Geology Reference
In-Depth Information
found in the biofilm on the tufa surface. The other
cyanobacteria lineages were also present in the
core layers (CL) below the biofilm, but in rather
low frequencies (5% or less), except for the lineages
'Unidentified C' (of which still 25% of the clones
were found in CL 1 - 3 and 5) and Pseudanabaena
sp. The latter even seemed to have a preference
for layers below the biofilm, because 80% of the
Pseudanabaena sp. clones were from CL 2 and 5;
only a single clone was found in the biofilm. A Pseu-
danabaena morphotype has also been reported for
the lower parts of the tufa biofilms at the DB (Arp
et al. 2001b). Also 'Unidentified F' was retrieved
only in a deeper layer, CL 5, but not found in the
biofilm at the WB5 site. From 'Unidentified A', a
single clone was even recovered from the
lowest layer CL 6. Likely, these two unidentified
cyanobacteria may exhibit an endo- or chasmolithic
life style.
Identifications of the environmental clones at the
generic level could be inferred only for three cyano-
bacteria lineages, i.e. where close relationships with
reference strains were resolved in the 16S rDNA
phylogeny which corresponded to high phenetic
sequence similarities. A large fraction (29%) of
the cyanobacterial clones from the WB5 site had a
very high phenetic sequence similarity of 99.9%
with Tychonema bourrellyi Anagnostidis &
Kom´rek strain CCAP 1459/11B which is the auth-
entic strain of the type (Suda et al. 2002). The full
sequence obtained for one of these clones was
together with T. bourrellyi CCAP 1459/11B in a
well supported clade (Fig. 7). Therefore, the
clones were identified as members of Tychonema
Anagnostidis & Kom´rek. Members of the 'Uniden-
tified A' lineage, however, were more distantly
related to strain T. bourrellyi CCAP 1459/11B in
the phylogeny corresponding to the 97.9 - 98.2%
phenetic sequence similarity level and thus were
not assigned to Tychonema. The majority of
clones sampled from the WB5 site (41%) formed a
lineage which could not be identified ('Unidentified
B' in Fig. 7). The lineage had a sister-group relation-
ship with the Tychonema/P. autumnale/Micro-
coleus clade in the 16S rDNA phylogeny (Fig. 7).
Environmental clones identified as Chamaesiphon
and Pseudanabaena formed well supported clades
with corresponding reference sequences (Fig. 5).
They had phenetic sequence similarities of at least
98.4% with C. subglobosus (Rostafinski) Lemmer-
mann PCC 7430 and 96.8% with Pseudanabaena
sp. PCC 6802, respectively. In contrast, for the
four lineages, 'Unidentified C', 'Unidentified D',
'Unidentified E' and 'Unidentified F', no closest
relatives were available and phenetic sequence simi-
larities with the next named sequences were below
96.5%. However, an uncultured unnamed cyano-
bacterium
number DQ181686; Taton et al. 2006) from Antarc-
tic lakes was the next closest relative with lineage
'unidentified C' (98.4% phenetic sequence simi-
larity) and an uncultured cyanobacterium (sequence
accession number EF111085) from the Bogota
River, Columbia, with lineage 'unidentified F'
(98.1% similarity). Neither the morphology nor
the taxonomic identity of these operational taxo-
nomic units (OTUs) is known yet. The two lineages,
'Unidentified D' and 'Unidentified E', did not even
group with any close relative in the phylogeny; they
were different at the 90 - 95% phenetic sequence
similarity level to the next neighbouring available
cyanobacteria sequences.
Cyanobacterial rDNA lineages and corresponding
morphotypes. Though the discussion of the
observed morphotypes in the light of the molecular
results is tempting, it remains still speculative at
present. Sequence comparisons of the cultured
biofilm cyanobacteria with their counterparts from
the environmental clone libraries will be essential
to link morphotypes and rDNA signatures. How-
ever, the same cyanobacterium may exhibit mor-
phological features in the resin-embedded hardpart
sections that are absent or different from those in
culture. For 'Unidentified A', which was recovered
from biofilms at almost all sites of both karstwater
streams, a cultured strain (WBK15) exhibited
features identical to those of the Lyngbya sp. mor-
photype (Fig. 6). Therefore, the Lyngbya sp. mor-
photype likely belongs to lineage 'Unidentified A'
(Fig. 7). However, the same morphological features
also match the description of Tychonema tenue
(Skuja) Anagostidis & Kom´rek (formerly Oscilla-
toria bornetii f. tenuis Skuja) as given in Kom´rek
& Anagnostidis (2005). A large fraction of the WB5
environmental clones were very close relatives with
T. tenue strain SAG 4.82 and T. bourrellyi strain
CCAP 1459/11B (Fig. 7). Interestingly, species of
Tychonema are not yet known from epilithic habitats
such as tufa surfaces. T. bourrellyi has been described
from phytoplankton (Anagnostidis & Kom´rek
1988; Suda et al. 2002) and T. tenue is a major com-
ponent of phytoplankton of the Norwegian fjord lake
Mjøsa (World Lakes Database of the International
Lake Environment Committee, www.ilec.or.jp/
database/database.html). From these findings it is
concluded that the same morphotype may even rep-
resent genetically distant cyanobacteria and the
closest relatives of tufa biofilm cyanobacteria may
be from quite different habitats. Furthermore, a
clear distinction of species and genera within the
Tychonema/P. autumnale/Microcoleus clade as
based on currently used morphological criteria
appears impossible, for example, the distinction of
Phormidium autumnale, T. bourrellyi and Microco-
leus spp. as discussed in Palinska & Marquardt
(clone
H-B02,
sequence
accession
Search WWH ::
Custom Search