Microbial diversity in ostrich ceca as revealed by 16S ribosomal RNA gene clone library and detection of novel Fibrobacter species

Anaerobe 16 (2010) 83–93

Contents lists available at ScienceDirect

Anaerobe journal homepage: www.elsevier.com/locate/anaerobe

Ecology/environmental microbiology

Microbial diversity in ostrich ceca as revealed by 16S ribosomal RNA gene clone library and detection of novel Fibrobacter species Hiroki Matsui*, Yuko Kato, Tohru Chikaraishi, Masanori Moritani, Tomomi Ban-Tokuda, Masaaki Wakita Graduate School of Bioresources, Mie University, 1577 Kurimamachiya-cho, Tsu, Mie 514-8507, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 March 2009 Received in revised form 2 July 2009 Accepted 7 July 2009 Available online 24 July 2009

The ostrich (Struthio camelus) is a herbivorous bird and although the hindgut is known as the site for fiber digestion, little is known about the microbial diversity in the ostrich hindgut. Our aim was to analyze the microbial diversity in ostrich ceca using a 16S ribosomal RNA gene (rDNA) clone library approach. A total of 310 clones were sequenced and phylogenetically analyzed and were classified into 110 operational taxonomy units (OTUs) based on a 98% similarity criterion. The similarity of the sequences ranged from 86 to 99% and 95 OTUs had less than 98% similarity to the sequences in the public databases. Coverage and the Shannon–Wiener index (H0 ) of the library were 83.9% and 4.29, respectively. The sequences were assigned to the following 6 phyla: Firmicutes (50.9% of the total number of sequences), Bacteroidetes (39.4%), Fibrobacteres (6.5%), Euryarchaeota (1.9%), Spirochaetes (1.0%), and Verrucomicrobia (0.3%); approximately 90% of the sequences were affiliated with Firmicutes and Bacteroidetes. The only OTU of Fibrobacteres (OTU 107), had 93 and 90% similarity to Fibrobacter succinogenes and F. intestinalis, respectively, suggesting a new species of Fibrobacter in ostrich ceca. Clostridium coccoides and C. leptum formed major groups within the Firmicutes. There was no OTU with high similarity (98%) to the 16S rDNA of cultivated fibrolytic bacteria in our library. Although two OTUs were affiliated with Euryarchaeota, no sequence was affiliated with methanogenic Archaea. This study presents the very complex ostrich cecal microbial community, in which the majority of the bacterial species have not yet been cultivated. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Cecal microbiota Fermentation Fibrobacter Herbivorous bird Ostrich 16S rDNA

1. Introduction The ostrich (Struthio camelus) is a large flightless, herbivorous bird. Unlike other avian species, the ostrich has a relatively long hindgut [37] with well developed, sacculated ceca, and a capacious, haustrated colon [4]. Microbes in the hindgut are known to actively ferment carbohydrates that have escaped digestion in the upper part of the gastrointestinal (GI) tract. In the hindgut, fermentation produces a high concentration of short chain fatty acids (SCFAs), most notably acetate, propionate and butyrate [33,37], which provide 76% of the metabolizable energy [37]. The hindgut is also known as the site for fiber digestion with a reported digestibility of cellulose and hemicellulose of 38% and 68%, respectively [36]. Soluble sugars resulting from hydrolysis of plant cell wall components are important substrates for fermentation in the hindgut.

* Corresponding author. Tel: þ81 59 231 9593; fax: þ81 59 231 9540. E-mail address: [email protected] (H. Matsui). 1075-9964/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.anaerobe.2009.07.005

High concentration of SCFAs and high digestibility of the cell wall components in the ostrich hindgut are comparable to those found in the ruminant foregut (reticulo-rumen) or equine hindgut, suggesting that the ostrich hindgut functions in a similar manner to non-ruminant herbivores. A highly diverse and abundant anaerobic microbiota are supposedly involved in cell wall digestion and fermentation in the ostrich hindgut. To date, fibrolytic bacteria from ostrich feces have been isolated and identified by 16S ribosomal DNA sequence analysis [43]. These include a strain having 97% similarity to Ruminococcus flavefaciens, a ruminal fibrolytic bacterium, and four cellulolytic strains having 97–99% similarity to Propionibacterium acnes. To our knowledge, no other microbiological study on the ostrich hindgut has been reported. Therefore, investigation of ostrich hindgut microorganisms is needed especially in relation to their physiological contribution to the health and nutrition of ostriches. Recent development of molecular techniques has provided an effective and thorough method to analyze microbial ecosystems without cultivation. The bacterial 16S ribosomal RNA gene

