http://orcid.org/0000-0002-2333-6127
Figures
Abstract
The grape yeast biota from several wine-producing areas, with distinct soil types and grapevine training systems, was assessed on five islands of Azores Archipelago, and differences in yeast communities composition associated with the geographic origin of the grapes were explored. Fifty-seven grape samples belonging to the Vitis vinifera grapevine cultivars Verdelho dos Açores (Verdelho), Arinto da Terceira (Arinto) and Terrantez do Pico (Terrantez) were collected in two consecutive years and 40 spontaneous fermentations were achieved. A total of 1710 yeast isolates were obtained from freshly crushed grapes and 1200 from final stage of fermentations. Twenty-eight species were identified, Hanseniaspura uvarum, Pichia terricola and Metschnikowia pulcherrima being the three most representative species isolated. Candida carpophila was encountered for the first time as an inhabitant of grape or wine-associated environments. In both sampling years, a higher proportion of H. uvarum in fresh grapes from Verdelho cultivar was observed, in comparison with Arinto cultivar. Qualitatively significant differences were found among yeast communities from several locations on five islands of the Archipelago, particularly in locations with distinctive agro-ecological compositions. Our results are in agreement with the statement that grape-associated microbial biogeography is non-randomly associated with interactions of climate, soil, cultivar, and vine training systems in vineyard ecosystems. Our observations strongly support a possible linkage between grape yeast and wine typicality, reinforcing the statement that different viticultural terroirs harbor distinctive yeast biota, in particular in vineyards with very distinctive environmental conditions.
Citation: Drumonde-Neves J, Franco-Duarte R, Lima T, Schuller D, Pais C (2017) Association between Grape Yeast Communities and the Vineyard Ecosystems. PLoS ONE 12(1): e0169883. https://doi.org/10.1371/journal.pone.0169883
Editor: Luca Cocolin, University of Torino, ITALY
Received: September 8, 2016; Accepted: December 24, 2016; Published: January 13, 2017
Copyright: © 2017 Drumonde-Neves et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: João Drumonde Neves is the recipient of a fellowship of the Azorean Government (M321/006/F/2008) and PROEMPREGO. This work was supported by the strategic programme UID/BIA/04050/2013 (POCI-01-0145-FEDER-007569) funded by national funds through the FCT I.P. and by the ERDF through the COMPETE2020 - Programa Operacional Competitividade e Internacionalização (POCI), and by national funds through FCT by the projects FCOMP-01-0124-008775, PTDC/AGR-ALI/103392/2008 and PTDC/AGR-ALI/121062/2010.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Traditionally, wines are produced by spontaneous fermentation carried out by the yeast biota naturally present in musts, having its origin on the grapes and/or winery equipment, and the process involves the sequential development of different yeast species. Strains of Saccharomyces cerevisiae, especially adapted, play the major role but the initial fermentation stages are usually carried out by non-Saccharomyces species [1]. The species Hanseniaspora uvarum (anamorph Kloeckera apiculata) is widely reported as predominant in initial stages of spontaneous fermentations, together with Candida spp. and Pichia spp. [2–9]. It is well known that the yeasts species/strains present during must fermentation affect wine’s flavor and aroma [10–14], because they act differently on musts, yielding different metabolites in different amounts [15–19]. Yeast biodiversity in vineyards are mainly affected by the grapevine cultivar [20–22], viticultural and oenological practices [23–29], macro and microclimatic conditions [1,30–32] and the geographic location of the vineyard [20,33–35]. It has been generally thought that different regions and grapevine cultivars, under different farm management practices–different viticultural terroirs–may harbor distinctive yeast communities and populations. Preserving the terroir characteristic of each wine has been one of the main concerns of the winemaking industry [21]. This led to an increased focus on the selection of autochthonous yeast, that might be better adapted to the fermentation of a particular grape must and might contribute to the typical oenological characteristics of a particular region [36–41]. Despite the suggestion that grape heath status is the main factor affecting the microbial ecology of grapes. [42] the idea that microbiological resources might be influenced by terroir aspects has long been suggested by several authors [21,22,43–45], and was recently demonstrated by using advanced short-amplicon sequencing approach by Bokulich et al. [32]. Moreover, the evidence for regional dispersion of vineyard-associated yeasts was previously described for cultivable yeast communities in New Zealand vineyards [46].
On the islands of the Azores Archipelago unique viticultural environments occur in several locations, resulting from the interaction between very particular macro and microclimatic conditions, autochthonous grapevine cultivars and local viticultural practices. Two important wine-producing regions in Azores Archipelago are Lajidos (PLJ) in Pico Island (classified by UNESCO as world heritage (http://whc.unesco.org/en/list/1117)) and Biscoitos (BCT), in Terceira Island, both corresponding to viticultural areas that are very distinct from the remaining locations in the archipelago.
Besides the vineyards installed in arable lands, grapevines were traditionally planted in poor soils (shallow or stony). Plants are placed in the irregularly distributed cracks of almost unmodified solidified lava flows (lajido) or in soils covered by thick layers of stones (biscoito). These particular types of soil ensure unique microclimates at the grape berry level, characterized by lower humidity and higher temperatures at maturation time. The grapevine training systems are mainly dependent on the soil type of the vineyard. Both in biscoito and in lajido, grapevines are not trellised, and the training system is unique for each plant depending on the area they have available. In the vineyards planted in arable soil, grapevines can be trellised or not.
The existence of different viticultural conditions in island environment, with well-defined borders, represents a suitable model to address the question of yeast biota / terroir association. Therefore, the aim of this study was to characterize the yeast biota from the Azorean traditional grape cultivars growing in several wine-producing areas from the Archipelago, and search for differences in grape yeast communities associated with those agro-ecological zones.
Material and Methods
Sampling and yeast isolation
The sampling areas were selected based on the combination of three agro-ecological criteria: i) type of soil (arable soil—arable; soil covered with stones—stony; solidified lava flows—lava); ii) grapevine cultivars (Verdelho: Verdelho dos Açores; Arinto: Arinto da Terceira; Terrantez: Terrantez do Pico); and iii) the grapevine training system (trellised grapevines, non-trellised grapevines; lying grapevines) (Fig 1).
A: not- trellised grapevine (NT) in soil covered with stones (SS); B: NT in solidified lava flows (SL); C: NT in arable soil (AS); D: trellised grapevines (TG) in AS.
