Effects of hydrostatic pressure on microbial alteration of sinking fecal pellets

1

Effects of hydrostatic pressure on microbial alteration of 1

sinking fecal pellets 2

Christian Tamburini1, Madeleine Goutx1, Catherine Guigue1, Marc Garel1, 3

Dominique Lefèvre1, Bruno Charrière1, Richard Sempéré1, Stéphane Pepa2, 4

Michael L. Peterson3,†, Stuart Wakeham4, and Cindy Lee5 5

1 Université de la Méditerranée, Centre d'Océanologie de Marseille, LMGEM UMR 6117 6 CNRS – INSU, Campus de Luminy, Case 901, 13288 Marseille, Cedex 9, France 7

2 Metro-Mesures, 47 Bd Charles de Gaulle, 91540 Mennecy, France 8 3 School of Oceanography, University of Washington, Box 357940, Seattle, WA 98195-5351, 9

USA 10 † Present Address: Ocean Science Consulting and Research, 9433 Olympus Beach Road NE, 11

Bainbridge Island, WA 98110 12 4 Skidaway Institute of Oceanography, 10 Ocean Science Circle, Savannah, GA 31411, USA 13 5 Marine Sciences Research Center, Stony Brook University, Stony Brook, NY 11794-5000, 14

USA 15

Running title: Sinking particle simulation 16

For submission to Deep Sea Research II MedFlux Issue. After 2nd review 17

18

19

Tamburini et al.

2

Abstract 1

We used a new experimental device called PASS (PArticle Sinking Simulator) during 2

MedFlux to simulate changes in in situ hydrostatic pressure that particles experience sinking 3

from mesopelagic to bathypelagic depths. Particles, largely fecal pellets, were collected at 4

200 m using a settling velocity NetTrap (SV NetTrap) in Ligurian Sea in April 2006 and 5

incubated in high-pressure bottles (HPBs) of the PASS system under both atmospheric and 6

continuously increasing pressure conditions, simulating the pressure change experienced at a 7

sinking rate of 200 m d-1. Chemical changes over time were evaluated by measuring 8

particulate organic carbon (POC), carbohydrates, transparent exopolymer particles (TEP), 9

amino acids, lipids, and chloropigments, as well as dissolved organic carbon (DOC) and 10

dissolved carbohydrates. Microbial changes were evaluated microscopically, using DAPI 11

stain for total cell counts and catalyzed reporter deposition – fluorescence in situ 12

hybridization (CARD-FISH) for phylogenetic distinctions. Concentrations (normalized to 13

POC) of particulate chloropigments, carbohydrates and TEP decreased under both sets of 14

incubation conditions, although less under the increasing pressure regime than under 15

atmospheric conditions. By contrast, dissolved carbohydrates (normalized to DOC) were 16

higher after incubation and significantly higher under atmospheric conditions, suggesting they 17

were produced at the expense of the particulate fraction. POC-normalized particulate 18

wax/steryl esters increased only under pressure, suggesting biochemical responses of 19

prokaryotes to the increasing pressure regime. The prokaryotic community initially consisted 20

of 43% Bacteria, 12% Crenarchaea and 11% Euryarchaea. After incubation, Bacteria 21

dominated (~90 %) the prokaryote community in all cases, with γ-Proteobacteria comprising 22

the greatest fraction, followed by the Cytophaga-Flavobacter cluster and α-Proteobacteria 23

group. Using the PASS system, we obtained chemical and microbial evidence that 24

Tamburini et al.

3

degradation by prokaryotes associated with fecal pellets sinking through mesopelagic waters 1

is limited by the increasing pressure they experience. 2

Key words: Hydrostatic pressure; particle sinking; organic matter degradation; prokaryotic 3

diversity; mesopelagic and bathypelagic waters; 4

5

Tamburini et al.

4

1. Introduction 1

Sinking particulate organic matter (POM) is the major conduit for the export of carbon 2

and energy from the sea surface to the deep ocean. Relatively little is known about the 3

microbiological factors that determine the composition of POM as it sinks through 4

mesopelagic and bathypelagic waters (Hedges et al., 2000; Wakeham et al., 1997), but 5

prokaryotes attached to particles must play an important biogeochemical role in oceanic 6

carbon flux by dissolving and remineralizing sinking particulate organic matter (POM) (Cho 7

and Azam, 1988; Turley and Mackie, 1994; Turley and Mackie, 1995). Senescence and death 8

of phytoplankton cells cause them to aggregate and sink faster through the water column 9

(Alldredge et al., 1995), and these aggregates are colonized by prokaryotes (Simon et al., 10

2002) that are adapted to low hydrostatic pressure. Yet, as particles sink, the associated 11

prokaryotes are subjected to increasing pressure, the rate of which depends on the settling 12

velocity of the aggregates. 13

The role of pressure on particle dissolution and preservation is poorly known. In spite of 14

numerous studies showing that incubation of deep-sea samples at atmospheric pressure may 15

lead to underestimates of prokaryotic activity rates (see reviews of Deming, 2001, and 16

Tamburini, 2006), degradation rates of sinking particles are usually measured under 17

atmospheric pressure (Bidle and Azam, 1999, 2001; Bidle et al., 2002; Goutx et al., 2007; 18

Panagiotopoulos et al., 2002; Sempéré et al., 2000). As a consequence, current 19

biogeochemical models do not take into account the possible effects of hydrostatic pressure 20

on decomposition during sinking. For example, modeling of the silica cycle is based on the 21

assumption that biogenic silica dissolution is controlled by temperature, zooplankton grazing 22

and diatom aggregation rate (Dugdale and Wilkerson, 1998; Nelson et al., 1995; Tréguer et 23

al., 1989) and implicitly assumes that pressure has no effect. 24

Tamburini et al.

5

The fate of sinking biogenic mineral particles depends on aggregation/disaggregation, 1

remineralization of the organic matrix and dissolution of the mineral matrix. Biogenic silica 2

(BSiO2) and calcium carbonate (CaCO3) are the major mineral components in marine 3

plankton and promote the sinking of less-dense organic matter. Minerals may also protect 4

organic matter from degradation (e.g., Hedges et al., 2001), or provide a matrix that holds 5

particles together in larger aggregates (Lee, 2004), allowing organic matter to penetrate 6

deeper into the ocean. Both degradation and dissolution can be mediated by microbial 7

activities. Milliman et al. (1999) proposed that dissolution of calcium carbonate might be 8

biologically-mediated, while other studies have demonstrated that bacteria accelerate diatom 9

opal dissolution by colonizing and enzymatically degrading the organic matrix of diatom 10

frustules (Bidle and Azam, 1999, 2001; Passow et al., 2003). Bidle et al. (2002) found that 11

temperature effects on marine bacteria that degrade diatom detritus can strongly influence the 12

coupling of biogenic silica and organic carbon preservation. Tamburini et al. (2006) applied 13

increasing pressure to an axenic, culture-generated phytodetritus and found a decrease 14

(relative to atmospheric pressure incubations) of the aminopeptidase activity of the natural 15

prokaryotic assemblage associated with the detritus, and consequently a decrease of the 16

dissolution of silica. 17

For the present study, we applied the approach of Tamburini et al. (2006) to 18

investigate the effect of continuously increasing pressure on organic matter degradation in 19

natural marine particles. We describe the apparatus that was used, namely the PArticle 20

Sinking Simulator (PASS), to simulate both the increase of hydrostatic pressure as particles 21

sink and corresponding changes in temperature. We compared atmospheric and pressurized 22

degradation rates of sinking particles, freshly recovered with a settling velocity NetTrap by 23

measuring changes in organic compound concentrations and prokaryotic community 24

structure. 25

Tamburini et al.

6

2. Materials and methods 1

2.1. PArticle Sinking Simulator (PASS) 2

To mimic the increase in pressure that occurs when particles sink, incubations were 3

carried out using high-pressure bottles (HPBs) where pressure was increased continuously 4

(linearly) using a piloted pressure generator. HPBs were rotated to maintain particles in 5

suspension, and incubated in water baths to reproduce the temperature variability with depth. 6

Fig. 1 presents a schematic diagram of the PASS system. 7

2.1.1. High-pressure bottles: description and preparation 8

High-pressure bottles (HPBs) have been used in previous studies to obtain 9

undecompressed deep-sea water samples (Bianchi et al., 1999; Tamburini et al., 2003). 10

HPBs are 500-ml APX4 stainless steel cylinders (75-mm OD, 58-mm ID and 505-mm total 11

length) with a 4-mm thick polyetheretherketone (PEEK®) coating; the bottles were 12

constructed by Metro-Mesures (Mennecy, France). The PEEK® floating piston (56-mm total 13

length) is fitted with two O-rings. The screw-top end-cap is covered with a sheet of PEEK to 14

avoid contact between the sample and the stainless steel. Viton® O-rings are used to ensure 15

that the system is pressure-tight; Viton® O-rings are chosen instead of nitrile O-rings to 16

eliminate possible carbon contamination of the sample. The screwed bottom end-cap is 17

connected, via a stainless steel tube, to the piloted pressure generator. Two HPB top end-caps 18

were modified to fit an O2 pulse electrode sensor (Langdon, 1984) in order to measure O2 19

concentrations over time during the incubations. 20

A cleaning protocol was developed to minimize organic carbon contamination and to 21

ensure the quality of DOC and organic compound analyses. The efficiency of cleaning was 22

evaluated by systematic DOC analysis at each step of the protocol. HPBs were dismantled 23

and all parts dipped in 0.5M HCl for 30 min. HPBs were reassembled without the floating 24

piston and rinsed with 1.5 l of MilliQ water. Then, HPBs were washed again for 2 h by 25

Tamburini et al.

7

recirculating 0.5 M HCl (at 0.2 ml sec-1) through them with a peristaltic pump and rinsing 1

with 1.5 l of MilliQ water. The floating piston was replaced, and the HPBs filled with MilliQ 2

water. Finally, HPBs were autoclaved and stored full until the experiment. Just before the 3

experiment, HPBs were emptied and filled with the sample under sterile conditions. 4

2.1.3. HPB rotation system 5

The rotation system (Metro-Mesures, Mennecy, France, Fig. 2) was composed of a 6

stainless steel frame (800 x 800 x 260 mm) to which two HPBs were attached, an electronic 7

control box with 3 positions (0°, 180°, automatic), and a stainless steel tube to connect HPBs 8

to the piloted pressure generator. Usually, one bottle is kept at atmospheric pressure (ATM), 9

while a second bottle is subjected to a pressure increase (HP) that mimics the chosen sinking 10

rate. The rotating stainless steel frame is raised on a compressed air jack and is piloted using 11

the electronic control box. In automatic mode, the time to activate a half-revolution of the 12

HPBs is between 15 seconds and several minutes; this rate was chosen to attempt to maintain 13

particles in suspension and to minimize turbulence (bubbles in the HPBs), collision and shear 14

between cells as discussed by Passow et al. (2003) both in increasing pressure regime (HP) 15

and atmospheric pressure (ATM) incubations. 16

2.1.4. Piloted Pressure Generator 17

To ensure maximum reliability and accuracy during the linear increase in hydrostatic 18

pressure, a piloted pressure generator was constructed (Metro-Mesures, Mennecy, France, 19

Fig. 1). In contrast to classical high-pressure pumps based on alternating movement of a 20

small piston associated with inlet and exhaust valves, the piloted pressure generator is based 21

on a step motor-driven syringe. A step motor with a high starting torque is connected to a 22

high precision worm screw operating the high-pressure syringe. Adjustment of the movement 23

of the syringe is controlled by a digital computer fitted with a high-power processor. The 24

software Metrolog (Metro-Mesures, Mennecy, France) allows an increase (or decrease) in 25

Tamburini et al.

