Aquatic Biosystems

Background: Prior research on the microorganisms associated with the brine shrimp, Artemia franciscana, has mainly been limited to culture-based identification techniques or feeding studies for aquaculture. Our objective was to identify bacteria and archaea associated with Artemia adults and encysted embryos to understand the role of microbes in the Artemia life cycle and, therefore, their importance in a hypersaline food chain. Results: We used small subunit (SSU) 16S ribosomal RNA gene sequencing to identify bacteria and archaea associated with adults and encysted Artemia embryos from one of their natural environments – Great Salt Lake (GSL), Utah, USA. We found that bacterial sequences most closely related to the genera Halomonas and Vibrio were commonly extracted from GSL adult Artemia, while bacterial sequences most similar to the genera Halomonas, Psychroflexus and Alkalilimnicola dominate in GSL water. Encysted embryos (cysts) yielded bacterial sequences from the genera Idiomarina and Salinivibrio, which were absent from adults and water. Common archaeal sequences in adults were most closely related to the genera Haloterrigena and Haloarcula, while all of the archaeal sequences from GSL water were most similar to the genus Halogeometricum. Cyst derived archaeal sequences were most closely related to the genera Halorubrum and Haloarcula. Conclusions: In addition to identifying microbial rRNA sequences that are specific to different stages of the Artemia life cycle, we observed striking differences in the sequences associated with the adult Artemia population in samples collected from GSL at different times and locations. While our study was limited in scope and the sample was small, our findings provide a foundation for future research into how the bacteria and archaea associated with Artemia influence the Artemia life cycle, and GSL food web. Background Artemia franciscana, the brine shrimp that inhabits many hypersaline environments including Great Salt Lake (GSL) in Utah, is an important food source for migrating birds and is used as fish food in aquaculture [1,2]. Most studies concerning Artemia and microbes are related to industrial production of Artemia [3]. These studies indicate that microorganisms may be involved in the Artemia life cycle as a food source [4] and for protection from pathogenic bacteria [5]. Using electron microscopy, intracellular symbiotic microbes, identified as spirochetes, have been detected in the epithelial * Correspondence: [email protected] Department of Biology and Great Salt Lake Institute, Westminster College, 1840 South 1300 East, Salt Lake City, UT 84105, USA Full list of author information is available at the end of the article © 2013 Riddle et al.; licensee BioMed Central Commons Attribution License (http://creativec reproduction in any medium, provided the or cells of the midgut in GSL Artemia [6]. Little is known, however, about the interaction between brine shrimp and the microbial community in their natural ecosystem. In this study, we present bacterial and archaeal 16S rRNA gene sequences isolated from Artemia adults and dormant encysted Artemia embryos (cysts) from GSL. We focused our efforts on Great Salt Lake (GSL) in Utah, which is one of the largest hypersaline lakes in the world. Artemia are abundant inhabitants of the moderately saline South Arm of GSL (~11-15% w/v sodium chloride equivalent in recent years), which is separated from the much saltier North Arm by a 50-year old railroad causeway [7]. The lake has a relatively simple food web involving roughly 250 million migrating birds that eat the two invertebrate inhabitants of GSL – the brine fly (Ephedra spp.) and Artemia [1]. GSL Artemia are Ltd. This is an Open Access article distributed under the terms of the Creative ommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and iginal work is properly cited. Riddle et al. Aquatic Biosystems 2013, 9:7 Page 2 of 11 http://www.aquaticbiosystems.org/content/9/1/7 thought to be supported by a rich phytoplankton community dominated by the species Dunaliella viridis [8] yet little is known regarding the bacteria and archaea that live in association with GSL Artemia or how these microbes contribute to the GSL food web. Previous studies that have identified microorganisms from adult brine shrimp in natural environments or from commercially harvested cysts have used culture techniques to identify a range of gram negative and