SEARCH

SEARCH BY CITATION

Keywords:

  • coralloid root;
  • cycad;
  • Cycas revoluta ;
  • DGGE ;
  • genetic diversity;
  • Nostoc

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The diversity of cyanobacterial species within the coralloid roots of an individual and populations of Cycas revoluta was investigated based on 16S rRNA gene sequences. Sixty-six coralloid roots were collected from nine natural populations of cycads on Kyushu and the Ryukyu Islands, covering the entire distribution range of the species. Approximately 400 bp of the 5′-end of 16S rRNA genes was amplified, and each was identified by denaturing gradient gel electrophoresis. Most coralloid roots harbored only one cyanobiont, Nostoc, whereas some contained two or three, representing cyanobiont diversity within a single coralloid root isolated from a natural habitat. Genotypes of Nostoc within a natural population were occasionally highly diverged and lacked DNA sequence similarity, implying genetic divergence of Nostoc. On the other hand, Nostoc genotypes showed no phylogeographic structure across the distribution range, while host cycads exhibited distinct north–south differentiation. Cycads may exist in symbiosis with either single or multiple Nostoc strains in natural soil habitats.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Many land plants can form symbiotic associations with nitrogen-fixing cyanobacteria. These cyanobacteria have symbiotic associations with fungi (Geosiphon and lichens), bryophytes (liverworts and hornworts), pteridophytes (Azolla), gymnosperms (Cycadaceae), and angiosperms (Gunnera) (Bergman et al., 1996; Rai et al., 2000).

Cycads are an ancient group of seed plants that have existed for approximately 300 million years. In general, cycads develop three types of roots: a tap root, which is equivalent to the primary root system found in most types of plants, lateral roots, and ‘coralloid roots’, which are highly specialized lateral roots containing cyanobionts (Costa & Lindblad, 2002). Mature coralloid roots are thick, finger-like structures usually embedded in soil at a shallow depth. Cyanobionts are restricted to the intercellular region of the cortex where they grow in a microaerobic and dark environment. Although not actively photosynthesizing, they maintain their clear green color in this environment (Costa & Lindblad, 2002). Nostoc spp. are the most common cyanobionts in Cycadaceae, whereas Anabaena and Calothrix species form symbiotic associations less frequently (Obukowicz et al., 1981; Zhu, 1982; Grobbelaar et al., 1986). All known cycad cyanobionts are of the botanical order Nostocales (Komarek & Anagnostidis, 1989). Several hypotheses of the mechanism of cyanobiont infection have been suggested, but the details are yet to be determined.

Costa et al. (1999) collected 11 coralloid roots from cycads cultivated in a botanical garden, including the genera Cycas, Encephalartos and Zamia. Only a single Nostoc strain was detected in each coralloid root, based on DNA sequence analyses of the tRNALeu (UAA) intron. They also determined Nostoc diversity within single or multiple coralloid root(s) of individual plants. Zheng et al. (2002) collected coralloid roots from indigenous cycad plants in a Chinese national park and found a large variety of cyanobacteria in a single coralloid root using polymerase chain reaction (PCR) fingerprinting. They also suggested that different strains of cyanobacteria could exist in the apical, middle and basal parts of a single coralloid root. However, Costa et al. (2004) suggested that the finding of multiple strains of cyanobacteria by Zheng et al. (2002) was likely due to a methodological limitation of fingerprinting: fingerprinting patterns may differ according to the DNA extraction method used, even with identical DNA samples. Costa et al. (2004) also pointed out that the results of Zheng et al. (2002) would be hard to corroborate, as no plant or bacterial controls were used. Thus, the number of cyanobiont strains that exist in a single coralloid root remains uncertain. Costa et al. (2004) also identified only a single strain of Nostoc within individual coralloid roots and in individual plants of indigenous cycads by sequencing of the tRNALeu (UAA) intron. They also found that plants growing in close proximity (female plants and their offspring) shared the same cyanobionts in their coralloid roots, suggesting a selective mechanism between the host plant and its cyanobionts. However, most previous studies used coralloid roots collected from botanical gardens and/or greenhouses. Zheng et al. (2002) used coralloid roots collected from cultivation in China and greenhouse specimens from Stockholm. The latter source exhibited lower cyanobiont species diversity. Gehringer et al. (2010) were the first to report the use of samples collected from their natural habitats. They collected coralloid roots from 31 Macrozamia species throughout their distribution range in Australia and sequenced the 16S rRNA genes. The data confirmed that coralloid roots harbored a single Nostoc strain, and the same strain was sometimes symbiotic with cycads in several locations. Host specialization of cyanobionts within the cycad genus Macrozamia was not observed in the wild.

In general, host specialization of the genus Nostoc is low (O'Brien et al., 2005). However, the water fern Azolla harbors unique cyanobionts: Papaefthimiou et al. (2008) analyzed the phylogeny of plural strains of symbiotic cyanobacteria and free-living cyanobacteria and found a robust cluster of Azolla symbionts, corroborating that the association of a unique cyanobacterial type with a host plant is strict and selective. They assumed that this may have been caused by coevolution and that Azolla symbiosis might have evolved in an aquatic environment.

