Microbiology of Seamounts Common Patterns Observed in Community Structure

Much interest has been generated by the discoveries of biodiversity associated with seamounts. The volcanically active portion of these undersea mountains hosts a remarkably diverse range of unusual microbial habitats, from black smokers rich in sulfur to cooler, diffuse, iron-rich hydrothermal vents. As such, seamounts potentially represent hotspots of microbial diversity, yet our understanding of the microbiology of seamounts is still in its infancy. Here, we discuss recent work on the detection of seamount microbial communities and the observation that specific community groups may be indicative of specific geochemical scenarios, such as iron and sulfur cycling. These observations are based on the metabolisms predicted by phylogenetic characteristics exhibited by the dominant populations found within these microbial communities as compared to the closest related isolate found in culture. Therefore, these studies combine the use of both cultivation-dependent and -independent analyses. Cultivation-independent studies were primarily completed using cloning and sequencing techniques targeting small subunit ribosomal gene (SSU rDNA) biomarkers along with similar biomolecular tools like terminal-restriction fragment length polymorphism (T-RFLP) and quantitative polymerase chain reaction (Q-PCR), which allow for the determination of phylotypes (analogous to species). We discuss the notion of Zetaproteobacteria and/or Epsilonproteobacteria being the most common members of hydrothermal habitats associated with seamounts exhibiting volcanic activity. Another noneruptive seamount scenario is also examined, for example, South Chamorro Seamount, an active forearc serpentinite mud volcano. iNtroductioN Microbial life is remarkable for its resilience to extremes of temperature, pH, and pressure, as well its ability to persist and thrive using an amazing number of organic or inorganic food sources. Nowhere are these traits more evident than in the deep ocean. Much of the deep seafloor consists of cold, relatively static sedimentary environments where heterotrophic microbes exist in significant numbers, but conditions remain relatively unchanged over long periods of time. Mid-ocean ridges associated with diverging plate boundaries are often sites of high-temperature hydrothermal venting and host many chemoautotrophic microbes; however, these systems are more homogeneous with respect to their physical and chemical properties than vent systems associated with volcanic arcs and backarcs (Takai et al., b y d aV i d e M e r s o N a N d c r a i g l . M o y e r common Patterns observed in community structure Th is aticle as been pulished in O ceagraphy, Vlum e 3, N um er 1, a qurterly jornal of Th e o ceagraphy soiety. © 2010 by Th e o ceagraphy soiety. a ll rhts rerved. Perm ision is gnted to cpy his aticle or se in teching nd rearch. replication, sstem m tic repruction, or coective redirbution of ny prtion of his aticle by phocopy m acine, repsting, or oher m eans is perm ited only w ith he aproval of Th e o ceagraphy soiety. send ll corrondence o: ifo@ tos.org or Th e o ceagraphy soiety, Po ox 131, rokville, M d 208-1931, u sa .

M o u N ta i N s i N t h e s e a Microbiology of seaMouNts abstract.Much interest has been generated by the discoveries of biodiversity associated with seamounts.The volcanically active portion of these undersea mountains hosts a remarkably diverse range of unusual microbial habitats, from black smokers rich in sulfur to cooler, diffuse, iron-rich hydrothermal vents.As such, seamounts potentially represent hotspots of microbial diversity, yet our understanding of the microbiology of seamounts is still in its infancy.Here, we discuss recent work on the detection of seamount microbial communities and the observation that specific community groups may be indicative of specific geochemical scenarios, such as iron and sulfur cycling.These observations are based on the metabolisms predicted by phylogenetic characteristics exhibited by the dominant populations found within these microbial communities as compared to the closest related isolate found in culture.Therefore, these studies combine the use of both cultivation-dependent and -independent analyses.Cultivation-independent studies were primarily completed using cloning and sequencing techniques targeting small subunit ribosomal gene (SSU rDNA) biomarkers along with similar biomolecular tools like terminal-restriction fragment length polymorphism (T-RFLP) and quantitative polymerase chain reaction (Q-PCR), which allow for the determination of phylotypes (analogous to species).We discuss the notion of Zetaproteobacteria and/or Epsilonproteobacteria being the most common members of hydrothermal habitats associated with seamounts exhibiting volcanic activity.Another noneruptive seamount scenario is also examined, for example, South Chamorro Seamount, an active forearc serpentinite mud volcano.

