Distinctly different bacterial communities in surface and oxygen minimum 1 layers in the Arabian Sea

12 Contributions of microbial communities to biogeochemical processes in oxygen minimum 13 oceanic zones are being realized through the applications of molecular techniques. To understand 14 seasonal and depth-wise variations in bacterial community structure (BCS) in the Arabian Sea 15 oxygen minimum region, extensive sampling and molecular analyses were carried out. 16S 16 rRNA gene sequencing was done to profile the BCS from five depths, surface (5m), deep 17 chorophyll maximum (43-50m, DCM), 250m, 500m and 1000m during Spring intermonsoon 18 (SIM), Fall intermonsoon (FIM), and Northeast monsoon (NEM) seasons. Sequencing of 19 743 chimera-free clones revealed a clear vertical partitioning of BCS between the surface 20 (surface + DCM) and OMZ (250 + 500 + 1000m) layers. There was no distinct seasonal 21 difference in the BCS. Most 16S rRNA gene sequences were affiliated to Gammaproteobacteria 22 (39.31%), Alphaproteobacteria (23.56%) and Cyanobacteria (20.2%). Higher diversity and 23 OTUs in OMZ predominantly consisting of Alteromonodales, Sphinogomonadales, 24 Rhodobacterales, Burkholderales, and Acidimicrobiales we observed might be due to their 25 microaerophilic metabolism, ability to degrade recalcitrant substrates and assimilate sinking 26 particulate matter. Further hitherto undescribed diversity both in surface and OMZ layers was 27 evidenced. Implicit role of extant bacterial community in denitrification and anammox and in 28 sulphur oxidation is highlighted. 29 30


Introduction
Poor ventilation of intermediate layers, higher microbial respiration and organic matter oxidation within the water column (Wyrtki 1962) lead to oxygen minimum zones (OMZ) with dissolved oxygen (DO) concentrations often below < 20µM L -1 (Lisa 2003).Major OMZs are found in the intermediate depths of eastern tropical North Pacific (ETNP, Wyrtki 1966), eastern tropical South Pacific (ETSP, Wyrtki 1966), eastern Arabian Sea (Wyrtki 1973;Madhupartap et al. 1996;Naqvi and Jayakumar 2000) and eastern South Atlantic (Karstensen et al. 2008).Within the OMZ, intense anaerobic and related processing of nitrogenous compounds (Stramma 2008) lead to loss of fixed nitrogen to the atmosphere.Denitrification (Naqvi 1994) and anaerobic oxidation of ammonia (anammox, Dalsgaard et al. 2012) are the major pathways for this nitrogen loss.Hypoxic conditions often select resilient microbes and restrict their vertical distribution (Wishner et al. 1995).
The OMZs, earlier believed to occur mostly in nutrient rich upwelling regions, are currently expanding and/or intensifying due to anthropogenic impacts (Diaz and Rosenberg 2008).The expansion of OMZs is affecting benthic ecosystems and marine fisheries due to habitat alterations and/or changes in nutrient cycling (Stramma et al. 2008).Further, the OMZ expansion is ascribed to increased production of climate active trace gases including carbon dioxide (CO 2 ), methane (CH 4 ) and nitrous oxide (N 2 O).Microbes are involved in all such processes but little is known about their community structure and metabolism within OMZs, in particular from the Arabian Sea-OMZ (AS-OMZ).
Coinciding with an active denitrification zone (Naqvi 1994;Naqvi et al. 1998) the AS-OMZ is among the largest anoxic regions in the oceans with DO levels < 20M, sometimes dipping as low as 0.1 μM (Naqvi 2006;Paulmier and Ruiz-Pino 2009).This alone accounts for 20% of the oceanic denitrification (Codispoti et al. 2001) and contributes to 40% of the global pelagic dinitrogen (N 2 ) production (Naqvi et al. 2008).Previous studies mostly focused on delineating the role of denitrifying and anammox bacteria in the loss of fixed nitrogen in the AS-OMZ (Jayakumar et

DNA extraction and PCR amplification of 16s rRNA genes
DNA was extracted from Sterivex filters using using the modified method of Ferrari & Hollibaugh (1999).The precipitated DNA was hydrated in 50 μl sterile deionised water.All DNA samples were subjected to PCR amplification using universal 16S rRNA primers 27F and 1492R (to confirm that the DNA was of PCR quality.The 16S rRNA gene was amplified following conditions mentioned by Sambrook (1989).PCR amplification was performed in a final volume of 50 μl in a thermocycler (Applied Biosystems, USA) and correct amplification was ensured by checking for the amplicons electrophoretically.

