Significant benthic fluxes of bioavailable dissolved amino acids to the ocean: Results from the East/Japan Sea

We measured dissolved organic carbon (DOC) and total dissolved amino acid (TDAA) in seawater and sediment porewater of the Ulleung Basin in the East/Japan Sea. The DOC and TDAA concentrations were 1.1‐ and 1.4‐fold higher in the euphotic zone, and 11‐ and 43‐fold higher in sediment porewater, respectively, than those in the deep ocean. Consequently, in the deep ocean, TDAA and DOC input fluxes from porewater were 2‐ and 0.4‐fold of those from the euphotic zone, respectively. This larger contribution of benthic flux for TDAA and its shorter residence time in the benthic boundary layer (BBL) (1.3 ± 0.9 yr) seem to result in steep TDAA increases in the BBL, although DOC concentrations remained relatively uniform throughout the entire deep ocean. AA‐derived indices also show enhanced bioavailability of dissolved organic matter in the BBL. Benthic inputs seem to supply a significant amount of bioavailable TDAA to the deep ocean, fueling microbial activity.

and decrease with depth, which is similar to the dissolved organic carbon (DOC) distribution pattern.In the deep ocean, TDAA concentrations decrease from the North Atlantic Ocean to the North Pacific Ocean as the age of the water mass increases (Kaiser and Benner 2009).
The concentrations and compositions of amino acids (AAs) are closely related to the source and transformation of organic matter (OM) (Dauwe and Middleburg, 1998;Cowie and Hedges 1994;Amon et al. 2001).During OM degradation, AAs are preferentially consumed by heterotrophic organisms.Some labile AAs are selectively degraded, while other AAs are preferentially produced or preserved.Thus, to evaluate the origin and degradation state of OM, various AA-derived indices, such as C-or N-normalized TDAA (TDAA-C% or TDAA-N%) (Cowie and Hedges 1994), ratios of protein precursor to non-protein degradation product (Cowie and Hedges 1994), reactivity index (RI) (Jennerjahn and Ittekkot 1997), and degradation index (DI) (Dauwe et al. 1999), have been utilized.Bacteria uniquely synthesize D-amino acid (D-AA) as a component of their cell wall peptidoglycan, and thus D-AA can be a useful indicator of bacterial contributions to OM (Tremblay and Benner 2009;Bourgoin and Tremblay 2010).
Relative to AA-derived index studies, studies on the TDAA flux in the ocean are very limited, particularly in the benthic boundary layer (BBL).Fejjar et al. (2021) studied the origin and transformation mechanisms of TDAA in the sedimentwater interface of the St. Lawrence estuary by quantifying TDAA in the seawater and sediment porewater.Other studies on AA concentration and characteristics in the porewater primarily focused on DFAA (Gardner and Hanson 1979;Henrichs et al. 1984;Burdige and Martens 1990).Henrichs et al. (1984) calculated the DFAA benthic flux in the Peru upwelling region.Burdige and Martens (1990) showed the seasonal variations of DFAA benthic flux in organic-rich sediments of Cape Lookout Bight, North Carolina, USA.Landén and Hall (2000) also calculated the DFAA benthic flux in the open Skagerrak based on its seawater and porewater gradients and laboratory experiments of sediment.However, there is a significant knowledge gap related to TDAA cycling in the ocean, including the differences in the sources and cycling between TDAA and DOC, the relative significance of TDAA benthic flux to its deep ocean pool, and the bioavailability and the residence time of sediment-borne TDAA in the BBL, especially in marginal seas, which are greatly affected by continental shelf waters and organic-rich sediments.
Therefore, the aim of this study was to investigate the source and cycling of TDAA and DOC in a marginal sea.For this purpose, we analyzed the distribution of DOC and TDAA, including D-AA and L-amino acid (L-AA), in the water column with fine-scale vertical resolution in the BBL and sediment porewater of the Ulleung Basin in the East/Japan Sea (hereafter East Sea).In addition, we calculated various AA-derived OM degradation indices, the vertical fluxes of TDAA and DOC in the upper and bottom boundary layers, and the residence time of sediment-borne TDAA.The East Sea is an ideal site to study the TDAA cycling and fluxes in the water column and sediment, since it is a marginal sea with high primary production and large lateral OM inputs through the continental shelf of the East China Sea.The sediment of the East Sea is characterized by high OM content (> 2.5% dry wt.; Cha et al. 2007), high OM mineralization rates (Lee et al. 2008;Lee et al. 2019), and high benthic microbial metabolic activity (Hyun et al. 2017).