Grape samples belonging to white Vitis vinifera cultivars Verdelho, Arinto and Terrantez were collected from 22 vineyards on 13 locations on five islands of the Azores Archipelago during the 2009 and 2010 harvests (30 and 27 grape samples, respectively) (Fig 2), always with the permission of the land owners. Locations were chosen within the existing vineyards, being the number of locations per island independent of its area. Each sample consisted of approximately 2–3 kg of rot-free grape bunches that were collected aseptically into sterile plastic bags and immediately transported to the laboratory under refrigerated conditions. For each location grape bunches were harvested in four different sampling points, separated by an average distance of ca. 10 m, in order to obtain a high diversity inside each harvest location. The berries were manually crushed inside the sterile bags, and from each sample 500 mL of must were obtained and fermented, using 500 mL Erlenmeyer flasks with a rubber stopper that was perforated with a syringe needle to allow CO2 release. Fermentations were performed at room temperature and progress was followed by daily weight loss determinations due to CO2 production. Immediately after the preparation of the must, diluted aliquots (10−1 to 10−5) were spread on plates containing YPD medium (yeast extract, 1% w/v; peptone, 1% w/v; glucose 2% w/v and agar 2%, w/v) supplemented with biphenyl (40 mg L-1). After incubation (2 days, 30°C), 30 colonies were randomly collected from plates containing between 30 and 300 colonies, which corresponded to a dilution of 10−2. When the weight loss of the must was about 65–70 g, corresponding to a stage close to the end of fermentation, must aliquots were again withdrawn, diluted and spread onto agar plates containing YPD medium (without biphenyl supplementation). Thirty colonies were randomly collected from plates containing between 30 and 300 colonies. Yeast isolates were stored in glycerol (30%, v/v) at -80°C.
Sampling sites (A) and summary of the samples collected (B) in 2009 and 2010 in 22 vineyards from 13 wine-producing areas on five islands of the Azores Archipelago, from the grapevine cultivars Verdelho dos Açores (Ver), Arinto da Terceira (Ar) e Terrantez do Pico (Tr) planted in arable soil (AS), soil covered with stone (SS) or solidified lava flows (LF), and trained as trellised grapevines (TG), non-trellised grapevines (NT) or lying grapevines (LG).
Molecular identification of the yeast isolates
DNA extraction was performed according to Drumonde-Neves et al. [47]. Molecular identification of isolates from freshly crushed grapes was performed by restriction fragment length polymorphism analysis (RFLP) and DNA sequencing. The 5.8-S ITS region was amplified using the primers ITS1 (5'-TCCGTAGGTGAACCTGCGG-3') and ITS4 (5'-TCCTCCGCTTATTGATATGC-3') [48]. PCR reaction was performed as follows: initial denaturation at 95°C for 6 min; 35 cycles of denaturing at 95°C for 20 s, annealing at 53°C for 20 s, extension at 72°C for 1 min; final extension at 72°C for 5 min. PCR amplification was carried out in a final volume of 10 μL of a reaction mix containing 20–50 ng of yeast DNA, 0.5 U Taq polymerase (MBI Fermentas), 1x Taq buffer (10 mM TrisHCl, 50 mM KCl, 0.08% Nonidet P-40), 0.4 pmol of each primer, 0.2 mM of each deoxynucleotides and 1.5 mM MgCl2. After dilution (1:4), 10 μL of the PCR products were digested with the restriction endonuclease HinfI (Fermentas) according to the supplier’s instructions. PCR products and their restriction fragments were mixed and separated in a 2% (w/v) agarose gel containing GelRedTM, (1x TAE Buffer, 100 V, 75 min). Identical electrophoretic profiles of each sample were considered as conspecific and grouped, and one representative isolate per group was selected for sequencing of the 5.8-S ITS region. The amplicons were obtained as described above and sequenced by the Sanger method [49, 50]. Sequence reactions were performed by use of the forward primer ITS1, and a BigDye Terminator Cycle Sequence Ready Reaction Kit version 3.1 (Applied Biosystems, Foster City, CA). After the sequence reaction, excess dye terminators were removed by gel filtration. Sequences were analyzed with an automated DNA sequencer 3730XL (Applied Biosystems). Species identity was determined using the BLASTN program [51] and GenBank reference sequences, considering an identity threshold of at least 98%.
Regarding final stages of fermentation, all isolates were analyzed by interdelta sequence typing [52,53]. Isolates that showed no interdelta pattern were considered to belong to non-Saccharomyces species, and were identified using the method described above.
Statistical analysis
Differences in community composition between islands/regions were tested by analysis of similarity (ANOSIM) on squared root transformed species incidences, using the informatics program PAST [54]. An analysis of similarity (ANOSIM) test makes no assumption about the normality of data and this multivariate test is classically used in community ecology. The test reports the probability of observing differences in community composition between islands/locations by chance using permutations of a rank Bray–Curtis similarity matrix to create null distributions [55]. The observed rank abundance difference (R) between islands/locations is reported, and this value ranges from -1 to +1. R-values above or below zero indicate that communities differ or not, respectively, between islands/locations. Null distributions were generated, that recalculated R under a framework that assumed no difference between regions, by randomizing the labels associated with samples across the entire data set one million independent times.
Results
Yeast species occurring on freshly crushed grapes
A total of 22 vineyards were sampled, spanning the geographic range of the wine-producing regions of the Azores Archipelago, as listed in Fig 2. Fifty-seven grape samples of the white grapevine cultivars Verdelho dos Açores, Arinto da Terceira and Terrantez do Pico were collected in 2009 and 2010, and 1710 yeast isolates were obtained from the freshly crushed grapes and analyzed by ITS-RFLP and by sequencing the 5.8-S ITS region. As summarized in Table 1 a total of 26 species were found in both years (14 in 2009 and 12 in 2010), corresponding to 19 different species since seven of them were found in both sampling years.
