Академический Документы
Профессиональный Документы
Культура Документы
1
Porto Alegre, Brazil
2
Curso de Pos-graduacao em Microbiologia Agricols e do Ambiente, Universidade
The 16S rDNA from 22 cyanobacteria isolated from biofilms on walls of modern and
historic buildings in Brazil was partially sequenced (~350 bp) using specific primers.
The cyanobacteria with the closest matching sequences were found using the BLAST
tool. The sequences were combined with 52 other cyanobacterial sequences already
deposited in public data banks and a dendrogram constructed, after deletion from each
sequence of one of the variable 16S rDNA regions (V I). The newly sequenced
organisms fitted well within their respective families, but their similarities to other
members of the groups were generally low, less than 96%. Close matches were found
only with one other terrestrial (hot dry desert) cyanobacterium, Microcoleus sociatus,
and with Anabaena variabilis. Phylogenetic analysis suggested that the deletion of the
studies. The results support the standard phylogenetic tree based on morphology, but
suggest that these terrestrial cyanobacteria are distant relatives of their equivalent
genera are present, and these grow in biofilms (microbial consortia) on stone and
painted surfaces (Anagnostidis et al., 1983; Ortega-Calvo et al., 1995; Gaylarde and
Gaylarde, 2000).
Calvo et al., 1995). Cyanobacteria can survive repeated drying and rehydration cycles
(Whitton, 1992) and high UV levels (Garcia-Pichel et al., 1992; Matsunaga et al.,
1993; Chazal and Smith, 1994) and for these reasons they are particularly important
microorganisms, which are found worldwide. They have long been recognised as
important soil and water organisms, where their activities of nitrogen fixation and
toxin production are of special interest. Only more recently has their role in the
Traditional methods for the detection of cyanobacteria are based on their culture on
specific growth media. These isolation techniques were developed in the area of
aquatic microbiology and later extended to terrestrial habitats (soil). They involve
solid medium (Rippka et al., 1979). However, these methods can result in the
detection of artificially low numbers and diversity, due, in part, to the presence of
inhibitory and predatory organisms, such as fungi, bacteria, and protozoa (Gaylarde
culture because of the activity of browsing and inhibitory organisms, and, of course,
many microorganisms present in the environment are not detected by current culture
Although 16S rRNA gene sequencing has been used to study the relationship between
cyanobacteria (Wilmotte et al., 1993; Nelissen et al., 1996; Turner, 1997), there has
Tomaselli et al. (2000) used amplified ribosomal DNA restriction analysis (ARDRA)
dendrogram that provided some idea of biofilm diversity. This approach did not,
population. Crispim et al. (2003) and Gaylarde et al. (2004) described a method to
does not rely on isolation and this is the approach used here to examine the
Sampling and microbiological analysis. Samples of biofilms were taken from the
external surfaces of modern and historic buildings in Porto Alegre, Ouro Preto,
Tiradentes and São Paulo, Brazil, using the adhesive tape method (Gaylarde and
Gaylarde, 1998; Shirakawa et al., 2002). All surfaces showed visible discoloration,
For microbiological analysis, tape samples were placed directly on solid Modified
Knop’s Medium (MKM; Gaylarde and Gaylarde, 2004) and incubated at 25oC in an
and in culture were identified by standard morphological methods (Holt et al., 1994;
DNA extraction. Genomic DNA was extracted from some cyanobacterial isolates
alcohol extraction (Gaylarde et al., 2004). Extracted DNA was amplified using the
DNA amplification For the “direct” PCR, microcolonies or filaments from the
rehydrated, or briefly incubated, biofilms and well separated filaments isolated from
temperature (23oC). After the final thawing, dNTPs, primers, MgCl 2, buffer and Taq
DNA polymerase were added directly in the appropriate amounts (Gaylarde et al.,
2004) to the Eppendorf to give a final volume of 25 µl. The PCR reaction was carried
Fullerton, CA) with the following programme: 92oC for 2 min, 30 cycles of 92 oC for
20 s, 55oC for 30 s, 72oC for 50 s and a final extension at 72 oC for 2 min. The
bromide staining.
Primers. 16S rRNA primers designed by Neilan et al. (1997, 2002) were used. The
is identical to A2 (Iteman et al., 1995). The reverse primer, 408R, is specific for
DNA sequencing. Purified PCR products were sequenced with the reverse primer
408R using Big Dye terminator technology (PE Applied Biosystems) and the
sequencing gels were run and analysed by the University of New South Wales
Genomic Analysis Facility. In some cases, both primers were used separately for
sequencing. The consensus sequences were submitted to the BLAST search tool
constructed for the sequences, along with others of similar groups selected randomly
from public databases, using the CLUSTAL-X program (Higgins et al., 1992) and
bootstrap with 1000 resampling events. An unrooted tree was constructed using a
heuristic approach (Swofford et al., 1996). Close inspection of the sequences revealed
that the hypervariable region, VI, differed not only in sequence but also in length and
interfered with the CLUSTAL alignment. This section was therefore removed, the
dendrogram reconstructed and the BLAST analyses repeated using these modified
sequences.