Clone library construction and DNA sequencing
PCR amplified 16S rRNA gene products were purified using Axyprep-96 PCR Clean up kit (Axygen, Biosciences), cloned into pCR4-TOPO vector using a TOPO-TA cloning kit for sequencing (Invitrogen, USA) and transformed by chemical transformation into TOP-10 cells as per manufacturer's instructions.As the transformation efficiency was low to moderate, at least thrice cloning trials were repeated to collect a minimum of 65 clones for further analyses.All positive clones/transformants from each sample were picked out, grown overnight at 37°C on LB plates and subjected to the colony-PCR with primers sets pucM13F/pucM13R using temperature conditions as per manufacturer's instructions.PCR products were purified with the Axyprep-96 PCR Clean up kit (Axygen, Biosciences) and then sequenced using an ABI 3130XL genetic analyzer (Applied Biosystems, USA).Clone libraries were constructed from particular depths keeping DO profile in mind.However, for the purpose of comparison, clone libraries from surface and DCM depth were considered as surface group (SIM, FIM and NEM).Similarly, clone libraries from 250, 500 and 1000m were considered as OMZ group for each season

Phylogenetic analysis of clone sequences
The sequences were assembled into contigs using DNA Baser sequence assembly software version 2 (DNA baser, USA).Vector contamination was removed from the sequences using the VecScreen tool (http:// www.ncbi.nlm.nih.gov /tools/vecscreen/).Only consensus sequences without vector and primer residues and with a quality score of 20 (which translates into more than 99.5% correct bases, Allex 1999), were used for further analyses.were submitted to the NCBI GenBank database and are available under accession numbers KJ589647 to KJ590044, KR269603 to KR269693, KR919859 to KR920002, KR919859 to KR920002 and KR673365 to KR819266.

Determination of OTUs
16S rRNA gene sequences were aligned using ClustalX and the alignment was curated using Gblocks (Castresana 2000) to remove poorly aligned positions and divergent regions.
Distance matrices were created from the curated alignments with Phylip 3.66.The sequences were then assigned into phylotypes (operational taxonomic units, OTUs) using MOTHUR by applying the average neighbor rule (Schloss and Westcott 2011).97% cut-off for sequence similarity was used to delimit an OTU.

Estimation of shared and unique OTUs
In order to elucidate seasonal and depth wise differences in bacterial phylotypes, the fraction of shared and unique OTUs was estimated using MOTHUR.Further, the numbers of shared and unique OTUs between all three seasons surface samples (SIM-surface, FIM-surface, NEM-surface) and all three seasons OMZ samples (SIM-OMZ, FIM-OMZ, NEM-OMZ) were also estimated.

Construction of Phylogenetic tree
A phylogenetic tree of representative shared OTUs from the clone libraries was constructed to visualize their relationship and affiliations with the closest relative sequences from the database (Kemble et al. 2011).The tree was constructed as per MEGA-6 using maximum composite likelihood as substitution model and bootstrap values were calculated using neighborjoining method with a resampling size of n = 500.

Statistical analysis of clone libraries
The clone libraries from the ASTS were analysed statistically using

Diversity and richness estimation
The indices for diversity (Simpson's and Shannon's indices), richness estimates (Jackknife, Chao 1 and ACE), rarefaction and collectors curve were performed using MOTHUR (Schloss et al. 2009).Total richness of the clone libraries was extrapolated from the observed number of OTUs using the three nonparametric richness estimators.