Globally, the most representative species was Hanseniaspora uvarum, found in the three grape cultivars and in all islands, corresponding to 41% and 87% of the total isolates obtained in 2009 and 2010, respectively. In the second sampling year the incidence of this species was considerably higher in comparison to the previous year, ranging between 53.3% (grape samples from Arinto cultivar collected in Biscoitos) and 100% (grape samples from Verdelho collected in the remaining locations in Pico). Metschnikowia pulcherrima was the second most representative yeast species. However it occurred only in 2009 and with a considerably higher incidence on Lajidos (95.6% and 83.3%, Arinto and Verdelho grape cultivars, respectively) compared to Graciosa (3.3% and 5.8%, Arinto and Verdelho grape cultivars, respectively) and S. Jorge (3.3%, Arinto grape cultivar). Pichia terricola was the third most representative species, corresponding to 15% and 3% of the isolates obtained in 2009 and 2010, respectively. In the first sampling year this species was isolated from Arinto grapes in all islands, ranging between 2.2% and 80.0% of the isolates from Lajidos and Biscoitos, respectively. From the grape cultivar Verdelho, P. terricola was isolated in 2009 only in Biscoitos and Lajidos (33.3% and 1.7% of the isolates, respectively). In grape samples from the cultivar Terrantez, P. terricola was found only in “remaining locations in Pico” both in 2009 (23.3%) and 2010 (6.7%). In 2010 this species was also found in grape samples from Arinto collected in Biscoitos (2.2%), Lajidos (2.2%), “remaining locations in Pico” (0.8%) and S. Jorge (5.0%) and in samples from Verdelho collected in S. Miguel (1.7%), Graciosa (7.8%) and Lajidos (15%). Similar patterns of distribution among the different islands and sampling years were also observed for the yeast species with a lower global representation such as Starmerella bacillaris (former Candida zemplinina) or Saccharomycopsis vini. To our knowledge, the species Candida carpophila was not previously found in grape or wine-associated environments, unlike the remaining species listed in Table 1.
When analyzing the global yeast biota isolated from the grape samples of each of the cultivars (Fig 3A), we observed that the percentage of H. uvarum isolated from the Verdelho samples was higher compared to the samples from Arinto cultivar (26,9% and 15,4%, in 2009 and 2010, respectively). This same tendency was observed when the comparison of the yeast biota isolated from each of the two grape cultivars was performed according to the islands (Fig 3B) or sampling locations where the varieties Verdelho and Arinto occurred simultaneously (Fig 3C). This was not observed in the vineyard 12, where the percentage of H. uvarum isolates from Arinto grapes was of 83% while this species was not isolated from grapes of the Verdelho variety. However, in this vineyard, 100% of the isolates collected from Verdelho cultivar corresponded to Hanseniaspora opuntiae, representing a higher proportion of the Hanseniaspora genus on Verdelho cultivar in comparison to Arinto.
Yeast communities composition (in percentage) from freshly crushed grapes of the grapevine cultivars Verdelho dos Açores (Ver) and Arinto da Terceira (Ar) collected in 2009 and 2010 on A) all sampled locations, B) in Lajidos (PLJ), the remaining locations of Pico (PRL), in Biscoitos (BCT) and in Graciosa (GRC) and C) in six vineyard (numbers 5, 12, 15, 16, 17 and 19; see Table 1).
Yeast species isolated from final stages of fermentations
From the 57 grape samples that were collected in five islands of the Archipelago, 40 spontaneous fermentations were achieved (Table 2). When the weight loss due to CO2 release during fermentation corresponded to 70 g L-1, a total of 1200 yeast isolates were collected at the final stages of fermentations (720 isolates from fermentations in 2009 and 480 from 2010). Thirty-one samples were predominantly fermented by S. cerevisiae (more than 75% of S. cerevisiae isolates in each sample). Ten fermentations were finished by non-Saccharomyces species (Candida boidinii, C. glabrata, C. humilis, C. railenensis, Starmerella bacillaris, Hanseniaspora vinae, Issatchenkia hanoiensis, Kazachstania exigua, Kluyveromyces thermotoerans, Meyerozyma guilliermondii, Pichia kudriavzevii, P. terricola, Saccharomycodes ludwigii, Zygosaccharomyces bailii). These species occurred together and in different proportions, with the exception of three fermentations that contained 100% of isolates of the species C. boidinii, C. railenensis or C. glabrata. The remaining 16 samples did not reach the final fermentative stage, since the weight loss was less than 65 g L-1 after 30 days of fermentation.
As shown in Table 2, higher species diversity was found in 2009 compared to 2010 (12 and 6, respectively). The species S. cerevisiae was predominant in both sampling years, corresponding to 64.5% and 87.5% of the total isolates obtained from the final stage of the fermentations performed in 2009 and 2010, respectively, independent of the grape cultivar and the island.
The species C. railenensis, S. bacillaris, M. guilliermondii and H. vineae occurred in two to five grape samples and each of these species was present only in two or three of the islands. The percentage of these species in each fermentation ranged between 6.7% and 100%. Less frequent species I. hanoiensis occurred in two fermentations from the same location (corresponding to 1.7% and 40% of the isolates). Some species occurred only in a single fermentation, representing between 66.7% and 100% of the isolates (C. boidinii, C. glabrata and Z. bailii) or between 10% and 24% of the isolates (K. exigua and Lachancea thermotolerans). Other species were rarely found, corresponding to one single isolate from a single fermentation (C. humilis, P. terricola and S. ludwigii).
From the 28 yeast species isolated in this study, listed in Tables 1 and 2, only five were found both in freshly crushed grapes and at the end of fermentations: C. railenensis, S. bacillaris, I. hanoiensis, H. vinae and P. terricola., For all but one of the grape samples, species occurring in the freshly crushed grapes did not occur at the end of the correspondent fermentation. One exception was observed for P. terricola that was isolated in both stages of the same fermentation.
Comparison between the different sampling locations
Differences in community composition between sampling locations were tested by analysis of similarity (ANOSIM) on squared root transformed species incidences. Fig 4 shows R-values, indicative of the observed rank abundance difference between locations, ranging this value from -1 to +1, with positive and negative values indicating that communities differ or not, respectively.
Statistically significant differences in yeast community composition observed between islands/locations (SMG: São Miguel, BCT: Biscoitos, GRC: Graciosa, PLJ: Lajidos, PRL: remaining locations in Pico, SJG: São Jorge). R-values are indicated in bold, and P-values in italic. Light grey and dark grey boxes refer to comparisons made with yeast species obtained in 2009 and 2010, respectively.