The cyanobacteria from our samples, and their similarities with public database
organisms, are shown in Table 1. The results are shown with and without the
sequence before base 101 (E. coli position), which includes VI (see Figure 1).
Analysis of the secondary structure indicated that the bases that did not match
with those of other organisms in the public databases increased in all cases after this
modification.
Our results demonstrate that the removal of the V I region of the genome improved the
similarity levels of the new sequences to those in public databases. Some workers
(Stackebrandt et al., 1991; Heuer et al., 1999) have suggested that phylogenetic
studies should be carried out on partial sequences of the 16S rRNA variable regions,
since they represent areas where evolutionary change may take place in an otherwise
variable regions after their identification with dedicated programs (Crosbie et al.,
2003).
Cyanobacteria are notable for the short length of their 16S rDNA genome compared
197 to 220 and 452 to 480 are consistently present. However, it is only in the case of
region 68 to 101 (VI) that the length of the deletion is variable (Figures 3a-c). The
sequences in this region, not found in related, but non-terrestrial, genera. We suggest
that variable regions, if they have any utility, should be used only for phylogenic
studies at the genus level, especially since alignment methods do not work for regions
of variable length.
The dendrogram produced from the modified sequences (Figure 2) shows that the 16S
morphology and that the cyanobacteria from buildings group well with their aquatic
taxonomy is inadequate to treat this group realistically. Turner (1997) stated that the
Epilithic cyanobacteria (including those from painted stone surfaces) must be well
adapted to survive frequent desiccation, high salt levels and extremes of temperature.
distinct population that differs from the better-studied aquatic members of this group.
Major differences include the fact that cyanobacteria found on solid surfaces include
fundamental character and that in the tropics (or at altitude) the cells frequently have
thick sheaths, which, like the cell cytoplasm, may be heavily pigmented. Figure 4(a-
d), in the web-based supplementary material, shows this feature, as well as the range
of morphological variation of Scytonema spp. ccg16, ccg25 and ccg26, which show a
very close relationship and cluster tightly in the dendrogram. These Scytonema spp.
differ from other environmental organisms that cluster with Scytonema hofmannii in
that dark pigments are present in the cell cytoplasm, false branches are more common
and consistently geminate and on media rich in nitrogen heterocysts are frequent. A
thick, pigmented sheath is normal in both groups in the environment and this usually
A similar grouping, with a tight cluster of terrestrial strains of Nostoc, is seen in the
dendrogram; these sequences are rooted in the same branch as deposited sequences of
Plectonema (Table 1). Boyer et al. (2003) studied 31 isolates of Microcoleus from
desert crusts and found the M. vaginatus clade was most similar to published
sequences from Trichodesmium and Arthrospira. They concluded that the genus
requires revision and our result emphasises this. The M. vaginatus morphotype
isolated from terrestrial environments must be regarded as a very diverse group with a
highly conserved 16S rDNA sequence. This analysis once again highlights the
organisms with very close 16S rDNA sequences may have very different appearances.
environmental conditions, and in the case of the heterocystous cyanobacteria and the
sheath, which produces false branches at some stage of growth. Since the branching
character includes a wide range of stable morphotypes, defined by the width and
length of the cells, their shape and colour, this is almost certain to represent several
genera, which will be phylogenically distant and, indeed, the existence of the genus
The dendrogram shows that these terrestrial cyanobacteria fit well to the taxonomic
positions of related genera, but also shows that their genetic distance from other
members of their morphologically identified genera is large. The table confirms this
result. The physiological adaptations for survival in the dry environment of terrestrial
surfaces would lead one to expect the emergence of novel taxa. The ribosomal
Acknowledgements
We wish to thank the Brazilian agency CNPq for funding for materials and a
postgraduate grant to CAC. BAN thanks the Australian Research Council for financial
support.
