Nucleic acid concentrations (DNA, RNA) in the continental and deep-sea sediments of the eastern Mediterranean: relationships with seasonally varying organic inputs and bacterial dynamics

*Corresponding author. Fax: 0039 71 2264650; e-mail: [email protected]

Deep-Sea Research I 46 (1999) 1077}1094

Nucleic acid concentrations (DNA, RNA) in thecontinental and deep-sea sediments of the eastern

Mediterranean: relationships with seasonallyvarying organic inputs and bacterial dynamics

R. Danovaro*,!, A. Dell'anno!, A. Pusceddu!, M. Fabiano"

!Istituto Scienze del Mare, Facolta% di Scienze, Universita% di Ancona, Via Brecce Bianche, 60131, Ancona, Italy"Istituto Scienze Ambientali Marine, Universita% di Genova, Genova, Italy

Received 2 January 1998; received in revised form 11 June 1998; accepted 11 September 1998

Abstract

In order to study temporal variations of the genetic material in the continental shelf anddeep-sea sediments of the extremely oligotrophic Cretan Sea, samples were collected on seasonalbasis from August 1994 to September 1995, with a multiple corer, at seven stations (from 40 to1540 m depth). Surface sediments (0}1 cm) were sub-sampled and analyzed for nucleic acidcontent (DNA, RNA) and bacterial density. DNA concentrations in the sediments were high(on annual average, 25.0 lg g~1) and declined with increasing water depth, ranging from 3.5 to55.2 lg g~1. DNA concentrations displayed wide temporal changes also at bathyal depthscon"rming the recent view of the large variability of the deep-sea environments. Also RNAconcentrations decreased with increasing water depth (range: 0.4}29.9 lg g~1). The ratio ofRNA to DNA did not show a clear spatial pattern but was characterized by signi"cant changesbetween sampling periods. DNA concentrations were signi"cantly correlated with protein andphytopigment concentrations in the sediment, indicating a possible relationship with the inputsof primary organic matter from the photic layer. Bacterial densities were generally high (range:0.9}4.6]108 cells g~1) compared to other deep-sea environments and decreased with increas-ing water depth. Estimates of the bacterial contribution to the sedimentary genetic materialindicated that bacterial-DNA accounted, on annual average, for a small fraction of the totalDNA pool (4.3%) but that bacterial-RNA represented a signi"cant fraction of the totalsedimentary RNA (26%). Bacterial contribution to nucleic acids increased, even thoughirregularly, with increasing depth. In deep-sea sediments, changes in RNA concentrationsappear to be largely dependent upon bacterial dynamics. Estimates of the overall livingcontribution to the DNA pools (i.e. microbial plus meiofaunal DNA) indicated that the large

0967-0637/99/$ } see front matter ( 1999 Elsevier Science Ltd. All rights reserved.PII: S 0 9 6 7 - 0 6 3 7 ( 9 8 ) 0 0 1 0 1 - 0

majority (about 90%) of the DNA in continental and deep-sea sediments of the easternMediterranean was detrital. The non-living DNA pools reach extremely high concentrations upto 0.41 g DNAm~2 cm~1. Thus, especially in deep benthic habitats, characterized by low inputsof labile organic compounds, detrital DNA could represent a suitable and high quality foodsource or a signi"cant reservoir of nucleic acid precursors for benthic metabolism. ( 1999Elsevier Science Ltd. All rights received.

Keywords: Nucleic acids; Deep-sea sediments; Organic matter; Benthic bacteria

1. Introduction

DNA pools in marine environments contain large amounts of non-living DNA. The"rst evidence of the mostly detrital nature of nucleic acids was provided by Holm-Hansen et al. (1968), who utilized particulate DNA concentrations as a possiblemeasure of the living planktonic biomass but concluded that only about 40% of thetotal particulate DNA pool was associated with living cells. Similar results wereobtained from studies carried out in di!erent marine systems, demonstrating that upto 90% of the particulate nucleic acid pool was associated with dead or dormant cellsor adsorbed to the particulate material (Winn and Karl, 1986). These results clearlydemonstrated that DNA concentrations can not be utilized as a measure of the livingbiomass of marine organisms.

Large amounts of particulate detrital DNA produced in the photic layer reach thebottom, providing inputs of allochthonous genetic material to deep-sea sediments(Baili! and Karl, 1991; Turley and Mackie, 1995). DNA supply to the bottom islargely represented by DNA-containing detrital particles that are rapidly colonized bydeep-sea bacteria (Lochte and Turley, 1988; Turley and Lochte, 1990). This &&rain'' ofparticulate DNA could represent an important source of labile organic N and P ornucleic acid precursors, which are energetically expensive to synthesize ex novo (Paulet al., 1987).

In marine sediments bacterial biomasses generally account for most of the livingbenthic biomass and their importance increases with depth (up to 90%; Pfannkuche,1993; Danovaro et al., in press-c). Due to their large biomass bacteria might contrib-ute signi"cantly to the living DNA pools in the sediment but, at the same time, canutilize detrital DNA as trophic source for their growth (Jorgesen and Jacobsen, 1996).Estimates of the bacterial contribution to the total DNA pool, together with theprotozoan and metazoan DNA, are essential to assess the actual detrital DNAfraction. In this regard, in the deep-sea sediments of the eastern Mediterranean,Danovaro et al. (1993) found that most of the sedimentary DNA ('90%) wasunaccounted from by bacterial standing stocks. Bacterial distribution was signi"-cantly correlated with nucleic acid concentrations, thus indicating that DNA poolsmay be a factor controlling bacterial distribution (Danovaro et al., 1993), but thequantitative role, ecological signi"cance and nature of the sedimentary DNA are stilluncertain (Baili! and Karl, 1991).

1078 R. Danovaro et al. / Deep-Sea Research I 46 (1999) 1077}1094

Previous studies on sedimentary nucleic acids revealed that DNA concentrations,due to their mostly detrital composition, did not change signi"cantly with depth; bycontrast RNA concentrations were found to decrease signi"cantly in deep waters(Danovaro et al., 1993). These results led some authors to hypothesize that the ratio ofRNA to DNA in deep-sea sediments may be related to rates of sediment communityoxygen consumption (SCOC; Duineveld et al., 1996).