Study area
The East Sea is a semi-enclosed marginal sea located in the northwestern Pacific Ocean, with a maximum depth of over 3500 m (Fig. 1).Water exchange between the East Sea and the North Pacific Ocean is limited through several shallow straits (< 150 m) (Talley et al. 2003).This sea has its own deep-water formation and meridional circulation, which is isolated from the North Pacific Ocean.The turnover time of deep water is approximately 100 yr (Watanabe et al. 1991).The East Sea consists of three major basins: Japan, Yamamoto, and Ulleung Basin.Among these basins, the Ulleung Basin has the highest annual primary production (273 g C m À2 yr À1 ; Kwak et al. 2013).

Sampling
Seawater and sediment porewater sampling was conducted from 04 November 2020 to 11 November 2020, on board the R/V Onnuri of the Korea Institute of Ocean Science and Technology.Temperature, salinity, and dissolved oxygen (DO) concentration were measured using a CTD (SBE 911 Plus, Seabird Electronics Inc.).Seawater samples were collected at five stations (KU2, KU4, KU6, KU8, and KU10) using a Niskin sampler.Sediment was collected at three stations (C1, C2, and KU10) using a box corer.Core tubes (8 cm internal diameter, 20 cm long) were carefully inserted into the sediment in the box corer to collect undisturbed sediment core samples.
Porewater was extracted using a soil moisture sampler (Rhizon SMS, Rizosphere Research Products) through a 1-cm interval predrilled hole under a glove bag filled with N 2 gas (Kim et al. 2020).
For DOC and AA analyses, 44-seawater and 16-porewater samples were filtered on board through precombusted (450 C, 4 h) 0.7-μm Whatman GF/F filters.The filtered samples were acidified to a pH of $ 2 using 6 N HCl to suppress bacterial activity and stored in precombusted glass ampoules (Dittmar 2008).

Chemical analysis
DOC samples were analyzed using a high-temperature catalytic oxidation method with a total organic carbon analyzer (TOC-L, Shimadzu), as previously published in our laboratory (Kim et al. 2015;Kim et al. 2017).A certified reference material of deep seawater ($ 43 μM DOC; University of Miami) (Hansell 2005) was measured in every batch to validate analytical accuracy.
TDAA samples were analyzed using a high-performance liquid chromatography system (Waters 2695, Waters) equipped with an Alltech Altima C18 column (5 μm, 4.6 Â 150 mm) and a Waters 2475 fluorescence detector (excitation: 330 nm, emission: 445 nm), following liquid-phase acid hydrolysis as previously reported by our laboratory (Yan et al. 2015;Park et al. 2022).We applied a linear gradient, as presented by Dittmar et al. (2009), with a flow rate of 1.1 mL min À1 .A mixture of 19 individual AA standards was used for calibration and quantification.Procedural blanks (< 6% of the average seawater sample concentrations) were subtracted from the measured sample concentrations.The effect of chemical racemization during acidic hydrolysis was corrected using the average rates determined for free and protein AAs by Kaiser and Benner (2005) under similar conditions.TDAA was taken as the sum of seventeen individual AAs, excluding two nonprotein AAs, β-alanine (β-Ala) and γ-aminobutyric acid (GABA).During hydrolysis, asparagine (Asn) and glutamine (Gln) were deaminated and converted to aspartic acid (Asp) and glutamic acid (Glu), respectively, and reported as the sum of Asp + Asn (Asx) and Gln + Glu (Glx).C-normalized TDAA (TDAA-C%) was calculated as the percentage contribution of TDAA-C to the total DOC (Cowie and Hedges 1994).D-AA% (mol%) was calculated as the molar ratio of D-AA to TDAA.The RI was calculated as the ratio of two aromatic AAs (tyrosine and phenylalanine) to two non-protein AAs (β-Ala and GABA) (Jennerjahn and Ittekkot 1997).The dataset is available in the Mendeley Data depository (Park et al. 2023).

