ebook img

Bacteria and Archaea diversity within the hot springs of Lake Magadi and Little Magadi in Kenya PDF

12 Pages·2016·2.46 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Bacteria and Archaea diversity within the hot springs of Lake Magadi and Little Magadi in Kenya

Kamburaetal.BMCMicrobiology (2016) 16:136 DOI10.1186/s12866-016-0748-x RESEARCH ARTICLE Open Access Bacteria and Archaea diversity within the hot springs of Lake Magadi and Little Magadi in Kenya Anne Kelly Kambura1, Romano Kachiuru Mwirichia2, Remmy Wekesa Kasili1, Edward Nderitu Karanja3, Huxley Mae Makonde4 and Hamadi Iddi Boga5* Abstract Background: Lake Magadi and little Magadi are hypersaline, alkaline lakes situated inthe southern part of Kenyan Rift Valley. Solutes are supplied mainly by a series of alkaline hot springs with temperatures as high as 86 °C. Previous culture-dependent and culture-independent studies have revealed diverse groups of microorganisms thriving under theseconditions. Previous culture independent studies were based on theanalysis of 16S rDNA but were done onless saline lakes. For thefirst time, this study combinedillumina sequencing and analysis of amplicons ofboth total community rDNA and 16S rRNA cDNA to determine thediversity and community structure ofbacteria and archaea within 3hot springs of L.Magadi and littleMagadi. Methods: Water, wet sediments and microbial mats were collectedfrom springs inthemain lake ata temperature of45.1 °C and from Little Magadi “Nasikie eng’ida”(temperature of 81 °Cand 83.6 °C). Total community DNA and RNA were extracted from samples using phenol-chloroform and Trizol RNA extractionprotocols respectively.The 16S rRNA gene variable region (V4– V7)of the extracted DNA and RNA were amplified and library construction performed following Illumina sequencing protocol. Sequenceswere analyzeddoneusing QIIME while calculation of Bray-Curtis dissimilarities between datasets,hierarchicalclustering, NonMetric Dimensional Scaling (NMDS) redundancy analysis (RDA) and diversity indiceswere carried out using theRprogramming language and the Vegan package. Results: Three thousand four hundred twenty-six and one thousand nine hundred thirteen OTUs were recovered from 16S rDNA and 16S rRNA cDNA respectively. Uncultured diversity accounted for 89.35 %16S rDNA and 87.61 % 16S rRNA cDNA reads. The most abundant phyla in both the 16S rDNA and 16S rRNA cDNA datasetsincluded: Proteobacteria (8.33–50 %), Firmicutes 3.52–28.92%, Bacteroidetes (3.45–26.44%), Actinobacteria (0.98–28.57 %) and Euryarchaeota (3.55–34.48 %)in allsamples. NMDS analyses of taxonomic composition clusteredthe taxa intothree groups according to sample types (i.e. wet sediments, mats and water samples) with evident overlap of clusters between wet sediments and microbial mats from thethreesampletypes in both DNA and cDNA datasets. The hot spring (45.1 °C) containedless diverse populations compared to those in LittleMagadi (81–83 °C). Conclusion: There were significant differences inmicrobialcommunitystructure at95 % level ofconfidence for both total diversity (Pvalue, 0.009) based on 16S rDNA analysis and active microbial diversity (P value, 0.01) based on16S rRNA cDNA analysis,within thethree hot springs. Differences inmicrobial composition and structure were observed as a function ofsampletype and temperature, with wet sediments harboring the highest diversity. Keywords: Lake Magadi, Hotsprings, 16S rDNA, 16S rRNAcDNA, Microbial diversity *Correspondence:[email protected] 5TaitaTavetaUniversityCollege,P.O.Box635-80300,Voi,Kenya Fulllistofauthorinformationisavailableattheendofthearticle ©2016TheAuthor(s).OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.0 InternationalLicense(http://creativecommons.org/licenses/by/4.0/),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinkto theCreativeCommonslicense,andindicateifchangesweremade.TheCreativeCommonsPublicDomainDedicationwaiver (http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,unlessotherwisestated. Kamburaetal.BMCMicrobiology (2016) 16:136 Page2of12 Background the total versus active microbial communities within the Extreme environment refers to any setting that exhibits hotspringsofLakeMagadiandLittleMagadiinKenya. life conditions detrimental to living organisms with re- spect to its physicochemical properties such as pH, Methods temperature, pressure, nutrient and saline concentration Researchauthorization [1]. Extreme physicochemical parameters