1995 2000 2005 2010 0.5 1.0 1.5 2.0 2.5 Year BP:PP ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 2000 2005 2010 10 15 20 25 30 35 40 45 Year DOC g m -2 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 2000 2005 2010 10 15 20 25 30 35 40 45 Year DOC g m -2 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Lake Fryxell West Lobe Bonney 2004 2006 2008 2010 0.00 0.05 0.10 0.15 Year BP:PP ● ● ● ● ● ● ● ● ● 1995 2000 2005 2010 0.05 0.10 0.15 0.20 0.25 0.30 Year BP:PP ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 2004 2006 2008 2010 50 52 54 56 58 60 62 Year DOC g m -2 ● ● ● ● ● ● ● ● North Palmer Region 0 2 4 6 Year PP g C m -2 d -1 2004 2006 2008 2010 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 10 20 30 40 50 60 Year PP mg C m -2 d -1 1995 2000 2005 2010 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 5 10 15 20 25 30 35 Year PP mg C m -2 d -1 1995 2000 2005 2010 Assessing microbial ecosystem function across two polar extremes: The Palmer (PAL) and McMurdo Dry Valley (MCM) LTERs * 1 Jeff S. Bowman, 2 Trista J. Vick-Majors, 3 Rachael Morgan-Kiss, 4 Christina Takacs-Vesbach, 1 Hugh W. Ducklow, 2 John C. Priscu 1 Lamont-Doherty Earth Observatory, 2 Montana State University, 3 Miami University, 4 University of New Mexico *[email protected] | www.polarmicrobes.org 2015 LTER ASM, Estes Park, CO Our hypothesis is that polar desert lakes are different from the coastal Antarctic marine ecosystem. As obvi- ous as the answer seems - one need only to consider the size of members of the top trophic levels to appre- ciate the huge ecological differences between these environments - making this comparison allows us to identify both common and unique ecological features of these sites associated with fundamental processes that might otherwise be overlooked. The 20+ year record of key ecosystem parameters at PAL and MCM provides a further opportunity to explore how these ecosystems respond to common events, such as the unusually warm austral summers in 2001-2002 and 2008-2009. To make our comparison we considered: 1. Records of bacterial production (BP), primary pro- duction (PP), and dissolved organic carbon (DOC). 2. Recent observations of microbial community structure. 3. Metabolic inference-based predictions of microbial metabolic potential. Function Trophic Level PAL MCM Pelagibacter Actinobacteria Whale/seal Rotifer/tardigrade The difference in the size of top predators at PAL (humpback whale, left) and MCM (rotifer, right) underscores major differences in ecosys- tem function. Such obvious differences, however, may mask function- al similarities that appear as we move toward basal trophic levels (left). Actinobacteria and Pelagibacter, the dominant bacteria at MCM and PAL respectively, for example, are both oligotrophic specialists with large functional overlaps. 1 m ELB WLB FRX NPAL SPAL Site/Region BP:PP 0.001 0.01 0.1 1 10 1 10 100 1000 10000 PP mg C m -2 day -1 BP mg C m -2 day -1 1 10 100 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● PP mg C m -3 day -1 0.001 0.01 0.1 1 0.01 0.1 1 10 BP mg C m -3 day -1 Fryxell East Lobe Bonney West Lobe Bonney North Palmer South Palmer 1:10 1:5 + + + + + The ratio of BP to PP provides insight into the functioning of the microbial loop. Values in the global pelagic marine environment tend to center around 1:10 (grey dotted lines). At a ratio of 1:5 (black dotted lines) PP is thought to pro- vide insufficient C to support both BP and respiration. Autochthonously fixed carbon must be subsidized by a source outside the photic zone. The three MCM lakes included in this investigation have BP:PP