Effects of ocean acidification on Antarctic marine microbes
Three experiments were performed at Davis Station, East Antarctica 68 degrees 35' S 77 degrees 58' E, to determine the effects of ocean acidification on natural assemblages of Antarctica marine microbes (bacteria, viruses, phytoplankton and protozoa). Incubation tanks (minicosms) were fill...
| Other Authors: | , |
|---|---|
| Format: | Dataset |
| Language: | unknown |
| Published: |
Australian Ocean Data Network
|
| Subjects: | |
| Online Access: | https://researchdata.ands.org.au/effects-ocean-acidification-marine-microbes/687212 https://data.aad.gov.au/metadata/records/TNE_8A_acidification_microbes http://data.aad.gov.au/aadc/portal/download_file.cfm?file_id=3285 https://secure3.aad.gov.au/proms/public/projects/report_project_public.cfm?project_no=40 http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=TNE_8A_acidification_microbes |
| Summary: | Three experiments were performed at Davis Station, East Antarctica 68 degrees 35' S 77 degrees 58' E, to determine the effects of ocean acidification on natural assemblages of Antarctica marine microbes (bacteria, viruses, phytoplankton and protozoa). Incubation tanks (minicosms) were filled on the 30/12/08, 20/01/09 and 09/02/09 and the microbial communities incubated for 10, 12 and 10 days, respectively. Minicosm System The minicosm system comprised a modified 21ft + refrigerated shipping container (reefer) that housed 6 x 650 l incubation tanks. The temperature of the minicosms was maintained at ambient +/- 0.2 degrees C. Each minicosm was illuminated on a 19 h light: 5 h dark cycle using 2, 150W HQI-TS metal halide lamps (Osram) the emission of which was screened using 1/4 CT blue (Arri). This gave an average PAR irradiance in minicosm of 16.97 M quanta.m2.d-1, approximating 50% of the daily clear-sky PAR irradiance at 5m depth at Davis Station (Davidson unpublished data). The contents of each tank was gently mixed by a shrouded auger that rotated at 15 rpm. Minicosm tanks were gas-tight allowing manipulation of pCO2 concentration and monitoring of O2 in the tanks with ambient, 1, 3, and 4 x pCO2 using a 4-channel optode and optical fibre array. (PreSense). Chemical Measurements pCO2 measurements Alkalinity: Alkalinity was measured using samples that were equilibrated to 25 +/- 0.2 degrees C (greater than 30 mins) in a temperature controlled water bath (Ratek). The alkalinity of the sea water was determined by titration against a titrant that was approximately isosmotic with sea water (33.0 ppt) and contained a known molarity of HCl (~0.6M), as determined by titration against a known concentration of Na2CO3. The titration was performed using a Metrohm 809 Titrando and single 800 Dosino auto-titrator. The pH probe was calibrated in millivolts to pHs 4.010 (Hanna high accuracy) and 7.000 (Hanna high accuracy) that were equilibrated to 25 +/- 0.2 degrees C as above. An aliquot of 99.71 mL of sea water at 25 plus or minus 0.2 degrees C (equilibrated as above) was then transferred to the titration vessel and the pH of the seawater adjusted to a potential of reached 170mV (~3.5 pH). The sample was then mixed and aerated for 360 seconds while the sample was bubbled to remove CO2. Titrant additions were made (dosing rate 0.56 mL/min) until the potential of the sample reached 215 mV (~ 3.0 pH). Millivolt measurements were converted to pH units and any drift in calibration was accounted for using the pH calibrations at the beginning and end of the sample analyses. pH was then regressed against the titration burette volume and the linear regression (r2 greater than 0.9998) then used to calculate alkalinity. Measurements of seawater alkalinity performed on standards were within 0.7% of measurements performed by Dickson et al. (S% = 33.390% and an AT = 2215.8 micro-mol/kg) or CSIRO (S% = 34.37% and an AT = 2280.8 micro-mol/kg). pH: The pH meter was calibrated using primary (A. Dickson, USA) or secondary (CSIRO, marine laboratories) standards for seawater alkalinity bubbled with either air or pure CO2 for greater than 30 minutes at ambient +/- 2 degrees C. The pH for seawater saturated with air and CO2 at the known alkalinity and salinity and measured temperature was calculated using an Excel macro based on "CO2SYS.BAS" of Lewis and Wallace (1998). The temperature was measured using temperature probes that were calibrated using a Guildline 9540, 3 decimal place, platinum resistance thermometer and atmospheric pressure (Davis meteorology). The pH meter was then manually calibrated to these calculated values using the air and CO2 saturated seawater, giving a regression of pH versus mV for the probe less than or equal to 5 %. Losses of dissolved gas were minimised by performing measurements of pH at ambient temperatures for the coastal sea water over summer Antarctic sea 2 +/- 2.5 degrees C and by housing the pH probe in a rubber bung that sealed the vessel of bubbled seawater during measurement. Samples were measured in the same apparatus. The alkalinity, temperature, salinity (measured using a LF197 conductivity meter (WTW)), pressure (provided by Davis meteorology) and pH of the sea water samples were then used to determine the ambient pCO2 of the seawater in the minicosms. It is noteworthy that the pH of seawater was commonly higher than the anticipated 8.05 for sea water at ~1 degrees C. This coincided with seawater that was super-saturated in O2, suggesting depletion of CO2 by primary production. TCO2 measurements: Samples were gently transferred into 250 ml glass bottles containing 100 micro litres of a saturated solution of mercuric chloride. Bottles were then tightly capped with lids that had a convex insert to exclude all air and retained at 4 degrees C and returned to Australia for determination of TCO2. Samples were analysed at CSIRO marine laboratories (need a ref here for method). Adjustment of seawater pH: The pCO2 in one of the tanks was maintained at ambient while the contents of the other 5 tanks were adjusted by adding CO2-saturated seawater. CO2-saturated water was generated by bubbling CO2 gas through an enclosed container of seawater for greater than or equal to one hour. Measurement of pH can be used as a saturation indicator. The alkalinity, salinity, temperature, and pressure of the seawater in the tanks was determined as above. The amount of CO2-saturated seawater required to reach the desired pCO2 was calculated using Lewis and Wallace (1998) at known physical characteristics (as above). The TCO2 values equated to approximately 1, 2, 3 and 4 times current atmospheric CO2 concentrations (~ 380, 760, 1140 and 1520 ppm CO2) with duplicate tanks at the 3x current pCO2 treatment. These values encompass the range of atmospheric CO2 concentrations predicted by IPCC by the end of the 21st century (Ref??). The desired pCO2 was achieved by carefully transferring known volumes (up to 1.1L at a time to a total volume of ~2 to 5L) of CO2-saturated seawater to medical saline infusion bags and allowing the contents of the bags to enter each minicosm over a period of ~ 2 h (flow rate of less than 40 mL CO2/min). The pH of each tank was measured following the initial addition of CO2-saturated seawater to ensure that the desired pCO2 was obtained. The pCO2 was maintained by measuring the pH and calculating the desired volume of CO2-saturated water to add on a daily basis. The daily dosing was performed using the same method. Nutrients Micronutrients I was not possible to perform the render the minicosm incubation tanks trace metal. This was not considered to be a problem as Antarctic coastal waters are not considered to be limited by trace metal availability. To avoid any variation in iron availability amongst minicosm tanks 5 nm final concentration iron was added to each of the tanks at the start of each experiment. Macronutirents The methods for macronutrient analysis are yet to be determined as these are yet to be performed by CSIRO Nitrate: Ammonium: Phosphate: Silicate?: Particulate and dissolved organic carbon All glassware and filters were muffled at 500 degrees C for 8 h prior to use. Forceps and other plastic ware were soaked in 10% Decon 90 detergent for greater than 2 d and thoroughly rinsed in MilliQ water. Forceps were left in 100% acetone between samples and rinsed with MilliQ prior to use. At each sample time, aliquots of less than or equal to 1 l of water from were obtained from each minicosm, transferred to a thermally insulated container and maintained on ice for less than 1 h. A known volume was filtered through a muffled 25 mm Sartorius Quartz microfibre disc (nominal pore size 0.8 microns) until the filter clogged. The resulting filtrate was collected directly into Whirlpaks (Nasco, U.K.) and retained frozen at -20 degrees C until analysis. Analysis of filtrate to determine concentrations of non-permeable dissolved organic carbon (DOC) was performed at Davis Station using a Shimadzu TOC 5000 total organic carbon analyser calibrated using freshly prepared solutions of potassium hydrogen phthalate in MilliQ water containing 0 to 10.0 mg C l-1. Each calibration solution was acidified with 45 micro litres 3 M HCl and sparged for 9 mins with high purity nitrogen to remove dissolved inorganic carbon (DIC). Whirlpaks containing frozen filtrate were thawed, 5 ml decanted into muffled sample vials the samples acidified and sparged as above. MilliQ blanks (unfiltered) and operational blanks (using MilliQ filtered as above) were performed at Davis Station, both of which gave DOC concentrations near the limits of detection. Operational blanks were subtracted from the DOC concentrations in the seawater samples. Analyses were performed using a that was calibrated A coefficient of variance less than 2% of the concentration was obtained over 3 to 5 replicate analyses of each sample. The 25 mm Sartorius Quartz microfibre disc (above) was retained to determine concentrations of particulate organic carbon (POC) and nitrogen (PON). Filters were folded in half, sample inward and frozen (as above) prior to analysis. A muffled glass