- Email: [email protected]
Diversity of symbiotic archaeal communities in marine sponges from Korea
Biomolecular Engineering 20 (2003) 299 /304 www.elsevier.com/locate/geneanabioeng
Diversity of symbiotic archaeal communities in marine sponges from Korea Eun-Young Lee a,1, Hong Kum Lee a, Yoo Kyung Lee a, Chung Ja Sim b, JungHyun Lee a,* a
Microbiology Laboratory, Korea Ocean Research and Development Institute, Ansan P.O. Box 29, Seoul 425-600, South Korea b Department of Biology, Hannam University, 133 Ojungdong, Taejeon 306-791, South Korea
Abstract A molecular analysis of archaeal communities in eight sponges collected along the coast of Cheju Island, Korea was conducted using terminal-restriction fragment length polymorphism (T-RFLP) in conjunction with sequencing analysis of 16S rDNA clones. The terminal-restriction fragment (T-RF) profiles showed that each sponge had a simple archaeal community represented by a single major peak of the same size except for one unidentified sponge (01CJ20). In order to identify the components of the community, 170 archaeal 16S rDNA clones were recovered from sponges and analyzed by RFLP typing. Sequences of 19 representative clones for all RFLP types found in each sponge were determined and phylogenetic analysis was carried out. Seventeen of these archaeal 16S rDNA clones showed a high similarity to marine group I, belonging to the crenarchaeotes. In the phylogenetic tree, 15 archaeal clones were grouped into five sponge-associated archaeal clusters. In the unidentified sponge sample (01CJ20), one major T-RF peak was represented by a single RFLP type (40 clones), which implied a specific relationship between the sponge and its symbiotic archaeal components. # 2003 Elsevier Science B.V. All rights reserved. Keywords: Marine sponge; Archaeal symbiont; T-RFLP; 16S rDNA analysis
1. Introduction Marine sponges are a haven to a wide variety of marine microorganisms. Some of these sponge-associated microorganisms are cyanobacteria, heterogeneous bacteria, and archaea [1 /5]. Host-specificity has been identified in some microbial symbiotic relationships [6]. Sponges commonly contain bioactive and biochemical metabolites [7]. These natural products present within sponges could be attributed to their symbiotic microbes [8,9]. Recent evidence suggests that a bacterial symbiont of the marine bryozoan Bugula neritina is responsible for the biosynthesis of bryostatins, a family of macrocyclic lactones with anti-cancer activity [10].
* Corresponding author. Tel.: /82-31-400-6243; fax: /82-31-4062495. E-mail address: [email protected] (J.-H. Lee). 1 Present address: Biological Resources Division, Biodiversity Research Department, National Institute of Environmental Research, Environmental Research Complex, Kyungseo-Dong, SeoGu, Incheon 404-170, South Korea.
The introduction of molecular approaches has identified the diversity of such microbes, and these techniques have allowed the recovery of whole communities of environmental microbes, including some that are unculturable [11]. Previous studies have used 16S rDNA sequences to clarify the symbiotic relationships between eubacteria and Halichondria [12,13]. One sponge can harbor diverse symbionts, including heterotrophic bacteria, cyanobacteria, green sulfur bacteria, and archaea [12,14,15]. Recently, Hentschel et al. [16] identified sponge-specific sequence clusters belonging to various eubacterial divisions, and host-specific eubacterial sequence clusters derived from three sponge species, including Aplysina, Theonella, and Rhopaloides. Recent investigations of microbial communities have focused primarily on eubacterial communities derived from environmental surroundings [13,16,17]. Relatively few studies of microbial communities have dealt with archaeal communities associated with sponges. Archaea is one of three major microbial domains and is generally divided into three phylogenetic lineages: Crenarchaeota,
1389-0344/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1389-0344(03)00034-0
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
300
Euryarchaeota, and marine archaea recovered from the deep sea [18 /21]. In contrast to Crenarchaeota, which are extreme thermophiles with limited habitats, Euryarchaeota include thermophiles, sulfur-metabolizing microorganisms, and methanogens. Recently, novel archaea have been found in more diverse aquatic and terrestrial environments, even from other living organisms, extending the biotopes of crenarchaeotes [17,22,23]. In the study of the unculturable archaeal community associated with sponges, Preston et al. [4] identified a symbiotic crenarchaeote, Cenarchaeum symbiosum , in a marine sponge, Axinella mexicana , using 16S rDNA sequences and fluorescence in situ hybridization (FISH). The current study aimed to survey and compare the archaeal community structure associated with various sponges, and to characterize each archaeal component of the communities using molecular analysis based on 16S rDNA sequences.