iNtroductioN
Microbial life is remarkable for its resilience to extremes of temperature, pH, and pressure, as well its ability to persist and thrive using an amazing number of organic or inorganic food sources.
Nowhere are these traits more evident vents are the extension of the thermal limit for life (Takai et al., 2008) as well as the discovery of a host of bacteria and archaea with novel anaerobic and aerobic metabolisms that take advantage of unique suites of energy sources (both autotrophic and heterotrophic) derived from the combination of geochemistry, temperature, and pressure at these sites (Karl, 1995;Nakagawa and Takai, 2008).
Another emerging aspect of microbiology relevant to seamount studies is the capacity for microbes to live in the deep subsurface.The organic and inorganic chemical reactions, as we as the geological processes associated with microbial populations in the deep subsurface biosphere, are interlinked, b little is known of the complexity of thos linkages (Gold, 1992;Whitman et al., 1998).Furthermore, we are only beginning to realize both the phylogenetic diversity of microbes, and the extent of their metabolic activities in these deep biosphere ecosystems.Drilling projects done in deep mines on land have revealed surprising microbial communi ties living kilometers below the surface and taking advantage of energy sources derived from interactions of water with basement rock minerals (White et al., 1998;Chivian et al., 2008).In the ocean deep-sea drilling into old sediments has revealed bacteria and archaea living up to hundreds of meters below the seafloor (Parkes et al., 2000;Jørgensen and Boetius, 2007).et al., 2004).These fluids may also feed both deep subsurface and/or benthic (bottom-dwelling) microbial communities (Cowen et al., 2003;Huber et al., 2006).Finally, even dormant seamounts that rise to appreciable depths above the ocean floor (i.e., > 1000 m) can impact ocean currents and mixing, potentially resulting in stimulation of microbial communities that feed off the enriched nutrients (see Lavelle and Mohn, 2010).
This latter possibility applies to active seamounts as well.
There are estimated to be over 100,000 seamounts that reach a kilometer or more above the seafloor (see Wessel et al., 2010).This large number illustrates the potential scope of seamounts to provide and impact habitats for marine microbes.The majority of seamounts are associated with tectonic processes taking place in Earth's crust, for example, island arcs that have chains of undersea mountains that result from converging plates.
Another source of seamounts is undersea siziNg uP seaMouNt Microbiology with a Microbial tour of seaMouNts Seawater circulation through the oceanic crust aided by seamount-supported permeable pathways is conducive to microbiological activity because of the large amounts of nutrient-rich fluids that flow through these relatively small volumes of permeable crust (Edwards et al., 2005;Staudigel et al., 2008).Microbial biofilms may develop on fresh basalts, or dense microbial mats (sometimes a meter or more thick) may form in close proximity to active hydrothermal venting at young seamounts (Emerson and Moyer, 2002;Edwards et al., 2005;Templeton et al., 2009).Thus, active submarine-arc volcanoes typify hydrothermal systems that provide abundant nutrient-rich fluids on which microbial communities can live and grow (Figure 1).However, with respect to the distribution of figure 1. cross section of a typical submarine arc volcano with an active hydrothermal system.as magma ascends, water and gases contained in magma exsolve and enter the deep hydrothermal system as pressure is released.cold seawater (blue arrows) permeates the volcano and becomes heated by the hot rock near the magma body.hybrid hydrothermal fluids (red arrows), derived from seawater and magmatic gas and fluid, buoyantly rise to the summit while altering the volcanic rock (white = alteration).fluids discharged at the seafloor rise tens to hundreds of meters before they spread laterally in a hydrothermal plume.local currents disperse the plume, which carries various dissolved and particulate chemical species derived from the magma and the volcanic rock.Courtesy of Cornel de Ronde, GNS Science microbes that thrive at seamounts are emerging.We take a brief tour of the seamounts that have been studied, all in the Pacific Ocean, to illustrate the diversity of habitats that exist and the diversity of microbes that live in those habitats (Figure 2).with multiple actively venting sites (Duennebier et al., 1997).The formation of luxuriant Fe-rich microbial mats, along with the conspicuous absence of large benthic macrofauna, is most likely due to the relatively high concentrations of dissolved CO 2 (300 mM) and enriched levels of ferrous iron (50 to 750 µM) in the vent effluent (Wheat et al., 2000).

