Microbial communities and chemical changes during fermentation of sugary Brazilian kefir

The microorganisms associated with sugary Brazilian kefir beverage were investigated using a combination of culture-dependent and -independent methods. A total of 289 bacteria and 129 yeasts were identified via phenotypic and genotypic methods. Lb. paracasei (23.8%) was the major bacterial isolate identified, followed by Acetobacter lovaniensis (16.31%), Lactobacillus parabuchneri (11.71%), Lactobacillus kefir (10.03%) and Lactococcus lactis (10.03%). Saccharomyces cerevisiae (54.26%) and Kluyveromyces lactis (20.15%) were the most common yeast species isolated. Scanning electron microscopy showed that the microbiota was dominated by lemon-shaped yeast cells growing in close association with Lactobacillus (long and curved). Some lactic acid bacteria detected by sequence analysis of DGGE (denaturing gradient gel electrophoresis) bands were not recovered at any time through fermentation by plating. Conversely, DGGE fingerprints did not reveal bands corresponding to some of the species isolated by culturing methods. The bacteria Acetobacter lovaniensis and the yeast Kazachstania aerobia are described for the first time in sugary kefir. During the 24 h of fermentation, the concentration of lactic acid ranged from 0.2 to 1.80 mg/ml, and that of acetic acid increased from 0.08 to 1.12 mg/ml. The production of ethanol was limited, reaching a final mean value of 1.24 mg/ml.     

Characterization and comparison of microbial community of different typical Chinese liquor Daqus by PCR-DGGE

Aims: To identify and compare microbiota in Chinese liquor Daqu, which were produced in the different regions using different production process. Methods and Results: The DNA exacted from Daqu samples was used as a template for PCR with universal primers of 16S rRNA, 26S rRNA and 18S rRNA, respectively. The amplicons were analysed using denaturing gradient gel electrophoresis (DGGE). It was observed that the bacterial DGGE profile indicated high diversity and predominance of lactic acid bacteria. The results showed that Saccharomycopsis fibuligera and Pichia anomal were dominant yeast species and that several non‐Saccharomyces yeasts including Hanseniaspora guilliermondii, Debaryomyces hansenii, Issatchenkia orientalis and Trichosporon asahii were also detected. As for fungal DGGE, Aspergillus oryzae and Absidia blakesleeana were the most common species amongst different samples. Based on the DGGE analysis, a few differences in community structure were found between Daqu samples. Conclusions: A variety of bacteria, yeast and moulds were identified in Daqu samples, in addition to the present knowledge obtained mainly through the traditional culture‐dependent methods. Moreover, production temperature played a more decisive role on the formation of micro‐organism composition in Daqu than geographical region. Significance and Impact of the Study: PCR–DGGE technique was used in this study to fully observe and asses all microbial community (including bacteria, yeast and mould) in Chinese liquor Daqu for the first time and proved to be effective in profiling Daqu microbial diversity.     

Yeast biodiversity in the guts of several pests on maize; comparison of three methods: classical isolation,cloning and DGGE

The yeast biodiversity in the guts of several pests (Diabrotica virgifera, Helicoverpa armigera, Ostrinia nubilalis) on maize from two isolation sources was assessed by cultivation-dependent and cultivation-independent methods. These yeasts are considered to bear a potentially high biotechnological relevance due to their potential ability to degrade several mycotoxins incorporated by their hosts. The 97 isolated yeast strains showed 21 different partial sequence types of the 26S rRNA gene which could be assigned to 10 different genera. The determined genera and species are discussed in terms of the meaning of their taxonomic status or their occurrence in nature. Two cultivation-independent methods, cloning and DGGE, were compared. We propose the combination of these methods as well as the combination of both cultivation-independent and cultivation-dependent approaches, for gaining better insights into fungal biodiversity. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Bacterial community in the intestine of the sea urchin Strongylocentrotus intermedius during digestion of Macrocystis pyrifera

Denaturing gradient gel electrophoresis (DGGE) was used to study the bacterial community changes in the intestine of the sea urchin Strongylocentrotus intermedius during the digestion of Macrocystis pyrifera. The distinct bands in DGGE gels were sequenced, and the results indicated that the bacterial community in the large and small intestine varied at different periods of digestion. Samples from the large intestine included six specific bands belonging to the genus Psychromonas, whereas samples from the small intestine included eight specific bands representing Psychromonas, Shewanella, Saccharophagus degradans, and Nitrosomonas eutropha. The bacterial flora differed at different periods of digestion. The increase in the microbial community species in the large intestine was not obvious compared with that in the small intestinal microbial community. Several microbes involved in degradation of M. pyrifera were found in the intestine of sea urchin.     