8

range of hydrostatic pressure with precise regulation (0.2%); it also allows constant or 1

variable flow injection so that the volume injected can be measured precisely. The working 2

pressure is up to 40 MPa; the working volume is 400 ml. The piloted pressure generator is 3

very safe using only hydrostatic pressure. 4

With the piloted pressure generator it is possible to mimic pressure changes 5

experienced as particles fall through the water column by increasing pressure up to 40 MPa 6

(~4000 m). To simulate a sinking rate of 200 m d-1, the programmable computer-driven 7

system of the piloted pressure generator was adjusted to 0.02315 kPa sec-1 (corresponding to 8

2.0 MPa d-1). It is also possible to easily transfer samples from one HPB to another without 9

any change of pressure during the manipulation. 10

2.1.5. Temperature regulation 11

Temperature was regulated using stainless steel water baths (995 x 305 x 700 mm; 12

316L; IMC Marindus, Marseille, France) and a copper coil at the bottom connected to a 13

temperature cooler (Huber, Unichiller UC009T, Fisher-Bioblock) with a Polystat cc3 14

controller. A RS232 connection links the Polystat cc3 controller with a computer running 15

Labworldsoft® 4.01 (IKA-Werke). An external probe (Pt100) allows complete temperature 16

control of the water bath. In this way, it is possible to reproduce the changes in temperature 17

with depth (or decrease in case of oceanic waters), further mimicking the fall of particles 18

through the water column. 19

2.2. Incubation of freshly recovered sinking particles 20

2.2.1. Sampling strategy and recovery of particles 21

Particles were collected at the French time-series DYFAMED station in the Ligurian 22

Sea, located at 28 nautical miles southeast of Nice, France (43°25' N, 07°52' E), during the 23

MedFlux-III cruise on the RV Endeavor (April 7-17, 2006). Sinking particles were collected 24

using a settling velocity NetTrap (SV NetTrap), a floating plankton net equipped with an 25

Tamburini et al.

9

indented rotating sphere (IRS) valved sediment trap that sorts particles into discrete classes as 1

a function of their settling velocity (Peterson et al., 2008; Peterson et al., 2005). The SV 2

NetTrap was deployed at 200 m depth for 24 hours (April 8-9); IRS valves were set to rotate 3

every 2 hours and particles were collected in glass tubes filled with 75 ml 0.2-µm filtered, 4

unpoisoned seawater collected from 200 m. Position 1 of the SV NetTrap is empty to allow 5

water to pass through during deployment and recovery. Sample tubes 2 and 3 were 6

eliminated because they contained material that differed visually from the other tubes. The 7

remaining sample tubes 4 to 8 were pooled and split using a MacLane WSD-10 wet sample 8

divider to prepare aliquots for simultaneous pressurized and atmospheric incubation 9

experiments. Visual examination and chloropigment composition of the particles showed 10

them to be mostly fecal pellets. 11

2.2.2. Sinking particle simulation 12

Particle splits were resuspended in 500 ml of filtered (GF/F filters of 0.7-µm nominal 13

pore size, combusted at 450°C for 6 h) seawater from 200 m depth. The samples were then 14

transferred into HPBs that were either maintained at atmospheric pressure (ATM) during 15

incubation or subjected to a linear increase of pressure (HP). For HP incubations, the piloted 16

pressure generator was adjusted to run at 0.02315 kPa sec-1 (i.e., at 2 MPa d-1) for the 156-h 17

(6.5 d) incubation, simulating a sinking velocity of 200 m d-1 and reaching a depth at the end 18

of incubation equal to 1500 m (Fig. 3). SV-NetTrap tubes 4 to 8 collected particles sinking at 19

70-326 m d-1 (Peterson et al., 2005, 2008). ATM and HP incubations were carried out in 20

triplicate simultaneously in 3 temperature-regulated water baths (a pair of ATM and HP 21

bottles in each bath). The aggregates were kept in suspension by rotating the HPBs half a 22

revolution every minute. The incubation temperature was programmed to mimic in situ 23

temperature profiles measured using a Sea-Bird CTD on the day of particle collection. The 24

initial temperature was 13.00°C, corresponding to the temperature at 200 m. Assuming a 25

Tamburini et al.

10

sinking rate of 200 m d-1, 13.00°C was maintained for 30 min, 13.00 - 13.70°C for 49 h, 13.70 1

- 13.30°C for 56.5 h, 13.30 - 13.45°C for 64 h, 13.45 - 13.35°C for 72 h, 13.35 - 13.55°C for 2

80 h, 13.55 - 13.20°C for 160 h, and then 13.20°C until the end of the incubation (Fig. 3). At 3

the end of the incubation period, the rotation was ended and HPB bottles were gently 4

decompressed taking around 5 min to reach ambient pressure. Sub-sampling was done using 5

a Peristaltic liquid dispenser (Jencons Scientific Ltd) for chemical and microbial analyses. 6

2.3. Chemical analyses 7

2.3.1. Post-incubation procedure 8

For organic analyses, sample aliquots (about 90 ml) were filtered under a low vacuum 9

(< 50 mm Hg) through pre-combusted GF/F filters (25 mm filter diameter). All glassware 10

was pre-combusted before the cruise and rinsed with 1 N HCl and Milli-Q water after each 11

sample. Plastic gloves were worn and care was taken to minimize contamination during 12

sampling. 13

2.3.2. POC and DOC 14

Sub-samples for particulate organic carbon (POC) were collected in pre-combusted 15

500-ml glass bottles, filtered, and the filters were acidified with acid fumes to remove 16

inorganic carbon. POC was measured according to Tan and Strain (1979) using a Carlo Erba 17

model 1602 CNS analyzer. Precision for OC is ± 2%, with a standard deviation for five 18

replicate samples of 5% at the start of the incubation. Experimental errors based on duplicate 19

analyses were estimated to be about 10% for POC concentration during the incubation 20

experiment. 21

Samples for DOC analysis were taken from the filtration apparatus above with 10 ml glass 22

pipettes, transferred to 10 ml glass ampoules, and sealed after addition of 10µl of H3PO4 as 23

preservative. Before analysis, samples were acidified to pH 1 with 85 % phosphoric acid and 24

bubbled for 10 minutes with CO2-free air to purge inorganic carbon. DOC was measured on 25

Tamburini et al.

11

3-4 replicates by high temperature catalytic oxidation using a Shimadzu TOC 5000 Analyzer 1

(Sohrin and Sempéré, 2005). Quantification was performed using a four point-calibration 2

curve with standards (0.5 to 2 mg C l-1) prepared by diluting potassium hydrogen phthalate in 3

Milli-Q water. 4

2.3.3. Carbohydrates 5

Particulate sugar (PCHO-C) and dissolved sugar (DCHO-C) samples were collected as 6

for POC and DOC above. Particulate sugar samples were analyzed as described previously 7

(Panagiotopoulos and Sempéré, 2005). Experimental errors based on duplicate analyses were 8

estimated to be about 10% for PCHOtot concentrations. The standard deviation estimated 9

from duplicate analysis for individual monosaccharides was lower than 8% for every sugar 10

except ribose (17%). Recoveries after desalting (Mopper et al., 1992), estimated using spiked 11

(20-100 nM) sodium chloride solutions, ranged from 72 to 79% for fucose, rhamnose, 12

arabinose, mannose, xylose and ribose, and between 84-100% for galactose and glucose, 13

respectively. Procedural blanks, run with desalted sodium chloride solutions, showed only a 14

small peak of glucose (~5 nM), but a systematic contamination peak induced by desalting 15

coeluted with fructose and prevented its quantification. The detection limit was 5-10 nM for 16

individual sugars. Concentrations of free dissolved sugars are corrected for blank values 17

(glucose) but not corrected for loss during desalting. For combined sugars, 4 ml of sample 18

were hydrolyzed (0.4 ml of 1 N HCl, at 110°C for 20h) and were neutralized (final pH ~ 4.5) 19

in a self-adsorbed resin bed (Sempéré et al., 2007). 20

2.3.4. Particulate amino acids 21

Total particulate hydrolyzed amino acid (PHAAtot) concentrations were measured on 22

GF/F particles filtered as described above and stored frozen until analysis. Amino acids were 23

analyzed by HPLC using pre-column o-phthaldialdehyde (OPA) derivatization after acid 24

hydrolysis as previously described (Lee and Cronin, 1982; Lee et al., 2000). An amino acid 25

Tamburini et al.

12

mixture (Pierce, Standard H) was used as the standard. The non-protein amino acids, β-1

alanine, and γ-aminobutyric acid (BALA and GABA), were added individually to the standard 2

mixture. Aspartic acid (ASP) and glutamic acid (GLU) measurements include the hydrolysis 3

products of asparagine and glutamine. Analytical errors determined from duplicate analysis 4

were <10%. 5

2.3.5. Particulate lipids 6

Particulate lipids on GF/F filters were extracted as described in Goutx et al. (2007). 7

Lipid extracts were stored in dichloromethane under nitrogen at -20°C until analysis. Lipids 8

were separated into classes of compounds on chromarods and quantified by an Iatroscan 9

model MK-6s (Iatron, Tokyo; H2 flow 160 ml min-1; air flow 2 l min-1) coupled to a PC 10

equipped with a Chromstar 6.1 integration system (Bionis, Paris). The separation scheme 11

involved five elution steps in solvent systems of increasing polarity according to a modified 12

procedure (Goutx et al., 1990) that separates neutral lipid classes (hydrocarbons, sterol esters 13

co-eluting with wax esters, ketone as internal standard, triacylglycerols, free fatty acids, 14

alcohols sterols and diglycerides), chloroplast lipids (pigments, glycolipids and 15

monoglycerides) and non-nitrogen containing phospholipids (diphosphatidylglycerides co-16

eluting with phosphatidylglycerides) from nitrogen containing phospholipids 17

(phosphatidylethanolamine and phosphatidylcholine). Under these conditions, the relative 18

standard deviation of replicate samples (n=3) was 3-11%. 19

Fatty acids were analyzed in part of the lipid extract as in Goutx et al. (2007). Free 20

and esterified fatty acids were derivatized to fatty acid methyl esters (FAMEs) using 21

BF3/methanol/toluene for 1 h at 70°C under N2. FAMEs were extracted from the aqueous 22

phase with hexane:ether (9:1, V:V) and purified on silica micro-columns, i.e., hydrocarbons 23

eluted first with hexane, FAMEs were then eluted with hexane:ethyl acetate (100:1), and 24

polar lipids remained on the silica. FAMEs were analyzed by capillary gas chromatography 25

Tamburini et al.

13

(Goutx et al., 2007). Quantification of individual components was based on an external 1

calibration using the response factor of tricosanoic methyl ester (Sigma). 2

2.3.6. Total chloropigments 3

Chloropigments (chlorophyll+pheopigments) and fucoxanthin were measured by 4

reverse-phase High Performance Liquid Chromatography (HPLC; Mantoura and Llewellyn, 5

1983; Bidigare et al., 1985; Lee et al., 2000). Pigments measured include chlorophyll a, 6

pheophorbide a, pyropheophorbide a, and pheophytin a. Monovinyl and divinyl chlorophylls 7

(Bidigare and Ondrusek, 1996) were not separated. Briefly, filters were extracted with 100% 8

acetone, diluted 20% with MilliQ water, and separated by HPLC using a 5-µm Adsorbosphere 9

C-18 column. Pigments were identified by fluorescence (excitation λ= 440 nm, emission λ = 10

660 nm); fucoxanthin was identified by absorbance (λ = 446 nm). Chloropigment and 11

fucoxanthin concentrations were identified by comparison of sample peaks and pigment 12

standards. Chlorophyll a was purchased from Turner Design, pheophorbide a from Porphyrin 13

Products, and fucoxanthin from DHI (Hørsholm, Denmark). Pyropheophorbide a and 14

pheophytin a were prepared from pheophorbide as in King (1993). Duplicate analyses of the 15

same extract agreed within 10 %. 16

2.3.7. Transparent exopolymer particles (TEP) 17

TEP was analyzed colorimetrically (Passow and Alldredge, 1995) from duplicate 4 ml 18

samples fixed with paraformaldehyde (2% final concentration), stored in liquid nitrogen until 19

stained with Alcian Blue, then filtered onto 0.2-µm polycarbonate filters. Replicate MilliQ 20

water samples were filtered (0.2-µm) for use as blanks. Analytical errors determined from 21

triplicate analyses were <10%. Our method differs slightly from that of Passow and 22

Alldredge (1995) who found that filter storage (at -20°C) is possible after Alcian blue staining 23

but results are more accurate without formaldehyde addition. Moreover, we used 0.2-µm 24

polycarbonate filters instead of 0.4-µm polycarbonate filters. Thus, it is not possible to 25

Tamburini et al.