Statistical analysis
The statistical significance of variations in DOC and TDAA concentrations and AA-derived indices between different depth layers of seawater (the surface vs. deep layer, the deep layer vs. BBL) and sampling locations (continental shelf vs. slope and rise) was assessed using either a parametric independent sample t-test or a non-parametric Mann-Whitney U test.The data were tested for normality and homogeneity using the Shapiro-Wilk test and Levene's test, respectively.All statistical analyses were conducted using SPSS Ver. 26 Software (IBM Corporation), with the significance level set at p < 0.05.

Distributions of DOC and AAs in the seawater of the Ulleung Basin
In the surface mixed layer (0-50 m), temperature and salinity ranged from 16 C to 21 C and 32.8 to 33.9, respectively.The temperature (0.18 AE 0.05 C) and salinity (34.07) were uniform below 1000 m in the water column.The DO concentrations were relatively uniform below 1000 m (178-187 μmol kg À1 ) but decreased slightly in the BBL (0approximately 250 m above the seafloor).The DOC concentrations, ranging from 51 to 76 μM (63 AE 6 μM), were relatively higher in the surface layer (0-100 m) (U-test, p < 0.05) than the uniformly lower concentrations (51-63 μM) in the entire deep ocean (> 100 m) (Fig. 2A).

Distribution characteristics of DOC and TDAA
In the East Sea, the DOC and TDAA concentrations were higher in the surface layer, associated with biological production.Especially, the TDAA concentrations were much higher in the surface layer of the Ulleung Basin, due to the large lateral inputs of OM and nutrients through the continental shelf.In the deep ocean, while the DOC concentrations were relatively uniform throughout the entire deep ocean, the TDAA concentrations showed steep increases in the BBL, which is comparable with those reported for the coastal and estuarine areas (Daumas 1976;Hébert and Tremblay 2017;Fejjar et al. 2021) and the Guatemala Basin (Lee and Bada 1975).This increasing pattern could be associated with larger benthic inputs via the sediment-water interface and/or more rapid TDAA degradation rates in the BBL as described in the next section.
Much higher DOC and TDAA concentrations were observed in the sediment porewater relative to the deep water of the ocean.These high concentrations in the porewater, similar to other productive coastal and estuarine regions (Coffin 1989;Colombo et al. 1998;Pedersen et al. 2001;Fejjar et al. 2021), could be due to high total organic carbon (TOC) contents of the sediment in this region (> 2.5% dry wt.; Fig. 3A).Higher DOC and TDAA concentrations in the surface sediment porewater relative to the deeper layer could result from higher TOC contents in the upper sediment (Fig. 3A).
In the porewater, TDAA-C% values (8.0%AE 2.4%) were higher than those in the seawater (Fig. 3C), indicating the presence of TDAA-enriched DOM in the porewater.However, D-AA% (15% AE 4%) (Fig. 3D), RI (0.4 AE 0.3) (Fig. 3E), and Asx/β-Ala (1.6 AE 0.9) (Fig. 3F) values were similar to those in the seawater.Similarly, Fejjar et al. (2021) reported remarkably high TDAA-C% values in the porewater of estuary sediments and the inconsistency of TDAA-C% values with the other indices.This inconsistency may be attributed to different sensitivities of indices at different degradation stages (Cowie and Hedges 1994;Davis et al. 2009), showing high TDAA-C% values during the early stages of DOM degradation.
The vertical fluxes of AAs and DOC in the upper and bottom boundary layers Much higher TDAA and DOC concentrations in the sediment porewater relative to those in the deep water of the ocean suggest the net production of AAs and DOC within the sediment and their subsequent upward diffusion via the sedimentwater interface.The benthic fluxes of AAs and DOC across the sediment-water interface can be calculated according to Fick's first law of diffusion (Berner 1980), with the assumption that the vertical transport across the sediment-water interface is solely due to molecular diffusion, where F is the benthic flux, ; is the sediment porosity, ∂C ∂z z¼0 is the AA or DOC concentration gradients across the sedimentwater interface, and D sed is the sediment diffusion coefficient corrected for tortuosity (Iversen and Jørgensen 1993).Then, the benthic fluxes of TDAA, L-AA, and D-AA are calculated to be 90 AE 54, 64 AE 36, and 13 AE 5 μmol m À2 d À1 , respectively.In comparison, the benthic flux of DOC is 1.0 AE 0.5 mmol m À2 d À1 .These TDAA benthic fluxes (90 AE 54 μmol m À2 d À1 ) are comparable with the results of the previously reported values (Table 1), although the information is only available at some estuarine and continental shelf sites only for DFAA.
To assess the relative contribution of the benthic flux to the deep ocean DOM pool, the downward fluxes of AAs and DOC from the euphotic zone (at a depth of 100 m) were calculated for comparison.The downward fluxes of AAs and DOC were determined by multiplying the AA or DOC concentration gradients by the eddy diffusivity based on the 228 Ra tracer estimated by Cho et al. (2022).The downward fluxes of TDAA, L-AA, and D-AA from the euphotic zone are estimated to be 39 AE 25, 29 AE 19, and 4.3 AE 2  its BBL inventory, while that of DOC is approximately 3% AE 2% of its BBL inventory.It is notable that TDAA to DOC production ratios in the sediment are much higher than those in the euphotic zone.
Based on the excess TDAA inventory in the BBL and the calculated benthic flux, we can estimate the residence time of sediment-borne TDAA in the BBL.Here, the excess TDAA inventory in the BBL is defined as depth-integrated amounts of TDAA in the BBL exceeding the average amounts of the deep ocean.Then, the residence time of TDAA is calculated to be 1.3 AE 0.9 yr, with a range of 0.2-1.6 yr, which is much shorter than the turnover time of deep water in the East Sea ($ 100 yr; Watanabe et al. 1991).
Overall, the larger contributions of TDAA benthic flux relative to its BBL inventory and its labile characteristics, relative to DOC, seem to cause steep TDAA increases in the BBL, although DOC concentrations remained relatively uniform (Fig. 4).The TDAA benthic flux may support microbial activity in the BBL, where vigorous biogeochemical alteration occurs.Hyun et al. (2022) observed 1.7-and 1.8-fold higher bacterial production and bacterial abundance, respectively, in the BBL of the Ulleung Basin, relative to those in the deep ocean.Furthermore, they observed significantly increased bacterial production in the Ulleung Basin bottom water when incubated after adding sediment porewater (Hyun et al. 2022).Overall, our findings underscore the crucial role of benthic inputs of TDAA from organic-rich sediments in marine biogeochemical cycles, which support microbial activity in the deep ocean.More extensive studies are necessary to understand the sources and cycling of TDAA in the BBL of the global ocean.

