Yeasts associated with Drosophila in tropical forests of Rio de Janeiro, Brazil PAULA B. MORAIS, ALLENN. HAGLER,' CARLOSA. ROSA, AND LEDAC. MENDONCA-HAGLER Instituto de Microbiologia, CCS, Bl. I, Ilha d o Fundclo, Universidade Federal d o Rio de Janeiro, Rio de Janeiro, Brasil 21941 AND
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
LOUIS
B.
KLACZKO~
Instituto de Biologia, CCS, Bl. A, Ilha d o Fund60, Universidade Federal d o Rio de Janeiro, Rio de Janeiro, Brasil 21941 Received March 2, 1992 Revision received May 11, 1992 Accepted June 4, 1992 MORAIS,P. B., HAGLER,A. N., ROSA, C. A., MENDONCA-HAGLER, L. C., and KLACZKO,L. B. 1992. Yeasts associated with Drosophila in tropical forests of Rio de Janeiro, Brazil. Can. J. Microbiol. 38: 1150-1155. The distribution and diversity of yeast species vectored by and from the crop of eight species groups of Drosophila is described for two rain forest sites and an urban wooded area in Rio de Janeiro, Brazil. The typical forest Drosophila groups guarani, tripunctata, and willistoni showed a higher diversity of yeasts than the cosmopolitan melanogaster species group, suggesting different strategies of utilization of substrates. Apiculate yeasts, including Kloeckera apis, Kloeckera javanica, and Kloeckera japonica, were the prevalent species. Geotrichum spp. and Candida citrea were also frequent isolates in the forest sites. Similarities between the yeasts from the external surfaces and crops of Drosophila suggested that the feeding substrates were the main source of the yeasts vectored by these flies. Most of the yeasts were strong fermenters and assimilated few compounds, usually sucrose, cellobiose, and glycerol. This indicated a preference of the flies for food sources such as fruits. Some yeasts were primarily isolated from one group of Drosophila; for example, Kloeckera javanica from the melanogaster group, Debaryomyces vanrijiae var. yarrowii from the tripunctata group, and Kluyveromyces delphensis from the willistoni group. These associations and differences in the yeast communities among the fly groups suggested a differentiation of diets and specialization of the yeast-Drosophila association in the tropical forests. Key words: Yeast communities, Drosophila, rain forest, tropical climate. MORAIS,P. B., HAGLER,A. N., ROSA, C. A., MENDONCA-HAGLER, L. C., et KLACZKO,L. B. 1992. Yeasts associated with Drosophila in tropical forests of Rio de Janeiro, Brazil. Can. J . Microbiol. 38 : 1150-1 155. La distribution et la diversite des espkces de levures dans un echantillonnage de huit groupes d'espkces de Drosophila vectrices de ces levures sont decrites pour deux sites d'une foret tropicale humide et une region boisee urbaine de Rio de Janeiro, Bresil. Les groupes de Drosophila : guarani, tripunctata et willistoni, typiques de la foret tropicale ont presente une diversite de levures plus elevee que les espkces du groupe melanogaster cosmopolitain, ce qui suggkre l'existence de strategies differentes d'utilisation des substrats. Les levures apiculees, incluant les Kloeckera apis, Kloeckera javanica et Kloeckera japonica, ont ete les espkces predominantes; toutefois, dans les sites forestiers, des espkces du genre Geotrichum et le Candida citrea ont aussi ete des isolats frequents. Les similarites entre les levures, de par leurs surfaces externes, et les jabots de Drosophila ont suggere que les substrats nutritifs etaient les principales sources de levures vChiculCes par ces mouches. La majorite des levures ont ete des agents de fermentation trks efficaces et n'ont assimile que peu de composes, habituellement du sucrose, du cellobiose et du glycerol. Ceci indique une preference des mouches pour des sources alimentaires telles que les fruits. Certaines levures ont ete isolees de groupes particuliers de Drosophila, par exemple le Kloeckera javanica du groupe melanogaster, le Debaryomyces vanrijiae var. yarrowii du groupe tripunctata et le Kluyveromyces delphensis du groupe willistoni. Ces associations et les differences dans les communautes de levures chez les groupes de mouches suggkrent une differenciation dans les diktes et une specialisation dans les associations levures-Drosophila dans les forets tropicales. Mots cles : communautes de levures, Drosophila, foret tropicale, climat tropical. [Traduit par la redaction]
Introduction Drosophila species are important vectors for yeasts in nature, and a mutualistic interaction exists in which the yeasts are a source of essential nutrients for the flies (Phaff and Starmer 1987). Starmer (1981) suggested that the characteristics of yeast communities partially describe the habitats of associated Drosophila. Different yeasts are associated with Drosophila species in tropical and temperate regions. The assimilation of only a few carbon sources 'corresponding author. 2 ~ r e s e n taddress: Departamento de Genetica e Evoluciio, Instituto de Biologia, UNICAMP, Cx. Postal 6109, Campinas, Brazil 13081. Prinled in Canada / lmprirne au Canada
and the intense fermentation of glucose is typical of yeasts isolated from tropical flies, suggesting the use of substrates rich in simple sugars such as fruits in decomposition (Do Carmo-Sousa 1969; Heed et al. 1976; Pignal and Lachaise 1979). The genera Pichia, Hanseniaspora, and Kloeckera predominate among yeasts of flies from a tropical forest in Brazil and Hanseniaspora valbyensis and Saccharomycopsis crataegensis are prevalent in African drosophilids (da Cunha et al. 1957; Pignal and Lachaise 1979). This contrasts with the higher frequency of species of Saccharomyces and Kluyveromyces in temperate regions (Carson et al. 1956; Heed et al. 1976; Phaff et al. 1956; Shehata et al. 1955). Adults of different Drosophila species in temperate regions probably share many feeding sites but
1151
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
MORAIS ET AL.
