Analysis of Yeast Flora Associated with Grape Sour Rot and of theChemical Disease Markers
ELISABETTA GUERZONI* AND ROSA MARCHETTI
Dipartimento di Protezione e Valorizzazione Agroalimentare, Sez. Microbiologica, Universita di Bologna,40126 Bologna, Italy
Received 6 August 1986/Accepted 9 December 1986
The frequency and the density of the species associated with grape sour rot in different cultivars were
determined. The most frequent species in the rotten grapes, Candida krusei, Kloeckera apiculata, andMetschnikowia pulcheryima, and a less frequent species, Issatchenkia occidentalis, when inoculated withSaccharomycopsis crataegensis were able to induce in vitro the symptoms of the disease. The gas chromato-graphic determination of the volatile compounds in the headspace was used to evaluate the metabolic role of thedifferent species associated with the disease. These analyses made it possible to presume that, whereas some
species, such as Candida krusei and Hanseniaspora uvarum, can be considered responsible for thesemodifications and in particular for the ethyl acetate production, others, such as Saccharomycopsis crataegensis,can promote the development of the former species.
Vine sour rot is a disease characterized by a typical andeasily recognized phenomenology-browning and disag-gregation of the internal tissues, detachment of the rottenberry from the pedicel, and a strong ethyl acetate smell (3,7). However, its etiology is ambiguous and controversialeven if all the authors (2, 3, 7, 8, 13) agree in associatingyeasts with the disease, at least when it manifests itselfsuddenly. The yeast species most frequently reported asactively proliferating in rotten berries are Hanseniasporauvarum, Torulopsis stellata (2, 3), Metshnikowia pulcher-rima, Candida spp., and Saccharomycopsis vini (syn. Endo-mycopsella Boedijn) (7, 13). The presence of acetic bacteriaor of the most common pathogenic fungi has only rarely beenobserved. The relationship of the disease to Botrytis cinereaor other grape diseases is controversial (3).Another component of the system is the Drosophila ge-
nus, individuals of which were always observed (3) on therotten grapes, even if the role of this insect has not beenstudied in depth. Sour rot can be considered a multifactorialdisease: its sudden appearance probably necessitates thecoming together of favorable combinations of particularphysiological conditions determined by the climate and thecultural practices. Moreover, it would seem that the cultivarcharacteristics of the host plants are of considerable impor-tance in disease incidence: experiments of in vivo diseaseinduction provoked a significant increase of the diseaseincidence only in certain cultivars which were generally themost susceptible ones (11).The etiological aspects of the phenomenon are intriguing
for a number of reasons. Apart from the fact that the yeasts,with few exceptions, are not capable of attacking planttissues and causing disease, the greatest increase in inci-dence was obtained by incubating blooming bunches withspecies such as Candida krusei, Candida steatolytica,Hanseniaspora uvarum, Saccharomycopsis vini, andMetschnikowia pulcherrima, which are not necessarily thedominant species when the disease appears (11).
* Corresponding author.
It is held that a comparative analysis of the data collectedover 4 years on the frequency of different yeast species inhealthy and rotten grapes from different localities andcultivars could be helpful in determining whether the dif-ferent groups have a specific or nonspecific role in thegenesis and manifestation of the disease. In addition, as themanifestation of sour rot is always accompanied by changesin the composition of the grape tissues, an evaluation of therelative contribution of the different species to the transfor-mation of such a complex ecosystem was attempted.
MATERIALS AND METHODSIsolation and counting. Estimates of the kinds and number
of yeasts present were obtained by suspending one(weighed) rotten or healthy grape in sterile water, sonifyingit by the method of Martini et al. (12), and plating variousdilutions on Sabouraud agar (Difco). After incubation atroom temperature, the distinct colony types were counted.A representative of each morphological colony type fromeach plant sampled was restreaked once or twice, and theresulting culture was subjected to the identification proce-dure. The yeast isolates were identified by the method ofLodder (10) and sent for confirmation to the CentraalbureauVoor Schimmelcultures (CBS). The taxonomic designationof the isolates was updated in accordance with Kreger-VanRij (8).