Concerning the year of 2009, our results show that Biscoitos and Lajidos locations harbor communities that differ from the remaining locations of the archipelago. As shown in Fig 4, yeast communities collected in 2009 from the Lajidos location differed greatly from all islands of the Archipelago, with R-values ranging between 0.5467 (P = 0.0087) and 0.9200 (P = 0.0085). When comparing the Lajidos location with the “remaning locations in Pico”, despite a less significant difference, a positive R-value (0.1982; P = 0.085) indicates that they harbor different communities. In the first sampling year, significant differences were also found between yeast communities from Biscoitos and S. Miguel islands (R = 0.4320; P = 0.0265). However with lower significance, only positive R-values were observed from the comparison between Biscoitos with all other locations (except S. Jorge). In the second sampling year, the only significant differences between yeast communities were observed when comparing the “remaining location in Pico” with S. Jorge and Biscoitos (R = 0.5000 and 0.2682; P = 0.0276 and 0.0453), reflecting the predominance of H. uvarum in all grape samples, and the considerable decrease in species richness. Using this biodiversity index, together with Shannon index, similar results were obtained, being this analysis present in Supplementary data (S1 Fig, S1 Table).
Discussion
A two-year sampling plan was designed and implemented in 13 locations, in five islands of the Azores archipelago. A total of 57 grape samples from three Azorean white grapevine cultivars were collected and 40 spontaneous fermentations were achieved. A total of 2910 yeast isolates were obtained and 28 yeast species were identified. We must caution that yeast isolates were obtained through selective conditions of growth that may differ from abiotic factors found in nature. Rarely occurring or slow-growing species may not have been detected, as the detection limit of our experimental approach is 3.3% (one species in 30 isolates). Also, when sequencing only one representative per grape sample (in a total of 450 profiles sequenced), some diversity could be lost. However, we consider that our approach allows a comparison of the yeast microbiota across vineyards and islands, even though it cannot provide a complete description of yeast community composition. Already in our previous work [56], we showed that this approach allowed the comparison between grape yeast communities. In adition, it was shown that the most representative genera belonging to these communities were comparable to the ones obtained in other continental areas. A fraction of the microbiota analyzed in this study was isolated after enrichment through must fermentation, allowing the comparison of the biodiversity of species enduring fermentation.
In agreement with most of the available bibliography, also in this study only non-Saccharomyces species were isolated from freshly crushed grapes and, for the first time, C. carpophila was encountered as an inhabitant of grape or wine-associated environments. To our knowledge so far, Pichia cecembensis has been isolated from grapes only in vineyards of the Azores Archipelago, from Pico and Terceira islands in one of our previous studies [57]. In the present study, this species was also found in Graciosa and S. Jorge. The wide occurrence of this species in the Azores (although in small proportions) may represent a particularity of the yeast biota from the vineyards of the archipelago. However, this species was only recently described [58], which might also explain the lack of previous findings in other grape ecosystems. The most representative species, H. uvarum and M. pulcherrima, are often reported as the predominant species on ripe grapes, in particular in regions with warmer climates [5,8,9,22,59–61]. P. terricola and S. bacillaris are also commonly reported as associated with grapes and wine environments [23,46,62–65]. However, in this study, both species occurred in higher proportions in comparison to other wine-producing areas. On the other hand, H. opuntiae has been less frequently reported as inhabitant of grape ecosystem [9,66–68], but in our study this species corresponded to 3.5% of the total isolates from freshly crushed grapes, despite being found only in 2 samples. Interestingly, in both samples H. opuntiae was predominant (more than 96% of the isolates) and H. uvarum was not found (while it was present in 82% of the remaining 55 grape samples). A similar correlation is reported by Settanni et al. [9], pointing to a possible antagonism between strains of both species. Globally, the remaining species (C. inconspicua, C. quercitrusa, C. railenensis, H. vineae, I. hanoiensis, P. fermentans, P. membranifaciens, S. crataegensis, S. vini and W. anomalus) were more rarely found on the archipelago, in accordance with most bibliographic references [5,65,69–72]. Pichia terricola was one of the most frequently found species in freshly crushed grapes, which is in agreement with recent reports that found this yeast in all stages of must fermentations [7,73]. In our study, only one isolate was obtained from final stages of the fermentations, suggesting that the isolated strains may have a low ethanol resistance or a weak ability to compete with other species. The remaining species isolated from finished fermentations were present in the correspondent freshly crushed grapes samples in proportions under the detection limit of our experimental approach (3.3%—one in 30 isolates). Only after enrichment through must fermentation could they be detected, with the frequencies and proportions generally found in similar studies on other geographic locations [4,5,7,46,63,74–76]. Among them, S. cerevisiae was predominant (in 75% of the fermentations), which is in agreement with the general observation that the isolation of S. cerevisiae from sound grapes, through direct plating methods has been rarely described (reviewed by [77]).
The effect of the non-Saccharomyces species on wine quality has been widely reported [71,78]. A possible linkage between yeast biota and a certain vintage, a specific grape cultivar and/or a particular wine-producing region may contribute to the typical sensory profile of the correspondent wines.
When considering the great differences found in the yeast biota between the two sampling years, we must stress that climatologically, the year of 2009 was characterized by the average conditions expected for the Archipelago, whereas unusually high frequency of precipitation was recorded in 2010. Rainfalls on the central and eastern islands were 20% and 60% higher, respectively, compared to the average values of the previous 30 years. In 2010, the adverse climatic conditions, led to a decrease of more than 80% of the average wine production. Globally, the yeast biota suffered a strong reduction in the species diversity from the first to the second sampling year (23 and 15 species respectively), which can be attributed to the pronounced climatic differences between the two sampling years. However, no data for the climatic variation between islands during grapes ripening period are available, which could impact those values obtained regarding yeast communities composition. In agreement with our results, some studies found a similar correlation between rainfall and yeast biodiversity [3,79], although other authors had reported a higher yeast biodiversity in rainy years [30,80]. This fact has been attributed to the effect of antifungal treatments that are carried out with greater intensity in years with higher precipitations [81]. In fact, some studies suggest that the usage of pesticides in vineyards decrease yeast biodiversity [82], however this is not universal [23,28,83,84]. Our previous studies [57] on yeast communities from abandoned vineyards suggested that the decrease of yeast diversity in rainy years is rather related to climatic conditions than to the usage of fungicides. Comparing both sampling years, a bigger decrease in species richness was observed in the yeast communities from the final stage of fermentations (from 12 to 6 species) than from the freshly crushed grapes (from 14 to 11 species).
