Edited by: Lasse Riemann, University of Copenhagen, Denmark
Reviewed by: Nicole Webster, Australian Institute of Marine Science, Australia; Claudia Dziallas, University of Copenhagen, Denmark
*Correspondence: Neus Garcias-Bonet, Department of Global Change Research, IMEDEA (CSIC-UIB), Miquel Marquès 21, 07190 Esporles, Spain. e-mail:
This article was submitted to Frontiers in Aquatic Microbiology, a specialty of Frontiers in Microbiology.
This is an open-access article distributed under the terms of the
Bacterial endophytes are crucial for the survival of many terrestrial plants, but little is known about the presence and importance of bacterial endophytes of marine plants. We conducted a survey of the endophytic bacterial community of the long-living Mediterranean marine angiosperm
Bacteria are commonly found living endophytically within plant tissues (e.g., Hallmann and Berg,
Seagrasses are marine clonal angiosperms that evolved from freshwater angiosperm ancestors that colonized the marine environment in the Cretaceous (den Hartog,
The interest in exploring the endophytic bacterial community of
Here we describe the bacterial communities associated with surface-sterilized tissues (roots, rhizomes, leaves) collected in summer in 26 meadows of
Surface-sterilized plant material (100 mg of fresh tissue) was ground with the help of a sterilized pestle. The total nucleic acid extraction was performed using a commercial kit specific for plant tissues (Partec®). Nucleic acid extracts were stored at −20°C until amplification. The DNA extract, containing plant and endophyte DNA when present, was amplified by standard PCR with primers 907R (5′-CCG TCA ATT CCT TTG AGT TT-3′) and 341F-GC containing a 40 bp GC clamp at the 5′ end (5′-CGC CCG CCG CGC CCC GCG CCC GGC CCG CCG CCC CCG CCC C/CC TAC GGG AGG GAG CAG-3′) specific for the bacteria domain (Muyzer and Smalla,
The amplification products of the fragment of the 16S ribosomal RNA gene (1 μg of PCR product) were separated by DGGE in a 6% polyacrylamide gel containing a gradient of denaturants ranging from 40 to 70% (where 100% is 7 M urea and 40% formamide). Gels were run for 18 h at 150 V in 1X TAE (Tris-Acetate-EDTA) buffer at 60°C in a CBS Scientific Co., DGGE system. Following electrophoresis, the gels were stained with SybrGold for 30 min in the dark and photographed using a G:BOX imaging system (Syngene). All the detectable bands were excised and stored frozen in autoclaved MiliQ water at −20°C for further processing.
The digital images of DGGE gels were analyzed by measuring the relative migration of each band, normalized to the migration of the 16S rDNA band corresponding to
Species accumulation curves (i.e., accumulated increase of the number of detected OTUs vs. number of samples) were constructed in R (R Development Core Team,
A binary matrix (presence/absence) was constructed for all of the identified OTUs in order to determine the similarity among samples and locations. Using the information about presence/absence of each OTU in different tissues of
Once the weighted bipartite networks for each location were constructed, we collapsed the networks of locations at the same island (Cabrera, Formentera, Ibiza, and Mallorca) in order to obtain the weighted bipartite network for each island. By collapsing the networks of all locations, we obtained the weighted bipartite network for the Balearic Archipelago.
We compared the bipartite networks of each island using the concept of the distance between networks with the same number of nodes, as described by Andrade et al. (
We used a bootstrap strategy to examine the robustness of the network analysis. We randomly removed one node of the networks to be compared and computed the distance between them. After repeating this procedure for each node, we calculated the average of the distances computed for each pair of networks. This average was considered as the best estimate of the average distance between any pair of networks, and the procedure was repeated for each pair of networks (each pair of locations).