lō`ihi seamount
In addition, the summit (presently at a depth of 956 m) is in the oxygen minimum zone, so seawater O 2 concen trations are only 10-15% of saturated seawater (Glazer and Rouxel, 2009).
The initial microbiology study  Mottl et al., 2003 at Lō`ihi, which used a cultivationindependent approach, was conducted prior to the 1996 eruption and showed that the class Epsilonproteobacteria (PVB OTU 2) dominated the surrounding microbial mats and that members of the newly discovered class Zetaproteobacteria (PVB OTU 4) were also present (Moyer et al., 1995) (Emerson and Moyer, 2002;Emerson et al., 2007).This conclusion was based on the isolation of a novel genus and species, Mariprofundus ferrooxydans, from an iron-rich mat at Lō`ihi.M. ferrooxydans is a chemoautotrophic, neutrophilic Fe-oxidizing bacterium that gains its energy for growth by oxidizing ferrous (Fe 2+ ) to ferric (Fe 3+ ) iron, and fixes CO 2 .Furthermore, as it grows, it produces a stalklike structure composed primarily of Fe-oxyhydroxides, the waste product of the Fe-oxidation reaction that remains behind after the cells are gone (Emerson et al., 2007).These stalklike structures are a common feature of many of the microbial mats at Lō`ihi, and are an easily identified diagnostic indicator that Fe-oxidizing bacteria are present (Figure 3).More recently, it has been shown that Zetaproteobacteria dominate vent-site colonies in and around Lō`ihi, though colonization by Epsilonproteobacteria is also possible, most notably at highertemperature (> 40°C) vents, demonstrating both spatial and temporal heterogeneity among these microbial mat communities (Rassa et al., 2009).(T-RFLP) has revealed that there appear to be two distinct community types, Lō`ihi Groups 1 and 2; the iron-based Group 1 is essentially dominated by Zetaproteobacteria and the sulfur-based Group 2 by Epsilonproteobacteria (Figure 4).Their distributions seem to be linked to both geochemistry and temperature, with Group 1 common at lower-temperature vents (median temperature = 23°C) where there are fewer overall community members and Group 2 found at higher-temperature vents (median temperature = 77°C) exhibiting a higher diversity of detectable populations.We have correlated these results with the Group 1 communities associated with vents that have higher levels of dissolved Fe 2+ and Group 2 communities with both high H 2 S and Fe 2+ concentrations (data not shown).We hypothesize that Group 1 communities are enhanced with members of Fe-oxidizing bacteria (Zetaproteobacteria), and that the Group 2 communities are enhanced with members of S-oxidizing bacteria (Epsilonproteobacteria) but still retain some of the Zetaproteobacteria. of this issue [Chadwick et al., 2010]; Rhodes et al., 1990).This seamount was intensely studied after an eruptive event in 1998 (Embley et al., 1999), and members of the Epsilonproteobacteria were shown to dominate diffuse flow vents (Huber et al., 2003).Microbial samples from other diffuse-flow vent sites in the caldera analyzed by light microscopy, scanning electron microscopy (SEM), and X-ray diffraction (XRD) showed distinct morphologies, for example, iron-encrusted stalks similar to those of Zetaproteobacteria (Kennedy et al., 2003).Multiple microbial mat samples from across Axial's caldera have been examined using T-RFLP with the majority dominated by Epsilonproteobacteria and a few dominated by Zetaproteobacteria (data not shown).Surprisingly, this fits the pattern seen at Lō`ihi Seamount in terms of two distinct groups of bacterial-dominated community types (e.g., Figure 4); however, in this location the Epsilonproteobacteria were acting as the primary colonizers at a majority of vent sites shortly after an eruptive event that occurred in 1998 (Moyer and Engebretson, 2002).