cap on the end of a syringe was used to punch greater than or equal to 5 replicate 2.69 mm diameter subsamples from each filter. All 5 punches were transferred to a single silver POC cup (Elemental Analysis Ltd), 20 micro litres of 2N HCl added to each up remove inorganic carbon and the cup then transferred to a 60 degrees C oven for 24-48 h to dry. When dry each cup was folded shut compressed and the concentration of POC and PON determined using a Carlo Erba Elemental Analyser at the University of Tasmania. Chlorophylls At each sampling time for each minicosm tank, a known volume of seawater was filtered though a 13 mm diameter Whatman GF/F filter Whatman GF/F filters (0.7 microns nominal pore size) using vacuum less than 0.5 atm, while protected from light. The filter was folded in half sample inwards, blotted dry to remove excess sea water and frozen in liquid N2 for returned to Australia (RTA'd) for analysis. In addition at the end of each minicosm experiment an bulk sample (up to ~4 l) was filtered onto a 47 mm Whatman GF/F filter to maximise the pigment retained on the filter and aid with pigment identification by HPLC where necessary. Pigments were extracted by a modification of the method of Mock and Hoch (2005). Filters were soaked in 300 micro litres dimethylformamide for 1 hr in the freezer (-18 degrees C) then shaken for 20 sec at 4800 cycles/min (Biospec Products Mini-BeadBeater) in the cryotube, with approx 0.6g zirconia beads (0.7mm dia., Biospec), following addition of 140 ng apo-8'-carotenal (Fluka) internal standard in 50 micro litres methanol. The extract was cleared of particulate matter by centrifugation, as follows: the base of the cryotube was pierced with a hot needle, the cryotube was placed on top of a second cryotube inside a 15 ml plastic centrifuge tube, and together they were centrifuged in a Heraeus Multifuge 3 in a swing-out rotor for 12 min at 2300 x g, and the supernatant was transferred to an amber autosampler vial with a 350 micro litres micro insert. Extracts (125 micro litres) were automatically diluted to 80% with water immediately before injection to improve peak shape (Wright and Jeffrey, 1997) and analysed by HPLC (Zapata et al., 2000) using a Waters 626 pump, Gilson 233xL autoinjector (with the sample stage refrigerated to -15 degrees C), Waters Symmetry C8 column (150 x 4.6mm, 3.5 microns packing, in a water bath at 30.0 +/- 0.1 degrees C), a Waters 996 diode array detector and a Waters 2475 and Hitachi FT1000 fluorescence detector. Pigments were identified by comparison of their retention times and spectra with those of mixed standards obtained from known cultures (Jeffrey and Wright, 1997) that were injected with each batch of samples. Peaks were integrated using Waters Empower 2 software. All peaks were checked manually and corrected where necessary, and quantified using the internal standard method (Mantoura and Repeta, 1997). Unidentified peaks were roughly quantified using the closest match from our spectral library. NPP and GPP; Oxygen Measurements Dissolved oxygen was measured at one minute intervals in the tanks using optically isolated fibre-optic minisensors connected to a four-channel (Oxy-4) meter system (PreSens GmbH). The sensors were calibrated using 0% and 100% oxygen standards. For the 0% standard, millipore water was bubbled with nitrogen gas for 15 minutes. For the 100% standard, wet cotton wool was placed in a closed Schott bottle (100 ml) to form water-vapour saturated air. For calibration at both 0% and 100% the minisensors were inserted through holes that were drilled into Schott bottle lids. A thermometer was also inserted through an additional hole so that temperature could be measured at the same time as calibration. Once calibrated, the sensors were inserted into the minicosm tanks and logging commenced. As there were only four minisensors rather than six, the tanks monitored were the x0.5, x2, x3A and x4 treatments. All data were temperature and salinity corrected. Throughout experimentation there were obvious spikes in the oxygen data caused through electrical activity in the container that housed the minicosm tanks. The data spikes were filtered by comparing each data point with the average of six surrounding data points and rejecting the data point if the difference was greater than or less than 0.3 micro mol O2 L-1. Data measured an hour before or after the lights switched between off/on were also rejected to avoid possible transition effects on measurements (e.g. through carry-over of cell physiology, or the possibility that the microsensors were not completely optically isolated). Rates of change in oxygen concentration with time (i.e. slopes) were determined through least-squares linear regression. Slopes measured during the day (19 hours light) represented net primary productivity (NPP). Negative slopes of oxygen consumption measured during the night (5 hours dark) represented respiration. These rates were dependent both on the rates of photosynthesis/respiration per cell, as well as the