We used the Girvan–Newman algorithm (Girvan and Newman,
The binary matrix was also used to generate a distance matrix based on Jaccard's coefficient as the basis for a non-metric multidimensional scaling (NMDS) diagram using package vegan in R. We performed an Analysis of Similarity (ANOSIM) using the vegan package (10,000 permutations), to test for the existence of differences in band patterns among tissue groups defined as roots, rhizomes, and leaves. The R value generated by ANOSIM test indicates the magnitude of difference among groups, where an
Finally, we performed an indicator species test (Dufrene and Legendre,
The detected and excised bands (OTUs) from the DGGEs were reamplified using the same pair of primers (907R and 341F-GC). The amplification products were cleaned and purified from primers and dNTPs by an enzymatic reaction with a mixture of Exonuclease I (1 U/reaction) and Alkaline Phosphatase (1 U/reaction) at 37°C during 60 min, followed by an enzyme denaturing step at 72°C for 15 min. The DNA was precipitated using isopropanol (66% final concentration), centrifuged (10,000× g, 15 min), washed with 66% isopropanol and resuspended in sterile water. The resulting DNA concentration was measured fluorometrically (Qubit®, Invitrogen) and 150 ng of the amplified product was used for the sequencing reaction using the reverse primer 907R. The sequencing was performed by Secugen, using the chemistry BigDye® Terminator v3.1. The sequences of about 500 bp were checked for existence of chimeras using the Bellerophon tool available at
The sequences obtained in this study have been deposited in Genbank under the accession numbers JF292432 to JF292446.
A total of 34 different OTUs were identified in DGGE profiles from all plant tissue samples (
The species accumulation curves (Figure
All | 186 | 34 | 34.12 ± 0.44 | 34.99 ± 0.99 | 35.04 ± 0.98 | 99.65 | 97.16 | 97.03 |
Leaves | 62 | 24 | 24.1 ± 0.38 | 24.98 ± 0.98 | 25.09 ± 1.28 | 99.59 | 96.06 | 95.64 |
Rhizomes | 57 | 28 | 28.29 ± 0.68 | 29.96 ± 1.38 | 29.89 ± 1.67 | 98.97 | 93.44 | 93.67 |
Roots | 67 | 28 | 40.5 ± 17.14 | 32.93 ± 2.61 | 30.08 ± 1.34 | 69.14 | 85.04 | 93.08 |
The bipartite network analysis showed differences in the band patterns among islands (Figure
The community analysis of the bipartite network of all the Balearic Islands studied, obtained by running the Girvan–Newman algorithm, and identified three different communities for each tissue type (Figure
Although NMDS did not show clear differences among tissues (data not shown), ANOSIM test confirmed that band patterns among tissues were different with statistically significance, although these differences were small, suggesting other variables play a role in the endophytic bacterial composition of
Moreover, the indicator species analysis identified some OTUs characteristic of each tissue (Table
OTU_33 | 5 | 9 | 13 | 27 | Leaves | 0.0994 | 0.186 |
OTU_1 | 0 | 0 | 2 | 2 | Leaves | 0.0323 | 0.209 |
OTU_5 | 0 | 0 | 2 | 2 | Leaves | 0.0323 | 0.216 |
OTU_7 | 6 | 1 | 7 | 14 | Leaves | 0.0579 | 0.232 |
OTU_16 | 5 | 5 | 8 | 18 | Leaves | 0.0571 | 0.482 |
OTU_6 | 12 | 7 | 12 | 31 | Leaves | 0.0756 | 0.63 |
OTU_11 | 3 | 5 | 6 | 14 | Leaves | 0.0408 | 0.698 |
OTU_31 | 0 | 2 | 0 | 2 | Rhizomes | 0.0351 | 0.074 |
OTU_23 | 1 | 3 | 0 | 4 | Rhizomes | 0.041 | 0.079 |
OTU_17 | 0 | 2 | 0 | 2 | Rhizomes | 0.0351 | 0.095 |