84

H. Matsui et al. / Anaerobe 16 (2010) 83–93

(16S rDNA) sequence has been widely used for this purpose. Analysis of the PCR-derived 16S rDNA clone libraries has shown that microbial communities in the GI tracts of various species of mammals and birds are highly diverse and complex [6,8,16,18,31,34,38,42,49]. A common feature of these studies is that a majority of the sequences in the libraries are not closely related to any known cultivated species, indicating an advantage in analyzing complex microbial communities. The purpose of the present study was to analyze the diversity of the microbial community in cecal digesta of ostrich using the 16S rDNA clone library. 2. Materials and methods 2.1. Morphometric measurement and digesta sampling of the gastrointestinal tract Freshly removed GI tracts, excluding the proventriculus and gizzard, were obtained from 3 ostriches (estimated body weight, w90 kg) slaughtered at a commercial facility (Toyohashi City, Aichi Prefecture, Japan). Birds were fed a commercial diet including maize, wheat, alfalfa meal, corn gluten meal, soybean meal, rice bran, animal fat vitamin/mineral/amino acid premix, calcium carbonate, calcium phosphate, sodium chloride, and betaine. The mean chemical composition of the diet on a dry matter basis was as follows: 15.5% crude protein, 14.0% crude fiber, 3.0% crude fat, 2.4% calcium and 0.5% phosphorus. The anterior end of the small intestine and the posterior end of the rectum were tied with strings. Each cecocolic junction of paired ceca was ligatured to prevent movement of the digesta. The length of the GI tracts was measured before separation of the cecum from the GI tract. The weights of the separated cecum and other parts of the GI tract were measured. The surface of each cecum was washed with sterilized physiological salt solution and cut with sterilized scissors. Cecal digesta was aseptically taken from the opening and divided into two parts. One part of the digesta was placed in a glass tube, and the pH was measured immediately with a pH meter before closing the tube tightly with a butyl rubber for later SCFA analysis. The second digesta sample was suspended in acetone at a final concentration of 70% to prevent DNA digestion [9]. Both digesta samples and the GI tract were cooled by placing on ice and transported to the laboratory, where they were stored at either 25  C (first sample) or 4  C (second sample) until analysis. The digesta remaining in the GI tract was washed out thoroughly with tap water and the weight of the GI tract was measured. 2.2. Short chain fatty acid concentration The cecal digesta samples were allowed to warm to room temperature and the concentration of SCFAs was measured with high-performance liquid chromatography (HPLC) as described by Yimiti et al. [50]. 2.3. DNA extraction and construction of the 16S rDNA clone library Cecal digesta was washed three times with sterile physiological salt solution to remove acetone by centrifugation (8950  g, 10 min, 4  C). DNA was extracted from the washed digesta samples by the bead-beating method with a FastPrep apparatus (Bio 101, Vista, CA, USA) as described by Godon et al. [10]. The crude DNA was purified with Genomic tip 100/G (QIAGEN, Hilden, Germany) and dissolved in sterilized water. The concentration of purified DNA was adjusted to 15 ng ml1. 16S rDNA fragments were amplified with a universal primer set (530F/1392R) that targets the small-subunit rDNA of 3 domains of organisms [15]. PCR was performed with the Ex Taq kit (TaKaRa, Otsu, Japan) with a PE480 Thermal Cycler (Perkin-Elmer,

Norwalk, CT, USA). The PCR reaction mixture (20 ml) contained 1 ml template DNA, 0.5 mM each primer, 1 Ex Taq reaction buffer, 200 mM each deoxynucleotide triphosphate (dNTP mixture), 0.5 unit Ex Taq DNA polymerase, and 0.4 mg ml1 bovine serum albumin. Thermal cycling conditions were as follows: an initial denaturation at 94  C for 2 min, followed by 15 cycles of 94  C for 30 s, 58  C for 30 s, and 72  C for 4 min with a final extension step at 72  C for 10 min. Following separation using 1.0% agarose gel electrophoresis (in TAE buffer), PCR products were confirmed by visualization with ethidium bromide staining and then cloned into TOP10 competent Escherichia coli with the TA cloning kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. Positive clones were randomly selected from the clone library. 2.4. DNA sequencing and analysis DNA sequencing of the cloned fragments was performed by BigDye Terminator kit ver. 3 (Applied Biosystems, Foster City, CA, USA) or DYEnamic ET Terminator Cycle Sequence kit (Amersham Bioscience, Piscataway, NJ, USA) with M13 P7 or M13 P8 primer according to the manufacturer’s protocol. Sequences were analyzed with ABI Prism 3700 or 3100 sequencer (Applied Biosystems). DNA sequences were searched for homology using BLAST [1]. Sequences forming chimera were determined by the CHECK_CHIMERA program of the Ribosomal Database Project (RDPII) [5] and omitted from analysis. Operational taxonomy units (OTUs), coverage, and the Shannon–Wiener index (H0 ) were calculated with the DOTUR program [30] based on a 98% sequence similarity criterion. DNA sequences were aligned with Clustal X ver. 2.0 [17] and phylogenetic trees were constructed using the Neighbor-Joining method [28]. The stability of branches was analyzed by bootstrapping (1000 resamplings). 2.5. Sequence accession numbers All nucleic acid sequences obtained in this study were deposited in the DDBJ, EMBL, and GenBank databases under the following accession numbers: AB385881–AB386190 . 3. Results 3.1. Relative length of ceca, wet weight and fermentation parameters of digesta in ostrich ceca The average length of the GI tracts without the proventriculus and gizzard was 1753  163 cm (mean  standard deviation, STD). The average length of two ceca was 152  10 cm, representing 8.7% of the total length of the GI tract. The total digesta weight in the whole GI tract was 2327  714 g. The wet digesta weight in the cecum was 310  199 g, representing 12.3% of total digesta weight. The pH of cecal digesta was 6.73  0.11 and the concentrations of SCFAs are shown in Table 1. A significant concentration of SCFAs was detected in ostrich cecal digesta. Acetate, propionate and n-butyrate were major components with molar ratios (%) relative to the total SCFAs of 69.8, 20.4 and 6.6, respectively. Trace amounts of formate, iso-butyrate and iso-valerate, and no n-valerate were detected. 3.2. 16S rDNA sequence diversity of ostrich cecal microbiota A total of 322 randomly selected clones isolated from the rDNA clone library of digesta were subjected to DNA sequence analysis; 12 chimeric sequences were omitted from further analysis. Based on the 98% sequence similarity criterion [10], 110 OTUs were found in the library (Table 2). Sequences ranged from 86 to 99% similarity