total biomass of phytoplankton and all other organisms within the tanks. Gross primary production (GPP) rates were calculated from the addition of NPP and respiration rates. Protists Autos/Hets - Sub-samples were taken from the minicosm tanks for the identification and enumeration of protist species by microscopy. Sub-samples of up to 1-2 l were concentrated over 47 mm, 0.8 microns polycarbonate filters and were fixed to a final concentration of 1% glutaraldehyde for SEM and TEM (below). After resuspension, cells were observed at 400 times magnification under Nomarski optics and blue epifluorescense (filter set 487909 with 450 to 490 nm exciter filter, 510 chromatic beam splitter and 520 nm barrier filter) using a Zeiss Axiovert inverted microscope to determine their trophic status based on the presence or absence of chlorophyll. In addition, samples to determine the species composition and discriminate auto- from heterotrophic taxa were collected at the beginning and end of each experiment. A known volume of water was filter-concentrated to approximately 5 ml above a 25 mm diameter 0.8 mm diameter black polycarbonate membrane filter (MicroAnalytix). The sample was agitated to resuspend cells and stained by adding 0.1 ml of 0.010 % w:v 4',6-diamidino-2-phenylindole (DAPI). The sample was incubated in the dark for 15 mins to stain filtered to dryness, a drop of immersion oil and a coverslip placed on the filter and the sample observed under UV (filter set . ) and blue epifluorescent excitation (as above) to discriminate protistan cells that lacked chlorophyll autofluorescence. Electron Microscopy - Identification of cryptic species was aided by electron microscopy. The remainder of the glutaraldehyde-fixed concentrate (above) was refrigerated at 4 degrees C and RTA'd scanning electron microscopy. For scanning electron microscopy (SEM) cells were settled onto a polylysine-coated glass coverslip, dehydrated with a graded series of acetone, critical point dried, sputter coated with gold and examined using a JEOL JSM 840 SEM. For transmission microscopy (TEM), a droplet of ~40 micro litres of cell concentrate (see above) was pipetted onto Parafilm "M" (American National, Canada) that was attached inside a petri dish and a TEM grid that had been coated in placed on polylysine treated formvar-coated copper grids (Marchant and Thomas, 1983) was positioned at the base of the droplet. The sample post-fixed for 60 s with 2% OsO4 vapour and cells allowed to sedimented onto the TEM grid for 30-40 mins. Each grid was then gently rinsed with MilliQ water excess water carefully blotted from the surface of the grid and the grid then air-dried and RTA'd. Grids were then shadow-cast with platinum/palladium or chromium and examined with a Phillips CM 100 TEM. For microzoopankton analyses (below), sub-samples of 960 ml were fixed with Lugol's Iodine. Protists were allowed to sediment for greater than or equal to 8?d and the supernatant then removed by aspiration. The resulting concentrated samples of ~250 ml were retained for subsequent counting, upon return to Australia samples were stored in the dark at 4 degrees C until counted. Further removal of the supernatant was performed to reduce the sample volume to ~ 100 ml. The final volume of the concentrate was measured and, depending on cell density, a sub-samples of 3-10 ml were settled in 10 ml sedimentation cylinder (Hydro-Bios, Keil) and allowed to sediment for 2 d. Counts were performed over 20 randomly selected fields of view at 400 and 1000 times magnification by inverted microscopy using Nomarski optics (as above). The mean and standard errors of the concentration of each protistan taxa/group and were calculated in each tank treatment. For assessment of the composition and abundance of all protists, permanent slides were prepared from Lugol's fixed material according to the 2-hydroxypropyl methacrylate (HPMA) (Sigma) method of Crumpton (1981), further modified by Sung-Ho Kang (KORDI, pers. comm.). Each sample was filtered at less than 5 kPa onto a 0.45 microns pore size, 25 mm diameter, GN-6 cellulosic membrane filter (Gelman). The filter was then rinsed by filtering a further 15 ml of MilliQ water, removed from the filtration apparatus and placed face down on a cover glass. A few drops of HPMA were placed on the back of the filter and the sample then transferred to a 60 degrees C cabinet for 12 to 24 h to clear the filter and polymerise the HPMA. Further drops of HPMA were then placed on the back of the filter, a slide placed on top and the sample again polymerised as above for 6-12 h. Protistan taxa were identified and counted in 20 randomly chosen quadrats (Whipple grids) for each HPMA slide using a Zeiss Axioscop equipped with Nomarski interference optics and epifluorescence. Counts were performed at 400 x magnification and blue light epifluorescent excitation (filter set 487909 with 450-490 nm exciter filter, 510 nm chromatic beam splitter and 520 nm barrier filter) was used to discriminate autotrophic cells due to chlorophyll