include acidity Research authorization was obtained from National (pH<5), alkalinity (pH>9), hyper salinity (salinity >35 %), Commission for Science, Technology and Innovation pressure (>0.1 MPa), high temperature (>40 °C), low (NACOSTI)on30thAugust2013inKenya,andpermission temperature (<5 °C), water stress (aw<0.80), and high- to conduct research in Lake Magadi was obtained from radiation environments [2]. The extreme environments KenyaWildlifeServices(KWS)on24thSeptember2013. are inhabited by organisms referred to as extremophiles thataresowell-adaptedthattheyreadilygrowand multi- Studysite ply[3].InKenya,thehaloalkalinesodalakesarecharacter- Lake Magadi is a hypersaline lake that lies in a naturally ized by exceptionally rich productivity rates presumably formed closed lake basin within the Southern part of the becauseofthehighambienttemperatures,highlightinten- Kenyan Rift Valley. It is approximately 2° 00′ 0″ S and 36° sities, availability of phosphates and unlimited access to 00′ 0″ E of the Equator at an elevation of about 600 m CO inthesecarbonaterichwaters[4,5].Salinitylevelscan above sea level [17]. The solutes are supplied mainly by a 2 be as high as 30 % to saturation in Lake Magadi, whereas seriesofalkalinespringswithtemperaturesashighas86°C thepHrangesbetween9and11.5[6].InLakeMagadi,sol- whicharelocatedaroundtheperimeterofthelake.Samples utesaresuppliedmainlybyaseriesofalkalinespringswith analyzedinthisstudywerecollectedfromthreehotsprings: temperaturesvaryingfrom33°Cto86°C[6,7]. onehotspringwithinthemainL.Magadi(02°00′3.7″S36° Previous culture dependent and culture independent 14′ 32″ E) at 45.1 °C and pH 9.8; and two hot springs studiesonLakeMagadihaverevealedadenseanddiverse withinLittleMagadi“Nasikieeng’ida”(01°43′28″S36°16′ population of aerobic, organotropic, halophilic, alkaliphi- 21″E), and (01° 43′ 56″ S 36° 17′ 11″ E) at elevations of lic, and haloalkaliphilic and alkalitolerant representatives 611 and 616 m, temperatures of 81 and 83.6 °C and pH of major bacterial phyla [8–15]. Although conventional rangeof9.2and9.4respectively(Table1). microbial cultivation methods have helped shape under- standing of physiology and metabolic functions of diverse Measurementsofphysicochemicalparameters organisms, they are laborious, time consuming, selective Geographicalpositionofeachsiteintermsoflatitude,lon- and biased for specific microbial growth. On the other gitude and elevation was taken using Global Positioning hand,culture-independent studiesdoneonsodalakesin System (GARMIN eTrex 20). The pH for each sampling Kenyahavebeenbasedontheanalysisofclonelibrariesof point was measured with a portable pH-meter (Oakton PCRamplifiedrDNA.Thismaynotrepresentanaccurate pH 110, Eutech Instruments Pty. Ltd) and confirmed with picture of prokaryotic diversity withina given community indicator strips (Merck, range 5–10). Temperature, Elec- due to low speed and coverage of a cloning and Sanger- trical Conductivity (EC), Total dissolved solids (TDS) and sequencing based approach, which gives a lower number dissolved oxygen (DO) were measured on site using Elec- ofampliconsequencescomparedtothemillionsofgener- trical Chemical Analyzer (Jenway - 3405) during sampling. ated by High Throughput Sequencing technologies such In situ temperature was recorded once for each study site asIlluminaSequencing[16].Thisisthefirstcultureinde- andassignedtoallthethreesampletypesforthatsite. pendent studyofthemicrobialcommunitywithinthehot springs located around the hypersaline Lakes Magadi and Samplecollection Little Magadi. This study employed Illumina Sequencing All samples were collected randomly in triplicates from ofPCRproductsof both16SrDNAand16SrRNAcDNA each hot spring. Water samples were collected using toobtainalessbiasedestimationofmicrobialcommunity sterile 500 ml plastic containers that had been cleaned within the hot springs’ ecosystem. The main objective of with 20 % sodium hypochlorite and UV-sterilized for this study was to analyze the targeted total community one hour. Wet sediments were collected by scooping rDNAandcDNAgeneratedfromrRNAsoastocompare with sterilized hand shovel into sterile 50 ml Falcon Table1Parametersofsamplingsitesanalyzedinthisstudy Parameter Latitude°S Longitude°E Elevation(m) Temperature°C pH EC(mS/cm) TDS(mg/l) DissolvedOxygen(mg/l) Hotspring1 02°00′3.7″ 36°14′32″ 603 45.1 9.8 0.03 1 12.4 Hotspring2 01°43′28″ 36°16′21″ 611 83.6 9.4 1 1 0.04 Hotspring3 01°43′56″ 36°17′11″ 616 81 9.2 1 1 0.71 Kamburaetal.BMCMicrobiology (2016) 16:136 Page3of12 tubes. The upper 5 mm from each microbial mat devel- Caporasoetal.