H. Matsui et al. / Anaerobe 16 (2010) 83–93 Table 1 Short chain fatty acid (SCFA) concentration (mM) in digesta from ostrich ceca. SCFA

Concentration (mM)

Formate Acetate Propionate iso-Butyrate n-Butyrate iso-Valerate n-Valerate

0.5 75.2 22.0 1.6 7.1 1.4 0.0

Total

      

0.3 5.5 3.2 1.5 1.4 0.2 0.0

107.8  9.1

to their nearest sequences in the databases (data not shown); 95 OTUs had less than 98% similarity, and 5 were less than 90%. Coverage and H0 value of the library was 83.9% and 4.29, respectively. 3.3. Phylogenetic analysis Table 2 shows the distribution of sequences and OTUs isolated from the ostrich ceca within prokaryotic phyla. Representative sequences of each OTU and reference sequences obtained from the databases were used to construct phylogenetic trees, as shown in Figs. 1–7. The analyzed sequences were assigned to the following 6 phyla; Bacteroidetes, Firmicutes, Fibrobacteres, Spirochaetes, Verrucomicrobia and Euryarchaeota. No sequence isolated from the ostrich ceca was assigned to Proteobacteria, Actinobacteria or Fusobacteria. 3.4. Detection of a novel uncultured species of genus Fibrobacter in the ostrich ceca OTU 107 was the only OTU that was affiliated with Fibrobacteres and consisted of 20 sequences, representing 6.5% of all sequences in the library (Table 2). This OTU, as well as the 16S rDNA sequences from groups 1 to 4 of F. succinogenes [22], was phylogenetically analyzed (Fig. 1). None of the sequences isolated from ostrich ceca were affiliated with F. succinogenes or F. intestinalis. OTU 107 was distantly related to the Fibrobacter succinogenes groups or F. intestinalis. Although OTU 107 clustered with an uncultured bacterial clone Fibro-B (clone 19-3) from pony cecum [23], these two sequences were obviously different from each other. Similarity (%) of the sequences affiliated with Fibrobacteres, including OTU 107,

Table 2 Number of operational taxonomy units (OTUs) and clones in the 16S rDNA library from ostrich ceca. The number in parenthesis shows the percentage relative to the total number of clones or OTUs.

Bacteroidetes Firmicutes

Fibrobacteres Spirochaetes Verrucomicrobia Euryarchaeota Total

Group

Clostridium coccoides group Clostridium leptum group Clostridiaceae Other cluster Mollicutes Sub total of Firmicutes

Fibro-B and the F. succinogenes groups 1–4, are shown in Table 3. The sequence of OTU 107 was more similar to Fibro-B (94.7%) than to any group of F. succinogenes (92.5–93.1%) or F. intestinalis (90.0%). 3.5. Firmicutes

Values are mean  standard deviation (n ¼ 3).

Phylum

85

Firmicutes dominated the ostrich cecal sequences, comprising 51% of the total number of analyzed sequences (Table 2, Figs. 2–4). Sequences within Firmicutes were further classified into the Clostridium coccoides group (Clostridium subcluster XIVa) (Fig. 2), the Clostridium leptum group (Clostridium subcluster IV) (Fig. 3), other Clostridium groups and Mollicutes (Fig. 4). None of the sequences were affiliated with the order Lactobacillales. The C. coccoides group was the most abundant group within Firmicutes, comprising 71 sequences that represent 23% of the total (Fig. 2 and Table 2). OTUs 035, 051, 054 and 059 showed more than 98% sequence similarity to following uncultured bacterial sequences from pig, lcy22, p-2592-9F5, p-2151-s959-3, and p-197-o5, respectively. OTUs 033, 034 and 035 were clustered with an uncultured bacterial clone lcy22. These OTUs had 92–93% similarity to the pectin degrading bacterium, Lachnospira pectinoschiza. OTUs 036, 037 and 038 were clustered with L. pectinoschiza having 95–96% similarity. OTUs 045, 046 and 047 were clustered with the homoacetogenic bacteria Ruminococcus schinkii, Ruminococcus productus and Ruminococcus hydrogenotrophicus. Although this group contains many of the fibrolytic Clostridium spp., Butyrivibrio spp., Ruminococcus spp., and Eubacterium spp., OTUs with high similarity (98%) to the 16S rDNA of cultivated fibrolytic bacteria were not found in this group. The second most abundant group within the Firmicutes was the C. leptum group (Table 2 and Fig. 3), comprising 12.9% of the total. OTUs 071, 077, 078 and 081 showed more than 98% similarity to an uncultured bacterial clone from the pig [18]. OTUs 072 and 074 showed more than 98% similarity to an uncultured bacterial clone from human feces [19,21]. OTU 078 also showed 98% similarity to Ruminococcus bromii. OTU 082 was clustered with R. flavefaciens strains having 94% similarity. Although the C. leptum group also contains fibrolytic bacteria such as R. flavefaciens and R. albus, none of the OTU showed high similarity to the 16S rDNA of these bacteria. A total of 33 sequences (10.6%) were classified into other Clostridium groups and the Mollicutes (Table 2, Fig. 4). OTU 068 showed 98% similarity to C. metallolevans and C. bartlettii. OTU 087 and 088 were affiliated with the ruminal Clostridia super cluster [35]. OTU 091 formed a deeply branched cluster with a rumen bacterium R-7 and an uncultured bacterial clone RL184_aao67h07 from human feces. Four OTUs comprising 4.5% of the total number of sequences were affiliated with the Mollicutes. OTU 101 consisted of 10 sequences and showed 93% similarity to Anaeroplasma varium.