2.2. Archaeal rDNA amplification and terminalrestriction fragment length polymorphism (T-RFLP) analysis Archaeal 16S rDNA was partially amplified with 21F (Sigma-Genosys, USA), and labeled at the 5? end with hexachlorofluorescein (HEX) or unlabeled 21F and 958R primers designed by DeLong [19]. The PCR reaction was performed in a thermal cycler (Biometra, Germany) with an initial denaturation step of 95 8C for 5 min followed by 30 cycles with the temperature profile, 95 8C for 1 min, 55 8C for 1 min, 72 8C for 1 min, ending with a final extension step of 72 8C for 7 min. Each amplicon was digested by three four-cutter restriction enzymes (Hha I, Hae III, Rsa I; Promega, WI), and analyzed with an automated sequencer (ABI377) equipped with the program GenScan (PE Applied Bioystems). GenScan 2500-Rox was used as an internal size standard and T-RFLP analysis for each sponge was done in triplicate. 2.3. Cloning and sequencing archaeal rDNA
2. Material and methods
2.1. Sponges and DNA extraction Eight sponges with different morphologies were collected at depths of 15/30 m on SCUBA dives on the coasts of Cheju and Mara Island in southern Korea, in June 2001. The sponges were classified as Spirastrella panis (01CJ08), Halichondria sp. (01CJ12), Halichondria okadai (01CJ19), Sarcotragus sp. (01CJ19), Petrosia sp. (01CJ24), and Erylus nobilis (01CJ25). Two sponges were unidentified and given the respective designations, 01CJ20 and 01CJ23 (Table 1). The sponges were transferred to the laboratory on ice and stored at / 80 8C. DNA was extracted from 2 g (wet weight) of each sponge, according to the modified methods of Junghans and Metzlaff [24].
Archaeal rDNA was amplified with the same DNA used in T-RFLP analysis. Amplified 16S rDNAs were cloned into pGEM T-Easy Vector (Promega, WI). A total of 170 16S rDNA clones were recovered from the eight sponges. All the clones were characterized by RFLP to select the different recombinant clones. A total of 19 clones (1 from a single 01CJ08 RFLP type, 2 from two 01CJ12 RFLP types, 5 from five 01CJ18 RFLP types, 3 from 01CJ19 RFLP types, 1 from a 01CJ20 RFLP type, 3 from 01CJ23 RFLP types, 3 from 01CJ24 RFLP types, and 1 from a 01CJ25 RFLP type) were selected from the 170 archaeal clones (Table 2). The sequences of the 19 representative RFLP clones were determined using an ABI377 (Applied Bioystems) automated sequencer. The sequences were analyzed using the CHECK_CHIMERA program at the Ribosomal Database Project II website to detect the chimeric rDNA clones [25]. None of the identified sequences were chimeric. 2.4. Phylogenetic analysis
Table 1 Total DNA yield from extraction Sponges
Classification
mg/g (wet weight)
01CJ08 01CJ12 01CJ18 01CJ19 01CJ20 01CJ23 01CJ24 01CJ25
S. panis Halichondria sp. H. okadai Sarcotragus sp. Unidentified sponge Unidentified sponge Petrosia sp. E. nobilis
5 8 0.8 3 5 15 5 14
Sequences of clones were edited using PHYDIT program ver. 3.2 [26], and a NCBI Blast search was performed to identify the nearest neighbor to the cloned sequence. Sequences were aligned with prokaryotic SSU rDNA data retrieved from the RDP website to create a matrix. Neighbor-joining analysis, based on a Kimura 2parameter matrix, was performed using both the collective sequence matrix and the program package PHYLIP to examine archaeal phylogenetic relationships [27]. To estimate the relative strength of a given clade, a model for equal change at all sites was employed and subjected to 1000 rounds of bootstrap resamplings [28].
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
301
Table 2 Description of representative archaeal clones based on RFLP/Hae III typing Sponge
RFLP type
Clone number
Representative clone
T-RF/Hae III
T-RF/Hha I
T-RF/Rsa I
Accession number
01CJ08 01CJ12
01CJ24
I I II I II III IV VI I II IV V I II IV I
19 15 4 7 8 2 1 1 16 3 1 40 8 3 5 26
01CJ25
II II
4 20
Ar08-1 Ar12-1 Ar12-16 Ar18-1 Ar18-8 Ar18-4 Ar18-3 Ar18-7 Ar19-1 Ar19-24 Ar19-36 Ar20-3 Ar23-13 Ar23-28 Ar23-23 Ar24-16 Ar24-9 Ar24-5 Ar25-51
213 213 213 213 213 241 241 46 213 213 213 153 213 213 213 213 213 213 213
326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326
262 262 262 262 262 262 262 262 262 262 262 295 262 262 262 262 262 262 262
AY192627 AY192628 AY192634 AY192644 AY192638 AY192635 AY192629 AY192637 AY192630 AY192639 AY192640 AY192631 AY192632 AY192642 AY192641 AY192643 AY192636 AY192645 AY192633
01CJ18
01CJ19
01CJ20 01CJ23
3. Results and discussion 3.1. DNA assemblages in the sponges Total genomic DNA extracted from all eight sponges ranged from 0.8 to 15 mg/g (wet weight of sponge) (Table 1). A 900 bp fragment of a partial archaeal 16S rDNA was amplified from each of the eight sponges using Hex21F, unlabeled 21F, and 958R (data not shown). Amplification of archaeal 16S rDNA sequences, using archaeal-specific primers, in all eight-sponge types provided further evidence of the extensive habitats that archaea can occupy, in addition to freshwater, coastal sediment, deep sea sediment and other organisms [29 / 31]. 3.2. T-RFLP analysis Archaeal community structures in the eight sponges were assessed using a T-RFLP analysis. Terminalrestriction fragment (T-RF) profiles were generated from the digestion of fluorescent-labeled PCR fragments using the restriction enzymes Hae III, Hha I, and Rsa I. The resulting profiles showed one major peak in all sponge samples (Fig. 1). In all eight sponges, a 326 bp fragment was identified with the restriction enzyme Hha I. A single major peak was also identified in most sponges using Rsa I and Hae III (262 and 213 bp, respectively). The only exception was observed in one unidentified sponge (01CJ20), which produced fragments of 295 (Rsa I) and 153 bp (Hae III) (Table 2, Fig. 1). In this survey of sponge-derived archaeal communities, the T-RF pattern with a single peak was
Fig. 1. T-RFLP profiles generated from a Hae III restriction digest and identification of T-RFLP peaks from a clone library. (A) S. panis (01CJ08); (B) Halichondria sp. (01CJ12); (C) H. okadai (01CJ18); (D) Sarcotragus sp. (01CJ19); (E) unidentified sponges (01CJ20); (F) unidentified sponges (01CJ23); (G) Petrosia sp. (01CJ24); and (F) E. nobilis (01CJ25).