[21].Amplificationproceededina30cycle oping on the hot spring water margins was collected PCR using the HotStarTaq Plus Master Mix Kit (Qiagen, into sterile500mlplasticjamjars. Allsampleswere pre- USA)withinitialdenaturationheatingat94°Cfor3 min, served on dry ice immediately after sampling, and trans- followedby28cyclesofdenaturationat94°Cfor30s,an- ported to the laboratory in Jomo Kenyatta University of nealingat53°Cfor40sandextensionat72°Cfor1min, Agriculture and Technology. Water for nucleic acid ex- and a final elongation at 72 °C for 5 min. The quality of traction (500 ml) was filtered through a 0.22 μM filter PCRproductswasassessedon2%agarosegeltodetermine membrane (Whatman) and all filter papers containing the success of amplification and the relative intensity of samples were stored at −80 °C. Pellets were obtained bands. Multiple samples, tagged with different barcodes, from water samples by re-suspending the filter papers in were pooled in equimolar ratios based on their DNA con- phosphate buffer solution, and centrifuging 5 ml of the centrationsfromthegelimages.Pooledsampleswerepuri- suspension at 13,000 rpm for 10 min. These were used fied using calibrated AmpureXPbeads (Beckman Coulter) fornucleicacidextraction. foruseinlibrarypreparation.ThepooledandpurifiedPCR products were used to prepare 16S rDNA and cDNA li- Nucleicacidextraction brary by following Illumina TruSeq DNA library prepar- Total community DNA was extracted from all the sam- ationprotocol[22].SequencingwasperformedatMRDNA ples in triplicates; pellets from water samples, 0.2 g of (www.mrdnalab.com, Shallowater, TX, USA) on a MiSeq sediment samples and 0.4 g of microbial mat samples, as 2x300bpVersion3followingthemanufacturer’sguidelines. described by as described by Sambrook et al. [18]. Total RNA was extracted from 0.25 g of sediment and mat Sequenceanalysis,taxonomicclassificationanddata samples, and pellets obtained from the water samples submission (describedabove),intriplicatesusingTrizolRNAextrac- Sequences obtained from the Illumina sequencing plat- tion protocol [19]. The respective nucleic acids extracted form were depleted of barcodes and primers using a from triplicate samples were pooled during the precipi- proprietary pipeline (www.mrdnalab.com, MR DNA, tation stage, the pellets were air dried and stored at Shallowater, TX) developed at the service provider’s la- −20 °C. The pellets were lyophilized to protect them boratory. Low quality sequences were identified by from degradation. denoising and filtered out of the dataset [23]. Sequences which were<200 base pairs after phred20- based quality SynthesisofcDNAfrom16SrRNA trimming, sequences with ambiguous base calls, and cDNA synthesis, amplification and sequencing were per- those with homopolymer runs exceeding 6 bp were re- formedatMolecularResearchDNALab(www.mrdnalab.- moved. Sequences were analyzed by a script optimized com,Shallowater,TX,USA).ThequalityoftotalRNAwas for high-throughput data to identify potential chimeras assessed using gel electrophoresis. The extracted RNA in the sequence files, and all definite chimeras were de- was dissolved in RNase-free water and subsequently pleted as described previously [24]. All this data filtering treated to remove DNA contaminants using the Amplifi- was done by the service provider using their pipeline. De cation Grade DNase I Kit (Sigma, MO) according to novo OTU clustering was done with standard UCLUST manufacturer’s instructions. cDNA first-strand and method using the default settings as implemented in second-strand synthesis was done using the Superscript QIIME pipeline Version 1.8.0 at 97 % similarity level. III First-Strand Synthesis SuperMix (Invitrogen, CA) and Taxonomy was assigned to each OTU using BLASTn the Second-strand cDNA Synthesis Kit (BeyoTime, against SILVA SSU Reference 119 database at default e- Jiangsu, China), respectively, following manufacturer’s in- value threshold of 0.001 in QIIME [25, 26]. Obtained se- structions.Single-strandreversetranscriptionwasdoneto quences were submitted to the NCBI Sequence Read provide template for amplicon libraries using Superscript Archive with SRP# Study accessions: SRP061805. These III (Invitrogen) according to the manufacturer’s protocol, included SRX1124606: RNA-Seq of Prokaryotes: Alka- random hexamer primed and with subsequent RNAse H line Hot springs and SRX1124607: DNA-Seq of Prokary- digestion. The Double stranded cDNA synthesis was car- otes: Alkaline Hot springs (Additional file 1: Table S1, riedoutasdescribedbyasdescribedbyUrichetal.