Amino acid biogeochemistry in the Laurentian Trough: vertical ¯uxes and individual reactivity during early diagenesis J.C. COLOMBO1*, N. SILVERBERG2 and J.N. GEARING3 1Environmental Chemistry and Biogeochemistry, Facultad de Cs. Naturales y Museo, Universidad Nacional de La Plata, Paseo del Bosque s/n, La Plata 1900, Argentina, 2Centro Interdisciplinario de Ciencias Marinas, Playa del Conchalito s/n, Ap. postal 2300, La Paz, B.C.S., Mexico and 3Department of Chemistry, University of Massachusetts at Dartmouth, North Dartmouth, MA 02747, USA (Received 7 October 1997; returned to author for revision 16 January 1998; accepted 18 June 1998) AbstractÐThe detailed composition of total hydrolysable (THAA), dissolved free (DFAA) and com- bined (DCAA) amino acids was studied in settling particles and the solid phase and porewaters of underlying sediments in the Laurentian Trough to evaluate their sources and individual reactivities during early diagenesis. Vertical ¯uxes of THAA measured at 150 m depth (234±980 mmol/m2/day) rep- resented 3.8% of the average daily primary production and 8±16% of total organic carbon (TOC) and 24±42% of total nitrogen (TN) ¯uxes. THAA concentrations decreased from 89239 to 3924.4 mmol/ g from settling particles to the top 3 cm sediments, with no signi®cant change of the %THAA-C and %THAA-N. However, these parameters decreased with depth in the sediments (10±13 to 7±8% and 30±45 to 22±28%, respectively) indicating a selective THAA removal. THAA composition of settling particles and sediments was relatively uniform and showed a marked enrichment in serine, threonine and glycine relative to fresh plankton which is ascribed to the selective preservation of diatom cell- walls. Serine was the more speci®c diatom tracer; it covaried with diatom lipid biomarkers, was rela- tively more abundant at a seaward site and increased downcore re¯ecting the selective preservation of diatom cell-walls. An increasing trend with sediment depth was also observed for aspartic acid whereas glutamic acid and histidine decreased. Porewater DFAA and DCAA accounted for 3±25% of total DOC and showed low levels in the surface zone of most intense solid phase THAA decay. Both frac- tions showed clear compositional di�erences related to the prevailing source material: DCAA, as solid phase THAA, were dominated by serine and threonine + glycine, whereas DFAA were enriched in glu- tamic (Glu) and b-aminoglutaric acids (bGlu), probably originating from bacteria. These patterns chan- ged with depth in the sediments: the proportion of serine and bGlu increased in DCAA and DFAA, respectively, whereas that of glutamine, alanine and Glu decreased in the DFAA pool. The preferential downcore decay and conversion of Glu into bGlu was re¯ected by a consistent increase of bGlu/Glu ratios, particularly at a landward station where the higher rates of sedimentation and OM burial favor the continued metabolism of bacteria in deeper sediment layers. # 1998 Elsevier Science Ltd. All rights reserved Key wordsÐAmino acids, Vertical ¯uxes, Solid phase, Porewater, Early diagenesis, Reactivity INTRODUCTION The conversion of inorganic nitrogen to amino acids, aliphatic amines, polyamines, pyrimidines and purines during photosynthesis is the major source of organic amino compounds in the sea (Lee, 1988). In some environments, the assimilation of inorganic nitrogen by bacteria can be also signi®- cant (Kirchman et al., 1991). A portion of these compounds is transported in the form of rapidly sinking particles to bottom sediments. The magni- tude of this ¯ux is closely related to primary pro- ductivity, increasing about 250-fold for every 10- fold increase in productivity (Lee and Cronin, 1984). Amino acids represent about 10±25% of total organic carbon (TOC) and 30±50% of total nitrogen (TN) in sediments and their degradation usually accounts for about 10±20% of TOC and 80% of TN remineralization (Wefer et al., 1982; Henrichs et al., 1984; Henrichs and Farrington, 1987; Burdige and Martens, 1988; Lee, 1988; Cowie and Hedges, 1992). Because of their important role in the cycle of or- ganic matter (OM) in the ocean, amino acids were included in our study of the organic composition and relative reactivity of di�erent OM fractions in settling particles and sediments of the Laurentian Trough, a 1200 km long, 300±450 m deep coastal environment. In previous papers the bulk compo- sition (carbon, nitrogen, carbohydrates, proteins, total amino acids, lipids and pigments), speci®c lipid biomarkers (fatty acids, sterols and hydrocar- bons), as well as the ¯ux and early diagenesis of Org. Geochem. Vol. 29, No. 4, pp. 933±945, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0146-6380/98 $ - see front matterPII: S0146-6380(98)00113-2 *Corresponding author. E-mail: colombo@isis.unlp. edu.ar. 933 OM in sinking particles and sediments were exam- ined (Colombo et al., 1996a,b,c, 1997). Overall, these studies indicated that the vertical ¯ux of OM in the St. Lawrence is dominated by terrigenous and zooplanktonic contributions with a lower inci- dence of fresh phytoplankton. The characterized fraction of total organic carbon consisted of lipids (17±37%), carbohydrates (7.9±16%), hydrolyzable amino acids (8.4±16%) and labile proteins (0.3± 2.6%). A clear land±sea gradient is apparent in terms of sedimentation rates (higher in the land- ward direction) and relative marine contribution. At a landward site, settling particles and sediments showed higher C/N ratios and a stronger contri- bution of vascular plant n-alkanes and fatty acids and petrogenic hydrocarbons, whereas marine lipids, including speci®c diatom tracers, were more abundant in the seaward direction. Very labile mar- ine lipids that were enriched in settling particles were rapidly lost near the sediment±water interface. The preferential decay of marine components con- tinued with depth in the cores resulting in similar residual lipid patterns at 35 cm depth. However, landward and seaward sediments could still be dis- criminated by C/N ratios and their fatty acid and hydrocarbon composition. In order to evaluate the behavior of amino acids under contrasting conditions of organic inputs and early diagenesis, in the present paper we report the detailed composition of amino acids in settling par- ticles and the solid phase and porewaters of under- lying sediments sampled in spring and mid-summer at two sites