OTU_21 | 1 | 2 | 0 | 3 | Rhizomes | 0.0246 | 0.199 |
OTU_25 | 2 | 3 | 2 | 7 | Rhizomes | 0.0241 | 0.675 |
OTU_9 | 1 | 2 | 2 | 5 | Rhizomes | 0.015 | 0.875 |
OTU_8 | 7 | 2 | 2 | 11 | Roots | 0.0635 | 0.104 |
OTU_32 | 3 | 0 | 0 | 3 | Roots | 0.0448 | 0.136 |
OTU_18 | 11 | 3 | 7 | 21 | Roots | 0.0818 | 0.157 |
OTU_20 | 11 | 3 | 9 | 23 | Roots | 0.0745 | 0.288 |
OTU_22 | 5 | 0 | 4 | 9 | Roots | 0.04 | 0.289 |
OTU_13 | 18 | 12 | 11 | 41 | Roots | 0.1099 | 0.405 |
OTU_28 | 5 | 4 | 3 | 12 | Roots | 0.0288 | 0.921 |
OTU_3 | 1 | 0 | 0 | 1 | Roots | 0.0149 | 1 |
We sequenced approximately 200 bands detected by DGGE analysis, trying to cover all identified OTUs. However, we only managed to obtain 12 different bacterial sequences. Totally 33.3% of the sequences analyzed belonged to Bacteroidetes, while the rest (66.7%) belonged to the class Proteobacteria: 41.7% were affiliated to the α-subclass, 16.7% to the γ-subclass, and 8.3% to the δ-subclass. More specifically, 15.4% of the sequences belonged to the
The results reported here provide a pioneering step toward the characterization of the endophytic bacterial community associated with tissues of a marine angiosperm, by both comparing DGGE band patterns and sequencing the main OTUs found. Our results show that endophytic bacteria are frequently present in tissues of
Whereas our study appeared to yield a thorough inventory of OTUs in tissues of
Our estimates of bacterial endophyte richness are influenced by the choice of DGGE in our survey. Much higher numbers of OTUs could be expected from large cloning efforts or from the use of massively parallel sequencing techniques. Webster et al. (
In contrast, most of the ribotypes reported by deep-sequencing studies are present in a very low abundance with only one or a few sequences out of several thousands or more (Webster et al.,
The comparison of patterns in endophytic bacterial communities between tissues suggested that bacteria associated with roots differ from those associated with rhizomes and leaves, similar to what was found among rice tissues (García de Salamone et al.,
The sequencing of the main OTUs detected by DGGE analysis allowed us to draw the first identification of the endophytic bacterial community in
The identification of bacteria similar in sequence to those found in diseased coral tissues opens a new and exciting research line, as there is no evidence, to our knowledge, of specific bacterial pathogens of seagrasses. However, demonstrating the pathogenicity of these organisms will require further research, involving the isolation of the potential causative agents and demonstrating that they fulfill Koch's postulates. Some of these bacteria found in diseased corals have been identified in association with macroalgae without relation to disease (Table
Similarly, we identify bacteria similar to sequences found endophytically in other plants and related to
In summary, this work is the first characterization of endophytic bacterial community in
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This study was funded by the “Fundación BBVA (PRADERAS),” the Spanish Ministry of Environment (project 055/2002) and the Spanish Ministry of Science (MEDEICG, reference No. CTM2009-07013). We thank T.S. Moore for his help in correcting the manuscript. Neus Garcias-Bonet was funded by a scholarship from the Government of the Balearic Islands.