[20]. Additional file 2: Table S2, Additional file 3: Table S3 andAdditionalfile4:TableS4). Ampliconlibrarypreparationandsequencing PCR amplification of the 16S rRNA geneV4 variable re- Statisticalanalysis gionwascarriedoutfromextractedDNAandcDNAgen- Diversity indices (Shannon, Simpson and Evenness) for erated from rRNA, using bacteria/archaeal primers 515 F each sample were calculated using vegan package ver- (GTGCCAGCMGCCGCGGTAA) that had barcode and sion 1.16-32 in R software version 3.1.3 [27]. Commu- 806R (GGACTACHVGGGTWTCTAAT) according to nity and Environmental distances were compared using Kamburaetal.BMCMicrobiology (2016) 16:136 Page4of12 Analysis of similarity (ANOSIM) test, based upon Bray- pH ranged from 9.2 to 9.8. The TDS was above the Curtis distance measurements with 999 permutations. measurement range for the Electrical Chemical Significance was determined at 95 % confidence interval Analyzer; hence all the readings appeared as one (1). (p=0.05). Calculation of Bray-Curtis dissimilarities be- The metadata collected before sampling is summarized tween datasets, hierarchical clustering, Non Metric Di- inTable1. mensional Scaling (NMDS), redundancy analysis (RDA) and parameter correlation were carried out using the R Compositionanddiversityofthemicrobialcommunities programming language [27] and theVegan package [28]. After denoising and demultiplexing, a total of 271,345 To support OTU-based analysis, taxonomic groups were and 214,663 sequence reads were generated from 16S derived from the number of reads assigned to each rDNA and 16S rRNA cDNA data respectively. Total taxonatallranksfromdomaintogenususingthetaxa_- OTU richness at 3 % distance amounted to 3502 and summary.txt output from QIIMEpipelineVersion1.8.0. 1915 OTUs respectively. 85 and 62 OTUs were shared acrossallhot springswhile82and45OTUswereshared Results and discussion across all sample types in the two data sets respectively. Sampling Figure 1a and b shows the distribution of OTUs across Three hot springs of Lake Magadi and Little Magadi hot springs and sample types. For 16S rDNA data, hot were selected based on different temperature and pH spring83.6°Cand45.1°Cshowedarelativelylargerover- levels. Temperatures ranged from 45.1 to 83.6 °C while lap (142 OTUs) than other hot springs, while 45.1 °C Fig.1aandbVenndiagramsshowingthedistributionofuniqueandsharedOTUswithinvarioussampletypesinthethreesamplingsites.The numberofOTUsineachhotspringisindicatedintherespectivecircle.SiteA,BandCrepresenthotsprings83.6°C,81°Cand45.1°Crespectively Kamburaetal.BMCMicrobiology (2016) 16:136 Page5of12 harbored most OTUs unique to one hot spring in both (3.03 %), Peptococcaceae (3.03 %) and uncultured gamma datasets. proteobacterium (3.03 %) in various samples as shown in Additionalfile6:TableS6. Bacterialtaxonomiccompositionanalysis Differences in relative abundance were seen as a func- 16S rDNA OTUs comprised of Bacteria (85.55 %) and tionofsampletypeandtemperature,withwet sediments Archaea (11.27 %) while 16S rRNA cDNA OTUs com- harboring the highest taxa. The dominant taxa corre- prised of Bacteria (83.22 %) and Archaea (13.95 %). sponded with those reported in previous studies con- These results suggest that bacteria are the most domin- ducted on deep sea and marine sediments community ant taxa inL. Magadi and Little Magadi hot springs. The composition [29–31]. For example, a review by Brown et groups with highest relative abundances at phylum level al. [30] on microbial life in extreme environments that belonged to members of Proteobacteria (16–31.4 %), compared metagenome analyses of different high ther- Firmicutes(2.8–28.9%),Bacteroidetes(3.4–23.8%), Acti- mal habitats, observed that microbes adapted to these nobacteria (1.0–14.4 %), Cyanobacteria (1.1–8.3 %, habitats, are different with respect to species abundance Chloroflexi (0.3–7.2 %), and Deinococcus-thermus (0.3– and community structure. However, some bacterial taxa 5.3 %) (Fig. 2). Other groups scoring high relative abun- such as Thermotoga, Deinococcus-Thermus and Proteo- dance in some samples include Gemmatimonadetes bacteria,were commonwithin the samples underreview (24.1 %) in water samples at 81 °C, Planctomycetes [30]. These bacterial