Identification of novel Archaea in bacterioplankton of a boreal forest lake by phylogenetic analysis

12 Pages·2000·0.82 MB·English

FEMSMicrobiologyEcology34(2000)45^56
www.fems-microbiology.org
Identi¢cation of novel Archaea in bacterioplankton of a boreal
forest lake by phylogenetic analysis and £uorescent in situ
hybridization1
German Jurgens a;*, Frank-Oliver Glo«ckner b, Rudolf Amann b, Aimo Saano a, D
o
Leone Montonen a, Markit Likolammi c, Uwe Mu«nster c wn
lo
a
d
a DepartmentofAppliedChemistryandMicrobiology,DivisionofMicrobiology,UniversityofHelsinki,P.O.Box56,Biocenter1A,Viikinkaari9, e
d
FIN-00014Helsinki,Finland fro
b Max-Planck-InstituteforMarineMicrobiology,Celsiusstrasse1,D-28359Bremen,Germany m
c LammiBiologicalStation,UniversityofHelsinki,FIN-16900Lammi,Finland h
ttp
s
Received24March2000; receivedinrevisedform7July2000; accepted11August2000 ://a
c
a
d
e
m
Abstract ic
.o
u
p
We report here on novel groups of Archaea in the bacterioplankton of a small boreal forest lake studied by the culture-independent .c
o
analysisofthe16SrRNAgenesamplifieddirectlyfromlakewaterincombinationwithfluorescentinsituhybridization(FISH).Polymerase m
chainreactionproductswereclonedand28ofthe160Archaeacloneswitharound900-bp-long16SrRNAgeneinserts,weresequenced. /fe
m
Phylogeneticanalysis,including642Archaeasequences,confirmedthatnoneofthefreshwatercloneswerecloselyaffiliatedwithknown s
e
culturedArchaea.TwelveArchaeasequences fromlakeValkeaKotinen(VAL)belongedtoGroupIofuncultivatedCrenarchaeotaand c/a
affiliatedwithenvironmentalsequencesfromfreshwatersediments,ricerootsandsoilaswellaswithsequencesfromananaerobicdigestor. rtic
EightoftheCrenarchaeotaVALclonesformedatightcluster.SixteensequencesbelongedtoEuryarchaeota.Fouroftheseformedacluster le
-a
togetherwithenvironmentalsequencesfromfreshwatersedimentsandpeatbogswithintheorderMethanomicrobiales.Fivewereaffiliated b
s
withsequencesfrommarinesedimentssituatedclosetomarineGroupIIandthreeformedanovelclusterVALIIIdistantlyrelatedtothe tra
c
orderThermoplasmales.Theremainingfourclonesformedadistinctcladewithinaphylogeneticradiationcharacterizedbymembersofthe t/3
ordersMethanosarcinalesandMethanomicrobialesonthesamebranchasriceclusterI,detectedrecentlyonricerootsandinanoxicbulk 4/1
soil of flooded rice microcosms. FISH with specifically designed rRNA-targeted oligonucleotide probes revealed the presence of /4
5
Methanomicrobiales inthestudiedlake. Theseobservations indicatea new ecologicalniche formany novel‘non-extreme’ environmental /6
1
Archaea inthe pelagic water of a boreal forest lake. (cid:223) 2000 Federation of European Microbiological Societies. Published by Elsevier 74
6
Science B.V. All rights reserved. 8
b
y
g
Keywords: Archaea; Bacterioplankton; Borealforestlake; 16SrRNA; Phylogeny; Fluorescentinsituhybridization u
e
s
t o
n
0
4
1. Introduction soil, sediments and water play an essential role in this A
p
scenario by their contribution to CO2 and CH4 cycling ril 2
Freshwaters and wetlands in boreal environments have and £uxes [4]. In addition to their global role in biogeo- 01
9
been identi¢ed as important sources for radiatively rele- chemical cycles [5], prokaryotes play a key role in the
vant trace gases [1,2] and can contribute by up to 34% transformation of natural organic matter, which in many
from total global wetlands CH £uxes [3]. Prokaryotes in borealfreshwatersisofprimarilyallochthonousoriginand
4
recalcitrant[6,7].However,itisstilllittleknownaboutthe
community structure and function of indigenous microor-
* Correspondingauthor.Tel.: [phone]; ganisms in boreal freshwaters [7]. Except for the abun-
Fax: [phone]; E-mail: german.jurgens@helsinki.¢ dance and the role of methanogens in some boreal lakes
[8,9] microorganisms in boreal freshwater are lumped to-
1 ThesequencesofenvironmentalArchaeafromfreshwaterSSUrDNA
gether and still treated as a black box. For a better under-
VAL clones obtained in this study have been submitted to the DDBJ/
standing of indigenous bacterioplankton in boreal fresh-
EMBL/GenBank database under accession numbers AJ131263^
AJ131278andAJ131311^AJ131322. waters, we are interested in the development of new
0168-6496/00/$20.00(cid:223)2000FederationofEuropeanMicrobiologicalSocieties.PublishedbyElsevierScienceB.V.Allrightsreserved.