Evaluation of fungal and yeast diversity in Slovakian wine-related microbial communities

Since the yeast flora of Slovakian enology has not previously been investigated by culture-independent methods, this approach was applied to two most common cultivars Frankovka (red wine) and Veltlin (white wine), and complemented by cultivation. Model samples included grapes, initial must, middle fermenting must and must in the end-fermentation phase. The cultured isolates were characterized by length polymorphism of rDNA spacer two region using fluorescence PCR and capillary electrophoresis (f-ITS PCR), and some were identified by sequencing. The microbial DNA extracted directly from the samples without cultivation was analysed by f-ITS PCR, amplicons were cloned and sequenced. The use of universal fungal primers led to detection of both yeasts and filamentous fungi. The amplicon of highest intensity and present in all the samples corresponded to Hanseniaspora uvarum. Other species demonstrated by both approaches included Saccharomyces sp., Metschnikowia pulcherrima or M. chrysoperlae, Candida zemplinina, Cladosporium cladosporioides, Botryotinia fuckeliana, Pichia anomala, Candida railenensis, Cryptococcus magnus, Metschnikowia viticola or Candida kofuensis, Pichia kluyveri or Pichia fermentas, Pichia membranifaciens, Aureobasidium pullulans, Alternaria alternata, Erysiphe necator, Rhodotorula glutinis, Issatchenkia terricola and Debaryomyces hansenii. Endemism of Slovakian enological yeasts was suggested on the level of minor genetic variations of the known species and probably not accounting for novel species. The prevalence of H. uvarum over Saccharomyces sp. in the samples was indicated. This is the first culture-independent study of Slovakian enology and the first time f-ITS PCR profiling was used on wine-related microbial communities.     

Phylogenetic diversity of nitrogen-fixing bacteria in mangrove sediments assessed by PCR-denaturing gradient gel electrophoresis

Culture-independent PCR–denaturing gradient gel electrophoresis (DGGE) was employed to assess the composition of diazotroph species from the sediments of three mangrove ecosystem sites in Sanya, Hainan Island, China. A strategy of removing humic acids prior to DNA extraction was conducted, then total community DNA was extracted using the soil DNA kit successfully for nifH PCR amplification, which simplified the current procedure and resulted in good DGGE profiles. The results revealed a novel nitrogen-fixing bacterial profile and fundamental diazotrophic biodiversity in mangrove sediments, as reflected by the numerous bands present DGGE patterns. Canonical correspondence analysis (CCA) revealed that the sediments organic carbon concentration and available soil potassium accounted for a significant amount of the variability in the nitrogen-fixing bacterial community composition. The predominant DGGE bands were sequenced, yielding 31 different nifH sequences, which were used in phylogenetic reconstructions. Most sequences were from Proteobacteria, e.g. α, γ, β, δ-subdivisions, and characterized by sequences of members of genera Azotobacter, Desulfuromonas, Sphingomonas, Geobacter, Pseudomonas, Bradyrhizobium and Derxia. These results significantly expand our knowledge of the nitrogen-fixing bacterial diversity of the mangrove environment.     

Polycyclic Aromatic Hydrocarbon-Induced Structural Shift of Bacterial Communities in Mangrove Sediment

Mangrove sediment is well known for its susceptibility to anthropogenic pollution, including polycyclic aromatic hydrocarbons (PAHs), but knowledge of the sediment microbial community structure with regards to exposure to PAHs is limited. The study aims to assess the effects of PAHs on the bacterial community of mangrove sediment using both 16s rDNA polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) and traditional enrichment methods. Both the exposure time and the PAH concentration reduced the microbial diversity, as determined by the DGGE bands. Although PAHs could act as carbon sources for microorganisms, PAHs, at a concentration as low as 20 mg l⁻¹, posed a toxic effect to the microbial community. Sequencing of DGGE bands showed that marine bacteria from the genera of Vibrio, Roseobacter, and Ferrimonas were most abundant after PAH exposure, which suggests that both marine and terrestrial bacteria coexisted in the mangrove sediment, but that the marine microbes were more difficult to isolate using the traditional culture method. DGGE determination further demonstrated that the consistency among triplicates of the enriched consortia was significantly less than that of the sediment slurries. The present study reveals that the mangrove sediment microbial structure is susceptible to PAH contamination, and complex microbial community interactions occur in mangrove sediment.     