taxa were also found within the (15.9 %) in water samples at 83.6 °C, Lentisphaerae samples from the hot springs of Lakes Magadi and Little 5.3 % in mat samples at 81 °C, Spirochaetae 4.5 % in wet Magadi. Species diversity in high temperature environ- sediment samples at 45.1 °C,Thermotogae (1.7 %) in wet ments has been shown to be relatively low [32, 33]. The sediment samples at83.6 °CandVerumicrobia (3.9%)in deep-sea hydrothermal vent chimneys have been found mat samplesat 45.1°C(Additionalfile5:Table S5). to harbor Proteobacteria [34, 35], Bacteroidetes and At family level, OTUs were distributed in 292 bacterial Planctomycetes[36]. families with the most abundant belonging to Rhodobac- teraceae that scored up to 9.43 % relative abundance in Archaealtaxonomiccompositionanalysis wet sediment samples at 45.1 °C, Cyanobacteria Family I The OTUs were distributed among three Archaeal phyla; (5.89 %), Desulfonatronaceae (4.55 %), Thermaceae Euryarchaeota(1.76–34.48%)acrossallsamples,Crenarch- (4.01%),Ectothiorhodospiraceae(3.71%),Spirochaetaceae aeota(upto2.46%)andThaumarchaeota(upto2.78%)in (3.34 %), Nitriliruptoraceae (3.19 %), Anaerolineaceae wetsedimentsamplesat81°C(Figs.2and3).Atthefamily Fig.2RelativeabundanceofthemostpredominantphylainvarioussamplescollectedfromthehotspringsofL.MagadiandLittleMagadi. ‘Others’representalltaxathatscoredarelativeabundanceofbelow1%acrossallsamplesinbothdatasets Kamburaetal.BMCMicrobiology (2016) 16:136 Page6of12 Fig.3RelativeabundanceofthemostabundantprokaryotictaxaatorderlevelinsamplesfromthehotspringsofL.MagadiandLittleMagadi. ‘Others’representalltaxathatscoredarelativeabundanceofbelow1%acrossallsamplesinbothdatasets level, OTUs were distributed in 24 families with Activediversitytaxonomiccompositionbased16SrRNA the most abundant belonging to Halobacteriaceae cDNAanalysis (27.06 %) in water samples at 81 °C, Deep Sea Hydro- Bacterialtaxonomiccomposition thermalVentGp6(DHVEG-6)(0.35–5.15%)acrossall The 1913 OTUs of 16S rRNA cDNA data set were domi- samples, South African Goldmine Gp (SAGMEG); Un- nated by bacterial phyla comprising Proteobacteria (8.3– cultured Archaeon (4.01 %) and Methanocaldococca- 50 %) and Actinobacteria (2.0–28.6 %) across all samples. ceae (2.47 %) in wet sediment samples at 81 °C The data confirm the predominance of previously defined (Additional file 6: Table S6). Crenarchaeota phyla groupsbelongingtoProteobacteria[38].Othergroupsscor- members identified belonged to the families Desulfuro- inghighrelativeabundanceinsomesamplesincludeBacter- coccaceae (0.93 %), Thermoproteaceae (0.93 %), Ther- oidetes (26.4 % in wet sediment samples at 45.1 °C), mofilaceae (0.61 %) in wet sediment samples at 81 °C Chloroflexi (5.3 % in water samples at 45.1 °C), Firmicutes and Sulfolobaceae (0.42 %) in wet sediment samples at (23.9 % in mat samples at 45.1 °C), Cyanobacteria (8 % in 83.6 °C, while Thaumarchaeota were mainly assigned matsamplesat45.1°C),Gemmatimonadetes(25%inwater to uncultured archaeon with up to 1.68 % relative samples at 81 °C), Lentisphaerae (7.1 % in wet sediment abundance in wet sediment samples at 83.6 °C samples at 83.6 °C) and Planctomycetes (6.6 % in mat sam- (Additional file 6: Table S6). Previous studies on ther- plesat83.6°C)(Fig.2). mal groundwater in a thermal field in Russia showed Unlike 16S rDNA dataset, the most abundant bacterial that Archaea is dominated by a novel division in the familiesfoundwithincDNAdatasetbelongedtoMethylo- phylum Euryarchaeota related to the order Thermo- philaceae;scoringupto8.3%relativeabundanceinsome plasmatales (39 % of all archaea) and by another abun- samples, Corynebacteriaceae, Dermatophilaceae, Micro- dant group (33 % of all archaea) related to the phylum coccaceae, Nocardioidaceae, Bacteroidaceae; ML635J-40 Crenarchaeota. Both groups are widely spread in hot aquatic group, Sphingomonadaceae, Burkholderiaceae, springs all over the world [37]. Some Archaeal taxa and Myxococcales; 0319-6G20, Alcanivoracaceae, Pseudo- such as Methanococcus,Thermoprotei and Thermococ- monadaceae (7.14 %), Desulfuromonadales; GR-WP33-58 cus were also common within the samples under re- (5.11 %), Cyanobacteria; Subsection III; Family I, Syntro- view by Brown et al. [30]. These are similar to the phomonadaceae(4.7%),Spirochaetaceae(3.7%), Anaero- classes obtained in this study, indicating that Archaea lineaceae(3.44%),Gemmatimonadetes;BD2-11terrestrial are well adapted to extreme conditions and could be