302
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
Fig. 2. A neighbor-joining tree based on partial archaeal 16S rDNA clones associated with the sponges. Bold letters indicate the determined archaeal sequences in this study. The boxes represent each sponge-associated archaeal cluster. Values near the branches are based on 1000 bootstrap replications.
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
consistent, showing archaeal diversity within and between the sponges to be similar and simplistic. This level of archaeal diversity was much lower than that observed in studies of marine fish [32] and chimney structure [33]. There were no clear differences in the composition of the archaeal community among the sponges studied due to the identification of the single major T-RF peak. Only the unidentified sponge (01CJ20) had a small peak shift, suggesting a distinct archaeal phylotype from those found in the other sponges. 3.3. Identification of archaeal components in the T-RFLP analysis Cloning and sequencing analysis was used to characterize each peak from the T-RF profiles. 16S rDNA clones of sponge-associated archaea were obtained for each sponge sample. A total of 170 clone libraries, including 16/40 clones from each sponge, were analyzed by RFLP typing using Hha I. The 170 clones resulted in six different RFLP patterns (Table 2). Each sponge had at least one RFLP type. The predominant RFLP type, type I (76 clones; 44.7%) was seen in six sponges. In the two unidentified sponges, 01CJ20 and 01CJ25, RFLP types V and II were dominant, respectively. Clones representing the different RFLP types of each sponge were sequenced as described in Table 2. The 16S rDNA sequences of the archaeal clones have been submitted to GenBank under accession numbers AY192627 to 192645 (Table 2). We analyzed T-RF peak data with sequence analysis and showed that the single peak was associated with different sequences of the same size. However, the single peak identified from each of Spirastrella (01CJ08), Erylus (01CJ25), and an unidentified sponge (01CJ20) had only one type of clone, (Ar08-1, Ar25-51, and Ar203, respectively) as a representative. Sequence analysis showed that most archaeal clones (89%; 17 clones) were affiliated with the crenarchaeotes, especially marine group I, except for clones Ar18-7 and Ar18-8, which belonged to the soil crenarchaeal group (Fig. 2). The neighbor-joining analysis grouped the 15 archaeal sequences into clusters of sponge-derived clones in marine group I. Within marine group I, based on the clusters, preliminary sponge archaeal groups were first designated as A /E (Fig. 2). Specifically, archaeal clones (Ar12-1, Ar18-3, Ar18-4, Ar20-3) in sponge groups C and E were strongly associated with previously known archaeotes associated with sponges. The other eleven clones were weakly associated with sponge groups A, B, and D. In sponge group C, clone Ar12-1 from sponge Halichondria was highly (100%) affiliated with the crenarchaeote derived from Rhopaloeides odorabile [34], showing non-host specificity. Sponge group E showed that C. symbiosum from the sponge Axinella
303
[4], which has been described as unique, grouped consistently (100%) with Ar20-3 and clustered significantly (91%) with Ar18-3 and Ar18-4 of Halichondria . Considering the monophyletic sponge group E clade, which was composed of Ar20-3, Ar18-3, and Ar18-4 archaeal sequences, the identification of three archaea similar to C. symbiosum on the Korean coast verifies the cosmopolitan distribution of this Cenarchaeum cluster, which includes the coast of California and Korea [4]. Margot et al. [35] collected filamentous archaea from Axinella sponges, which are phylogenetically closely related to C. symbiosum , from the Spanish Mediterranean coast. Sponge archaeal groups C and E appeared to be clusters with sponge-specific sequences, while sponge groups A, B, and D were associated with archaeal clones derived from five different sponges (Erylus , Halichondria , Sarcotragus , Petrosia , unidentified 01CJ23 sponge). Sponge groups A, B, and D were unstable clusters with low boostrap support (B/50%) and may be artifacts. Further study with more archaeal sequences is necessary to evaluate the support for these sponge clusters. The presence of clones (Ar18-7 and Ar18-8) within the soil cluster group suggests contamination by sediments passing through the sponges. Similarly, two Ar08-1 and Ar19-17 clones, which were not identified within any sponge groups, may be contaminants of seawater archaea (Fig. 2). Host-specific archaeal sequence clusters were not found in this study.