Physico-chemical characteristics of the sampling site
Physico-chemical characteristics (i.e.temperature, salinity, dissolved oxygen, pH, nitrite, nitrate, ammonium, silicate, phosphate and total organic carbon) at the ASTS are described

Bacterial Community Structure
Seasonal distribution pattern of clone sequences from surface and OMZ (Figure 2) indicate that maximum numbers of clone sequences were affiliated to three major phylogenetic groups.These are Gammaproteobacteria (39.31%),Alphaproteobacteria (23.56%) and Cyanobacteria (20.2%).These groups represented up to 82% of the usable sequences.The relative proportions of sequences in these groups varied temporally, as well as spatially, with depth.The percentages of Gamma-and Alphaproteobacteria were much higher in the OMZ than in the surface layers.During NEM the percentages of Cyanobacteria and Gammaproteobacteria were the highest at the surface and in the DCM.Notably, the percentage of Gammaproteobacteria within OMZ (250, 500 and 1000m) did not vary much between seasons.
The distribution of Alphaproteobacteria in OMZ was quite similar during FIM and NEM.The highest proportion of unclassifiable bacterial sequences in the surface layers, as well as OMZ depths, was observed during SIM, followed by FIM and NEM.

Shared and Unique OTUs
The shared and unique OTUs between all three seasons in the surface and OMZ samples were divided into four categories:

Phylogenetic affiliation and phylogenetic tree of shared OTUs
The relative abundance and phylogenetic affiliation of the OTUs shared by all seasons and any of the two seasons in surface and OMZ are listed in Table 1(a) and Table 1

Comparison of clone libraries
Using the RDP classifier, it was seen that at Class level, the surface samples during all three seasons had more representatives of Cyanobacteria than in the OMZ samples.This was entirely different in the OMZ where Alphaproteobacteria, Gammaproteobacteria and Marinimicrobia were the dominant groups (Table 2).Further, from the RDP libcompare it was clear that the abundance of Gammaproteobacteria, (Altermonodales), Alphaproteobacteria (Spingomondales), and SAR11 differed significantly between seasons in the surface clone libraries.In contrast, the abundance of Cyanobacteria and Alphaproteobacteria was significantly lower and different in the OMZ (Table 3).The LIBSHUFF (p < 0.0016) analysis, used to note that bacterial community, was significantly different between surface and OMZ during all three seasons.The bacterial community in surface layers differed significantly between seasons (LIBSHUFF, p < 0.0001), whereas no such significant seasonal difference was found in the bacterial community extant in the OMZ (Table 4).5).However, during each season, higher diversity indices were evident in the OMZ than in the surface layers.The nonparametric Jackknife, Chao 1 and ACE estimators also revealed that the estimated OTUs are much higher in the OMZ than the observed number of OTUs during all the seasons.Rarefaction (Figure 6a) and collector curves (Figure 6b) implied that no saturation was reached either at sequence or OTU level at an evolutionary distance of 1% and 3%, respectively.However, at Class or Family level, at distances of 10%, the rarefaction implied some saturation.The SIM is a transition period from winter to summer and is generally known to be a period of low primary productivity caused by water column stratification and oligotrophic conditions (Madhupratap et al. 1996).During this period, the bacterial community is mainly ).We find that the bacterial diversity at the ASTS varies seasonally in the surface layers and in OMZ and is at its highest during SIM.High rates of denitrification might be among the many possible causes for this higher diversity.Further, the high proportion of unclassified bacteria may also contribute to the increased diversity to some extent as observed during SIM.
Though over 60 clones were finally available from each depth for sequencing, they do not capture the full bacterial diversity in the AS-OMZ.This notwithstanding, our efforts are useful to suggest that overall bacterial diversity at Phylum and OTU level in the ASTS is higher in OMZ than in the surface layers.Interestingly, the depth-dependent variation of OMZ bacterial diversity is not consistent among different studies (Ganesh et al.Based on diversity and richness indicators, the estimated total numbers of OTUs for all samples from the AS-OMZ were much higher than the observed number of OTUs, indicating that additional sampling would have revealed greater diversity.The same trend was observed when rarefaction and collector curves were calculated as no stationarity was reached at levels below Phylum approximated by an evolutionary distance of 10%.Together, the richness estimators ACE and Chao 1, and the three different diversity indices indicate that there is a high degree of undescribed diversity in both surface and OMZ, a large fraction of which likely belong to clusters with sequences whose taxonomic resolution is finer than that defined for species.
Results of this study demonstrate that the OMZ bacterial community is diverse yet distinctively different communities predominate in the surface and OMZ layers.Many of the groups that have adapted to low oxygen are likely to have important roles in C, N and S cycling within the OMZs.Organisms contributing to high species-richness may act as regulators of fluxes of C, N and S into and out of the OMZ.Their rates of cycling and metabolism in the OMZ seem to be stable.While seasonal variation in the OMZ bacterial community is minimal, the diversity in the OMZ depths is greater than that in the surface layers, with deep-branching groups and many novel and as yet uncultivated clades.Also, the sulphur oxidation appears to be operated by certain microbes in OMZs exemplied by sparse yet obvious presence of Thiotrichales and Desulfobacteriales in the AS-OMZ.Prevalence of denitrification, anammox and sulphur oxidation brings forth the role of microbes in the AS-OMZ ecosystem functioning.