along the terrestrial±marine gradient. MATERIALS AND METHODS Sampling A detailed description of the study area and sampling methods was presented in earlier papers (Colombo et al., 1996a,b). Brie¯y, settling particles were intercepted with two unpoisoned, free-drifting sediment traps suspended at 150 m depth, at land- ward (L) and seaward (S) sites in the Laurentian Trough (Fig. 1) during May and July 1988. After 8±30 h of deployment, the traps were recovered and the samples collected into 2 l glass jars. The water was decanted, swimmers and intact copepods were eliminated and the material was split for micro- Fig. 1. Drifting sediment trap trajectories and coring sites in the Laurentian Trough. J. C. Colombo et al.934 scopic examination, for the determination of total particle ¯ux and for chemical analysis. Plankton samples were collected in horizontal and vertical tows made at both stations with 75 and 333 mm mesh nets. Undisturbed bottom sediments were recovered during each period of sediment trap deployment using a 0.1 m2 box corer. The cores were immediately subsampled in a glove box under a forced ¯ow of N2 by scraping successive 12 mm thick layers in the ®rst cm and then taking 1 cm thick slices every 1 to 5 cm depth and at 5 cm inter- vals down to 35 cm depth. Porewater from each subsample was subsequently extracted and ®ltered through precombusted Whatman glass ®ber ®lters (1.2 and 0.7 mm) with a Reeburgh-style squeezer and collected in acid-washed, precombusted amber bottles. All samples were stored at ÿ20 or ÿ408C until analysis. Amino acid analysis Amino acid composition was determined by pre- column o-phthaldialdehyde (OPT) derivatization and separation of the components by reversed- phase high performance liquid chromatography (HPLC) followed by ¯uorescence detection (Lindroth and Mopper, 1979; Dawson and Liebezeit, 1983). Dissolved free amino acids (DFAA) were determined by direct injection of porewaters into the HPLC after reaction with the OPT-mercaptoethanol reagent (10 min after mixing with borate bu�er at pH = 13.5). For dissolved combined amino acid (DCAA) analyses, porewaters were evaporated at 100±1108C, cooled, acidi®ed with 6 N HCl, ¯ushed with N2 and hydrolysed for 22±24 h at 1108C. Samples were neutralized with NaOH, reacted with the OPT reagent and injected into the HPLC system. The concentration of DCAA was obtained by substraction of DFAA levels, after correction for the recovery yields. Total hydrolysable amino acids (THAA) were determined in 60±200 mg of dried trap material and about 1 g of wet sediments using the same hydrolysis pro- cedure with the addition of a 30 min-sonication step with 6 N HCl. After hydrolysis, the samples were centrifuged, diluted and reacted with the ¯uor- escent tag in a similar manner. Analyses were performed on a Waters 600 multi- solvent delivery system equipped with a Waters U6K injector, a 4.6 mm�15 cm, 3 mm Supelcosil LC-18 column and a Perkin-Elmer LS-5 ¯uor- escence spectrophotometer operated at 340 nm (ex- citation) and 450 nm (emission). The mobile phase components were aqueous methanol and 34 mM phosphate bu�er at pH 6.8. The gradient program varied the methanol concentration from 15 to 47% in 10 min and then increased the concentration to 77% in 20 min. The column equilibration time was 12 min and the ¯ow rate was 1 ml/min. A reference mixture of 23 authentic amino acid standards (Sigma) containing aspartic acid (Asp), glutamic acid (Glu), b-aminoglutaric acid (bGlu), muramic acid (Mur), asparagine (Asn), serine (Ser), gluta- mine (Gln), histidine (His), glycine (Gly), threonine (Thr), arginine (Arg), tyrosine (Tyr), alanine (Ala), b-aminobutyric acid (bAbu), a-aminobutyric acid (aAbu), tryptophan (Trp), methionine (Met), valine (Val), phenylalanine (Phe), isoleucine (Ile), leucine (Leu), ornithine (Orn) and lysine (Lys), was used to assign the identities. This mixture was resolved into 21 peaks, with an incomplete separation of Gln and His and the co-elution of Thr and Gly (T + G) under the speci®ed analytical conditions. Recovery assays carried out with the standards indicated that during the hydrolysis step, Asn and Gln were almost quantitatively converted to Asp and Glu whose recoveries were 170±180%, while Trp and Met experienced severe losses. Thus, the reported Asp and Glu values for DCAA and THAA include the contribution from Asn and Gln. Amino acid data have been corrected for blank values (negli- gible for THAA, 110% and 116% average signal for DFAA and DCAA, respectively) and for the recovery yields (average 79219). The recovery was optimal for DFAA analyses (mean: 97%). Due to their erratic responses, Orn and Lys were omitted. Mur, bAbu and aAbu were always below the detec- tion limits. The analytical precision of the method averaged28.1% (RSD). RESULTS AND DISCUSSION Solid phase amino acids: THAA ¯uxes, concen- trations and early diagenesis The concentration of THAA in settling particles ranged between 40 and 145 mmol/g, with higher values at the seaward site and during the summer at both stations (Table 1). These trends were followed by all organic components and re¯ect the di�erent contribution of OM-poor terrestrial detritus (high- est at L in May), and of organic-rich marine ma- terial (highest at S and during July), according to the position of the stations along the land±sea gra- dient and the seasonal cycle of primary production (Colombo et al., 1996a). The amino acid yields nor- malized to TOC were relatively low, more constant at S (208±216 mmol THAA/100 mg TOC) and increasing in July at L (168 to 225 mmol THAA/100 mg TOC), re¯ecting the shift from terrestrial runo� in the spring to more marine primary production in the summer. The calculated ¯uxes of THAA varied from 234 to 980 mmol/m2/day (Table 1) and were higher at station L in July. THAA accounted for an average of 13% of the TOC and 37% of the TN ¯uxes. The values increased in July at L due to the higher con- tribution of biogenic material, and remained more or less constant at S (Table 1). The grand mean THAA carbon ¯ux (28 mg/m2/d) represents 3.8% Amino acid biogeochemistry in the Laurentian Trough 935 of the average daily primary production calculated for a 153-day ice-free period (745 mg C/m2/d), simi- lar to the values reported for other moderately pro- ductive coastal areas (Lee and Cronin, 1982). The THAA ¯uxes measured at 150 m depth in the St. Lawrence are