do not necessarily harbor yeasts from them. In contrast, yeasts of tropical Drosophila appear to be species isolated from the probable feeding substrates of these flies (da Cunha et al. 1957). Yeasts from the intestinal tracts of various species of tropical Drosophila are different from each other, demonstrating food preferences and diversity of diets among species of these flies (Carson et al. 1956; da Cunha 195 1; da Cunha et al. 1957; Dobzhansky and da Cunha 1955). The yeasts isolated from wild Drosophila species differ from those encountered in the intestinal tract of Drosophila melanogaster, a species associated with anthropogenic habitats (Phaff et al. 1978). Few studies have focussed on the interaction between Drosophila and yeasts in forest sites, and only one was done in Brazil. This was carried out in an Atlantic forest area in the state of Siio Paulo (da Cunha et al. 1957). Improvements in yeast taxonomy since then allow more accurate identification of yeasts. The comparison of different tropical forest habitats could help to elucidate the ecological specificity of yeast communities and the niche uses of the associated flies. Yeast communities from a habitat have nutritional requirements corresponding to the characteristics of the environment (Phaff et al. 1978; Lachance and Starmer 1986). Thus, their physiological profiles can be used to study the relationships among different Drosophila species groups occupying the same habitat. In this paper we describe the diversity, distribution, and physiological profiles of yeast communities associated with Drosophila species groups in an urban wooded area and two Atlantic rain forest sites. Materials and methods Fruit flies of the genus Drosophila were collected from three sites near the city of Rio de Janeiro in December 1989 and January, August, and September 1990. Catalfio Point is an abandoned farm area with fruit trees, mostly mango (Mangifera indica). This site is located at the Universidade Federal do Rio de Janeiro campus on Ilha do Fundfio and is influenced by human activities. Collections at Catalso Point were made when mango fruits were ripe. The Mairink Chapel area is within the Tijuca National Park forest, a 122-km2area of Atlantic rain forest. The Pau da Fome site is in a preserved primary rain forest area with restricted access near a suburb west of the city. Insects were attracted by means of baits made of mashed bananas fermented overnight with commercial baker's yeast. Bait was placed in 2-L plastic beakers covered with several layers of sterile gauze to avoid contamination of flies by yeasts from the bait. Three beakers of bait were placed in a triangular pattern 1.5 m apart in the field for 3-5 h during the collection. The flies were aspirated aseptically as they approached or landed on the gauze. Yeasts were isolated from the external parts of 228 adult flies that were allowed to walk 8-12 h on plates containing YMA (0.3% malt extract, 0.3% yeast extract, 0.5% peptone, 1% glucose, 2% agar adjusted to pH 4.0 with 1 M HC1, and supplemented with 100 mg/L of chloramphenicol). These flies were then recovered for identification. Another 286 adult flies were surface sterilized in the field by immersion in 70% ethanol for 1 min and transported to the laboratory within 2 h on ice in tubes containing sterile water. These were identified and crops dissected from them were streaked on YMA. Plates were incubated for 2-7 days at 25°C until colonies appeared. A representative of each type of colony morphology from each plate was brought into pure culture by successive plating on YMA. Yeast isolates were characterized and identified according to standard methods and keys (Barnett et al. 1983; Kreger van Rij 1984) and a DBB test was done as described by Hagler and Mendon~a-Hagler(1991). Multiple cultures identified as the
TABLE1. Number of flies belonging to eight Drosophila species groups in two forest sites and an urban wooded area in Rio de Janeiro Species groups
Sex
Catalao Point
Tijuca forest
Pau da Fome forest
Total
cardini dreyfusi guarani melanogaster repleta saltans tripunctata willistoni
Not groupeda Total NOTE: M, male; F, female. "Flies not identified as belonging to any Drosophila group.
same species from a single sample were considered a single isolate. Statistical analyses used to describe the relative occurrence of yeasts in either fly species groups or habitats were based on Starmer et al. (1990). Diversity indexes of yeasts were calculated for Drosophila and for the three collection sites. Geographic diversity and host diversity were calculated for yeast species that occurred in two or more sites or groups of flies. Physiological profiles for yeast communities were calculated by multiplying the number of isolates of each species by the species physiological capability and dividing by the total number of isolates. The physiological capability was calculated as described in Heed et al. (1976) and Lachance and Starmer (1986). Physiological profiles of the total yeast community and each group of flies and habitat were constructed on compounds that were frequently utilized as in Heed et al. (1976).
Results The 524 flies collected from the three sites were identified as eight groups of species of Drosophila listed in Table 1. The melanogaster species group dominated at Cataliio Point, willistoni and tripunctata groups dominated at the Tijuca forest site, and guarani group dominated at the Pau da Fome forest site. The other Drosophila species groups occurred in low numbers and only in the forests. The frequencies of isolation of 72 yeast species from different Drosophila species groups in the three collection sites are presented in Table 2. Habitat and host diversity indexes for the yeasts are also presented in Table 2. Apiculate yeasts, especially of the species Kloeckera apis, Kloeckera japonica, and Kloeckera javanica, were the most frequent isolates. Saccharomyces cerevisiae was represented by only three isolates, all of which differed from the baker's yeast used in the bait. The bait yeast differed from our isolates by assimilating trehalose but failing to form ascospores, grow at 37"C, ferment maltose, or assimilate galactose. Few strains of Pichia and Williopsis were isolated from Cataliio Point, but Pichia angusta, Pichia beckii, Pichia carsonii,
CAN. J . MICROBIOL. VOL. 38, 1992
1152
TABLE2. Isolation frequencies of yeasts from Drosophila of three sites in Rio de Janeiro Cataliio Point
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Yeasts
Tijuca forest
Pau da Fome forest
Host Geographical Me1 Tri Wil NG Gua Me1 Tri Wil Others NG Gua Tri Wil Others NG diversity" diversityb
A ureobasidiurn pullulans Candida antarctica Candida apicola Candida apis var. galacta Candida berthetii Candida birnundalis var . arnericana Candida boleticola Candida bornbicola Candida citrea 3 Candida colliculosa 2 Candida diversa Candida fructus 1 Candida guillierrnondii Candida hurnicola 1 "Candida hurnicola like" ''Candida kara waiewii like" Candida lambica Candida rnagnoliae 1 Candida rnelinii 2 2 Candida norvegensis 1 "Candida norvegensis like" 1 1 Candida pulcherrirna Candida quercitrusa "Candida sorbosa like" ''Candida tepae like" Candida variabilis ''Candida zeylanoides like" Candida sp. A Candida sp. B Candida sp. C Cryptococcus infirrno-rniniatus Cryptococcus rnacerans 1 Debaryornyces rnelissophilus 2 Debaryornyces vanrijiae var . yarro wii 1 Geotrichurn spp. 1 Hanseniaspora occidentalis 2 ~anseniasporauvarurn 6 2 ~anseniasporavalbyensis 1 1 Hanseniaspora vineae Kloeckera africana 30 10 Kloeckera apis Kloeckera corticis Kloeckera japonica 16 Kloeckera javanica 95 Kluyverornyces delphensis 2 Kluy verornyces rnarxianus var . drosophilarurn Kluy verornyces rnarxianus var . rnarxianus Kluyverornyces therrnotolerans 2 Kluy verornyces wickerharnii Pichia abadiae Pichia acaciae Pichia angusta Pichia anornala Pichia beckii Pichia carsonii Pichia dryadoides Pichia holstii Pichia kluyveri
4 1
4 1 5 2
3
1
3 9
9 4
2
1 2 2 1 1 1 1 2
1
10 5
2
2 4
1
2 1
1
1 20 4 1 3
2
1 1 3
8
3 1 1 3
4
1
6 1
4
2 1 5
5
2
3
1
1 2 1 5 5
2
2
1
3 3
1 1 2 1 1 1 1 1 5 14 5
4
7
3 6
1153
MORAIS ET AL.
TABLE2 (Concluded) Catalfio Point
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Yeasts
Tijuca forest
Pau da Fome forest
Host Geographical Me1 Tri Wil NG Gua Me1 Tri Wil Others NG Gua Tri Wil Others NG diversityo diversityb
Pichia silvicola Pichia toletana Rhodotorula araucariae Rhodotorula aurantiaca Rhodotorula fujisanensis Rhodotorula glutinis Rhodotorula minuta Rhodotorula rubra Saccharomyces cerevisiae Torulaspora pretoriensis Williopsis californica Williopsis saturnus Total NOTE:The number of flies in e a ~ h ' ~ r oatueach ~ site is noted in Table 1. Species names in the genera Pichia and Williopsis are according to Kurtzman (1984 and 1991). Gua, guarani group; Mel, melanogaster group; Tri, tripunctata group; Wil, willktoni group; others, cardini, dreyfusii, repleta, and sakans groups; NG, flies not belonging to any group. Host diversity calculated as 1 - (070 in each group/row total12. Geographic diversity calculated as 1 - (070 in each site/row t o t a ~ ) ~ .