In the present study we investigated the depth-related changes of the sedimentaryDNA and RNA concentrations and benthic bacterial dynamics on a seasonal basis inrelation to the seasonally varying organic inputs. The aims of this study are: (1) toprovide quantitative information on temporal changes in sedimentary genetic mater-ial and (2) to evaluate bacterial DNA and RNA contribution to the total pools ina highly oligotrophic system.

2. Materials and methods

2.1. Study area and sampling

Sediment sampling was carried out on the continental shelf and slope and in thedeep basin of the Cretan Sea (South Aegean, Eastern Mediterranean) north of Iraklio.This area is one of the world's most oligotrophic, characterized by high bottomtemperatures (13}14.53C), low sedimentation rates (total mass #ux 50 mgm~2d~1),strong water column strati"cation and extremely low primary productivity(20}30 g Cm~2 yr~1; Dugdale and Wilkerson, 1988).

Undisturbed sediment samples were collected using a multicorer (Mod. Maxicorer;i.d. 9.0 cm, depth penetration '20 cm) on August 1994, February}March 1995, Mayand September 1995 at seven stations along a transect from 40 to 1540 m depth: 40,100, 200, 500, 700, 940 and 1540 m depth (Fig. 1). For sedimentary analyses andnucleic acid determination we utilized the top 1 cm of two replicate cores, which weresubsequently homogenized and deep frozen at !203C. From the same cores utilizedfor nucleic acid analysis we collected, with sterile cut-o! syringes, 3 to 5 replicatesubsamples (0.63 cm3) for bacterial analyses. The subsamples were "xed with 0.2 lm-"ltered seawater containing 2% bu!ered formalin and stored at 43C until analysis(within 20 days).

2.2. Organic matter biochemical composition and chloroplastic pigments

Total carbohydrates (TCHO) were analyzed according to Gerchakov and Hatcher(1972). Acid soluble carbohydrates were extracted in 0.1 N HCl (2 h, 503C). Anotherset of replicates was extracted in NaOH 0.1 M (4 h, ambient temperature). Solublecarbohydrate determinations were carried out after extraction as for TCHO. Proteindetermination (PRT) was carried out following an extraction with NaOH (0.5 M, 4 h)and determined according to Hartee (1972) modi"ed by Rice (1982) to compensate forphenol interference. Lipids (LIP) were extracted from sediment samples by directelution with chloroform and methanol. Analyses were carried out using the methods

R. Danovaro et al. / Deep-Sea Research I 46 (1999) 1077}1094 1079

Fig. 1. Sampling area and station locations.

of Bligh and Dyer (1959) and Marsh and Weinstein (1966). For each biochemicalanalysis, blanks were made using the same sediments previously treated in a mu%efurnace (5503C, 4 h). Chlorophyll-a (Chl-a) and phaeopigment concentrations weredetermined according to Lorenzen and Je!rey (1980). All analyses were carried out on3}5 replicates and normalized to sediment dry weight.

2.3. Nucleic acid analysis

Before analysis larger macroscopic organisms were removed from the sedimentsamples. Nucleic acid analyses were performed using material treated to avoid nu-clease contamination: glassware, glass Pasteur pipettes and pipette tips were carefullycleaned by soaking in 1 NNaOH, 10% HCl and milliQ water. Nucleic acid extraction


and measurement followed the procedures of Zachleder (1984) as applied byDanovaro (1996), with few modi"cations to enhance DNA extraction from marinesediment (Dell'Anno et al., 1998). Brie#y, 1 g of wet sediment (three replicates) wastreated with 3.0 ml of 0.5 N perchloric acid, stirred for 3 min and sonicated three timesfor 1 min (with intervals of 30 s). Nucleic acids were hydrolyzed at 753C for 30 minunder continuous stirring (Burton, 1968). After centrifugation (3000]g, 10 min), theabsorbance of the total nucleic acid content (TNA) in the supernatant was measuredat 260 nm. DNA absorbance was determined with a diphenylamine (2% in acetic acid)light-activated reaction (40 W, 12 h) at 598 nm, and converted into concentrationusing standard solutions of DNA Type I from calf thymus. DNA concentration wasthen expressed as equivalents of absorbance at 260 nm in order to calculate bydi!erence the absorbance due to RNA:

ABSRNA

"ABSTNA

!ABSDNA

RNA absorbance (260 nm) was then converted into concentration using standardsolutions of RNA Type III from baker's yeast.

The extinction coe$cient, calculated from calibration curves of nucleic acid stan-dard solutions (ranging from 1 to 50 lg ml~1), was 0.02 and 0.025 for DNA and RNA,respectively. These values are the same as those extrapolated from Sanbrook et al.(1989). Since TNA absorbance at 260 nm might be a!ected by interference due toinorganic compounds, we used sediment subsamples, previously treated in a mu%efurnace (5503C, 4 h), as blanks.

Internal standards of calf thymus DNA and baker's yeast RNA were added toreplicate subsamples before extraction. The "nal yield of the internal standards ofDNA and RNA was, on average, 60 and 85%, respectively. The reasons for suchdi!erence are unknown but similar results were obtained by Kemp et al. (1993). Nocorrections for extraction e$ciency have been applied to our data because we cannotbe sure that RNA and DNA in sediment samples will be extracted with the samee$ciency as added nucleic acids. Sensitivity of the method has been tested usinginternal standards of calf thymus DNA and baker's yeast RNA (accuracy$1.0 lg)and appeared to be adequate for "eld investigations (Dell'Anno et al., 1998). Datawere normalized to sediment dry weight after desiccation (603C, constant weight).