similar to those reported for the Brans®eld Strait (462 mmol/m2/d at 323 m depth; Liebezeit and von Bodungen, 1987) and are lower than those published for the Gulf of Trieste (390±1385 mmol/m2/d at 16 m; Faganeli, 1989); Peru Upwelling (1053 mmol/m2/d at 52 m; Lee and Cronin, 1982) and Dabob Bay (280±1625 mmol/m2/ d at 60 m; Cowie and Hedges, 1992). In these shal- lower traps, THAA accounted for a higher fraction of TOC (10±31%) and TN (40±68%) re¯ecting the greater `freshness' of this material collected in or close to the euphotic zone. The partial decay of our settling organic material is re¯ected by the %THAA-N which is considered to be a good quali- tative indicator of the degree of OM alteration (Whelan, 1977; Henrichs et al., 1984; Cowie and Hedges, 1994). According to Cowie and Hedges (1992), values of %THAA-N below 38% indicate diagenetic alteration and this parameter averaged 3625.3% in our traps. The composition of THAA gives a further indication of partial amino acid alteration (see below). The THAA levels in the 0±3 cm sediment layer ranged from 31 to 47 mmol/g (Table 2) with higher values at the landward station (4222.6 versus 3522.7 mmol/g at S) re¯ecting the stronger OM ¯uxes, higher sedimentation rates and deeper bio- turbation observed at this site (Colombo et al., 1996b). This di�erence disappeared at the bottom of the cores where THAA levels declined to 15±18 mmol/g (Fig. 2). The amino acid yield of the sedi- ments in the top 3 cm averaged 177±201 mmol THAA/100 mg TOC, similar to the traps, but only 111±127 mmol THAA/100 mg TOC at 35 cm depth. These changes of the THAA concentrations from settling particles to surface and deeper sediments re¯ect an intense amino acid removal. In the ®rst stage of diagenetic alteration, from settling particles to the top 0±3 cm sediments, the drop of THAA levels (89239 to 3924.4 mmol/g) is proportional to that of TOC and TN as indicated by the simi- larity of the %THAA-C and THAA-N in trap par- ticles (1222.6 and 3625.3%) and sediments (1121.5 and 3925.3%, respectively). Deeper in the sediments, the decrease of the %THAA-C (10± 13 to 7±8%) and %THAA-N (30±45 to 22±28%) from the top 0±3 to 35 cm depth indicates preferen- tial removal of THAA relative to TOC and TN (Fig. 3). An important proportion of the total sedimentary decay of THAA occurs in the ®rst 5 cm of the cores (39% at L and 53% at S; Fig. 2). The loss of THAA over the entire length of the cores represents 21% of the decrease of TOC and 67±80% of that of TN, the higher percentage of TN loss corre- sponding to the seaward site. This greater THAA utilization is probably related to the predominance of marine inputs and lower sedimentation rates at this station (Colombo et al., 1996b). The pro- portions of TOC and TN remineralized as THAA in the Laurentian Trough are comparable to those reported for Buzzards Bay and Cape Lookout Bight, 11±27% of TOC and 82% of TN (Henrichs and Farrington, 1987; Burdige and Martens, 1988). Solid phase amino acids: THAA composition, sources and individual reactivity Figure 4 shows the relative composition of THAA in settling particles and sediments compared to that of fresh plankton samples. The most con- spicuous changes of THAA patterns occurs between fresh plankton and settling particles, e.g. reduction of the proportion of Asp and Glu and strong increase of Ser and T + G. Only subtle di�erences are observed between particles, surface and 30 cm- Table 1. Total concentration, ¯uxes and relative composition of total hydrolyzable amino acids measured at 150 m depth at a landward (L) and seaward (S) stations in the Lower St. Lawrence Estuary Mole% Sample Total (mmol/g) THAA- C (%) THAA- N (%) Flux (mmol/ m2/d) Asp Glu Ser His T + G Arg Tyr Ala Val Phe Ile Leu L, May 12 40.3 9.6 34.1 423.0 12.2 11.2 21.8 ± 22.4 4.0 1.8 12.7 2.1 2.2 4.2 5.3 L, May 12 57.9 10.3 36.2 666.0 10.0 9.7 18.5 2.3 29.0 2.1 1.5 11.5 3.6 2.9 4.0 5.0 L, May 13 48.5 10.8 37.8 262.0 11.8 12.0 20.0 1.7 17.8 4.9 2.7 12.2 4.3 2.6 5.1 4.9 Mean 48.9 10.2 36.0 450.3 11.3 11.0 20.1 1.3 23.1 3.7 2.0 12.1 3.3 2.6 4.4 5.1 L, July 15 86.7 13.8 39.4 980.0 9.4 9.4 30.5 0.8 23.2 ± 1.3 11.4 4.2 3.6 3.2 2.7 S, May 9 93.7 13.8 37.0 646.0 9.9 11.0 29.6 1.9 20.0 2.3 1.4 6.5 4.2 4.2 8.8 ± S, May 10 124.0 15.9 42.0 421.0 8.0 9.2 28.5 4.8 22.1 3.1 1.6 9.0 3.5 3.4 3.1 3.8 S, May 10 66.1 8.4 24.0 245.0 9.3 8.9 29.3 ± 20.5 0.9 2.2 10.1 4.7 4.0 4.9 5.4 Mean 94.6 12.7 34.3 437.3 9.0 9.7 29.1 2.7 20.9 2.1 1.7 8.5 4.1 3.9 5.6 3.0 S, July 17 145.0 13.6 37.1 268.0 8.2 8.8 22.9 4.0 22.6 5.5 3.2 11.9 3.8 2.9 2.5 3.8 S, July 18 137.0 12.8 37.5 234.0 10.2 11.1 25.1 ± 17.8 2.9 2.8 13.7 4.2 3.2 3.8 5.0 Mean 141.0 13.2 37.3 251.0 9.2 10.0 24.0 2.0 20.2 4.2 3.0 12.8 4.0 3.0 3.1 4.4 ±: below detection limits. J. C. Colombo et al.936 deep sediments, whose amino acid spectra are domi- nated by Ser and T + G (20±30% THAA each) fol- lowed by Asp (10±15%), Ala (110%) and Glu (110%) whereas in plankton, Ala, Asp and Glu are more abundant. As has been observed by other authors (Bamstedt, 1986; Cowie and Hedges, 1992), the amino acid patterns of phytoplankton and zoo- plankton are very similar, except perhaps for a sig- ni®cantly higher proportion of Asp in phytoplankton (Fig. 4). Relatively invariant THAA distributions have often been reported for biotic and abiotic compart- ments of di�erent environments re¯ecting the poor performance of these compounds as biomarkers of OM sources (Siezen and Mague, 1978; Rosenfeld, 1979; Lee et al., 1983; Henrichs et al., 1984; Henrichs and Williams, 1985; Montani and Okaichi, 1985; Henrichs and Farrington, 1987; Burdige and Martens, 1988; Cowie and Hedges, 1992). More speci®c amino acid signatures have been reported for bacteria (Mur, Glu and perhaps bGlu), marine invertebrates (taurine and Gly; Henrichs et al., 1984; Henrichs and Farrington, 1987) and the cell-wall protein of diatoms which is enriched in Gly, Ser and Thr (Hecky et al., 1973). The abundance of these latter amino acids as well as their relative increase in deeper suspended ma- terial, sinking particles and sediments have been attributed to the selective preservation of this resist- ant complex (Siezen and Mague, 1978; Lee and Table 2. Total concentrations and major components of total hydrolyzable (THAA), dissolved combined (DCAA) and dissolved free (DFAA) amino acids in sediment cores collected at a landward (L) and seaward (S) stations in the lower