Pichia kluy veri, Williopsis californica, and Williopsis saturnus were represented by multiple isolates at the forest sites. With the exception of Candida citrea and Candida colliculosa, the genus Candida was restricted to the forest sites. Candida apicola, Candida apis, Candida citrea, Candida colliculosa, Candida fmctus, Candida pulcherrima, and Candida quercitrusa were the most frequent isolates of this genus. Geotrichum was frequently isolated from forest sites, especially from the guarani group at Pau da Fome. Kluveromyces delphensis was associated with the willistoni group from which all but 1 of 34 strains were isolated. Debaryomyces vanrijiae var. yarrowii was the predominant yeast associated with the tripunctata group. The black yeast Aureobasidium pullulans was isolated in low numbers from all three sites and Basidiomycete yeasts and their anamorphs were represented by only a few isolates of Candida, Cryptococcus, and Rhodotorula species. Procedures for the isolation of yeasts from Drosophila surfaces yielded similar species to those for isolation of ingested yeasts, although more isolates were obtained from surfaces (557 cultures) than from crops (305 cultures). The most common yeasts from both procedures were Kloeckera apis, Kloeckera japonica, Kloeckera javanica, Candida citrea, and Kluy veromyces delphensis. Debaryomyces vanrijiae var. yarrowii and Pichia beckii were also frequently isolated from the external parts of flies. The willistoni and tripunctata groups had high yeast diversity indexes (0.93 and 0.89, respectively). The forest sites, where these flies were predominant, also showed high yeast diversity indexes (0.86 for Tijuca forest and 0.92 for Pau da Fome forest). The Catalso Point site had the lowest yeast diversity index (0.77) in accord with the abundance of melanogaster flies, which had the lowest diversity of yeasts (0.69). Kloeckera apis was the yeast with the highest host diversity, since it was frequently isolated from all groups of flies except dreyfusii and the crop of tripunctata flies (Table 2). Host diversity was high for Candida apicola, Candida melinii, and Candidapulcherrima, but they were rare in all host flies. Debaryomyces vanrijiae var. yarrowii and Kloeckera javanica showed the lowest values for host diver-
sity. The indexes of geographic diversity were relatively low for all yeasts, the highest values being for Kloeckera apis and Candida melinii and the lowest values for Klockera javanica and Kloeckera africana. The yeast biotypes isolated frequently failed to assimilate most of the carbon sources considered variable in the standard descriptions of their species (Table 3). The assimilative abilities were mostly restricted to a few compounds that included glycerol, D-xylose, sucrose, cellobiose, mannitol, and succinate. Fermentative species made up 84.2% of the isolates, but most of these fermented only glucose although rapidly and vigorously. Some biotypes differed from the standard descriptions of species by growing at higher temperatures than expected. Among the apiculate yeasts, 68 strains of Kloeckera apis grew at 40°C, and 117 strains of Kloeckera javanica and 4 strains of Kloeckera africana grew at 37OC. Some strains of Candida citrea grew at 37OC and strains of Candida berthetii, Candida guilliermondii, Pichia acaciae, and an unidentified Candida species positive for the DBB test grew at 43OC.
Discussion Yeast species are often associated with specific insects or plants (Phaff and Starmer 1987). Feeding preferences of different Drosophila species for different yeasts have been demonstrated (da Cunha 1951; Dobzhansky and da Cunha 1955; Dobzhansky et al. 1956; Lachaise 1979). The low values of host and geographical diversity obtained for the yeasts in the present study are in agreement with the existence of specialized associations of yeasts with Drosophila in tropical forest environments. Although the general characteristics of the yeast communities were similar, some species were associated predominantly with particular fly groups. This suggested that some specific associations occurred between yeast species and Drosophila species groups and that differentiation of diets is common among these groups of flies. Some of the yeasts isolated from Rio de Janeiro were associated with two or more groups of Drosophila. Pichia beckii and Williopsis californica were isolated from the willistoni and tripunctata groups. In the Pau da Fome site, the majority of the yeasts isolated from the guarani group
CAN. J . MICROBIOL. VOL. 38, 1992
1154
TABLE3. Physiological profiles of yeast isolates obtained in forest sites and an urban wooded area in Rio de Janeiro and from flies from temperate, desert, and tropical regions (% utilization of carbon compounds) Site of collectiona
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Tests
Cataliio
Tijuca
Pau da Fome
Drosophila groups
Tri
Will
Me1
Gua
Tropical fliesc
Tropical fliesd
Temperate fliese
Desert fliese
All known speciesy
Fermentation Glucose Assimilation Galactose Sucrose Maltose Trehalose Cellobiose Salicin D-XY~OS~ Glycerol Mannitol Lactate Succinate "Profile of the yeast species isolated from the three sites of collection. b ~ r o f i l eof the yeast communities isolated from the four most frequent groups of flies: Tri, tripunctata group; Will, willistoni group; Mel, melanogaster group; Gua, guarani group. 'Profile of all species isolated (our data). d ~ r o mHeed et 01. (1976), based on da Cunha et 01. (1957). eFrom Heed et 01. (1976). /From Lachance and Starmer (1986), based on Kreger van Rij (1984).
were also found in the willistoni group. This suggested the use of some common resources by these flies. Similar yeast populations and consequently similar feeding spectra have been noted for different African drosophilids (Lachaise et al. 1979). The willistoni and tripunctata groups had the highest yeast species diversities. This may have resulted from their occurrence in all three sites, in contrast with the anthropogenic habitat-associated melanogaster group that predominated only at Cataliio Point. The high yeast diversity in some fly species groups may indicate their use of various food sources. For the willistoni group this could be an effect of the differentiation of diets among species of the group as pointed out by da Cunha et al. (1957). The melanogaster group presented the lowest yeast diversity, probably reflecting a low yeast diversity at Cataliio Point where it was predominant. Apparently the melanogaster group utilized ripe mango fruits, the most abundant food source in this environment. Their reproductive characteristics and resource utilization pattern indicate a strong capacity to colonize and suggested an opportunistic relation to its resources (Lachaise 1979). The yeast populations from flies of Cataliio Point were similar to those reported previously for mango fruits in Brazil (De Oliveira 1980; Joly 1955). These fruits have a high sucrose content (Chan and Kwok 1975). Sucrose is not assimilated by most apiculate yeasts but is by Kloeckera javanica, the prevalent yeast associated with the melanogaster group. The high yeast diversity and predominant Drosophila groups of the forest sites were probably the result of more diversified food sources in forests environments. A higher diversity of food sources is also supported by the physiological profiles of the yeast communities associated with guarani group that predominated at the Pau da Fome site (Table 3). The yeast population associated with a substrate should reflect the carbon sources available in it (Lachance and
Starmer 1986). The similarity of external and crop yeast populations suggested that the flies carried yeasts from their feeding substrates and is in accord with the observations of da Cunha et al. (1957). The lower frequency of yeast isolated from crops may reflect the attraction of starved flies by the baits as suggested by Ganter et al. (1986). The majority of yeasts associated with Drosophila groups in our study were characterized by the most limited assimilative profile of each species. A rapid and vigorous fermentation of glucose, limited assimilative spectrum, and formation of pseudomycelia and ascospores were characteristic of these communities. This physiological profile is characteristic of yeasts associated with fruits and other substrates with high sugar content (Phaff et al. 1978). Wickerham (195 1) considered Hansenula species with low assimilative abilities to be highly evolved in association with plants and insects vectors. Heed et al. (1976), based on data from da Cunha et al. (1957), noted the inability of yeasts from tropical flies to utilize the majority of tested compounds, and this was confirmed by our results. Similar characteristics were also found for yeasts associated with Drosophila of an African savanna (Pignal and Lachaise 1979). Our results differed from those analyzed by Heed et al. (1976) for yeasts associated with tropical Drosophila in having a higher value for most assimilation tests, including sucrose and cellobiose, and lower values for assimilation of lactate and salicin. Salicin is a glucoside commonly found in bark of Salix spp. (Heed et al. 1976) but probably not common in substrates utilized by the flies in our study. The high profiles for sucrose and cellobiose and low profile for lactate assimilation were probably the result of a high proportion of apiculate yeasts. Many of our isolates, especially apiculate yeasts, from Drosophila in Rio de Janeiro exceeded the known temperature limits for growth (Morais et al. 1992). These results differ from the characteristics pointed out previously (da Cunha et al. 1957; Heed et al. 1976) for yeasts of a Brazilian tropical region
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
MORAIS ET AL.