PII: S0168-6496(00)00073-8
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
46 G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56
molecular tools and approaches to follow their role in some of these new probes. In the future we intend to
nutrient cycling, biodegradation and microbial food web follow their fate and role in the microbial food web of
interactions. Recently, we reported the phylogenetic posi- this lake and to place them in a phylogenetic and func-
tions of novel groups of Archaea obtained from boreal tional context.
forests in southern Finland [10,11]. There may be phylo-
genetic and functional links between the boreal terrestrial
and aquatic environments and we are now searching for 2. Materials and methods
similararchaealgroupsin boreallake waterbydeveloping
phylogenetic probes in order to trace them under in situ 2.1. Basic physico-chemical and biological aspects in
conditions in their speci¢c habitats. Valkea Kotinen
Most phylogenetic studies on freshwater Archaea have
so far been made from sediment samples. It has been Valkea Kotinen is a typical small boreal lake located in
shown that some of the 16S rDNA sequences of Archaea the Evo forest area in southern Finland (61‡24PN,
D
belong to di¡erent sub-clusters of Group I uncultured 24‡07PE). It has a water surface area of 3.6 ha, a catch- o
w
Crenarchaeota [12,13]. Others were a⁄liated with the ment/surfaceratioof8.3,amaximumdepthof7mandan nlo
a
orders Methanosarcinales [14] and Methanomicrobiales average depth of 3 m. d
e
[A15rc,1h6a]e.aOhnalvyeabefeenwdnesocnr-icbueldtubreyd(cid:210)ivnrdeiafigseneotuasl.p[l9a]nkintotnhice anTdhcihsemmeiscoahlustmraicti¢focraetsiotnlapkaethtearsnstywpiictha,lsoenasaovnearalgthe,eram1a^l d from
meromictic lake Saelenvannet (Norway) with a salinity 2-m-deep epilimnion, a 0.5^1-m-deep metalimnion and a h
ttp
gradient in the chemocline. deeper (3^4 m deep) anoxic hypolimnion, from early May s
In addition to polymerase chain reaction (PCR)-based until late September. The thermal stability of the water ://a
c
a
techniques, £uorescent in situ hybridization (FISH) with column can vary from May until September between 5 d
e
rRNA-targeted £uorescently labeled oligonucleotide and 25 J m32 (Birgean wind work), and 2 and 18 J m32 m
ic
probes has been widely used to study microbial commun- (Schmidtthermalstability),respectively[21],withveryfew .o
u
p
ities in various environments [17,18]. However, the num- mixing events during the summer stagnation [22]. The .c
o
ber of detected Archaea cells often remained below the water was slightly acid with a pH=4.9^6.2, a low bu¡er m
detection limit (i.e. 1% of the cells stained by DAPI capacity (alkalinity 60.2 meq m33) and a watercolor of /fem
(4P,6-diamidino-2-phenylindole)) when e.g. the microbial 100^250 g platinum units m33. The total organic carbon se
c
community composition of the Wadden Sea sediments varied between 7.8 and 17.4 g C m33 and the DOC be- /a
from the German North Sea coast were investigated [19]. tween 7.5 and 16.5 g C m33 indicating a high level of rtic
le
In a study on seasonal variation in the community DOC sources, which were mainly of allochthonous origin. -a
b
s
structure and cell morphology of pelagic prokaryotes in Inorganic nutrients were low and the phosphorus and ni- tra
a high mountain lake in Austria, it has been shown that trogen resources were available primarily as organic re- ct/3
Archaea represented from less than 1% to a maximum of sources. Phytoplankton biomass (Chl-a) varied between 4
/1
5% of all DAPI-stained cells and reached their highest 2.2 and 75.4 mg m33 with a major contribution from /4
5
densities in September and November. A distinct maxi- Raphidiophyceae[21].Gonyostomumsemenisthedominant /6
1
7
mum of a ¢lamentous fraction of Archaea was observed phytoplankton speciesin this lakefrom mid-July untillate 4
6
during September and October, which was probably an September, which covers on average 40^90% of all the 8 b
y
e¡ect of grazing pressure by phagotrophs [18]. All these Chl-a [23]. The ratio of net primary production to com- g
u
investigations were made by using probe ARCH915 [20], munity respiration (P:R) was on average 1.1:1 and in late e
s
whichisspeci¢cforallArchaea.Onlyafewattemptshave summer and autumn decreased even below 1, which is a t o
n
yet been made to develop phylogenetic trees of indigenous majore¡ectofbacterioplanktonactivityonthewholelake 0
4
planktonic Archaea in boreal freshwater and to use them carbon metabolism [21]. A
p
as a basis for probe design in FISH. ril 2
In our studies on the phylogenetic relatedness and dis- 2.2. Sampling and nucleic acid extraction 01
9
tribution of Archaea in the bacterioplankton of the boreal
forest lake Valkea Kotinen we used two basic molecular Lake water was taken with a Limnos sampler from the
approaches: phylogenetic analysis of the ampli¢ed 16S surface (0.1 m) down to the deepest (7 m) water layers in
rRNA gene sequences and FISH. In this paper we report 0.5-m intervals from the pelagial zone.
the phylogenetic placement of new freshwater clones For microbial DNA extraction, isolation and puri¢ca-
among numerous recently discovered environmental 16S tion, integrated water samples were taken from the pela-
rDNA Archaea sequences. New sequences could be used gial zone in July and storedin pre-cleaned, sterileNalgene
todesignprobes,suitableforthestudyoftheseindigenous bottles in the dark on crushed ice prior to ¢ltration. The
Archaea under in situ conditions. We show here that it cells from 250 ml of water were then collected on a 0.2-
is possible to detect the abundance, cell size and morpho- Wm-pore-size polycarbonate (hydrophilic) ¢lter (diameter,
type of novel Archaea in the plankton by FISH, using 47 mm; Cyclopore1 Track Etched Membrane, Whatman,
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56 47
UK) and immediately frozen at 325‡C until nucleic acid 1-m depths), the metalimnion (1^3-m depths) and the hy-
extraction. polimnion (from 3 to 7 m, near the bottom sediment).