St. Lawrence Estuary THAA DCAA DFAA mole % mole % mole % Depth (cm) Sample total (mmol/g) Asp Glu Ser T+G Ala total (mM) Asp Glu Ser T+G Ala total (mM) Asp Glu bGlu Ser T+G Ala 0±0.6 L, May 43.0 11.2 8.9 23.2 25.1 9.8 2.5 2.8 ± 20.2 49.3 ± 2.2 4.5 24.5 4.1 7.7 5.9 24.1 L, July 44.0 12.5 11.8 23.1 24.8 9.5 2.7 ± ± 17.8 23.3 4.1 3.1 5.2 41.6 11.9 4.5 2.3 15.6 S, May 37.2 10.5 9.2 23.0 22.7 11.6 4.4 0.2 ± 14.6 85.3 ± 5.8 3.3 38.3 2.9 6.4 2.9 34.5 S, July 30.6 12.3 10.7 25.0 22.7 9.8 2.3 3.0 ± 15.7 31.3 6.7 1.8 3.9 45.0 12.8 3.9 4.4 7.2 0.6±1 L, May 38.8 11.4 7.3 21.9 21.6 10.8 0.8 13.8 35.0 22.2 3.4 26.4 2.4 4.2 42.5 12.5 6.7 4.6 10.0 L, July 43.1 12.2 13.1 24.0 24.1 10.0 4.0 3.3 ± 24.4 41.1 11.5 5.2 3.8 41.9 17.1 2.5 2.3 15.6 S, May 33.4 11.7 9.8 24.7 20.5 11.4 1.1 ± ± 31.7 62.7 ± 1.6 9.4 45.0 13.1 6.9 3.8 15.0 S, July 36.2 12.7 10.9 25.3 22.3 9.6 2.0 ± ± 22.7 49.9 9.9 4.0 5.0 47.0 18.0 2.5 3.5 6.8 1±2 L, May 39.7 11.7 8.9 23.3 23.4 11.7 3.4 ± ± 4.0 87.6 3.6 3.5 4.3 45.7 12.9 4.6 3.1 10.9 L, July 41.0 10.2 10.9 21.3 28.1 8.5 2.1 4.8 ± 26.1 28.1 11.9 8.8 2.1 35.7 13.2 1.6 1.3 35.5 S, May 33.4 14.4 12.9 28.2 17.3 11.4 3.5 6.3 12.9 23.4 32.3 13.5 4.0 4.2 37.8 10.0 4.5 5.5 23.3 S, July 36.2 12.1 12.1 25.1 20.4 10.9 5.9 4.2 ± 12.9 78.6 0.6 6.6 5.2 34.9 11.7 3.8 8.3 10.8 2±3 L, May 46.9 12.0 10.0 25.3 22.4 11.3 3.5 7.4 25.7 19.7 26.7 9.1 3.0 3.7 43.7 16.0 6.7 4.7 10.0 L, July 41.7 11.2 12.6 24.9 21.8 10.1 4.8 5.8 ± 27.4 39.4 ± 12.6 3.3 37.1 16.3 4.1 5.2 9.2 S, May 35.6 12.9 10.6 25.5 19.9 11.2 ± 2.0 4.5 51.5 17.0 6.5 3.5 11.5 S. July 39.5 10.6 9.6 24.1 26.6 11.1 8.7 6.6 ± 17.0 43.9 11.5 4.2 7.1 49.0 19.5 1.7 3.1 6.9 3±4 L, May 33.5 12.8 9.9 27.2 20.0 11.3 2.6 0.4 ± 24.1 45.2 6.4 3.2 3.7 52.2 21.9 2.8 1.6 6.3 L, July 34.1 10.6 9.8 24.3 24.3 10.0 5.9 8.8 ± 25.9 32.9 1.8 9.1 2.5 43.2 21.9 2.6 2.9 11.6 S, May 35.4 13.4 10.8 25.5 20.9 9.8 6.7 4.5 4.6 19.8 34.5 8.6 2.1 4.3 50.0 16.7 6.2 4.3 12.4 S, July 37.3 11.4 9.9 24.8 22.3 10.2 24.1 1.4 ± 4.5 37.6 21.5 9.4 6.4 41.2 15.3 3.7 5.1 7.1 4±5 L, May 34.0 11.6 9.3 26.3 22.3 11.1 2.6 9.2 ± 32.6 44.0 ± 4.1 2.9 47.9 24.5 2.6 3.1 6.9 L, July 30.8 11.2 12.8 22.8 23.2 9.6 0.5 ± ± ± 100.0 ± 11.0 2.8 42.9 28.2 2.9 3.1 4.2 S, May 22.0 13.4 10.4 25.7 22.4 11.2 4.2 5.7 ± 29.1 41.9 9.9 4.0 6.4 46.4 21.0 4.6 2.1 6.9 S, July 28.3 13.6 12.6 25.7 22.0 9.6 13.1 6.9 ± 22.7 47.3 10.8 11.1 5.4 42.4 17.8 3.1 3.7 7.8 9±10 L, May 33.4 12.1 9.6 25.6 21.8 9.9 20.3 6.2 ± 13.2 63.0 7.8 10.4 2.9 31.7 21.8 5.5 6.3 7.4 L, July 28.2 9.7 11.5 21.5 33.9 9.4 25.1 11.9 ± 19.2 39.0 12.2 15.9 3.6 31.4 16.7 4.1 7.3 7.5 S, May 20.0 14.0 9.9 27.2 23.2 10.3 16.2 9.0 5.2 28.9 39.4 9.5 5.3 3.8 28.3 18.5 10.4 6.8 13.4 S, July 22.2 11.9 10.3 26.7 26.7 9.9 13.4 4.2 ± 27.4 49.3 11.4 5.0 3.6 45.2 31.0 1.8 2.4 4.2 14±15 L, May 25.9 13.3 9.8 23.0 22.1 11.6 5.0 17.6 12.2 14.5 15.3 3.6 8.4 4.3 28.0 21.4 13.1 7.7 9.2 L, July 21.9 11.2 12.0 24.3 26.2 9.4 11.6 5.6 ± 28.8 30.7 13.5 5.3 2.4 45.3 39.8 1.7 2.3 2.3 S, May 18.2 15.0 7.7 28.2 17.6 12.1 17.6 6.9 8.6 31.1 34.6 10.8 6.1 5.6 22.6 17.5 17.5 9.5 8.2 S, July 24.6 10.9 8.4 23.6 30.5 9.1 32.2 9.2 ± 23.9 36.0 12.4 10.1 3.9 28.5 19.7 4.5 7.2 6.1 19±20 L, May 26.1 10.2 7.1 24.6 30.9 10.8 8.5 5.2 1.4 26.1 46.8 12.2 7.2 4.3 23.6 24.4 17.9 7.9 6.7 L, July 20.7 10.9 12.5 23.9 23.1 10.8 10.5 7.6 1.1 31.9 30.4 7.7 4.2 1.7 41.9 44.3 1.7 2.6 1.9 S, May 14.4 13.5 8.4 25.9 19.0 7.5 12.6 5.9 3.8 37.8 23.4 6.3 3.4 5.0 28.5 23.5 10.9 14.1 7.1 S, July 18.1 14.5 10.1 28.0 25.3 8.0 9.8 5.1 ± 42.4 32.4 8.3 4.9 4.1 35.3 32.2 1.6 6.9 1.8 24±25 L, May 19.2 12.7 9.3 24.0 22.8 12.7 6.9 5.5 ± 21.7 44.4 12.9 2.9 3.1 31.5 35.0 11.9 5.8 5.0 L, July 20.2 13.1 10.9 27.0 20.7 10.9 15.6 8.3 5.5 20.9 27.6 12.4 2.3 2.7 36.3 36.3 3.0 3.3 6.0 S, May 23.8 14.4 10.0 26.4 19.3 9.8 15.1 5.3 8.7 41.1 28.6 11.7 2.8 6.9 20.7 16.9 22.0 10.0 10.2 S, July 18.6 12.9 9.2 27.3 20.7 10.8 16.8 7.1 2.3 28.1 32.4 9.0 3.7 2.7 38.6 34.8 3.4 4.1 3.8 29±30 L, May 21.5 12.1 8.7 25.1 21.3 10.1 8.7 4.5 ± 18.7 53.5 7.5 2.9 1.4 35.9 37.2 8.6 3.4 4.1 L, July 22.9 11.1 10.7 23.5 28.7 8.6 11.2 7.1 2.5 33.9 37.2 6.5 2.3 3.0 38.7 44.3 5.7 4.8 2.6 S, May 19.1 13.9 9.3 27.9 19.0 11.3 11.4 5.2 2.1 16.4 50.0 17.3 2.8 3.9 30.0 34.6 7.9 5.7 3.6 S, July 18.0 13.4 8.2 28.7 27.6 13.0 10.1 4.0 ± 41.2 33.3 12.4 3.7 9.7 33.8 28.4 3.2 3.2 4.9 34±35 L, May 2.7 4.2 ± 23.8 51.0 10.5 2.7 1.5 38.1 46.3 3.7 1.5 2.2 L, July 16.8 13.2 9.6 27.4 26.6 8.5 2.7 6.7 ± 41.3 29.5 6.8 2.7 4.4 42.6 45.6 4.4 ± 0.4 S, May 14.9 14.6 8.1 26.3 23.0 12.7 1.4 4.1 0.5 44.4 31.5 8.4 1.4 2.1 36.4 32.9 9.3 5.0 4.3 S, July 17.8 14.3 9.1 28.9 25.0 11.9 2.7 4.1 ± 32.2 34.3 8.1 2.8 4.6 33.9 26.4 4.6 6.4 5.7 Amino acid biogeochemistry in the Laurentian Trough 937 Cronin, 1984; MuÈ ller et al., 1986; Burdige and Martens, 1988; Cowie and Hedges, 1992). The marked abundance of Ser and T + G in our trap and sediment samples relative to fresh plank- ton, which is the main amino acid source (see below), would then re¯ect the selective preservation of the diatom cell-wall complex. This further sup- ports the interpretation of a partial degradation of Fig. 2. Sediment porewater (DFAA and DCAA) and solid phase (THAA) pro®les of amino acids at both stations and sampling periods. Fig. 3. Proportion of carbon and nitrogen present as THAA in sediments and settling particles (above 0 depth). Average conversion factors are 0.44 for C and 0.14 for N (n= 17). J. C. Colombo et al.938 the trap material. From the three amino acids enriched in diatom cell-walls, Ser appears as the most sensitive tracer: its relative abundance covar- ied with those of diatom lipid biomarkers (heneico- sahexaene, fatty acid 16:1) and the chlorophyll a/ Pheopigment ratio (Colombo et al., 1996c), which presented highest values in the traps of L July and specially S May (Fig. 4) when an early diatom bloom was sampled. This signal is also detectable in underlying sediments: as has been observed for marine lipids and diatom biomarkers (Colombo et al., 1997), the proportion of Ser is generally higher at the seaward site, particularly in the core collected at S in July during the spring diatom bloom (Fig. 5, Table 2). This provides good ®eld evidence for a diatom source of Ser. The increasing proportion of Ser with sediment depth indicates the selective pres- ervation of the diatom protein±silica complex. The poorer sensitivity of Thr and Gly as diatom tracers might be related to their lower enrichment in the cell-walls. In the ®ve species