Several studies suggest that particular yeast communities may be associated with certain grape cultivars [22,25,46,61,73,75,85,86], however no clear patterns or robust conclusions were achieved. Other authors did not find any correlation between grape cultivar and yeast communities composition [79,87]. We herein show that the yeast species H. uvarum tends to occur in higher proportions on grape samples from Verdelho than from Arinto cultivars. As described in the previous section, this was observed in both sampling years and in several locations and vineyards (Fig 3A–3C). Similar cultivar-specific yeast occurrences have been linked to varietal factors such as the thickness of the berry skin [88], that might be related to differences in berry composition, grape ripeness and sanitary condition at the sampling moment. However, this observation corresponds to a general tendency when the yeast biota was globally analyzed. The confirmation of a linkage between higher incidence of H. uvarum and the cultivar Verdelho needs further investigation. No differences were apparent between grape cultivars concerning the predominance of other species or concerning the presence of S. cerevisiae.
The geographical location and microclimatological conditions affects the composition of yeast communities isolated from grape and fermentations [22,46]. In agreement with this, our study revealed qualitative and quantitative variations in the yeast flora composition, between several locations from five islands of the Azores Archipelago. The species richness varied among geographical location, independently from the grape cultivar, sampling year, or fermentation stage. The highest yeast diversity was observed in the vineyards from S. Jorge, Biscoitos and Lajidos locations, having very particular environmental characteristics, corresponding to distinctive terroirs. The soils on these locations, covered by basaltic stone or solidified lava flows, have a strong effect on the microclimate at the grape level [89]. These environmental characteristics, together with an intensive viticultural activity, might explain the observed higher species richness. Differences were also observed concerning the occurrence of the predominant species on freshly crushed grapes from these locations. For example, M. pulcherrima showed a much higher incidence (Arinto: 95.6%; Verdelho: 83.3%) in the wine region Lajidos in comparison with the “remaining locations in Pico” (26%) in the sampling year 2009. The same was observed in the Biscoitos region of the Terceira Island concerning the proportion of P. terricola, which was considerably higher compared to the other islands of the archipelago. The statistical analysis of the yeast community composition revealed significant differences between those two locations and the rest of the archipelago. These differences were particularly accentuated in 2009, whereas the occurrence of atypical climatic conditions strongly affected grapevine growth and grapes development in the second sampling year. This may have affected yeast communities composition, and therefore differences between locations were less apparent.
This study is the first report on autochthonous yeast communities from the grapevine cultivars used in production of Azorean geographical indications wines. Considerable climate-associated variations were observed between vintages. Differences between cultivars were apparent and significant differences in yeast community composition were found between locations. In 2009 yeast communities found on freshly crushed grapes that were sampled from the locations of Lajidos and Biscoitos diverged greatly from the remaining locations of the archipelago.
Our observations strongly support the existence of a linkage between grape yeast communities and vineyards´ associated ecology. This fact is particularly true in vineyards with very distinctive environmental conditions.
Supporting Information
S1 Fig. PCA visualization obtained using average values of Shannon’s index and number of yeast species per sample or fermentation determined in six wine-producing areas of the Azores Archipelago (SMG–S. Miguel, BCT–Biscoitos, GRC–Graciosa, PLG–Lajidos, PRL–“Pico remaining locations”, SJG–S. Jorge).
https://doi.org/10.1371/journal.pone.0169883.s001
(DOCX)
S1 Table. Shannon´s index and average number of species per sample calculated for each grape sample or fermentation, in 2009 and 2010 and six locations (SMG–S. Miguel, BCT–Biscoitos, GRC–Graciosa, PLG–Lajidos, PRL–“Pico remaining locations”, SJG–S. Jorge).
https://doi.org/10.1371/journal.pone.0169883.s002
(DOCX)
Acknowledgments
The authors would like to thank all wine-producers that kindly supported grape samples collection.
Author Contributions
- Conceptualization: JDN TL DS.
- Data curation: JDN RFD.
- Formal analysis: JDN RFD.
- Funding acquisition: TL DS CP.
- Investigation: JDN RFD CP.
- Methodology: JDN RFD DS.
- Project administration: TL DS CP.
- Resources: DS CP.
- Software: RFD.
- Supervision: TL DS CP.
- Validation: JDN RFD.
- Visualization: JDN RFD.
- Writing – original draft: JDN RFD DS CP.
- Writing – review & editing: JDN RFD CP.
References
- 1.
Fleet GH. The yeast ecology of wine grapes. In: Ciani M, editor. Biodiversity and Biotechnology of wine yeasts. Research Signpost; 2002. pp. 1–19.
- 2. Beltran G, Torija MJ, Novo M, Ferrer N, Poblet M, Guillamón JM, et al. Analysis of yeast populations during alcoholic fermentation: a six year follow-up study. Syst Appl Microbiol. 2002;25: 287–93. pmid:12353885
- 3. Jolly NP, Augustyn OHPH, Pretorius IS. The occurrence of non-Saccharomyces cerevisiae yeast species over three vintages in four vineyards and grape musts from four production regions of the Western Cape, South Africa. South African J Enol Vitic. 2003;24: 35–42.
- 4. Combina M, Elia A, Mercado L, Catania C, Ganga A, Martinez C, et al. Dynamics of indigenous yeast populations during spontaneous fermentation of wines from Mendoza, Argentina. Int J Food Microbiol. 2005;99: 237–43. pmid:15808358
- 5. Di Maro E, Ercolini D, Coppola S. Yeast dynamics during spontaneous wine fermentation of the Catalanesca grape. Int J Food Microbiol. 2007;117: 201–210. pmid:17512625
- 6. Sun HH, Ma HQ, Hao ML, Pretorius IS, Chen SW. Identification of yeast population dynamics of spontaneous fermentation in Beijing wine region, China. Ann Microbiol. 2009;59: 69–76.
- 7. Ocon E, Gutierrez AR, Garijo P, Tenorio C, Lopez I, Lopez R, et al. Quantitative and qualitative analysis of non-Saccharomyces yeasts in spontaneous alcoholic fermentations. Eur Food Res Technol. 2010;230: 885–891.
- 8. Li E, Liu A, Xue B, Liu Y. Yeast species associated with spontaneous wine fermentation of Cabernet Sauvignon from Ningxia, China. World J Microbiol Biotechnol. 2011;27: 2475–2482.