Lytic activity. The proteolytic activity of the isolates wasdetermined on casein by the method of Ahearn et al. (1),lipolytic activity was determined on synthetic solid mediumwith Bacto-lipase reagent (Difco) added, and the pectolyticactivity was evaluated by the method of Winborne andRichard (15).Frequency and enzymatic activity data processing. Compar-
isons between yeast communities were made on the stan-dardized proportion (number of isolates per number of plantssampled) of species by the methods of Starmer (13) andLachance and Starmer (9). Comparisons were made on thebasis of the presence or absence of a particular species ofyeast in a sample. A simple measure of similarity betweencommunities of different habitats was expressed as the
TABLE 1. Frequency and density of different yeast species in healthy and rotten grapes belonging to different cultivarsaSangiovese (Forli, Tebano, 1981, 1982) Rossiola (Volania, 1985)
a For each cultivar, the locality and year are given in parentheses. Abbreviations: H, healthy grapes; R, rotten grapes; F, frequency (no. of isolates per plant);D, density (log1o yeasts, average of four samples).bNo. of plants tested.
average of the absolute difference in the proportion of eachspecies. This was calculated as the following distancemeasure:
1nDim I IPIr-Pmjl
n j=1
The distance DIm represents the distance between commu-nities of habitats I and m; the proportion Plj represents theproportion of yeast species (I) found in habitat 1. The value ofthe distance can be any value between 0 (close or identicalcommunities) and 1 (distant or least similar).As far as the lytic activities are concerned, the distance in
the lytic activity of a yeast community from two habitats, 1and m, was calculated as:
1 n8lm = - qlj qml
n j=1The proportion qlj represents the proportion of isolates inhabitat I which show a particular enzyme activity, where Iand m are the communities of healthy and rotten grapes,respectively.GLC analyses of HS vapors. The isolates and the collection
(CBS) strains were inoculated in 100 ml of must in cylindricalcontainers with an empty/full ratio of 0.17. After 5 days ofincubation at 28°C, 5 ml of filtrate was collected, and 5 g ofdry Na2SO4 was then added. After conditioning at 60°C for10 min, 500 ,ul of headspace (HS) vapors was injected.For the grape samples, 500 pLI of HS vapors in equilibrium
in sealed plastic bags was injected 6 h after harvesting.Artificial inoculations of detached berries were done by
means of microinjuries on table grapes after sterilizationwith sodium hypochlorite. The grapes were then incubatedin sealed vials at 28°C. The gas-liquid chromatography(GLC) analyses, injecting 500 ,ul of HS vapors in equilibriumin the vials, were performed after 7 days. Analytical condi-tions: Carlo Erba 4160 HRGC GLC unit with glass capillarycolumn (40 m by 0.3 mm) coated with 0.1- to 0.15-p.m thick
FFAP film; chamber temperature, 110°C; detector (FID)temperature, 210°C; the spectrum area was measured by aSpectra Physics (model 4100B) integrator.
RESULTS
Frequency and density of different species in diseased andhealthy grapes. The relative complexity and variability of theyeast population occurring on rotten grapes have made itdifficult to establish the existence of a close relationshipbetween species and disease manifestation. To evaluatewhether the yeast community associated with rotten grapespresents a homogeneous structure, even as the habitatcharacteristics microenvironment and characteristics ofthe host plant-vary, various cultivars of table and winegrapes from different localities and different years wereconsidered. The cultivars, in fact, were seen to have animportant role in disease incidence, independent of thegeographic zone.Some cultivars such as Rossiola, Uva d'oro, Perlette, and
Delight presented an elevated incidence of the disease (up to80% of attacked plants observed in 1981, 1982, 1984, and1985); others, such as Fiesta and Sultana moscato, notconsidered in this study are only rarely attacked. In Table 1the mean density and frequency of each species in rotten andhealthy grapes are compared.The samples of healthy and diseased grapes of the same
cultivar collected from the same field and, where possible,from the same plant were analyzed on the same date. Thefrequency of the different species was calculated from thenumber of samples reported in the table. The density of cellsof the same species (log1o of the number of viable cells pergram) was determined for four plants for the rotten grapesand all the plants indicated in the table for the healthy ones.The species having greater density and frequency on thediseased grapes were Candida krusei, Hanseniasporauvarum, Kloeckera apiculata, Saccharomycopsis vini and S.