studied by Hecky et al. (1973), the di�erence between cell-wall and cell con- tents averaged 1.621.4% for Thr, 4.824.2% for Gly and 9.026.7% for Ser. Estimates of contribution of marine and terres- trial inputs to THAA in the sediments can be made Fig. 4. Individual amino acid composition of THAA in fresh plankton, trap material, surface and deep sediments. Amino acid biogeochemistry in the Laurentian Trough 939 by using the amino acid yields of Cowie and Hedges (1992) for phytoplankton, zooplankton and terrestrial vascular plants (55±112, 65±137 and 0.4± 17.4 mg THAA/100 mg OC, respectively) and cal- culations from our previous work (Colombo et al., 1996a) indicating 50% of TOC from terrestrial sources (equivalent to 22.3 mg TOC/g) and 29% from phytoplankton (12.9 mg TOC/g) in the trap material. This calculation shows that most THAA are of marine origin even if the relevant contri- bution of zooplankton is not estimated, i.e. from 50 to 100% of THAA would come from algae and only about 13% from terrestrial plants. In coastal environments, the composition of THAA generally shows no clear change with depth in the sediment (Henrichs et al., 1984; Henrichs and Farrington, 1987; Cowie and Hedges, 1992). However, some authors reported a preferential removal of acidic versus the more stable basic amino acids (Rosenfeld, 1979; Gonzalez et al., 1983; Burdige and Martens, 1988). In the Laurentian Trough some subtle compositional changes can be discerned, notably the relative increase of Ser and Asp and slight decrease of Glu and His (Fig. 5). The other amino acids showed very weak and not signi®cant trends (p= 0.05): Arg, Val, Phe, Ile and Leu tended to decrease, T + G and Tyr increased slightly and Ala showed more constant pro®les (Table 2). Serine seems to be the most stable amino acid. Its relative increase with depth in the sediments is more pronounced than that of Asp (overall concen- tration versus depth regression slopes = 0.08 and 0.05; r= 0.45 and 0.39, n = 47, respectively). In contrast, the proportion of Glu, which shows sur- face values similar to Asp, decreases with depth (slope =ÿ 0.05; r =ÿ 0.37). The protection of the cell-wall matrix may also explain the unusual rela- tive persistence of Asp. Although Ser is one of the most abundant amino acids in diatom cell-walls, Fig. 5. Variation of the relative abundance of selected solid phase (THAA) and porewater amino acids (DCAA and DFAA) with sediment depth. Trap values are indicated for THAA. Filled points of THAA SER correspond to the cores collected at the seaward site. J. C. Colombo et al.940 Asp is also an important component, and in some cases it is enriched, i.e. Cyclotella cryptica, whereas Glu is more abundant in the cell contents (Hecky et al., 1973). Porewater amino acids: DFAA and DCAA concen- trations and early diagenesis The concentrations of DFAA in the sediment porewaters ranged from 1.4 to 16 mM, with the highest levels occurring in the 5±15 cm section. DCAA levels were low and similar to those of DFAA in the top 5 cm and also showed sub-surface maxima (5±32 mM) at 10±15 cm depth (Fig. 2; Table 2). Both amino acid fractions combined rep- resented 2±32% of porewater DOC. DFAA accounted for 2±17% DOC in the ®rst 10 cm and 1.4% at the bottom of the cores. DCAA contri- bution was higher at 10 cm (10±19% DOC) and decreased at the top and bottom of the cores. These DFAA concentrations are similar to those reported for the Limfjord, Denmark (1±21 mM) and Georgia salt marsh soils (0.4±8.8 mM), where the pro®les also exhibited sub-surface maxima at 4±6 and 10± 15 cm, respectively (Gardner and Hanson, 1979; Jorgensen et al., 1981). Much higher DFAA levels have been reported for the Peru Upwelling region (1±5 to 10±200 mM), Buzzards Bay (5 to 30±60 mM) and Cape Lookout Bay (2±5 to 20±60 mM), where DFAA generally decreased exponentially with core depth (Henrichs et al., 1984; Henrichs and Farrington, 1987; Burdige and Martens, 1990). These exponentially decreasing pro®les are inter- preted in the context of a general model where DFAA and DCAA are important intermediates in the mineralization of OM (Henrichs et al., 1984; Burdige and Martens, 1988, 1990). Near the oxic±a- noxic interface, the hydrolysis of peptides and pro- teins from the solid phase results in higher concentrations of DFAA and DCAA in the pore- waters. These amino acids are then converted by fermentative processes to volatile fatty acids, H2 or methylated amines which are utilized by sulphate reducing and methanogenic bacteria. DFAA and DCAA can also be incorporated back to the solid phase by abiotic reactions (adsorption, conden- sation) or by inclusion into bacterial biomass (Burdige, 1989, 1991). In sediments from the Laurentian Trough, how- ever, the low levels of DFAA and DCAA coincide with the sur®cial zone of most intense solid phase decay (Fig. 2). This pattern indicates that removal mechanisms (i.e. consumption, irrigation, di�usion) are very intense in this suboxic±anoxic sur®cial sedi- ment layers (the oxygenated layer is restricted to top 0.5 cm). The presence of clear near-surface (2±4 cm) minima was also observed in the pro®les of total DOC and porewater carbohydrates and has been attributed to an intense consumption of DOM, probably enhanced by biologically mediated exchange of solutes within the sediment, and to bioirrigation to overlying waters (Colombo et al., 1996b). The mineralization of DFAA by bacteria is very intense and accounts for a signi®cant fraction of the ammonium production in anoxic sediments (Burdige, 1991). Tracer experiments indicated that more than 70% of added alanine was metabolized in 2±4 h with turnover times of 5±9 min (Christensen and Blackburn, 1980). In addition to the catabolic uptake of DFAA, their anabolic in- corporation into bacterial biomass can also play a signi®cant role. This process was suspected to account for 20±40% of the added amino acids in sediment slurry experiments (Burdige, 1991). DCAA also constitute a source of carbon and nitrogen for bacteria (Co�n, 1989; Rosenstock and Simon, 1993), although DCAA usually show slower turnover times. Another potentially important sink of DFAA is the transepidermal uptake by benthic organisms. This process occurs even against strong concen- tration gradients and could cover a major part of the organism energy requirements (Jorgensen, 1976; Stephens et al., 1978). In Buzzards Bay, uptake by