Acknowledgements This work was supported by funds received from the Life Phenomenon and Function Research Project (M10016-00-0003) of the Korea Ministry of Science and Technology. We thank two anonymous referees for their critical review of the manuscript and their comments. The first author gratefully acknowledges Seung-Seup Bae, Jisun Yu, Kyoung-Hee Cho, and Hyo-Won Kim for technical assistance.
References [1] Larkum AWD, Cox GC, Hiller RG, Parry DL, Dibbayawan TP. Mar Biol 1987;95:1 /13. [2] Santavy DL, Willenz P, Colwell RP. Appl Environ Microbiol 1990;56:1750 /62. [3] Douglas AE. Symbiotic interactions. Oxford: Oxford University Press, 1994:148. [4] Preston CM, Wu KY, Molinski TF, DeLong EF. Proc Natl Acad Sci USA 1996;93:6241 /6. [5] Lee YK, Lee J-H, Lee HK. J Microbiol 2001;39:254 /64. [6] Wilkinson CR, Nowak M, Austin B, Colwell RR. Microb Ecol 1981;7:13 /21. [7] Bewley CA, Faulkner DJ. Angew Chem Ausg B 1998;37:2162 /78.
304
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
[8] Faulkner DJ, Harper MK, Salomon CE, Schmidt EW. Mem Queensl Mus 1999;44:167 /73. [9] Schmidt EW, Obraztsova AY, Davidson SK, Faulkner DJ, Haygood MG. Mar Biol 2000;136:969 /77. [10] Davidson SK, Allen SW, Lim GE, Anderson CM, Haygood MG. Appl Environ Microbiol 2001;67:4531 /7. [11] Amann RI, Ludwig W, Schleifer K-H. Microbiol Rev 1995;59:143 /69. [12] Althoff K, Schu¨tt C, Steffen R, Batel R, Mu¨ller WEG. Mar Biol 1998;30:529 /36. [13] Mu¨ller WEG. Mar Biol 1999;134:213 /5. [14] Bewley CA, Holland ND, Faulkner DJ. Experientia 1996;52:716 / 22. [15] Webster NS, Hill RT. Mar Biol 2001;138:843 /51. [16] Hentschel U, Hopke J, Horn M, Friedrich AB, Wagner M, Hacker J, Moore BS. Appl Environ Microbiol 2002;68:4431 /40. [17] Takai K, Moser DP, DeFlaun M, Onstott TC, Fredrickson JD. Appl Environ Microbiol 2001;67:5750 /60. [18] Woese CR, Kandler O, Whiilis ML. Proc Natl Acad Sci USA 1990;87:4576 /9. [19] DeLong EF. Proc Natl Acad Sci USA 1992;89:5685 /9. [20] Fuhrman JA, McCallum K, Davis AA. Nature 1992;371:695 /7. [21] Fuhrman JA, Davis AA. Mar Ecol Prog Ser 1997;150:275 /85. [22] Bintrim SB, Donohue TJ, Handelsman J, Robers GP, Goodman RM. Proc Natl Acad Sci USA 1997;94:277 /82.
[23] Friedrich MW, Schmitt-Wagner D, Lueders T, Brune A. Appl Environ Microbiol 2001;67:4880 /90. [24] Junghans H, Metzlaff M. BioTechniques 1990;8:176. [25] Maidak BL, Cole JR, Lilburn TG, Parker CT, Jr, Saxman PR, Farris RJ, Garrity GM, Olsen GJ, Schmidt TM, Tiedje JM. Nucleic Acids Res 2001;29:173 /4. [26] Chun J. Computer-assisted classification and identification of actinomycetes: PhD thesis. University of Newcastle upon Tyne, Newcastle upon Tyne, United Kingdom, 1995. [27] Felsenstein J. PHYLIP, phylogenetic inference package. Version 3.5: University of Washington, Seattle, USA, 1993. [28] Felsenstein J. Evolution 1985;39:783 /91. [29] Kato C, Li L, Tamaoka J, Horikoshi K. Extremophiles 1997;1:117 /23. [30] Munson M, Nedwell DB, Embley TM. Appl Environ Microbiol 1997;63:4729 /33. [31] Schleper C, Holben W, Klenk H-P. Appl Environ Microbiol 1997;63:321 /3. [32] van der Maarel MJEC, Artz RRE, Haanstra R, Forney LJ. Appl Environ Microbiol 1998;64:2894 /8. [33] Takai K, Komatsu T, Inagaki F, Horikoshi K. Appl Environ Microbiol 2001;67:3618 /29. [34] Webster NS, Watts JEM, Hill RT. Mar Biotechnol 2001;3:600 /8. [35] Margot H, Acebal C, Toril E, Amils R, Puentes JLF. Mar Biol 2002;140:739 /45.