After 3 days storage at 325‡C, extraction began with Water samples were taken in 0.5-m intervals with a 1-l
the addition of 5 ml of lysis bu¡er (1 mg ml31 lysozyme, Limnos sampler in the pelagial of the lake at the deepest
40 mM EDTA, 50 mM Tris^HCl, 0.75 M sucrose, point. Lake water samples from the epi-, meta- and hypo-
pH=8.0) and incubation at 37‡C for 30 min [24]. Then limnetic sub-samples were pooled and stored in sterile 5-l
proteinase K (0.5 mg ml31) and sodium dodecyl sulfate Nalgene bottles immersed in crushed ice during sampling
(1% SDS w/v) were added and the ¢lter was incubated at and transportation to the laboratory. Within 2^3 h after
55‡C for 2 h. The lysate was recovered from the ¢lter to a samplingthemicrobialcellswereconcentratedfrom15^20
fresh tube. The ¢lter was rinsed with an additional 2 ml ml of water on a 0.2-Wm-pore-size polycarbonate ¢lter
lysis bu¡er, incubated for 10 min at 55‡C and the lysates (diameter, 47 mm; Cyclopore1 Track Etched Membrane,
were pooled. To the pooled lysate, 5 M NaCl (¢nal con- Whatman,UK)byvacuum¢ltrationat 6100Pm32.The
centration0.7M)and hexadecyltrimethyl ammoniumbro- ¢lters were prepared and ¢xed for FISH as described pre-
D
mide (¢nal concentration 1% w/v in the presence of 0.7 M viously [26,27] and stored at 325‡C until further process- o
w
NaCl) wereadded.Themixturewasincubatedat65‡Cfor ing. nlo
a
20 min and then extracted with chloroform^isoamyl alco- d
e
haoflr(e2s4h:1t)u.bTeheanudppDerNwAatewr-aDsNpArecpiphiatsaetewdasbycoalldecdtiendgin0t.o6 2.4. PCR ampli¢cation and cloning d from
volume of isopropanol. The pellet was washed with 70% A region of the 16S rRNA gene was ampli¢ed using a h
ttp
(v/v) ethanol, dried and dissolved in 100 Wl of water. This nested PCR approach with primers described in Table 1. s
crude DNA sample was puri¢ed using a Wizard DNA As a ¢rst step, PCR was performed using two sets of ://a
c
a
clean up kit (Promega, USA) [25] and the resulting primers: (i) Ar4F^Un1492R and (ii) Ar4F^Ar958R, am- d
e
m
DNA was used as a template for the PCR. plifying respective regions of the 16S rRNA gene of
ic
around 1500 and 1000 bp in length. As a second step, .o
u
p
2.3. Sampling for in situ hybridization both PCR products were used separately in new PCR re- .c
o
actions using an Ar3F^Ar9R ‘internal nested’ primer set. m
SamplesforFISHwerecollectedduringtheperiodfrom This pair of Archaea-speci¢c primers was located inside /fem
MarchtoSeptemberfromthreedepths: theepilimnion(0^ the products from the ¢rst round and ampli¢ed a 16S se
c
/a
Table1 rticle
PCRprimersandFISHprobesusedinthisstudy -a
b
s
Name Sequence(5P^3P) Comments Refs. Fd tra
c
PCRprimers t/3
Ar4Fa (8^25)b TCYGGTTGATCCTGCCRG forwardprimer,speci¢ctoArchaea16SrRNAgene [12] 4/1
Ar958Ra (958^967) YCCGGCGTTGAVTCCAATT reverseprimer,speci¢ctoArchaea16SrRNAgene [53] /4
5
Ar9R(906^927) CCCGCCAATTCCTTTAAGTTTC reverse‘internalnested’primer,speci¢ctoArchaea [10] /6
1
16SrRNAgene 7
4
Ar3F(7^26) TTCCGGTTGATCCTGCCGGA forward‘internalnested’primer,speci¢ctoArchaea [10] 6
8
16SrRNAgene b
y
Un1492R(1492^1510) GGTTACCTTGTTACGACTT universalreverseprimer,speci¢ctoalltypesof [53] g
u
16SrRNAgene e
s
FISHprobesc t o
ARCH915(915^934) GTGCTCCCCCGCCAATTCCT speci¢ctomostArchaeasequences [20] 20 n 0
EURY498(498^511) CTTGCCCRGCCCTT speci¢ctopartofEuryarchaeotasequences [33] 0 4
A
CREN499(499^515) CCAGRCTTGCCCCCCGCT speci¢ctopartofCrenarchaeotasequences [33] 0 p
CREN512(512^527) CGGCGGCTGACACCAG speci¢ctomostCrenarchaeotasequencesinthe thiswork 0 ril 2
updatedARBdatabase 01
CREN569(569^585) GCTACGGATGCTTTAGG speci¢ctomostenvironmentalCrenarchaeota thiswork 0 9
sequencesintheupdatedARBdatabase
EURY496(496^512) GTCTTGCCCGGCCCTTTC speci¢conlytoaMethanomicrobialesgroupand thiswork 20
newVAL47,VAL1,VAL9,VAL78clones
EURY499(499^514) CGGTCTTGCCCGGCCCT speci¢ctoMethanosarcina,Methanosaeta, thiswork 20
MethanomicrobialesgroupsandnewVAL47,
VAL1,VAL9,VAL78clones
EURY514(514^528) GCGGCGGCTGGCACC speci¢ctomostEuryarchaeotasequencesinthe thiswork 20
updatedARBdatabase
aModi¢cationsinprimersAr4FandAr958RcomparedtoreferenceprimersweremadebyG.Jurgens.