Table 1 (a)
al. 2009; Bulow et al. 2010; Pitcher et al. 2011; Newell et al. 2011; Bouskill et al. 2012).However, detailed attempts to understand bacterial community structure and/or bacterial diversity of the AS-OMZ and its overlying surface waters are a few and far in between Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.(Fuchs et al. 2005; Jain et al. 2014).The present study thus aimed to delineate phylogenetic diversity of the overall bacterial communities in the AS-OMZ, as well as the diversity of bacterial phylotypes contributing to temporally stable bacterial community structure (BCS) in the OMZ (vide Jain et al. 2014).For this, we analyzed small subunit ribosomal RNA gene (16S rRNA or SSU rRNA) clone libraries prepared from water samples collected from five depths from the Arabian Sea Time Series station (ASTS; 17°0.126'N,67°59.772'E),during three different seasons.ASTS is akin to well known HOTS (Hawaii Ocean Time-Series) in the Pacific Ocean and BATS (Bermuda Atlantic Time-Series) in the Altantic Ocean.Water samples were collected under the SIBER program in May 2012 (Spring intermonsoon [SIM]), September 2012 (Fall intermonosson [FIM]), and February 2013 (Northeast monsoon [NEM]), as described previously by Jain et al (2014).In brief, samples for DNA extractions were collected using pre-cleaned Niskin bottles on a CTD rosette.Samples from surface (0.5m), deep chlorophyll maxima (DCM) (~35-50m), intense denitrification zone (250m, 500m) and deep denitrification zone (1000m) at the ASTS station (17°0.126'N, 67°59.772'E)were strained through 200µm pore sized bolting silk, immediately after collection, 2.5 L of seawater sample from each depth was filtered peristatically through a Sterivex cartridge fitted with 0.22 μm pore size membrane filter (Millipore, USA).The Sterivex cartridge was then filled with 1.8 ml of lysis buffer (50 mM 131 Tris pH 8.3, 40 mM EDTA and 0.75 M sucrose), sealed, and stored frozen at -80°C until nucleic acid extraction was performed in the laboratory.

1 .
earlier in Jain et al. (2014) and vertical distribution of dissolved oxygen (DO), nitrite (NO 2 ) and nitrate (NO 3 ) during three seasons is shown in Figure In general, surface and DCM depths (43-50 m) are well oxygenated followed by a steep oxycline between DCM and 250 m.The average DO concentration ranged from 185.24 ± 31.1 μM L −1 at DCM to 5.56 ± 5.5 μM L −1 at 250m.The DO concentrations were slightly more at 1000 m.The nitrite concentration was 0.29 in the upper thermocline region (50m) with higher oxygen concentrations (during FIM) and 2.5 μM L −1 in the intermediate depths (250m) with low oxygen (during NEM).Surface waters during SIM and FIM were devoid of nitrate and it was quite low during the NEM.Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.