Diversity of symbiotic archaeal communities in marine sponges from Korea Eun-Young Lee a,1, Hong Kum Lee a, Yoo Kyung Lee a, Chung Ja Sim b, JungHyun Lee a,* a
Microbiology Laboratory, Korea Ocean Research and Development Institute, Ansan P.O. Box 29, Seoul 425-600, South Korea b Department of Biology, Hannam University, 133 Ojungdong, Taejeon 306-791, South Korea
Abstract A molecular analysis of archaeal communities in eight sponges collected along the coast of Cheju Island, Korea was conducted using terminal-restriction fragment length polymorphism (T-RFLP) in conjunction with sequencing analysis of 16S rDNA clones. The terminal-restriction fragment (T-RF) profiles showed that each sponge had a simple archaeal community represented by a single major peak of the same size except for one unidentified sponge (01CJ20). In order to identify the components of the community, 170 archaeal 16S rDNA clones were recovered from sponges and analyzed by RFLP typing. Sequences of 19 representative clones for all RFLP types found in each sponge were determined and phylogenetic analysis was carried out. Seventeen of these archaeal 16S rDNA clones showed a high similarity to marine group I, belonging to the crenarchaeotes. In the phylogenetic tree, 15 archaeal clones were grouped into five sponge-associated archaeal clusters. In the unidentified sponge sample (01CJ20), one major T-RF peak was represented by a single RFLP type (40 clones), which implied a specific relationship between the sponge and its symbiotic archaeal components. # 2003 Elsevier Science B.V. All rights reserved. Keywords: Marine sponge; Archaeal symbiont; T-RFLP; 16S rDNA analysis
1. Introduction Marine sponges are a haven to a wide variety of marine microorganisms. Some of these sponge-associated microorganisms are cyanobacteria, heterogeneous bacteria, and archaea [1 /5]. Host-specificity has been identified in some microbial symbiotic relationships [6]. Sponges commonly contain bioactive and biochemical metabolites [7]. These natural products present within sponges could be attributed to their symbiotic microbes [8,9]. Recent evidence suggests that a bacterial symbiont of the marine bryozoan Bugula neritina is responsible for the biosynthesis of bryostatins, a family of macrocyclic lactones with anti-cancer activity [10].
* Corresponding author. Tel.: /82-31-400-6243; fax: /82-31-4062495. E-mail address: [email protected] (J.-H. Lee). 1 Present address: Biological Resources Division, Biodiversity Research Department, National Institute of Environmental Research, Environmental Research Complex, Kyungseo-Dong, SeoGu, Incheon 404-170, South Korea.
The introduction of molecular approaches has identified the diversity of such microbes, and these techniques have allowed the recovery of whole communities of environmental microbes, including some that are unculturable [11]. Previous studies have used 16S rDNA sequences to clarify the symbiotic relationships between eubacteria and Halichondria [12,13]. One sponge can harbor diverse symbionts, including heterotrophic bacteria, cyanobacteria, green sulfur bacteria, and archaea [12,14,15]. Recently, Hentschel et al. [16] identified sponge-specific sequence clusters belonging to various eubacterial divisions, and host-specific eubacterial sequence clusters derived from three sponge species, including Aplysina, Theonella, and Rhopaloides. Recent investigations of microbial communities have focused primarily on eubacterial communities derived from environmental surroundings [13,16,17]. Relatively few studies of microbial communities have dealt with archaeal communities associated with sponges. Archaea is one of three major microbial domains and is generally divided into three phylogenetic lineages: Crenarchaeota,
1389-0344/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1389-0344(03)00034-0
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
300
Euryarchaeota, and marine archaea recovered from the deep sea [18 /21]. In contrast to Crenarchaeota, which are extreme thermophiles with limited habitats, Euryarchaeota include thermophiles, sulfur-metabolizing microorganisms, and methanogens. Recently, novel archaea have been found in more diverse aquatic and terrestrial environments, even from other living organisms, extending the biotopes of crenarchaeotes [17,22,23]. In the study of the unculturable archaeal community associated with sponges, Preston et al. [4] identified a symbiotic crenarchaeote, Cenarchaeum symbiosum , in a marine sponge, Axinella mexicana , using 16S rDNA sequences and fluorescence in situ hybridization (FISH). The current study aimed to survey and compare the archaeal community structure associated with various sponges, and to characterize each archaeal component of the communities using molecular analysis based on 16S rDNA sequences.