2.4. Bacterial analyses

For bacterial analyses, subsamples were sonicated 3 times (Soni"er Branson 2200,60 W for 1 min), diluted 100 times, stained with Acridine Orange and "ltered on blackNuclepore 0.2 lm "lters (Fry, 1990). The "lters were analyzed as described byDanovaro (1996) under epi#uorescence microscopy (Zeiss Universal Microscope).Bacterial DNA (B-DNA) and bacterial RNA (B-RNA) contribution to the totalDNA and RNA pool were calculated assuming a conversion factor of 3.3]10~15 gDNAcell~1 (Simon and Azam, 1989) and 4.2]10~15 g RNAcell~1 (Fuhr-man and Azam, 1982). These conversion factors were selected because they werecalculated for cells of the same average size encountered in this study. Data werenormalised to sediment dry weight after desiccation (603C, constant weight).


3. Results

3.1. Phytopigments and biochemical composition of sediment organic matter

Data on phytopigment, protein, carbohydrate and lipid concentrations in the top1 cm of the sediments are summarized from Tselepides et al. (1996) and reported inTable 1. Chlorophyll-a concentrations were generally extremely low and stronglydecreased from continental shelf to bathyal depths. Phaeopigment concentrationswere generally characterized by highest values in February and May 1995 and showedsimilar spatial patterns strongly decreasing with water depth. Total carbohydrates(TCHO) represented the main biochemical component of the organic matter, andtheir concentrations were almost constant with depth. Total carbohydrate content ofthe sediments showed clear temporal changes (ANOVA, F"24.9, p(0.001), withhighest values in August 1994 and lowest in May and September 1995. By contrast,soluble carbohydrates (SCHO; i.e. the sum of NaOH and HCl soluble carbohydrates)decreased from continental shelf to bathyal depths and accounted on average for lessthan 6% of TCHO. SCHO contribution to TCHO was highest in May and September1995 (8 and 6%, respectively) and lowest in August 1994 and February 1995 (5 and1%, respectively). Proteins, the second main biochemical class of organic compounds,generally decreased from continental to bathyal sediments (on average up to 7 times).Highest values were observed in February and lowest in August 1994 and September1995. Lipid concentrations typically decreased from the continental shelf to bathyaldepths and were characterized by highest values in August 1994 and February 1995and lowest values in May 1995.

3.2. Spatial and temporal changes in nucleic acid concentrations

Spatial and temporal changes in DNA and RNA concentrations and RNA :DNAratio are illustrated in Fig. 2a}c. DNA concentrations showed highest values on thecontinental shelf at 40 m depth (55.2$4.4 lg g~1 sed. DW in August 1994) and lowestconcentration at 1540 m depth (3.5$0.0 lg g~1 sed. DW in May 1995). Signi"cantdi!erences in DNA concentrations were observed among stations (ANOVA, F"3,p(0.05). Similar spatial patterns were evident for RNA concentrations, which werecharacterized by a signi"cant decrease with increasing water depth (ANOVA, F"8.8,p(0,001). Highest RNA concentrations were observed at 40 m depth (29.9$5.0 lg g~1 sed. DW in September 1995) and the lowest at 1540m depth (0.4$0.7 lg g~1 sed. DW in September 1995). RNA : DNA ratio did not show signi"cantchanges with depth, although lowest values were observed at 940 m depth ((0.1 inSeptember 95).

Temporal changes in DNA concentrations were characterized by highest values inFebruary 1995 and September 1995 (on average 32 and 31 lg g~1, respectively) andlowest concentrations in August 1994 and May 1995 (on average 21 and 17 lg g~1).RNA concentrations were characterized by highest values in May 1995 and Septem-ber 1995 (on average 15 and 11 lg g~1, respectively) and lowest concentrations inAugust 1994 and February 1995 (on average 7 and 9 lg g~1). RNA :DNA ratios were


characterized by highest values in May 1995 (on average 0.7) and lowest in February95 (on average 0.3).

3.3. Benthic bacteria

Bacterial density in the sediments of the Cretan Sea displayed highest values at 100and 500 m depth (4.6]108 cells g~1, both in February 95) and lowest values at 900 mdepth (0.9]108 cells g~1 in August 1994; Table 2). Bacteria showed signi"cant di!er-ences between sampling periods (ANOVA, F"3.2, p(0.05), with highest values inFebruary 1995 (on average 2.8]108 cells g~1) and lowest densities in August 1994 (onaverage 1.3]108 cells g~1).

The relative contributions of bacterial-DNA and bacterial-RNA to the total DNAand RNA pools are reported in Table 2. Bacterial DNA contribution increased, onannual average, with depth from 1.7% at 40 m depth to 7.4% at 700 m depth.Bacterial DNA contribution increased from August 1994 to May 1995 (from 3.0 to8.0%, respectively). Bacterial RNA contribution strongly increased with increasingdepth with values ranging from 2.2% at 40 m depth in August 94 up to 100% at1540 m depth in September 95. Bacterial RNA contribution showed evident temporalchanges with highest value in September 1995 (on average 43.6%) and lowest in May1995 (on average 8.5%).

4. Discussion

Previous studies have shown that the sediments of the Cretan Sea represent aprimary organic matter depleted ecosystem (Tselepides et al., 1996). Chlorophyll-aconcentrations (utilized as a measure of the fresh material inputs) were indeedextremely low and characterized by a strong decrease with increasing water depth.Chloroplastic pigments followed a similar spatial pattern, and phaeopigments ac-counted for more than 60% of total chloroplastic pigment concentrations on thecontinental shelf and for 90% in deep-sea sediments. Strong depth gradients wereidenti"ed in terms of protein and soluble carbohydrate concentrations, as well as inthe nutritional quality of the sedimentary organic matter. Carbohydrates were charac-terized by a highly refractory composition (overall soluble fraction less than 6%). Theratio of proteins to carbohydrates decreased with depth indicating an increasing&&protein dilution'' in deeper stations. The overall picture suggests that the continentalshelf was characterized by relatively high concentrations of labile organic compounds(i.e. proteins, lipids and soluble carbohydrates). By contrast, the continental slope anddeep-basin were characterized by a highly refractory OM composition (as indicatedby the low contribution of soluble carbohydrates to the total carbohydrate concentra-tion) and by negligible inputs of primary organic matter. All sedimentary organiccompounds were found to vary strongly between sampling periods, even at greaterdepths. Protein content increased up to 3 times from August 1994 to February 1995 todecrease in spring and summer. By contrast, total carbohydrate concentrationsshowed highest values in August 1994 and lower values in the other sampling periods.