Cycas revoluta Thunb. grows on rocky and sandy coasts of southern Japan, ranging from southern Kyushu to the most southern edge of the Ryukyu Islands (Fig. 1). Thus, distribution of C. revoluta is mostly isolated and scattered on islands and its range is wide, approximately 900 km in a northeast–southwest direction. This plant is a 1.5–5-m-tall dioecious tree that produces large seeds approximately 4 cm in length and 3 cm in width (Yamazaki, 1995). Because fertile seeds sink in seawater, seed dispersal of Crevoluta is thought to be limited, and dispersal over the sea is likely problematic. Phylogeography of this cycad is highly geographically structured: chloroplast and mitochondrial DNA haplotype analysis indicated two major areas, comprising northerly and southerly populations that were demarcated north and south of Okinoerabujima (Kyoda & Setoguchi, 2010). In addition, leaflet margin morphology also exhibited north–south differentiation (Setoguchi et al., 2009). The phylogeographic structure of C. revoluta is assumed to be determined by the formation and division of a landbridge between mainland Asia and Kyushu via the Ryukyu Islands during the Quaternary climatic oscillations (Kizaki & Oshiro, 1977, 1980; Ujiie, 1990; Kimura, 1996, 2000). However, no phylogeographic study of cyanobionts in Cycas coralloid roots has been conducted. Nostoc can be common in the environment, but the cycad cyanobionts might also exhibit a phylogeographic structure across the distribution range if they share the host plant's migration history. In contrast, multiple Nostoc strains can occur in the environment, and so cycad coralloid roots may use indigenous strain(s). In the latter case, no phylogeographic structure would be observed among cycad cyanobionts in the habitats of the Ryukyu Islands and Kyushu.

image

Figure 1. Coralloid root sampling locations. Locations of the nine sampling sites on the Ryukyu Islands and Kyushu. Population name abbreviations correspond to those in Table 1. (Inset) Location of the Ryukyu Islands and Kyushu.

Download figure to PowerPoint

We aimed to investigate the genetic diversity of cyanobionts within single or multiple coralloid root(s) of individual plants, among the coralloid roots of individual plants and within populations of C. revoluta. All materials were collected from natural populations. In addition, we studied the phylogeography of the coralloid root cyanobionts in relation to their host cycads across their distribution range in the Ryukyu Islands and Kyushu. We amplified cyanobacterial 16S rRNA genes, and genotyping was conducted by a combination of denaturing gradient gel electrophoresis (DGGE) and sequencing.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Sampling

The coralloid roots used in this study were collected from natural C. revoluta habitats across its entire distribution range. Details of sampling localities, coordinates, numbers of samples, and abbreviations of localities are presented in Fig. 1 and Table 1. Coralloid roots were excavated from the ground at less than 10 cm in depth and stored at 4 °C until DNA extraction.

Table 1. Sampling localities, number of coralloid roots and genotypes
NameLocalitiesLatitude (N)Longitude (E)Number of samplesNumber of genotypes
MtKushima City, Toi Cape (Miyazaki Pref.)31°21′131°20′2615
AaAmamioshima Isl., Ayamaru Cape (Kagoshima Pref.)28°28′129°42′129
AsAmamioshima Isl., Setouchi (Kagoshima Pref.)28°09′129°17′21
TkTokunoshima Isl., Kanami Cape (Kagoshima Pref.)27°53′128°58′41
OhOkinawa Isl., Hedo Cape (Okinawa Pref.)26°51′128°14–15′56
OsOkinawa Isl., Sezokojima (Okinawa Pref.)26°38′127°51′12
MyMiyakojima Isl., Ikema (Okinawa Pref.)24°55′125°14′52
IkIshigakijima Isl., Kabira (Okinawa Pref.)24°28′124°06′54
YkYonagunijima Isl., Kubura (Okinawa Pref.)24°27′122°56′61

At population Mt (Miyazaki: Toi Cape), two and nine coralloid roots were excavated from the same cycad individual and in a very narrow area (approximately < 4 m2), respectively, to detect genotype polymorphisms. In total, 66 coralloid roots were collected from nine populations.

DNA extraction and genotyping

The collected coralloid roots were initially washed in running water, then the surfaces were washed again for 10 min in distilled sterilized water using supersonic waves. Next, coralloid roots were sectioned at the apical and middle parts (see Fig. 2) following Zheng et al. (2002). Each part was cut in round slices using a sterile scalpel; then, the epidermal layers (including the nongreen part of the outer layer of the cortex) were removed with a sterile scalpel under a stereomicroscope. Finally, only the green-colored layer of the cortex (presumably containing cyanobionts) was recovered with a sterile needle.

image

Figure 2. Morphology of coralloid roots of Cycas revoluta. (a) Gross morphology and (b) tip of a coralloid root. Cyanobionts were collected from the apical (a) and middle (m) positions.