and suggested a selection by the high temperatures frequently encountered in this environment. Most yeasts associated with tropical Drosophila in forests were not unique but were similar to yeast communities associated with fruits and other fermentable substrates. There were differences among yeast communities from different fly groups, and their composition was influenced by the site of collection. There is little data available on substrate use and niche occupation of flies in tropical forests, but the yeast community profile indicated the preferential utilization of fruits and a probable differentiation of diets among groups of flies. Acknowledgements We thank Newton Carlos Marciel Gomes for dedicated technical assistance and M. B. Martins and C. A. C. Andrade for the identification of the flies. This work was supported by the Conselbo Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) and Financiadora de Estudos e Projetos (FINEP). Barnett, J.A., Payne, R.W., and Yarrow, D. 1983. Yeasts, characteristics and identification. Cambridge University Press, Cambridge, Cambridge. Carson, H.L., Knapp, E.P., and Phaff, H.J. 1956. Studies on the ecology of Drosophila in the Yosemite Region of California. 111. The yeast flora of the natural breeding sites of some species of Drosophila. Ecology, 37: 538-544. Chan, H.T., Jr., and Kwok, S.C.M. 1975. Identification and determination of sugars in some tropical fruits. J. Food Sci. 40: 4 19-420.
da Cunha, A.B. 1951. Modifications of the adaptative values of chromosomal types in Drosophila pseudobscura by nutritional variables. Evolution, 5: 395-404. da Cunha, A.B., Shehata, A.M. El Tabey, and Oliveira, W. 1957. A study of the diets and nutritional preferences of tropical species of Drosophila. Ecology, 38: 98-106. De Oliveira, R.B. 1980. Taxonomia de leveduras associadas com cajus e mangas. Masters thesis, Universidade Federal do Rio de Janeiro, Rio de Janeiro. Dobzhansky, Th., and da Cunha, A.B. 1955. Differentiation of nutritional preferences in Brazilian species of Drosophila. Ecology, 36: 34-39. Dobzhansky, Th., Cooper, D.M., Phaff, H.J., Knapp, E.P., and Carson, H.L. 1956. Studies on the ecology of Drosophila in the Yosemite region of California. IV. Differential attraction of species of Drosophila to different species of yeasts. Ecology, 37: 544-550.
Do Carmo-Sousa, L. 1969. Distribution of yeasts in nature. In The yeasts. Vol. 1. Biology of yeasts. Edited by A.H. Rose and J .S. Harrison. Academic Press, London. pp. 79-106. Ganter, P.F., Starmer, W.T., Lachance, M.A., and Phaff, H.J. 1986. Yeast communities from host plants and associated Drosophila in Southern Arizona: new isolations and analysis of the relative importance of host and vectors on community composition. Oecologia, 70: 386-392.
1155
Hagler, A.N., and Mendon~a-Hagler,L.C. 1991. A diazonium Blue B test for yeasts grown three days on yeast carbon base urea agar. Rev. Microbiol. (Sao Paulo), 22: 71-74. Heed, W.B., Starmer, W.T., Miranda, M., Miller, M.W., and Phaff, H. J. 1976. An analysis of the yeast flora associated with cactiphilic Drosophila and their host plants in the Sonoran Desert and its relations to temperate and tropical associations. Ecology, 57: 151-160. Joly, S. 1955. Contribui~aoao estudo de leveduras ocorridas em frutos maduros. Doctoral thesis, Universidade de Siio Paulo, Piracicaba, Siio Paulo. Kreger van Rij, N.J.W. 1984. The yeasts - a taxonomic study. 3rd. ed. Elsevier Science Publishers B. V., Amsterdam. Kurtzman, C.P. 1984. Synonymy of the yeast genera Hansenula and Pichia demonstrated through comparisons of deoxyribonucleic acid relatedness. Antonie van Leeuwenhoek, 50: 209-2 17. Kurtzman, C .P. 199 1. DNA relatedness among saturn-spored yeasts assigned to the genera Williopsis and Pichia. Antonie van Leeuwenhoek, 60: 13-19. Lachaise, D. 1979. Le concept de niche chez les drosophilides. Terre Vie Rev. Ecol. 33: 425-456. Lachaise, D., Pignal, M.C., and Rouault, J. 1979. Yeast flora partitioning by drosophilid species inhabiting a tropical African savanna of the Ivory Coast (Diptera). Ann. Soc. Entomol. Fr. 15: 659-680. Lachance, M.-A., and Starmer, W.T. 1986. The community concept and the problem of non-trivial characterization of yeast communities. Coenoses, 1: 21-28. Morais, P.B., Rosa, C.A., Mendon~a-Hagler,L.C., and Hagler, A.N. 1992. High growth temperatures for apiculate yeasts isolated from Drosophila in Rio de Janeiro. Rev. Microbiol. (Sao Paulo), 23(3). In press. Phaff, H.J., and Starmer, W.T. 1987. Yeasts associated with plants, insects and soils. In The yeasts. Vol. 1. Biology of the yeasts. 2nd ed. Edited by A.H. Rose and J.S. Harrison. Academic Press, New York. Phaff, H.J., Miller, M.W., Recca, J.A., Shifrine, J., and Mrak, E.M. 1956. Studies on the ecology of Drosophila in the Yosemite Region of California. Ecology, 37: 533-538. Phaff, H.J., Miller, M.W., and Mrak, E.M. 1978. The life of yeasts. 2nd ed. Harvard University Press, Cambridge, Mass. Pignal, L.C., and Lachaise, D. 1979. Les levures des drosophilides de savanne d'Afrique intertropicale (Savannes de Lamto, C8te d'Ivoire). Mycopathologia, 68: 155-165. Shehata, A.M.T., Mrak, E.M., and Phaff, H.J. 1955. Yeasts isolated from Drosophila and from their suspected feeding places in Southern and Central California. Mycologia, 47: 799-881. Starmer, W.T. 198 1. A comparison of Drosophila habitats according to the physiological attributes of the associated yeast communities. Evolution, 35: 35-52. Starmer, W.T., Lachance, M.-A., Phaff, H.J., and Heed, W.B. 1990. The biogeography of yeasts associated with decaying cactus tissue in North America, the Caribbean, and Northern Venezuela. Evol. Biol. 24: 253-296. Wickerham, L.J. 1951. Taxonomy of yeasts. U.S. Dep. Agric. Tech. Bull. 1029: 1-56.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
LOUIS
B.