Tab

le1

Photo

synt

hetic

pig

men

ts(c

hloro

phyl

l-a

and

pha

eopi

gmen

ts)

and

bio

chem

ical

com

position

ofth

ese

dim

enta

ryorg

anic

mat

ter

(as

tota

lca

rboh

ydra

tes,

Tota

l-C

HO

;so

lubl

eca

rboh

ydra

tes,

Solu

ble

CH

O}

asth

esu

mof

NaO

Han

dH

Clso

luble

carb

ohyd

rate

s;lip

ids;

and

prot

eins

)in

the

Cre

tan

Sea

(0}1

cm).

Sta

nda

rddev

iations

are

repor

ted.

Dat

aar

eex

pre

ssed

aslg

g~1

sedi

men

tD

W

Stat

ion

Chl

oro

phyl

l-a

Phae

opig

men

tsTota

lC

HO

Sol

ubl

eC

HO

Lip

ids

Pro

tein

s

lgg~

1D

WSt

dlg

g~1

DW

Std

lgg~

1D

WSt

dlg

g~1

DW

Std

lgg~

1D

WSt

dlg

g~1

DW

Std

D1

Aug

ust

1.5

0.4

5.8

0.1

3289

506

144

1847

232

1049

158

(40

m)

Feb

ruar

y1.

30.

32.

80.

511

2426

221

327

023

1554

337

May

3.1

0.9

5.2

0.8

388

2793

711

124

1261

201

Sept

ember

2.6

1.2

3.0

0.3

580

211

110

2335

848

605

102

Ann

ualav

g.2.

10.

74.

20.

413

4525

292

1330

332

1117

199

D2

Aug

ust

0.3

0.2

5.7

0.5

3220

850

333

9468

914

373

823

2(1

00m

)Feb

ruar

y0.

10.

27.

11.

623

3856

334

399

5221

7221

1M

ay0.

50.

27.

60.

187

227

264

1513

322

1075

258

Sept

ember

0.5

0.1

4.7

0.1

1543

366

5910

297

5374

820

7A

nnua

lav

g.0.

30.

26.

30.

619

9338

612

231

380

6811

8322

7

D3

Aug

ust

0.0

0.0

3.0

0.7

3752

850

177

106

336

4363

636

(200

m)

Feb

ruar

y0.

10.

26.

91.

229

2953

728

645

415

232

6558

7M

ay0.

10.

23.

90.

585

098

5811

647

1038

129

Sept

ember

0.2

0.1

2.5

0.2

789

4771

1511

321

520

37A

nnua

lav

g.0.

10.

14.

10.

720

8038

383

3524

256

1365

197

D4

Aug

ust

0.3

0.2

0.8

0.2

4026

982

6763

370

2540

515

2(5

00m

)Feb

ruar

y0.

20.

32.

71.

229

6242

810

220

055

1371

362

May

0.0

0.0

2.0

0.1

849

5641

779

394

479

Sept

ember

0.1

0.1

0.9

0.2

463

2839

815

912

268

30A

nnua

lA

vg.

0.1

0.2

1.6

0.4

2075

373

3920

202

2474

715

6

D5

Aug

ust

0.1

0.1

0.9

0.3

7864

1582

525

297

388

5342

363

(700

m)

Feb

ruar

y0.

00.

02.

00.

534

7268

710

114

122

1081

152

May

0.1

0.1

2.9

0.4

782

5935

549

460

590

Sept

ember

0.0

0.0

0.6

0.4

1052

164

323

124

1821

823

Ann

ualA

vg.

0.1

0.1

1.6

0.4

3293

623

150

7617

624

582

82

D6

Aug

ust

0.4

0.1

0.1

0.1

7476

1101

446

185

155

4220

510

7(9

40m

)Feb

ruar

y0.

00.

00.

70.

225

6457

56

185

1361

488

May

0.2

0.3

0.7

0.4

515

7529

646

438

483

Sept

ember

0.0

0.0

0.6

0.4

576

104

162

7011

116

9A

nnua

lA

vg.

0.2

0.1

0.5

0.3

2783

464

124

4889

1833

072

D7

Aug

ust

0.1

0.1

0.7

0.3

6266

563

3912

325

8931

317

2(1

540

m)

Feb

ruar

y0.

00.

00.

70.

121

5679

88

160

859

780

May

0.2

0.2

1.0

0.5

393

130

176

174

519

26Se

ptem

ber

0.1

0.1

0.7

0.7

803

155

215

105

317

259

Ann

ualA

vg.

0.1

0.1

0.8

0.4

2405

411

216

127

2640

084


Fig

.2.

Nuc

leic

acid

conce

ntra

tion

sin

these

dim

ents

oft

he

Cre

tan

Sea

:(a)

spat

ialp

atte

rnsof

DN

Aco

nce

ntra

tions

;(b)s

pat

ialp

atte

rnsofR

NA

conce

ntr

atio

ns;

(c)

spat

ialpat

tern

softh

eR

NA

/DN

Ara

tio.

Stan

dar

ddev

iation

sar

ere

port

ed(n"

3).D

ata

are

expre

ssed

aslg

g~1

ofse

dim

ent

DW

.