2.2. Archaeal rDNA amplification and terminalrestriction fragment length polymorphism (T-RFLP) analysis Archaeal 16S rDNA was partially amplified with 21F (Sigma-Genosys, USA), and labeled at the 5? end with hexachlorofluorescein (HEX) or unlabeled 21F and 958R primers designed by DeLong [19]. The PCR reaction was performed in a thermal cycler (Biometra, Germany) with an initial denaturation step of 95 8C for 5 min followed by 30 cycles with the temperature profile, 95 8C for 1 min, 55 8C for 1 min, 72 8C for 1 min, ending with a final extension step of 72 8C for 7 min. Each amplicon was digested by three four-cutter restriction enzymes (Hha I, Hae III, Rsa I; Promega, WI), and analyzed with an automated sequencer (ABI377) equipped with the program GenScan (PE Applied Bioystems). GenScan 2500-Rox was used as an internal size standard and T-RFLP analysis for each sponge was done in triplicate. 2.3. Cloning and sequencing archaeal rDNA
2. Material and methods
2.1. Sponges and DNA extraction Eight sponges with different morphologies were collected at depths of 15/30 m on SCUBA dives on the coasts of Cheju and Mara Island in southern Korea, in June 2001. The sponges were classified as Spirastrella panis (01CJ08), Halichondria sp. (01CJ12), Halichondria okadai (01CJ19), Sarcotragus sp. (01CJ19), Petrosia sp. (01CJ24), and Erylus nobilis (01CJ25). Two sponges were unidentified and given the respective designations, 01CJ20 and 01CJ23 (Table 1). The sponges were transferred to the laboratory on ice and stored at / 80 8C. DNA was extracted from 2 g (wet weight) of each sponge, according to the modified methods of Junghans and Metzlaff [24].
Archaeal rDNA was amplified with the same DNA used in T-RFLP analysis. Amplified 16S rDNAs were cloned into pGEM T-Easy Vector (Promega, WI). A total of 170 16S rDNA clones were recovered from the eight sponges. All the clones were characterized by RFLP to select the different recombinant clones. A total of 19 clones (1 from a single 01CJ08 RFLP type, 2 from two 01CJ12 RFLP types, 5 from five 01CJ18 RFLP types, 3 from 01CJ19 RFLP types, 1 from a 01CJ20 RFLP type, 3 from 01CJ23 RFLP types, 3 from 01CJ24 RFLP types, and 1 from a 01CJ25 RFLP type) were selected from the 170 archaeal clones (Table 2). The sequences of the 19 representative RFLP clones were determined using an ABI377 (Applied Bioystems) automated sequencer. The sequences were analyzed using the CHECK_CHIMERA program at the Ribosomal Database Project II website to detect the chimeric rDNA clones [25]. None of the identified sequences were chimeric. 2.4. Phylogenetic analysis
Table 1 Total DNA yield from extraction Sponges
Classification
mg/g (wet weight)
01CJ08 01CJ12 01CJ18 01CJ19 01CJ20 01CJ23 01CJ24 01CJ25
S. panis Halichondria sp. H. okadai Sarcotragus sp. Unidentified sponge Unidentified sponge Petrosia sp. E. nobilis
5 8 0.8 3 5 15 5 14
Sequences of clones were edited using PHYDIT program ver. 3.2 [26], and a NCBI Blast search was performed to identify the nearest neighbor to the cloned sequence. Sequences were aligned with prokaryotic SSU rDNA data retrieved from the RDP website to create a matrix. Neighbor-joining analysis, based on a Kimura 2parameter matrix, was performed using both the collective sequence matrix and the program package PHYLIP to examine archaeal phylogenetic relationships [27]. To estimate the relative strength of a given clade, a model for equal change at all sites was employed and subjected to 1000 rounds of bootstrap resamplings [28].
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
301
Table 2 Description of representative archaeal clones based on RFLP/Hae III typing Sponge
RFLP type
Clone number
Representative clone
T-RF/Hae III
T-RF/Hha I
T-RF/Rsa I
Accession number
01CJ08 01CJ12
01CJ24
I I II I II III IV VI I II IV V I II IV I
19 15 4 7 8 2 1 1 16 3 1 40 8 3 5 26
01CJ25
II II
4 20
Ar08-1 Ar12-1 Ar12-16 Ar18-1 Ar18-8 Ar18-4 Ar18-3 Ar18-7 Ar19-1 Ar19-24 Ar19-36 Ar20-3 Ar23-13 Ar23-28 Ar23-23 Ar24-16 Ar24-9 Ar24-5 Ar25-51
213 213 213 213 213 241 241 46 213 213 213 153 213 213 213 213 213 213 213
326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326 326
262 262 262 262 262 262 262 262 262 262 262 295 262 262 262 262 262 262 262
AY192627 AY192628 AY192634 AY192644 AY192638 AY192635 AY192629 AY192637 AY192630 AY192639 AY192640 AY192631 AY192632 AY192642 AY192641 AY192643 AY192636 AY192645 AY192633
01CJ18
01CJ19
01CJ20 01CJ23
3. Results and discussion 3.1. DNA assemblages in the sponges Total genomic DNA extracted from all eight sponges ranged from 0.8 to 15 mg/g (wet weight of sponge) (Table 1). A 900 bp fragment of a partial archaeal 16S rDNA was amplified from each of the eight sponges using Hex21F, unlabeled 21F, and 958R (data not shown). Amplification of archaeal 16S rDNA sequences, using archaeal-specific primers, in all eight-sponge types provided further evidence of the extensive habitats that archaea can occupy, in addition to freshwater, coastal sediment, deep sea sediment and other organisms [29 / 31]. 3.2. T-RFLP analysis Archaeal community structures in the eight sponges were assessed using a T-RFLP analysis. Terminalrestriction fragment (T-RF) profiles were generated from the digestion of fluorescent-labeled PCR fragments using the restriction enzymes Hae III, Hha I, and Rsa I. The resulting profiles showed one major peak in all sponge samples (Fig. 1). In all eight sponges, a 326 bp fragment was identified with the restriction enzyme Hha I. A single major peak was also identified in most sponges using Rsa I and Hae III (262 and 213 bp, respectively). The only exception was observed in one unidentified sponge (01CJ20), which produced fragments of 295 (Rsa I) and 153 bp (Hae III) (Table 2, Fig. 1). In this survey of sponge-derived archaeal communities, the T-RF pattern with a single peak was
Fig. 1. T-RFLP profiles generated from a Hae III restriction digest and identification of T-RFLP peaks from a clone library. (A) S. panis (01CJ08); (B) Halichondria sp. (01CJ12); (C) H. okadai (01CJ18); (D) Sarcotragus sp. (01CJ19); (E) unidentified sponges (01CJ20); (F) unidentified sponges (01CJ23); (G) Petrosia sp. (01CJ24); and (F) E. nobilis (01CJ25).