(Emerson and Moyer, 2002) figure 4. unweighted Pair group Method with arithmetic mean (uPgMa)/productmoment cluster analysis of 55 terminal-restriction fragment length polymorphism bacterial community fingerprints using eight restriction digest treatments on each sample.two groups of communities were identified and designated "lō`ihi group 1" and "lō`ihi group 2. " colored labels indicate samples taken in the same year.ambient temperature for summit area of lō`ihi seamount is 4.0°c.scale bar is the Pearson product moment correlation r-value X 100.Staudigel et al., 2006).Microbial samples from Vailulu'u have been analyzed using both cultivation-dependent and -independent approaches.Standard clone library analysis with small subunit ribosomal gene (SSU rDNA) biomarkers found that the majority of clones represented Epsilonproteobacteria, though Zetaproteobacteria were also detected (Sudek et al., 2009).In addition, a diverse fungal community, including several species of yeast, was found associated with iron mats as well as with basaltic rock surfaces at Vailulu'u (Connell et al., 2009).Unlike Lō`ihi, the complex circulation patterns created by Vailulu'u's summit crater provide habitat for a much more diverse and abundant macrofaunal community, including shrimp and eels.One result of these flow patterns is the existence of a "moat of death" where fish and other larger animals become entrapped in anoxic water and suffocate, providing a food source for anaerobic bacteria.The specific nutrient dynamics that support the macrofauna, including the role of bacterial primary production, have yet to be fully understood at Vailulu'u.

suiyo seamount
Suiyo Seamount is an active submarine volcano located on the volcanic front of the Izu-Bonin Arc, western Pacific Ocean, near Japan.It has a 200-300-m deep caldera at its summit with many active hydrothermal vents rich in hydrogen sulfide and methane that have recently been accessed through shallow drilling (Ishibashi et al., 2007).
A number of novel thermophilic bacteria and archaea have been isolated from this seamount, including new chemoautotrophic species of bacteria that can use hydrogen as an energy source (Takai et al., 2003;Nakagawa et al., 2003), as well as an archaeon that can grow anaerobically with hydrogen and thiosulfate (Mori et al., 2008).Other microbial studies have included an examination of the hydrothermal plume within the caldera, which identified two dominant phylotypes of putative sulfuroxidizing bacteria, the SUP01 group, belonging to Epsilonproteobacteria, and the SUP05 group, in the class Gammaproteobacteria (Sunamura et al., 2004).These phylotypes were distinct from those common in the surrounding seawater and accounted for up to 66% of the bacteria in the hydrothermal plume.
This large percentage suggests that the hydrodynamics of the vent plume within the caldera create an incubator for these two groups of microbes.Another microbial study at Suiyo used a series of novel catheter-type in situ growth chambers inserted into subsurface boreholes after excavation with a portable submarine driller.In this case, a diverse array of phylotypes from Epsilonproteobacteria were found (Higashi et al., 2004).An extensive examination of borehole and vent fluids detected Gammaproteobacteria (including Thiomicrospira and Alcanivorax) in all samples, while multiple samples also exhibited Epsilonproteobacteria as well as a diverse array of potentially seawaterentrained bacterial phylotypes (Kato et al., 2009).Another study compared the diversity and distribution of integrase and integron genes (generally thought to be responsible for lateral gene transfer between unrelated bacterial groups) at Suiyo Seamount and at the Fryer site (aka Pika site) of the Mariana backarc (Elsaied et al., 2007).Many novel members of these gene families were discovered at both sites, suggestin that gene exchange between different bacteria at these sites could potentially be very common.Interestingly though, the results indicated that the two different vent communities harbored quite different sets of integron genes tha were not necessarily compatible with one another.This observation suggests that lateral gene transfer may take place between populations within a vent site, but that gene transfer between geograph ically separated vent communities woul not happen as readily.This lack of latera gene transfer is in contrast with antibiotic-resistant genes carrying integrons, where disparate environmental populations have been shown to have common integron functions (Elsaied et al., 2007) [Chadwick et al., 2010]; Embley et al., 2007).In a recent microbial study of 25 microbial communities from 18 different hydrothermal systems located at seven Mariana Arc volcanoes, a remarkable diversity of vent ecosystem environments