- 9. Settanni L, Sannino C, Francesca N, Guarcello R, Moschetti G. Yeast ecology of vineyards within Marsala wine area (western Sicily) in two consecutive vintages and selection of autochthonous Saccharomyces cerevisiae strains. J Biosci Bioeng. Elsevier Ltd; 2012;114: 606–14. pmid:22877686
- 10. Lambrechts MG, Pretorius IS. Yeast and its importance to wine aroma—a review. S Afr J Enol Vitic. 2000;21: 97–129.
- 11. Cordente AG, Swiegers JH, Hegardt FG, Pretorius IS. Modulating aroma compounds during wine fermentation by manipulating carnitine acetyltransferases in Saccharomyces cerevisiae. FEMS Microbiol Lett. 2007;267: 159–166. pmid:17156120
- 12. Ciani M, Comitini F, Mannazzu I, Domizio P. Controlled mixed culture fermentation: a new perspective on the use of non-Saccharomyces yeasts in winemaking. FEMS Yeast Res. 2010;10: 123–33. pmid:19807789
- 13. King ES, Kievit RL, Curtin C, Swiegers JH, Pretorius IS, Bastian SEP, et al. The effect of multiple yeasts co-inoculations on Sauvignon Blanc wine aroma composition, sensory properties and consumer preference. Food Chem. Elsevier Ltd; 2010;122: 618–626.
- 14. Hyma KE, Saerens SM, Verstrepen KJ, Fay JC. Divergence in wine characteristics produced by wild and domesticated strains of Saccharomyces cerevisiae. FEMS Yeast Res. 2011;11: 540–51. pmid:22093681
- 15. Romano P, Fiore C, Paraggio M, Caruso M, Capece A. Function of yeast species and strains in wine flavour. Int J Food Microbiol. 2003;86: 169–180. pmid:12892932
- 16. Swiegers JH, Pretorius IS. Yeast modulation of wine flavor. Adv Appl Microbiol. 2005;57: 131–175. pmid:16002012
- 17. Swiegers JH, Pretorius IS. Modulation of volatile sulfur compounds by wine yeast. Appl Microbiol Biotechnol. 2007;74: 954–60. pmid:17262212
- 18. Howell KS, Cozzolino D, Bartowsky EJ, Fleet GH, Henschke PA. Metabolic profiling as a tool for revealing Saccharomyces interactions during wine fermentation. FEMS Yeast Res. 2006;6: 91–101. pmid:16423074
- 19. Anfang N, Brajkovich M, Goddard MR. Co-fermentation with Pichia kluyveri increases varietal thiol concentrations in Sauvignon Blanc. Aust J Grape Wine Res. 2009;15: 1–8.
- 20. Martini A, Frederichi F, Rosini G, Fattal B, Katzenelson E, Federici F. A new approach to the study of yeast ecology on natural substrates. Can J Microbiol. 1980;26: 856–859.
- 21. Pretorius IS, van der Westhuizen TJ, Augustyn OHP. Yeast biodiversity in vineyards and wineries and its importance to the South African wine industry. South African J Enol Vitic. 1999;20: 61–74.
- 22. Raspor P, Milek DM, Polanc J, Možina SS, Čadež N. Yeasts isolated from three varieties of grapes cultivated in different locations of the Dolenjska vine-growing region, Slovenia. Int J Food Microbiol. 2006;109: 97–102. pmid:16626833
- 23. Cadez N, Zupan J, Raspor P. The effect of fungicides on yeast communities associated with grape berries. FEMS Yeast Res. 2010;10: 619–630. pmid:20491940
- 24. Tofalo R, Schirone M, Telera GC, Manetta AC, Corsetti A, Suzzi G. Influence of organic viticulture on non-Saccharomyces wine yeast populations. Ann Microbiol. 2010;61: 57–66.
- 25. Cordero-Bueso G, Arroyo T, Serrano A, Valero E. Influence of different floor management strategies of the vineyard on the natural yeast population associated with grape berries. Int J Food Microbiol. Elsevier B.V.; 2011;148: 23–9. pmid:21620506
- 26. Cordero-Bueso G, Arroyo T, Serrano A, Tello J, Aporta I, Vélez MD, et al. Influence of the farming system and vine variety on yeast communities associated with grape berries. Int J Food Microbiol. 2010/12/28. Elsevier B.V.; 2011;145: 132–139. pmid:21185102
- 27. Setati ME, Jacobson D, Andong U-C, Bauer F. The vineyard yeast microbiome, a Mixed Model Microbial Map. Gilbert JA, editor. PLoS One. 2012;7: e52609. pmid:23300721
- 28. Milanović V, Comitini F, Ciani M. Grape berry yeast communities: Influence of fungicide treatments. Int J Food Microbiol. Elsevier B.V.; 2013;161: 240–246. pmid:23337124
- 29. Oliva J, Martínez-Gil A, Lorenzo C, Cámara M, Salinas M, Barba A, et al. Influence of the use of fungicides on the volatile composition of Monastrell red wines obtained from inoculated fermentation. Food Chem. 2015;170: 401–406. pmid:25306363
- 30. Longo E, Cansado J, Agrelo D, Villa TG. Effect of climatic conditions on yeast diversity in grape musts from northwest Spain. Am J Enol Vitic. 1991;42: 141–144.
- 31. Jackson DI, Lombard PB. Environmental and Management Practices Affecting Grape Composition and Wine Quality—A Review. Am J Enol Viticul. 1993;44: 409–430.
- 32. Bokulich N a, Thorngate JH, Richardson PM, Mills D a. Microbial biogeography of wine grapes is conditioned by cultivar, vintage, and climate. Proc Natl Acad Sci U S A. 2014;111: E139–48. pmid:24277822
- 33. Fleet GH, Lafonlafourcade S, Ribereaugayon P. Evolution of Yeasts and Lactic-Acid Bacteria During Fermentation and Storage of Bordeaux Wines. Appl Environ Microbiol. 1984;48: 1034–1038. pmid:16346661
- 34. Rosini G, Frederichi F, Martini A. Yeast flora of grape berries during ripening. Microbiol Ecol. 1982;8: 83–89.