( 1 )
OTUs common to all seasons, (2) OTUs common to two seasons, (3) season-specific OTUs having >1 sequence, and (4) season-specific OTUs having only one sequence (singleton).Three OTUs common to all season's within surface (Common to all season's surface-OTUs, CTASS-OTU) was represented by a minor fraction (11-17 %) of the sequences in individual seasons (Figure 3).Further, 20 OTUs (Common to 2 seasons surface, CT2SS-OTU) were shared within surface during two of the three seasons and represented by 19 to 38 % of the sequences in individual seasons.The proportion of season-specific OTUs (containing only one sequence) in the surface was the largest (43-73%) and represented by 19-57% of sequences in individual season.A total of eight OTUs were common within OMZ (Common to all seasons OMZ, CTASO-OTU) during three seasons and represented by 14-37% of individual season sequences.Further, 20 OTUs (Common to 2 seasons OMZ, CT2SO-OTU) shared within OMZ during two of the three seasons represented 15-20 % of the individual season sequences.The proportion of Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.season-specific OTUs in the OMZ was the largest (50-63%) and represented by 28-46% of the individual sequences from any of the seasons.More than half (63%) of season-specific OTUs in the OMZ are represented by single sequences.

10 3. 6
Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.Analysis of clone diversity and richness Shannon and Simpson indices, as well as rarefaction curves, clearly indicate that at an evolutionary distance of 3% the bacterial diversity in both surface and OMZ was the highest during SIM followed by NEM and FIM (Table Traditionally the OMZs were seen as regions dominated by heterotrophic denitrification fueled by sinking of organic matter produced via photosynthesis in the sunlit surface ocean.They were also considered to process a fundamentally different microbial community and operate on a different biogeochemistry.The discovery Anammox, and active but cryptic sulfur cycle in anoxic OMZs have significantly shifted the old paradigms.For almost a decade now the OMZ gene surveys have focused extensively on microbes performing denitrification and anaerobic ammonia oxidation (anammox) with less emphasis on the overall microbial community.Microbial communities within OMZs play central roles in ocean, yet we still lack a fundamental understanding of how microbial biodiversity is distributed across the OMZs.In this regard, our efforts are useful in providing some novel insights.Arabian Sea is modulated seasonally by upwelling, winter cooling(Prasanna Kumar et al. 2001) and semi-annual reversal of monsoonal winds(Madhupratap et al. 1996).This greatly influences primary production, organic carbon concentration and flux (Hansell and Peltzer 1998) and bacterial abundance (Ramaiah et al. 1996, 2000; Jain et al. 2014).In spite of the global biogeochemical and climatic importance of AS-OMZ, spatio-temporal variation (Riemann et al. 1999; Fuchs et al. 2005; Jain et al. 2014) and phylogenetic diversity of the microorganisms inhabiting therein are only sparsely addressed (Riemann et al. 1999; Jayakumar et al. 2009; Divya et al. 2010, 2011).Oxygen deficient waters in the intermediate depths is a perennial feature of the Northeastern Arabian Sea which gets intensified during winter or NEM season due to poor water circulation and high surface Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.productivity (Naqvi et al. 1990; Prasanna Kumar and Prasad 1996).Lower DO level and higher nitrite concentration at the intermediate depth during NEM season signifies intense denitrification processes and the presence of nitrite in the thermocline region signifies the nitrification process (Sen Gupta et al. 1976).During intermonsoon (SIM and FIM) periods, the intense solar heating and weak winds stratify the Northeastern Arabian Sea surface layer, leading to depletion of nitrate in the upper euphotic zone (Muraleedharan and Prasanna Kumar 1996).