P.o_rhiz_3b (JF292432) | Rhizome/C.Sahona | 494 | 99 | Uncultured bacteria (EU181012) | Isolated from seagull fecal sample | Lu et al., |
Non-Identified Coral BBD isolates/α-proteobacteria |
99 | Uncultured bacterium clone (GU118071) | Isolated from the coral |
Sunagawa et al., |
||||
99 | Uncultured bacterium clone (FJ202069) | Isolated from the coral |
Sunagawa et al., |
||||
99 | Uncultured bacterium clone (EF123439) | Isolated from Black Band Diseased tissues of the coral |
Sekar et al., |
||||
99 | Rhizobiales; Cohaesibacteraceae. Isolated from sediment of a seawater pond used for sea cucumber culture | Qu et al., |
|||||
P.o_rhiz_6b (JF292433) | Rhizome/Figueral | 433 | 97 | Uncultured bacteria (EU181012) | Isolated from seagull fecal sample | Lu et al., |
Non-Identified Coral BBD isolates/α-proteobacteria |
97 | Uncultured bacterium clone (GU118071) | Isolated from the coral |
Sunagawa et al., |
||||
97 | Uncultured bacterium clone (FJ202069) | Isolated from the coral |
Sunagawa et al., |
||||
97 | Uncultured bacterium clone (EF123439) | Isolated from Black Band Diseased tissues of the coral |
Sekar et al., |
||||
97 | Rhizobiales; Cohaesibacteraceae. Isolated from sediment of a seawater pond used for sea cucumber culture | Qu et al., |
|||||
P.o_rhiz_23b (JF292434) | Rhizome/Talamanca | 518 | 92 | Uncultured δ-proteobacteria (AY133092) | Isolated from TCE-contaminated site | Carrol and Zinder, Unpublished | |
92 | Uncultured bacterium clone (GU118736) | Isolated from the coral |
Sunagawa et al., |
||||
91 | Bacterium enrichment culture clone (HQ622261) | Polluted estuarine sediment | Abed et al., Unpublished | ||||
91 | Desulfarculales; Desulfarculaceae; Desulfarculus | Lucas et al., Unpublished | |||||
P.o_root_15c (JF292435) | Root/C.Marmacen | 471 | 96 | Rhodobacterales; Rhodobacteraceae; Roseovarius | Jeanthon et al., Unpublished | ||
96 | Rhodobacterales; Rhodobacteraceae; Pelagibaca | Yuan et al., Unpublished | |||||
96 | Rhodobacterales bacterium (HQ537377 + HQ537273) | Isolated from 75 m depth on C-MORE BLOOMER cruise, Hawaii Ocean Time Series (HOT) station ALOHA“ | Sher et al., Unpublished | ||||
96 | Uncultured Rhodobacterales bacterium (GU474886) | Isolated from Hawaii Oceanographic Time-series study site ALOHA“ | Rich et al., |
||||
96 | Rhodobacterales; Rhodobacteraceae; Marinovum | Pradella et al., |
|||||
P.o_root_26c (JF292436) | Root/Magalluf | 520 | 94 | Nitrogen Fixing Bacteria isolated from stuarine grasses |
Cramer et al., |
||
93 | Alteromonadales. Halophilic denitrifying bacteria isolated from water brine in Siberian permafrost | Shcherbakova et al., Unpublished | |||||
92 | Uncultured |
Alteromonadales; Alteromonadaceae; Agarivorans | Dahle et al., |
||||
92 | Alteromonadales; Alteromonadaceae; Agarivorans. Isolated from surface of seaweeds | Du et al., |
|||||
P.o_rhiz_0BDSB (JF292437) | Rhizome/C.Torreta | 391 | 91 | Uncultured bacterium clone (GU946163) | Isolated from agricultural soil | Ros et al., Unpublished | |
90 | Bacteroidetes; Sphingobacteria; Sphingobacteriales; Sphingobacteriaceae; Pedobacter. Isolated from lake water | Berg et al., |
|||||
90 | Bacteroidetes; Sphingobacteria; Sphingobacteriales; Sphingobacteriaceae; Pedobacter. Isolated from fresh water | Baik et al., |
|||||
P.o_rhiz_0BDSD (JF292438) | Rhizome/C.Torreta | 412 | 96 | Uncultured bacterium clone (HM125351) | Isolated from soils | Bissett, Unpublished | |
95 | Uncultured bacterium (AM158409) | Isolated from |
Saenz de Miera et al., Unpublished | ||||