Download figure to PowerPoint

Cyanobionts from each section were collected in 555 μL of cetyltrimenthyl ammonium bromide (CTAB: 0.02% w/v, pH 8.0 with 1 M Tris-HCl buffer). They were homogenized, and DNA was extracted using the CTAB method (Doyle & Doyle, 1990). The extracted DNA was dissolved in 50 μL of TE buffer.

In addition, we extracted DNAs from ten C. revoluta leaves from ten populations from the Ryukyu Islands to examine DNA sequences of the cyanobacterial 16S rRNA gene using the same primers below (as controls).

The cyanobacterial 16S rRNA gene was used for genotyping of symbiotic cyanobionts in C. revoluta coralloid roots. In the first step, we conducted genotyping by DGGE analysis using ~ 400 bp of the 5′-side of the 16S rRNA gene. Mixtures of DNA from the apical and middle parts of coralloid roots were used to detect polymorphisms within single coralloid roots. In cases of polymorphic genotypes, we amplified and reanalyzed the same region using DNA from the apical and middle parts separately. Each band with different electrophoretic mobility was excised and subjected to direct sequencing. In addition, several samples were subjected to sequencing of ~700 bp of the 5′-side of the 16S rRNA gene, as described later.

In the first step, the cyanobacterial 16S rRNA gene was amplified using the forward primer 16S353F with a GC clamp at its 5′-end and reverse primers 16S781R(a) and 16S781R(b) (see Fig. 3 for primer positions). Primer information is presented in Table 2. Primers 16S359F and 16S781R, reported by Nübel et al. (1997), have been commonly used for specific amplification of the cyanobacterial 16S rDNA fragment (e.g. Boutte et al., 2006, 2008). PCR was conducted in a total reaction volume of 20 μL containing 6.9 μL of autoclaved ion-exchanged water, 10 μL of Ampdirect Plus (Shimadzu, Kyoto, Japan), 0.1 μL of Ex Taq (Takara Bio, Ohtsu, Japan), 1.0 μL of each primer, and 1.0 μL of template DNA. PCR was performed using the following protocol 1: 95 °C for 1 min, 20 cycles at 95 °C for 30 s, 65 °C (one cycle, this temperature fell 0.5 °C) for 30 s, and 72 °C for 60 s; 10 cycles at 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 60 s; and a final elongation cycle at 72 °C for 10 min. PCR products were then visualized on 0.5× TAE-agarose gels strained with ethidium bromide.

image

Figure 3. Positions of primers used for PCR and sequencing of the 16S rRNA gene and adjacent regions. The black and white triangles indicate forward and reverse primers, respectively.

Download figure to PowerPoint

Table 2. Primers information used in this study
PrimerSequence (5′–3′)Target siteReference
  1. Target site indicates Escherichia coli numbering of 16S or 23S rRNA nucleotides.

  2. 16S353F is reconstructed a forward primer CYA359F.

16S353FAGCAGTGGGGAATTTTCCGC353–372Ohkubo et al. (2006)
GC-16S353FGCclamp-AGCAGTGGGGAATTTTCCGC353–372Ohkubo et al. (2006)
16S359FGGGGAATTTTCCGCAATGGG359–378Nübel et al. (1997)
16S781R(a)GACTACTG G GGTATCTAATCCCATT781–805Nübel et al. (1997)
16S781R(b)GACTACAGGGGTATCTAATCCCTTT781–805Nübel et al. (1997)
16S1371RaGTTRCRGTAAYGACTTCGGGCRTGA1371–1395Murakami et al. (2004)
16S1371RbGTTRCRGTAAYGACTTCGGGCGTKG1371–1395Murakami et al. (2004)
16S1371RcGTTRCRGTAAYGACTTCGGGCWTGG1371–1395Murakami et al. (2004)
23S30RCTTCGCCTCTGTGTGCCTAGGT30–52Murakami et al. (2004)
GC clampCGCCCGCCGCGCCCCGCGCCGGTCCC
GCCGCCCCCGCCCG

The PCR products were used for DGGE analysis on a Dcode Universal Mutation Detection System (Bio-Rad Laboratories Inc., Hercules, CA), as described by Boutte et al. (2006). Each PCR product (15 μL) was applied directly onto a 9% (w/v) polyacrylamide gel in 0.5× TAE buffer (40 mM Tris-base, 20 mM acetic acid, 10 mM EDTA, pH 8.0) with a linear 30% to 60% denaturant gradient [100% denaturant solution was defined as 7 M urea and 40% (v/v) formamide]. The gels were run at a constant temperature of 60 °C at 100 V for 13 h. PCR products were visualized with SYBR Gold (Invitrogen, Carlsbad, CA). Bands that were visible by the naked eye were excised from the DGGE gel on a blue light transilluminator (UVP, Upland, CA).