sustained by the slow-to-degrade dissolved organic carbon (DOC) from earlier phytoplankton blooms of the NEM(Ramaiah et al. 2000).Despite these environmental conditions we observed higher bacterial diversity during SIM (both in surface as well as OMZ).Most microbial habitats are spatially heterogeneous(Kassen and Rainey 2004) and can contain a large number of potential niches.Previous studies reported a positive correlation between habitat heterogeneity ("patchiness") and the phylogenetic diversity of bacteria(Korona et al. 1994;Rainey et al. 2000;Kassen et al. 2000;Zhou et al. 2002 2014).Further,Bryant et al (2012) reported a consistent decline in the bacterial diversity within the ETSP-OMZ.Similarly,Zaikova et al (2010) observed low diversity associated with the seasonal OMZ off British Columbia.In contrast,Stevens and Ulloa (2008), based on 16S rRNA clone libraries identified higher OTU diversity at the ETSP-OMZ, a pattern consistent with the observation in our study.Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.Similarly, Brown et al (2009) and Kemble et al (2011) showed elevated OTU richness to coincide with the zone of minimum oxygen at the HOTS.Bryant et al (2012) and Jain et al (2014) attributed lower diversity in the OMZ to competition for limited resources, environmentalfiltering, and lower redox potential and less readily available organic matter.On the other hand, higher diversity in the OMZ has been linked to the use of a wider range of terminal oxidants compared to the oxic depths where oxygen is the dominant electron acceptor(Stevens and Ulloa 2008).Most clades of Gammaproteobacteria are known to denitrify using nitrate as electron acceptor(Miller et al. 2010).Among them, the predominant Alteromonodales in the AS-OMZ appears to be a proficient denitrifier group.Bacterial community structure at the ASTS is dominated by phylotypes affiliated with three major phylogenetic groups viz., Gammaproteobacteria, Alphaproteobacteria and Cyanobacteria.The dominance of Alphaproteobacteria and Gammaproteobacteria, has been reported earlier from a variety of pelagic marine environments(Giovannoni and Rappé 2000), including the OMZ of the ETSP (Stevens and Ulloa 2008;Ganesh et al. 2014) and the Southern Arabian Sea(Fuchs et al. 2005).Our results suggest seasonal variation in the abundance of certain major bacterial groups mainly in the surface layers than in the OMZ depths.As Hansell & Peltzer (1998) suggested, the organic carbon concentration and changes in primary productivity patterns in the surface layers may be responsible for variation in the abundance of Gammaproteobacteria (Altermonodales) and Alphaproteobacteria (Spingomondales)and SAR11 in the OMZ libraries might be supported by microaerophilic metabolism, particle associated lifestyle and/or their ability to use abundantly available nitrate as their terminal electron acceptor for energy generation.The total number of common OTUs shared within surface clone libraries was significantly lower than that in the OMZ.All common surface OTUs constitute only a minor fraction (13 %) of the total surface sequences and were mostly affiliated with Cyanobacteria, more precisely to the genus Synechococcus and Prochlorococcus.Interestingly, majority of surface OTUs (shared between two of the three clone libraries) were also affiliated with Cyanobacteria, specifically to Prochlorococcus.Cyanobacteria of the genera Prochlorococcus Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.and Synechococcus have been found inhabiting the upper lit part of OMZs (Johnson et al. 1999; Goericke et al. 2000; Ulloa et al. 2006; Galán et al. 2009).However, phylotypes of Synechococcus are often restricted to the upper part of the euphotic zone and do not show a clear vertical partitioning with depth as Prochlorococcus does.Moreover Prochlorococcus is dominant in the oligotrophic waters (Olson et al. 1990; Li 1995; Liu et al. 1997a; Grob et al. 2007).The dominance of Prochlorococcus over Synechococcus in the OMZ is presumably a result of specific physiological adaptation to the prevailing light and nutrient conditions within the region.Even though few Prochlorococcus ecotypes are able to utilize oxidized forms of nitrogen, particularly nitrite (Moore et al. 2002), the biological processes of nitrate and nitrite reduction carried out by the local OMZ microbial community (Farías et al. 2007) could support their dominant presence in the AS-OMZ upper stratum.It is apparent that the higher number of common OTUs affiliated with Alteromonodales (16% of the total OMZ sequences) and Sphinogomonadales (5% of the total OMZ sequences) might contribute to the temporally stable BCS in the OMZ.However, it should be noted that proportions of phylotypes affiliated with Alteromonodales and Sphinogomonadales also varied seasonally in the surface layers.Alteromonas macleodii clustered into two major genotypic groups or ecotypes, one found in the upper water column and another in the deep water column.Martinez et al (2008) reported that the deep ecotype is better suited to microaerophilic conditions and for the degradation of recalcitrant compounds and colonizes relatively large particles that sink rapidly to meso and bathypelagic depths.On the other hand, the surface ecotype, a typical rstrategist (investing most energy