was found, including black smokers (220°C), white smokers rich in CO 2 , vents with high sulfide concentrations, as well as iron-rich vents (Davis and Moyer, 2008).Molecular fingerprinting analysis found that the community composition at each vent was unique, indicating a substantial degree of diversity in the microbial populations that inhabit these ecosystems; however, the majority of the communities examined were showed that these microbial communities contained no more than 2% archaea (Nakagawa et al., 2006).Combined, both of these microbial studies show a high degree of phylogenetic diversity as well as putative metabolic potential across the Epsilonproteobacteria.These data also fit the general pattern that either Epsilonproteobacteria or Zetaproteobacteria dominate the majority of seamount vent microbial communities.Figure 5  with Fe-oxyhydroxide mineral crusts (Stoffers et al., 2006).Samples from both locations were composed primarily of two-line ferrihydrite, and transmission electron microscopy (TEM) revealed structures similar to those formed by M. ferrooxydans, supporting the assumption that these are biogenic iron oxides (Langley et al., 2009).Molecular microbial analysis of similar samples from both Volcanoes 1 and 19 detected the presence of multiple Zetaproteobacteria phylotypes, representing a high degree of diversity within this group.In addition, these Fe-oxidizing bacteria were most abundant at Volcano 1, where sediments were richer in Fe and contained more crystalline forms of Fe oxides (Forget et al., in press).

Kermadec arc seamounts
The Kermadec-Tonga Arcs, together the longest submarine arc system on the planet, contain a series of recently explored seamounts (deRonde et al., 2001(deRonde et al., , 2007)).Microbial samples have been collected at Brothers Volcano, from the caldera vent field, and at the lower cone site.At the caldera vent field, the exteriors of high-temperature (up to 290°C) black smoker chimneys were dominated by variable communities of bacteria, including a majority of putative sulfur-oxidizing Epsilonproteobacteria and Gammaproteobacteria (Takai et al., 2009).A single piece of dacite lava covered with elemental sulfur and white microbial mats was collected at the lower cone site from 67°C vent effluent and analyzed using standard molecular microbial techniques.In this study, the most abundant phylotypes detected were from the class Deltaproteobacteria (most closely related to sulfatereducing bacteria).However, 5% of this clone library was represented by Epsilonproteobacteria phylotypes (Stott et al., 2008).Another recent microbial study from two more seamounts (Clark and Tangaroa) on the southern Kermadec Arc, revealed extensive zones of flocculent iron oxides on the order of tens to hundreds of square meters.
Morphologically, these microbial mats resembled the biogenic oxides found at Lō`ihi and elsewhere.Phylotypes of the Zetaproteobacteria were detected in each of the three mats examined; however, the clone libraries were dominated (67% of the total number of clones) by members of the Epsilonproteobacteria.

T-RFLP fingerprinting analysis of five
Fe-floc and two sediment samples showed that a majority of these samples contained detectable Zetaproteobacteria peaks.These findings from the southern Kermadec Arc suggest that we may not fully understand the story of iron-oxidizing bacterial communities associated with seamounts.Either there may be a group of as-of-yet uncultured Fe-oxidizers that are Epsilonproteobacteria (facilitating the formation of Fe-floc microbial mats), or perhaps the Fe-oxidizing Zetaproteobacteria are only active in rapidly accreting mats (Hodges and Olsen, 2009).

south chamorro and other Mariana forearc seamounts
The Mariana forearc is comprised of a series of serpentine mud volcanoes formed between the Mariana Trench and the volcanic arc that lies to its west (Fryer et al., 1999).Dehydration  (Fryer et al., 2006).In addition, springs often form (as a byproduct of the upwelling fluid, attributed to the sediment compaction and dehydration process) and deliver water to the seafloor that is typically fresher than seawater and highly alkaline, with pH up to 12.5 (see Spotlight 9 on page 174 of this issue [Wheat et al., 2010]; Mottl et al., 2003Mottl et al., , 2004)).