302
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
Fig. 2. A neighbor-joining tree based on partial archaeal 16S rDNA clones associated with the sponges. Bold letters indicate the determined archaeal sequences in this study. The boxes represent each sponge-associated archaeal cluster. Values near the branches are based on 1000 bootstrap replications.
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
consistent, showing archaeal diversity within and between the sponges to be similar and simplistic. This level of archaeal diversity was much lower than that observed in studies of marine fish [32] and chimney structure [33]. There were no clear differences in the composition of the archaeal community among the sponges studied due to the identification of the single major T-RF peak. Only the unidentified sponge (01CJ20) had a small peak shift, suggesting a distinct archaeal phylotype from those found in the other sponges. 3.3. Identification of archaeal components in the T-RFLP analysis Cloning and sequencing analysis was used to characterize each peak from the T-RF profiles. 16S rDNA clones of sponge-associated archaea were obtained for each sponge sample. A total of 170 clone libraries, including 16/40 clones from each sponge, were analyzed by RFLP typing using Hha I. The 170 clones resulted in six different RFLP patterns (Table 2). Each sponge had at least one RFLP type. The predominant RFLP type, type I (76 clones; 44.7%) was seen in six sponges. In the two unidentified sponges, 01CJ20 and 01CJ25, RFLP types V and II were dominant, respectively. Clones representing the different RFLP types of each sponge were sequenced as described in Table 2. The 16S rDNA sequences of the archaeal clones have been submitted to GenBank under accession numbers AY192627 to 192645 (Table 2). We analyzed T-RF peak data with sequence analysis and showed that the single peak was associated with different sequences of the same size. However, the single peak identified from each of Spirastrella (01CJ08), Erylus (01CJ25), and an unidentified sponge (01CJ20) had only one type of clone, (Ar08-1, Ar25-51, and Ar203, respectively) as a representative. Sequence analysis showed that most archaeal clones (89%; 17 clones) were affiliated with the crenarchaeotes, especially marine group I, except for clones Ar18-7 and Ar18-8, which belonged to the soil crenarchaeal group (Fig. 2). The neighbor-joining analysis grouped the 15 archaeal sequences into clusters of sponge-derived clones in marine group I. Within marine group I, based on the clusters, preliminary sponge archaeal groups were first designated as A /E (Fig. 2). Specifically, archaeal clones (Ar12-1, Ar18-3, Ar18-4, Ar20-3) in sponge groups C and E were strongly associated with previously known archaeotes associated with sponges. The other eleven clones were weakly associated with sponge groups A, B, and D. In sponge group C, clone Ar12-1 from sponge Halichondria was highly (100%) affiliated with the crenarchaeote derived from Rhopaloeides odorabile [34], showing non-host specificity. Sponge group E showed that C. symbiosum from the sponge Axinella
303
[4], which has been described as unique, grouped consistently (100%) with Ar20-3 and clustered significantly (91%) with Ar18-3 and Ar18-4 of Halichondria . Considering the monophyletic sponge group E clade, which was composed of Ar20-3, Ar18-3, and Ar18-4 archaeal sequences, the identification of three archaea similar to C. symbiosum on the Korean coast verifies the cosmopolitan distribution of this Cenarchaeum cluster, which includes the coast of California and Korea [4]. Margot et al. [35] collected filamentous archaea from Axinella sponges, which are phylogenetically closely related to C. symbiosum , from the Spanish Mediterranean coast. Sponge archaeal groups C and E appeared to be clusters with sponge-specific sequences, while sponge groups A, B, and D were associated with archaeal clones derived from five different sponges (Erylus , Halichondria , Sarcotragus , Petrosia , unidentified 01CJ23 sponge). Sponge groups A, B, and D were unstable clusters with low boostrap support (B/50%) and may be artifacts. Further study with more archaeal sequences is necessary to evaluate the support for these sponge clusters. The presence of clones (Ar18-7 and Ar18-8) within the soil cluster group suggests contamination by sediments passing through the sponges. Similarly, two Ar08-1 and Ar19-17 clones, which were not identified within any sponge groups, may be contaminants of seawater archaea (Fig. 2). Host-specific archaeal sequence clusters were not found in this study.