benthic organisms was considered insigni®cant because the calculated potential turnover time of amino acids was higher than the values currently measured in sediments (Henrichs and Farrington, 1987). A di�erent conclusion is obtained for the Laurentian Trough. Using the same rate of uptake utilized for Buzzards Bay (0.5 mmol/g ww/h) and considering only the polychaete biomass in the ®rst 13 cm (1220 individuals/m2*0.5 g each = 110 g ww/m2; Ouellet, 1982), the rate of uptake would be 10 nmol/cm3/d and the turnover time 19 h, in the range of those reported for marine sediments (<1 to 24 h). This suggests that macrofaunal uptake could be a complementary sink of DFAA in the Laurentian Trough. This conclusion would be re- inforced if more recent, higher polychaete abun- dances (Nehr, 1991) and the presence of echinoderms (8±40/m2), crustaceans (120/m2) and bivalves (119/m2) are taken into account. Below 3±5 cm, increasing levels of dissolved amino acids (Fig. 2) indicate that the rates of pro- duction outstrip their removal by biological uptake, biodi�usion or adsorption. The presence of sub-sur- face maxima of DFAA has been related to enhanced microbial activity due to sulphate re- duction or to the transition from sulphate reduction to methanogenesis (Jorgensen et al., 1981; Burdige and Martens, 1990). In the Laurentian Trough the data do not allow any ®rm conclusion: at S the sub-surface amino acid maxima coincide with the depth (17 cm) of maximum sulphate reduction reported for an intermediate site (Edenborn et al., 1987) but at station L the amino acid maxima seem Amino acid biogeochemistry in the Laurentian Trough 941 too shallow (10 cm) to be related with a peak of sulphate reduction. At the bottom of the cores, microbial reminerali- zation and adsorption onto particles probably pre- dominate, resulting in lower amino acid levels. The consistently higher concentrations of DCAA indi- cate that they are produced at higher rates, adsorbed to a lesser degree or, more likely, that DCAA are removed more slowly than DFAA. This latter assumption is supported by the greater chemi- cal complexity of DCAA. For estuarine waters, three di�erent DCAA fractions have been pro- posed: labile proteins (<10%, bacterial turnover times = 428 h), protein±carbohydrate conden- sation products (150%; turnover times = 2882344 h) and a nonproteinaceous fraction bound to col- loids or small particles (Keil and Kirchman, 1993). Porewater amino acids: DFAA and DCAA compo- sition, sources and individual reactivity The composition of DFAA and DCAA show remarkably di�erent patterns (Table 2). DCAA re- semble most the patterns of the source biopolymers from the solid phase (THAA) and show a marked abundance of Ser and T + G. In contrast the com- position of DFAA show a predominant contri- bution of Glu and bGlu, probably originating from bacteria. This indicates that although DCAA may be considered as precursors of DFAA, the compo- sition of this more reactive fraction is strongly in¯u- enced by other factors such as the abundance or activity of bacteria. Previous studies in salt marshes and marine sedi- ments reported similar Glu and bGlu-dominated DFAA patterns (Gardner and Hanson, 1979; Jorgensen et al., 1981; Henrichs et al., 1984; Henrichs and Farrington, 1987; Burdige and Martens, 1990). The most probable source of Glu in porewaters is bacteria because Glu is usually the major component of bacterial free amino acid pools. b-Aminoglutaric acid has also been found among the free amino acids of some marine bac- teria (Henrichs et al., 1984; Henrichs and Farrington, 1987). The high porewater levels of Glu have also been attributed to its preferential mobiliz- ation from the solid phase (Burdige and Martens, 1990). These authors suggested that the selective utilization of Glu and Ala and preservation of Gly in the solid phase (indicated by their opposing mol% trends in THAA) could explain the relative enrichment of Glu and Ala and the depletion of Gly in porewaters relative to THAA. Our amino acid patterns also show an enrichment of Glu and Ala and a depletion of T + G in sur®cial pore- waters relative to THAA, but only Glu showed a moderate decrease in the solid phase (Table 2). Moreover, if DCAA are intermediaries in the for- mation of DFAA, a selective mobilization of Glu from the solid phase would also produce an enrich- ment of Glu in the DCAA pool and the proportion of Glu in this fraction is very low (Table 2). Another factor which could in¯uence the distri- bution of DFAA is adsorption onto particles. However, this process would only contribute to decreasing porewater levels of basic amino acids (Lys, His, Arg, Orn) which react more rapidly with OM and are preferentially adsorbed by clay min- erals (Rosenfeld, 1979; Hedges and Hare, 1987; Henrichs and Farrington, 1987; Henrichs and Sugai, 1993). The selective uptake or release of amino acids by macrofauna could be another relevant process. The transepidermal absorption of amino acids involves the existence of speci®c membrane carriers, and a preferential uptake of Ala and Gly has been reported for polychaetes, echinoderms and mollusca (Stephens et al., 1978; Stewart, 1979) whereas acidic amino acids (Glu and Asp) show very low or null rates of transport. Thus, when macrofaunal absorp- tion is signi®cant, as it seems to be for the Laurentian Trough, the selective uptake of amino acids could contribute to the abundance of Glu relative to Ala or Gly. In contrast to THAA, which showed only subtle downcore trends, the composition of porewater amino acids showed marked changes with depth in the sediments, notably the relative increase of Ser in DCAA and of bGlu in the DFAA fraction and the decrease of Gln, Ala and Glu in the DFAA pool. The relative contribution of Ser to DFAA increases between 9±25 cm and decreases towards the bottom of the cores (Table 2). A clearer down- core relative increase of Ser is observed in the DCAA fraction (Fig. 5) suggesting that this is the most stable amino acid, both in the solid phase and porewaters. The mol% contributions of Gln and Ala to DFAA show a signi®cant decrease with sedi- ment depth. Glutamine is seldom detected below 20 cm, whereas Ala shows a moderate decrease in the core collected at S in July (7±11 to 5±6%) and a strong decay in the other three cores (16±35 to 0.4± 4%; Fig. 5). The marked decrease of Ala is consist- ent with its rapid biological