Evidence for Polyphyletic Origin of the Members of the Subsection IV Cyanobacteria as Determined by 16S rRNA Analysis

16S rRNA 분석에 의한 Subsection IV cyanobacteria 균주들의 다계통성 기원의 증거

Abstract

Unicellular cyanobacterial strains of Subsections I and II and filamentous cyanobacterial strains of Subsection III have been shown to be polyphyletic, heterocystous strains of Subsections IV and V, both of which were previously reported to be monophyletic. In this study, the small subunit ribosomal RNA (16S rRNA) sequences of 13 strains of cyanobacteria - one strain, Oscillatoria nigro-viridis PCC7112, of the Subsection III, 6 strains including genus Anabaena, Nostoc, Tolypothrix, Calothrix and Scytonema of the Subsection IV, and 6 strains including genus Hapalosiphon, Fischerella and Chlorogloeopsis of the Subsection V - were determined. The phylogenetic analysis of cyanobacteria was carried out using the 16S rRNA sequences. The results of the phylogenetic analyses of 16S rRNA sequences, based on Neighbour-joining, maximum-parsimony, and maximum-likelihood methods, indicated that the members of Subsection IV were not monophyletic but polyphyletic. In addition, the phylogenetic results strongly indicated that the genus Scytonema in Subsection IV could be a common ancestor of heterocystous cyanobacteria in Subsection IV and V. Furthermore, the phylogenetic analyses revealed that the genus Anabaena could be phylogenetically diverse and that cyanobacterial strains in Subsection IV might be polyphyletic, whereas those in Subsection V could be monophyletic, as reported before. The results for the genus Anabaena indicate that it should be reclassified.

Subsection I과 II의 시아노박테리아 균주들은 단세포성이며, Subsection III의 시아노박테리아 균주들은 섬유상의 다계통성, 이형 사이토시스 형성성 균주들인 반면, Subsections IV와 V는 단일계통성으로 보고되어있다. 본 연구에서 13 균주의 시아노박테리아의 the small subunit rRNA (16S rRNA) 염기서열들이 - Subsection III의 Oscillatoria nigro-viridis PCC7112, Subsection IV에 속하는 Anabaena, Nostoc, Tolypothrix, Calothrix 및 Scytonema속을 포함한 6 균주, Subsection V에 속하는 Hapalosiphon, Fischerella and Chlorogloeopsis 속의 6 균주 - 결정되었다. 결정된 16S rRNA 염기서열을 이용하여 시아노박테리아의 분자계통분석을 수행하였다. 그러나, 16S rRNA의 염기서열 결정을 근거로 한 본 연구의 계통분석결과 Subsection IV는 단일 계통성이 아닌 다계통성이며, 반면 Subsection V는 이전에 보고되어진 것처럼 단일 계통성임을 나타내었다. 또한, 본 연구 결과는 Scytonema속이 이형 사이토시스 형성성 시아노박테리아인 Subsection IV 및 V의 공통 조상일 수 있음을 강력하게 나타낸다. 부가적으로, 본 연구의 분자계통 분석을 통해 Anabaena속은 다계통성으로 계통학적으로 다양한 종들로 구성되어 있음을 나타내고 있다. 본 연구 결과는 Anabaena속이 좀 더 세밀하게 재분류 되어져야 함을 나타낸다.