ratios exceeding PAL, despite their lower DOC concentrations. Lake Fryxell has extreme values of BP:PP suggesting a large subsidy from DOC-rich water below the photic zone. The relationship between BP and PP Temporal trends The concentration of DOC and rates of BP and PP change with time. Interestingly, while the concentration of DOC has increased at PAL in recent years, concentrations have decreased since 2000 in the photic zone of Lake Fryxell and Lake Bonney. Some of the dynamics in BP:PP appear to be driven by extreme events. The summer of 2001-2002 was unusually warm and windy, leading to increased glacial melt and heightened lake levels at MCM. This influx of lake water may have suppressed PP due to increased turbidity, while stimulating BP with a new source of labile DOC. Nutrients carried by the meltwater enhanced PP in subsequent years after turbidity re- turned to normal. The summer of 2008-2009 was also unusually warm and had a major impact at both PAL and MCM, with BP:PP de- creasing at both sites as a result of elevated PP. This mutual uncoupling of BP and PP may have been driven by en- hanced krill grazing at PAL and increased particle export at MCM. Diverging community structure and converging function fall_frx_9_b.1 fall_frx_9_b.2 summer_frx_6_b.1 summer_frx_6_b.2 fall_frx_6_b.1 fall_frx_6_b.2 summer_wlb_13_b.2 summer_frx_9_b.2 summer_frx_9_b.1 fall_wlb_13_b.2 summer_wlb_13_b.1 fall_wlb_13_b.1 summer_wlb_18_b.2 summer_wlb_18_b.1 fall_wlb_18_b.2 fall_wlb_18_b.1 summer_nw_shallow_b.2 summer_nw_shallow_b.1 summer_sw_deep_b.1 summer_sw_deep_b.2 winter_ne_shallow_b.2 winter_ne_shallow_b.1 summer_ne_shallow_b.1 summer_ne_shallow_b.2 summer_sw_shallow_b.1 summer_se_shallow_b.1 summer_se_shallow_b.2 summer_sw_shallow_b.2 summer_ne_deep_b.1 summer_se_deep_b.1 summer_se_deep_b.2 summer_nw_deep_b.2 summer_nw_deep_b.1 Candidatus Pelagibacter ubique HTCC1062 Tropheryma whipplei Acidothermus cellulolyticus 11B Actinobacteria Syntrophomonas wolfei Goettingen Owenweeksia hongkongensis DSM 17368 Polaribacter MED152 Francisella Alcanivorax Pelagibacter Candidatus Amoebophilus asiaticus 5a2 Cytophagia Acidimicrobidae bacterium YM16 304 Hyphomonas neptunium ATCC 15444 Candidatus Cardinium hertigii Thermodesulfobium narugense DSM 14796 Caldisericum exile AZM16c01 Parvibaculum lavamentivorans DS 1 Clavibacter michiganensis nebraskensis NCPP Polaromonas JS666 Fluviicola taffensis DSM 16823 Clavibacter michiganensis NCPPB 382 Polaromonas naphthalenivorans CJ2 Candidatus Pelagibacter IMCC9063 Octadecabacter Muricauda ruestringensis DSM 13258 Glaciecola nitratireducens FR1064 Teredinibacter turnerae T7901 Robiginitalea biformata HTCC2501 alpha proteobacterium IMCC1322 665 0 summer_wlb_13_b.2 fall_wlb_18_b.1 fall_wlb_18_b.2 summer_wlb_18_b.2 summer_wlb_18_b.1 fall_wlb_13_b.2 summer_wlb_13_b.1 fall_wlb_13_b.1 fall_frx_9_b.1 fall_frx_6_b.1 fall_frx_6_b.2 summer_frx_9_b.2 summer_frx_9_b.1 fall_frx_9_b.2 summer_frx_6_b.2 summer_frx_6_b.1 summer_nw_shallow_b.2 summer_nw_shallow_b.1 summer_ne_shallow_b.1 summer_ne_shallow_b.2 summer_sw_shallow_b.1 summer_se_shallow_b.2 summer_se_shallow_b.1 winter_ne_shallow_b.1 winter_ne_shallow_b.2 summer_ne_deep_b.1 summer_sw_deep_b.1 summer_sw_deep_b.2 summer_sw_shallow_b.2 summer_se_deep_b.2 summer_se_deep_b.1 summer_nw_deep_b.1 summer_nw_deep_b.2 formate oxidation to CO2 IRUPDOGHK\GH R[LGDWLRQ ,, JOXWDWKLRQHíGHSHQGHQW pyruvate fermentation to acetone QLWUDWH UHGXFWLRQ , GHQLWULILFDWLRQ pyruvate fermentation to lactate VXOILWH R[LGDWLRQ , VXOILWH R[LGRUHGXFWDVH IRUPDOGHK\GH R[LGDWLRQ ,9 WKLROíLQGHSHQGHQW formaldehyde oxidation I IRUPDOGHK\GH