bTargetsite(positionnumbersrefertotheE.coli16SrRNA).
cFISHprobeoverlapsshowninbold.
dFormamideconcentrationin%(v/v)fortheinsituhybridizationbu¡er.
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
48 G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56
rRNA gene region around 900 bp long. The reampli¢ed struction. Phylogenetic trees were inferred by performing
PCR products were puri¢ed from agarose gel using Prep- maximum likelihood (fastDNAml) [30], maximum parsi-
A-Gene DNA Puri¢cation Kits (Bio-Rad Laboratories, mony [31] and neighbor-joining with Jukes^Cantor dis-
USA) and were separately cloned in the pGEM-T-Easy tance correction method [32] included in the ARB pack-
vector (Promega, USA). age. Similarities between 16S rDNA sequences were
In order to ensure the archaeal origin of the 160 clones determined by using the multiple sequence comparison
obtained, Southern hybridization with an Archaea-speci¢c tool of the ARB program package.
probewasperformed[11].InsertsweredigestedwithAvaII
and MspI restriction enzymes to determine restriction 2.6. In situ hybridization and probe description
fragment length polymorphism [10] patterns and as a re-
sult 28 of them were chosen for sequencing. We have not FISH followed the protocol of Amann [26] and Glo«ck-
found any speci¢c di¡erences between the inserts reampli- ner et al. [17]. Oligonucleotides labeled with carbocyanine
¢ed from the two initial PCR products and therefore con- dye CY3 were purchased from Interactiva (Germany). A
D
sider all clones independent and equivalent. list of tested probes, their speci¢city, nucleotide sequences o
w
and respective formamide concentrations in the in situ hy- nlo
a
2.5. Phylogenetic analysis bridization bu¡ers are summarized in Table 1. d
e
The phylogenetic analysis was performed using the pro- scrTiboedde1te6cSt ArRrcNhAaeageinnethperosbaemspEleUsRthYe49p8r,evCioRusElyN4d9e9- d from
gram package ARB [28]. In order to ensure the correct [33] and ARCH915 [20] were used as well as ¢ve probes, h
ttp
phylogenetic placement of the obtained sequences and to newly designed using the ARB ‘Probe_design’ program s
give a comprehensive up-to-date evaluation of the Ar- [28]. ://a
c
a
chaea phylogeny, new sequences from freshwater (VAL) The probe EURY498 was previously thought to be spe- d
e
m
were added to the ARB database consisting of 642 com- ci¢c to most Euryarchaeaota [33]. However, since it was
ic
plete or partial Archaea 16S rRNA sequences from public designed, many new indigenous Archaea sequences were .o
u
p
databases.Onlythesequenceswiththesu⁄cientlengthfor found. After inclusion into the ARB database recently .c
o
phylogeny (minimum of 400 nucleotides) were included in published archaeal 16S rRNA sequences, it appeared m
the analysis.The ARB_EDIT tool was used for automatic that EURY498 was not able to detect more than 50% of /fem
sequence alignment and then the sequences were corrected the Euryarchaeota sequences. This probe does not cover se
c
manually. Regions of ambiguous alignment were excluded sequences obtained in this study from VAL-clones: VAL /a
from the analysis. A 50% invariance criterion for the in- 47, VAL 1, VAL9 and VAL78, related to Methanomicro- rtic
le
clusion of individual nucleotide sequence position in the biales, as well as many other sequences of uncultured Ar- -a
b
analysis was used to avoid possible treeing artifacts. The chaea. Therefore, in addition to the EURY498 probe, stra
program CHIMERA_CHECK version 2.7 from Riboso- three new probes for Euryarchaeota (EURY514, ct/3
mal Database Project [29] was used to detect possible chi- EURY496 and EURY499) were designed taking into ac- 4
/1
mera among studied sequences. countnewArchaeasequences,includingVALclones.Two /4
5
Many clones in the database were sequenced only par- newprobes, CREN512andEURY514, targetedsequences /6
1
7
tiallyandsomeofthemdonotoverlapwiththeothers.To at the kingdom level and were speci¢c to kingdoms Cren- 4
6
avoid reconstruction artifacts arising from partial sequen- and Euryarchaeota, respectively. The new probe 8 b
ces we constructed two ‘backbone’ maximum-likelihood CREN569 was speci¢c to most environmental sequences y g
u
trees: one consisting of 84 Crenarchaeota and the other of Crenarchaeota. New probes CREN512, EURY499 and e
s
of 79 Euryarchaeota clones for which sequence data be- EURY514 were tested for the optimal formamide concen- t o
n
tweenEscherichiacoli16SrDNApositions7and927were tration on pure cultures of Crenarchaeota (Pyrobaculum 0
4
available. All the VAL sequences were already included at neutrophilum, Sulfobococcus zilligii, Thermoproteus tenax) A
p
this stage of the analysis, therefore their placement was andEuryarchaeota(Methanosarcinabarkery,Methanobac- ril 2
determined by the more robust and sophisticated maxi- terium formicium). Members of the Bacteria (E. coli and 01
9
mum-likelihood method. The partial sequences were then Sinorhizobium) were used as negative controls for all Ar-
addedusingtheparsimonyoptionoftheprogrampackage chaea probes. Probe-speci¢c cell counts were calculated
ARB to those trees without changing the overall tree taking into consideration counts obtained with the nega-
topology. tive control NON338 [34].