14

directly compare our results with other published data or to use the carbon conversion factor 1

proposed by Engel (2004). However, the processing of T0, HP and ATM samples was 2

internally consistent, making between-treatment comparisons valid. 3

2.4. Microbiological analyses 4

Samples were fixed with 0.2-µm filtered formaldehyde (2% final concentration), kept 5

at room temperature for 15 min, and subsequently stored at 4°C in the dark for around 12 h. 6

2.4.1. DAPI counts 7

Total prokaryotic counts were obtained by filtration onto 0.2-µm-pore-size 8

polycarbonate filters supported by 0.45-µm-pore-size cellulose nitrate filters. To obtain the 9

fraction <2µm, samples were sequentially filtered by gravitation onto 2-µm-pore-size 10

polycarbonate filters and then 0.2-µm-pore-size polycarbonate filters supported by 0.45-µm-11

pore-size cellulose nitrate filters. Cells were stained with diamidinophenylindole (DAPI) 12

according to Porter and Feig (1980). The fraction >2µm was calculated by subtracting values 13

of the fraction <2µm from the total fraction. 14

2.4.2. CARD-FISH 15

Samples were filtered onto 0.2-µm-pore-size polycarbonate filters supported by 0.45-16

µm-pore-size cellulose nitrate filters, washed twice with 0.2-µm filtered MilliQ water, air-17

dried, and stored in scintillation tubes at -20°C until analysis. Filters for counts of catalyzed 18

reporter deposition coupled with fluorescence in situ hybridization (CARD-FISH) cells were 19

embedded in low-gelling point 0.1% agarose (Sigma), dried at 37°C for 10 min, and 20

dehydrated with 95% ethanol. For detection of Bacteria, embedded cells were permeabilized 21

by subsequent treatments with lysozyme (10 mg ml-1, Sigma) in 0.5M EDTA, 0.1M Tris-HCl 22

(pH 8.0) for 60 min at 37°C (Amann et al., 1995) and with achromopeptidase (60U in 0.01M 23

NaCl, 0.01M Tris-HCl, pH 8.0) for 30 min at 37°C (Sekar et al., 2003). Filters were then 24

rinsed 3 times with 0.2-µm filtered MilliQ water and placed in 0.01M HCl at room 25

Tamburini et al.

15

temperature (to denature endogenous peroxydase). Following acid treatment, filters were 1

washed with 0.2-µm filtered MilliQ water and dipped in 95% ethanol and air-dried. Filters 2

were stored at -20°C until analysis. For detection of Archaea, embedded cells were 3

permeabilized with proteinase K (0.2 µl ml-1, Fluka) according to Teira et al. (2004). Filters 4

were cut in sections and hybridized with 5'-horseradish peroxydase (HRP)-labeled 5

oligonucleotide probes as described by Pernthaler et al. (2002) with some modifications. 6

Probes used in this work are listed in Table 1. Eub-II and Eub-III were mixed in equal 7

proportions with Eub338 to target the Planctomycetales and Verrucomicrobia missed by 8

Eub338 (Daims et al., 1999). We used the same concentration for unlabeled competitor 9

probes, i.e., unlabeled Gam42a with labeled Bet42a, and unlabeled Bet42a with labeled 10

Gam42a (Manz et al., 1992). For each probe a filter piece (cell-adherent side) was placed on 11

a Parafilm-covered glass slide overlaid with a 30 µl drop of hybridization solution with HRP-12

probe using a final DNA concentration of 1 ng µl-1. The glass slides were placed in a closed 13

50-ml tube, and incubated overnight at 35°C. After the hybridization and amplification steps, 14

pieces of filter were mounted in Citifluor:Vectashield:DAPI mix. Slides were examined 15

under an Olympus BX61 microscope equipped with a 100-W Hg-lamp and appropriate filter 16

sets for DAPI and Alexa488. The fraction of CARD-FISH+ (CARD-FISH-stained) cells was 17

quantified for at least 1000 DAPI-stained cells per sample. Prior to counting, the slides were 18

stored at –20°C up to several days without loss of fluorescence intensity. Negative control 19

counts (hybridization with HRP-Non338) averaged 1% and were always below 5% of DAPI-20

stained cells. 21

3. Results 22

3.1. Experimental protocol and general parameters (P, T, O2) 23

Particles recovered from the 2006 SV NetTraps were almost exclusively copepod (and 24

some salp) fecal pellets as indicated by visual observation under microscopy and confirmed 25

Tamburini et al.

16

by pigment composition. Indeed, pigment compositions show that the chloropigments were 1

dominated by pyropheophorbide (>90%, Table 6), a zooplankton grazing product (Shuman 2

and Lorenzen, 1975; Ziegler et al., 1988). Satellite data (SeaWiFS data for Chl and AVHRR, 3

TERRA, and AQUA datasets for SST; Jay O’Reilly and Teresa Ducas, NOAA/NMFS, 4

Narragansett, RI, USA, personal communication) show that our sampling occurred before the 5

spring bloom, so that diatom aggregates were not present. This contrasted with previous 6

MedFlux cruises when high chlorophyll concentrations showed that the spring bloom was 7

underway and during which NetTraps collected more diatom aggregates (Lee et al., 2008). 8

Incubation experiments using freshly recovered particles were conducted in the PArticle 9

Sinking Simulator (PASS) with a linear increase in pressure simulating sinking from 200-10

1500 m and a minor variation in temperature (in the range of 13.0-13.8°C) that simulated 11

CTD-derived water column temperatures (Fig. 3). O2 concentrations decreased from ~ 410 12

µM to ~ 300 µM after 156 hours of incubation under both atmospheric and high pressure 13

conditions, so incubations remained oxic. 14

3.2. Chemical and microbiological composition of particles and dissolved fractions 15

3.2.1. Chemical results 16

Bulk organic compounds. Concentrations of TOC (total organic carbon, or the sum 17

of POC and DOC, Table 2) were not significantly different under increasing pressure regime 18

(HP, mean±SE: 111±17 µM) and under atmospheric (ATM, 126±23 µM) conditions at the 19

end of the incubation but both higher than initially (88±2 µM). 20

Patterns of OC-normalized concentrations of dissolved and particulate carbohydrates 21

were opposite in the HP and ATM incubations (Fig. 4a,b). DCHO:DOC increased in both 22

treatments from initial values, but the endpoint values in ATM samples were significantly 23

higher (n=6, p=0.0361, Student Test t0.05=2.57) than in T0 or HP samples (Fig. 4a). On the 24

other hand, PCHO:POC decreased over time in both cases, but the endpoint values in HP 25

Tamburini et al.

17

samples (4.8-6.5%) were significantly higher (n=6, p=0.0145, Student Test t0.05=2.57) than in 1

ATM samples (3.0-3.8%, Fig. 4b). 2

Although we could not convert TEP to TEP-C, we looked at the pattern of TEP 3

concentration normalized to POC. The pattern of TEP:POC was similar to that of 4

PCHOtot:POC, with HP values about the same or slightly higher than the initial value, but 5

significantly higher (n=6, p=0.0500, Student Test t0.05=2.57) under HP than under ATM 6

pressure conditions (Fig. 4c). 7

The sum total of particulate organic compounds that we identified (i.e., PHAAtot, 8

PLiptot, PPigtot and PCHOtot) accounted for 37±13, 43±20 and 37±16 % of POC in T0, HP and 9

ATM bottles, respectively (Fig. 4). Initially, particulate total carbohydrates (PCHOtot ,Fig. 10

4b), particulate total amino acids (PHAAtot, Fig. 4d), and total lipids (PLiptot,Fig. 4e) 11

comprised equal proportions of POC (10-13% each), whereas particulate total pigments 12

(PPigtot) accounted for only 1.9 % (Fig. 5e). After 156 h of incubation, the proportion of POC 13

made up by PHAAtot (~20%) and PLiptot (~22%) was about 50% higher and was not 14

significantly different between the HP and ATM bottles. In contrast, proportions of PCHOtot 15

and PPigtot in POC both decreased by about half in HP samples (0.8% and 5.5%, respectively) 16

and even more in ATM samples (0.4 and 3.5%, respectively) after 156 h of incubations (Fig. 17

4b and Fig. 4e). 18

Organic composition of the particulate fraction. Particles were initially enriched in 19

several PHAAtot, PCHOtot, PLiptot and PPigtot biomarkers (see Tables 3, 4, 5 and 6 20

respectively) that together indicate the presence of fresh phytoplankton organic matter 21

embedded in zooplankton fecal pellets. After 156 h of incubation, mole percents of most 22

amino acids did not change significantly, but glucosamine and glycine decreased somewhat 23

(Table 3). Among particulate carbohydrates (Table 4), galactose decreased while fructose 24

(for both incubation conditions) and ribose (for ATM conditions) increased after 156 h of 25

Tamburini et al.

18

incubations. However, as total classes neither PHAAtot nor PCHOtot showed much difference 1

between ATM and HP conditions. The proportion of PUFA in total fatty acids decreased by a 2

factor of 2 (on average) from 6.4±0.9 at T0 vs. 3.0± 1.1 in HP and 3.9± 1.4 % in ATM 3

samples (Table 5). In contrast, wax and steryl esters (WSE) were significantly higher (n=6, 4

p=0.0005, Student Test t0.05=2.57) after incubation under increasing pressure (mean =8.3, 5

S.D.=1.3) compared to ATM (mean =1.6, S.D.=1.6) pressure conditions (Fig. 4g and Table 6

5). Wax ester concentrations in POC increased ~2- fold in HP bottles after incubation (Fig. 7

4g). 8

3.2.2. Microbiological results 9

Total abundance of prokaryotes increased from 0.5±0.1 x 106 cells ml-1 (mean±SE) at 10

T0 to 2.0±0.2 and 2.1±0.2 x 106 cells ml-1 (mean±SE) after 156 h of incubation under 11

increasing pressure and under atmospheric pressure, respectively (Fig. 5a). Both free-living 12

(fraction <2µm) and attached (fraction >2µm) prokaryotes increased with time. The fact that 13

both size fractions behaved in a similar manner suggests that both are involved in organic 14

matter degradation to similar extents. 15

The prokaryotic community at T0 consisted of 43% Bacteria, 12% Crenarchaea and 16

11% Euryarchaea (Fig. 5). At T156, Bacteria were ~90% of prokaryotes, a ~2-fold increase 17

over T0. In contrast, Archaea (Crenarchaea and Euryarchaea) decreased to <10% of total 18

prokaryotes by the end of the incubation. 19

Among the Bacteria, γ-Proteobacteria group dominated under both increasing 20

pressure (HP) and ATM conditions, with average values of 66 ± 6 and 62 ± 8% of Bacteria, 21

respectively. The percentage of Cytophaga-Flavobacter cluster was slightly lower under HP 22

(20 ± 3%) than under ATM pressure conditions (25 ± 2%). The percentage of α-23

Proteobacteria group was clearly lower under HP (15 ± 4%) than under ATM conditions (24 24

± 3%). Regardless of pressure, β-Proteobacteria abundance was below detection limits. 25

Tamburini et al.