95 | Bacterium (FJ654260) | Isolated from soil | Kim et al., Unpublished | ||||
95 | Uncultured bacterium clone (GU946163) | Isolated from agricultural soil | Ros et al., Unpublished | ||||
95 | Bacteroidetes; Sphingobacteria; Sphingobacteriales; Sphingobacteriaceae; Pedobacter | Im, Unpublished | |||||
P.o_rhiz_0BDSG (JF292439) | Rhizome/C.Torreta | 368 | 92 | Unidentified bacterium clone (EF606109) | Isolated from rhizosphere soil from former arable field sown with low seeds diversity | Kielak et al., |
|
92 | Uncultured bacterium clone (HM066499) | Isolated from environmental sample | Gray and Engel, Unpublished | ||||
92 | Uncultured bacterium clone (AB583099) | Isolated from soybean leaf | Ikeda et al., |
||||
92 | Rhizobiales; Rhizobiaceae; Rhizobium/Agrobacterium group; Agrobacterium. Isolated from a lake | Sahay et al., Unpublished | |||||
92 | Rhizobiales; Rhizobiaceae; Rhizobium/Agrobacterium group; Agrobacterium. Isolated from a plant, endophytic microbiota | Zheng and Feng, Unpublished | |||||
P.o_rhiz_0BDSJ (JF292440) | Rhizome/Porto Colom | 404 | 97 | Rhodobacterales; Rhodobacteraceae; Nautella. Bacteria associated with sponges | Feby and Nair, Unpublished | ||
97 | Rhodobacterales; Rhodobacteraceae; Nautella. Isolated from surface seawater | Cho and Hwang, |
|||||
95 | Uncultured bacterium clone (GU472165) | Isolated from BBD affected corals | Arotsker et al., Unpublished | ||||
97 | Rhodobacterales; Rhodobacteraceae; Ruegeria. Isolated from surface of the red macroalgae, |
Case and Kjelleberg, Unpublished | |||||
97 | Rhodobacteraceae bacterium (FJ937900) | Rhodobacterales; Rhodobacteraceae. Isolated from |
Li et al., Unpublished | ||||
97 | Uncultured bacterium clone (FJ202604) | Isolated from |
Sunagawa et al., |
||||
P.o_leaf_0BDVV (JF292441) | Leaf/Pujols | 436 | 80 | Uncultured gamma proteobacterium clone (DQ269096) | Isolated from surface of marine macro-alga |
Longford et al., Unpublished | Sulfur-Oxidizing-Symbionts/γ-proteobacteria |
80 | Uncultured gamma proteobacterium clone (FJ205337) | Isolated from deep marine sediments | Dong and Shao, Unpublished | ||||
80 | Uncultured bacterium clone (EU491600 + EU491489 + EU491463) | Isolated from seafloor lavas from the East Pacific Rise | Santelli et al., |
||||
80 | Uncultured gamma proteobacterium (AB611274) | Isolated from abdominal setae of galatheid crab ( |
Yoshida-Takashima et al., Unpublished | ||||
80 | Uncultured gamma proteobacterium clone (AY534017) | Isolated from oxic surface sediments of eastern Mediterranean Sea | Polymenakou et al., |
||||
P.o_leaf_0BDVW (JF292442) | Leaf/Pujols | 467 | 95 | Sphingobacteriales bacterium (FJ952766) | Isolated from healthy tissue of coral |
Rypien et al., |
|
95 | Isolated from |
Zhang et al., Unpublished | |||||
95 | Isolated from natural subtidal biofilm | Huang et al., Unpublished | |||||
95 | Uncultured bacterium clone (EF433127) | Isolated from |
Barneah et al., |
||||
95 | Bacteroidetes; Cytophagia; Cytophagales; Flammeovirgaceae; Flammeovirga. | Lu, Unpublished | |||||
Isolated from a marine gastropod mollusk |
|||||||
P.o_leaf_0BDVX (JF292443) | Leaf/Pujols | 447 | 87 | Isolated from |
Zhang et al., Unpublished | ||
86 | Sphingobacteriales bacterium (FJ952766) | Isolated from healthy tissue of coral |
Rypien et al., |
||||
86 | Isolated from natural subtidal biofilm | Huang et al., Unpublished | |||||
86 | Uncultured bacterium clone (EF433134) | Isolated from |
Barneah et al., |
||||
86 | Isolated from a marine gastropod mollusk |
Zhao, Unpublished |