South Chamorro Seamount has active summit springs (Figure 6).This area was the focus of an Ocean Drilling Program (ODP) investigation (Leg 195, Site 1200) into the structure and diversity of the microbial populations existing near the seamount surface and at depth below the seafloor.In general, these highly alkaline fluids also contain elevated levels of dissolved sulfide with nearly no sulfate, which is attributed to both anaerobic methane oxidation (AMO) and sulfate reduction in a microbial habitat overwhelmingly dominated by archaea (Mottl et al., 2003) provides an excellent opportunity for microbial investigation of phylogenetic diversity and ecological significance in an extreme subsurface environment representing the upper pH limit for life (Takai et al., 2005).(Curtis et al., 2009).

eMergiNg PatterNs regardiNg coMMuNity structure
There is growing evidence, as summarized in Table 1, that Zetaproteobacteria and/or Epsilonproteobacteria are the most common members of surface hydrothermal habitats associated with volcanically active seamounts.
Somewhat surprisingly, archaea tend not to be abundant in the surface habitats sampled at seamounts, although they are far more abundant in the interior of high-temperature chimneys and anaerobic subsurface fluids at volcanically active seamounts (Huber et al., 2002;Takai et al., 2009).A notable exception are the mud volcanoes from Mariana forearc, which represent some of the most extremely alkaline habitats on Earth and appear to be dominated by unusual archaea (Curtis et al., 2009).Overall, microbial community patterns must reflect the underlying geochemistry that is driven by seamount geologic processes; for example, neither Epsilonproteobacteria or Zetaproteobacteria are common members of the seawater surrounding seamounts.
As mentioned earlier, the Epsilonproteobacteria are comprised of a large breadth and depth of phylogenetic diversity.Significant effort has gone into cultivating representatives for each of the various groups that have been described using cultivationindependent approaches based on the analysis of SSU rDNA clone libraries.
A recent review finds that all cultures characterized as hydrothermal-ventassociated Epsilonproteobacteria either oxidize reduced sulfur or reduce sulfur while oxidizing hydrogen or formate.
Although most are autotrophs, they can also be mixotrophs and occasionally heterotrophs (Campbell et al., 2006).

The emerging discovery of
Zetaproteobacteria associated with diffuse vents that are rich in available Fe 2+ is intriguing.On the basis of morphology, it has been recognized sinc the 1980s that conspicuous iron-oxideencrusted stalks, presumed to be of biogenic origin, were commonly found in flocculent iron oxyhydroxide deposit collected from modern and ancient hydrothermal sites (Juniper and Fouque 1988).It was the isolation of M. ferrooxy dans and analysis of its SSU rDNA (as well as other evolutionary conserved genes) that suggested it represented a novel class of Proteobacteria (Emerson et al., 2007).While this group of microbes was initially thought to be ver rare, follow-up studies are suggesting th Zetaproteobacteria can be quite abundant at seamount habitats rich in iron.
The fact that M. ferrooxydans is in cultu " …seaMouNts PoteNtially rePreseNt hotsPots of Microbial diVersity, yet our uNderstaNdiNg of the Microbiology of seaMouNts is still iN its iNfaNcy.

"
does not mean that we have described all the important Zetaproteobacteria.
Indeed, it appears that M. ferrooxydans may represent a lineage that is relatively minor at most sites (Figure 3A).It is also quite possible, perhaps even likely, that not all Zetaproteobacteria are Fe-oxidizers, but that some might use other substrates for growth.

oPPortuNities for Microbial research at seaMouNts abouNd
In general, our understanding of the forces that shape microbial community structure within a given environment are in a state of flux, be it the human gut, salt marsh sediment, or the ocean's water column.The same can be said even more emphatically of seamounts, because they are poorly studied compared to these other environments.
Nonetheless, seamounts have the potential to make good natural laboratories for studying basic problems having to do with microbial ecology and evolution.