Several trees di¡ering in the set of alignment positions Forhybridization,asectionofthe¢lterwasplacedona
as well as reference sequences were used during tree con- glass slide, overlaid with 20 Wl of hybridization solution
C
Fig. 1.Phylogeneticanalysis oftheVAL freshwater Crenarchaeota clonesfromlake ValkeaKotinen(shownasbold).Scalebar: 0.1,estimatednumber
ofsubstitutionspernucleotideposition.
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56 49
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/fe
m
s
e
c
/a
rtic
le
-a
b
s
tra
c
t/3
4
/1
/4
5
/6
1
7
4
6
8
b
y
g
u
e
s
t o
n
0
4
A
p
ril 2
0
1
9
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
50 G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56
[0.9 M NaCl, 20 mM Tris/HCl (pH 7.4), 0.01% SDS w/v, Table2
0^35% (v/v) formamide (for the reference see Table 1) and Environmental sequences of Group I Crenarchaeota included in the
phylogeneticanalysisandshownonthetreeasnumbersinboxes
50 ng of probe] and incubated at 46‡C for 2 h in an equil-
ibrated humid chamber [27]. Optimal formamide concen- Nameofthebox Sequencename AccessionNo.
trations for the new probes were determined empirically GroupI.1a TS AF052943,AF052946,AF052948,
by formamide series from 0 to50% (v/v) with a 10% (v/v)- AF052949
FIN AF052954
step.
ANT12 U11043
The ¢lters were transferred to a pre-warmed (48‡C) vial
WHARQ M88079
containing 50 ml of washing solution [70 mM NaCl, 20 SBAR M88075,M88076
mMTris/HCl(pH7.4),5mMEDTAand0.01%SDSw/v] pB1-100 U86488
and incubated freely £oating at 48‡C for 15 min. Filters JBT,JTA AB015274^AB015278
Mariana D87348^D87350
were dried on Whatman paper and overlaid with 50 Wl of
JM L24195^L24201
DAPIsolution(1mgml31 inwater)for5^10minatroom
PM,C U71109,U71110,U71112^U71118 D
temperature in the dark. Subsequently, they were shortly p712 U81531 ow
washed in 70% (v/v) ethanol and then in distilled water, Fosmid4B7 U39635 nlo
dried on Whatman paper and mounted on slides with C.symbiosum U51469 ad
e
Cesictei£nuceormAicFr1os(cCoiptiy£uwoarsLctadr.r,ieCdanotuertbausryd,eUscKri)b.eEdpbi£eufoorre- NPLVMHA4A_9O-9TU ZUU148167565771387^^UU4867658200 d from
[27] using a Zeiss Axioplan equipped with speci¢c ¢lter SB95 U78197^U78200 http
sets (DAPI: Zeiss01 and CY3: Chroma HQ41007, Chro- GroupI.1b P17,M17 U68650,U68604 s
ma Tech. Corp., Brattleboro, VT, USA). KBS AF058719^AF058730 ://a
c
SCA U62811^U62820 ad
pLemA8 U59987 em
pGrfA U59968^U59982,U59999 ic
3. Results and discussion GroupI.1c FFS X96688^X96692,X96694^X96696, .ou
p
AJ006919^AJ006922 .c
o
3.1. Microbial community aspects in Valkea Kotinen m
/fe
m
Due to its thermal stability the Valkea Kotinen hypo- as those found for marine waters, lakes and other studied se
c
limnetic water is anaerobic during the whole growing sea- Finnish forest lakes [7,40]. /a
son. About 43^94 WM methane has been measured in rtic
le
epilimnetic water and 140^13000 WM methane in hypo- 3.2. Phylogenetic placement of archaeal clones recovered -a
b
s
limnetic water from spring to winter, respectively [35]. from lake Valkea Kotinen among other environmental tra
Signi¢cant PCR products were obtained from hypolim- sequences ct/3
netic water layers with mcr-primers [36] coding for the 4
/1
methyl-coenzyme reductase (MCR) in methanogens [37]. A total of 28 clones (named VAL) recovered from lake /4
5
This could imply that methanogens make up a consider- Valkea Kotinen were sequenced. We found (i) that VAL /6
1
7
able part of the archaeal population, and play a crucial sequences contained representatives of both Cren- and 4
6
role in the C1-carbon £ux that via grazing phagotrophs Euryarchaeota, (ii) no cultivated Archaea were closely re- 8 b
y
¢nallyends upin zooplanktonassuggested byJones et al. lated to the VAL sequences and (iii) the archaeal sequen- g
u
[38]. ces most closely related to the VAL clones (with similarity e
s
Bacterial cell numbers in the lake varied between 1 and around 96%) inhabited a very wide range of environ- t o
n
5.7U106 cells ml31 with a bio-volume between 0.009 and ments: marine and lake sediments, di¡erent types of soils, 0
4
0.04 Wm3 cell31. The generally low variation of bacterial rice roots and an anaerobic digestor. Two separate trees, A
p
numbersandtheobservedsmallsizeofbacteriawasprob- one for Crenarchaea and one for Euryarchaea, were gen- ril 2
ably a major e¡ect of grazing pressure by phagotrophs erated (Figs. 1 and 2). Sequences listed in Tables 2 and 3 01
9
[39]. Bacterial production measured as [14C]leucine incor- represent uncultured Archaea indicated on tree ¢gures as
porationvariedonaveragebetween15.4and19.7WgCl31 numbers in the boxes. Tables 2 and 3 were made in order
d31 all over the water column [21]. A mean value of 15.4 to provide detailed information on all environmental se-
WgCl31 d31 wasmeasuredinepilimneticwater,18.5WgC quences used in tree reconstruction. The names of the
l31 d31 in metalimnetic water and 19.7 Wg C l31 d31 in clones, references and environments are given in Table 4.