Introduction

The cyanobacteria are morphologically diverse bacteria that perform oxygenic photosynthesis and possess chlorophyll a. Cyanobacteria have also been recognized as a model system for studying the photosynthesis and endosymbiotic origins of chloroplasts in plants [1]. At present, cyanobacteria are classified into five sections and formally recognized primarily according to their morphology and developmental characteristics [2-4, 15-17]. Unicellular cyanobacteria are classified into the Subsection I and II in terms of their different division patterns. The division pattern of Subsection I show a binary fission or budding, but that of the Subsection II is expressed by multiple fission which results in the formation of baeocytes [16]. The Subsection III is composed of filamentous cyanobacteria reproducing by trichome breakage [16]. The Subsection IV and V comprise the filamentous and heterocytous cyanobacteria reproducing by hormogonia formation [16], and are able to develop heterocysts and akinetes. These two sections are separated on more one plane. Members of the Subsection IV are divided on only one plane, and members of the Subsection V can be divided on more than one plane [16]. The Subsection IV and V are composed of eight genera (Anabaena, Calothrix, Cylindropermum, Nostoc, Nodularia, Scytonema, Chlorogloeopsis, and Fischerella) that are common to most classification system [2-4, 9, 16], as well as additional genera (Tolypothrix and Hapalosiphon) in more recent classification systems [2-4]. Cyanobacterial classifications based on morphological characteristics remain controversial and may not reflect true phylogenetic relationships [5, 8, 9, 11, 18, 22, 26].

Recently, phylogenetic analyses of cyanobacteria based on small subunit rRNA (16S rRNA) have been performed. Several studies based on 16S rRNA have reported that in contrast to unicellular and filamentous non-heterocytous cyanobacterial strains composed of Subsection I, II, and III, that do not form clusters to be consistent with their classification, the heterocytous members of cyanobacterial Subsection IV and V are monophyletic [8, 11, 21, 23, 24].

In this study, we investigated the phylogenetic relationships of the cyanobacterial strains of the Subsection IV and V based on 16S rRNA and compared the phylogenetic position of cyanobacterial strain with each of them.

 

Materials and Methods

Cyanobacterial strains and growth conditions

The cyanobacterial strains investigated and media are listed in Table 1. All were cultured at 20℃, 12:12 hr L: D (light : dark) with fluorescent lamp illumination of 500 lux.

Table 1.Sequences in bold were determined in this study. T: type strain. aRippka and Herdman, 1992 [6]; b IAM Catalogue of strains, 1998 [26].

The preparation of cell lysate

Cells were harvested from 1.5 ml culture broth by centrifugation, washed with 1 ml of deionized water, and suspended in 100 μl of 20 mM Tris (pH 8.0) containing 0.1 mM EDTA, 0.5% Tween 20, and 0.1% Non-iodet P-40 (Boehringer Mannheim, Germany). Final lysis was achieved by the addition of 10 μg of Proteinase K and incubation for 20 min at 60℃.

PCR amplification

The almost complete 16S rDNA from the genomic DNA of the respective strains was amplified by PCR using oligonucleotide primers of 1R (forward primer: 5′-AGAGTTTGATCCTGGCTCAG-3′) and 16C (reverse primer: 5′-AAGGAGGTGATCCAGCCGCA-3′) [27]. PCR was performed at 3 min at 94℃ and then at 30 cycles with the following features: 1 min at 94℃, 1 min at 55℃, and 2 min at 72℃, followed by a final elongation step for 10 min at 72℃.

Sequencing of 16S rRNA

The refined PCR products were directly sequenced with the use of a BigDye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, USA). Primers used for the cycle sequencing were 1R, 320R (5′-CTGCTGCCTCCCGATA-3′), 520F (5′-CAGCAGCCGCGGTAATAC-3′), 704R (5′-TCTACGCATTTCACCGCTAC-3′), 926F (5′-AAACTCAAAGGAATTGACGG-3′), 1100R (5′-GGGTTGCGCTVGTTG-3′) (V, G or C or A), and 16C. The cycle sequencing reaction was performed with a PCR Thermal CyclerMP (Takara, TP3080, Japan). The sequencing reaction was performed for 5 min at 96℃ at first and consisted of 25 cycles of the following: 10 sec at 96℃, 10 sec at 52℃, and 2 min at 60℃. The cycle sequencing products were purified with use of CENTRI-SEP Spin Columns (Applied Biosystems, USA). DNA sequences were analyzed using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems, USA).