gram positive bacterial species [9-16], with one exception that examined bacterial Ribosomal RNA (rRNA) gene sequences associated with Artemia from salterns in Isreal [17]. No archaeal species have been identified. Since microbes from aquatic environments are often difficult to culture or are unculturable in a laboratory setting [18,19], the results of these previous cultured-based studies may be dominated by microbes that grew more successfully in culture rather than represent a sample of the natural populations [20,21]. rRNA gene sequence has been used as a cultureindependent method to identify bacteria and archaea that inhabit natural environments from oceans to deserts, inhospitable places such as rocks and salt crystals, as well as microbes that are associated with other living organisms [22-26]. While culture-independent techniques may have their own biases [27,28], we used this method to identify a different set of microorganisms present in Artemia samples than previously reported in the literature (with the exception of members of the genus Vibrio identified by several studies and members of the genera Halomonas, Salinivibrio, and Roseovarius [17]), including the first report of archaea associated with Artemia. Our initial step toward understanding how microorganisms affect the Artemia life cycle and surrounding food web was to identify microbial 16S rRNA genes associated with adult Artemia. We then compared these data to sequences from GSL water to test the hypothesis that Artemia harbor some microbes at much higher concentrations than the surrounding water. We also hypothesized that some microbes may be specific to encysted embryos, and that there may be some microbes in common between cysts and adults. In order to test this hypothesis, we identified microbial 16S rRNA sequences associated with Artemia cysts from GSL and compared them to the sequences from adults and GSL water. And finally, we hypothesized that microbes found in association with GSL Artemia may also be found in populations from other hypersaline ecosystems. Therefore, we expected to find similarities in the 16S rRNA gene sequences that we isolated from GSL Artemia cysts and Artemia cysts from the “San Francisco Bay” (SFB) strain harvested from another location. Results Sequences from adult Artemia We collected 75 sequences from GSL adult Artemiaderived clones, 37 were amplified with the bacterial primer set (see Methods and Table 1A) and 48 with the archaeal primer set (see Methods and Table 2A). The bacterial sequences represent seven different genera based on RDP classification: Vibrio, Halolactibacillus, Halomonas, Roseovarius, Lutibacter, Alkalilimnicola, and Caulobacter (Table 1A). Based on BLAST analysis, six of the eight distinct sequences most closely matched sequences isolated from uncultured microbes. The results of RDP Classifier analysis indicated that the archaeal sequences represent five genera: Haloterrigena, Haloarcula, Natronobacterium, Halogeometricum, and Halovivax, and according to BLAST analysis three of these groups most closely matched uncultured clones (Table 2A). The most abundant archaeal sequence isolated from adults (AAC1) was most identical to Haloterrigena limicola by BLAST. It represents over half of the adult derived clones and was not found in our water or encysted Artemia embryo samples. However, we did isolate a single sequence (GAU1) from GSL cysts that was classified as the same genus as AAC1 but matched Haloterrigena saccharevitans by BLAST, and when aligned with AAC1 was only 93% identical so it was considered to represent a different organism. Of the bacterial clones amplified from adult Artemia, 21 clones were from the sample collected in the Fall of 2006 at Black Rock and 15 clones were from the sample collected in Spring 2007 at DWR3. The number of sequences that made up some contigs varied between these two samples. For example, contig ABC1 consisted of clones that were all from the Spring 2007/DWR3 sample, suggesting that this sequence was much more abundant at that time and location. This sequence was classified as being from the genus Vibrio according to RDP Classifier and BLAST. Sequence ABC2 was isolated with similar frequency in both samples, while ABC3 (identified as Halomonas by RDP and BLAST) was more abundant in the adult Artemia from the Fall 2006/Black Rock sample. Percent abundance of clones from each sample was determined for each sequence and is graphed in Figure 