Excised bands were kept in 50 μL of TE buffer at −30 °C for 1 h. Frozen liquid was kept at 95 °C for 10 min and the DNA eluted from the gel (gel and liquid containing DNA can be separated in a sampling tube). DNA was re-amplified using the primers 16S353F and 16S781R(a) or 16S781R(b) with protocol 1. PCR products were visualized on 0.5× TAE-agarose gels stained with ethidium bromide. The PCR products were sequenced using the standard methods of the BigDye Terminator Cycle Sequence Ready Reaction Kit (Applied Biosystems, Foster City, CA, USA) using the primers 16S353F, 16S359F, or 16S781R(a) and 16S781R(b) on an ABI 3130 Genetic Analyzer (Applied Biosystems).

The sequences obtained together with those of representative reference strains were aligned using clustal x (Thompson et al., 1997); end gaps were removed, and the ~ 300–350-bp sequences were realigned. A phylogenetic tree based on neighbor-joining (NJ) was generated from ~ 310-bp sequences using clustal x (Thompson et al., 1997) with 1000 bootstrap resampling events.

Identification of cyanobionts exhibiting unique DNA sequences

Because some samples were suggested in the phylogenetic analysis to nest outside of the Nostoc clade (Aa 10-7, Ik 1-17, Ik 5-20, Mt 11-26, and Mt 15-28), they were further amplified to obtain longer sequences using the forward primer 16S353F and the reverse primer 23S30R. PCR was conducted in a total reaction volume of 10 μL containing 1.6 μL of autoclaved ion-exchanged water, 5.0 μL of Multiplex (Qiagen, Hilden, Germany), 1.0 μL of each primer, and 1.4 μL of template DNA. PCR was performed using the following protocol: 95 °C for 15 min, 20 cycles at 95 °C for 30 s, 65 °C (by one cycle, this temperature fell 0.5 °C) for 90 s, and 72 °C for 2 min; 15 cycles at 95 °C for 30 s, 55 °C for 90 s, and 72 °C for 2 min; and a final elongation cycle at 72 °C for 10 min. PCR products were visualized on 0.5× TAE-agarose gels stained with ethidium bromide.

Amplified DNA was used as the template in the following nested PCR. PCR products were diluted with autoclaved ion-exchanged water (about 10 times volume) and subjected to nested PCR using the forward primer 16S353F and the reverse primers 16S1371Ra or 16S1371Rb, or 16S1371Rc [Primers 16S1371R(a), 16S1371R(b), and 16S1371R(c) are cyanobactrial specific primers that are different in four bases at 3′-end (see Table 2). Initially, we attempted 16S1371R(a) and if not succeeded in the amplification, we used 16S1371R(b) or 16S1371R(c). PCR was conducted in a total reaction volume of 11.05 μL containing 4.0 μL of autoclaved ion-exchanged water, 5.0 μL of Ampdirect Plus, 0.05 μL of Ex Taq, 0.5 μL of each primer, and 1.0 μL of template DNA. PCR was performed using the following protocol: 95 °C for 1 min; 20 cycles at 95 °C for 30 s, 65 °C (by one cycle, this temperature fell 0.5 °C) for 30 s, and 72 °C for 90 s; 10 cycles at 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 90 s; and a final elongation cycle at 72 °C for 10 min. After this, PCR products were visualized on 0.5× TAE-agarose gels stained with ethidium bromide. PCR products were sequenced from the both sides using the standard methods of the BigDye Terminator Cycle Sequence Ready Reaction Kit on an ABI 3130 Genetic Analyzer.

Phylogenetic analysis

The cyanobacterial 16S rRNA gene sequences obtained, together with those of representative reference strains (Papaefthimiou et al., 2008), were aligned using clustal x (Thompson et al., 1997). End gaps were removed, and the ~ 300–350-bp sequences were realigned. Chroococcidiopsis thermalis was selected as an outgroup based on phylogenies of nostocacean cyanobacteria (Papaefthimiou et al., 2008). In addition, we added 66 taxa as more distantly related outgroups according to molecular phylogeny of cyanobacteria by Schirrmeister et al. (2011). We further removed end gaps, and ~ 310-bp sequences were used for generating phylogenetic trees based on the NJ method using clustal x (Thompson et al., 1997) with 1000 replicates of bootstrap analysis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Identification of genotypes by DGGE

DGGE analysis of the ~ 420-bp PCR products (including ~ 390 bp of the cyanobiont 16S rRNA gene and GC clamped primers; that is, amplified using primers GC clamped-16S359F and 16S781R) exhibited a diverse fingerprinting pattern (Fig. 4a–d). Almost all samples shared a DNA fragment at the bottom of the electrophoretogram (these fragments were identified as chloroplast 16S rRNA genes, as described later). Most samples also exhibited one to three clear bands representing the cyanobiont 16S rRNA gene, while four samples (As 1, and My 2, 3, and 5) contained no such band. As the coralloid root tissue was green colored, suggesting the presences of cyanobionts, this result could have been because of PCR failure due either to low-quality cyanobacterial DNA or contamination by PCR inhibitors.

image

Figure 4. DGGE electrophoretic patterns of the 16S rRNA gene PCR products from cyanobionts collected from Cycas revoluta coralloid roots. The bands that were excised and sequenced are indicated by numbers adjacent to each band. The localities of coralloid roots are shown. The samples (Aa, Yk, Oh, Ik, Os), (Mt), (Tk), and (As, My) were analyzed separately, and their electrophoretic patterns are presented in a, b, c, and d, respectively. At site Mt, a star indicates that samples were collected from the same point. Double stars indicate that samples were collected from the same plant. A band of Mt-14 (b) was very obscure by the naked eye and we omitted to analyze.