KLACZKO~
Instituto de Biologia, CCS, Bl. A, Ilha d o Fund60, Universidade Federal d o Rio de Janeiro, Rio de Janeiro, Brasil 21941 Received March 2, 1992 Revision received May 11, 1992 Accepted June 4, 1992 MORAIS,P. B., HAGLER,A. N., ROSA, C. A., MENDONCA-HAGLER, L. C., and KLACZKO,L. B. 1992. Yeasts associated with Drosophila in tropical forests of Rio de Janeiro, Brazil. Can. J. Microbiol. 38: 1150-1155. The distribution and diversity of yeast species vectored by and from the crop of eight species groups of Drosophila is described for two rain forest sites and an urban wooded area in Rio de Janeiro, Brazil. The typical forest Drosophila groups guarani, tripunctata, and willistoni showed a higher diversity of yeasts than the cosmopolitan melanogaster species group, suggesting different strategies of utilization of substrates. Apiculate yeasts, including Kloeckera apis, Kloeckera javanica, and Kloeckera japonica, were the prevalent species. Geotrichum spp. and Candida citrea were also frequent isolates in the forest sites. Similarities between the yeasts from the external surfaces and crops of Drosophila suggested that the feeding substrates were the main source of the yeasts vectored by these flies. Most of the yeasts were strong fermenters and assimilated few compounds, usually sucrose, cellobiose, and glycerol. This indicated a preference of the flies for food sources such as fruits. Some yeasts were primarily isolated from one group of Drosophila; for example, Kloeckera javanica from the melanogaster group, Debaryomyces vanrijiae var. yarrowii from the tripunctata group, and Kluyveromyces delphensis from the willistoni group. These associations and differences in the yeast communities among the fly groups suggested a differentiation of diets and specialization of the yeast-Drosophila association in the tropical forests. Key words: Yeast communities, Drosophila, rain forest, tropical climate. MORAIS,P. B., HAGLER,A. N., ROSA, C. A., MENDONCA-HAGLER, L. C., et KLACZKO,L. B. 1992. Yeasts associated with Drosophila in tropical forests of Rio de Janeiro, Brazil. Can. J . Microbiol. 38 : 1150-1 155. La distribution et la diversite des espkces de levures dans un echantillonnage de huit groupes d'espkces de Drosophila vectrices de ces levures sont decrites pour deux sites d'une foret tropicale humide et une region boisee urbaine de Rio de Janeiro, Bresil. Les groupes de Drosophila : guarani, tripunctata et willistoni, typiques de la foret tropicale ont presente une diversite de levures plus elevee que les espkces du groupe melanogaster cosmopolitain, ce qui suggkre l'existence de strategies differentes d'utilisation des substrats. Les levures apiculees, incluant les Kloeckera apis, Kloeckera javanica et Kloeckera japonica, ont ete les espkces predominantes; toutefois, dans les sites forestiers, des espkces du genre Geotrichum et le Candida citrea ont aussi ete des isolats frequents. Les similarites entre les levures, de par leurs surfaces externes, et les jabots de Drosophila ont suggere que les substrats nutritifs etaient les principales sources de levures vChiculCes par ces mouches. La majorite des levures ont ete des agents de fermentation trks efficaces et n'ont assimile que peu de composes, habituellement du sucrose, du cellobiose et du glycerol. Ceci indique une preference des mouches pour des sources alimentaires telles que les fruits. Certaines levures ont ete isolees de groupes particuliers de Drosophila, par exemple le Kloeckera javanica du groupe melanogaster, le Debaryomyces vanrijiae var. yarrowii du groupe tripunctata et le Kluyveromyces delphensis du groupe willistoni. Ces associations et les differences dans les communautes de levures chez les groupes de mouches suggkrent une differenciation dans les diktes et une specialisation dans les associations levures-Drosophila dans les forets tropicales. Mots cles : communautes de levures, Drosophila, foret tropicale, climat tropical. [Traduit par la redaction]
Introduction Drosophila species are important vectors for yeasts in nature, and a mutualistic interaction exists in which the yeasts are a source of essential nutrients for the flies (Phaff and Starmer 1987). Starmer (1981) suggested that the characteristics of yeast communities partially describe the habitats of associated Drosophila. Different yeasts are associated with Drosophila species in tropical and temperate regions. The assimilation of only a few carbon sources 'corresponding author. 2 ~ r e s e n taddress: Departamento de Genetica e Evoluciio, Instituto de Biologia, UNICAMP, Cx. Postal 6109, Campinas, Brazil 13081. Prinled in Canada / lmprirne au Canada
and the intense fermentation of glucose is typical of yeasts isolated from tropical flies, suggesting the use of substrates rich in simple sugars such as fruits in decomposition (Do Carmo-Sousa 1969; Heed et al. 1976; Pignal and Lachaise 1979). The genera Pichia, Hanseniaspora, and Kloeckera predominate among yeasts of flies from a tropical forest in Brazil and Hanseniaspora valbyensis and Saccharomycopsis crataegensis are prevalent in African drosophilids (da Cunha et al. 1957; Pignal and Lachaise 1979). This contrasts with the higher frequency of species of Saccharomyces and Kluyveromyces in temperate regions (Carson et al. 1956; Heed et al. 1976; Phaff et al. 1956; Shehata et al. 1955). Adults of different Drosophila species in temperate regions probably share many feeding sites but
1151
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
MORAIS ET AL.