Yeast communities associated with artisanal mezcal fermentations from <Emphasis Type="Italic">Agave salmiana</Emphasis>

The aims of this work were to characterize the fermentation process of mezcal from San Luis Potosi, México and identify the yeasts present in the fermentation using molecular culture-dependent methods (RFLP of the 5.8S-ITS and sequencing of the D1/D2 domain) and also by using a culture-independent method (DGGE). The alcoholic fermentations of two separate musts obtained from Agave salmiana were analyzed. Sugar, ethanol and major volatile compounds concentrations were higher in the first fermentation, which shows the importance of having a quality standard for raw materials, particularly in the concentration of fructans, in order to produce fermented Agave salmiana must with similar characteristics. One hundred ninety-two (192) different yeast colonies were identified, from those present on WL agar plates, by RFLP analysis of the ITS1-5.8S- ITS2 from the rRNA gene, with restriction endonucleases, HhaI, HaeIII and HinfI. The identified yeasts were: Saccharomyces cerevisiae, Kluyveromyces marxianus, Pichia kluyveri, Zygosaccharomyces bailii, Clavispora lusitaniae, Torulaspora delbrueckii, Candida ethanolica and Saccharomyces exiguus. These identifications were confirmed by sequencing the D1-D2 region of the 26S rRNA gene. With the PCR-DGGE method, bands corresponding to S. cerevisiae, K. marxianus and T. delbrueckii were clearly detected, confirming the results obtained with classic techniques.     

Microbial Diversity of a Camembert-Type Cheese Using Freeze-Dried Tibetan Kefir Coculture as Starter Culture by Culture-Dependent and Culture-Independent Methods

The biochemical changes occurring during cheese ripening are directly and indirectly dependent on the microbial associations of starter cultures. Freeze-dried Tibetan kefir coculture was used as a starter culture in the Camembert-type cheese production for the first time. Therefore, it''s necessary to elucidate the stability, organization and identification of the dominant microbiota presented in the cheese. Bacteria and yeasts were subjected to culture-dependent on selective media and culture-independent polymerase chain reaction (PCR)-denaturing gradient gel electrophoresis (DGGE) analysis and sequencing of dominant bands to assess the microbial structure and dynamics through ripening. In further studies, kefir grains were observed using scanning electron microscopy (SEM) methods. A total of 147 bacteria and 129 yeasts were obtained from the cheese during ripening. Lactobacillus paracasei represents the most commonly identified lactic acid bacteria isolates, with 59 of a total of 147 isolates, followed by Lactococcus lactis (29 isolates). Meanwhile, Kazachstania servazzii (51 isolates) represented the mainly identified yeast isolate, followed by Saccharomyces cerevisiae (40 isolates). However, some lactic acid bacteria detected by sequence analysis of DGGE bands were not recovered by plating. The yeast S. cerevisiae and K. servazzii are described for the first time with kefir starter culture. SEM showed that the microbiota were dominated by a variety of lactobacilli (long and curved) cells growing in close association with a few yeasts in the inner portion of the grain and the short lactobacilli were observed along with yeast cells on the exterior portion. Results indicated that conventional culture method and PCR-DGGE should be combined to describe in maximal detail the microbiological composition in the cheese during ripening. The data could help in the selection of appropriate commercial starters for Camembert-type cheese.     