Because they tend to be geographically isolated from one another, there may be more of a barrier for movement of microorganisms between seamounts than is the case in the more contiguous environments of a mid-ocean ridge.
than in the deep ocean.Much of the deep seafloor consists of cold, relatively static sedimentary environments where heterotrophic microbes exist in significant numbers, but conditions remain relatively unchanged over long periods of time.Mid-ocean ridges associated with diverging plate boundaries are often sites of high-temperature hydrothermal venting and host many chemoautotrophic microbes; however, these systems are more homogeneous with respect to their physical and chemical properties than vent systems associated with volcanic arcs and backarcs (Takai et al., b y d aV i d e M e r s o N a N d c r a i g l .M oy e r common Patterns observed in community structure volcanism driven by hotspots or mantle plumes within Earth's crust (see Staudigel and Clague, 2010).The Hawaiian Islands in the Pacific Ocean and the Azores in the Atlantic Ocean are spectacular examples of hotspots.Because our knowledge about the role of seamounts as biogeochemical foci for microbial activity is in its infancy, it is instructive to consider another seamount-associated microbial habitat that derives from fundamentally different geological processes.The importance of hydrothermal vents associated with deep ocean crustal spreading centers as havens for microbial activity has fundamentally changed how we view life on Earth, and enlightened our thinking about other places in our solar system where life might also exist.The iconic features of these sites are black smokers billowing superheated water charged with iron sulfide minerals, as well as dense communities of macrofauna largely fueled by symbiotic relationships with bacteria that provide energy for their hosts using sulfide-rich waters emanating from the vents.The discovery of these deep-sea ecosystems has reshaped our thinking and research agenda regarding extreme oases of life in the deep sea.Some of the most notable findings from over three decades of research on hydrothermal Lō`ihi is a seismically active submarine hotspot volcano and the youngest seamount in the Hawaiian Island chain; it is located 35 km southeast of the big island of Hawai`i with its summit rising nearly 4 km above the seafloor (see Spotlight 3 on page 72 of this issue[Staudigel et al., 2010]).Karl et al. (1988) made the first discovery of seamount-associated, Fe-rich microbial mats, showing bacteria with a dominant sheathlike morphology occurring near the summit at Pele's Vents.An eruption at Lō`ihi in 1996 led to the destruction of Pele's Vents and the formation of Pele's Pit, a 300-m-deep pit crater figure 2. Pacific ocean bathymetric map showing the locations where seamounts have been microbiologically investigated.These include (north to south): axial seamount, suiyo seamount, lō`ihi seamount, the Mariana arc (n = 8), south chamorro seamount and Mariana forearc (n = 7), Vailulu'u seamount, the tonga arc (n = 2), and the Kermadec arc (n=3).Image reproduced with permission from the GEBCO world map (http://www.gebco.net/) axial seamount Although Axial Seamount is located along the central portion of the Juan de Fuca Ridge (a mid-ocean ridge spreading center), it also lies at the eastern end of the Cobb-Eickelberg seamount chain.Axial originated at the Cobb hotspot, which generates lavas without a typical deep mantle chemical signature (see Spotlight 1 on page 38 figure 3. Morphotypes of fe-oxidizing bacteria commonly observed at seamounts.(a) a light micrograph of Mariprofundus ferrooxydans, showing the twisted iron oxyhydroxide stalk with two cells, denoted by arrows, growing at the ends.(b) a light micrograph of an unknown organism that forms tubular iron oxide structures; the fluorescent dye image shows two cells present at the apical end of the structure, which is formed as the cells grow.(c) a transmission electron microscopy image of the same organism as shown in b; note that each time the cell divides the structure bifurcates.(d) a sheath-forming fe-oxidizing bacterium; most of the sheaths are empty tubes of iron oxyhydroxides formed by filaments of cells that grow within them.(D) is reproduced with permission(Emerson and Moyer, 2002) photosynthetic encrusting red algae and coral at Mat City near the summit of Ea Diamante (at a depth of only 200 m), an sampled white flocculent microbial mat next to white smoker chimneys venting liquid CO 2 droplets at Champagne Ven on Northwest Eifuku (see Spotlight 1 on page 38 of this issue[Chadwick et al., Mariana Arc study, dense microbial mats were observed covering white-smoker sulfur chimneys in TOTO caldera, of Paton-Masala Seamount, with temperatures up to 170°C.Samples collected from the chimney surface and interior, and from an in situ colonization system indicated that the microbial communities were dominated by Epsilonproteobacteria that were either hydrogen and/ or sulfur oxidizers or anaerobic hydrogen-oxidizing, sulfur-reducing chemoautotrophs.This study also and 5B) or those dominated by Epsilonproteobacteria (Figure 5C and 5D).Overall, the Mariana Arc and backarc support ecosystems that represent a bacterial biodiversity hotspot are thought to result from the complex geochemistry of these hydrothermal systems driven by the heterogeneity of the subducting source material coupled with the enhanced volcanic activity of the region (Davis and Moyer, 2008).tonga arc seamounts Manned submersible dives along the Tonga Arc revealed two shallowwater volcanoes.Volcano 19 exhibited relatively high-temperature venting (245-265°C) in a depth range of 385-540 m and in some cases chimneys that had a distinctive "flamelike" discharge caused by an instantaneous phase separation upon the release of vent fluids.Volcano 1 exhibited more widespread diffuse venting (30-70°C), though both had extensive areas with filamentous microbial mats associated

figure 5 .