group; uncultured bacterium 3.3 %, Rhodospirillaceae responsible for various functional processes within the (3.3%)andLentisphaerae(ML1228J-2;unculturedbacter- ecosystem. ium)(2.6%)(Additionalfile7:TableS7). Kamburaetal.BMCMicrobiology (2016) 16:136 Page7of12 Archaealtaxonomiccomposition taxonomic level of both 16S rDNA and 16 rRNA cDNA 16S rRNA cDNA OTUs were distributed among three indicated that significant differences between samples Archaeal phyla, similar to those obtained from the 16S based on alpha diversity indices whose values obtained rDNA dataset. These were Euryarchaeota (6.6–28.4 %) were as follows: Shannon diversity index (H’); wet sedi- across all samples, Crenarchaeota (up to 1.38 %) in ment 83.6 °C (7.9 vs 3.8), water 81 °C (9.1 vs 3.6) and water samples at 81 °C and Thaumarchaeota (up to Simpson (1/D); wet sediment 83.6 °C (34.54 vs 12.2); 1.15 %) in wet sediment samples at 45.1 °C (Fig. 3). The water 81 °C (7.76 vs 6) and water 45.1 °C (28.53 vs 10.3) most abundant families belonged to Halobacteriaceae respectively (Table2). (21.31 %), Deep Sea Hydrothermal Vent Gp 6 (DHVEG- Total microbial diversity based on 16S rDNA and 16S 6) (8.59 %), and Marine Benthic Group D and DHVEG-1 rRNA cDNA ANOSIM at order level showed that there (2.29 %). Crenarchaeota phyla members identified weresignificantdifferencesinmicrobialcommunitystruc- belonged to the families Desulfurococcaceae (1.11 %) in ture in the samples at 95 % level of confidence (P value, wet sediment samples at 81 °C and Sulfolobaceae 0.009), and 0.383 R statistic value while active microbial (0.28 %) in wet water samples at 45.1 °C, while Thau- diversity based on 16S rRNA cDNA had (P value, 0.01), marchaeota were mainly assigned to uncultured and0.333Rstatisticvalue.Thiscouldbeattributedtodif- archaeon with up to 1.15 % relative abundance in wet ferencesintemperatureandpHofthespecificsitesduring sediment samples at 45.1 °C (Additional file 7: Table sampling.Samplesfrom45.1°Charboredmorecloselyre- S7). Previously, several 16S rRNA gene sequences re- lated populations because it had less extreme conditions lated to novel Archaea (Euryarchaeota) were retrieved ascomparedtothetwootherhotsprings. from the alkaline saltern at Lake Magadi [13]. Distance based redundancy analysis showed that the mi- Haloalkaliphilic Archaea related to Natronomonas, crobial community evenness significantly differed from Natrialba, Natronolimnobius and Halorubrum spp. each site for both 16S rDNA and 16S rRNA cDNA. 16S have also been isolated from Lake Magadi and Inner rDNA ordination of the three sample types showed a sig- Mongolian soda lakes [39]. nificance of 0.017 and while 16S rRNA cDNA dataset showed a significance of 0. 011. Samples from each site Microbialrichnessanddiversityindices clusteredclosetoeachotherinseparatequadrantsindicat- Using rarefaction, the same number of sequences from inghighbetadiversitybetweenthethreesamplingsites. each sample was used in comparison of community NMDS analyses supported by OTU and taxonomic alpha and beta diversity measures. Paired t-tests at class composition, divide the datasets into three ellipses: one Table2DiversityindicescomputedonallOTU-basedmicrobialtaxonomicunitswithin16SrDNAand16SrRNAcDNAdatasets SampleID Dataset No.ofsequences No.ofOTUs Shannon(H’) Simpson(1/D) Evenness Wetsediment(83.6°C) 16SrDNA 38753 238 5.5 34.54 0.658 16SrRNAcDNA 1625 14 2.6 12.2 0.975 Mats(83.6°C) 16SrDNA 7575 66 4.2 27.41 0.475 16SrRNAcDNA 21451 212 5.3 32 0.632 Water(83.6°C) 16SrDNA 19186 680 6.5 14.88 0.356 16SrRNAcDNA 5124 151 5.0 18.1 0.594 Wetsediment(81°C) 16SrDNA 36047 324 5.8 22.57 0.427 16SrRNAcDNA 67235 361 5.9 53.6 0.636 Mats(81°C) 16SrDNA 24252 282 5.6 34.55 0.61 16SrRNAcDNA 33751 349 5.8 34.3 0.538 Water(81°C) 16SrDNA 36126 568 6.3 7.76 0.278 16SrRNAcDNA 638 12 2.5 6 0.866 Wetsediment(45.1°C) 16SrDNA 52115 509 6.2 35.09 0.577 16SrRNAcDNA 11471 87 4.4 27.7 0.784 Mats(45.1°C) 16SrDNA 37446 382 6.0 22.45 0.785 16SrRNAcDNA 48578 375 6.0 37.4 0.572 Water(45.1°C) 16SrDNA 19789 377 6.0 28.53 0.461 16SrRNAcDNA 24790 352 5.9 10.3 0.346 Kamburaetal.BMCMicrobiology (2016) 16:136 Page8of12 Fig.4aNon-metricmultidimensionalscaling(NMDS)basedonBray-Curtisdissimilaritiesbetweenmicrobialcompositionsof16SrDNAdataset, groupedaccordingtosamplingsites.SiteA,BandCrepresenthotsprings83.6ºC,81ºCand45.1ºCrespectively.Theboxes1-3representwet