19

4. Discussion 1

4.1. Particle sinking simulator 2

It is difficult to measure in situ degradation of organic matter and dissolution of 3

mineral ballast (e.g. silica, calcium carbonates and dust) of marine particles sinking 4

throughout the water column. With the PASS, however, it is possible to manipulate various 5

experimental parameters and to obtain a first-order simulation of pressure effects on 6

degradation. The small volume of HPBs (500 ml) limits the analyses that can be performed to 7

a few chemical parameters in a time-series mode, or somewhat more parameters when only an 8

initial and final time are sampled. Replicates can be performed by increasing the number of 9

water baths, temperature coolers, piloted pressure generators and HPBs, such as in the 10

triplicate systems used here. 11

An important limitation of our experimental setup is that the HPB is a closed system 12

containing water and particles from 200 m while we apply temperature and pressure 13

variations that mimic a 1500-m water column. Particles actually sinking through the water 14

column cross water masses with different physical, chemical, and biological characteristics. 15

They come into contact with gradients of ambient prokaryotic communities that may affect 16

degradation and dissolution in different ways. Whereas it is technically possible in future 17

experiments to add aliquots of seawater and their microbial communities from different 18

depths while maintaining in situ temperatures and pressures ; we started in these first 19

simulations with a simpler system and more well-defined materials. The incubations we 20

performed thus mimic to an imperfect degree what happens in the vicinity of a particle 21

sinking through the water column. Exchange with other prokaryotes may occur during 22

sinking, but this exchange might be important to degradation if particle sinking velocities are 23

too high compared to the generation times of prokaryotes in meso- and bathypelagic waters. 24

For the fast sinking fecal pellets we used, exchange might be minimal. Another complication 25

Tamburini et al.

20

that we can not evaluate at present might be interactions between particles and the walls of the 1

pressure vessels, potentially resulting in fragmentation of the experimental particles. We 2

hope that the slow rotation of the HPBs minimizes wall contact and subsequent particle 3

disaggregation, but we can not rule it out. In any case, increasing pressure regime (HP) and 4

atmospheric pressure conditions were done in the exact same manner except for the pressure 5

so differences between HP and ATM are exclusively due to the pressure incubation 6

differences. 7

To our knowledge, the only similar work on pressure effects is by Turley (1993). She 8

incubated sinking particles and their associated microbial assemblages, collected over 48 h at 9

200-m depth in a sediment trap that did not sort particles according to sinking velocity 10

(Lampitt, 1992), in sealed bags within pressure vessels held at a constant temperature of 5°C. 11

Pressures of 0.1, 10, 20, 30 and 43 MPa were applied within 30 min and maintained constant 12

for 4 h. She also incubated whole seawater samples from depths of 10 and 40 m under similar 13

conditions, evaluating microbial activity in both the seawater and sediment trap samples via 14

leucine and thymidine incorporation methods. The aim of her work was 1) to explain how 15

relatively fresh aggregates are found on the deep-sea floor that still contain sufficient organic 16

carbon to fuel the rapid growth of benthic micro-organisms, and 2) to indicate that pressure 17

effects on microbial processes may be important in oceanic biogeochemical cycles. So, 18

Turley’s work is relevant to understanding the capacity of prokaryotes that originate from the 19

sub-surface and are carried down with sinking particles to degrade organic matter reaching the 20

deep-sea bed. In contrast, our study focused on the prokaryotic processes and particle 21

degradation in the mesopelagic zone just after particles exit the euphotic zone and before their 22

arrival to the deep-sea floor at their actual settling velocity. Here, we examined the chemical 23

and microbial components of the incubated particles using long-term incubations (156 h); 24

Turley used short-term (4-h) incubations. Moreover, we applied a continuous increase in 25

Tamburini et al.

21

pressure and maintained corresponding in situ temperatures, thus mimicking the P/T 1

conditions of descent, while Turley (1993) applied constant pressures (10, 20, 30 and 40 2

MPa) representing multiple depths, relevant for the different processes occurring at the deep 3

water-sediment interface rather than processes occurring throughout the water column. 4

The continuous and linear change in pressure and minor variations in temperature 5

applied to water samples during our incubations corresponded to those encountered in situ at 6

this location by particles sinking at ~200 m d-1. Particle sinking rates typically range from 1 - 7

510 m d-1 for phytoplankton (Smayda, 1970), 43 - 95 m d-1 for marine snow (Shanks and 8

Trent, 1980), and an average of 105 m d-1 for sinking particles (McCave, 1975). Other field 9

and experimental studies report values ranging from 1 - 370 m d-1 (Alldredge et al., 1995 and 10

references therein). Armstrong et al. (2008) calculated an average sinking rate of 242 m d-1 for 11

particles at the DYFAMED site during 2003 and 2005. Temperature may also change as 12

particles sink. During our experiment, we varied the temperature according to the CTD data 13

measured during the cruise, but vertical variations in temperature in the Mediterranean Sea 14

are not large, ~13°C below 200 m. However, changes in temperature with depth can be 15

important elsewhere in deep ocean waters where temperatures can be as low as 2°C or even 16

lower in deep polar basins. When prokaryotes are adapted to high pressure and low 17

temperature, their physiology (e.g., membrane lipid and protein content and composition) 18

responds to these extreme conditions (Yayanos, 1995). Using the PASS system, it would also 19

be possible to mimic how both temperature and pressure affect the physiology and activity of 20

prokaryotes. 21

4.2. Effects of increasing pressure on organic composition of sinking particles 22

The particulate carbon introduced into the pressure bottles initially increased by 5-fold 23

the ratio of POC to DOC usually found in seawater from the DYFAMED site (Van Wambeke 24

et al., 2001). The contribution of the biochemicals we measured to POC was 37±13% of OC. 25

Tamburini et al.

22

For comparison, at DYFAMED site in 1995 Goutx et al. (2000) accounted for 30-40% of OC 1

in drifting traps as the sum of protein-C, lipid-C and glucosamine-C. During the 2003 2

MedFlux field program, Wakeham et al. (2008) found that summed amino acid-C, lipid-C and 3

pigment-C constituted ~35% of OC in moored trap samples and Goutx et al. (2007) accounted 4

for 57-85% of OC as amino-acid-C, lipid-C, and carbohydrate-C in SV NetTrap samples. 5

ATM POC concentrations were substantially higher after 156 h (6.5 d) than initial values and 6

than HP POC concentrations, possibly due to an increase in prokaryotic biomass in ATM vs 7

HP samples. 8

Carbohydrate distributions showed a significant difference between increasing high 9

pressure (HP) and atmospheric (ATM) incubations. DCHO increased slightly in HP samples 10

over initial values, but increased by 20 fold in the ATM experiments (Table 2). PCHOtot and 11

TEP concentrations decreased much more under ATM than under HP regimes by T156 12

compared to T0 (Fig. 4b and c), suggesting that complex carbohydrates were degraded faster 13

under ATM conditions. Pigments also were subjected to more degradation under ATM than 14

under HP incubation conditions although the differences were not significant statistically 15

because of higher degradation in HP3 pigments compared to HP1 and HP2 (Fig. 4e). 16

This pattern of preservation of HP organic matter is consistent with previous results 17

showing that the prokaryotic capacity to degrade particulate organic matter under high-18

pressure is lower than at atmospheric pressure (Lochte and Turley, 1988; Tamburini et al., 19

2006; Turley, 1993). Turley (1993) showed that rates of leucine and thymidine incorporation 20

in both seawater and sediment-trap samples were lower at the higher pressure tested, while 21

there was no significant influence of pressure on cell numbers. Tamburini et al. (2006) 22

demonstrated that aminopeptidase activity was significantly lower (by nearly 5 fold) after a 23

period of continuously increasing pressure compared to atmospheric pressure. Lower 24

aminopeptidase activity under increasing pressure led to a reduction in the decrease in 25

Tamburini et al.

23

biogenic silica dissolution, at least for the first 800 m (depth-simulated) of the water column. 1

Clearly then, incubation experiments with sinking particles conducted at atmospheric pressure 2

overestimate the rate of organic matter decay. This pressure effect of decreasing 3

decomposition and dissolution reinforces the observations that fast settling particles are less 4

degraded in the water column and may therefore provide a more labile food resource for 5

bathypelagic and epibenthic communities (Goutx et al., 2007; Honjo et al., 1982; Wakeham, 6

1983). 7

However, under ATM conditions, the increase in DCHO relative to t0 was associated 8

with a decrease of PCHO and TEP (Fig. 4a, b and c). Interestingly, Engel et al. (2004) 9

reported abiotic transfer of DCHO to TEP during a bloom experiment with Emiliania huxleyi 10

in surface conditions giving rise to the production of particles rich in dissolved carbohydrates. 11

These authors showed that polysaccharide aggregation was driven by DCHO-TEP. However, 12

when TEP-rich particles are degraded, they may release DCHO. In natural environments, 13

sinking particles may be trailed by a plume of nutrients, carbon and microbes that can create 14

hotspots of growth and carbon cycling by free-living prokaryotes (Kiørboe and Jackson, 15

2001; Long and Azam, 2001). Our results would suggest that heterotrophic mechanisms in 16

ATM pressure conditions are more efficient at degrading carbohydrate-rich particles than 17

under increasing pressure conditions. This highlights the importance for studying further 18

POM/DOM exchanges during fast settling particle incubation experiments. 19

Adaptations of cultivated bacteria to high pressure are characterized by changes in the 20

composition of cellular fatty acids (Bartlett, 1999; DeLong and Yayanos, 1985; Kato et al., 21

1998; Nogi et al., 1998; Sinninghe-Damste et al., 2002). In our study, the large initial input 22

of plankton-derived fatty acids from particles did not allow us to evaluate the effect of 23

pressure on bacterial fatty acids as they were minor components. The most striking feature of 24

the lipid class pattern, however, was the higher contributions of wax and steryl esters (WSE) 25

Tamburini et al.

24

to POC in HP samples compared to ATM and T0 samples (Fig. 4g and Table 5). WSE and 1

their constituent fatty acids are usually attributed to zooplankton and fish (Lee and Hirota, 2

1973), but in fact they may also have bacterial sources. The accumulation of WSE in cells 3

and extracellular medium has been reported for metabolically stressed marine prokaryotes 4

growing either under nitrogen depleted conditions (e.g., on carbon-rich oleyl and hydrocarbon 5

substrates ; Fixter et al., 1986 ; Goutx et al., 1990; Kaneshiro et al., 1996) or in N-starved 6

conditions (Ishige et al., 2003; Rontani et al., 1999; Silva et al., 2007), and low temperature 7

(Russell and Volkman, 1980). The higher abundances of WSE at the end of our incubations 8

with increasing pressure (compared to atmospheric pressure) are consistent with the above-9

cited studies and probably reflected the metabolic response of bacteria to high pressure. This 10

is also consistent with the measurement of an increase in prokaryotic cells (by around 7-fold) 11

at the end of the incubations. Hence, our results suggest that hydrostatic pressure enhances 12

the biosynthesis of WSE in prokaryotes, although this still has to be clearly demonstrated. 13

Finally, as WSE are commonly used as tracers of organic matter zooplankton origin in marine 14

sinking particles (Wakeham, 1983), if pressure does trigger WSE biosynthesis in prokaryotes, 15

then this process must be taken into account and the interpretation of WSE biomarkers as 16

exclusively zooplankton-derived be reconsidered in sediment traps deeper than 1000 meter 17

depth. 18

Overall our results confirm that under increasing pressure (i.e., as when particles and 19

attached-prokaryotes are sinking through the water column), the prokaryotic capacity to 20

degrade particulate organic matter at depth is lower than at atmospheric pressure (Lochte and 21

Turley, 1988; Tamburini et al., 2006). Turley (1993) incubated sinking particles and their 22

associated microbial assemblages collected at 200-m depth in the NE Atlantic with 3H-23

thymidine and 3H-leucine to measure DNA and protein synthesis under varying pressures to 24

simulate sinking. She concluded that DNA and protein synthesis by free-living and particle-25

Tamburini et al.