autofluorescence. Molecular biology samples: Known volumes of sample from each minicosm tank were filtered on 25mm GF/F Whatman filters at the beginning and end of each minicosm incubation and stored in 2ml cryotubes in liquid N2 for analysis in Hobart. Subsequent RNA analyses were performed to establish gene libraries and discriminate the taxonomic content and affinities of the microbial community. In addition, pyrosequencing was performed to determine the gene expression of the microbial communities amongst the treatments to discriminate differences in activity and characterise whether specific expressions were assoicted with taxa that were inhibited or enhanced by high pCO2 concenrtations. This work is to be performed by Nikki Clargo, a PhD student co-supervised by myself, Simon Jarman and Andrew McMinn. Lipids All sampling equipment was treated as specified for POC/DOC sample collection. Known volumes (~ 2 l) from each minicosm tank was filtered through pre-weighed Whatman 47mm GF/F filters (cleaned with CHCl3 and MEoH) for lipid analysis. Samples were filtered to dryness, folded sample inward and stored in aluminium satchel at -20 degrees C for RTA and subsequent analysis. Subsequent lipid analyses are to be performed in association with Cathryn Wynne-Edwards, a PhD student co-supervised by myself, So Kawaguchi and others Rates of Production 1 degrees C Production Primary productivity in each minicosm tank was determined every 2 days during the incubation of the microbial community and PAM measurements were conducted every second day during the minicosm incubations. For sampling, 400 ml of water was obtained from each tank at each sampling time and stored in darkened polycarbonate jars in a sea-water cooled insulated container, until the commencement of measurements. Primary productivity incubations were conducted according to the method of Griffiths et al. (1999), based on the small bottle 14C technique of Lewis and Smith (1983). 6.327 x 106 Bq (0.171 mCi) NaH14CO3 were added to 162 ml of sample to produce a working solution of 39.183 x 103 Bq per ml (1.1 micro C ml-1). Seven ml aliquots of working solution were then added to transparent glass scintillation vials and incubated for 1 hour at 21 light intensities ranging from 0 to 485 micro mol m-2 s-1. The temperature of the light incubator was controlled by water-baths set to within +/- 0.1 degrees C of in situ values. After 1 hour, 250 micro litres of 6M HCl was added to each vial and they were then agitated for 3 hours to ensure that all inorganic carbon was removed. Gaseous 14CO2 was trapped in carbosorb cartridges after being pumped through silica gel to ensure the air was dry. For radioactive counts, 10 ml Aquasure scintillation fluid was added to each vial and shaken. Samples were then counted using a Beckman LS6500 scintillation counter with the maximum counting time set at 5 min. In addition, Time 0 counts were taken to determine background radiation and 100% counts were used to determine the specific activity of the working solution. For Time 0 counts, 7 ml aliquots of working solution were subjected to acid addition without any exposure to light, and counted after shaking for 3 hours. For 100%'s, 100 micro litres of working solution was added to 7 ml NaOH (0.1 M) and immediately counted following the addition of scintillation fluid. Carbon uptake rates were corrected for in situ chlorophyll a concentrations measured using HPLC (Wright and van den Enden 2000) and for total dissolved inorganic carbon availability (see above). Photosynthesis-irradiance (P-I) relationships were then plotted and analysed using SYSTAT 8.0. The equation of Platt et al. (1980) was used to fit curves to the data using least squares non-linear regression. Photosynthetic parameters determined included light-saturated photosynthetic rate [Pmax, mg C (mg chl a)-1 h-1], initial slope of the light-limited section of the P-I curve [delta, mg C (mg chl a)-1 h-1 (micro mol m-2 s-1)-1], light intensity at which carbon-uptake became maximal calculated as Pmax/ delta (Ek, micro mol m-2 s-1), intercept of the P-I curve with the carbon uptake axis [c, mg C (mg chl a)-1 h-1] , and the rate of photoinhibition where applicable [beta, mg C (mg chl a)-1 h-1 (micro mol m-2 s-1)-1]. Bacterial Production Bacterial production in each minicosm tank was also determined every 2 days as above using the micro-centrifuge method of Kirchman (2001). The concentration of [14C]leucine added to production samples was determined using saturation experiments conducted on the 10th Dec 2008. Results showed that that bacterial uptake of [14C]leucine was saturated at concentrations greater than or equal to 35? nM. ?? ???Vol of of 5 micro Ci mL-1 working stock solution of [14C]-Leucine was added to 5 microcentrifuge tubes for each minicosm tank. Ninety micro litres 100% TCA (5% final concentration) was then added to two of the tubes which will act as killed controls, 1.7 mls of sea water pipetted into the microfuge tube, the tube capped and immediately mixed. All tubes