sediments,matsandwatersamplesfrom83.6ºC,4-6representwetsediments,matsandwatersamplesfrom81ºCand7-9representwetsed- iments,matsandwatersamplesfrom45.1ºC.bNon-metricmultidimensionalscaling(NMDS)basedonBray-Curtisdissimilaritiesbetweenmicro- bialcompositionsof16SrRNAcDNAdatasetsgroupedaccordingtosamplingsites.SiteA,BandCrepresenthotsprings83.6ºC,81ºCand45.1 ºCrespectively.Theboxes1-3representwetsediments,matsandwatersamplesfrom83.6ºC,4-6representwetsediments,matsandwater samplesfrom81ºCand7-9representwetsediments,matsandwatersamplesfrom45.1ºC for each hot spring. Some microbial taxa were shared be- into six well-separated habitats with relatively few OTUs tween habitats in both 16S rDNA and 16S rRNA cDNA that were shared between more than one or two habitats derived datasets. This scenario was more pronounced in [41]. The taxa were also observed to cluster according to 16S rDNA - derived dataset, indicating that DNA pool sampletypes(i.e.wetsediments,microbialmatsandwater containeda“seedbank”ofinactiveandsporulatingorgan- samples). There was an overlap of taxonomic clusters be- isms [40], while fewer taxa were active within the ecosys- tween wet sediments and microbial mats from the three tem as shown in the 16S rRNA cDNA derived dataset sampletypesinboth16SrDNAand16SrRNAcDNAde- (Fig. 4). Similar results were observed in a study on Ethi- rived datasets. However, water samples formed separate opian soda lakes where NMDS analyses supported both clusters from the other two sample types in both 16S byOTUandtaxonomiccomposition;dividedthedatasets rDNAand16SrRNAcDNAdatasets(Figs.4and5). Fig.5aNon-metricmultidimensionalscaling(NMDS)basedonBray-Curtisdissimilaritiesbetweenmicrobialcompositionsof16SrDNAdataset- groupedaccordingtosampletypes.Eachellipserepresentsasetofthethreesampletypescollectedfromeachhotspring.Theboxes1-3repre- sentwetsediments,matsandwatersamplesfrom83.6ºC,4-6representwetsediments,matsandwatersamplesfrom81ºCand7-9 representwetsediments,matsandwatersamplesfrom45.1ºC.bNon-metricmultidimensionalscaling(NMDS)basedonBray-Curtisdissimilarities betweenmicrobialcompositionsof16SrRNAcDNAdatasetgroupedaccordingtosampletypes.Eachellipserepresentsasetofthethreesample typescollectedfromeachhotspring.Theboxes1-3representwetsediments,matsandwatersamplesfrom83.6ºC,4-6representwetsedi- ments,matsandwatersamplesfrom81ºCand7-9representwetsediments,matsandwatersamplesfrom45.1ºC Kamburaetal.BMCMicrobiology (2016) 16:136 Page9of12 Fig.6aHierarchicalclusteringof16SrDNAsamplescollectedfromthethreehotspringsofL.MagadiandLittleMagadi.Phylumlevelwaschosentobe usedinhierarchicalclusteringtoassesstherelationshipsbetweensamplesandtaxa.“Sed”representwetsedimentsamplesfromrespectivetemperature.b Hierarchicalclusteringof16SrRNAcDNAsamplescollectedfromthethreehotspringsofL.MagadiandLittleMagadi.Phylumlevelwaschosentobe usedinhierarchicalclusteringtoassesstherelationshipsbetweensamplesandtaxa.“Sed”representwetsedimentsamplesfromrespectivetemperature Kamburaetal.BMCMicrobiology (2016) 16:136 Page10of12 Hierarchical clustering between samples collected andMethanomicrobiahavebeenreportedinoilfields,while from lake Magadi and Little Magadi revealed samples Halobacteria and Thermoprotei have been reported in pet- from the two hot springs in Little Magadi “Nasikie roleum reservoirs [45]. Some Halobacteria members are eng’ida” were closer than samples from the hot spring in important in organic fertilizer production industry as Lig- the main lake (Fig. 6 (a and b) and (Additional file 8: nindecomposers[46].However,mostofthegeneraidenti- FigureS1andAdditionalfile9:FigureS2). fiedinthisstudyareknowntobeheterotrophsresponsible fortheprimarydegradationoforganicmatter[39].Theac- Conclusion tualfunctionofmicrobialtaxareportedinthisstudycould The combined findings of this study, show that estimated further be explored and established using culture diversity and richness within the hot spring samples were dependentmethodsaswellasmRNAtranscripts. foundtobeashighas thosefoundinotherenvironments In conclusion, this study presented microbial diversity such as soil and deep-sea hydrothermal environments analysis of samples collected from the hot springs of L. [42].Theresultsconfirmthatdifferentgroupsofmicroor- Magadi and Little Magadi based on both DNA and RNA, ganisms have the capacity to adapt and thrive even in the usingIllumina SequencingTechnology. The results showed most hostile environments. Some of these groups (Acido- comparable profiles of microbial community using 16S bacteria; Blastocatella, Bryobacter and Telmatobacter