Table 2Bacterial contribution to the nucleic acid pools: bacterial density ($std), bacterial-DNA and RNAconcentrations (B-DNA and B-RNA) and their relative contribution (expressed as %) to the nucleic acidpools (B-DNA/DNA and B-RNA/RNA). nd"not determined

Bacterial densityDepth B-DNA B-RNA B-DNA/DNA B-RNA/RNA

Season Station (m) n]108 Std (lg g~1) (lg g~1) (%) (%)

August 94 D1 40 1.5 0.4 0.5 0.6 0.9 2.2D2 100 1.6 0.2 0.5 0.7 2.6 5.5D3 200 1.9 0.3 0.6 0.8 7.3 ndD4 500 1.1 0.1 0.4 0.5 2.6 14.5D5 700 1.2 0.1 0.4 0.5 1.7 35.1D6 940 0.9 0.1 0.3 0.4 1.9 33.5D7 1540 1.2 0.2 0.4 0.5 5.9 25.2Average 1.3 0.4 0.6 3.3 16.6

February 95 D1 40 1.5 0.3 0.5 0.6 1.6 2.3D2 100 4.6 0.3 1.5 1.9 3.1 18.3D3 200 1.8 0.1 0.6 0.8 1.8 100.0D4 500 4.6 0.7 1.5 1.9 5.7 18.7D5 700 2.1 0.3 0.7 0.9 1.7 48.4D6 940 3.3 0.4 1.1 1.4 4.1 43.6D7 1540 1.5 0.2 0.5 0.6 2.9 8.0Average 2.8 0.9 1.2 3.0 34.2

May 95 D1 40 2.2 0.5 0.7 0.9 2.4 3.2D2 100 1.7 0.3 0.6 0.7 3.7 6.1D3 200 2.7 0.3 0.9 1.1 4.5 5.1D4 500 2.3 0.2 0.8 1.0 5.7 9.4D5 700 3.5 1.0 1.2 1.5 22.9 19.0D6 940 2.0 0.5 0.7 0.9 2.3 5.4D7 1540 1.5 0.3 0.5 0.6 14.3 11.1Average 2.3 0.7 1.0 8.0 8.5

September 95 D1 40 2.3 0.2 0.8 1.0 1.9 3.3D2 100 3.5 0.4 1.1 1.5 3.0 7.7D3 200 3.4 1.1 1.1 1.4 3.0 7.0D4 500 2.9 0.6 1.0 1.2 5.7 23.8D5 700 3.3 0.3 1.1 1.4 3.2 63.7D6 940 2.2 0.5 0.7 0.9 1.9 100.0D7 1540 1.0 0.1 0.3 0.4 2.7 100.0Average 2.7 0.9 1.1 3.1 43.6

The di!erent temporal patterns of the various components suggest a di!erent com-position or origin of the OM inputs in di!erent seasons.

4.1. Nucleic acid distribution and temporal variability in relation to environmentalconstraints

It is generally expected that DNA and RNA concentrations in the sediments arerelated to the trophic state, benthic biomass and metabolism (Danovaro et al., 1993).


Table 3Comparison of DNA, RNA concentrations and RNA : DNA ratio in marine sediments of di!erent areas

Depth DNA RNALocation (m) (lg g~1) (lg g~1) RNA : DNA Authors

Spartina Marsh nd nd 4.7}16.0! nd Moran et al. (1993)(7.9)

Continental Shelf 36 nd 1.9! nd ''

SE-USPosidonia Bed 4 7.0}16.4 5.0}30.7 0.4}1.7 Danovaro (1996)W-Mediter. (11.4) (9.5) (0.8)Ligurian Sea 10 3.0}9.0 5.1}13.8 0.8}3.2 Danovaro et al. (1995)W-Mediter. (4.9) (7.6) (1.7)Ionian Sea 550}2401 2.4}4.2 (2.0}10.6 0.3}4.5 Danovaro et al. (1993)

(3.5) (5.7) (1.9)Aegean Sea 110}204 7.2}14.5 (2.0}2.5 0.1}0.6 ''

(11.7) (1.8) (0.3)Aegean Sea 533}1840 6.9}52.0 (2.0}7.1 0.1}1.0 ''

(28.4) (0.6) (0.1)Adriaric Sea 15}60 18.0}145.0 nd nd Fabiano et al. (1997)

(57.0)Kenian Slope 70}2200 19.0}74.0 2.5}23.0 0.1}0.3 Duineveld, unpubl.Upwelling Coast (38.3) (8.1) (0.2)Indian OceanCretan Sea 40}200 8.7}55.2 (1.0}29.9 (0.1}1.8 Present studyEastern Med. (31.5) (19.2) (0.7)

500}1540 3.5}40.3 (1.0}15.9 (0.1}0.9 ''

(20.1) (4.9) (0.3)North Atlantic 4850 52.2}60.8 7.6}28.2 0.1}0.5 Rice (1997)

(56.5) (17.9) (0.3)

Note: Mean values are reported in parenthesis.!data reported to sediment wet weight.

Table 3 summarizes DNA and RNA concentrations from di!erent marine sediments.DNA concentrations in the Cretan Sea appear to be comparable to those observed inother deep-sea sediments, such as the Porcupine Abyssal Plain (at 4850 m depth, Rice,1997) and in the southern Aegean. Highest DNA concentrations were reported fromthe eutrophic Northern Adriatic (up to 145 lg DNAg~1 sed dry wt) in#uenced by Poriver out#ow. These data suggest that the amounts of sedimentary DNA are closelyrelated to the productivity of the systems. A further con"rmation to this conclusion isprovided by the analysis of the mesoscale distribution of DNA concentrations in theLigurian Sea (NW Mediterranean). At two sites 1 mile apart at similar depths, DNAconcentrations in the seagrass sediments (Posidonia oceanica) were about double ofthose reported in well sorted sandy sediments outside the meadows (Danovaro, 1996;Danovaro et al., 1995).

Table 3 shows that a signi"cant di!erence in DNA concentrations is evident onlywhen clearly di!erent trophic conditions are encountered. This study was designed to


Fig. 3. Patterns of nucleic acid distribution with increasing depth: (a) DNA distribution; (b) RNAdistribution. Data are relative to all sampling periods.

investigate sediments characterized by a strong trophic gradient from the continentalshelf to the deep sea. The di!erent trophic conditions previously observed for othersedimentary parameters (i.e. phytopigments, proteins and soluble carbohydrates)between continental shelf and bathyal sediments are re#ected by a general decrease inDNA and RNA concentrations (Fig. 3a, b). By contrast neither the DNA concentra-tions along the Kenyan slope (Indian Ocean, Duineveld, pers. comm.) or in the deepIonian and Aegean Sea (Eastern Mediterranean, Danovaro et al., 1993) showed a cleardepth-related trend. In the Ionian and the Aegean Sea DNA concentrations weresigni"cantly correlated to total sedimentary carbohydrates (Danovaro et al., 1993).