uptake and turnover in marine sediments (Christensen and Blackburn, 1980). However, in the solid phase or in DCAA, Ala did not show any signi®cant trend (Table 2), suggesting that polymerized Ala is not as readily bioavailable. The opposing trends followed by the relative pro- portions of Glu (slope =ÿ 0.28, r =ÿ 0.42) and bGlu (slope = 0.74, r= 0.80, n = 48) in the DFAA pool (Table 2) strongly suggest the preferen- tial degradation of Glu and its conversion into bGlu. According to Henrichs et al. (1984), the rela- tive increase of bGlu in deeper porewaters could be related to (1) changes in the excretion of DFAA by bacteria due to varying chemical conditions of the sediments, (2) di�erences in the decomposition rates J. C. Colombo et al.942 of both amino acids or (3) the biologically-mediated formation of bGlu from Glu. The fact that the bGlu/Glu values increase smoothly with depth in our cores regardless of whether the concentrations increase (0±10 cm) or decrease (10±35 cm; Fig. 6), suggests that preferential decay and conversion are coupled, i.e. some fraction of Glu is converted to bGlu during the utilization of the former. This is in agreement with sediment slurry experiments, where the addition of Glu lead to a small (1%) transient build-up of bGlu (Burdige, 1989). The evolution of the bGlu/Glu values with sedi- ment depth also indicates a geographical di�erence. The ratios show similar initial values and rates of increase in the ®rst two cm at both stations. Below this depth, the ratios increase more rapidly at L and a clear divergence is observed below 10±15 cm (Fig. 6). These pro®les suggest a deeper activity of bacteria with extended selective degradation and conversion of Glu at the landward site. These trends could be related to the higher sedimentation rates and bioturbation observed at the landward station which favor the downward transport of still undegraded OM (Colombo et al., 1996b), thus allowing the continued metabolism of bacteria in deeper sediment layers. CONCLUSIONS Total hydrolysable (THAA), dissolved free (DFAA) and combined (DCAA) amino acids were analyzed in settling particles and the solid phase and porewaters of underlying sediments in the Laurentian Trough to evaluate their sources and in- dividual reactivity during early diagenesis. THAA ¯uxes measured at mid-depth in the water column (234±980 mmol/m2/d) represented 3.8% of the aver- age daily primary production, comparable to other moderately productive deep coastal areas. The settling organic material showed relatively low %THAA-C and THAA-N indicating partial degra- dation. Sur®cial sediments presented an average 56% decrease of THAA levels but similar %THAA-C and %THAA-N compared to settling particles suggesting that TOC, TN and THAA decompose at similar rates. Deeper in the cores, however, THAA were preferentially degraded and accounted for 21% of TOC reduction and 67±80% of TN loss. THAA composition in settling particles and sedi- ments was relatively uniform and showed an enrich- ment of serine and threonine + glycine relative to fresh plankton, re¯ecting the selective preservation of diatom cell-walls. Serine was the most speci®c diatom tracer and its proportion covaried with that of diatom lipid biomarkers. Its relative abundance was higher at a seaward site, which receive stronger inputs of marine OM and increased downcore re¯ecting selective preservation of the diatom cell- wall matrix. Glutamic acid and histidine were pre- ferentially degraded with depth in the sediments. Porewater DFAA and DCAA accounted for 3± 25% of total DOC. Low levels in the surface zone of most intense solid phase THAA decay point to strong biological uptake. Deeper in the cores, DFAA were preferentially consumed relative to DCAA. Both fractions showed remarkably di�erent compositions: DCAA were dominated, as was solid phase THAA, by serine and threonine + glycine whereas DFAA were enriched in glutamic and b- aminoglutaric acids of bacterial origin. In contrast to the subtle trends observed for solid phase THAA, the relative abundance of porewater amino acids showed marked changes with sediment depth, re¯ecting their greater reactivity. bGlu increased and Ala and Glu decreased in the DFAA pool and Ser increased in DCAA. Increases of bGlu/Glu values with sediment depth are attributable to pre- ferential degradation and conversion of Glu into bGlu. These processes were more intense at a land- ward station, where the higher sedimentation rates and bioturbation favor the burial of relatively unde- graded OM, allowing the continued metabolism of bacteria in deeper sediment layers. Associate EditorÐS. Wakeham REFERENCES Bamstedt, U., 1986. Chemical composition and energy content. In: Corner, E.D.S., O'Hara, S.C.M. (Eds.), The Biological Chemistry of Marine Copepods. Clarendon Press, pp. 1±58. Fig. 6. Evolution of the ratio b-aminoglutaric acid (bGLU)/glutamic acid (GLU) in porewaters from both stations and sampling periods. Amino acid biogeochemistry in the Laurentian Trough 943 Burdige, D. J. (1989) The e�ects of sediment slurring on microbial processes, and the role of amino acids as sub- strate for sulphate reduction in anoxic marine sedi- ments. Biogeochemistry 8, 1±23. Burdige, D. J. (1991) Microbial processes a�ecting alanine and glutamic acid in anoxic marine sediments. FEMS Microbiol. Ecol. 85, 211±232. Burdige, D. J. and Martens, C. S. (1988) Biogeochemical cycling in an organic-rich coastal marine basin. 10. The role of amino acids in sedimentary carbon and nitrogen cycling. Geochim. Cosmochim. Acta 52, 1571±1584. Burdige, D. J. and Martens, C. S. (1990) Biogeochemical cycling in an organic-rich coastal marine basin. 11. The sedimentary cycling of dissolved, free amino acids. Geochim. Cosmochim. Acta 54, 3033±3052. Christensen, D. and Blackburn, T. H. (1980) Turnover of tracer (14C, 3H labelled) alanine in inshore marine sedi- ments. Mar. Biol. 58, 97±103. Co�n, R. B. (1989) Bacterial uptake of dissolved free and combined amino acids in estuarine waters. Limnol. Oceanogr. 34, 531±542. Colombo, J. C., Silverberg, N. and Gearing, J. N. (1996a) Biogeochemistry of organic matter in the Laurentian Trough. I. Composition and vertical ¯uxes of rapidly settling particles. Mar. Chem. 51, 277±293. Colombo, J. C., Silverberg, N. and Gearing, J. N. (1996b) Biogeochemistry of organic matter in the Laurentian Trough. II. Bulk composition of the sediments and rela- tive reactivity of major components during early diagen- esis. Mar. Chem. 51, 295±314. Colombo, J. C., Silverberg, N. and Gearing, J. N. (1996c) Lipid biogeochemistry in the Laurentian Trough. I. Fatty acids, sterols and aliphatic hydrocarbons in rapidly settling particles. Org. Geochem. 