hypolimnetic water. These data are within a similar range We found that 12 freshwater VAL-sequences fall within
C
Fig. 2.Phylogeneticanalysis oftheVAL freshwaterEuryarchaeotaclonesfromlakeValkea Kotinen(shownasbold).Scalebar: 0.1,estimatednumber
ofsubstitutionspernucleotideposition.NumbersintheboxesindicatehowmanyofthesequencesarefromunculturedArchaea.
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56 51
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/fe
m
s
e
c
/a
rtic
le
-a
b
s
tra
c
t/3
4
/1
/4
5
/6
1
7
4
6
8
b
y
g
u
e
s
t o
n
0
4
A
p
ril 2
0
1
9
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
52 G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56
a subcluster (named ‘freshwater cluster’ [41] or Group I.3 Table4
[13]ofGroupIunculturednon-thermophilicCrenarchaeo- Archaeasequencesbytheirenvironments
ta (Fig. 1). Eight of the sequences form a tight VAL I Clonename Environmentdescription
lineage inside of the cluster which also includes sequences Hotenvironments
previously detected in lake sediments [12,15], rice roots pSL hotspring,YellowstoneNationalPark,USA
and soil [42], and in an anaerobic digestor [43]. pJP hotspring,YellowstoneNationalPark,USA
PVA_OTU hydrothermalventmicrobialmat,Paci¢cOcean,
The analysis of 16 VAL sequences identi¢ed four dis-
Hawaii
tinct lineages related to di¡erent orders of Euryarchaeota
Marineandsaltyenvironments
(Fig. 2). Four clones were a⁄liated with environmental ANT,OARB marinepicoplankton,ArthurHarbor,Antarctica
sequences previously detected in anoxic sediments from WHARQ, marinepicoplankton,Atlanticocean,USA
Rotsee (Switzerland) [44], post-glacial profundal fresh- WARN
SBAR,SB95 marinepicoplankton,St.BarbaraChannel,USA
pB1 marinepicoplankton,Atlanticocean
Table3 p712,pN1 marineplankton,Paci¢cocean D
o
Environmental sequences included in the analysis and shown on the PM,C marineplankton,Atlanticocean w
Euryarchaeotatreeasnumbersinboxes Fosmid4B7 marinepicoplankton,Paci¢cOcean nlo
Nameof Sequence AccessionNo. NH marinepicoplankton,Paci¢cOcean,USA ade
TS suspendedparticulatematter,NorthSea,Netherlands d
thebox name FIN £ounderdigestivetract fro
Methano- ARR AJ227936,AJ227944,AJ227945,AJ227922, FF £ounderfeces m
sarcina AJ227940 GIN492 greymulletdigestivetract http
ABS Y15394,Y15387 JM seacucumbermidgut,Atlanticocean s
vadin U81777,U81776 C.symbiosum marinespongetissue,Paci¢cOcean ://a
c
LMA U87515,U87516 JBT,JTA,JTB marinesedimentcold-seeparea,JapanTrench ad
ARC AF029207^AF029211 Mariana marinesediment,MarianaTrench em
2MT,2C, AF015986,AF015987,AF015974, BBA marinesediment,CapeCod,USA ic
1MT AF015975,AF015972,AF015966, 1MT,2MT,2C saltmarshsediments,UK .ou
p
AF015991,AF015977,AF015986 Eel-BA,Eel-TA marinesediment,Eelriver,CA,USA .c
JTA AB015279 Freshwaterenvironments om
Eel-BA, AF134393,AF134384 pLaw freshwaterlakesediment,LawrenceLake,USA /fe
Eel-TA LMA freshwaterlakesediment,LakeMichigan,USA ms
Methano- ABS, Y15391^Y15393,Y15386,Y15389,Y15398, pLem freshwaterlakesediment,LakeLemon,USA ec
saeta A2M5.T1-,A2C YA1F5031956990,AF015980,AF015993, pWGinrfarc ffrreesshhwwaatteerrllaakkeesseeddiimmeenntt,,LLaakkeeGWriin¡dye,rmUeSrAe,UK /article
AF015973 Rot freshwaterlakesediment,LakeRotsee,Switzerland -a
b
Rot Y18082,Y18083,Y18087,Y18090, VAL freshwaterlakewater,ValkeaKotinenLake,Finland s
Y18091,Y18094 Soils tra
c
RiceclusterI ARR AJ227924,AJ227950,AJ227923,AJ227930, P17,M17 forestsoil,Brazil t/3
AJ227928,AJ227937,AJ227942,AJ227943, FFSB forestsoil,Finland 4/1
AJ227933,AJ227926,AJ227921,AJ227927, FFSA,FFSC forestsoil,Finland /4
5
AJ227920,AJ227935 KBS agriculturalsoil,Michigan,USA /6
1
ANME-1 Eel-BA, AF134380-AF134382,AF134383, SCA agriculturalsoil,Wisconsin,USA 7
4
Eel-TA AF134385^AF134387,AF134391, ABS,A5.1 anoxicricepaddysoil 6
8
AF134392 Arc. deepsubsurfacepaleosol,WashingtonState,USA b
y
Extreme 2C,1MT, AF015984,AF015967,AF015976, Otherenvironments g
u
halophiles 2MT AF015988,AF045182,AF045164 R Peatbogs e
s
AF045189 ARR riceroots t o