figure 5. Photographs of microbial mats collected by remotely operated vehicle.(a) iron-oxide-encrusted microbial mat at yellow top Vent, Northwest eifuku (b) small chimneys and iron oxide microbial mats inside Pele's Pit crater at upper hiolo Vents, lō`ihi.(c) white smoker vents covered with microbial mats at champagne Vents, Northwest eifuku.(d) white microbial mats covering rocks at iceberg Vent, Northwest rota-1.
figure 6. summit springs at south chamorro seamount showing bathymodiolid mussels present.Photo taken from submersible Shinkai 6500 view port.Courtesy of Patty Fryer, University of Hawai`i

A
recent molecular microbial study revealed new populations of archaea that exist at the seafloor in serpentinized muds from South Chamorro Seamount.Archaeal communities from seven Mariana forearc seamounts were assessed with T-RFLP, while two sites atop South Chamorro were also investigated using clone library analysis.Though a bacterium (Marinobacter alkaliphilus) was previously isolated from ODP Site 1200 seafloor muds (Takai et al., 2005), no bacteria were detected using PCR-based molecular methods.Novel Crenarcheaota phylotypes, from within Marine Group I and Marine Benthic Group B, as well as Euryarcheaota phylotypes from the Methanobacteria and Methanosarcinales, were detected.T-RFLP fingerprinting revealed that similar archaeal communities from Big Blue and Blue Moon seamounts were nearly identical to the South Chamorro ODP Hole 1200E site, which was hypothesized to correlate with active summit spring habitats figure 7. Maximum-likelihood phylogenetic tree showing the evolutionary placement of phylotypes belonging to Proteobacteria with emphasis on zetaproteobacteria and epsilonproteobacteria.only bootstrap values above 50 are shown.scale bar represents five nucleotide substitutions per 100 positions.����� Thus, seamounts could represent repositories for studying the biogeography of microbial populations.In addition, many other basic questions remain:• Are there fundamental differences in microbial communities at seamounts where hydrothermal processes prevail compared to mid-ocean ridges?Is iron-oxidation a more important process at seamounts than at ridgeaxis spreading centers?• Are seamounts windows into a subsurface world?How important are subseafloor circulation paths for the redistribution of microbes?Can microbes themselves influence these patterns by altering the subsurface flow regimes?• To what extent do water-column hydrodynamics associated with seamounts influence microbial community structure and dynamics through their influence on ocean dynamics and changes in nutrient distribution patterns?• How important are seamounts to global marine geomicrobiology and what is their role in altering the crust and ocean chemistry, including iron and sulfur cycling in the ocean?How might seamounts affect ocean acidification and the global carbon cycle?These types of questions illustrate both how limited our understanding is of undersea microbial ecosystems, and also how much potential remains for significant new discoveries in marine microbiology.In addition to contributing new knowledge about fundamental processes at the ocean floor and in the subsurface, it is possible that novel compounds, enzymes, or organisms could be discovered from seamounts that would have practical applications in the biotechnology, bioenergy, or bioremediation sectors.It is even possible that microbes from seamounts could provide important clues about the origin of life on Earth and its potential to exist in other habitable zones in the solar system acKNowledgMeNts .