Number of clones

Number of OTUs

3.6. Bacteroidetes

122 (39.4) 71 (22.9)

28 (25.5) 37 (33.6)

40 (12.9)

15 (13.6)

33 (10.6)

19 (17.3)

14 (4.5) 158 (50.9)

4 (3.6) 75 (68.2)

Bacteroidetes was the second most represented phylum (Table 2, Fig. 5), consisting of 39% of the clone sequences. OTUs 003 and 004 were clustered with Bacteroides. OTUs 008 and 009 were clustered with the Prevotella species, P. albensis, P. brevis, P. ruminicola and P. bryantii isolated from the rumen. OTUs 005, 006, 007, 019 and 020, as well as uncultured bacterial clones from other studies, formed a deeply branched cluster in the phylogenetic tree showing lower similarity to any known bacterial species. The nearest relative of OTU 015 was an uncultured rumen bacterial clone T28H60F14 [41] with 99% similarity. OTU 016 showed 98% similarity to uncultured bacterial clone p-2158-s959-3 from the pig [18]. OTUs 016 to 018 were related to no cultivated bacterium. OTUs 021 to 025 formed a novel and distinct cluster comprising 4.5% of

20 3 1 6 310

(6.5) (1.0) (0.3) (1.9)

1 3 1 2

(0.9) (2.7) (0.9) (1.8)

110

86

H. Matsui et al. / Anaerobe 16 (2010) 83–93

Fibrobacter succinogenes S85 (AJ496032) 718

Fibrobacter succinogenes B1 (M62684) 982

Fibrobacter succinogenes A3C (M62683)

1000

Group 1

UB12-5 (AF018454) Fibrobacter succinogenes BL2 (AJ505937)

1000

Fibrobacter succinogenes REH9-1 (M62682) Fibrobacter succinogenes GC5 (M62688)

Group 2

FIBRO-A (clone 5-19) from pony (L35548) 996

Fibrobacter succinogenes MB4 (M62692) 1000

Fibrobacter succinogenes MM4 (M62694)

Group 3

Fibrobacter succinogenes HM2 (AJ496186) Fibrobacter succinogenes MC1 (M62693)

Group 4

Fibrobacter intestinalis JG1 (M62690) OTU107 (AB385905) [OSTca026, 20] 956

Fibro-B (clone 19-3) from pony (L35547) Aquifex pyrophilus (M83548)

0.02 Fig. 1. Phylogenetic tree of sequences affiliated with Fibrobacteres from ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

the total number of sequences and were weakly related to Barnesiella viscericola from the chicken cecum [29]. These OTUs showed low similarity (<92%) to any sequence in the public databases. OTU 005 was the most numerous among all OTUs, with 22 sequences (7.1%), and weakly related to an uncultured bacterial clone YT58 from fresh equine feces [49]. 3.7. Euryarchaeota OTUs 109 and 110 were clustered within the phylum Euryarchaeota (Fig. 6), comprising 1.9% of the total number of the sequences (Table 2). Both OTUs were clustered with an uncultured soil clone Thermoplasma sp. XT107 and an uncultured rumen archaeon M2 [40], but not with methanogenic Archaea. OTU 109 showed 97% similarity to an uncultured archaeon CSIROQld09 from sheep rumen [47] and OTU 110 showed 97% similarity to an uncultured euryarchaeote PE-CAN.12 from cattle rumen [48], while both OTUs showed 95 and 97% similarities to an uncultured rumen archaeon M2 from sheep, respectively. 3.8. Spirochaetes and Verrucomicrobia The phylogenetic distribution of sequences within the phyla Spirochaetes and Verrucomicrobia is shown in Fig. 7. Three OTUs comprising 1.0% of the total sequences (Table 2) were affiliated with Spirochaetes. OTUs 104 and 105 showed 96 and 97% similarities, respectively, to an equine manure clone wet198. These OTUs were distantly related to a cultivated Treponema. OTU 106 showed high similarity to an uncultured rumen clone T33H60F71 (99%) [41] and 97% similarity to Treponema bryantii. Only one OTU (OTU 108, 1 sequence, Table 2) was affiliated with Verrucomicrobia and with uncultured Verrucomicrobia clones from the rumen [27,38], fresh equine feces [49] and wild gorilla feces [8]. OTU 108 showed low similarity to 16S rDNA sequences from known bacterial species. 4. Discussion The total length of the GI tracts of ostriches obtained in this study was comparable to adult wild ostrich [33] and the relative lengths of paired ceca and weights of cecal digesta were similar to