应用变性梯度凝胶电泳和16SrDNA序列分析对kefir粒中细菌多样性的研究

以开菲尔（Kefir）粒为材料,经过DNA抽提和16SrDNA V3区PCR扩增,扩增产物经变性梯度凝胶电泳（DGGE）分离并切割电泳条带进行序列测定,并与现有的数据库进行了比较,对Kefir粒的细菌多样性进行分析。结果表明,DGGE图谱中可检测到的8条带的16SrDNA基因序列中有7个基因序列与GenBank数据库登录的相关序列的相似性大于98％,余下的1个基因序列的相似性也大于96％。相似性大于98％的7个克隆中,有3个属于鞘氨醇杆菌属（Sphingobacterium）,2个属于乳杆菌属（Lactobacillus）,其它2个分别属于肠杆菌属（Errterobacter）和不动杆菌属（Acinetobacter）。首次报道了鞘氨醇杆菌作为优势菌群存在开菲尔Kefir粒中。     

Yeast Microflora Isolated From Brazilian Cassava Roots: Taxonomical Classification Based on Molecular Identification

A rapid molecular identification technique was applied on microbial microflora isolated from Brazilian cassava roots given a yeast profile presented in the samples analyzed. A total of 24 strain isolated from cassava were initially grouped and identified in five groups using restriction-fragment length polymorphism (RFLPs) of 5.8S-ITS rDNA region. Sequencing analysis of the domains D1 and D2 of the 26S rRNA gene or 5.8S rRNA-ITS region were used to identify different groups of yeasts. Representative colonies of yeasts of each group were isolated and identified as Debaromyces hansenii, Kodamaea ohmeri, Candida glabrata, C. haemulonii, and Pichia gullhermondii. It is hoped that these results will contribute toward selecting yeast from this microflora capable to degrade cassava starch in the near future.     

Community Structure of Actively Growing Bacterial Populations in Plant Pathogen Suppressive Soil

The bacterial community in soil was screened by using various molecular approaches for bacterial populations that were activated upon addition of different supplements. Plasmodiophora brassicae spores, chitin, sodium acetate, and cabbage plants were added to activate specific bacterial populations as an aid in screening for novel antagonists to plant pathogens. DNA from growing bacteria was specifically extracted from the soil by bromodeoxyuridine immunocapture. The captured DNA was fingerprinted by terminal restriction fragment length polymorphism (T-RFLP). The composition of the dominant bacterial community was also analyzed directly by T-RFLP and by denaturing gradient gel electrophoresis (DGGE). After chitin addition to the soil, some bacterial populations increased dramatically and became dominant both in the total and in the actively growing community. Some of the emerging bands on DGGE gels from chitin-amended soil were sequenced and found to be similar to known chitin-degrading genera such as Oerskovia, Kitasatospora, and Streptomyces species. Some of these sequences could be matched to specific terminal restriction fragments on the T-RFLP output. After addition of Plasmodiophora spores, an increase in specific Pseudomonads could be observed with Pseudomonas-specific primers for DGGE. These results demonstrate the utility of microbiomics, or a combination of molecular approaches, for investigating the composition of complex microbial communities in soil.     

Molecular Characterization of Fecal Microbiota in Patients with Viral Diarrhea

The study provides molecular analyses of fecal microbiota of diarrhea patients infected with four different types of viruses. Fecal specimens from 52 patients with viral diarrhea (13 each of adenovirus, norovirus, rotavirus, and astrovirus) and six healthy individuals were collected and etiological viral agent was confirmed by enzyme immunoassay and specific PCR. To assess the changes in microbial diversity in patients with viral diarrhea, DNA from stool were extracted and characterized by PCR-denaturing gradient gel electrophoresis (DGGE) with universal primers specific for the V3 region of 16S rRNA gene. The strongest bands of the DGGE profiling were excised and sequenced to identify the dominant groups. Bacteroides vulgatus, Bifidobacterium, and Lactobacillus genera were also enumerated by real time PCR. The results revealed that bacterial diversity and similarity in feces from viral diarrhea groups were significantly lower (mean H′/ H _max^￠ H_{ \max }^{\prime } 0.89–0.94, 29–43, respectively) as compared with those of healthy individuals (mean H′/ H _max^￠ H_{ \max }^{\prime } 1.36, 59, respectively). Sequencing of dominant bands affirmed that diarrhea groups were mainly comprised of phylum Firmicutes, such as genera Enterococcus, Peptostreptococcaceae incertae sedi, Streptococcus, Weissella, and Clostridium, and opportunistically pathogenic genus Shigella, while dominant group in healthy individuals was phylum Bacteroidetes. Copy number of Bacteroides vulgatus, Bifidobacterium, and Lactobacillus genera was also reduced significantly in viral diarrhea groups as compared to healthy group. It is concluded that opportunistic pathogens increases, while other species of commensal microbiota decrease significantly in the viral diarrhea patients and dysbacteriosis is dependent on type of virus infection.     