GroupII TS AF052944,AF052945AF052947 Vadin bio¢lmofa£uidized-bedanaerobicdigestor n 0
FF AF052953,AF052951 ARC bovinerumen 4 A
FIN AF052952 Cd30 intestinalmicro£oraofthetermitegut p
GIN492 AF052950 ril 2
pB1 U86502,U86508 01
pN1,p712 U86455,U81547 9
PVA_OTU_1 U46677
water sediment [16] and peat bogs [36] within the order
ANT5 U11044
Methanomicrobiales. Another four clones formed a group
WARN M88078
SB95,SBAR U78206,M88074,M88077 among sequences inhabiting marine sediments [45,46], rice
OARB U11042 roots and soil [42] and the deep-sea [47] (see Group III on
Methano- ABS,A5.1-B Y15388,Y15385,Y15390,Y15399 Fig. 2). Clone VAL147 clustered together with clone Eel-
bacteriales
TA1c9 found by Hinrichs et al. [48] in marine sediments
ARR AJ227934,AJ227938,AJ227947,AJ227949,
related to a decomposing methane hydrate, which is very
AJ227939
vadinDC06 U81775 convincingly, albeit based on circumstantial evidence, pro-
ARC AF029172^AF029205 posed to represent methanotrophic anaerobes.
Cd30 AB008900 The remaining seven clones formed two distinct lines of
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56 53
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/fe
m
s
e
c
/a
rtic
le
-a
b
s
tra
c
Fig. 3. Identi¢cation of Archaea in awater sample taken from lake Valkea Kotinen by FISH. Combination of hybridization with CY3-labeled, rRNA- t/3
targeted oligonucleotide probes EURY496 (A) and EURY499(B) andDAPI staining. Identical microscopic ¢elds have been visualized with an epi£uor- 4
/1
escencemicroscopeusing¢ltersetspeci¢cforCY3(right,upanddown)andDAPI(left,upanddown).Bar,10Wm. /4
5
/6
1
7
descent which all three methods (maximum likelihood, ranged from 71.3% (Thermoplasma acidophilum, Thermo- 4
6
maximum parsimony and neighbor-joining) of phyloge- plasmales) to 76.3% (Methanothermus fervidus, Methano- 8 b
y
netic analysis placed in the kingdom Euryarchaeota, bacteriales). The rice cluster V, which was most close to g
u
although the exact placement within this kingdom could the three VAL III clade sequences, is believed to comprise e
s
vary because of the large phylogenetic distances between non-methanogenic anaerobic Archaea [42]. t o
n
members of these clades and the rest of the analyzed se- We found that the new VAL sequences shared major 0
4
quences. Clade VALII, consisting of four clones, was only parts of the archaeal domain signatures at homologous A
p
distantly related to the Methanosaeta group. Sequence and nonhomologous positions [49,50]. However, clones ril 2
VAL33-1 allowed placement of this clade on the same with unstable placement (VAL90, VAL35-1, VAL31-1, 01
9
branch with the rice root and soil cluster I [42] using all VAL84, VAL125, VAL112) showed non-archaeal signa-
methodsonthebasisof85.9%similaritybetweensequence tures for the several strong signature nucleotides at the
and cluster. Levels of rDNA similarity between the other positions338,367,393and867(E.colinumbering).Intra-
three VAL II clade sequences and sequences from the domain signature analysis con¢rmed a⁄liation of 12 VAL
closest rice cluster I were extremely low (65.8 to 69.2%). sequences with the Crenarchaeota and 16 VAL sequences
A similar situation was observed for three clones from with the Euryarchaeota kingdom of Archaea.
cluster VAL III, which were related to the order Thermo-
plasmales, including two marine environmental Groups II 3.3. FISH in Valkea Kotinen
and III [13] and rice cluster V. We found that the level of
similarity between members of this clade, compared to In this study, we developed new archaeal probes to be
members of Thermoplasmales and Methanobacteriales, used with the FISH technique. This enabled us to quanti-
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart
54 G.Jurgensetal./FEMSMicrobiologyEcology34(2000)45^56
tatively study the indigenous Archaea distribution in the mental sequences in pelagic interfaces as shown in this
water column and its potential contribution to the micro- study was surprising.