1. We found a significant difference in the bacterial sequence distribution between the two samples (p < .001). The archaeal data from adults were also analyzed for variation between samples as described above (Figure 2). A total of 31 clones from the Fall 2006/Black Rock sample and 17 clones from the Spring 2007/DWR3 sample were sequenced. The number of AAC1 (Haloterrigena) clones from the Fall 2006/Black Rock sample was higher than the number of contributing clones from the Spring Table 1 Phylogenetic affiliations of the uncultured bacteria based on 16S rDNA analysis Sequence Clones Closest BLASTn match to NCBI nr database Identities (%) Genus [confidence value] Genbank Accession A. Sequences obtained from GSL Adult Artemia ABC1 9 DQ068937.1 Vibrio metschnikovii 627/631 (99) Vibrio[100%] KC696895-KC696903 ABC2 5 AB362696.1 Halolactibacillus halophilus 615/631 (97) Halolactibacillus[82%] KC696904-KC696906 ABC3 15 DQ351910.1 Uncultured bacterium clone JH-WH17 592/603 (98) Halomonas[100%] KC696872-KC696886 ABC4 4 AM691100.1 Uncultured “Rhodobacteraceae” 582/589 (98) Roseovarius[100%] KC696887-KC696892 ABU1 1 EU245085.1 Uncultured organism clone MAT-CR-H1-G02 629/661 (95) Alkalilimnicola[16%] KC696869 ABU4 1 AM420114.1 Uncultured alpha proteobacterium 532/532 (100) Caulobacter[98%] KC696870 ABU6 1 DQ396185.1 Uncultured organism clone ctg_NISAA81 581/587 (98) Halomonas[100%] KC696871 ABU7 1 DQ154838.1 Uncultured bacterium clone GN01-0.012 513/534 (96) Lutibacter [15%] KC696893-KC696894 B. Sequences obtained from GSL Artemia cysts GBRC1 14 DQ462298.1 Uncultured bacterium clone e41 525/531 (98) Idiomarina[100%] KC703296-KC703308 GBRC2 4 X95527.1 SCRR701T S. costicola (strain NCIMB 701-T) 496/497 (99) Salinivibrio[100%] KC70329-KC703295 GBRU1 1 EU287134.1 Uncultured bacterium clone P13-41 406/430 (94) Marinimicrobium[96%] KC703293 GBRU2 1 EF190068.1 Uncultured Psychroflexus sp. clone GSX1 543/587 (92) Gramella[41%] KC703291 GBRU3 1 DQ157009.1 “Marinobacter haloterrigenus” strain FP2.5 488/518 (94) Marinobacter[54%] KC703292 C. Sequences obtained from SFB strain Artemia cysts SBR1 15 X95527.1 SCRR701T S. costicola (strain NCIMB 701-T) 618/622 (99) Salinivibrio[100%] KC696910-KC696930 SBR2 3 EF177666.1 Idiomarina sp. Y24 564/565 (99) Idiomarina[100%] KC696931-KC696933 SBU1 1 EU135665.1 Lysobacter sp. YIM C734 566/596 (94) Lysobacter[75%] KC696909 SBU2 1 AF505746.1 Gamma proteobacterium UMB21A 562/566 (99) Psychrobacter[100%] KC696907 SBU3 1 AM084035.1 Acidovorax sp. R-25076 564/564 (100) Acidovorax[100%] KC696908 D. Sequences obtained from GSL water HBRC2 4 AM691089.1 Uncultured Psychroflexus sp. isolate EG26 554/559 (99) Psychroflexus[100%] KC696938-KC696940, KC696944-KC696945 HBRC3 4 AM691089.1 Uncultured Psychroflexus sp. isolate EG26 531/546 (97) Psychroflexus[99%] KC696935-KC696937, KC696941-KC696943 HBRC4 9 EF554887.1 Halomonas sp. G27 512/527 (97) Halomonas[100%] KC696946-KC696952, KC696963-KC696966 HBRC5 7 AY862776.2 Uncultured proteobacterium clone At18AugB10 492/492 (100) Alkalilimnicola[74%] KC696953-KC696962 HBUN1 1 AY862797.2 Uncultured actinobacterium clone At12OctB10 413/418 (98) Agrococcus [59%] KC696934 Sequence names, number of clones, BLASTn matches with percent identity, and genus assignment according to RDP Classifier with confidence values in brackets are listed for all of the bacterial sequences. Part A includes sequences derived from GSL Artemia adults, Part B is sequences from GSL Artemia cysts, Part C is sequences from SFB strain Artemia cysts, and Part D shows sequences from GSL water. Riddle et al. Aquatic Biosystems 2013, 9:7 Page 3 of 11 http://www.aquaticbiosystems.org/content/9/1/7 2007/DWR3 sample. This suggests that sequence AAC1 was more abundant in the Fall 2006/Black Rock sample. Sequence AAC3 (Haloarcula) was constructed of clones only from the Spring 2007/DWR3 sample. There was a significant difference in sequence distribution between these two samples (p < .001). Sequences from encysted Artemia embryos To continue our search for microbes associated with Artemia, we amplified bacterial and archaeal 16S rRNA genes from encysted Artemia embryos from two different sources of commercially harvested dry cysts – one identified as “San Francisco Bay” (SFB) strain and one from Great Salt Lake (GSL) (see Methods). A total of 45 clones isolated from GSL cysts were sequenced, 20 from the bacterial primer set (Table 1B) and 25 from the archaeal set (Table 2B). A total of 36 clones