25

attached bacteria were significantly and adversely affected by pressure but that pressure did 1

not affect bacterial cell numbers. Furthermore it is unlikely that bacteria attached to particles 2

originating at the surface would have much impact on degradation in the bathypelagic zone. 3

Turley’s (1993) observations helped to explain how labile organic matter could reach the sea 4

floor with minimal degradation, and that pressure was an important factor that had not been 5

adequately examined. Tamburini et al. (2006) later demonstrated that aminopeptidase 6

activity was significantly lower (by nearly 5 fold) after a period of continuously increasing 7

pressure compared to atmospheric pressure. Lower aminopeptidase activity under increasing 8

pressure also reduced biogenic silica dissolution, at least for the first 800 m (depth-simulated) 9

of the water column. Clearly then, incubation experiments with sinking particles conducted at 10

atmospheric pressure overestimate the rate of organic matter decay. This pressure effect of 11

decreasing decomposition and dissolution reinforces the observations that fast settling 12

particles are less degraded in the water column and may therefore provide a more labile food 13

resource for bathypelagic and epibenthic communities (Goutx et al., 2007; Honjo et al., 1982; 14

Turley, 1993; Wakeham and Lee, 1993). 15

4.3. Effect of increasing pressure on prokaryotic community structure 16

While the prokaryotic community initially consisted of 43% Bacteria, 12% 17

Crenarchaea and 11% Euryarchaea, after 156 h, Bacteria were dominant (close to 90% of 18

total prokaryotes) whereas Archaea (Crenarchaea plus Euryarchaea) represented less than 19

10% of total prokaryotes. Organic matter in fresh fecal pellets can be a substrate for 20

heterotrophic Bacteria, but Archaea do not play a significant role in degrading particulate 21

organic matter (Bidle and Azam, 2001; Simon et al., 2002; Tamburini et al., 2006). 22

Incubations of axenic diatom detritus as the source of organic matter showed that Archaea 23

decreased in all incubations as soon as the experiment started (Tamburini et al., 2006) and 24

may help explain why Archaea are not enriched on particles in surface waters (Simon et al., 25

Tamburini et al.

26

2002). The high densities of free-living Archaea found in the deep sea (Church et al., 2003; 1

DeLong et al., 1999; Herndl et al., 2005; Karner et al., 2001; Tamburini et al., 2008; Teira et 2

al., 2004) are thus not related to vertical transport by sinking particles. 3

Among the Bacteria in our incubations, γ-Proteobacteria dominated, followed by 4

much lower numbers of Cytophaga-Flavobacter and α-Proteobacteria. No significant 5

difference between HP and ATM conditions was found for the relative abundance of γ-6

Proteobacteria, but the relative abundances of Cytophaga-Flavobacter and α-Proteobacteria 7

were lower under HP than under ATM pressure conditions. Cytophaga-Flavobacter, α-8

Proteobacteria and γ-Proteobacteria clusters appear to dominate communities associated 9

with marine snow in typical oceanic systems (DeLong et al., 1993; Ploug and Grossart, 1999; 10

Simon et al., 2002) and in the Mediterranean Sea (Acinas et al., 1999; Moeseneder et al., 11

2001). Tamburini et al. (2006) found an increase in the relative abundance of the Cytophaga-12

Flavobacter cluster and of the γ-Proteobacteria when natural prokaryotic assemblages were 13

supplemented with an input of fresh diatom detritus, but the taxonomic composition of 14

prokaryotic communities was not affected by increasing pressure in that experiment. Thus 15

bacterial compositions in our present experiment support other studies suggesting that 16

whereas the Cytophaga-Flavobacter cluster and the γ-Proteobacteria group play a major role 17

in the degradation of high molecular weight organic carbon (Cottrell and Kirchman, 2000b; 18

Kirchman, 2002), α-Proteobacteria appear to dominate uptake of low molecular weight 19

material (Cottrell and Kirchman, 2000a; Malmstrom et al., 2005). Actually, the fraction of 20

the α-Proteobacteria after incubating under increasing pressure (HP) was the same or slightly 21

higher than at time zero, in agreement with previous results with axenic diatom detritus 22

(Tamburini et al., 2006), but clearly lower (15 ± 4%) than under ATM pressure conditions (24 23

± 3%) after 156 h (Fig. 5b). This difference between HP and ATM conditions deviates from 24

Tamburini et al. (2006). In the present study, more dissolved carbohydrate (Fig. 4a) is 25

Tamburini et al.

27

available under ATM conditions as substrate for α-Proteobacteria (Fig. 5b). We can 1

hypothesize then, that in our experiment, Cytophaga-Flavobacter clusters were more sensitive 2

to an increase in pressure and hydrolyzed less particulate carbohydrate; thus less dissolved 3

carbohydrate was produced at HP for α-Proteobacteria to consume, leading to the lower 4

abundance of these organisms as pressure increased. 5

6

5. Conclusion 7

In previous work, Tamburini et al. (2006) showed that continuously increasing 8

pressure to simulate the transit between 200 m and 800 m (depth-simulated) resulted in 9

reduced rates of silica dissolution and thus organic matrix hydrolysis of freshly prepared 10

diatom detritus relative to rates measured under atmospheric pressure conditions. In this 11

study, using the PASS system, we can now show that at least some biochemical fractions of 12

naturally collected sinking particles, composed primarily of fecal pellets, were less degraded 13

by prokaryotes when pressure was continuously increased, simulating descent from 200 to 14

1500 m, than when left at atmospheric pressure. Pressure decreased the number of 15

prokaryotes attached to particles and the apparent activity of free-living prokaryotes. Our 16

results demonstrate the utility of a new tool for studying pressure effects in the ocean and 17

shed new light on how hydrostatic pressure affects microbial alteration of particulate organic 18

matter as particles sink. In doing so, we help to explain why fast sinking particles such as 19

fecal pellets, but possibly also including fast sinking marine snow aggregates, can fall through 20

the water column with minimal degradation. 21

Acknowledgements. 22

This work was funded by the ANR-POTES (ANR-05-BLAN-0161-01, see 23

http://www.com.univ-mrs.fr/LMGEM/potes) program supported by the Agence Nationale de 24

Tamburini et al.

28

la Recherche (France) and the MedFlux program (http://www.msrc.sunysb.edu/MedFlux) 1

supported by the U.S. National Science Foundation Chemical Oceanography Program. We 2

thank Olivier Wyss for help with the cleaning protocol for HPBs, Jennifer Szlosek and Anja 3

Engel for advice on TEP analysis, Lynn Abramson for help with pigment and amino acid 4

analyses, David Hirschberg for CHN analyses, and two anonymous reviewers and Uta Passow 5

for helpful comments on the manuscript. CT thanks E. Teira and G. Herndl for instruction in 6

CARD-FISH methodology. SGW thanks the Hanse-Wissenschaftskolleg/Institute for 7

Advanced Studies (Delmenhorst, Germany) for a Fellowship during the completion of this 8

manuscript. This is MedFlux contribution No. 26 and MSRC contribution No. 1364. 9

10

6. References 11

Acinas, S.G., Anton, J., Rodriguez-Valera, F., 1999. Diversity of free-living and 12

attached bacteria in offshore Western Mediterranean waters as depicted by analysis of genes 13

encoding 16S rRNA. Applied and Environmental Microbiology 65 (2), 514-522. 14

Alldredge, A.L., Gotschalk, C., Passow, U., Riebesell, U., 1995. Mass aggregation of 15

diatom blooms: Insights from a mesocosm study. Deep-Sea Research II 42 (1), 9-27. 16

Amann, R., Ludwig, W., Schleifer, K., 1995. Phylogenetic identification and in situ 17

detection of individual microbial cells without cultivation. Microbiol. Rev. 59 (1), 143-169. 18

Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., Stahl, D.A., 19

1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for 20

analyzing mixed microbial populations. Applied and Environmental Microbiology 56 (6), 21

1919-1925. 22

Armstrong, R.A., Lee, C., Peterson, M.L., Cochran, J.K., Wakeham, S.G., 2008. Sinking 23

velocity spectra and the ballast ratio hypothesis. Deep-Sea Research II, this volume. 24

Tamburini et al.

29

Bartlett, D.H., 1999. Microbial adaptations to the psychrosphere/piezosphere. Journal of 1

Molecular Microbiology and Biotechnology 1 (1), 93-100. 2

Bianchi, A., Garcin, J., Tholosan, O., 1999. A high-pressure serial sampler to measure 3

microbial activity in the deep sea. Deep-sea Research I 46 (12), 2129-2142. 4

Bidigare, R.R., Kennicutt, M.C., Brookes, J.M., 1985. Rapid determination of 5

chlorophylls and their degradation products by high performance liquid chromatography. 6

Limnology and Oceanography 30, 432–435. 7

Bidigare, R.R., Ondrusek, M.E., 1996. Spatial and temporal variability of phytoplankton 8

pigment distributions in the central equatorial Pacific Ocean. Deep-Sea Research II 43: 809–9

834. 10

Bidle, K.D., Azam, F., 1999. Accelerated dissolution of diatom silica by marine 11

bacterial assemblages. Nature 397, 508-512. 12

Bidle, K.D., Azam, F., 2001. Bacterial control of silicon regeneration from diatom 13

detritus: significance of bacterial ectohydrolases and species identity. Limnology and 14

Oceanography 46 (7), 1606-1623. 15

Bidle, K.D., Manganelli, M., Azam, F., 2002. Regulation of oceanic silicon and carbon 16

preservation by temperature control on bacteria. Science 298 (5600), 1980-1984. 17

Cho, B.C., Azam, F., 1988. Major role of bacteria in biogeochemical fluxes in the 18

ocean's interior. Nature 332, 441-443. 19

Church, M.J., Delong, E.F., Ducklow, H.W., Karner, M.B., Preston, C.M., Karl, D.M., 20

2003. Abundance and distribution of planktonic Archaea and Bacteria in the waters west of 21

the Antarctic Peninsula. Limnology and Oceanography 48 (5), 1893-1902. 22

Cottrell, M.T., Kirchman, D.L., 2000a. Community composition of marine 23

bacterioplankton determined by 16S rRNA gene clone libraries and fluorescence in situ 24

hybridization. Applied and Environmental Microbiology 66 (12), 5116-5122. 25

Tamburini et al.

30

Cottrell, M.T., Kirchman, D.L., 2000b. Natural assemblages of marine proteobacteria 1

and members of the Cytophaga-Flavobacter cluster consuming low- and high-molecular-2

weight dissolved organic matter. Applied and Environmental Microbiology 66 (4), 1692-3

1697. 4

Daims, H., Brühl, A., Amann, R., Schleifer, K.-H., Wagner, M., 1999. The domain-5

specific probe EUB338 is insufficient for the detection of all Bacteria: development and 6

evaluation of a more comprehensive probe set. Systematic and Applied Microbiology 22, 7

434-444. 8

DeLong, E.F., Franks, D.G., Alldredge, A.L., 1993. Phylogenetic diversity of aggregate-9

attached vs. free-living marine bacteria assemblages. Limnology and Oceanography 38 (5), 10

924-934. 11

DeLong, E.F., Taylor, L.T., Marsh, T.L., Preston, C.M., 1999. Visualization and 12

enumeration of marine planktonic archaea and bacteria by using polyribonucleotide probes 13

and fluorescent in situ hybridization. Applied and Environmental Microbiology 65 (12), 14

5554-5563. 15

DeLong, E.F., Yayanos, A.A., 1985. Adaptation of the membrane lipids of a deep-sea 16

bacterium to changes in hydrostatic pressure. Science 228 (4703), 1101-1113. 17

Deming, J.W., 2001. Unusual or extreme high-pressure marine environments. In: Hurst, 18

C.J., Knudsen, G.R., McInerney, M.J., Stetzenbach, L.D., Walter, M.V. (Eds.), ASM Manual 19

of Environmental Microbiology. ASM press, Washington DC, pp. pp. 478-490. 20

Dugdale, R.C., Wilkerson, F.P., 1998. Silicate regulation of new production in the 21

equatorial Pacific upwelling. Nature 391, 270-273. 22

Engel, A., 2004. Distribution of transparent exopolymer particles (TEP) in the northeast 23

Atlantic Ocean and their potential significance for aggregation processes. Deep Sea Research 24

I 51 (1), 83-92. 25

Tamburini et al.