bialfoodweb.Initially,weusedtheprobesARCH915[20], It has been shown earlier that representatives of Cren-
CREN499 and EURY498 [33] in the FISH experiments. archaeota inhabit boreal forest soils [10,11]. In this paper,
With the general Archaea-speci¢c probe ARCH915 we we have shown that they also occupy pelagic habitats in a
were able to detect 7(cid:254)2% of the DAPI-stained cells (data boreal forest lake. We found that in both biotopes, Cren-
for hypolimnium). Interestingly, no signi¢cant di¡erences archaeota were diverse, forming clusters that do not fall
were detected between the samples taken at di¡erent times into those known from other environments. Crenarchaeo-
during the sampling period from spring to autumn. ta found in the boreal forest lake water di¡ered signi¢-
Bright positive hybridization signals and cell morpholo- cantly from Crenarchaeota found in the boreal forest
gies (Fig. 3) received with four Euryarchaeotal probes soilsamples.Thismaybeduetothedi¡erentgeographical
(EURY498, EURY514, EURY496 and EURY499) were location of the two sites in Finland or due to speci¢c
identical to the signals obtained with the general Archaea habitatpropertiessuchasthedi¡erentnutrientavailability
D
probeARCH915.Themostspeci¢cofalltestedEuryarch- and/or selection- and grazing pressure in aquatic environ- o
w
aeota probes was EURY496, which covers the order ments compared to forest soil. However, a possible phy- nlo
a
Methanomicrobiales and related environmental sequences. logenetical^ecological link between Crenarchaeota in bor- d
e
Ttehcteerdefoinret,heweFIsSuHspeecxtpetrhiamtetnhtes bEeulorynagrctohatehoetaorcdleorneMsedthe-- estauldfioerdestthrsoouilgahnsdamboprleinagl faotrevsatrliaokues wsitaetserfrroemmalianksetoVable- d from
anomicrobiales. The size and the morphotypes of detected kea Kotinen and from soils within its water catchment h
ttp
cells (Fig. 3) show similarities to some methanogens (for area. Euryarchaeota detected in our experiments occupy s
example Methanospirillum hungatei). Such morphotypes a very speci¢c place in the environment. As many of them ://a
c
a
could have ecological aspects in the microbial food web seem to be very distantly related to any of the main meta- d
e
m
as was discussed by Pernthaler et al. [18] who showed that bolic subgroups of Archaea, namely, the methanogens or
ic
Archaea represented 1^5% of all DAPI-stained cells in thehalophiles,theirwayoflivingandmetabolicproperties .o
u
p
lake Gossenko«llesee, Austria. Despite a high background remain unclear. .c
o
£uorescence, maximum intensity of the signals and max- The outcome of this study, together with other recent m
imum cell counts for the EURY496 probe (7(cid:254)2% of the work on Archaea in marine and terrestrial environments, /fem
DAPI-stained cells) were observed in samples taken from supports our idea and working hypothesis that these or- se
c
the hypolimnion, which is in agreement with our methane ganisms inhabit a wide variety of niches and are ecologi- /a
measurement data and the results for the mcr-gene. Cell cally much more successful and important then previously rtic
le
counts for epilimnion were below the detection limit (1%). thought.Wesuggestthattheyplayaspeci¢croleinboreal -a
b
In comparison to the success with Euryarchaeota aquatic food webs and in the biochemical/geochemical cy- stra
probes, none of the Crenarchaeota probes including cling of carbon in their habitats. We expect that a variety ct/3
CREN499, as well as newly designed CREN512 and of new physiological phenotypes will be discovered, as 4
/1
CREN569 probes, gave positive hybridization signals. soon as some of these organisms can be successfully culti- /4
5
The negative FISH result is in contrast to our obtained vated and their metabolic and genetic potentials will be /6
1
7
PCR products but could be due to low numbers of Cren- available for further characterization. 4
6
archaeota cells in the samples and/or low ribosome con- 8 b
y
tents in the bacterial cells [51,52]. Other reasons could g
u
arise from physical^chemical aspects, i.e. low penetration Acknowledgements e
s
of probe material through the cell wall and the secondary t o
n
structure of the rRNA molecule. We thank Bernhard Fuchs, Wilhelm Scho«nhuber, Ka- 0
4
In summary, the results of the phylogenetic analyses trin Ravenschlag and Jakob Pernthaler (Max Plank Insti- A
p
and FISH suggest that a large number of previously ‘un- tute, Bremen, Germany) for their great support and help- ril 2
known’ Archaea were found in non-extreme habitats, i.e. ful advice, and Robin Sen (Biocenter, University of 01
9
in the pelagic water of a boreal forest lake, which has Helsinki, Finland) for revising the English. This work
temperature amplitudes between 4 and 20‡C and partial was supported by a three-month grant from Max Planck
oxic/anoxic conditions over the water column. Sequences Institute, Bremen, Germany, a grant from the Mai and
detected in lake Valkea Kotinen by the rRNA approach Tor Nessling Foundation, Finland, a grant from the Uni-
belong to members of the Crenarchaeota and Euryarch- versity of Helsinki, Finland and by the Finnish Academy
aeota branch of the domain Archaea. The existence of within the Finnish Biodiversity Programme (FIBRE).
Archaea in the freshwater environment was expected, if
we take into consideration the proven presence of meth-
anogens in freshwater ecosystems. However, such large References
diversity and low 16S rRNA sequence similarity of the
obtained new freshwater Archaea to the existing environ- [1] IPCC Report (1997) The regional impacts of climate change: an
FEMSEC116918-10-00 Cyaan Magenta Geel Zwart