do not necessarily harbor yeasts from them. In contrast, yeasts of tropical Drosophila appear to be species isolated from the probable feeding substrates of these flies (da Cunha et al. 1957). Yeasts from the intestinal tracts of various species of tropical Drosophila are different from each other, demonstrating food preferences and diversity of diets among species of these flies (Carson et al. 1956; da Cunha 195 1; da Cunha et al. 1957; Dobzhansky and da Cunha 1955). The yeasts isolated from wild Drosophila species differ from those encountered in the intestinal tract of Drosophila melanogaster, a species associated with anthropogenic habitats (Phaff et al. 1978). Few studies have focussed on the interaction between Drosophila and yeasts in forest sites, and only one was done in Brazil. This was carried out in an Atlantic forest area in the state of Siio Paulo (da Cunha et al. 1957). Improvements in yeast taxonomy since then allow more accurate identification of yeasts. The comparison of different tropical forest habitats could help to elucidate the ecological specificity of yeast communities and the niche uses of the associated flies. Yeast communities from a habitat have nutritional requirements corresponding to the characteristics of the environment (Phaff et al. 1978; Lachance and Starmer 1986). Thus, their physiological profiles can be used to study the relationships among different Drosophila species groups occupying the same habitat. In this paper we describe the diversity, distribution, and physiological profiles of yeast communities associated with Drosophila species groups in an urban wooded area and two Atlantic rain forest sites. Materials and methods Fruit flies of the genus Drosophila were collected from three sites near the city of Rio de Janeiro in December 1989 and January, August, and September 1990. Catalfio Point is an abandoned farm area with fruit trees, mostly mango (Mangifera indica). This site is located at the Universidade Federal do Rio de Janeiro campus on Ilha do Fundfio and is influenced by human activities. Collections at Catalso Point were made when mango fruits were ripe. The Mairink Chapel area is within the Tijuca National Park forest, a 122-km2area of Atlantic rain forest. The Pau da Fome site is in a preserved primary rain forest area with restricted access near a suburb west of the city. Insects were attracted by means of baits made of mashed bananas fermented overnight with commercial baker's yeast. Bait was placed in 2-L plastic beakers covered with several layers of sterile gauze to avoid contamination of flies by yeasts from the bait. Three beakers of bait were placed in a triangular pattern 1.5 m apart in the field for 3-5 h during the collection. The flies were aspirated aseptically as they approached or landed on the gauze. Yeasts were isolated from the external parts of 228 adult flies that were allowed to walk 8-12 h on plates containing YMA (0.3% malt extract, 0.3% yeast extract, 0.5% peptone, 1% glucose, 2% agar adjusted to pH 4.0 with 1 M HC1, and supplemented with 100 mg/L of chloramphenicol). These flies were then recovered for identification. Another 286 adult flies were surface sterilized in the field by immersion in 70% ethanol for 1 min and transported to the laboratory within 2 h on ice in tubes containing sterile water. These were identified and crops dissected from them were streaked on YMA. Plates were incubated for 2-7 days at 25°C until colonies appeared. A representative of each type of colony morphology from each plate was brought into pure culture by successive plating on YMA. Yeast isolates were characterized and identified according to standard methods and keys (Barnett et al. 1983; Kreger van Rij 1984) and a DBB test was done as described by Hagler and Mendon~a-Hagler(1991). Multiple cultures identified as the
TABLE1. Number of flies belonging to eight Drosophila species groups in two forest sites and an urban wooded area in Rio de Janeiro Species groups
Sex
Catalao Point
Tijuca forest
Pau da Fome forest
Total
cardini dreyfusi guarani melanogaster repleta saltans tripunctata willistoni
Not groupeda Total NOTE: M, male; F, female. "Flies not identified as belonging to any Drosophila group.
same species from a single sample were considered a single isolate. Statistical analyses used to describe the relative occurrence of yeasts in either fly species groups or habitats were based on Starmer et al. (1990). Diversity indexes of yeasts were calculated for Drosophila and for the three collection sites. Geographic diversity and host diversity were calculated for yeast species that occurred in two or more sites or groups of flies. Physiological profiles for yeast communities were calculated by multiplying the number of isolates of each species by the species physiological capability and dividing by the total number of isolates. The physiological capability was calculated as described in Heed et al. (1976) and Lachance and Starmer (1986). Physiological profiles of the total yeast community and each group of flies and habitat were constructed on compounds that were frequently utilized as in Heed et al. (1976).
Results The 524 flies collected from the three sites were identified as eight groups of species of Drosophila listed in Table 1. The melanogaster species group dominated at Cataliio Point, willistoni and tripunctata groups dominated at the Tijuca forest site, and guarani group dominated at the Pau da Fome forest site. The other Drosophila species groups occurred in low numbers and only in the forests. The frequencies of isolation of 72 yeast species from different Drosophila species groups in the three collection sites are presented in Table 2. Habitat and host diversity indexes for the yeasts are also presented in Table 2. Apiculate yeasts, especially of the species Kloeckera apis, Kloeckera japonica, and Kloeckera javanica, were the most frequent isolates. Saccharomyces cerevisiae was represented by only three isolates, all of which differed from the baker's yeast used in the bait. The bait yeast differed from our isolates by assimilating trehalose but failing to form ascospores, grow at 37"C, ferment maltose, or assimilate galactose. Few strains of Pichia and Williopsis were isolated from Cataliio Point, but Pichia angusta, Pichia beckii, Pichia carsonii,
CAN. J . MICROBIOL. VOL. 38, 1992
1152
TABLE2. Isolation frequencies of yeasts from Drosophila of three sites in Rio de Janeiro Cataliio Point
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Yeasts
Tijuca forest
Pau da Fome forest
Host Geographical Me1 Tri Wil NG Gua Me1 Tri Wil Others NG Gua Tri Wil Others NG diversity" diversityb
A ureobasidiurn pullulans Candida antarctica Candida apicola Candida apis var. galacta Candida berthetii Candida birnundalis var . arnericana Candida boleticola Candida bornbicola Candida citrea 3 Candida colliculosa 2 Candida diversa Candida fructus 1 Candida guillierrnondii Candida hurnicola 1 "Candida hurnicola like" ''Candida kara waiewii like" Candida lambica Candida rnagnoliae 1 Candida rnelinii 2 2 Candida norvegensis 1 "Candida norvegensis like" 1 1 Candida pulcherrirna Candida quercitrusa "Candida sorbosa like" ''Candida tepae like" Candida variabilis ''Candida zeylanoides like" Candida sp. A Candida sp. B Candida sp. C Cryptococcus infirrno-rniniatus Cryptococcus rnacerans 1 Debaryornyces rnelissophilus 2 Debaryornyces vanrijiae var . yarro wii 1 Geotrichurn spp. 1 Hanseniaspora occidentalis 2 ~anseniasporauvarurn 6 2 ~anseniasporavalbyensis 1 1 Hanseniaspora vineae Kloeckera africana 30 10 Kloeckera apis Kloeckera corticis Kloeckera japonica 16 Kloeckera javanica 95 Kluyverornyces delphensis 2 Kluy verornyces rnarxianus var . drosophilarurn Kluy verornyces rnarxianus var . rnarxianus Kluyverornyces therrnotolerans 2 Kluy verornyces wickerharnii Pichia abadiae Pichia acaciae Pichia angusta Pichia anornala Pichia beckii Pichia carsonii Pichia dryadoides Pichia holstii Pichia kluyveri
4 1
4 1 5 2
3
1
3 9
9 4
2
1 2 2 1 1 1 1 2
1
10 5
2
2 4
1
2 1
1
1 20 4 1 3
2
1 1 3
8
3 1 1 3
4
1
6 1
4
2 1 5
5
2
3
1
1 2 1 5 5
2
2
1
3 3
1 1 2 1 1 1 1 1 5 14 5
4
7
3 6
1153
MORAIS ET AL.
TABLE2 (Concluded) Catalfio Point
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Yeasts
Tijuca forest
Pau da Fome forest
Host Geographical Me1 Tri Wil NG Gua Me1 Tri Wil Others NG Gua Tri Wil Others NG diversityo diversityb
Pichia silvicola Pichia toletana Rhodotorula araucariae Rhodotorula aurantiaca Rhodotorula fujisanensis Rhodotorula glutinis Rhodotorula minuta Rhodotorula rubra Saccharomyces cerevisiae Torulaspora pretoriensis Williopsis californica Williopsis saturnus Total NOTE:The number of flies in e a ~ h ' ~ r oatueach ~ site is noted in Table 1. Species names in the genera Pichia and Williopsis are according to Kurtzman (1984 and 1991). Gua, guarani group; Mel, melanogaster group; Tri, tripunctata group; Wil, willktoni group; others, cardini, dreyfusii, repleta, and sakans groups; NG, flies not belonging to any group. Host diversity calculated as 1 - (070 in each group/row total12. Geographic diversity calculated as 1 - (070 in each site/row t o t a ~ ) ~ .
