H (1992)Evolution de la flore d'affinage sur des fromages deGruyère. Schweiz Mi/chwirtsch Forschung 21, 3-5

Greez N, Dack GM (1961) Taxonomically significantcolor reactions of Brevibacterium linens. J Bacteriol82,241-246

Hug-Michel C, Barben S, Kaufmann U (1989) Hemmungvon Listeria spp durch Mikroorganismen aus derRinde von Hotschmiereweichkàse. Schweiz Mi/ch-wirtsch Forschung 18, 46-49

IDF (1990) Milk and milk products enumeration of yeastsand moulds colony count technique at 25°C. Stan-dard 94 B

Keller A, Puhan Z (1985) Entsauerung der Kâse mitSchmiere am Beispiel des Tilsiters. Schweiz Mi/ch-wirtsch Forschung 14,3-11

Mayer W (1990) Einfluf3faktoren auf die Oberflâchen-flora von Rotschmierekâsen - Versuchsergebnisse.IX. Kâsereitechnologischer Sonderlehrgang, StaatlLehr- und Versuchsanstalt Kempten/AlIgâu

Mulder EG, Adamse AD, Antheunisse J, Deinema MH,Woldendorp JW, Zevenhuizen LPTM (1966) Therelationship between Brevibacterium Iinens and bac-teria of the genus Arthrobacter. J Appl Bacteriol29,44-71

Piton C, Fontanier C (1990) Caractérisation d'une col-lection de souches de bactéries corynéformes de lamorge du gruyère de Comté. Lail70, 383-398

Piton-Malleret C, Gorrieri M (1992) Nature et variabilitéde la flore microbienne dans la morge des fromagesde Comté et de Beaufort. Lait 72, 143-164

Bacterial flora of Tilsit cheese

Rapp M (1974) Indikatorzusâtze zur Keimdifferenzierungauf Würze- und Malzextrakt-Agar. Milchwissenschaft29,341-344

Sauter H (1986) Die Oberflâchenflora von Weinkâse.Reifungsfehler und ihre Ursachen. Thesis, TechnicalUniversity, Munich, Germany

Seiler H (1983) Identification key for coryneform bacte-ria derived by numerical taxonomie studies. J GenMicrobio/129,1433-1471

Seiler H (1986) Identification of cheese-smear coryne-form bacteria. J Dairy Res 53, 439-449

583

Seiler H, Braatz R (1988) Computerprogramme MCOM·PRE und CLUSUM zur Coryneformenidentifizierung.Bakteriologisches Institut, südd Versuchs- undForschungsanstalt für Milchwirtschaft Weihenstephan

Seiler H, Braatz R, Ohmayer G (1980) Numerical elus-ter analysis of the coryneform bacteria from acti-vated sludge. Zb/ Bakterio/ Mikrobio/ Hyg Ab! / Org C1,357-375

Valdés-Stauber N, Seiler H (1990) Identifizierung vonCoryneformen mit Mikrotiterplatten. Bakteriologis-ches Institut, südd Versuchs- und Forschungsanstaltfür Milchwirtschaft Weihenstephan