31

Engel, A., Thoms, S., Riebesell, U., Rochelle-Newall, E., Zondervan, I., 2004. 1

Polysaccharide aggregation as a potential sink of marine dissolved organic carbon. Nature 2

428, 929-932. 3

Fixter, L.M., Nagi, M.N., McCormack, J.G., Fewson, C.A., 1986. Structure, distribution 4

and function of wax esters in Acinetobacter calcoaceticus. Journal of General Microbiology 5

132, 3147–3157. 6

Goutx, M., Acquaviva, M., Bertrand, J.C., 1990. Cellular and extracellular 7

carbohydrates and lipids from marine bacteria during growth on soluble substrates and 8

hydrocarbons. Marine Ecology Progress Series 61, 291-296. 9

Goutx, M., Momzikoff, A., Striby, L., Andersen, V., Marty, J.C., Vescovali, I., 2000. 10

High-frequency fluxes of labile compounds in the central Ligurian Sea, northwestern 11

Mediterranean. Deep-Sea Research I 47, 533-556. 12

Goutx, M., Wakeham, S.G., Lee, C., Duflos, M., Guigue, C., Liu, Z., Moriceau, B., 13

Sempéré, R., Tedetti, M., Xue, J., 2007. Composition and degradation of sinking particles 14

with different settling velocities. Limnology and Oceanography 52, 1645-1664. 15

Hedges, J.I., Baldock, J.A., Gelinas, Y., Lee, C., Peterson, M., Wakeham, S.G., 2001. 16

Evidence for non-selective preservation of organic matter in sinking marine particles. Nature 17

409, 801-804. 18

Hedges, J.I., Eglinton, G., Hatcher, P.G., Kirchman, D.L., Arnosti, C., Derenne, S., 19

Evershed, R.P., Kögel-Knabner, I., de Leeuw, J.W., Littke, R., Michaelis, W., Rullkötter, J., 20

2000. The molecularly-uncharacterized component of nonliving organic matter in natural 21

environments. Organic Geochemistry 31 (10), 945-958. 22

Herndl, G.J., Reinthaler, T., Teira, E., van Aken, H., Veth, C., Pernthaler, A., Pernthaler, 23

J., 2005. Contribution of Archaea to total prokaryotic production in the deep Atlantic Ocean. 24

Applied and Environmental Microbiology 71 (5), 2303-2309. 25

Tamburini et al.

32

Honjo, S., Manganini, S.J., Cole, J.J., 1982. Sedimentation of biogenic matter in the 1

deep ocean. Deep-Sea Research 29, 609−625. 2

Ishige, T., Tani, A., Sakai, Y., Kato, N., 2003. Wax ester production by bacteria. 3

Current Opinion on Microbiology 6 (3), 244-250. 4

Kaneshiro, T., Nakamura, L.K., Nicholson, J.J., Bagby, M.O., 1996. Oleyl oleate and 5

homologous wax esters synthesized coordinately from oleic acid by Acinetobacter and 6

coryneform strains. Current Microbiology 32, 336–342. 7

Karner, M., Fuhrman, J., 1997. Determination of active marine bacterioplankton: a 8

comparison of universal 16S rRNA probes, autoradiography, and nucleoid staining. Applied 9

and Environmental Microbiology 63 (4), 1208-1213. 10

Karner, M.B., DeLong, E.F., Karl, D.M., 2001. Archaeal dominance in the mesopelagic 11

zone of the Pacific Ocean. Nature 409 (6819), 507-510. 12

Kato, C., Li, L., Nogi, Y., Nakamura, Y., Tamaoka, J., Horikoshi, K., 1998. Extremely 13

barophilic bacteria isolated from the Mariana Trench, Challenger Deep, at a depth of 11,000 14

meters. Applied and Environmental Microbiology 64 (4), 1510-1513. 15

King, L. 1993. Chlorophyll diagenesis in the water column and sediments of the Black 16

Sea. Ph.D. Thesis, W.H.O.I./M.I.T. Joint Program. 17

Kiørboe, T., Jackson, G.A., 2001. Marine snow, organic solute plumes, and optimal 18

chemosensory behavior of bacteria. Limnology and Oceanography 46 (6), 1309-1318. 19

Kirchman, D.L., 2002. The ecology of Cytophaga-Flavobacteria in aquatic 20

environments. FEMS Microbiology Ecology 39, 91-100. 21

Lampitt, R.S., 1992. The contribution of deep-sea macroplancton to organic 22

remineralization: results from sediment trap and zooplankton studies over the Madeira abyssal 23

plain. Deep Sea Research 49, 221-233. 24

Tamburini et al.

33

Langdon, C., 1984. Dissolved oxygen monitoring system using a pulsed electrode: 1

design, performance, and evaluation. Deep-Sea Research 31 (11), 1357-1367. 2

Lee, C., 2004. Transformations in the "Twilight Zone" and beyond. Marine Chemistry 3

92 (1-4), 87-90. 4

Lee, C., Cronin, C., 1982. The vertical flux of particulate organic nitrogen in the sea: 5

Decomposition of amino acids in the Peru upwelling area and the equatorial Atlantic. Journal 6

of Marine Research 41, 227–251. 7

Lee, C., Wakeham, S.G., Hedges, J.I., 2000. Composition and flux of particulate amino 8

acids and chloropigments in equatorial Pacific seawater and sediments. Deep-Sea Research I 9

47, 1535-1568. 10

Lee, C., Wakeham, S.G., Peterson, M.L., Cochran, J.K., Miquel, J.C., Armstrong, R.A., 11

Fowler, S., Hirschberg, D., Beck, A., Xue, J., 2008. Particulate matter fluxes in time-series 12

and settling velocity sediment traps in the northwestern Mediterranean Sea. Deep-Sea 13

Research II. this issue. 14

Lee, R.F., Hirota, J., 1973. Wax esters in tropical zooplankton and nekton and the 15

geographical distribution of wax esters in marine copepods. Limnology and Oceanography 16

18, 227-239. 17

Lochte, K., Turley, C.M., 1988. Bacteria and cyanobacteria associated with 18

phytodetritus in the deep sea. Nature 333, 67-69. 19

Long, R.A., Azam, F., 2001. Microscale patchiness of bacterioplankton assemblage 20

richness in seawater. Aquatic Microbial Ecology 206 (2), 103-113. 21

Malmstrom, R.R., Cottrell, M.T., Elifantz, H., Kirchman, D.L., 2005. Biomass 22

Production and Assimilation of Dissolved Organic Matter by SAR11 Bacteria in the 23

Northwest Atlantic Ocean. Applied and Environmental Microbiology 71 (6), 2979-2986. 24

Tamburini et al.

34

Mantoura, R.F.C., Llewellyn, C.A., 1983. The rapid determination of algal chlorophyll 1

and carotenoid pigments and their breakdown products in natural waters by reverse phase 2

high performance liquid chromatography. Analytical and Chemical Acta 151, 297–314. 3

Manz, W., Amann, R., Ludwig, W., Vancanneyt, M., Schleifer, K.H., 1996. Application 4

of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the 5

phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology 142 ( Pt 6

5), 1097-1106. 7

Manz, W., Amann, R., Ludwig, W., Wagner, M., Scheiler, K.-H., 1992. Phylogenetic 8

oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and 9

solution. Systematic and Applied Microbiology 15, 593-600. 10

McCave, I.N., 1975. Vertical flux of particles in the Ocean. Deep-Sea Research 22, 491-11

502. 12

Milliman, J.D., Troy, P.J., Balch, W.M., Adams, A.K., Li, Y.-H., Mackenzie, F.T., 13

1999. Biologically mediated dissolution of calcium carbonate above the chemical lysocline? 14

Deep-Sea Research I 46 (10), 1653-1669. 15

Moeseneder, M.M., Winter, C., Herndl, G.J., 2001. Horizontal and vertical complexity 16

of attached and free-living bacteria of the eastern Mediterranean Sea, determined by 16S 17

rDNA and 16S rRNA fingerprints. Limnology and Oceanography 46 (1), 95-107. 18

Mopper, K., Schultz, C.A., Chevolot, L., Germah, C., Revuelta, R., Dawsong, R., 1992. 19

Determination of sugars in unconcentrated seawater and other natural waters by liquid 20

chromatography and pulsed amperometric detection. Environmental Science and Technology 21

26, 133-137. 22

Nelson, D.M., Tréguer, P., Brzezinski, M.A., Leynaert, A., Quéguiner, B., 1995. 23

Production and dissolution of biogenic silica in the ocean: revised global estimates, 24

Tamburini et al.

35

comparison with regional data and relationship with biogenic sedimentation. Global 1

Biogeochemical Cycles 9, 359-372. 2

Nogi, Y., Masui, N., Kato, C., 1998. Photobacterium profundum sp. nov., a new, 3

moderately barophilic bacterial species isolated from a deep-sea sediment. Extremophiles 2 4

(1), 1-7. 5

Panagiotopoulos, C., Sempéré, R., 2005. Analytical methods for the determination of 6

sugars in marinesamples: A historical perspective and future directions. Limnology and 7

Oceanography: Methods 3, 419-454. 8

Panagiotopoulos, C., Sempéré, R., Obernosterer, I., Striby, L., Goutx, M., Van 9

Wambeke, F., Gautier, S., Lafont, R., 2002. Bacterial degradation of large particles in the 10

southern Indian Ocean using in vitro incubation experiments. Organic Geochemistry 33, 985-11

1000. 12

Passow, U., Alldredge, A.L., 1995. A dye-binding assay for the spectrophotometric 13

measurement of transparent exopolymer particles (TEP). Limnology and Oceanography 40 14

(7), 1326-1335. 15

Passow, U., Engel, A., Ploug, H., 2003. The role of aggregation for the dissolution of 16

diatom frustules. FEMS Microbiology Ecology 46 (3), 247-255. 17

Pernthaler, A., Pernthaler, J., Amann, R., 2002. Fluorescence in situ hybridization and 18

catalyzed reporter deposition for the identification of marine bacteria. Applied and 19

Environmental Microbiology 68 (6), 3094-3101. 20

Peterson, M.L., Fabres, J., Wakeham, S.G., Lee, C., Miquel, J.C., 2008. Sampling the 21

vertical particle flux in the upper water column using a large diameter free-drifting NetTrap 22

adapted to an Indented Rotating Sphere Settling Velocity trap. . Deep-Sea Research II. This 23

volume. 24

Tamburini et al.

36

Peterson, M.L., Wakeham, S.G., Lee, C., Askea, M.A., Miquel, J.C., 2005. Novel 1

techniques for collection of sinking particles in the ocean and determining their settling rates. 2

Limonology and Oceanography: Methods 3, 520-532. 3

Ploug, H., Grossart, H.-P., 1999. Bacterial production and respiration in suspended 4

aggregates-a matter of the incubation method. Aquatic Microbial Ecology 20 (1), 21-29. 5

Porter, K.G., Feig, Y.S., 1980. The use of DAPI for identifying and counting aquatic 6

microflora. Limnology and Oceanography 25, 943-948. 7

Rontani, J.F., Bonin, P.C., Volkman, J.K., 1999. Production of wax esters during 8

aerobic growth of marine bacteria on isoprenoid compounds. Applied and Environmental 9

Microbiology 65 (1), 221-230. 10

Russell, N.J., Volkman, J.K., 1980. The effect of growth temperature on wax ester 11

composition in the psychrophilic bacterium Micrococcus cryophilis ATCC 15174. Journal of 12

General Microbiology 118, 131–141. 13

Sekar, R., Pernthaler, A., Pernthaler, J., Warnecke, F., Posch, T., Amann, R., 2003. An 14

improved protocol for quantification of freshwater Actinobacteria by fluorescence in situ 15

hybridization. Applied and Environmental Microbiology 69 (5), 2928-2935. 16

Sempéré, R., Tedetti, M., Charrière, B., Duhamel, S., Raimbault, P., Van Wambeke, F., 17

2007. Molecular distribution and bacterial availability of dissolved sugars in the South East 18

Pacific Ocean in relation to photochemical oxidation reactions Biogeosciences, in prep. 19

Sempéré, R., Yoro, S.O., Van Wambeke, F., Charrière, B., 2000. Microbial 20

decomposition of large organic particles in the northwestern Mediterranean Sea: an 21

experimental approach. Marine Ecology Progress Series 198, 61-72. 22

Shanks, A.L., Trent, J.D., 1980. Marine snow: sinking rates and potential role in 23

vertical flux. Deep-Sea Research 27A, 137-143. 24

Tamburini et al.