those reported by Swart et al. [37]. Moreover, the observed cecal pH was similar to the pH 6.9–7 reported by Skadhouge et al. [33] and Swart et al. [37]. Although a significant amount of SCFAs (108 mM) was detected in cecal digesta, the amount was lower than those reported by Skadhouge et al. [33] (170 mM) or Swart et al. [37] (140 mM), suggesting a difference in feed composition or sampling time after feeding. Molar ratios of acetate, propionate and butyrate in ceca were similar to that found in the rumen [26] or horse colon [12]. However, Swart et al. [37] detected more acetate (90% of the total concentration of acetate, propionate and butyrate) in ostrich ceca, possibly due to differences in diet components such as fiber. This study aimed to provide information on the microbial diversity within the cecum of ostriches and is the first attempt to analyze the ostrich cecal microbiota of by culture-independent molecular methodology. We obtained a high H0 value of the microbial community in ostrich ceca (4.29). H0 values of microbiota in the GI tracts of several animals have been previously published, including H0 values of 4.87 calculated from the sequence data from the rumen deposited in the public databases [7] and 4.25 from the rumen of reindeer [35]. Although the H0 value obtained in the present study was similar to these previously reported values, not all H0 values may be directly comparable due to the differences in DNA extraction methods, primers used for construction of the library and the number of sequences analyzed; that is, higher H0 values have been calculated for cecal microbiota of turkeys (6.7) [31] and pig intestine (6.88) [18]. Even though a large number of cloned sequences isolated from the GI tracts of various kinds of mammals and avian species have been deposited in the public databases, over 85% of OTUs had less than 98% similarity to all deposited sequences. This result suggests that the majority of the sequences were from uncultured bacteria specific to the ostrich GI tract. We detected novel uncultured species of Fibrobacter in the library. To the best of our knowledge, this is the first report of Fibrobacter detected in the avian GI tract. From phylogenetic analysis and sequence similarity, the Fibrobacter sequences are most likely novel Fibrobacter species. Bacteria that belong to the genus Fibrobacter have been previously isolated from the rumen [45], the large intestine of pig [44] and rat [25], and detected in the large intestine of the horse [14,23,49] by culture-independent methods. Fibrobacter has also been detected in the feces of wild gorilla by using F. succinogenes

Fig. 2. Phylogenetic tree of the Clostridium coccoides group of Firmicutes found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

Fig. 3. Phylogenetic tree of the Clostridium leptum group of Firmicutes found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

Fig. 4. Phylogenetic tree of other Firmicutes found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

Fig. 5. Phylogenetic tree of Bacteroidetes found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

H. Matsui et al. / Anaerobe 16 (2010) 83–93

91

Fig. 6. Phylogenetic tree of Euryarchaeota found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.02 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

specific primer [8]. More recently, McDonald et al. [24] have isolated a 16S rDNA fragment of Fibrobacter from a landfill site in the United Kingdom, indicating the presence of F. succinogenes and novel clusters closely related to Fibrobacter outside the gut ecosystem. Furthermore, bacteria of Fibrobacteres subphylum 2 have been specifically detected in the gut of termites by fluorescence in situ hybridization [11]. These and our findings indicate that bacteria of Fibrobacter or related genera propagate in a wide range of anaerobic environments. Although the predominance of Fibrobacter in the rumen has been established, their nucleic acid sequences are often poorly represented in 16S rDNA clone libraries constructed using a universal primer set [38], possibly due to poor amplification efficiencies of Fibrobacter 16S rDNA [39]. Nevertheless, Fibrobacter

sequences comprised a relatively high percentage of the total number of sequences in the ostrich cecal library, suggesting that the population density of the novel Fibrobacter species is relatively high and play an important role in ostrich ceca. F. succinogenes is known to play an important role in fiber digestion in the rumen [15,32] but since it is unclear whether the Fibrobacter species in ostrich ceca has similar fibrolytic ability. Isolation, characterization and enumeration of the novel Fibrobacter species are needed to clarify their function in ostrich ceca. One distinct property of the ostrich hindgut is the strong ability to digest fiber [36], supporting the view that the 16S rDNA sequences detected in the ceca originated from known fiber digesting bacteria. However, no sequence with 98% similarity to any known fibrolytic

Fig. 7. Phylogenetic tree of Spirochaetes and Verrucomicrobia found in ostrich ceca. Accession number of each sequence is shown in parentheses. Representative clone name starting with ‘‘OSTca’’ and number are shown in brackets. The scale bar represents 0.05 substitutions per nucleotide position. Bootstrap values for 1000 trees are shown at nodes; only values of 700 are shown.

92

H. Matsui et al. / Anaerobe 16 (2010) 83–93

Table 3 Similarity (%) among Fibrobacteres OTUs from ostrich ceca, an uncultured Fibrobacter clone from the pony, and the F. succinogenes groups 1–4. OTU107 Fibro-B Group 1 Group 2 Group 3 Group 4 F. intestinalis OTU107 100.0 94.7 Fibro-Ba 92.5 Group 1b 93.1 Group 2b Group 3b 93.0 92.9 Group 4b c 90.0 F. intestinalis

100.0 92.9 92.6 93.4 93.0 90.4

100.0 97.1 97.6 97.9 93.0

100.0 97.2 97.4 92.0

100.0 98.1 92.7

100.0 93.7

100.0

a

Fibrobacter sp. Fibro-B (clone 19-3; accession no., L35547) from pony cecum [23]. b Strains of F. succinogenes [22] and accession numbers of their respective rRNA genes are as follows: Group 1, S85 (AJ496032); Group 2, REH9-1 (M62682); Group 3, HM2 (AJ496186); and Group 4, MC1 (M62693). c Accession no.: M62690.