Alignment and phylogenetic analysis

The nucleotide sequences of 16S rRNA were aligned using the CLUSTAL W computer program, version 1.81 [20]. 16S rRNA sequences were aligned based on their secondary structures with a selection of cyanobacterial reference sequences obtained from the DNA Data Bank of Japan (DDBJ), and the alignment was manually corrected.

Neighbour-Joining (NJ)

The phylogenetic tree was constructed from the evolutionary distance matrix calculated by the neighbour-joining method [19] with Kimura`s two-parameter method [12]. Neighbour-joining analysis was performed using the MEGA2 program [13]. All gaps in the alignment were excluded in order to draw the tree. Bootstrap analyses were performed with 1,000 replicates.

Maximum-likelihood (ML) and Maximum-parsimony (MP)

A total of 100 bootstrap samples for alignment was produced by using the program SEQBOOT from PHYLIP version 3.6 [6, 7], and phylogenetic trees were inferred from each bootstrap sample by using maximum-likelihood (DNAMLK software in PHYLIP version 3.6) and maximum-parsimony (DNAPARS software in PHYLIP). The resulting trees were combined to yield a consensus tree (CONSENSE software in PHYLIP). A matrix of evolutionary distances was also derived from bootstrap alignment by using the DNADIST software in PHYLIP. Trees were inferred from the matrices by using the FITCH software in PHYLIP. The resulting trees were visualized to yield consensus trees (CONSENSE software in PHYLIP) using the TREEVIEW version 1.6.6 software [14].

 

Results and Discussion

The new 16S rDNA sequences were deposited in the DDBJ; accession numbers for each cyanobacterial strain used in this study and reference strains are given in Table 1. Positions with gaps and undetermined and ambiguous sequences were removed. A total of 1335 sites were used for the phylogenetic analysis.

Constructed phylogenetic trees, the NJ, the ML, and MP trees, reveal that all the numbers of cyanobacterial Subsection V were monophyletic (Fig. 1). However, as shown in Figure 1, the cyanobacterial strains of Subsection IV are clearly separated into two clusters. All members of cyanobacterial Subsection IV, except the genus Scytonema, consist of a major cluster on phylogentic trees, supported by bootstrap values of 99% of NJ, 95% of ML, and 75% of MP (Fig. 1). The cluster of the genus Scytonema is placed externally to those of Subsections IV and V, forming a well-supported cluster with bootstrap values of 95% of NJ, 87% of ML, and 88% of MP (Fig. 1). In some reports published earlier, although unicellular cyanobacterial strains of Subsections I and II, and filamentous cyanobacterial strains of Subsection III were shown to be polyphyletic, heterocystous strains of Subsections IV and V were shown to be monophyletic [8, 10, 11, 21, 23, 24]. However, interestingly, our phylogenetic analyses based on sequence determination of the 16S rDNA indicate Subsection IV was not monophyletic.

Fig. 1.Phylogenetic tree of cyanobacterial 16S rRNA sequences. Numbers at nodes are bootstrap values. NJ, ML, and MP bootstrap values are indicated as NJ/ML/MP. A total of 1335 unambiguously aligned positions were used. Local bootstrap probabilities are indicated at nodes if larger than 60. The strains of which the 16S rRNA gene sequences were determined in this study are indicated in bold. Starred strains were used as outgroups. T: type strain. Bar, 0.01 substitution per nucleotide position.

Our phylogenetic analyses indicate that the genus Scytonema could be a common ancestor of cyanobacterial Subsections IV and V. The results reveal that the divergence of the genus Scytonema is earlier than any other genus of Subsections IV and V. The results of analyses based on NJ, ML, and MP strongly support this (Fig. 1). Actually, in Bergey’s Manual of Systematic Bacteriology, the Scytonema hofmanni PCC 7110 was described as the following: “Scytonema hofmanni PCC 7110 has no close relatives in the phylogenetic trees with the recent addition of new sequences, however, some interesting relationships are emerging” [25]. Our results may constitute evidence to help explain the evolutional relationships of cyanobacterial Subsection IV and V.