Download figure to PowerPoint

In this study, genotypes of the cyanobacterial 16S rRNA gene were identified based on the position of each band and on their nucleotide sequence. We were able to determine sequences for all bands of cyanobacteria except band 10 of Yk2 and YK4, band 23 of Mt2 and Mt4, and band 31 of Mt19 and Mt21. As a result, almost all bands from the same position exhibited the same sequence, while bands 19 (Ik4), 20 (Ik5), and 21 (Os1) showed different genotypes (Fig. 4a). Each cyanobiont genotype is indicated as a number above the appropriate band in Fig. 4. Each DNA sequence was deposited in DDBJ/EMBL/GenBank under accession numbers AB612916AB612956.

In total, 41 genotypes were detected from 66 coralloid roots. Thirty-nine of the 66 (59.1%) coralloid roots contained one genotype, while eight (12.1%) contained two genotypes. Two samples from Okinawa Island (Oh 2 and 3; 3.0%) had three genotypes that are endemic to this population.

Although sampling size varied among the populations, most populations exhibited genotype diversity, while only one genotype was found on Yonagunijima Island (Yk 1-6). In contrast, different genotypes were identified from coralloid roots of the same cycad individual (Mt 16 and 17). In addition, nine neighboring coralloid roots collected within a narrow area (Mt 7-15) also contained five genotypes, suggesting the presence of polymorphic symbiotic cyanobionts. On the other hand, we found nine sequences for Cycas rRNA genes that appeared in the lowermost position in each electrophoretic pattern (Aa1, Yk1, Oh1, IK1, Os1, Mt1, Tk1, As1, and My1).

PCRs were also conducted using the same primer sets (GC clamped-16S359F and 16S781R) for DNAs extracted from 10 leaf samples, and single DNA fragment was located at the bottom at each sample in the DGGE electrophoretogram. The DNA sequences were identical with C. revoluta 16S rRNA gene sequences.

Genotype variation within a coralloid root and within a narrow habitat

DGGE fingerprinting suggested that 10 of the 66 samples exhibited polymorphic genotypes within a coralloid root. DGGE fingerprinting was conducted separately on DNA from the apical and middle parts of eight samples (Aa 2, Aa 5, Aa 11, Oh 1, Oh 2, Oh 3, Oh 5, and Os 1). DGGE fingerprinting patterns were identical between the apical (left) and middle (right) parts in six samples (Aa 5, Oh 1, Oh 2, Oh 3, Oh 5, and Os 1), while two samples (Oh 2 and Oh 3) harbored three genotypes at both the apical and middle parts of the coralloid roots (Fig. 5).

image

Figure 5. DGGE electrophoretic patterns of 16S rRNA gene PCR products from tips of coralloid roots. The left and right lanes of each sample show the apical (‘a' in Fig. 2) and middle (‘m’ in Fig. 2) positions, respectively.

Download figure to PowerPoint

Nine coralloid roots excavated from a narrow, dense cycad habitat harbored five cyanobacterial 16S rDNA genotypes, corroborating the presence of polymorphisms in symbiotic cyanobacteria within a natural population of host cycads.

Sequence and phylogenetic analysis

DNA fragments representing the 41 genotypes of the cyanobacterial 16S rDNA (~ 310 bp of the 5′-end) were excised from DGGE gels and sequenced. Phylogenetic trees (NJ trees with 1000 bootstrap replicates) using the obtained genotypes and addition of representative reference strains (strains of Nostoc, Anabaena, Cylindrospermopsis, Calothrix, and Trichormus) and chloroplast DNAs of C. revoluta, chlorella, tobacco, and black pine (Fig. 6) suggested no geographical structure within the 41 cyanobacterial 16S rDNA genotypes. Two clades in Fig. 6 (above bootstrap values of 90%) comprised allopatric genotypes spread across the Ryukyu Islands and Kyushu (clades I and II): for example, clade I encompassed the most northern and southern habitats [Toi Cape (Mt) to Ishigakijima Island (Ik), respectively] along with the islands of Amamioshima and Okinawa.

image

Figure 6. Neighbor-joining phylogenetic tree (divided to a and b) based on 16S rRNA gene sequences of cyanobionts obtained in this study and those from GenBank. CpDNA indicates a monophyletic clade comprising 16S rRNA sequences of chloroplast DNA of tobacco, black pine and Cycas revoluta and two isolated genotypes in this study. Red indicates the sequences obtained in this study, while blue indicates symbiotic Nostoc sequences from different hosts.