rDNAand16SrRNAcDNAderiveddatasets,henceindicat- genera, Bacteroidetes; Bacteroidales, Rhodothermaceae, ingthattheobserveddiversityisreal.Thefindingsshoweda Flavobacteriaceae, Sphingobacteriales and Chloroflexi; broad microbial distribution with water from the spring at Dehalococcoidales and Thermomicrobia) also obtained 83.6°Cfoundtobetherichestsample,constituting680ob- from previous similar studieswerereportedtohavea fer- served species. Despite the fact that the sampling environ- mentative ability [41]. It was observed that from the ment is multi-extreme due to high pH, temperature, and cDNA dataset, photosynthetic taxa were represented by salinity,thisstudyshowsthattherearestableandactivemi- Cyanobacterial genera Leptolyngbya and Lyngbya, among crobialcommunitiesthathaveadaptedtothisenvironment. other uncultured groups. Primary production within the Culturedependentstudiesinfuturewillhelpusunravelthe hotspringsisprobablysupportedbysomegroupsofnon- survivalmechanismsusedbythesepolyextremophiles. sulfur purple bacteria from the family Rhodobacteraceae (specificallythegeneraRoseobacterthatscored1.1%rela- tive abundance),and purple sulfur bacteria from the fam- Additional files ily Ectothiorhodospiraceae present across samples at different relative abundance. The presence of Planctomy- Additionalfile1:TableS1.Totaldiversityenvironmentaldatasheetof thesamplescollectedfromthehotspringsofL.MagadiandLittle cetes within samples could be an indicator that anaerobic Magadi(XLS25kb) ammonium oxidation may be another metabolic pathway Additionalfile2:TableS2.Activediversityenvironmentaldatasheet supporting primary production in the low-oxygen, saline ofthesamplescollectedfromthehotspringsofL.MagadiandLittle environment, since the dissolved oxygen concentration of Magadi(XLS24kb) the sampling sites ranged between 0.04 and 12.4 mg/l. Additionalfile3:TableS3.TotaldiversitySequenceReadArchive metadatasheetofthesamplescollectedfromthehotspringsofL. Actinobacteria and Firmicutes are believed to have adap- MagadiandLittleMagadi(XLSX50kb) tiveadvantageunderlow-nutrientconditionsofthehighly Additionalfile4:TableS4.ActivediversitySequenceReadArchive alkaline, saline hot springs hence their high relative metadatasheetofthesamplescollectedfromthehotspringsofL. abundance levels. The presence of sulfate reducers in MagadiandLittleMagadi.(XLSX50kb) the family Desulfohalobiaceae (mainly Desulfonatrono- Additionalfile5:TableS5.Relativeabundanceofmicrobialtaxaat phylumlevelinvarioussamplescollectedfromthethreehotsprings vibrio), suggested an internal sulfur cycle within the (83.6,81and45.1°C)ofL.MagadiandLittleMagadi.‘Noblasthit/Others’ lake, as previously suggestedforKhadin,Tuva,Russiaand representunclassifiedsequencesinbothdatasets.(XLSX17kb) Natron soda lakes [42, 43]. Taxa typical for highly special- Additionalfile6:TableS6.Totaldiversityrelativeabundanceof ized metabolisms that were encountered in this study in- microbialtaxaatfamilylevelinvarioussamplescollectedfromthethree hotsprings(83.6,81and45.1°C)ofL.MagadiandLittleMagadi.‘Noblast clude Nitriliiruptor, known for their ability to catabolize hit/Others’representunclassifiedsequencesinbothdatasets.(XLSX45kb) nitriles or cyanides [44] and heterotrophic Oceanospirilla- Additionalfile7:TableS7.Activediversityrelativeabundanceof ceae (Marinospirillum). Other functional taxa encountered microbialtaxaatfamilylevelinvarioussamplescollectedfromthethree include aerobic heterotrophs (e.g. Bacteroidetes, Marini- hotsprings(83.6,81and45.1°C)ofL.MagadiandLittleMagadi.‘Noblast hit/Others’representunclassifiedsequencesinbothdatasets.(XLSX32kb) cella) and fermentative anaerobes such as Thermoplasma- Additionalfile8:FigureS1.Comparativeanalysis(UPGMAsimilarity tales among other uncultured groups. Euryarchaeota tree)oftotalmicrobialdiversityofvarioussampletypeswithinhot members were clustered into the classes, Halobacteria, springsofL.MagadiandLittleMagadi.(DOCX55kb) Methanobacteria, Methanomicrobia, Methanococci, Ther- Additionalfile9:FigureS2.Comparativeanalysis(UPGMAsimilaritytree) mococci and Thermoplasmata while Crenarchaeota phyla ofactivemicrobialdiversityofvarioussampletypeswithinhotspringsofL. MagadiandLittleMagadi.(DOCX56kb) comprisedThermoproteiclass.Previously,Methanobacteria

Description:
Total community DNA and RNA were extracted from samples using phenol-chloroform and Trizol RNA extraction protocols respectively. The 16S
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.