Pichia kluy veri, Williopsis californica, and Williopsis saturnus were represented by multiple isolates at the forest sites. With the exception of Candida citrea and Candida colliculosa, the genus Candida was restricted to the forest sites. Candida apicola, Candida apis, Candida citrea, Candida colliculosa, Candida fmctus, Candida pulcherrima, and Candida quercitrusa were the most frequent isolates of this genus. Geotrichum was frequently isolated from forest sites, especially from the guarani group at Pau da Fome. Kluveromyces delphensis was associated with the willistoni group from which all but 1 of 34 strains were isolated. Debaryomyces vanrijiae var. yarrowii was the predominant yeast associated with the tripunctata group. The black yeast Aureobasidium pullulans was isolated in low numbers from all three sites and Basidiomycete yeasts and their anamorphs were represented by only a few isolates of Candida, Cryptococcus, and Rhodotorula species. Procedures for the isolation of yeasts from Drosophila surfaces yielded similar species to those for isolation of ingested yeasts, although more isolates were obtained from surfaces (557 cultures) than from crops (305 cultures). The most common yeasts from both procedures were Kloeckera apis, Kloeckera japonica, Kloeckera javanica, Candida citrea, and Kluy veromyces delphensis. Debaryomyces vanrijiae var. yarrowii and Pichia beckii were also frequently isolated from the external parts of flies. The willistoni and tripunctata groups had high yeast diversity indexes (0.93 and 0.89, respectively). The forest sites, where these flies were predominant, also showed high yeast diversity indexes (0.86 for Tijuca forest and 0.92 for Pau da Fome forest). The Catalso Point site had the lowest yeast diversity index (0.77) in accord with the abundance of melanogaster flies, which had the lowest diversity of yeasts (0.69). Kloeckera apis was the yeast with the highest host diversity, since it was frequently isolated from all groups of flies except dreyfusii and the crop of tripunctata flies (Table 2). Host diversity was high for Candida apicola, Candida melinii, and Candidapulcherrima, but they were rare in all host flies. Debaryomyces vanrijiae var. yarrowii and Kloeckera javanica showed the lowest values for host diver-
sity. The indexes of geographic diversity were relatively low for all yeasts, the highest values being for Kloeckera apis and Candida melinii and the lowest values for Klockera javanica and Kloeckera africana. The yeast biotypes isolated frequently failed to assimilate most of the carbon sources considered variable in the standard descriptions of their species (Table 3). The assimilative abilities were mostly restricted to a few compounds that included glycerol, D-xylose, sucrose, cellobiose, mannitol, and succinate. Fermentative species made up 84.2% of the isolates, but most of these fermented only glucose although rapidly and vigorously. Some biotypes differed from the standard descriptions of species by growing at higher temperatures than expected. Among the apiculate yeasts, 68 strains of Kloeckera apis grew at 40°C, and 117 strains of Kloeckera javanica and 4 strains of Kloeckera africana grew at 37OC. Some strains of Candida citrea grew at 37OC and strains of Candida berthetii, Candida guilliermondii, Pichia acaciae, and an unidentified Candida species positive for the DBB test grew at 43OC.
Discussion Yeast species are often associated with specific insects or plants (Phaff and Starmer 1987). Feeding preferences of different Drosophila species for different yeasts have been demonstrated (da Cunha 1951; Dobzhansky and da Cunha 1955; Dobzhansky et al. 1956; Lachaise 1979). The low values of host and geographical diversity obtained for the yeasts in the present study are in agreement with the existence of specialized associations of yeasts with Drosophila in tropical forest environments. Although the general characteristics of the yeast communities were similar, some species were associated predominantly with particular fly groups. This suggested that some specific associations occurred between yeast species and Drosophila species groups and that differentiation of diets is common among these groups of flies. Some of the yeasts isolated from Rio de Janeiro were associated with two or more groups of Drosophila. Pichia beckii and Williopsis californica were isolated from the willistoni and tripunctata groups. In the Pau da Fome site, the majority of the yeasts isolated from the guarani group
CAN. J . MICROBIOL. VOL. 38, 1992
1154
TABLE3. Physiological profiles of yeast isolates obtained in forest sites and an urban wooded area in Rio de Janeiro and from flies from temperate, desert, and tropical regions (% utilization of carbon compounds) Site of collectiona
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
Tests
Cataliio
Tijuca
Pau da Fome
Drosophila groups
Tri
Will
Me1
Gua
Tropical fliesc
Tropical fliesd
Temperate fliese
Desert fliese
All known speciesy
Fermentation Glucose Assimilation Galactose Sucrose Maltose Trehalose Cellobiose Salicin D-XY~OS~ Glycerol Mannitol Lactate Succinate "Profile of the yeast species isolated from the three sites of collection. b ~ r o f i l eof the yeast communities isolated from the four most frequent groups of flies: Tri, tripunctata group; Will, willistoni group; Mel, melanogaster group; Gua, guarani group. 'Profile of all species isolated (our data). d ~ r o mHeed et 01. (1976), based on da Cunha et 01. (1957). eFrom Heed et 01. (1976). /From Lachance and Starmer (1986), based on Kreger van Rij (1984).
were also found in the willistoni group. This suggested the use of some common resources by these flies. Similar yeast populations and consequently similar feeding spectra have been noted for different African drosophilids (Lachaise et al. 1979). The willistoni and tripunctata groups had the highest yeast species diversities. This may have resulted from their occurrence in all three sites, in contrast with the anthropogenic habitat-associated melanogaster group that predominated only at Cataliio Point. The high yeast diversity in some fly species groups may indicate their use of various food sources. For the willistoni group this could be an effect of the differentiation of diets among species of the group as pointed out by da Cunha et al. (1957). The melanogaster group presented the lowest yeast diversity, probably reflecting a low yeast diversity at Cataliio Point where it was predominant. Apparently the melanogaster group utilized ripe mango fruits, the most abundant food source in this environment. Their reproductive characteristics and resource utilization pattern indicate a strong capacity to colonize and suggested an opportunistic relation to its resources (Lachaise 1979). The yeast populations from flies of Cataliio Point were similar to those reported previously for mango fruits in Brazil (De Oliveira 1980; Joly 1955). These fruits have a high sucrose content (Chan and Kwok 1975). Sucrose is not assimilated by most apiculate yeasts but is by Kloeckera javanica, the prevalent yeast associated with the melanogaster group. The high yeast diversity and predominant Drosophila groups of the forest sites were probably the result of more diversified food sources in forests environments. A higher diversity of food sources is also supported by the physiological profiles of the yeast communities associated with guarani group that predominated at the Pau da Fome site (Table 3). The yeast population associated with a substrate should reflect the carbon sources available in it (Lachance and
Starmer 1986). The similarity of external and crop yeast populations suggested that the flies carried yeasts from their feeding substrates and is in accord with the observations of da Cunha et al. (1957). The lower frequency of yeast isolated from crops may reflect the attraction of starved flies by the baits as suggested by Ganter et al. (1986). The majority of yeasts associated with Drosophila groups in our study were characterized by the most limited assimilative profile of each species. A rapid and vigorous fermentation of glucose, limited assimilative spectrum, and formation of pseudomycelia and ascospores were characteristic of these communities. This physiological profile is characteristic of yeasts associated with fruits and other substrates with high sugar content (Phaff et al. 1978). Wickerham (195 1) considered Hansenula species with low assimilative abilities to be highly evolved in association with plants and insects vectors. Heed et al. (1976), based on data from da Cunha et al. (1957), noted the inability of yeasts from tropical flies to utilize the majority of tested compounds, and this was confirmed by our results. Similar characteristics were also found for yeasts associated with Drosophila of an African savanna (Pignal and Lachaise 1979). Our results differed from those analyzed by Heed et al. (1976) for yeasts associated with tropical Drosophila in having a higher value for most assimilation tests, including sucrose and cellobiose, and lower values for assimilation of lactate and salicin. Salicin is a glucoside commonly found in bark of Salix spp. (Heed et al. 1976) but probably not common in substrates utilized by the flies in our study. The high profiles for sucrose and cellobiose and low profile for lactate assimilation were probably the result of a high proportion of apiculate yeasts. Many of our isolates, especially apiculate yeasts, from Drosophila in Rio de Janeiro exceeded the known temperature limits for growth (Morais et al. 1992). These results differ from the characteristics pointed out previously (da Cunha et al. 1957; Heed et al. 1976) for yeasts of a Brazilian tropical region
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF MINNESOTA LIBRARIES on 05/08/13 For personal use only.