- 35. Zarraonaindia I, Owens S, Weisenhorn P, West K, Hampton-Marcell J, Lax S, et al. The Soil Microbiome Influences Grapevine-Associated Microbiota. MBio. 2015;6: e02527–14. pmid:25805735
- 36. Versavaud A, Courcoux P, Roulland C, Dulau L, Hallet JN. Genetic diversity and geographical distribution of wild Saccharomyces cerevisiae strains from the wine-producing area of Charentes, France. Appl Environ Microbiol. 1995/10/01. 1995;61: 3521–9. pmid:7486988
- 37. Khan W, Augustyn OPH, Westhuizen TJ Van Der, Lambrechts MG, Pretorius IS. Geographic Distribution and Evaluation of Saccharomyces cerevisiae Strains Isolated from Vineyards in the Warmer, Inland Regions of the Western Cape in South Africa. South African J Enol Vitic. 2000;21: 17–31.
- 38. Pramateftaki P V, Lanaridis P, Typas MA. Molecular identification of wine yeasts at species or strain level: a case study with strains from two vine-growing areas of Greece. J Appl Microbiol. 2000;89: 236–48. pmid:10971755
- 39. Torija MJ, Rozès N, Poblet M, Guillamón JM, Mas A, Rozes N, et al. Yeast population dynamics in spontaneous fermentations: comparison between two different wine-producing areas over a period of three years. Antonie Van Leeuwenhoek. 2001;79: 345–52. pmid:11816978
- 40. Lopes CA, van Broock M, Querol a, Caballero a C, Broock M Van. Saccharomyces cerevisiae wine yeast populations in a cold region in Argentinean Patagonia. A study at different fermentation scales. J Appl Microbiol. 2002;93: 608–615. pmid:12234344
- 41. Capece A, Romaniello R, Siesto G, Pietrafesa R, Massari C, Poeta C, et al. Selection of indigenous Saccharomyces cerevisiae strains for Nero d’Avola wine and evaluation of selected starter implantation in pilot fermentation. Int J Food Microbiol. 2010;144: 187–92. pmid:20933292
- 42. Barata A, Malfeito-Ferreira M, Loureiro V. The microbial ecology of wine grape berries. Int J Food Microbiol. Elsevier B.V.; 2012;153: 243–59. pmid:22189021
- 43. Vezinhet F, Hallet J-N, Valade M, Poulard A. Ecological survey of wine yeast strains by molecular methods of identification. Am J Enol Vitic. 1992;43: 83–86.
- 44. Martínez C, Cosgaya P, Vásquez C, Gac S, Ganga A, Martinez C, et al. High degree of correlation between molecular polymorphism and geographic origin of wine yeast strains. J Appl Microbiol. 2007;103: 2185–95. pmid:18045401
- 45. Valero E, Cambon B, Schuller D, Casal M, Dequin S. Biodiversity of Saccharomyces yeast strains from grape berries of wine-producing areas using starter commercial yeasts. FEMS Yeast Res. 2007;7: 317–29. pmid:17040482
- 46. Gayevskiy V, Goddard MR. Geographic delineations of yeast communities and populations associated with vines and wines in New Zealand. ISME J. 2011/12/23. Nature Publishing Group; 2011;6: 1281–90. pmid:22189497
- 47. Drumonde-Neves J, Vieira E, Lima MT, Araujo I, Casal M, Schuller D. An easy, quick and cheap high-throughput method for yeast DNA extraction from microwell plates. J Microbiol Meth. Elsevier B.V.; 2013;93: 206–208.
- 48. Esteve-Zarzoso B, Belloch C, Uruburu F, Querol a. Identification of yeasts by RFLP analysis of the 5.8S rRNA gene and the two ribosomal internal transcribed spacers. Int J Syst Bacteriol. 1999;49 Pt 1: 329–37.
- 49. Sanger F, Coulson a R. A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J Mol Biol. 1975;94: 441–8. pmid:1100841
- 50. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74: 5463–5467. pmid:271968
- 51. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, et al. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucleic Acids Res. 1997;25: 3389–3402. pmid:9254694
- 52. Schuller D, Valero E, Dequin S, Casal M. Survey of molecular methods for the typing of wine yeast strains. FEMS Microbiol Lett. 2004;231: 19–26. pmid:14769461
- 53. Franco-Duarte R, Mendes I, Gomes AC, Santos M a S, de Sousa B, Schuller D. Genotyping of Saccharomyces cerevisiae strains by interdelta sequence typing using automated microfluidics. Electrophoresis. 2011;32: 1447–1455. pmid:21630290
- 54. Hammer Ø, Harper DAT, Ryan PD. PAST: paleontological statistics software package for education and data analysis. Palaeontol Electron. 2001;4: 1–9.
- 55. Clarke KR. Non-parametric multivariate analyses of changes in community structure. Aust J Ecol. 1993;18: 117–143.
- 56. Drumonde-Neves J, Franco-Duarte R, Lima T, Schuller D, Pais D. Yeast biodiversity in vineyards environments is increased by human intervention. Plos One. 2016; 11(8): e0160579. pmid:27500638
- 57. Drumonde-Neves J, Franco-Duarte R, Lima T, Schuller D, Pais C. Yeast Biodiversity in Vineyard Environments Is Increased by Human Intervention. PLoS One. 2016;11: e0160579. pmid:27500638
- 58. Bhadra B, Sreenivas Rao R, Naveen Kumar N, Chaturvedi P, Sarkar PK, Shivaji S. Pichia cecembensis sp. nov. isolated from a papaya fruit (Carica papaya L., Caricaceae). FEMS Yeast Res. 2007;7: 579–84. pmid:17316366
- 59. Pretorius IS. Tailoring wine yeast for the new millennium: novel approaches to the ancient art of winemaking. Yeast. 2000;16: 675–729. pmid:10861899
- 60. Combina M, Mercado L, Borgo P, Elia a, Jofre V, Ganga A, et al. Yeasts associated to Malbec grape berries from Mendoza, Argentina. J Appl Microbiol. 2005;98: 1055–1061. pmid:15836474
- 61. Clavijo A, Calderón IL, Paneque P, Calderon IL. Diversity of Saccharomyces and non-Saccharomyces yeasts in three red grape varieties cultured in the Serrania de Ronda (Spain) vine-growing region. Int J Food Microbiol. 2010/09/21. Elsevier B.V.; 2010;143: 241–245. pmid:20851489
- 62. Baffi MA, Santos Bezerra C, Arévalo-Villena M, Isabel Briones-Pérez A, Gomes E, Silva R. Isolation and molecular identification of wine yeasts from a Brazilian vineyard. Ann Microbiol. 2010;61: 75–78.