Download figure to PowerPoint

Additionally, these 41 genotypes were paraphyletic (indicated in red in Fig. 6). Twelve of the 41 genotypes and the lowermost bands in the DGGE electrophoretic patterns (indicated as Cycas rRNA in Fig. 6a) were clustered with chloroplast DNA sequences (tobacco, black pine, and C. revoluta) and located outside of the Nostoc clade (bootstrap support was at 96.8%). In particular, genotypes Oh 1-11 and Oh 1-12 were embedded within a robust cluster of cpDNA, while nine genotypes made an independent cluster. These nine genotypes were distinguished from C. revoluta chloroplast DNA sequences by between15 substitution changes (Ik 1-17 and Mt 15-28) and 19 substitution changes (As 10-7), with sequence similarities ranging from 93.9% to 95.2%. The 16S rRNA gene sequences for genotypes Ik 5-20 and Aa 10-7 (692 bp and 819 bp, respectively) were further determined using primers 23S30R and 16S1371R, and their sequence homology, with the C. revoluta chloroplast DNA, was found to range between 89.6% and 90.4%. The branch length of the combined clade (cpDNA denoted with a single asterisk in Fig. 6a) was longer than those of other clades within the phylogenetic tree.

The other 29 genotypes (71%), however, were embedded within the tree among symbiotic Nostoc (in blue in Fig. 6b) and free-living Nostoc, Anabaena, Trichormus, Cylindrospermopsis, Calothrix, and Nodularia. In particular, 20 of the 29 genotypes were clustered with free-living Nostoc (including soil, sand, and meadow) in Clade II. Multiple strains of symbiotic cyanobacteria from single coralloid roots were included (e.g. Oh 2-13, Oh 2-14, and Oh 2-15). In addition, genotypes from a single coralloid root (Aa 2-2 and Aa 2-3) clustered with both symbiotic Nostoc and free-living Nostoc clades.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Diversity of symbiotic cyanobacteria within and among coralloid roots

In this study, DGGE analyses and DNA sequence analysis of the partial 16S rRNA gene suggested the presence of diversity within single coralloid roots of C. revoluta obtained from natural habitats. As described previously, the existence of polymorphism within single coralloid roots has been controversial (Zheng et al., 2002; Costa et al., 2004). Thus, it is noteworthy that the present study confirmed that a single coralloid root can harbor multiple symbiotic cyanobacterial strains; a maximum of three Nostoc strains were detected in a single coralloid root. Two hypotheses may explain the cyanobacterial diversity observed in this study. First, cyanobionts were sampled from cycads growing in their natural habitat. In previous studies, most cyanobionts were collected from cycads growing in gardens or greenhouses. Zheng et al. (2002) indicated that symbiotic cyanobacteria from cultivated cycads exhibited low genotypic diversity. Second, the coexistence of multiple symbiotic cyanobacterial strains in a single coralloid root may be unusual. We detected a polymorphism rate of 15.1% among 66 coralloid roots although some genotypes were included within the cpDNA clade. In the case of cycad populations at Ayamaru Cape, only one of the 12 coralloid roots examined (8.3%) harbored multiple cyanobionts. On the other hand, previous studies analyzed only a few coralloid roots at each locality, suggesting that polymorphisms may have been overlooked.

We also detected multiple strains of symbiotic cyanobacteria from plural coralloid roots of the same individual. Coralloid roots of Mt 16 and 17 (Toi Cape) developed on the same root system of a cycad tree; however, these exhibited different genotypes, indicating that the symbiotic cyanobionts of coralloid roots of the same host plant sometimes differ. In addition, plural strains of symbiotic cyanobacteria were present within a very narrow area. Nine coralloid roots excavated from a small population in a narrow area (< 4 m2) harbored four symbiotic Nostoc strains (because many cycads were growing in this small area, we could not determine the host individuals). These genotypes were apparently paraphyletic, implying that plural strains of symbiotic cyanobionts inhabited the root systems of the C. revoluta population (genotypes Mt 7-24, Mt 8(9,10)-25 and Mt 13-27 were clustered within the Nostoc clade). Costa et al. (2004) found that neighboring cycad trees shared the same Nostoc strain. They estimated that the identical cyanobiont genotype accounts for the lower diversity of Nostoc strains in sandy soil environments and kinship among the host trees. However, our data did not support their conclusion, although we could not estimate the kinship of cycad trees in the studied area based on genetic markers.

Contrary to this finding, coralloid roots from different cycads at different points within Toi Cape (Mt 1, Mt 2, Mt 3, Mt 4, and Mt 5) shared the same cyanobiont genotype (Fig. 4b). This indicates that the same Nostoc strain was sometimes symbiotic with coralloid roots in different locations, as suggested by Gehringer et al. (2010), corroborating the wide distribution of a single genotype across the islands.