in were then incubated for a known duration (30-60 min) in the dark at 2 degrees C and incubations were begun within 30 mins of sample collection. Following incubation samples were killed by adding 90 micro litres 100% TCA the samples were mixed using a vortex mixer and allowed to and wait 10 minutes then centrifuged at 14 500 for 10 min at 4 degrees C and the supernatant removed by aspiration. The sample was then rinsed by adding 1.7 ml 5 % ice-cold TCA, mixing, centrifuging and removing the supernatant (as above). Then repeating the rinse using 80% ice-cold ethanol and the ethanol allowed to evaporate. One ml of Ultima Gold scintillation cocktail was then added to each tube, the microfuge tube housed in a 20 ml glass scintillation vial and the activity measured using a Beckman LS 6500 liquid scintillation counter. Flow Cytometry Seawater for analysis by flow cytometry (FCM) was collected through a 50 micron mesh every second day. Fresh samples for nano-protists were analysed using a FACScan FCM (Becton Dickinson) and fixed samples for bacteria and virus-like particles (VLP) were analysed using a FACSCalibur (Becton Dickinson). Both flow cytometers used a 488 nm argon laser. Samples were weighed to +/- 0.0001 g before and after each run to determine the volume analysed. Concentrations of microbes were then calculated using event counts from bivariate scatter plots and the volume analysed. Ten microlitres of PeakFlow Green 2.5 micron beads (Molecular Probes) was added to all samples as an internal standard. Nanoplankton As picoplankton are rare in the Indian Ocean Sector of the Southern Ocean south of 60o S (Kosahi et al. 1985), cells enumerated here are defined as nanoplankton (2 - 20 microns). Fresh samples for nanophytoplankton were housed in a beaker containing ice and analysed for 10 min at a flow rate of 60 micro litres min-1. Nanophytoplankton were discriminated using their red autofluorescent in bivariate scatter plots of red versus orange fluorescence (FL3 and FL2, respectively). Lysotracker Green-stained heterotrophic nanoflagellates (HNF) (Rose et al. 2004; Thomson et al. 2009) were discriminated from red autofluorescent nanophytoplankton in bivariate scatter plots of green versus red fluorescence (FL1 and FL3, respectively). A working solution of Lysotracker Green (Molecular Probes) was prepared daily by diluting the 1 mM commercial stock 1:10 with 0.22 microns filtered seawater. Ten ml of seawater from each treatment was stained with 7.5 micro litres of the working solution (75 nM final concentration) and incubated in the dark and on ice for 10 min. Following incubation, a 1 ml sub-sample was transferred to a sterile 5 ml Falcon tube. Samples were housed in a beaker containing ice and run for 10 min on high flow rate (approximately 60 micro litres min-1) using 0.22 microns filtered seawater as sheath fluid. Bacteria Concentrations of total bacteria and those with high and low DNA (HDNA and LDNA respectively) were determined from fixed samples. Samples were fixed to a final concentration of 0.5% glutaraldehyde for 1 hr, stained for 20 mins using 1:10000 final dilution of SYBR-Green I (Molecular Probes) and analysed by FACSCalibur (Marie et al. 1999). Samples were run for 5 min on medium flow rate (approximately 20 micro litres min-1) using MilliQ water sheath. Bacterial numbers were determined from bivariate scatter plots of SSC verses FL1 and the bacterial abundance calculated from the volume analysed. Viruses One ml samples for VLP were collected for at each sampling time. Samples were fixed with 0.5% (final concentration) of glutaraldehyde in the dark at 0 degrees C for 15 minutes, frozen in liquid nitrogen and transported to Australia (Brussaard, 2004). After rapid thawing at 37 degrees C, samples were diluted (1:10) in 0.02 microns filtered TE Buffer (10 mM Tris, 1mM EDTA, pH 8), stained with SYBR-Green I solution (1:5000 dilution) and incubated at 80 degrees C in the dark for 10 minutes (Brussard, 2004). Acquisition was run for 2 min and 400 to 800 events s-1 were collected at ~40 micro litres -1 min-1. VLP were discriminated in bivariate plots of FL1 and SSC according to Brussaard (2004). Fluorescent beads of 1 micron diameter (Molecular Probes) were added to all samples, as an internal standard. FCM sorting Nanophytoplankton populations identified by FCM on the FL3 versus FL2 bivariate scatter plots were sorted using the FACSCalibur. The sheath fluid for sorting was 4 degrees C, 0.22 microns filtered seawater and the sorted fluid was collected in 50 mL centrifuge tubes containing 5 mL of frozen sheath. The identified populations were sorted at a flow rate of 20 micro litres min-1. Up to 25 mL subsamples of the sorted fluid were fixed with Lugols iodine for later identification and the remaining fluid of approximately 20 mL was dispensed into 50 mL culture flaks and incubated at 4 oC under light (light intensity/cycle?). Microzooplankton and HNF grazing The dilution technique, modified according to Gallegos et al. (1996), was used to quantify rates of