37

Shuman, F.R., Lorenzen, C.J., 1975. Quantitative degradation of chlorophyll by a 1

marine herbivore. Limnology and Oceanography 20, 580-586. 2

Silva, R.A., Grossi, V., Alvarez, H.M., 2007. Biodegradation of phytane (2,6,10,14-3

tetramethylhexadecane) and accumulation of related isoprenoid wax esters by Mycobacterium 4

ratisbonense strain SD4 under nitrogen-starved conditions. FEMS Microbiology Letters 272 5

(2), 220-228. 6

Simon, M., Grossart, H.-P., Schweitzer, B., Ploug, H., 2002. Microbial ecology of 7

organic aggregates in aquatic ecosystems. Aquatic Microbial Ecology 28, 175-211. 8

Sinninghe-Damste, J.S., Rijpstra, W.I.C., Hopmans, E.C., Prahl, F.G., Wakeham, S.G., 9

Schouten, S., 2002. Distribution of membrane lipids of planktonic Crenarchaeota in the 10

Arabian Sea. Applied and Environmental Microbiology 68 (6), 2997-3002. 11

Smayda, T.J., 1970. The suspension and sinking of phytoplankton in the sea. 12

Oceanography and Marine Biology Annual Reviews 8, 353-414. 13

Sohrin, R., Sempéré, R., 2005. Seasonal variation in total organic carbon in the 14

Northeast Atlantic in 2000-2001. Journal of Geophysical Reseach 110 (C10S90), 15

doi:10.1029/2004JC002731 16

Tamburini, C., 2006. Life under pressure. Deep-sea microbial ecology. Life as we know 17

it. Series: Cellular Origin and Life in Extreme Habitats and Astrobiology. Springer, 18

Dordrecht, The Netherlands pp. 125-143. 19

Tamburini, C., Garcin, J., Bianchi, A., 2003. Role of deep-sea bacteria in organic matter 20

mineralization and adaptation to hydrostatic pressure conditions in the NW Mediterranean 21

Sea. Aquatic Microbial Ecology 32 (3), 209-218. 22

Tamburini, C., Garcin, J., Grégori, G., Leblanc, K., Rimmelin, P., Kirchman, D.L., 23

2006. Pressure effects on surface Mediterranean prokaryotes and biogenic silica dissolution 24

during a diatom sinking experiment. Aquatic Microbial Ecology 43 (3), 267-276. 25

Tamburini et al.

38

Tamburini, C., Garel, M., Al Ali, B., Mérigot, B., Kriwy, P., Charrière, B., Budillon, G., 1

2008. Distribution and activity of Bacteria and Archaea in the different water masses of the 2

Tyrrhenian Sea. Deep-Sea Research II in revision. 3

Tan, F.C., Strain, P.M., 1979. Organic Carbon Isotope Ratio in Recent Sediments in the 4

St-Lawrence Estuary and the Gulf of the St-Lawrence. Estuarine and Coastal Marine Science 5

8, 213-225. 6

Teira, E., Reinthaler, T., Pernthaler, A., Pernthaler, J., Herndl, G.J., 2004. Combining 7

catalyzed reporter deposition-fluorescence in situ hybridization and microautoradiography to 8

detect substrate utilization by bacteria and Archaea in the deep ocean. Applied and 9

Environmental Microbiology 70 (7), 4411-4414. 10

Tréguer, P., Kamatani, A., Gueneley, S., Quéguiner, B., 1989. Kinetics of dissolution of 11

Antarctic diatom frustules and the biogeochemical cycle of silicon in the Southern Ocean. 12

Polar Biology 9, 397-403. 13

Turley, C.M., 1993. The effect of pressure on leucine and thymidine incorporation by 14

free-living bacteria and by bacteria attached to sinking oceanic particles. Deep-Sea Research I 15

40 (11-12), 2193-2206. 16

Turley, C.M., Mackie, P.J., 1994. Biogeochemical significance of attached and free-17

living bacteria and the flux of particles in the NE Atlantic Ocean. Marine Ecology Progress 18

Series 115, 191-203. 19

Turley, C.M., Mackie, P.J., 1995. Bacterial and cyanobacterial flux to the deep NE 20

Atlantic on sedimenting particles. Deep Sea Research I 42 (8), 1453-1474. 21

Van Wambeke, F., Goutx, M., Striby, L., Sempéré, R., Vidussi, F., 2001. Bacterial 22

dynamics during the transition from spring bloom to oligotrophy in the nothwestern 23

Mediterranean Sea: relationships with detritus and dissolved organic matter. Marine Ecology 24

Progress Series 212, 89-105. 25

Tamburini et al.

39

Wakeham, S.G., Farrington, J.W., Volkman, J.K., 1983. Fatty acids, wax esters, 1

triacylglycerols and alkyldiacylglycerols associated with particles collected in sediment traps 2

in the Peru upwelling. In: Bjorøy, M., et al. (Ed.), Advances in Organic Geochemistry, 3

Chichester, John Wiley and Sons, pp. pp. 185–197. 4

Wakeham, S.G., Lee, C., 1993. Production, transport, and alteration of particulate 5

organic matter in the marine water column. In: M. Engel and S. Macko, e. (Ed.), Organic 6

Geochemistry. Plenum Press, New York, pp. 145-169. 7

Wakeham, S.G., Lee, C., Hedges, J.I., Hernes, P.J., Peterson, M.L., 1997. Molecular 8

indicators of diagenetic status in marine organic matter. Geochimica et Cosmochimica Acta 9

61 (24), 5363-5369. 10

Wakeham, S.G., Lee, C., Peterson, M.L., Liu, Z., Szlosek, J., Putnam, I.F., Xue, J.,. 11

2008. Organic compound composition and fluxes in the Twilight zone – Time series and 12

settling velocity sediment traps during MedFlux. Deep-Sea Research II, this volume. 13

Yayanos, A.A., 1995. Microbiology to 10,500 meters in the deep-sea. Annual Review of 14

Microbiology 49 (1), 777-805. 15

Ziegler, R., Blaheta, A., Guha, N., Schönegge, B., 1988. Enzymatic formation of 16

pheophorbide and pyropheophorbide during chlorophyll degradation in a mutant of Chlorella 17

fusca SHIHIRA et KRAUS. Plant Physiology 132, 327-332. 18

19

20

Tamburini et al.

40

Figure captions 1

Fig. 1. Schematic diagram of the PArticle Sinking Simulator (PASS). PASS is composed of 2

(1) several water baths where high-pressure bottles (HPB) are rotated to maintain 3

particles in suspension, (2) a cooler to control the temperature of water baths and (3) a 4

programmable computer-driven: piloted pressure generator. In this study, 3 HPBs were 5

maintained at atmospheric pressure (ATM), while 3 others were connected to the piloted 6

pressure generator, which continuously increased the hydrostatic pressure to simulate fall 7

through the water column (HP). See details in the text. 8

Fig. 2. Photograph and schematic diagram of the high-pressure bottles rotation system. 9

Fig. 3. Pressure and temperature data measured during the incubation of freshly recovered 10

sinking particles to simulate increase of pressure and variation of temperature through the 11

water column (between 200 and 1500m depth). Temperatures were chosen according to 12

the CTD profile obtained during the MedFlux III cruise in April 2006. 13

Fig. 4. Percent of (a) total dissolved carbohydrates (DCHO-C) versus dissolved organic 14

carbon (DOC) ; (b) total particulate carbohydrates (PCHOtot-C) versus particulate organic 15

carbon (POC) ; (c) transparent exopolymers particles (TEP) versus POC ; (d) total 16

particulate hydrolyzed amino acids (PHAAtot) versus POC ; (e) total particulate pigments 17

(PPigtot) versus POC ; (f) total particulate lipids (PLiptot) versus POC and (g) wax ester 18

(WE) versus POC initially (T0) and after 156 hours of incubation under increasing 19

pressure (HP-1, HP-2 and HP-3) and under atmospheric pressure (ATM-1, ATM-2 and 20

ATM-3). Error bars represent experimental errors based on duplicates. 21

Fig. 5. (a) Abundance of total, attached (fraction>2µm) and free-living (fraction<2µm) 22

prokaryotes (DAPI counts) initially (T0) and after 156 hours of incubation under 23

increasing pressure (HP-1, HP-2 & HP-3) and under atmospheric pressure ( ATM-1, 24

Tamburini et al.

41

ATM-2 & ATM-3). (b) Percent of total prokaryotes (DAPI stained cells) detected by 1

catalyzed reporter deposition coupled with fluorescence in situ hybridization (CARD-2

FISH) initially (T0) and after 156 hours of incubation under increasing pressure (HP-1, 3

HP-2 & HP-3) and under atmospheric pressure (ATM-1, ATM-2 & ATM-3). Eub mix 4

(Eub338, Eub-II, Eub-III): Bacteria ; CF319a: Cytophaga-Flavobacter cluster of the 5

Cytophaga-Flavobacter-Bactroides division ; Alf968, Gam42a:α-subclass and γ-subclass 6

of Proteobacteria ; Cren537: Crenarchaea ; Eury806: Euryarchaea. 7

8

9

Tamburini et al.

42

Table 1. 16S rRNA-targeted oligonucleotide probes used in this study.

Probe Sequence (5' to 3') of probe Target organisms % formamide References

Eub338/I GCT GCC TCC CGT AGG AGT Domain of Bacteria 55 (Amann et al., 1990)

EubII GCA GCC ACC CGT AGG TGT Domain of Bacteria 55 (Daims et al., 1999)

EubIII GCT GCC ACC CGT AGG TGT Domain of Bacteria 55 (Daims et al., 1999)

ALF968 GGT AAG GTT CTG CGC GTT Most of α-subclass of Proteobacteria 55 (Manz et al., 1992)

BET42aa GCC TTC CCA CTT CGT TT γ-subclass of Proteobacteria 55 (Manz et al., 1992)

GAM42ab GCC TTC CCA CAT CGT TT β-subclass of Proteobacteria 55 (Manz et al., 1992)

CF319a TGG TCC GTG TCT CAG TAC Cytophaga-Flavobacter cluster 55 (Manz et al., 1996)

Cren537 TGA CCA CTT GAG GTG CTG Crenarchaea 20 (Teira et al., 2004)

Eury806 CAC AGC GTT TAC ACC TAG Euryarchaea 20 (Teira et al., 2004)

NegControl TAG TGA CGC GCT CGA For non-specific probe binding 55 (Karner and Fuhrman, 1997) aIncluding an unlabeled competitor probe Gam42a (5’- GCCTTCCCACATCGTTT-3’), see Manz et al. (1992) for details bIncluding an unlabeled competitor probe BET42a (5’-GCCTTCCCACTTCGTTT-3’), see Manz et al. (1992) for details

Tamburini et al.

43

Table 2. Organic matter composition initially (T0) and after 156 hours of incubation under increasing pressure regime (HP-1, HP-2 and HP-3)

and under atmospheric pressure (ATM-1, ATM-2 and ATM-3). Abbreviations : total organic carbon (TOC as the sum of DOC and POC), dissolved organic carbon (DOC), total dissolved carbohydrates (DCHOtot), particulate organic carbon (POC), total particulate hydrolyzed amino acids (PHAAtot), total particulate pigments (PPigtot), total particulate lipids (PLiptot) and total particulate carbohydrates (PCHOtot). ±: standard error of measurements, s.d; standard deviation (n=3).