APPL. ENVIRON. MICROBIOL.572 GUERZONI AND MARCHETTI
crataegensis, and Candida steatolytica. These are, with the possible role played by different injuries, especially thoseexception of Saccharomycopsis spp. and Candida krusei, deriving from a previous attack ofBotrytis cinerea and thoseconsidered normal resident species of the vineyard (5, 6). caused by insects. The lipolytic, proteolytic, and pectolyticZygosaccharomyces spp. presented a high density, but were activities of the strains isolated on the rotten and healthypresent in only one cultivar. grapes were determined.The most frequent and abundant species on the healthy The distance between the two communities of the rotten
grapes were Aureobasidium pullulans and Rhodotorula spp., and healthy grapes, determined as indicated in Materials andwhich were always absent in the rotten grapes. Methods, was 0.41, 0.22, and 0.32 for lipolytic, proteolytic,The distance (D) between communities of the healthy and and pectolytic activities, respectively. It is evident that the
rotten grapes, calculated on the basis of the frequency of the lipolytic yeasts were more frequent in the rotten samplesvarious species in the two groups of samples (last column in than in the healthy ones (Table 2). This activity was in factTable 1), was 0.17. particularly linked to S. vini and S. crataegensis; the per-The distances between the 10 pairs of cultivars were centage of proteolytic and pectolytic strains was higher
calculated separately for healthy and rotten grapes. The among the healthy fruits. These activities were particularlycommunity actively growing on the rotten samples was only linked to A. pullulans, frequent in the healthy grapes.slightly more homogeneous than that associated with the GLC determinations. An interpretation based exclusivelyhealthy ones; in fact, the average distance was 0.40 and 0.34 on the frequency of the different species does not take intofor healthy and diseased grapes, respectively. account the lack of homogeneity in terms of cell density ofThe examination of species frequency has shown certain the two series of samples: the healthy grapes were colonized
differences between diseased and healthy samples, but the by epiphytic yeasts, whose density was on the order of 104mere presence of the species does not provide an explana- cells per g; an actively growing population of about 107 cellstion of their role. per g was present in the rotten tissues. In the latter samples
It was thought important to consider the possible direct it was difficult to evaluate on the basis of pure presence thecontribution of the yeasts to penetration of the berry tissues role of the single species. Occurrence, even of a singleby a special enzymatic system. This does not exclude the species alone, does not necessarily imply that this species is
TABLE 2. Distribution of lytic activities in isolates in different habitats
Lytie Proportion of isolates showing activity Distance betweenactivity Grapesa H and R
TABLE 3. Total area and average relative percentage of the individual volatile compounds in equilibrium, in standard conditions, in theHS with naturally infected grape samples belonging to different cultivarsa
CultivarNo. of Ethyl formate Acetalde- Ethyl acetate Ethanol Isobutanol Total area
Cultivar No. of hydesamples x SD x SD x SD x SD x SD x SD
a The percent relative standard deviation was calculated on the basis of the absolute area values; when a value is not reported, the number of repetitions was in-sufficient for its determination.
the primary agent of disease diffusion or the one responsiblefor the chemical modifications associated with it.
In fact one or more organisms could have been able,because of favorable opportunities, to drastically modify theinternal tissues of the grapes so that other organisms couldno longer grow and favored organisms could become domi-nant.Thus, the evaluation of the suitability of some volatile
metabolites associated with rotten fruits as markers of themetabolic contribution to the disease manifestation relativeto the different species was attempted.