growth and grazing mortality of phytoplankton (Landry and Hassett 1982) and bacterioplankton (Tremaine and Mills 1987). Rates growth and mortality were determine in the beginning middle and end of the minicosm incubations in the tanks with ambient, 3 and 4 times pCO2 (Tanks 1, 4 and 6). At each sample time, approximately 45 l of sea water from each was gently transferred to 2, 25 l acid washed and MilliQ rinsed black polythene carbuoys. The contents of on of the was then gravity filtered through large capacity capsule 0.2 microns Supor (Gelman) inline filters as recommended by Lavrentyev et al. (2004) to minimise physical damage to plankton (thus enrichment of dilutent) during filtration. The diluent was then transferred into 2.4 l acid washed and MilliQ rinsed polycarbonate bottles to create a dilution gradient of triplicate bottles containing 0, 30, 60, 90 and at high microbial biomass 99% filtered dilutent and one bottle was filled with 100% dilutent as a blank. The bottles were then gently filled with sea water. As the minicosms had been filled with water that had been filtered through a 200 microns mesh to remove metazooplankton, this was not repeated in preparation of the grazing dilution bottles. A further 3 polycarbonate bottles were filled with undiluted seawater and sacrificed to estimate time zero concentrations of Chl a, bacteria and heterotrophic nanoflagellates (HNF) (below). Triplicate subsamples of undiluted, 200 microns filtered, sea water (~280 ml) were also obtained from the reservoir, fixed with Lugol's iodine (1% final concentration) and refrigerated at ~ 4 degrees C to determine the initial abundance of microzooplankton in incubated samples (below). The remaining triplicate bottles containing 100, 70, 40, 10 % sea water, and single bottle containing 100 % dilutent, were placed on a shaker table rotating at 70 rpm in a cold room at 1 degrees C, exposed to the same light irradiance as that in the minicosm and incubated for 24 h. Following incubation, subsamples (~500-1000 ml) were obtained from each bottle to determine concentrations of Chl a, bacteria, and HNF. Samples were also obtained from each undiluted bottle, fixed with Lugol's iodine and the abundance of microzooplankton determined by inverted microscopy (below). Concentrations of macronutrients are not regarded as limiting microbial productivity in Antarctic waters (e.g. Odate and Fukuchi 1995). Thus, in contrast to Landry and Hassett (1982), nutrients were not added to the grazing dilution bottles prior to incubation. Chlorophyll concentrations A known volume (250-1000 ml) of sea water from each of the polycarbonate bottles (above) was size fractionated using a 20 microns mesh and a GF/F filter (0.7 microns particle retention) in succession. This gave particle separation of pico- and nano-sized Chl a fraction (0.7 to 20 microns) from the micro-sized (20-200 microns) fraction. Filters were folded, blotted, transferred into cryovials and stored in liquid nitrogen. Upon return to Australia, samples were removed from liquid nitrogen and stored in an ultra low freezer at -135 degrees C until pigment analysis by HPLC (see above). Bacterial and heterotrophic nanoflagellate abundance Flow cytometry was used to determine concentrations of total bacteria and heterotrophic nanoflagellates (HNF) from unpreserved samples following staining by SYTO 13 and LysoTracker Green, respectively. See Thompson et al. (above) for methods. Microzooplankton abundance Microzooplankton (20-200 microns) abundance was determined by inverted microscopy. A measured 250 ml from each of the triplicate Lugol's-preserved bottles (as above) was transferred into a measuring cylinder, allowed to sediment for greater than or equal to 4 d, and the supernatant removed by aspiration to give a sample volume of 20 ml. Duplicate 10 ml subsamples of concentrate were transferred to separate sedimentation chambers and the number of microzooplankton in each of 20 randomly chose fields of view determined using a Zeiss Axioskop inverted microscope at 200 and 400 x magnification and their identity determined using Scott and Marchant (2007). Growth and grazing rates From each experiment conducted, the apparent growth rates (mu) in each bottle were calculated from the change in concentration of Chl a or bacteria at each dilution, using the equation: mu= 1/t ln(Nt/N0) (Landry and Hassett 1982) where Nt and N0 were the bacterial/Chl a concentrations at the beginning and end of the grazing experiment respectively, and t was the duration of the experiment in days. Coefficients of growth (mu) and microzooplankton grazing (g) were determined from least squares linear regression analysis (using Microsoft Excel) of apparent growth rates versus the fraction of unfiltered water in each bottle; the regression slope representing the rate of grazing mortality and the y-intercept representing the growth rate of the prey. The percentage of standing stock grazed was calculated after Safi et al. (2007): % standing stock grazed = (1-e-g) x 100 % production grazed = 100*(1-e-g)/(e mu -1) where e = base e Linear feeding kinetics were corrected for the growth of grazers during the incubation by dividing linear regression slopes by the relative geometric mean predator density (GMPD) as described by Gallegos et al. (1996). Growth rates of HNF and microzooplankton were estimated based on their initial and final concentrations in undiluted bottles. Lewis, E., and D. W. R. Wallace. 