Dissolved Particulate Sample TOC

(µM) DOC (µM)

DCHOtot (µM-C)

POC (µM)

PHAAtot (µM-C)

PPigtot (µM-C)

PLiptot (µM-C)

PCHOtot (µM-C)

T0 88.1±1.8 54.8±1.7 0.27±0.02 33.3±0.7 4.3±0.4 0.64±0.06 4.0±0.2 3.4±0.2

HP-1 94.1±1.2 56.0±0.9 0.91±0.06 38.1±0.8 5.0±0.5 0.41±0.04 7.9±3.4 2.5±0.2

HP-2 113.0±2.8 82.3±2.7 0.50±0.04 30.8±0.6 4.8±0.5 0.29±0.03 7.2±3.4 1.7±0.1

HP-3 124.5±1.9 75.0±1.7 0.90±0.07 49.5±1.0 14.4±1.4 0.25±0.02 10.4±3.4 2.4±0.2

Mean HP 110.5 71.1 0.79 39.5 8.1 0.32 8.5 2.2

s.d. 16.5 13.6 0.25 9.4 5.5 0.08 1.7 0.4

ATM-1 148.5±3.1 65.6±2.6 7.1±0.5 82.9±1.7 7.9±0.8 0.22±0.02 19.0±1.2 3.1±0.2

ATM-2 94.3±6.4 52.9±6.3 8.2±0.6 41.4±0.8 5.8±0.6 0.22±0.02 8.4±1.2 1.3±0.1

ATM-3 135.2±12.9 74.7±12.9 3.8±0.3 60.5±1.2 10.4±1.0 0.33±0.03 15.4±1.2 2.1±0.1

Mean ATM 126.0 64.4 6.4 61.6 8.1 0.26 14.2 2.2

s.d. 23.5 10.9 2.3 20.7 2.3 0.06 5.4 0.9

Tamburini et al.

44

Table 3. Percentage of particulate individual amino acids (mole% C) in total particulate hydrolyzed amino acids initially (T0) and after 156 hours of incubation under increasing pressure regime (HP-1, HP-2 and HP-3) and under atmospheric pressure (ATM-1, ATM-2 and ATM-3). Abbreviations: aspartic acid (ASP), glutamic acid (GLU), serine (SER), histidine (HIS), glucosamine (GCSM), glycine (GLY), threonine (THR), arginine (ARG), β-alanine (BALA), alanine (ALA), γ-aminobutyric acid (GABA), methionine (MET), valine (VAL), phenylalanine (PHE), isoleucine (ILE), leucine (LEU) and lysine (LYS).

Sample ASP GLU SER HIS GCSM GLY THR ARG BALA ALA GABA TYR MET VAL PHE ILE LEU LYS

T0 5.0 11.5 6.6 1.7 8.2 8.3 0.4 6.7 3.1 6.1 0.3 6.4 0.3 6.3 7.1 5.9 9.4 6.5

HP-1 3.8 9.8 6.8 0.9 5.5 7.6 0.3 7.0 5.3 7.6 0.3 7.2 0.4 6.4 8.0 6.3 10.6 6.1

HP-2 5.7 14.2 6.0 2.2 7.1 5.5 0.4 6.8 1.1 8.0 0.9 8.5 0.2 6.6 8.6 4.7 9.7 3.8

HP-3 5.0 15.7 5.1 2.0 3.6 5.2 0.4 5.7 0.7 7.5 0.1 6.2 0.3 7.4 8.3 6.6 10.7 9.6

Mean HP 4.8 13.2 6.0 1.7 5.4 6.1 0.4 6.5 2.4 7.7 0.4 7.3 0.3 6.8 8.3 5.9 10.3 6.5

s.d. 1.0 3.1 0.9 0.7 1.8 1.3 0.1 0.7 2.5 0.3 0.4 1.2 0.1 0.5 0.3 1.0 0.6 2.9

ATM-1 5.0 14.1 5.8 1.4 4.1 6.2 0.3 6.7 2.6 6.0 0.3 6.6 0.3 7.0 8.2 6.8 11.2 7.3

ATM-2 5.5 11.2 5.6 1.5 4.9 5.3 0.4 6.3 3.3 6.7 0.2 7.2 0.7 7.2 8.2 6.7 10.6 8.6

ATM-3 5.1 10.8 5.8 1.9 4.9 5.9 0.4 6.2 0.8 7.6 0.1 6.5 0.2 6.7 8.0 6.8 11.5 10.9

Mean ATM 5.2 12.0 5.7 1.6 4.6 5.8 0.4 6.4 2.2 6.8 0.2 6.8 0.4 7.0 8.1 6.8 11.1 8.9

s.d. 0.3 1.8 0.1 0.3 0.5 0.5 0.1 0.3 1.3 0.8 0.1 0.4 0.3 0.3 0.1 0.1 0.5 1.8

Tamburini et al.

45

Table 4. Percentage of particulate individual carbohydrates (mole% C) in total particulate carbohydrates initially (T0) and after 156 hours of incubation under increasing pressure regime (HP-1, HP-2 and HP-3) and under atmospheric pressure (ATM-1, ATM-2 and ATM-3). Abbreviations: fucose (fuc), rhamnose (rha), arabinose (ara), galactosamine (galaN), glucosamine (glcN), galactose (gala), glucose (glu), mannose (man), xylose (xyl), fructose (fru) and ribose (rib).

Sample fuc rha ara galaN glcN gala glu man xyl fru rib

T0 5.6 25.0 3.2 0.0 1.9 24.1 15.6 8.4 11.0 2.0 3.2

HP-1 6.4 25.8 1.4 0.0 18.3 19.6 5.0 9.4 12.4 1.7 0.0

HP-2 6.2 22.5 0.0 0.0 0.0 18.6 18.0 5.0 9.6 20.2 0.0

HP-3 6.5 14.5 5.0 0.0 6.5 12.8 20.6 6.9 14.9 5.0 7.3

Mean HP 6.4 20.9 2.1 0.0 8.3 17.0 14.5 7.1 12.3 9.0 2.4

s.d. 0.2 5.8 2.6 0.0 9.3 3.7 8.4 2.2 2.7 9.9 4.2

ATM-1 8.6 17.1 0.2 0.0 2.2 16.3 19.2 7.4 13.4 11.1 4.5

ATM-2 6.7 23.0 0.0 0.0 0.0 16.2 12.3 4.6 11.0 16.7 9.6

ATM-3 7.0 17.5 3.4 0.0 4.4 17.1 16.7 7.2 14.4 6.3 6.0

Mean ATM 7.4 19.2 1.2 0.0 2.2 16.5 16.1 6.4 12.9 11.4 6.7

s.d. 1.0 3.3 1.9 0.0 2.2 0.5 3.5 1.6 1.7 5.2 2.6

Tamburini et al.

46

Table 5. Percentage of particulate individual lipid classes (mole% C) in total particulate lipids, and fatty acid classes as a percent of total fatty acids (mole% C) initially (T0) and after 156 hours of incubation under increasing pressure regime (HP-1, HP-2 and HP-3) and under atmospheric pressure (ATM-1, ATM-2 and ATM-3). Abbreviations: wax and steryl esters (WSE), triacylglycerols (TG), free fatty acids (FFA), alcohols (ALC), 1,3-diglycerides (DG), sterols (ST), 1,2-diglycerides (DG), monoglycerides (MG), chloroplast lipids (CL), phosphatidylglycerides (PG), phosphatidylethanolamines (PE), saturated fatty acids (FAsat), monounsaturated fatty acids (FAmonounsat), polyunsaturated fatty acids (PUFA).

Sample WSE TG FFA ALC 1,3DG ST 1,2DG MG CL PG PE FAsat FAmonounsat PUFA

T0 6.4 11.3 13.9 11.4 0.0 1.9 4.6 9.1 18.0 12.2 11.2 74.4 19.0 6.4

HP-1 6.8 14.9 18.8 12.6 0.0 7.8 0.0 4.6 23.5 7.5 3.5 81.4 15.6 2.8

HP-2 9.1 8.9 10.6 18.1 0.0 12.2 0.0 6.6 27.3 3.6 3.6 85.8 12.0 2.0

HP-3 9.0 31.8 10.9 7.7 0.0 6.8 3.1 5.6 12.1 5.8 7.2 75.8 20.0 4.1

Mean HP 8.3 18.5 13.4 12.8 0.0 8.9 1.0 5.6 21.0 5.6 4.8 81.0 15.9 3.0

s.d. 1.3 11.9 4.7 5.2 0.0 2.9 1.8 1.0 7.9 2.0 2.1 5.0 4.0 1.1

ATM-1 1.6 0.0 25.8 5.3 5.3 6.7 10.7 15.0 14.7 12.4 2.5 85.4 12.1 2.3

ATM-2 3.2 38.0 7.8 5.4 0.0 6.1 0.0 4.1 32.9 2.6 0.0 68.5 26.2 4.5

ATM-3 0.0 50.8 6.2 4.4 0.0 4.1 0.0 5.1 17.5 7.1 4.8 75.3 19.6 4.9

Mean ATM 1.6 29.6 13.3 5.0 1.8 5.6 3.6 8.1 21.7 7.4 2.4 76.4 19.3 3.9

s.d. 1.6 26.4 10.9 0.6 3.1 1.4 6.2 6.0 9.8 4.9 2.4 8.5 7.1 1.4

Tamburini et al.

47

Table 6. Percentage of individual pigments (mole% C) in total particulate pigments initially (T0) and after 156 hours of incubation under increasing pressure regime (HP-1, HP-2 and HP-3) and under atmospheric pressure (ATM-1, ATM-2 and ATM-3). Abbreviations: Chlorophyll a (Chl. a), Pheophorbide (Phide), Pyropheophorbide (Pyrophide), Pheophytin (Phytin) and Fucoxanthin (Fuco). Sample Chl a Phide Pyrophide Phytin Fuco

T0 0.31 2.18 94.70 0.71 2.10

HP-1 0.38 2.65 92.25 3.39 1.32

HP-2 0.31 2.72 94.50 1.53 0.94

HP-3 0.43 3.31 92.46 2.59 1.21

Mean HP 0.4 2.9 93.1 2.5 1.2

s.d. 0.1 0.4 1.2 0.9 0.2

ATM-1 0.55 4.13 91.65 1.54 2.13

ATM-2 0.62 3.86 93.18 0.43 1.91

ATM-3 0.48 3.40 90.20 2.67 3.24

Mean ATM 0.6 3.8 91.7 1.5 2.4

s.d. 0.1 0.4 1.5 1.1 0.7

Tamburini et al.

Figures 1

Fig. 1 2

3) Piloted pressure generator

1) HPBs fitted within the revolving system incubated in water bath

2) Temperature cooler

ATM HP

3

Tamburini et al.

1

Fig. 2 2

3 4 5 6

Drivingsystem

Stainlesssteel frame

Framefor HPBs

Drivingsystem

Stainlesssteel frame

Framefor HPBs

7 8

Tamburini et al.

1 2 3

Fig. 3 4

5

6

Tamburini et al.

0

2

4

6

8

10

12

14

16

T0 HP-1 HP-2 HP-3 ATM-1 ATM-2 ATM-3

DC

HO

-C/D

OC

(%)

0

2

4

6

8

10

12

14

16

T0 HP-1 HP-2 HP-3 ATM-1 ATM-2 ATM-3

PCH

O/P

OC

(%)

a)

b)

c)

0

20

40

60

80

100

120

140

TEP

/ PO

C

T0 HP-1 HP-2 HP-3 ATM-1 ATM-2 ATM-3

05

101520253035

T0 HP-1 HP-2 HP-3 ATM1 ATM2ATM3

PHAA

/PO

C (%

)

0,0

0,5

1,0

1,5

2,0

2,5

T0 HP-1 HP-2 HP-3 ATM1 ATM2 ATM3

PPig

/PO

C (%

)

d)

e)

f)

0

5

10

15

20

25

30

t0 HP-1 HP-2 HP-3 ATM1 ATM2 ATM3

PLip

/PO

C (%

)

0,0

0,5

1,0

1,5

2,0

2,5

3,0

T0 HP-1 HP-2 HP-3 ATM1 ATM2 ATM3

WE/

POC

(%)

g)

1

Fig. 4 Tamburini et al.

106

cells

ml-1

Total f ractionFraction <2µmFraction >2µm

a)

Eub mixCF319aGam42aAlf968CREN537EURY806

0

10

20

30

40

50

60

70

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

T0 HP-1 HP-2 HP-3 ATM-1 ATM-2 ATM-3

80

90

100

T0 HP-1 HP-2 HP-3 ATM-1 ATM-2 ATM-3

Per

cent

of p

roka

ryot

es

b)

1

Fig. 5 Tamburini et al.