bacteria was detected in our library. Although 3 of the OTUs that belong to the C. coccoides group showed similarity to pectinolytic bacteria, whether the sequences were isolated from bacteria capable of degrading pectin is inconclusive. The previously published phylogenetic composition of gut microbiota in various animals is summarized in Table 4. The majority of the clones from the large intestine of the horse are affiliated with Bacteroidetes and Firmicutes [6], which correlates well with previous reports on ruminants and humans and suggests that bacteria in these phyla play important roles in fermentation in the large intestine. In the present study, the majority of the sequences isolated from the ostrich cecum are also affiliated with these two phyla: Firmicutes (51%) and Bacteroidetes (39%). Similar trends were observed in microbiota in the bovine rumen [38,46], yak rumen [2], pig feces [18], Japanese native horse feces [49] and human feces [34]. All these results emphasize the importance of these phyla in microbial fermentation in the gut. Clones that are affiliated with Firmicutes comprised 72% of the total isolated clone population in the equine large intestine [6]. The abundance of Firmicutes suggests that the phylum is the most important functional group within intestinal ecosystems, as this phylum contains many fibrolytic bacteria [6]. In chicken ceca, the vast majority of the sequences are affiliated with Firmicutes (71.7–93.9%) and only a small number of the sequences are affiliated with Bacteroidetes (1.9–4.7%) [3,16,51]. Therefore, the abundance of Firmicutes species

Table 4 Distribution of clones among Bacteroidetes, Firmicutes and other phyla in the gastrointestinal tracts of mammals and birds. Source

Bacteroidetes

Firmicutes

Others

Reference

Turkey ceca (average of wild and domestic) Equine large intestine (Hokkaido native horse, Japan) Pig feces Jinnan cattle rumen Ostrich ceca Cow rumen Human feces Yak rumen Cow rumen Equine large intestine (United Kingdom) Reindeer rumen Mouse feces Dugong feces Chicken ceca Chicken ceca Chicken ceca Wild gorilla feces

54.0

30.0

16.0

[31]

47.4

36.8

15.8

[49]

47.2 39.6 39.4 32.0 31.0 30.9 27.0 20.7

45.1 22.3 50.9 62.0 64.1 54.1 67.5 71.5

7.7 38.1 9.7 6.0 4.9 14.9 5.4 7.7

[18] [2] This study [38] [34] [2] [46] [6]

22.9 18.7 15.0 4.7 4.4 1.9 1.1

74.6 75.9 83.1 93.9 92.0 71.7 71.0

2.5 5.5 1.9 1.8 3.5 26.4 27.9

[35] [13] [42] [16] [3] [51] [8]

does not necessarily correlate with fibrolytic ecosystems. The results shown in Table 4 suggest that the abundance of Bacteroidetes species is an important factor for the fibrolytic and/or actively fermenting microbial ecosystem in the GI tracts. In contrast, the number of clones affiliated with Bacteroidetes is high in the ceca of wild and domestic turkeys, which have a non-fibrolytic ecosystem [31], while the number of Bacteroidetes clones is extremely low (1.1%) and Verrucomicrobia relatively high in (17.2%) the feces of wild gorilla, which have a fibrolytic microbial ecosystem [8]. Recently, the diversity of gut microbes has been studied in 60 mammalian species, including humans, to understand the co-evolution of mammals and their indigenous microbial communities [20]. The comparison of gut bacterial phylogeny in herbivores showed that herbivores are generally clustered into two groups, foregut and hindgut fermenters. This suggests that as mammals undergo convergent evolution in the morphological adaptation of their guts to herbivory, their microbiota arrive at similar compositional configurations in unrelated hosts with similar gut structures [19]. However, whether this hypothesis is applicable to microbiota in the avian GI tracts is still unclear. In the present study, 2 OTUs that are related to Thermoplasma and an uncultured rumen archaeon M2 [40] were found in the Euryarchaeota cluster. Related sequences were also retrieved from sheep (CSIROQld09) [47] and cattle (PE-CAN.120) [48]. These three rumen clones are considered to comprise a novel uncultivated archaeal group in the rumen [40,47,48]. As no cultivated isolate is available yet, metabolic function of these clones from our ostriches and from the rumen remains unclear. Although Proteobacteria, Actinobacteria, Fusobacteria and methanogenic Archaea were not detected in this study, the absence of these types of bacteria in ostrich ceca is inconclusive. However, PCR biases, such as primer pairs used for analysis or amplification efficiency of each sequence, may affect the composition of the sequences analyzed in PCR-generated clone libraries. The ostrich cecal microbial community is very complex and the majority of the bacterial species have not been cultivated. A novel Fibrobacter species unique to ostrich ceca has been detected. Despite the limited number of sequences analyzed, our results provide valuable insight into a poorly understood microbial ecosystem and form the basis for further studies into microbial functions affecting ostrich nutrition and health. Efforts to isolate and identify fibrolytic or other functionally important bacteria are necessary. Acknowledgements The authors thank Professor Kazunari Ushida and his colleagues (Kyoto Prefectural University, Japan) for their technical instruction of DNA extraction and Professor Masakazu Goto (Mie University, Japan) for his instruction of HPLC analysis, as well as Professor Yasuo Kobayashi (Hokkaido University, Japan) for his revisions to the manuscript and valuable discussion. The authors also thank Dr. Graeme Attwood (AgResearch, New Zealand) for his correction of English grammar of the manuscript. The present study was financially supported in part by Grants-in-Aid for Scientific Research, Japan Society for the Promotion of Science (17380157). DNA sequencing was carried out in part at Life Science Research Center of the Center for Molecular Biology and Genetics in Mie University. References [1] Altschul SF, Madden TL, Scha¨ffer AA, Zhang J, Zhang Z, Miller W, et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997;25:3389–402. [2] An D, Dong X, Dong Z. Prokaryote diversity in the rumen of yak (Bos grunniens) and Jinnan cattle (Bos taurus) estimated by 16S rDNA homology analyses. Anaerobe 2005;11:207–15.