1998. Program Developed for CO2 System Calculations. ORNL/CDIAC-105. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tennessee. Three experiments were performed at Davis Station, East Antarctica 77 degrees 58' E, 68 degrees 35' S to determine the effects of ocean acidification on natural assemblages of Antarctica marine microbes (bacteria, viruses, phytoplankton and protozoa). Incubation tanks (minicosms) were filled on the 30/12/08, 20/01/09 and 09/02/09 with sea water that was filtered through 200 microns mesh to remove metazoan grazers. The pH of the contents of each tank was then adjusted by adding calculated amounts to CO2 saturated sea water to achieve and maintain CO2 concenrtations that encompassed atmospheric concenrtations from pre-industrial to post-2100. As 6 tanks were available the 3 x current CO2 treatment was duplicated to indicate the variance among replicate tanks. Instead, responses were analysed to determine trends among concentrations. The microbial communities were incubated for 10, 12 and 10 days, in experiments 1, 2 and 3 respectively. Chemical and biological parameters were measured every second day to determine concentrations of macronutrients, particulate and dissolved organic carbon, pigment composition, dissolved oxygen, concentrations of phytoplankton, protozoa, bacteria (and viruses) using flow cytometry, light and electron microscopy, lipids, rates of primary, bacterial production and microzooplankton grazing. These data have been collected as part of ASAC project 40 (ASAC_40), and Terrestrial Nearshore Ecosystems project 8A. The excel spreadsheet contains: Separate sheets reporting the results from each of the 3 experiments run at Davis Station in the 2008/09 summer. Abbreviations are as follows: Nutrients: NO3 =nitrate, PO4 = Phosphate, Si = silicate Primary production and respiration were determined from oxygen microelectrodes: net photosynthesis from oxygen increase during exposure to light and respiration determined from net decrease in oxygen in the absence of light. Photosynthetic parameters were also measure using 14C bicarbonate as a trace for Carbon uptake, these being: maximum photosynthetic rate) Pmax, Photosynthetic efficiency (Alpha) and saturating light intensity (Ek). Flow cytometry was used to count 7 microbial parameters: pico phytoplankton (Picos) nanophytoplankton in two regions (Nano R2 and Nano R3). Cryptophytes, high DNA bacteria (HDNA_bact) and low DNA bacteria (LDNA_bact). Microscope cell counts identified a range of taxa/groups that comprised greater than 1% of the total phytoplankton abundance: unidentified nanoplankton (UNAN), small pennate diatoms (Pennate less than 10 microns) and other taxa as specified. Organic material measurements including: Particulate organic carbon (POC), Particulate organic nitrogen (PON) particulate carbon to nitrogen ratio (C:N), Dissolved organic carbon (DOC) Intermittent measurements were also made of rates of herbivory and bacterivory and rates of phytoplankton and bacterial growth in 3 of the 6 tanks. Photosynthetic pigments were measured and are given only for experiment 1 so far (other to come later): Beta-Beta carotene (BB carotene), Chlorophylls c1 (Chl c1), c2 (Chl c2), c3 Chl c3), a (Chl a), b (Chl b), Chlorophyllide a (Chlidea), diadinoxanthin (Ddx), Diatoxanthin (dtx), Chl a epimer (epi), Fucoxanthin (Fuc), 19'-hexanoyloxyfucoxathin (Hex), Methyl Chlorophyllide a (MeChlidea), Magnesium divinyl pheaoporphyrin monomethyl ester (MgDVP), Phaeophytin (Phaeo), Violaxanthin (viola) and total pigment concentration. CHEMTAX will also be performed using these pigments to study CO2-induced changes in phytoplankton community structure. Antarctica is likely to be amongst the first regions of the earth to be affected by ocean acidification due to the high solubility of CO2 in cold water and the upwelling of high CO2 water off the Antarctic coast. Yet little is known regarding the effect of ocean acidification on Antarctic biota, especially marine microbes and particularly non-calcifying taxa. Virtually without exception, studies that have looked at such effects have done so by bubbling the samples with at the target CO2 concentration. Though easy and convenient as a means of adjusting sea water pCO2 it is a practice not recommended by the EPOCA guidlines for experiments studying the biological effects of ocean acidification experiments. This study aims to determine the responses of natural communities of Antaerctic marine microbes to changes in pH from the pre-industrial era (1800s) to concentrations predicted for the atmosphere by the end of this century 2100. |
|---|