The results of this study seem to indicate the presence of spatial and temporalvariations of sedimentary DNA also at bathyal depths. Such a "nding might con"rmthe recent view of the deep seas as systems characterized by evident temporal changesand clear interannual variability (Rice et al., 1994). Spatial and temporal changes innucleic acid concentrations are likely to be related to variations in the concentrationof labile organic compounds (particularly phytopigments and proteins). In this regard,although correlation analyses do not allow cause-e!ect relations to be inferred,signi"cant relationships between nucleic acids and phytopigments (n"28, r"0.43,p(0.05; r"0.61, p(0.01, respectively, for DNA and RNA) and between DNA andproteins (n"28, r"0.38, p(0.05) were found.

Changes in the sedimentary DNA concentrations may be directly related to a DNAinput from the photic layer or indirectly through a benthic response (in terms ofincreased biomass and therefore increased DNA content) to OM inputs. Previousstudies have shown that large amounts of DNA may be supplied to the benthos fromparticle sedimentation. Baili! and Karl (1991) reported a DNA #ux of2}4 mgm~2d~1 in the Antarctic Peninsula region. Turley and Mackie (1995) esti-mated a bacterial-DNA #ux of 0.01 mg m~2d~1 in the NE Atlantic at 4465 m depth.From synoptic measurements carried out in the Cretan Sea at 1500 m depth a bacter-ial-DNA #ux to the sediments of about 0.005 mgDNAm~2d~1 was reported(Danovaro et al., 1999a). Unfortunately, no measurements of the total DNA #ux areavailable, but bacterial-DNA #ux alone appears negligible when compared to theconcentrations reported in the sediments (see below).


Fig. 4. Bacterial contribution to the nucleic acid pools. Illustrated are: (a) bacterial contribution to the totalDNA pool; (b) bacterial contribution to the total RNA pool. Data reported at each depth are means of thefour sampling periods (expressed as %).

At 1540 m depth, both DNA concentrations and bacterial densities increased fromAugust 1994 to February 1995 in response to increased #uxes. By contrast noresponse (in terms of biomass) was observed for other benthic components such as#agellates and meiofauna (Danovaro et al., 1998, 1999b). Estimates of the bacterial-DNA contribution to the total DNA pool revealed that bacteria contributed 6% inAugust 1994 and 3% in February 1995. Therefore it seems that bacteria were notresponsible for the increased sedimentary DNA levels. These data led the authors toconclude that changes in sedimentary DNA concentrations are due to DNA inputsfrom the photic layer more than to an increased benthic biomass.

4.2. Bacterial contribution to sedimentary nucleic acid concentrations

Bacterial contribution to the DNA pools increased, even though irregularly,with increasing depth (Fig. 4a). This pattern is coarsely opposite to the one dis-played by sedimentary chloropigments and indicates that bacterial contribution to the


Fig. 5. Seasonal changes in the contribution of the living-DNA to the total DNA pools. Reported are:Bacterial-DNA (Bact-DNA), heterotrophic #agellate-DNA (Flag-DNA), Meiofaunal-DNA (Meio-DNA)and the remaining detrital fraction (Detrital-DNA). Data reported are expressed as % and represent themean value of the seven stations sampled.

sedimentary DNA increases with the decreasing of allochthonous DNA input fromthe photic layer.

The decrease of RNA concentrations along the Cretan shelf and slope re#ectsa drop in the benthic metabolic activity. Duineveld et al. (1996), from synopticmeasurements on sediment community oxygen consumption (SCOC), demonstrateda clear decrease of the respiratory activity with increasing depth. Since bacteriarepresent the large majority of the total benthic biomass in deep-sea sediments (seebelow), it may be hypothesized that most of this activity is due to the smallest sizegroups. These results are consistent with our estimates of bacterial contribution to theRNA pools. Bacterial RNA alone, on average, accounted for about 26% of the totalsedimentary RNA pool. Such contribution increased signi"cantly with depth from3% at 40 m depth to an average of 41% below 700 m depth (Fig. 4b). These resultsindicate that, during all sampling periods, changes of RNA concentrations in deepersediments may be largely dependent upon bacterial dynamics.

Data reported in this study indicate that DNA pools in the sediments of the EasternMediterranean are mostly composed of detrital material. In fact, bacterial-DNAaccounted on average for 4% of the total DNA pool (ranging from 3 to 8% inFebruary and May 1995, respectively). Since meiofaunal biomass was equivalent toonly 14}24% of the bacterial biomass (Danovaro et al., 1999b), and heterotrophicnano#agellates (Danovaro et al., 1998) accounted only for 0.1}0.2% of the total DNApool (based on an average density of 67.7]103 cells g~1, and assuming 0.05 pgDNAcell~1; Turk et al., 1992), a total living contribution to the DNA pool rangingfrom 3 to 10% may be estimated (Fig. 5). These estimates indicate that at all samplingperiods more than 90% of the total DNA extracted from the sediments was detrital.These values must obviously be viewed with caution, because they may be biased bythe use of conversion factors. Conversion factors depend upon cell biomass, whereaswe estimated DNA content from cell number (even though from similar cell size).


Moreover, bacterial assemblages display seasonal changes in their DNA content (Leeand Kemp, 1994) not considered in this study. Our conversion factor of 3.3 fg DNAcell~1 is about 1/3 of the factor (9.8 fg DNA cell~1) utilized by Paul and Carlson(1984). This means that the bacterial contribution, utilizing the latter factor, couldincrease up to about 13%.

Nonetheless, these data suggest that the large majority (more than 80% evenassuming the highest conversion factors for all benthic components) of the DNA poolin the sediments of the eastern Mediterranean is detrital. Little information is actuallyavailable for comparison: Winn and Karl (1986) at various Paci"c Ocean locationsand Baili! and Karl (1991) in the Antarctic peninsula region, provided the "rstestimates for the relative signi"cance of particulate detrital DNA (75}90% and0}93%, respectively). Bacterial contribution to DNA pools reported in this studymatch exactly the value reported by Danovaro et al. (1993) in the Ionian and AegeanSea (7%, using a conversion factor of 5 fg DNAcell~1).