BOYER, S.L., JOHANSEN, J.R., FLECHTNER, V.R. and HOWARD G.L. (2003).
based on sequence analysis of the 16S rRNA gene and associated 16S-23S ITS
D'SOUZA, L.M., DU, Y., FENG, B., LIN, N., MADABUSI, L.V., MÜLLER, K.M.,
PANDE, N., SHANG, Z., YU, N. and GUTELL, R.R. (2002) The Comparative RNA
Systematic Bacteriology, Boone, D.R., Castenholz, R.W., Garrity, G.M. (eds.), New
species from Java that is resistant to high light intensity and UV. Microbiol. 140, 183-
189.
and morphological diversity of cyanobacteria in soil desert crusts from the Colorado
for ultraviolet sunscreen role of the extracellular pigment scytonemin in the terrestrial
GAYLARDE, P.M. and GAYLARDE, C.C. (1998). A rapid method for the detection
GAYLARDE C.C. and GAYLARDE P.M. (2004a). A comparative study of the major
monuments in Latin America: microorganisms and mechanisms. Corr. Rev. 22, 395.
(1999). Polynucleotide probes that target a hypervariable region of 16S rRNA genes
8, 189-191.
HOLT, J.G., KRIEG, N.R., SNEATH, P.H.A., STALEY, J.T. and WILLIAMS, S.T.
and Wilkins.
(1995). Use of molecular tools for the study of genetic relationships of heterocystous
39, 250-253
NEILAN, B.A., JACOBS, D., DEL DOT, T., BLACKALL, L.L., HAWKINS, P.R.,
COX, P.T. and GOODMAN, A.E. (1997). rRNA sequences and evolutionary
STANIER, R.Y. (1979). Generic assignments, strain histories and properties of pure
buildings and the effect of biocide. FEMS Microbiol. Ecol. 39, 165-173.
SUZUKI, M., RAPPE, M.S. and GIOVANNONI, S.J. (1998). Kinetic bias in
Phylogenetic inference. In: Molecular Systematics, Hillis, D.M., Moritz, C. & Mable,
B. (eds.), Sinauer Ass. Ltd., Sunderland, MA, Chap. 11, pp. 407-514.
WARD, D.M., WELLER, R. and BATESON, M.M. (1990). 16S rRNA sequences
65.
WHITTON, B.A. (1992). Diversity, ecology and taxonomy of the cyanobacteria. In:
Mann, N.H., Carr, N.G. (Eds.), Photosynthetic Prokaryotes, Plenum Press, New York,
pp 1-51.
(Mastigoclaudus laminosus HTF') strain PCC 7518 and phylogenetic analysis. FEBS
Figure 1
The structure of E. coli 16S rRNA from base 28 to base 408 is shown with variable
regions VI and VII marked. The simplified structure is modified from Cannone et al.,
(2002), in such a way that the E. coli 16S rRNA will show variation with respect to
cyanobacteria; the modification has been made with reference to all bacterial
sequences. Suzuki et al. (1998) state that VI runs from 72 to 101, inclusive, whereas
our observations show that it runs from 69 to 100 in all cyanobacteria. The pairing
VII runs from 176 to 221 (Suzuki et al., 1998). Two helical regions, denominated
Helix 1 (136 to 142 and 221 to 227, inclusive) and Helix 2 (154 to 156 and 165 to
highly conserved. Within VII, bases 195 to 197 are conserved in all the bacteria from
to 180-182 in cyanobacteria. Two regions, V IIa and VIIb, show features characteristic
of cyanobacteria. VIIa runs from 183 to 193 and its variability is very large. VIIb (198
to 205 and 214 to 219) is deleted in cyanobacteria. This same deletion is shared with
bacteria of the genus Clostridium and its length varies in other bacterial genera.
Wide bars show canonical pairing (G:C and T:U), narrow bars show G:U pairs and
dots mark other non-canonical pairs which contribute to the secondary structure.
Lines mark the E. coli positions, which are numbered at every tenth base.
Dendrogram constructed from sequences starting at E. coli position 101. The position
of Leptolyngbya sp. ccg40 is marked to indicate that this short sequence clusters with
its nearest relatives, but that the genetic distance is increased because of the short
sequence used to give the match. This indicates that bootstrapping is robust for
sequences of variable length, but shows that the degree of relatedness cannot be
web material), black = subgroup I, red = subgroup II, blue = subgroup III and green =
heterocystous cyanobacteria. The colours group well together, even though the stems
Figure 3
This figure shows the variation in length of V I in cyanobacteria. The diagrams show
that when helical regions are present they are well paired.
Scytonema ccg26 (4c and d), all grown on MKM. The bar markers are 10µm.
This summarises the morphological identity of the organisms sequenced and the
nearest matches of the sequences to deposited organisms in the BLAST database. The
nearest matches to the morphological identifications are also given. The similarities
are for the full sequence and for the short sequence from E. coli position 101 onward,
together with the percentage similarities with and without the sequence preceding
base 101.
(a)
(b)
(c)
(a) (b)
(c)
(d)
POA = Porto Alegre; OP = Ouro Preto; SP = Sao Paulo; T = Tiradentes; n.a. = not available (short sequence deposited in data bank)