Phylogeny of genotypes and geographical structure

Phylogenetic analysis of 41 cyanobiont genotypes suggested the presence of two clades among representative reference strains. Twelve of the 41 genotypes and the lowermost bands in DGGE electrophoretic patterns (indicated as Cycas rRNA in Fig. 6a) were clustered with chloroplast DNA sequences (tobacco, black pine, and C. revoluta) and located outside of the Nostoc clade. In particular, genotypes Oh 1-11 and Oh 1-12 were almost identical with sequences of C. revoluta cpDNA, possibly due to amplification of the cpDNA 16S rRNA gene. PCR products using DNAs from leaves were also identical with the cpDNA 16S rRNA gene. Thus, we should identify these genotypes and lowermost bands of each DGGE electrophoretic pattern as 16S rRNA genes of C. revoluta chloroplast DNA. The nine genotypes in clade I (Fig. 6a) may be chimeric DNA fragments, that is, hybrid products of chloroplast and cyanobacterial sequences.

Twenty-nine of the 41 genotypes (71%) were embedded within the phylogenetic tree, with positions among both symbiotic Nostoc and free-living Nostoc, Anabaena, Trichormus, Cylindrospermopsis, Calothrix, and Nodularia. In particular, 20 of the 29 genotypes were clustered with free-living Nostoc (including soil, sand, and meadow) in Clade II (Fig. 6b). Multiple strains of symbiotic cyanobacteria from single coralloid roots were included in this clade (e.g. Oh 2-13, Oh 2-14 and Oh 2-15). In addition, genotypes Aa 2-2 and Aa 2-3, from a single coralloid root, clustered with both symbiotic Nostoc and free-living Nostoc clades. Thus, symbiotic Nostoc in C. revoluta may exhibit inapparent relationship with Nostoc strains (e.g. symbiotic or free-living), while Azolla harbors specific strains of symbiotic Nostoc (Papaefthimiou et al., 2008).

Most genotypes were identified as Nostoc strains. The phylogenetic tree suggests conversion of Nostoc strains between symbiotic and free-living individuals in soil and water, and the 29 cyanobiont strains in Fig. 6b were scattered among these reference strains. Land plants, including C. revoluta, may therefore take up Nostoc strains from the soil at random, with no specificity between cycads and cyanobionts, as has been discussed elsewhere (Lindblad et al., 1989; Lotti et al., 1996; Costa et al., 1999, 2004; Zheng et al., 2002; Papaefthimiou et al., 2008; Gehringer et al., 2010). The Nostoc flora in each environment may be diverse with a wide global distribution.