isolated from SFB strain cysts were sequenced, 21 from the bacterial primer set (Table 1C) and 15 from the archaeal primer set (Table 2C). Some of the SFB strain cyst derived sequences were classified as the same genus by RDP, yet were dissimilar enough to have different closest match sequences according to BLAST analysis so were considered separately. Sequences that were most identical to Salinivibrio costicola by BLAST were found in both SFB strain and GSL cysts. When the sequences (GBRC2 and SBR1) were compared they were 99% identical which indicates that they are likely to be the same species or very closely related. Also, sequences classified as Idiomarina were found in both SFB strain and GSL cysts (GBRC1 and SBR2, 99% identical). Table 2 Phylogenetic affiliations of the uncultured archaea based on 16S rDNA analysis Sequence Clones Closest BLASTn match to NCBI nr database Identities (%) Genus [confidence value] Genbank Accession A. Archaeal sequences obtained from GSL adult Artemia AAC1 26 DQ367241.1 Haloterrigena limicola strain AX-7 481/485 (99) Haloterrigena[100%] KC696970-KC696994 AAC3 17 EF645686.1 Haloarcula japonica strain JCM7785 502/504 (99) Haloarcula[100%] KC696995-KC697011 AAC4 2 AJ969886.1 Uncultured archaeon, clone ss_014 330/354 (93) Natronobacterium[28%] KC697012-KC697013 AAC5 2 AJ969890.1 Uncultured archaeon, clone ss_010a 498/503 (99) Halogeometricum[53%] KC696968-KC696969 AAU9 1 EF690637.1 Uncultured haloarchaeon clone TX4CA_82 497/520 (95) Halovivax[57%] KC696967 B. Archaeal sequences obtained from GSL Artemia cysts GAC1 10 AY510707.1 Halorubrum xinjiangense 522/524 (99) Halorubrum[100%] KC697020-KC697029 GAC2 8 EF645686.1 Haloarcula japonica strain JCM7785 522/524 (99) Haloarcula[100%] KC697030-KC697037 GAC3 6 AM269467.1 Halovivax ruber type strain XH-70 T 507/519 (97) Halovivax[89%] KC697015-KC697019 GAU1 1 AY820137.2 Haloterrigena saccharevitans strain AB14 499/524 (95) Haloterrigena[35%] KC697014 C. Archaeal sequences obtained from SFB strain Artemia cysts SAC1 2 EF468473.1 Halorubrum tebenquichense strain JCM12290 521/524 (99) Halorubrum[100%] KC697047-KC697048 SAC2 3 AY820137.2 Haloterrigena saccharevitans strain AB14 518/524 (99) Haloterrigena[75%] KC697049-KC697051 SAC4 4 AY570917.1 “Natrinema ajinwuensis” strain AJ12 499/524 (95) Natronorubrum[81%] KC697041-KC697044 SAC5 2 DQ103672.1 Uncultured euryarchaeote clone ArcH05 479/509 (94) Halogeometricum[26%] KC697045-KC697046 SAU1 1 AJ969890.1 Uncultured archaeon clone ss_010a 378/409 (99) Halogeometricum[50%] KC697039 SAU4 1 DQ431096.1 Uncultured archaeon clone A10 458/478 (96) Halogeometricum[70%] KC697038 SAU5 1 EF632847.1 Uncultured archaeon clone Hua-w-29 300/302 (99) Halorubrum[100%] KC697040 SAU7 1 EF077641.1 “Halorubrum alimentarium” strain B43 180/181 (99) Halorubrum[97%] unavailable D. Archaeal sequences obtained from GSL water HAC1 29 AF196290.1 Archaeon JDS-1 503/504 (99) Halogeometricum[49%] KC697052-KC697086 Sequence names, number of clones, BLASTn matches with percent identity, genus assignment according to RDP Classifier with confidence values in brackets, and Genbank accession numbers are listed for all of the archaeal sequences. Part A includes sequences derived from GSL Artemia adults, Part B is sequences from GSL Artemia cysts, Part C is sequences from SFB strain Artemia cysts, and Part D shows sequences from GSL water. Riddle et al. Aquatic Biosystems 2013, 9:7 Page 4 of 11 http://www.aquaticbiosystems.org/content/9/1/7 The most abundant archaeal sequence found in our GSL cyst derived clones (Table 2B, GAC1, 40% of sequences) was most identical to Halorubrum xinjiangense by BLAST. Several different sequences classified as Halorubrum by RDP were also isolated from SFB strain Figure 1 Variation of bacterial 16S rRNA gene sequences from differe was calculated by dividing the number of clones with that sequence in the season. Only contigs that were constructed from five or more total clones closely matched is shown for reference. ABC1 was only present in the Sprin cysts at low abundances. When the sequence of GAC1 was compared to either SAU5 or SAU7 each pair was 97% identical so they were considered separately. No sequences closely related to the genus Halorubrum were found in GSL water or adult samples. Other nt GSL adult Artemia samples. Percent abundance of a sequence season being analyzed by the total number of all clones from that were included in the analysis. The genus that each sequence most