Thirty-two samples of rotten grapes from different locali-ties and belonging to all the susceptible cultivars were
considered. For each variety samples having the same
infection level (30 to 40% rotten grapes) were analyzed.Preliminary analyses have shown that in general the totalarea of volatile compounds determined in standard condi-tions (temperature, bunch volume/empty volume) in the HSincreases as the percentage of rotten grapes increases,though not in the same proportion.
In Table 3 the data on total area and relative percentage ofthe single compounds in equilibrium in the HS are reported.The information coming from the comparison of the data canbe summarized as follows.
(i) The volatile compounds present in the HS are neitherphysiological products of the fruit nor metabolites resultingfrom development of Botrytis cinerea. (ii) All the samplesshowing the sour rot symptoms presented qualitatively iden-tical GLC profiles; even if the variability for every peak washigh, the compounds, alcohols or esters, detectable inthese conditions are all common primary metabolites of theyeasts.
In some cultivars, such as Rossiola and Delight, themanifestation of the disease was associated with an amountof volatile compounds (total area) in the HS which onaverage was, at the same infection degree, three times that ofsamples of Fiesta, Montuni, and Perlette. Generally speak-ing, however, the GLC pattern can be considered qualita-tively homogeneous and independent of the cultivar.Thus, it might be ascertained whether the yeast species
isolated were able to produce a GLC pattern quantitativelyand qualitatively compatible with that of infected grapes.Therefore the level of the individual volatile compound andthe total area of the volatile compounds in equilibrium in theHS of standard cultures of 46 species frequently or sporad-ically isolated on rotten grapes were determined. The totalareas, corresponding to all the volatile compounds in equi-librium in the HS of liquid medium, are not quantitatively
TABLE 4. Relative percentages of the individual volatile compounds' and their total area in equilibrium in the HS with artificiallyinfected grapes% of total Relative area in HS (%)
comparable to those of natural samples. However, it waspossible to select five species able to produce a GLC patternqualitatively similar to that of natural samples.
In particular, the yeast species showing a ratio betweenthe two most important metabolites, ethanol and ethylacetate, closest to that of infected samples (ranging from1.42 to 13.7) were H. uvarum, K. apiculata, M. pulcherrima,Pichia membranaefaciens, and C. krusei. Hansenulaanomala presented a content of ethyl acetate that was toohigh, 30 to 40% of the total volatile compounds, and wasonly sporadically isolated in the field. Isolates belonging tothese species, both alone and in pairs, were thus inoculatedinto sterilized grapes contained in sealed vials. Other strainsbelonging to the species Issatchenkia occidentalis and S.crataegensis, frequent in the rotten grapes, were inoculatedas controls.
K. apiculata AV33 and C. krusei 12181 alone and somecombinations-M. pulcherrima 721B plus S. crataegensisAR 43 and Issatchenkia sp. strain AR 41 plus S.crataegensis-developed inside the grapes after 4 days,giving rise in only a few cases to the ethyl acetate odorcharacteristic of the disease, even though the symptomswere not exactly the same as in vivo.The results of the GLC analyses of the volatiles in
equilibrium in the HSs in the vials containing the infectedberries are reported in Table 4. This table also shows theareas of the single peaks produced from Issatchenkia oc-cidentalis, M. pulcherrima, and S. crataegensis inoculatedalone; they grew in the grapes, as indicated by the volatiletotal area which ranged from 2,000 to 9,000, but did not giverise to the symptoms of sour rot. It is interesting that S.crataegensis plus M. pulcherrima and S. crataegensis plus I.occidentalis inoculated together reproduced the phenomenaand chemical symptoms of the disease in 5.6 and 6.8% ofcases, respectively.M. pulcherrima and K. apiculata, when inoculated to-
gether, showed the typical peak of the disease but with avery low total volatile area.These results seem to substantiate the hypothesis of the
different roles of the various species and in particular theroles of a species such as S. crataegensis, which does notproduce the typical GLC pattern but increases the probabil-ity of success of the others except when inoculated with K.apiculata. A comparison appears in Table 5 of the GLCpattern of volatile compounds in equilibrium with standardcultures of the isolates from infected samples identified by usand by CBS and of collection (CBS) strains of the samespecies. Even taking into account individual variabilitywithin the different strains, it can be observed that the sourrot isolates always presented a higher percentage of ethylacetate than the collection strains belonging to the samespecies.