H. Matsui et al. / Anaerobe 16 (2010) 83–93 [3] Bjerrum L, Engberg RM, Leser TD, Jensen B, Finster K, Pedersen K. Microbial community composition of the ileum and cecum of broiler chicken as revealed by molecular and culture-based techniques. Poult Sci 2006;85:1151–64. [4] Cho P, Brown R, Anderson M. Comparative gross anatomy of ratites. Zoo Biol 1984;3:133–44. [5] Cole JR, Chai B, Marsh TL, Farris RJ, Wang Q, Kulam SA, et al. The ribosomal database project (RDP-II): previewing a new autoaligner that allows regular updates and the new prokaryotic taxonomy. Nucleic Acids Res 2003;31:442–3. [6] Daly K, Stewart CS, Flint HJ, Shirazi-Beechey SP. Bacterial diversity within the equine large intestine as revealed by molecular analysis of cloned 16S rRNA genes. FEMS Microbiol Ecol 2001;38:141–51. [7] Edwards JE, McEwan NR, Travis AJ, Wallace RJ. 16S rDNA library-based analysis of ruminal bacterial diversity. Antronie van Leeuwenhoek 2004;86:263–81. [8] Frey JC, Rothman JM, Pell AN, Nizeyi JB, Cranfield MR, Angert ER. Fecal bacterial diversity in a wild gorilla. Appl Environ Microbiol 2006;72:3788–92. [9] Fukatsu T. Acetone preservation: a practical technique for molecular analysis. Mol Ecol 1999;8:1935–45. [10] Godon J, Zumstein E, Dabert D, Habouzit F, Moletta R. Molecular microbial diversity of an anaerobic digester as determined by small-subunit rDNA sequence analysis. Appl Environ Microbiol 1997;63:2802–13. [11] Hongo Y, Deevong P, Hattori S, Inoue T, Noda S, Noparatnaraporn N, et al. Phylogenetic diversity, localization, and cell morphologies of members of the candidate phylum TG3 and a subphylum in the phylum Fibrobacteres, recently discovered bacterial groups dominant in termite guts. Appl Environ Microbiol 2006;72:6780–8. [12] Hussein HS, Vogedes LA, Fernandez GC, Frankeny RL. Effects of cereal grain supplementation on apparent digestibility of nutrients and concentrations of fermentation end-products in the feces and serum of horses consuming alfalfa cubes. J Anim Sci 2004;82:1986–96. [13] Kibe R, Sakamoto M, Hayashi H, Yokota H, Benno Y. Maturation of the murine cecal microbiota as revealed by terminal restriction fragment length polymorphism and 16S rRNA gene clone libraries. FEMS Microbiol Lett 2004;235:139–46. [14] Koike S, Shingu Y, Inaba H, Kawai M, Kobayashi Y, Hata H, et al. Fecal bacteria in Hokkaido native horses as characterized by microscopic enumeration and competitive polymerase chain reaction assays. J Equine Sci 2000;11:45–50. [15] Koike S, Yoshitani S, Kobayashi Y, Tanaka K. Phylogenetic analysis of fiberassociated rumen bacterial community and PCR detection of uncultured bacteria. FEMS Microbiol Lett 2003;229:23–30. [16] Lan PTN, Hayashi H, Sakamoto M, Benno Y. Phylogeneic analysis of cecal microbiota in chicken by the use of 16S rDNA clone libraries. Microbiol Immunol 2002;46:371–82. [17] Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. Clustal W and Clustal X version 2.0. Bioinformatics 2007;23:2947–8. [18] Leser TD, Amenuvor JZ, Jensen TK, Lindecrona RH, Boye M, Møller K. Cultureindependent analysis of gut bacteria: the pig gastrointestinal tract microbiota revisited. Appl Environ Microbiol 2002;68:673–90. [19] Ley RE, Hamady M, Lozupone C, Turnbaugh PJ, Ramey RR, Bircher JS, et al. Evolution of mammals and their gut microbes. Science 2008;320:1647–51. [20] Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Microbial ecology: human gut microbes associated with obesity. Nature 2006;444:1022–3. [21] Li M, Wang B, Zhang M, Rantalainen M, Wang S, Zhou H, et al. Symbiotic gut microbes modulate human metabolic phenotypes. Proc Natl Acad Sci USA 2008;105:2117–22. [22] Lin C, Flesher B, Capman WC, Amann RI, Stahl DA. Taxon specific hybridization probes for fiber-digesting bacteria suggests novel gut-associated Fibrobacter. Syst Appl Microbiol 1994;17:418–24. [23] Lin C, Stahl DA. Taxon-specific probes for the cellulolytic genus Fibrobacter reveal abundant and novel equine-associated populations. Appl Environ Microbiol 1995;61:1348–51. [24] McDonald JE, Lockhart RJ, Cox MJ, Allison HE, McCarthy AJ. Detection of novel Fibrobacter populations in landfill sites and determination of their relative abundance via quantitative PCR. Environ Microbiol 2008;10:1310–9. [25] Montgomery L, Macy JM. Characterization of rat cecum cellulolytic bacteria. Appl Environ Microbiol 1982;44:1435–43. [26] Nagaraja TG, Newbold CJ, van Nevel CJ, Demeyer DI. Manipulation of ruminal fermentation. In: Hobson PN, Stewart CS, editors. The rumen microbial ecosystem. 2nd ed. London, Great Britain: Blackie Academic and Professional; 1997. p. 523–632. [27] Ozutsumi Y, Tajima K, Takenaka A, Itabashi H. The effect of protozoa on the composition of rumen bacteria in cattle using 16S rRNA gene clone libraries. Biosci Biotechnol Biochem 2005;69:499–506.