MORAIS ET AL.
and suggested a selection by the high temperatures frequently encountered in this environment. Most yeasts associated with tropical Drosophila in forests were not unique but were similar to yeast communities associated with fruits and other fermentable substrates. There were differences among yeast communities from different fly groups, and their composition was influenced by the site of collection. There is little data available on substrate use and niche occupation of flies in tropical forests, but the yeast community profile indicated the preferential utilization of fruits and a probable differentiation of diets among groups of flies. Acknowledgements We thank Newton Carlos Marciel Gomes for dedicated technical assistance and M. B. Martins and C. A. C. Andrade for the identification of the flies. This work was supported by the Conselbo Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) and Financiadora de Estudos e Projetos (FINEP). Barnett, J.A., Payne, R.W., and Yarrow, D. 1983. Yeasts, characteristics and identification. Cambridge University Press, Cambridge, Cambridge. Carson, H.L., Knapp, E.P., and Phaff, H.J. 1956. Studies on the ecology of Drosophila in the Yosemite Region of California. 111. The yeast flora of the natural breeding sites of some species of Drosophila. Ecology, 37: 538-544. Chan, H.T., Jr., and Kwok, S.C.M. 1975. Identification and determination of sugars in some tropical fruits. J. Food Sci. 40: 4 19-420.
da Cunha, A.B. 1951. Modifications of the adaptative values of chromosomal types in Drosophila pseudobscura by nutritional variables. Evolution, 5: 395-404. da Cunha, A.B., Shehata, A.M. El Tabey, and Oliveira, W. 1957. A study of the diets and nutritional preferences of tropical species of Drosophila. Ecology, 38: 98-106. De Oliveira, R.B. 1980. Taxonomia de leveduras associadas com cajus e mangas. Masters thesis, Universidade Federal do Rio de Janeiro, Rio de Janeiro. Dobzhansky, Th., and da Cunha, A.B. 1955. Differentiation of nutritional preferences in Brazilian species of Drosophila. Ecology, 36: 34-39. Dobzhansky, Th., Cooper, D.M., Phaff, H.J., Knapp, E.P., and Carson, H.L. 1956. Studies on the ecology of Drosophila in the Yosemite region of California. IV. Differential attraction of species of Drosophila to different species of yeasts. Ecology, 37: 544-550.
Do Carmo-Sousa, L. 1969. Distribution of yeasts in nature. In The yeasts. Vol. 1. Biology of yeasts. Edited by A.H. Rose and J .S. Harrison. Academic Press, London. pp. 79-106. Ganter, P.F., Starmer, W.T., Lachance, M.A., and Phaff, H.J. 1986. Yeast communities from host plants and associated Drosophila in Southern Arizona: new isolations and analysis of the relative importance of host and vectors on community composition. Oecologia, 70: 386-392.
1155
Hagler, A.N., and Mendon~a-Hagler,L.C. 1991. A diazonium Blue B test for yeasts grown three days on yeast carbon base urea agar. Rev. Microbiol. (Sao Paulo), 22: 71-74. Heed, W.B., Starmer, W.T., Miranda, M., Miller, M.W., and Phaff, H. J. 1976. An analysis of the yeast flora associated with cactiphilic Drosophila and their host plants in the Sonoran Desert and its relations to temperate and tropical associations. Ecology, 57: 151-160. Joly, S. 1955. Contribui~aoao estudo de leveduras ocorridas em frutos maduros. Doctoral thesis, Universidade de Siio Paulo, Piracicaba, Siio Paulo. Kreger van Rij, N.J.W. 1984. The yeasts - a taxonomic study. 3rd. ed. Elsevier Science Publishers B. V., Amsterdam. Kurtzman, C.P. 1984. Synonymy of the yeast genera Hansenula and Pichia demonstrated through comparisons of deoxyribonucleic acid relatedness. Antonie van Leeuwenhoek, 50: 209-2 17. Kurtzman, C .P. 199 1. DNA relatedness among saturn-spored yeasts assigned to the genera Williopsis and Pichia. Antonie van Leeuwenhoek, 60: 13-19. Lachaise, D. 1979. Le concept de niche chez les drosophilides. Terre Vie Rev. Ecol. 33: 425-456. Lachaise, D., Pignal, M.C., and Rouault, J. 1979. Yeast flora partitioning by drosophilid species inhabiting a tropical African savanna of the Ivory Coast (Diptera). Ann. Soc. Entomol. Fr. 15: 659-680. Lachance, M.-A., and Starmer, W.T. 1986. The community concept and the problem of non-trivial characterization of yeast communities. Coenoses, 1: 21-28. Morais, P.B., Rosa, C.A., Mendon~a-Hagler,L.C., and Hagler, A.N. 1992. High growth temperatures for apiculate yeasts isolated from Drosophila in Rio de Janeiro. Rev. Microbiol. (Sao Paulo), 23(3). In press. Phaff, H.J., and Starmer, W.T. 1987. Yeasts associated with plants, insects and soils. In The yeasts. Vol. 1. Biology of the yeasts. 2nd ed. Edited by A.H. Rose and J.S. Harrison. Academic Press, New York. Phaff, H.J., Miller, M.W., Recca, J.A., Shifrine, J., and Mrak, E.M. 1956. Studies on the ecology of Drosophila in the Yosemite Region of California. Ecology, 37: 533-538. Phaff, H.J., Miller, M.W., and Mrak, E.M. 1978. The life of yeasts. 2nd ed. Harvard University Press, Cambridge, Mass. Pignal, L.C., and Lachaise, D. 1979. Les levures des drosophilides de savanne d'Afrique intertropicale (Savannes de Lamto, C8te d'Ivoire). Mycopathologia, 68: 155-165. Shehata, A.M.T., Mrak, E.M., and Phaff, H.J. 1955. Yeasts isolated from Drosophila and from their suspected feeding places in Southern and Central California. Mycologia, 47: 799-881. Starmer, W.T. 198 1. A comparison of Drosophila habitats according to the physiological attributes of the associated yeast communities. Evolution, 35: 35-52. Starmer, W.T., Lachance, M.-A., Phaff, H.J., and Heed, W.B. 1990. The biogeography of yeasts associated with decaying cactus tissue in North America, the Caribbean, and Northern Venezuela. Evol. Biol. 24: 253-296. Wickerham, L.J. 1951. Taxonomy of yeasts. U.S. Dep. Agric. Tech. Bull. 1029: 1-56.