DISCUSSIONThe results obtained have made it possible to make certain
conclusions. (i) The population actively growing on diseasedgrapes differs from that which colonizes the surface ofhealthy grapes in density and species frequency. (ii) Isolates,but not collection strains, belonging to the species C. krusei,H. uvarum, M. pulcherrima, and P. membranaefaciens areable to produce, in liquid medium and when inoculated insterilized grapes, the same volatile metabolites, especiallyethyl acetate, in relative ratios similar to those of naturalsamples.
S. crataegensis, which is incapable of producing thetypical GLC pattern when inoculated in liquid medium,
would seem to play an important role in determining thesuccess of the infection, on the basis also of previousexperiments with field inoculation at blooming (11). Isolatesbelonging to this species present, in addition, lipolytic activ-ity. It may therefore be hypothesized that several specieswith different roles act together. C. krusei, H. uvarum, M.pulcherrima, and P. membranaefaciens can be indicated aspossibly being responsible for the biosynthesis of diseasechemical markers and probably for the grape tissue disag-gregation due to internal CO2 produced, as indicated by theelevated alcohol content and pressure. S. crataegensis inturn could play a role in the creation of conditions favorableto the selective development of the others. It is not to beruled out that in different habitats other species may fulfillthis role, which could also be attributed to Botrytis cinereaor mechanical injuries.
Ethyl acetate can be considered a disease indicator, asmay its precursor acetic acid, ranging in the grapes from 0.3to 12.5 mg/ml (7). Nevertheless the mechanisms whichdetermine such an elevated production are not altogetherclear. In fact the isolates belonging to the species C. krusei,H. uvarum, and P. membranaefaciens were seen to producethis metabolite in lower quantities in liquid medium thanwhen inoculated into detached and sterilized grapes. Inaddition, it was not possible to create a combination ofchemicophysical conditions which were able to induce in thestrains of these species a production of acetic acid as high asthat found in the field samples (11). Moreover, diseaseisolates belonging to the species C. krusei, P. membranaefa-ciens, and M. pulcherrima, which are not considered ethylacetate producers, give rise, in the same conditions, to amuch greater quantity of this ester than do the collectionstrains of the same species. This behavior, which is stableafter 2 years of isolation, may be attributed to the geneticvariability within the species. Furthermore, as regards theinduction of such a high synthesis of acetic acid and ethylacetate in vivo in these yeasts, it may be hypothesized thatthe anaerobic conditions inside the grapes-or in any casethe redox potential or its variations-may determine situa-tions similar to those that induce Dekkera and Bret-tanomyces strains to increase the production of acetic acidand ethyl acetate (4, 14).Even if the general conditions which facilitate the appear-
ance of the disease in natural conditions have been deter-mined, it is not yet clear what triggers the explosive yeastdevelopment, nor is it clear why the disease has onlyrecently been reported. Moreover, it may not be ignored thatthe vineyard environments have been treated for many yearswith pesticides belonging to different fungicide families andhaving a wide spectrum. A study of the action in vitro on 80strains belonging to different species of several antifungalproducts used over the last 30 years has indicated that thespecies S. vini and S. crataegenesis, C. krusei, and M.pulcherrima are inhibited by the antifungal agents that wereused up until the 1970s, but they are not inhibited by the
latest generation of products (M. E. Guerzoni, R. Marchetti,and M. Gentile, Giornate Fitopatologiche Proc. 2:145, 1984).Furthermore, the efficient chemical control of Botrytiscinerea exercised in recent years may have eliminated thisstrong competitor of the yeasts that is insensitive to thenewer and widely used compounds, the cyclic dicarboxy-mides, and could have modified their role in the vineyardcomplex microflora.