Thus, our data suggest that the genotypes of symbiotic cyanobionts in C. revoluta coralloid roots exhibit no geographical structure, although the host plant exhibits north–south differentiation across its distribution range (Setoguchi et al., 2009; Kyoda & Setoguchi, 2010). Cycas revoluta can establish symbiosis with one or several strains of Nostoc within each habitat. Note that 20 of the 29 genotypes of C. revoluta-symbiotic Nostoc were included in one clade that also comprised six free-living Nostoc from soil (clade II in Fig. 6b). Nevertheless, this clade was not supported by a high bootstrap value. Most free-living Nostoc were isolated from soil in tropical regions (Southeast Asia, South America, and Africa), suggesting that some strains of C. revoluta-symbiotic Nostoc may be derived from free-living Nostoc in soil. Further study of the infection mechanisms of cyanobionts during the early developmental stages of coralloid roots are needed to elucidate their genetic signature.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the anonymous reviewers for their helpful comments on the manuscript. We also thank Mr. Yu Akita for permission to work within the national park at Toi Cape, and Mr. Y. Maeda for allowing us to collect coralloid roots on Amamioshima Island. This study was supported by Grants-in-Aid for Scientific Research (#22405013 and #21247005) from the Ministry of Education, Culture, Science, Sports and Technology, Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Bergman B , Ray AN & Rasmussen U (1996) Chemical signaling in cyanobacterial-plant symbioses. Trends Plant Sci 1: 191197.
  • Boutte C , Grubisitic S , Balthasart P & Wilmotte A (2006) Testing of primers for the study of cyanobacterial molecular diversity by DGGE. J Microbiol Methods 65: 542550.
  • Boutte C , Mankiewicz-Boczek J , Komarkova J , Grubisic S , Izydorczyk K , Wautelet F , Jurczak T , Zalewski M & Wilmotte A (2008) Diversity of planktonic cyanobacteria and microcystin occurrence in Polish water bodies investigated using a polyphasic approach. Aquat Microb Ecol 51: 223236.
  • Costa JL & Lindblad P (2002) Cyanobaceria in symbiosis with cycads. Cyanobacteria in Symbiosis ( Ray AN , Bergman B & Rasmussen U , eds), pp. 195205. Kluwer Academic Publishers, Dordrecht.
  • Costa JL , Paulsrud P & Lindblad P (1999) Cyanobiont diversity within coralloid roots of selected cycad species. FEMS Microbiol Ecol 28: 8591.
  • Costa JL , Romero EM & Lindblad P (2004) Sequence based data supports a single Nostoc strain in individual coralloid roots of cycads. FEMS Microbiol Ecol 49: 481487.
  • Doyle JJ & Doyle JL (1990) Isolation of plant DNA from fresh tissue. Focus 12: 1315.
  • Gehringer MM , Pengelly JJL , Cuddy WS , Fieker C , Forster PI & Neilan BA (2010) Host selection of symbiotic cyanobacteria in 31 species of the Australian cycad genus: Macrozamia (Zamiaceae). Mol Plant Microbe Interact 23: 811822.
  • Grobbelaar N , Scott WE , Hattingh W & Marshall J (1986) The occurrence of coralloid roots on the South African species of the Cycadales and their ability to fix nitrogen symbiotically. S Afr J Bot 52: 467471.
  • Kimura M (1996) Quaternary paleogeography of the Ryukyu Arc. J Geogr 105: 259285.
  • Kimura M (2000) Paleogeography of the Ryukyu Islands. Tropics 10: 524.
  • Kizaki K & Oshiro I (1977) Palaeogeography of the Ryukyu Islands (in Japanese with English summary). Mar Sci 9: 542549.
  • Kizaki K & Oshiro I (1980) The origin of the Ryukyu Islands. Natural History of the Ryukyus (in Japanese) ( Kizaki K , ed), pp. 137. Tsukiji-shokan, Tokyo.
  • Komarek J & Anagnostidis K (1989) Modern approach to the classification system of Cyanophytes 4-Nostocales. Arch Hydrobiol Suppl 82: 247345.
  • Kyoda S & Setoguchi H (2010) Phylogeography of Cycas revoluta Thunb. (Cycadaceae) on the Ryukyu Islands: very low genetic diversity and geographical structure. Plant Syst Evol 288: 177189.
  • Lindblad P , Haselkorn R , Bergman B & Nierzwicki-Bauer SA (1989) Comparison of DNA restriction fragment length polymorphisms of Nostoc strains in and from cycads. Arch Microbiol 152: 2024.
  • Lotti F , Giovanetti L , Margheri MC , Ventura S & Materassi R (1996) Diversity of DNA methylation pattern and total DNA restriction pattern in symbiotic Nostoc . World J Microbiol Biotechnol 12: 3842.
  • Murakami A , Miyashita H , Iseki M , Adachi K & Mimuro M (2004) Chlorophyll d in an epiphytic cyanobacterium of red algae. Science 303: 1633 (Supporting online material) www.sciencemag.org/cgi/content/full/303/5664/1633/DC1.
  • Nübel U , Garcia-Pichel F & Muyzer G (1997) PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl Environ Microbiol 63: 33273332.
  • O'Brien H , Miadlikowska J & Lutzoni F (2005) Assessing host specialization in symbiotic cyanobacteria associated with four closely related species of the lichen fungus Peltigera . Eur J Phycol 40: 363378.
  • Obukowicz M , Schaller M & Kennedy GS (1981) Ultrastructure and phenolic histochemistry of the Cycas revoluta-Anabaena symbiosis. New Phytol 87: 751754.
  • Ohkubo S , Miyashita H , Murakami A , Takeyama H , Tsuchiya T & Mimuro M (2006) Molecular detection of epiphytic Acaryochloris ssp. on marine macroalgae. Appl Environ Microbiol 72: 79127915.
  • Papaefthimiou D , Hrouzek P , Mugnai MA , Lukesova A , Turicchia S , Rasmussen U & Ventura S (2008) Differential patterns of evolution and distribution of symbiotic behaviour in nostocacean cyanobacteria. Syst Evol Microbiol 58: 553564.
  • Rai AN , Sonderback E & Bergman B (2000) Cyanobacterium-plant symbioses. New Phytol 147: 449481.
  • Schirrmeister BE , Antonelli A & Bagheri HC (2011) The origin of multicellularity in cyanobacteria. BMC Evol Biol 11: 45.
  • Setoguchi H , Kyoda S , Maeda Y & Nomura N (2009) Geographical variation in leaflet of Cycas revoluta (Cycadaceae) on the Ryukyu Islands. J Phytogeogr Taxon 57: 16.
  • Thompson JD , Gibson TJ , Plewniak F , Jeanmougin F & Higgins DG (1997) The CLUSTALX windows interface: flexible strategies for multiple sequence alignment aided by quality Analysis tools. Nucleic Acids Res 25: 48764882.
  • Ujiie H (1990) Geological history of the Ryukyu Island Arc. Nature of Okinawa Geomorphology and Geology ( Ujiie H , ed), pp. 251255. Hirugisha, Naha.
  • Yamazaki T (1995) Cycadaceae. Flora of Japan, Vol. 1 ( Iwatsuki K , Yamazaki T , Boufford DE & Ohba H , eds), pp. 263. Kodansha, Tokyo.
  • Zheng W , Song T , Xiaodong B , Bergman B & Rasmussen U (2002) High cyanobacterial diversity in coralloid roots of cycads revealed by PCR fingerprinting. FEMS Microbiol Ecol 40: 215222.
  • Zhu C (1982) Fine structure of blue-green algae and the cells lined along the endophyte cavity in the coralloid root of Cycas . Acta Bot Sin 24: 109114.