in multiplying fast) has more potential for regulation and degradation of sugars and amino acids, and specializes in colonizing smaller particulate organic matter with much slower sinking rates (Perez et al. 2012).Therefore, it is possible that the persistence of Alteromonodales throughout the year in the AS-OMZ could be due to its microaerophilic metabolism and due to its ability to degrade recalcitrant compounds as well as to be able to degrade large sinking particulate matter, for eg.transparent exopolymeric particles (Dileep kumar et al. 1998 and Ramaiah et al. 2005).Alteromonadales, particularly in the genera Idiomarina and Alteromonas are known to metabolize both labile and semi-labile high molecular weight dissolved organic matter (McCarren et al. 2010).Presence of Sphingomonadales in the upper OMZ has been reported earlier by Riemann et al (1999), and although their ecological Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.significance in OMZ is not clear, Sphingomonadales has been documented to have wide metabolic capabilities (Miller et al. 2010) and can degrade aromatic compounds (Fredrickson et al. 1995).Thus, their abundance, especially of Novosphingobium and Sphingobium, in the AS-OMZ may be driven by high concentrations of phosphate and ammonia as postulated by Liu et al (2015).The SAR11 group is known to limit the flux of sulfur from the ocean by demethylating dimethylsulfoniopropionate (DMSP), which is usually degraded to volatile dimethyl sulfide(Moran et al. 2003;Howard et al. 2006).Demethylated DMSP can serve as a substrate in sulfur oxidation in the deeper layer.This group is also known to assimilate labile amino acids, thereby playing a significant role in C, N, and S cycling(Malmstrom et al. 2004).The other common OMZ-OTUs were those Burkholderilaes, Acidimicrobiales, uncultured Gammaproteobacteria, and Rhodobacterales seems to contribute to the stable BCS in the AS-OMZ.It is interesting to note that only a few of the common OMZ OTUs affiliated to known denitrifiers, suggesting that denitrification is spatially heterogenous (niche segregation).Many of the other shared OMZ OTUs (between any two seasons) and season specific OMZ OTUs were affiliated with Nitratireducter.sp, a known denitrifying bacterium.One of the most significant findings in OMZ biogeochemistry and microbiology has been the recognition of a cryptic pelagic sulphur cycle.Traditionally, it was believed that sulphur reduction would not initiate until the oxygen and nitrate/nitrite is fully consumed(Froelich et al. 1979).Surprisingly, recent metagenomic studies have discovered an abundant and diverse sulfuroxidizing microbial community in the OMZ layers.This community is particularly enriched in Deltaproteobacteria related to sulfur-oxidizing symbionts of deep sea bivalves (Canfield et al. 2010).Radiolabeled sulfate tracer experiments from the OMZ off the Chilean coast revealed significant sulfate reduction in the upper reaches of the OMZ water (Canfield et al. 2010).Sulphur oxidising bacterial clades, such as Thiotrichales and Desulfobacteriales, were present in the libraries made from 250m and 500m, although represented by only a few sequences.These are the very first indications of the existence of sulphur metabolizing bacterial community in the AS-OMZ.Biogeosciences Discuss., doi:10.5194/bg-2016-147,2016 Manuscript under review for journal Biogeosciences Published: 3 May 2016 c Author(s) 2016.CC-BY 3.0 License.
Percentage of OTUs from Arabian Sea Time Series (ASTS) location being common to all three seasons and in two 812 seasons in surface during different season and the phylogenetic affiliation of the representative sequence from each OTU.

Table 1 (b).
Percentage of OTUs from Arabian Sea Time Series (ASTS) location being common to all three seasons and in two 814 seasons in OMZ during different season and the phylogenetic affiliation of the representative sequence from each OTU.

Table 2 .
RDP library comparison results for the groups of significantly differing (p <0.01) sequences between surface and OMZ clone 842 libraries and their phylogenetic affiliations according to RDP library compare (Cole et al 2007).The more abundant group is shown in 843 bold.844 RDP
855 sequences within surface and OMZ clone libraries obtained during different seasons and their 856 phylogenetic affiliation according to RDP library compare (Cole et al 2007).The more abundant

Table 5 .
Comparison of bacterial diversity in clone libraries constructed from the surface layers and OMZ during spring intermonsoon (SIM), fall intermonsoon (FIM), and northeast monsoon (NEM) at the Arabian Sea Time Series (ASTS) location.