LITERATURE CITED1. Ahearn, D., S. P. Meyers, and R. H. Nichols. 1968. Extracellular
proteinases of yeast and yeast-like fungi. Appl. Microbiol.16:1370-1374.
2. Bisiach, G., G. Minervini, and F. Zerbetto. 1986. Possibleintegrated control of grapevine sour rot. Vitis 25:118-128.
3. Bisiach, M., G. Minervini, and M. C. Salomone. 1982.Recherches experimentales sur la pourriture acide de la grappeet sur ses rapports avec la pourriture grise. Bull. OEPP 12:5-28.
4. Carrascosa, J. M., M. D. Vignera, N. Castro, and W. A.Scheffers. 1981. Metabolism of acetaldehyde and Custers effectin the yeast Brettanomyces abstinens. Antonie van Leeuwen-hoek J. Microbiol. Serol. 47:203-215.
5. Davenport, R. R. 1974. Microecology of yeasts and yeast-likeorganisms associated with an English vineyard. Vitis 13:123-130.
6. Davenport, R. R. 1976. Distribution of yeasts and yeast-likeorganisms from aerial surfaces of developing apples and grapes,p. 325-359. In C. H. Dickinson and T. F. Preece (ed.), Micro-biology of aerial plant surfaces. Academic Press, Inc., NewYork.
7. Guerzoni, M. E., and R. Marchetti. 1982. Microflora associata almarciume acido della vite e modificazioni indotte dalla malattiasulla composizione di uve e mosti. Difesa Piante 4:231-246.
8. Kreger-van Rij, N. J. W. 1984. The yeasts: a taxonomic study,3rd ed. Elsevier Science Publishers, New York.
9. Lachance, M. A., and W. T. Starmer. 1982. Evolutionarysignificance of physiological relationships among yeast commu-nities associated with trees. Can. J. Bot. 60:285-293.
10. Lodder, J. 1970. General classification of the yeasts, p. 35-107.In J. Lodder (ed.), The yeasts: a taxonomic study. North-Holland Publishing Co., Amsterdam.
11. Marchetti, R., M. E. Guerzoni, and M. Gentile. 1984. Recherchesu l'etiologie d'une nouvelle maladie de la grappe: la pourritureacide. Vitis 23:55-65.
12. Martini, A., F. Federici, and G. Rosini. 1980. A new approach tothe study of yeast ecology of natural substrates. Can. J. Micro-biol. 26:854-860.
13. Starmer, W. T. 1980. The evolutionary ecology of yeasts foundin the decaying stems of cacti, p. 493-498. In G. G. Stewart andI. Russell (ed.), Current developments in yeast research.Pergamon Press, Toronto.
14. Wijsman, M. R., J. P. van Diken, B. H. A. van Kleefe, andW. A. Scheffers. 1984. Inhibition of fermentation and growth inbatch cultures of the yeast Brettanomyces intermedius upon ashift from aerobic to anaerobic conditions. Antonie vanLeeuwenhoek J. Microbiol. Serol. 50:183-192.
15. Winborne, M. P., and P. A. D. Richard. 1978. Pectinolyticactivity of Saccharomyces fragilis cultured in controlled envi-ronments. Biotechnol. Bioeng. 20:231-242.
![Detectionof benzo[a]pyrene DNAadductsin · 2005. 4. 22. · Benzo[a]pyrene(B[a]P) is aubiquitous chemicalcarcinogen foundin tobaccosmoke,atmosphericpollution duetoburn-ing offossil](https://static.documents.pub/doc/80x56/61266effb3a25756793990f8/detectionof-benzoapyrene-dnaadductsin-2005-4-22-benzoapyrenebap-is.jpg)
![Topological superconductivity in metallic nanowires ...digital.csic.es/bitstream/10261/95684/1/Topological... · SinceKitaev’s proposal[1]in 2001that Majoranafermions couldbe foundin](https://static.documents.pub/doc/80x56/5edde9a0ad6a402d66692579/topological-superconductivity-in-metallic-nanowires-sincekitaevas-proposal1in.jpg)