Acknowledgements This work was supported by funds received from the Life Phenomenon and Function Research Project (M10016-00-0003) of the Korea Ministry of Science and Technology. We thank two anonymous referees for their critical review of the manuscript and their comments. The first author gratefully acknowledges Seung-Seup Bae, Jisun Yu, Kyoung-Hee Cho, and Hyo-Won Kim for technical assistance.
References [1] Larkum AWD, Cox GC, Hiller RG, Parry DL, Dibbayawan TP. Mar Biol 1987;95:1 /13. [2] Santavy DL, Willenz P, Colwell RP. Appl Environ Microbiol 1990;56:1750 /62. [3] Douglas AE. Symbiotic interactions. Oxford: Oxford University Press, 1994:148. [4] Preston CM, Wu KY, Molinski TF, DeLong EF. Proc Natl Acad Sci USA 1996;93:6241 /6. [5] Lee YK, Lee J-H, Lee HK. J Microbiol 2001;39:254 /64. [6] Wilkinson CR, Nowak M, Austin B, Colwell RR. Microb Ecol 1981;7:13 /21. [7] Bewley CA, Faulkner DJ. Angew Chem Ausg B 1998;37:2162 /78.
304
E.-Y. Lee et al. / Biomolecular Engineering 20 (2003) 299 /304
[8] Faulkner DJ, Harper MK, Salomon CE, Schmidt EW. Mem Queensl Mus 1999;44:167 /73. [9] Schmidt EW, Obraztsova AY, Davidson SK, Faulkner DJ, Haygood MG. Mar Biol 2000;136:969 /77. [10] Davidson SK, Allen SW, Lim GE, Anderson CM, Haygood MG. Appl Environ Microbiol 2001;67:4531 /7. [11] Amann RI, Ludwig W, Schleifer K-H. Microbiol Rev 1995;59:143 /69. [12] Althoff K, Schu¨tt C, Steffen R, Batel R, Mu¨ller WEG. Mar Biol 1998;30:529 /36. [13] Mu¨ller WEG. Mar Biol 1999;134:213 /5. [14] Bewley CA, Holland ND, Faulkner DJ. Experientia 1996;52:716 / 22. [15] Webster NS, Hill RT. Mar Biol 2001;138:843 /51. [16] Hentschel U, Hopke J, Horn M, Friedrich AB, Wagner M, Hacker J, Moore BS. Appl Environ Microbiol 2002;68:4431 /40. [17] Takai K, Moser DP, DeFlaun M, Onstott TC, Fredrickson JD. Appl Environ Microbiol 2001;67:5750 /60. [18] Woese CR, Kandler O, Whiilis ML. Proc Natl Acad Sci USA 1990;87:4576 /9. [19] DeLong EF. Proc Natl Acad Sci USA 1992;89:5685 /9. [20] Fuhrman JA, McCallum K, Davis AA. Nature 1992;371:695 /7. [21] Fuhrman JA, Davis AA. Mar Ecol Prog Ser 1997;150:275 /85. [22] Bintrim SB, Donohue TJ, Handelsman J, Robers GP, Goodman RM. Proc Natl Acad Sci USA 1997;94:277 /82.
[23] Friedrich MW, Schmitt-Wagner D, Lueders T, Brune A. Appl Environ Microbiol 2001;67:4880 /90. [24] Junghans H, Metzlaff M. BioTechniques 1990;8:176. [25] Maidak BL, Cole JR, Lilburn TG, Parker CT, Jr, Saxman PR, Farris RJ, Garrity GM, Olsen GJ, Schmidt TM, Tiedje JM. Nucleic Acids Res 2001;29:173 /4. [26] Chun J. Computer-assisted classification and identification of actinomycetes: PhD thesis. University of Newcastle upon Tyne, Newcastle upon Tyne, United Kingdom, 1995. [27] Felsenstein J. PHYLIP, phylogenetic inference package. Version 3.5: University of Washington, Seattle, USA, 1993. [28] Felsenstein J. Evolution 1985;39:783 /91. [29] Kato C, Li L, Tamaoka J, Horikoshi K. Extremophiles 1997;1:117 /23. [30] Munson M, Nedwell DB, Embley TM. Appl Environ Microbiol 1997;63:4729 /33. [31] Schleper C, Holben W, Klenk H-P. Appl Environ Microbiol 1997;63:321 /3. [32] van der Maarel MJEC, Artz RRE, Haanstra R, Forney LJ. Appl Environ Microbiol 1998;64:2894 /8. [33] Takai K, Komatsu T, Inagaki F, Horikoshi K. Appl Environ Microbiol 2001;67:3618 /29. [34] Webster NS, Watts JEM, Hill RT. Mar Biotechnol 2001;3:600 /8. [35] Margot H, Acebal C, Toril E, Amils R, Puentes JLF. Mar Biol 2002;140:739 /45.