25, 211±225. Colombo, J. C., Silverberg, N. and Gearing, J. N. (1997) Lipid biogeochemistry in the Laurentian Trough. II. Changes in composition of fatty acids, sterols and ali- phatic hydrocarbons during early diagenesis. Org. Geochem. 26, 257±274. Cowie, G. L. and Hedges, J. I. (1992) Sources and reactiv- ities of amino acids in a coastal marine environment. Limnol. Oceanogr. 37, 703±724. Cowie, G. L. and Hedges, J. I. (1994) Biochemical indi- cators of diagenetic alteration in natural organic matter mixtures. Nature 369, 304±307. Dawson, R., Liebezeit, G., 1983. Determination of amino acids and carbohydrates. In: Grassho�, K., Ehrhardt, K., Kremling, K. (Eds.) Methods of Seawater Analysis, 2nd ed. Verlag Chemie, pp. 319±340. Edenborn, H. M., Silverberg, N., Mucci, A. and Sundby, B. (1987) Sulphate reduction in deep coastal marine sediments. Mar. Chem. 21, 329±345. Faganeli, J. (1989) Sedimentation of particulate nitrogen and amino acids in shallow coastal waters (Gulf of Trieste, Northern Adriatic). Mar. Chem. 26, 67±80. Gardner, W. S. and Hanson, R. B. (1979) Dissolved free amino acids in interstitial waters of Georgia salt marsh soils. Estuaries 2, 113±118. Gonzalez, J. M., Grimalt, J. and Albaige s, J. (1983) Amino acid composition of sediments from a deltaic en- vironment. Mar. Chem. 14, 61±71. Hecky, R. E., Mopper, K., Kilham, P. and Degens, E. T. (1973) The amino acid and sugar composition of dia- tom cell-walls. Mar. Biol. 19, 323±331. Hedges, J. I. and Hare, P. E. (1987) Amino acid adsorp- tion by clay minerals in distilled water. Geochim. Cosmochim. Acta 51, 255±259. Henrichs, S. M. and Farrington, J. W. (1987) Early dia- genesis of amino acids and organic matter in two coastal marine sediments. Geochim. Cosmochim. Acta 51, 1±15. Henrichs, S. M. and Sugai, S. F. (1993) Adsorption of amino acids and glucose by sediments of Resurrection Bay, Alaska, USA: functional group e�ects. Geochim. Cosmochim. Acta 57, 823±835. Henrichs, S. M. and Williams, P. M. (1985) Dissolved and particulate amino acids and carbohydrates in the sea surface microlayer. Mar. Chem. 17, 141±163. Henrichs, S. M., Farrington, J. W. and Lee, C. (1984) Peru upwelling region sediments near 158S. 2. Dissolved free and total hydrolysable amino acids. Limnol. Oceanogr. 29, 20±34. Jorgensen, C. B. (1976) August PuÈ tter, August Krogh, and modern ideas on the use of dissolved organic matter in aquatic environments. Biol. Rev. 51, 291±328. Jorgensen, N. O. G., Lindroth, P. and Mopper, K. (1981) Extraction and distribution of free amino acids and am- monium in sediment interstitial waters from the Limfjord, Denmark. Oceanol. Acta 4, 465±474. Keil, R. G. and Kirchman, D. L. (1993) Dissolved com- bined amino acids: chemical form and utilization by marine bacteria. Limnol. Oceanogr. 38, 1256±1270. Kirchman, D. L., Suzuki, Y., Garside, C. and Ducklow, H. W. (1991) High potential turnover rates of dissolved organic carbon during a spring phytoplankton bloom. Nature 352, 612±614. Lee, C., 1988. Amino acid and amine biogeochemistry in marine particulate material and sediments. In: Blackburn, T.H., Sorensen, J. (Eds.), Nitrogen Cycling in Coastal Marine Environments. John Wiley and Sons, pp. 125±141. Lee, C. and Cronin, C. (1982) The vertical ¯ux of particu- late organic nitrogen in the sea: decomposition of amino acids in the Peru upwelling area and equatorial Atlantic. J. Mar. Res. 40, 227±251. Lee, C. and Cronin, C. (1984) Particulate amino acids in the sea: e�ects of primary productivity and biological decomposition. J. Mar. Res. 42, 1075±1097. Lee, C., Wakeham, S. G. and Farrington, J. W. (1983) Variations in the composition of particulate organic matter in a time-series sediment trap. Mar. Chem. 13, 181±194. Liebezeit, G. and von Bodungen, B. (1987) Biogenic ¯uxes in the Brans®eld Strait: planktonic versus macroalgal sources. Mar. Ecol. Prog. Ser. 36, 23±32. Lindroth, P. and Mopper, K. (1979) High performance liquid chromatographic determination of subpicomole amounts of amino acids by precolumn ¯uorescence deri- vatization with o-phthaldialdehyde. Anal. Chem. 51, 1667±1674. Montani, S., Okaichi, T., 1985. Amino acid variations in marine particles during sinking and sedimentation in Harima-Nada, The Seto Inland Sea, Japan. In: Sigleo, A.C., Hattori, A. (Eds.), Marine and Estuarine Geochemistry. Lewis Publishers, pp. 15±27. MuÈ ller, P. J., Suess, E. and Ungerer, C. A. (1986) Amino acids and amino sugars of surface particulate and sedi- ment trap material from waters of the Scotia Sea. Deep- Sea Res. 6, 819±838. Nehr, M.O., 1991. Etude des echelles de variabilite spatiale de la macrofaune benthique du chenal de l'estuaire Maritime du St-Laurent (Quebec). Rapport de D.E.A.. Universite d'Aix-Marseilles II, France, 51 p. Ouellet, G., 1982. Etude de l'interaction des animaux benthiques avec les se diments du Chenal Laurentien. M.Sc. Thesis, Universite du Que bec aÁ Rimouski. Rosenfeld, J. K. (1979) Amino acid diagenesis and adsorp- tion in nearshore anoxic sediments. Limnol. Oceanogr. 24, 855±866. Rosenstock, B. and Simon, M. (1993) Use of dissolved combined and free amino acids by planktonic bacteria of Lake Constance. Limnol. Oceanogr. 38, 1521±1531. J. C. Colombo et al.944 Siezen, R. J. and Mague, T. H. (1978) Amino acids in sus- pended particulate matter from oceanic and coastal waters of the Paci®c. Mar. Chem. 6, 215±231. Stephens, G. C., Volk, M. J., Wright, S. H. and Backlund, P. S. (1978) Transepidermal accumulation of naturally occurring amino acids in the sand dollar, Dendraster excentricus. Biol. Bull. 154, 335±347. Stewart, M. G. (1979) Absorption of dissolved organic nutrients by marine invertebrates. Oceanogr. Mar. Biol. Ann. Rev. 17, 163±192. Wefer, G., Suess, E., Balzer, W., Liebezeit, G., MuÈ ller, P. J., Ungerer, C. A. and Zenk, W. (1982) Fluxes of bio- genic components from sediment trap deployment in cir- cumpolar waters of the Drake Passage. Nature 299, 145±147. Whelan, J. K. (1977) Amino acids in a surface sediment core of the abyssal plain. Geochim. Cosmochim. Acta 41, 803±810. Amino acid biogeochemistry in the Laurentian Trough 945