应用变性梯度凝胶电泳和16S rDNA序列分析对kefir粒中细菌多样性的研究

以开菲尔(Kefir)粒为材料,经过DNA抽提和16S rDNA V3区PCR扩增,扩增产物经变性梯度凝胶电泳(DGGE)分离并切割电泳条带进行序列测定,并与现有的数据库进行了比较,对Kefir粒的细菌多样性进行分析。结果表明,DGGE图谱中可检测到的8条带的16S rDNA基因序列中有7个基因序列与GenBank数据库登录的相关序列的相似性大于98%,余下的1个基因序列的相似性也大于96%。相似性大于98%的7个克隆中,有3个属于鞘氨醇杆菌属(Sphingobacterium),2个属于乳杆菌属(Lactobacillus),其它2个分别属于肠杆菌属(Enterobacter)和不动杆菌属(Acinetobacter)。首次报道了鞘氨醇杆菌作为优势菌群存在开菲尔Kefir粒中。     

Changes in Community Structure of Sediment Bacteria Along the Florida Coastal Everglades Marsh–Mangrove–Seagrass Salinity Gradient

Community structure of sediment bacteria in the Everglades freshwater marsh, fringing mangrove forest, and Florida Bay seagrass meadows were described based on polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) patterns of 16S rRNA gene fragments and by sequencing analysis of DGGE bands. The DGGE patterns were correlated with the environmental variables by means of canonical correspondence analysis. There was no significant trend in the Shannon–Weiner index among the sediment samples along the salinity gradient. However, cluster analysis based on DGGE patterns revealed that the bacterial community structure differed according to sites. Not only were these salinity/vegetation regions distinct but the sediment bacteria communities were consistently different along the gradient from freshwater marsh, mangrove forest, eastern-central Florida Bay, and western Florida Bay. Actinobacteria- and Bacteroidetes/Chlorobi-like DNA sequences were amplified throughout all sampling sites. More Chloroflexi and members of candidate division WS3 were found in freshwater marsh and mangrove forest sites than in seagrass sites. The appearance of candidate division OP8-like DNA sequences in mangrove sites distinguished these communities from those of freshwater marsh. The seagrass sites were characterized by reduced presence of bands belonging to Chloroflexi with increased presence of those bands related to Cyanobacteria, γ-Proteobacteria, Spirochetes, and Planctomycetes. This included the sulfate-reducing bacteria, which are prevalent in marine environments. Clearly, bacterial communities in the sediment were different along the gradient, which can be explained mainly by the differences in salinity and total phosphorus.     

Denaturing gradient gel electrophoresis (DGGE) analysis of bacterial community composition in deep-sea sediments of the South China Sea

The South China Sea, which is one of the largest marginal seas in the world, is predicted to have suitable accumulation conditions and exporting prospects for natural gas hydrate. The aim of this study was to explore the bacterial community composition of deep-sea sediments from such an ecosystem. DNA was extracted by five different methods and used as templates for PCR amplification of the V3 regions of the 16S rRNA gene. Denaturing gradient gel electrophoresis (DGGE) was used to separate the amplified products and analyse the 16S rRNA gene diversity of sediment samples. The results of DGGE indicated that the bacterial community composition is influenced by DNA extraction methods. Sequencing dominant bands demonstrated that the major phylogenetic groups identified by DGGE belong to Proteobacteria, Bacteroidetes, gram-positive bacteria and Archaea. Integrating different DNA extraction procedures are needed to analyse the actual bacterial diversity from environment when the amplification of 16S rRNA gene and construction of representative clone library were adopted.     