- 63. Brežná B, Zenišová K, Chovanová K, Chebeňová V, Kraková L, Kuchta T, et al. Evaluation of fungal and yeast diversity in Slovakian wine-related microbial communities. Antonie Van Leeuwenhoek. 2010;98: 519–29. pmid:20556654
- 64. Ciani M, Comitini F. Non-Saccharomyces wine yeasts have a promising role in biotechnological approaches to winemaking. Ann Microbiol. 2010;61: 25–32.
- 65. Li S-S, Cheng C, Li Z, Chen J-Y, Yan B, Han B-Z, et al. Yeast species associated with wine grapes in China. Int J Food Microbiol. Elsevier B.V.; 2010;138: 85–90. pmid:20116124
- 66. Nisiotou A a, Nychas G-JE. Yeast populations residing on healthy or botrytis-infected grapes from a vineyard in Attica, Greece. Appl Environ Microbiol. 2007;73: 2765–8. pmid:17293525
- 67. Bovo B, Andrighetto C, Carlot M, Corich V, Lombardi A, Giacomini A. Yeast population dynamics during pilot-scale storage of grape marcs for the production of Grappa, a traditional Italian alcoholic beverage. Int J Food Microbiol. Elsevier B.V.; 2009;129: 221–8. pmid:19136177
- 68.
Assis MO. Identificação das leveduras isoladas de uvas cultivadas no vale do submédio São Francisco e sua utilização na fermentação do mosto de uva. Universidade Federal da Bahia. 2012.
- 69. Sinigaglia M, Rosaria Corbo M, Ciccarone C. Influence of temperature, pH and water activity on “in vitro” inhibition of Penicillium glabrum (Wehmer) Westling by yeasts. Microbiol Res. Gustav Fischer Verlag; 1998;153: 137–143. pmid:9760746
- 70. Subden RE, Husnik JI, van Twest R, van der Merwe G, van Vuuren HJJ. Autochthonous microbial population in a Niagara Peninsula icewine must. Food Res Int. 2003;36: 747–751.
- 71. Jolly NP, Augustyn OPHP, Pretorius IS. The role and use of non-Saccharomyces yeasts in wine production. South African J Enol Vitic. 2006;27: 15–39.
- 72. Zhang HY, Lee SA, Bradbury JE, Warren RN, Sheth H, Hooks DO, et al. Yeasts isolated from New Zealand vineyards and wineries. Aust J Grape Wine Res. 2010;16: 491–496.
- 73. Clemente-Jimenez JM, Mingorance-Cazorla L, Martínez-Rodríguez S, Heras-Vázquez FJ Las, Rodríguez-Vico F. Molecular characterization and oenological properties of wine yeasts isolated during spontaneous fermentation of six varieties of grape must. Food Microbiol. 2004;21: 149–155.
- 74. Amerine BMA, Kunkee RE. Microbiology of winemaking. Annu Rev Microbiol. 1968;22: 323–358. pmid:4879519
- 75. Chavan P, Mane S, Kulkarni G, Shaikh S, Ghormade V, Nerkar DP, et al. Natural yeast flora of different varieties of grapes used for wine making in India. Food Microbiol. Elsevier Ltd; 2009;26: 801–808. pmid:19835764
- 76. Gonzalez SS, Barrio E, Querol a, González SS. Molecular identification and characterization of wine yeasts isolated from Tenerife (Canary Island, Spain). J Appl Microbiol. 2007;102: 1018–1025. pmid:17381745
- 77. Fleet GH. Yeast interactions and wine flavour. Int J Food Microbiol. 2003;86: 11–22. pmid:12892919
- 78. Ciani M, Maccarelli F. Oenological properties of non-Saccharomyces yeasts associated with wine-making. World J Microbiol Biotechnol. 1998;14: 199–203.
- 79. Rementeria A, Rodriguez JA, Cadaval A, Amenabar R, Muguruza JR, Hernando FL, et al. Yeast associated with spontaneous fermentations of white wines from the “Txakoli de Bizkaia” region (Basque Country, North Spain). Int J Food Microbiol. 2003;86: 201–207. pmid:12892935
- 80. de La Torres MJ, Millan MC, Perez-Juan MP, Morales J, Ortega JM. Changes in the microbiota during ripening of two Vitis vinifera grape varieties grown in southern Spain. Microbos. 1998;96: 165–176.
- 81. Parish ME, Carroll DE. Indigenous yeasts associated with muscadine (Vitis rotundifolia) grapes and musts. Am J Enol Vitic. 1985;36: 165–169.
- 82. Calhelha RC, Andrade J V, Ferreira IC, Estevinho LM. Toxicity effects of fungicide residues on the wine-producing process. Food Microbiol. 2006;23: 393–8. pmid:16943029
- 83. Comitini F, Ciani M. Influence of fungicide treatments on the occurrence of yeast flora associated with wine grapes. Ann Microbiol. 2008;58: 489–493.
- 84. Cabras P, Angioni A, Garau VL, Pirisi FM, Farris GA, Madau G, et al. Pesticides in fermentative processes of wine. J Agr Food Chem. 1999;47: 3854–7.
- 85. Valero E, Schuller D, Gambon B, Casal M, Dequin S, Cambon B. Dissemination and survival of commercial wine yeast in the vineyard: A large-scale, three-years study. FEMS Yeast Res. 2005;5: 959–969. pmid:15975857
- 86. Schuller D, Cardoso F, Sousa S, Gomes P, Gomes AC, Santos MAS, et al. Genetic diversity and population structure of Saccharomyces cerevisiae strains isolated from different grape varieties and winemaking regions. PLoS One. 2012;7: e32507. pmid:22393409
- 87. Ahavi TZ, Roby SD, Ohen LC, Eiss BW, Rie RBEN. Characterization of the yeast flora on the surface of grape berries in Israel. Vitis. 2002;41: 203–208.
- 88.
Bisson LF, Kunkee RE. Microbial interactions during wine production. In: Zeikus J.G. J, editor. Mixed Cultures in Biotechnology. New York: McGraw-Hill; 1991. pp. 37–68.
- 89. Medeiros CA. Contribuição para o estudo da vinha e do vinho nos Açores. Finisterra. 1994;58: 199–229.