If we assume that about 90% of the sedimentary DNA is detrital, it is possible tocoarsely estimate (assuming a speci"c sediment density of 1.8 and 50% of watercontent in the top-1cm layer of the sediments), on average, about 0.26 gDNAm~2 cm~1 are available throughout the year. Such estimate might range from0.23 to 0.28 g DNAm~2 cm~1 if we assume respectively the highest and the lowestcontribution of the living components to the total DNA pool.

The amount of detrital DNA calculated assuming an average living contribution of10%, ranged from 0.22 to 0.34 g DNAm~2 cm~1 (in August 1994 and February 1995,respectively) and decreased with depth, ranging from 0.41 (at 40 m depth) to 0.13 gDNAm~2 cm~1 (at 1540 m depth).

The quantitative relevance of the sedimentary detrital DNA appears particularlyevident when compared to bacterial (+70 mgCm~2 cm~1) and meiofaunal biomass(+40 mgCm~2 cm~1; Danovaro et al., 1999b, c). Such detrital DNA may representan important source of organic N and P or a reservoir of nucleic acid precursors thatare energetically expensive to synthesize ex novo (Paul et al., 1987). However, withoutdetailed information on detrital-DNA #ux and utilization rates by heterotrophs it isdi$cult to evaluate the ecological role and the balance (input vs. output) of thiscomponent in deep-sea sediments.

The use of the RNA : DNA ratio in bacteria and phytoplankton has long beenthought to be proportional to levels of their growth rates (Dortch et al., 1983}1985).However, in natural samples, the use of RNA : DNA ratio may be biased by thepresence of detrital DNA. This appears evident from the lack of clear RNA :DNApatterns with increasing water depth (Fig. 2c). As the large majority of DNA pools inthe sediments of the Cretan Sea is detrital, the use of this ratio as a measure of themicrobial benthic growth is precluded.

Acknowledgements

We would like to thank the crew and the o$cers of the R/V PHILIA for help duringsampling. We thank Dr. A. Tselepides and Prof. A. Eleftheriou (Institute of Marine


Biology of Crete) for great collaboration and providing laboratory facilities. Prof.Della Croce provided "nancial support and facilities and greatly stimulated thediscussion of the topic dealt within this study. Thanks are due also to two anonymousreferees for their relevant and useful comments. This research was undertaken in theframework of the CINCS program. We acknowledge the support from the EuropeanCommission's Marine Science and Technology Program (MAST II) under Contractn. MAS2-CT-940092.

References

Baili!, D.M., Karl, D.M., 1991. Dissolved and particulate DNA dynamics during a spring bloom in theAntarctic Peninsula region, 1986}87. Deep Sea Research 38, 1077}1095.

Bligh, E.G., Dyer, W., 1959. A rapid method for total lipid extraction and puri"cation. Canadian JournalBiochemistry Physiology 37, 911}917.

Burton, K., 1968. Determination of DNA concentration by diphenylamine. Methods Enzymology 12B,163}166.

Danovaro, R., 1996. Detritus-Bacteria-Meiofauna interactions in a seagrass bed Posidonia oceanica) of theNW Mediterranean. Marine Biology 127, 1}13.

Danovaro, R., Fabiano, M., Della Croce, N., 1993. Labile organic matter and microbial biomasses indeep-sea sediments (Eastern Mediterranean Sea). Deep-Sea Research I 40, 953}965.

Danovaro, R., Fabiano, M., Albertelli, G., 1995. Possible use of RNA :DNA ratio for detecting oil induceddisturbance: a "eld report. Chemistry and Ecology 11, 1}10.

Danovaro, R., Marrale, D., Della Croce, N., Dell'Anno, A., Fabiano, M., 1998. Heterotrophic nano#agel-lates, bacteria and labile organic compounds in continental shelf and deep-sea sediments of the EasternMediterranean. Microbial Ecology 35, 244}255.

Danovaro, R., Della Croce, N., Dell'Anno, A., Fabiano, M., Marrale, D., Martorano, D., 1999a. Seasonalchanges and biochemical composition of the labile organic matter #ux in the Cretan Sea. Progress inOceanography, in press.

Danovaro, R., Tselepides, A., Otegui, A., Della Croce, N., 1999b. Dynamics of meiofaunal assemblages onthe continental shelf, slope and deep-sea sediments of the Cretan Sea: relationships with seasonalchanges in food supply. Progress in Oceanography, in press.

Danovaro, R., Marrale, D., Dell'Anno, A., Della Croce, N., Tselepides, A., Fabiano, M., 1999c. Bacterialresponse to seasonal changes in labile organic composition on the continental shelf and bathyalsediments of the Cretan Sea. Progress in Oceanography, in press.

Dell'Anno, A., Fabiano, M., Duineveld, G.C.A., Kok, A., Danovaro, R., 1998. Nucleic acid (DNA, RNA)and RNA/DNA ratio determination in marine sediments: comparison of spectrophotometric,#uorometric and High Performance Liquid Chromatography methods and estimation of detrital DNA.Applied and Environmental Microbiology 64, 3238}3245.

Dortch, Q.F., Roberts, T.L., Clayton, J.R., Ahmed, S.I., 1983. RNA : DNA ratios and DNA concentrationsas indicators of growth rate and biomass in planktonic organisms. Marine Ecology Progress Series 13,61}71.

Dortch, Q.F., Clayton, J.R., Thoresen, S.S., Ahmed, S.I., 1985. Nitrogen storage and the use of biochemicalindices to assess nitrogen de"ciency and growth rate in natural plankton populations. Journal ofMarine Research 43, 437}464.

Dugdale, R.C., Wilkerson, F.R., 1988. Nutrent sources and primary production in the Eastern Mediterra-nean. Oceanologica Acta 9, 178}184.