DVVLPLODWLRQ ,, 5X03 &\FOH pyruvate fermentation to ethanol I pyruvate fermentation to acetate II VXOIDWH UHGXFWLRQ ,9 GLVVLPLODWRU\ methanol oxidation to formaldehyde II DPPRQLD R[LGDWLRQ , DHURELF GLPHWK\O VXOILGH GHJUDGDWLRQ ,, R[LGDWLRQ VXOILGH R[LGDWLRQ ,, VXOILGH GHK\GURJHQDVH methanol oxidation to formaldehyde I glycerol degradation III formate to dimethyl sulfoxide electron transfer QLWUDWH UHGXFWLRQ 9,, GHQLWULILFDWLRQ Bifidobacterium shunt K\GURJHQ R[LGDWLRQ ,,, DQDHURELF 1$'3 NADH to dimethyl sulfoxide electron transfer IRUPDWH WR WULPHWK\ODPLQH 1íR[LGH HOHFWURQ WUDQVIHU nitrite oxidation sulfur reduction I QLWUDWH UHGXFWLRQ ,,, GLVVLPLODWRU\ hydrogen production V mixed acid fermentation DUVHQLWH R[LGDWLRQ ,, UHVSLUDWRU\ pyruvate fermentation to ethanol II sulfite oxidation IV sulfite oxidation III pyruvate fermentation to ethanol III K\GURJHQ R[LGDWLRQ ,, DHURELF 1$' hydrogen production II hydrogen production VI VXOIDWH UHGXFWLRQ 9 GLVVLPLODWRU\ QLWUDWH UHGXFWLRQ ,9 GLVVLPLODWRU\ hydrogen production III reductive monocarboxylic acid cycle IRUPDOGHK\GH DVVLPLODWLRQ , VHULQH SDWKZD\ K\GURJHQ R[LGDWLRQ , DHURELF VXOILGH R[LGDWLRQ , VXOILGHíTXLQRQH UHGXFWDVH &DOYLQí%HQVRQí%DVVKDP F\FOH VXOIXU UHGXFWLRQ ,, YLD SRO\VXOILGH 784 0 180 0 90 W 90 E PAL MCM We used PAPRICA to conduct a metabolic inference, matching 16S rRNA gene reads from PAL and MCM with the closest related completed genomes and associated metabolic pathways. Despite the marine origin of the MCM Lakes, the composition of the water column microbial community is distinct from the microbial community at PAL. In particular the oligotrophic specialists best represented by the complete genomes of Tropheryma and Pe- lagibacter are phylogenetically very distant although they may occupy a similar niche. These clades are non-mo- tile opportunists, and are likely to posses alternate energy acquisition strategies such as proteorhodopsins. Taxa that may be associated with particles, such as Polaribacter MED152 and Syntrophomonas Wolfei Goettingen are shared between these environments, as are their associated metabolisms. This suggests that microniches, such as regions of low oxygen within particles, may be more important in determining community function than large-scale environmental factors (such as a terrestrial or marine location). Community composition as determined by phylogenetic placement Key metabolic pathways predicted by metabolic inference 1. While phytoplankton directly provide the carbon for BP at PAL, this is demphasized at MCM. 1a. Phytoplankton derived carbon at MCM is highly subsidized by allochthonous sources. 1b. PP is inherently decoupled from BP at MCM, particularly in Lake Fryxell. 1c. DOC is produced below the chemocline in Lake Fryxell, and diffuses across the chemocline to support BP. 2. Differences in water column structure may drive some of the differences in carbon utilization, with the shallow chemo- cline limiting particle degradation in the photic zone of MCM lakes. 3. Differences in trophic structure may also account for some differences, with biomass being disproportionately chan- neled to krill and the higher trophic levels at PAL in some years. 4. As a result of these differences PAL and MCM can have different responses to major perturbations, although sometimes the response is the same - but for very different reasons! 5. Although the composition of the microbial communities diverge sharply, ecological similarities at the microbial level allow for strong functional similarities. Conclusions View poster online Email presenting author 50 μm Image Wei Li Image JSB