The Development and Stability of the Genus <Emphasis Type="Italic">Bacteriodes</Emphasis> from Human Gut Microbiota in HFA Mice Model

Human flora-associated (HFA) mice are frequently applied in studying the ecology and metabolism of human gut microbiota. However, the development and stability of the genus Bacteriodes, a prominent bacteria group of human gut microbiota, in HFA mice have not yet fully been examined. In this study, PCR-denaturing gradient gel electrophoresis (DGGE) analysis was employed to monitor the Bacteriodes community in the fecal microbiota of six HFA Kunming mice during a period of 3 weeks. Based on the DGGE banding patterns, the majority of prominent bands in the HFA mice DGGE profile were also typical bands in the human DGGE profile, despite the absence of three bands (corresponding to two different B. thetaiotaomicron strains and one B. intestinalis strain) from the human DGGE profile. The Dice coefficient of similarity for the fecal microbiota of HFA mice in comparison to the human donor sample ranged between 74 ± 6% and 81 ± 7%. The phylogeny of bands in the DGGE profile showed that the dominant Bacteriodes species in the fecal microbiota of HFA mice were B. thetaiotaomicron, representing 66.7% of all bands. Our results indicate that the genus Bacteriodes in the fecal microbiota of HFA mice was selected from the human donor and could remain relatively stable over time.     

PCR-based dENATURING GRADIENT GEL ELECTROPHORETICtypification and identification of the microbial consortium present in Kefir grains

Kefir grains have a complex microbiological composition that makes it difficult to obtain an optimal and constant starter culture necessary for the production of a quality Kefir beverage. The microbes present in the grains have in the past been identified using traditional methods such as growth on selective media and morphological and physiological characteristics. The aim of this study was to typify and identify the complex microbial community present in mass cultured, traditionally cultured and Irish Kefir grains by PCR amplification of a variable part of the ribosomal RNA (rRNA) genes in Eubacteria and yeasts and resolving these PCR fragments by denaturing gradient gel electrophoresis (DGGE). Unique PCR-based DGGE fingerprints were obtained for the Eubacterial and yeast species present in the three different grain types. A part of the Eubacterial and yeast rRNA genes were sequenced and compared to sequences available on NCBI. The phylogenetic relatedness of the amplified and sequenced lactobacilli was determined. Different bands in the Eubacterial and the yeast DGGE profiles of the mass cultured grains were identified to species level. A DGGE marker was constructed providing a quick reference for the identification of the members of the Eubacterial microbial population in mass cultured Kefir grains.     

Analysis of the fungi community in multiple- and single-grains <Emphasis Type="Italic">Zaopei</Emphasis> from a <Emphasis Type="Italic">Luzhou</Emphasis>-<Emphasis Type="Italic">flavor</Emphasis> liquor distillery in western China

The fungi communities of both multiple-grains and single-grains Zaopei were analyzed by denaturing gradient gel electrophoresis (DGGE) and traditional identification methods. The results of the DGGE fingerprint and the dendrogram based on banding patterns showed the diversified community structure and the phylogenetic affiliation of different Zaopei samples. The results indicated that the fungi community was affected by both raw materials and fermentation location. The genera of Debaryomyces, Pichia and Candida were dominant communities in multiple-grains Zaopei suggested by the results of both DGGE and the traditional methods, the DGGE result suggested Candida dominant in single-grains Zaopei was much different from the results by traditional method. Additionally, DGGE results showed the existence of thermophilic fungi (Thermomyces lanuginosus and Thermoascus aurantiacus) which were not detected by the traditional method. This work may contribute to further understanding of the brewing in Chinese Luzhou-flavor liquor.     

The costs and benefits of killer toxin production by the yeast Pichia kluyveri

Numerous yeast species in many genera are able to produce and excrete extracellular toxic proteins (mycocins) that can kill other specific sensitive yeasts. Natural distributions of killer yeasts suggest that they may be important in maintaining community composition and provide a benefit to the toxin producing cells. The fact that not all yeasts are killers and that polymorphisms exist within some killer species suggests there may be a cost associated with killer toxin production. This study focuses on the costs and benefits associated with toxin production by the yeast Pichia kluyveri. Strains differing in their ability to kill were obtained by tetrad dissection. One parent strain produced spores that exhibited a trade-off between killing ability and intrinsic growth rate. A killer clone from this strain was able to maintain a higher proportion of cells than a non-killer when grown with the same sensitive yeast under laboratory-simulated natural conditions. On the other hand, when grown with a yeast not sensitive to Pichia kluyveri toxin, the non-killer maintained a higher proportion of the total community than did the killer clone. The data support the hypothesis that there are both costs and benefits to producing killer toxin, and based on this, selection may favor different phenotypes in different conditions.