93

[28] Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987;4:406–25. [29] Sakamoto M, Lan PT, Benno Y. Barnesiella viscericola gen. nov., sp. nov., a novel member of the family Porphyromonadaceae isolated from chicken caecum. Int J Syst Evol Microbiol 2007;57:342–6. [30] Schloss PD, Handelsman J. Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness. Appl Environ Microbiol 2005;71:1501–6. [31] Scupham AJ, Patton TG, Bent E, Bayles DO. Comparison of the cecal microbiota of domestic and wild turkeys. Microb Ecol 2008;56:322–31. [32] Shinkai T, Kobayashi Y. Localization of ruminal cellulolytic bacteria on plant fibrous materials as determined by fluorescence in situ hybridization and realtime PCR. Appl Environ Microbiol 2007;73:1646–52. [33] Skadhauge E, Waru¨i CN, Kamau JMZ, Maloiy GMO. Function of the lower intestine and osmoregulation in the ostrich: preliminary anatomical and physiological observations. Q J Exp Physiol 1984;69:809–18. [34] Suau A, Bonnet R, Sutren M, Godon JJ, Gibson GR, Collins MD, et al. Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species within the human gut. Appl Environ Microbiol 1999;65:4799–807. [35] Sundset MA, Prasteng KE, Cann IKO, Mathiesen SD, Mackie RI. Novel rumen bacterial diversity in two geographically separated sub-species of reindeer. Microb Ecol 2007;54:424–38. [36] Swart D, Mackie RI, Hayes JP. Influence of live mass, rate of passage and fibre digestion in the ostrich (Struthio camelus var. domesticus). S Afr J Anim Sci 1993a;23:119–26. [37] Swart D, Mackie RI, Hayes JP. Fermentative digestion in the ostrich (Struthio camelus var. domesticus), a large avian species that utilizes cellulose. S Afr J Anim Sci 1993b;23:127–35. [38] Tajima K, Aminov RI, Nagamine T, Ogata K, Nakamura M, Matsui H, et al. Rumen bacterial diversity as determined by sequence analysis of 16S rDNA libraries. FEMS Microbiol Ecol 1999;29:159–69. [39] Tajima K, Aminov RI, Nagamine T, Matsui H, Nakamura M, Benno Y. Dietdependent shifts in the bacterial population of the rumen revealed with realtime PCR. Appl Environ Microbiol 2001a;67:2766–74. [40] Tajima K, Nagamine T, Matsui H, Nakamura M, Aminov RI. Phylogenetic analysis of archaeal 16S rRNA libraries from the rumen suggests that existence of a novel group of archaea not associated with known methanogens. FEMS Microbiol Lett 2001b;200:67–72. [41] Tajima K, Nonaka I, Higuchi K, Takusari N, Kurihara M, Takenaka A, et al. Influence of high temperature and humidity on rumen bacterial diversity in Holstein heifers. Anaerobe 2007;13:57–64. [42] Tsukinowa E, Karita S, Asano S, Wakai Y, Oka Y, Furuta M, et al. Fecal microbiota of a dugong (Dugong dugong) in captivity at Toba Aquarium. J Gen Appl Microbiol 2008;54:25–38. [43] van Gylswyk KO, Russouw T, James MD, Thomson JA. Bacterial cellulose degradation in the ostrich digestive tract. In: Ohmiya K, Hayashi K, Sakka K, Kobayashi Y, Karita S, Kimura T, editors. Genetics, biochemistry and ecology of cellulose degradation. Tokyo, Japan: Uni Publishers, Co. Ltd.; 1998. p. 602–10. [44] Varel VH, Fryda SJ, Robinson IM. Cellulolytic bacteria from pig large intestine. Appl Environ Microbiol 1984;47:219–21. [45] Varel VH, Dehority BA. Ruminal cellulolytic bacteria and protozoa from bison, cattle–bison hybrids, and cattle fed three alfalfa-corn diets. Appl Environ Microbiol 1989;55:148–53. [46] Whitford MF, Foster RJ, Beard CE, Gong J, Teather RM. Phylogenetic analysis of rumen bacteria by comparative sequence analysis of cloned 16S rRNA genes. Anaerobe 1998;4:153–63. [47] Wright AD, Toovey AF, Pimm CL. Molecular identification of methanogenic archaea from sheep in Queensland, Australia reveal more uncultured novel archaea. Anaerobe 2006;12:134–9. [48] Wright AD, Auckland CH, Lynn DH. Molecular diversity of methanogens in feedlot cattle from Ontario and Prince Edward Island, Canada. Appl Environ Microbiol 2007;73:4206–10. [49] Yamano H, Koike S, Kobayashi Y, Hata H. Phylogenetic analysis of hindgut microbiota in Hokkaido native horses compared to light horses. Anim Sci J 2008;79:234–42. [50] Yimiti W, Hiraoka H, Yahaya MS, Takeda M, Kim W, Takahashi T, et al. Effects of phenylalanine fermentation byproduct and sugarcane molasses on fermentation quality and rumen degradation of whole crop barley (Hordeum vulgare L.) silage in situ. Grassland Sci 2006;52:73–9. [51] Zhu XY, Zhong T, Pandya Y, Joerger RD. 16S rRNA-based analysis of microbiota from the cecum of broiler chickens. Appl Environ Microbiol 2002;68:124–37.