Duineveld, G.C.A., Tselepides, A., Berghuis, E.M., Van Derwelle, J., Kok, A., 1996. Oxygen consumptionby the sediment communities in the oligotrophic Cretan Sea. In: Tselepides, A., Papadopoulou,K.-N., Polychronaki, T. (Eds.), CINCS: Pelagic benthic coupling in the oligotrophic Cretan Sea,pp. 145}153.


Fry, J.C., 1990. Direct methods and biomass estimation. Methods in Microbiology, vol. 22. Academic Press,New York, pp. 41}85.

Fabiano, M., Danovaro, R., Manini, E., 1997. Organic matter composition, bacterial production andenzymatic activities in the sediments of the northern and middle Adriatic (June}July 1996). TechnicalReport, PRISMA Programme, CNR.

Fuhrman, J.A., Azam, F., 1982. Thymidine incorporation as a measure of heterotrophic bacterioplanktonproduction in marine surface waters: evaluation and "eld results. Marine Biology 66, 109}120.

Gerchakov, S.M., Hatcher, P.G., 1972. Improved technique for analysis of carbohydrates in sediments.Limnology and Oceanography 17, 938}943.

Hartee, E.F., 1972. Determination of protein: a modi"cation of the Lowry method that gives a linearphotometric response. Analytical Biochemistry 48, 422}427.

Holm-Hansen, O., Sutcli!e, W.H., Sharp, J., 1968. Measurement of deoxyribonucleic acid in the ocean andits ecological signi"cance. Limnology and Oceanography 13, 507}514.

Jorgesen, N.O.G., Jacobsen, C.S., 1996. Bacterial uptake and utilization of dissolved DNA. AquaticMicrobial Ecology 11, 263}270.

Kemp, P.F., Lee, S., LaRoche, J., 1993. Estimating the growth rates of slowly growing marine bacteria fromRNA content. Applied and Environmental Microbiology 59, 2594}2601.

Lee, S., Kemp, P.F., 1994. Single-cell RNA content of natural marine planktonic bacteria measured byhybridization with multiple 16S rRNA-targeted #uorescent probes. Limnology and Oceanography 39,869}879.

Lochte, K., Turley, C.M., 1988. Bacteria and cyanobacteria associated with phytodetrituts in the deep-sea.Nature 333, 67}69.

Lorenzen, C., Je!rey, S.W., 1980. Determination of chlorophyll in sea water. Unesco Technical Papers inMarine Science 35, 1}20.

Marsh, J.B., Weinstein, W.J., 1966. A simple charring method for determination of lipids. Journal of LipidResearch 7, 574}576.

Moran, M.A., Torsvik, V.L., Torsvik, T., Hodson, R.E., 1993. Direct extraction and puri"cation of rRNAfor ecological studies. Applied and Environmental Microbiology 59, 915}918.

Paul, J.H., Carlson, D, 1984. Genetic material in the marine environment implication for bacterial DNA.Limnology and Oceanography 29, 1091}1097.

Paul, J.H., Je!rey, W.H., DeFlaun, M.F., 1987. Dynamics of extracellular DNA in the marine environment.Applied and Environmental Microbiology 53, 170}179.

Pfannkuche, O., 1993. Benthic response to the sedimentation of particulate organic matter at the BIOT-RANS station, 473N, 203W. Deep-Sea Research II 40, 135}149.

Rice, D.L., 1982. The detritus nitrogen problem: new observations and perspectives from organic geochem-istry. Marine Ecology Progress Series 9, 153}162.

Rice, A.L., 1997. High resolution temporal and spatial study of the benthic biology and geochemistry ofa north-eastern Atlantic abyssal locality (BENGAL): Annual Report, p. 94.

Rice, A.L., Thurston, M.H., Bett, B.J., 1994. The IOSDL DEEPSEAS programme: introduction andphotographic evidence for the presence and absence of a seasonal input of phytodetritus at contrastingabyssal sites in the northeastern Atlantic. Deep-Sea Research I41, 1305}1320.

Sanbrook, J., Fritsch, E.F., Magnatis, T., 1989. Molecular Cloning: a Laboratory Manual, 2nd ed. ColdSpring Harbor Laboratory Press, U.S.A.

Simon, M., Azam, F., 1989. Protein content and protein synthesis rates of planktonic bacteria. MarineEcology Progress Series 51, 201}213.

Tselepides, A., Danovaro, R., Della Croce, N., Duineveld, G., Polychronaki, T., Dell'Anno, A., Dafnomili,E., Plaiti, W., Akoumianaki, I., Martorano, D., 1996. Seasonal variability of chloroplastic pigments,TOC, TON, ATP and labile organic matter (carbohydrates, lipids, proteins and nucleic acids) over thecontinental margin and deep-sea sediments of the oligotrophic Cretan Sea. In: Tselepides, A.,Papadopoulou, K.-N., Polychronaki, T. (Eds.), CINCS: Pelagic } Benthic Coupling in the OligotrophicCretan Sea. pp. 113}135.

Turk, V., Rehnstam, A.S., Lundbegr, E., 1992. Release of bacterial DNA by marine nano#agellates, anintermediate step in phosphorus regeneration. Applied Environmental Microbiology 58, 3744}3750.


Turley, C.M., Lochte, K., 1990. Microbial response to the input of fresh detritus to the deep-sea bed.Paleogeography, Paleoclimatology, Paleoecology (Global, Planetary Change Section) 89, 3}23.

Turley, C.M., Mackie, P.J., 1995. Bacterial and cyanobacterial #ux to the deep NE Atlantic on sedimentingparticles. Deep-Sea Research I 42, 1453}1474.

Winn, C.D., Karl, D.M., 1986. Diel nucleic acid synthesis and particulate DNA concentrations: con#ictswith division rate estimates by DNA accumulation. Limnology and Oceanography 31, 637}645.

Zachleder, V., 1984. Optimization of nucleic acids assay in green and blue-green algae: extraction proced-ures and the light-activated diphenylamine reaction for DNA. Archive Hydrobiologie 67, 313}328.


Nucleic acid concentrations (DNA, RNA) in the continental and deep-sea sediments of the eastern Mediterranean: relationships with seasonally varying organic inputs and bacterial dynamics

Documents