Fig. 1 .
Fig. 1.Location of sampling stations in the East/Japan Sea.Solid circles and open squares represent the seawater and porewater sampling sites, respectively.

Fig. 2 .
Fig. 2. Vertical distributions of (A) DOC, (B) TDAA, (C) L-AA, and (D) D-AA in the East Sea.The vertical distributions of DOC, TDAA, L-AA, and D-AA in the northern East Sea are plotted together for comparison(Kim et al. 2017).We collected and measured seven samples from a station (latitude: 41 29.815 0 N, longitude:132 19.907 0 E) in the northern part.The data agree within 88% with those fromKim et al. (2017) at the nearby location.
Fig. 4. A schematic diagram of the distributions and vertical fluxes of DOC and TDAA in the East Sea.The downward fluxes from the euphotic zone are based on the concentration gradients and the eddy diffusivity estimated by Cho et al. (2022).The benthic fluxes are estimated based on Fick's first law of diffusion.The inventories of DOC and TDAA in the deep ocean (Inv Deep,DOC and Inv Deep,TDAA ) and BBL (Inv BBL,DOC and Inv BBL,TDAA ) are estimated based on the depth-integrated amounts of DOC and TDAA in the deep ocean (from a depth of 100 m to the bottom layer) and BBL (0-approximately 250 m above the seafloor), respectively.The residence time of sediment-borne TDAA (τ) is estimated based on the excess TDAA inventory in the benthic boundary layer divided by the calculated TDAA benthic flux.

Table 1 .
Comparison of the benthic fluxes of dissolved AAs in the East Sea with those from other estuarine and continental shelf environments.TDAA represents total dissolved amino acid, and DFAA represents dissolved free amino acid.Negative fluxes denote the net flux into the sediment.Converted based on the assumption of an average of 3.6 carbon atoms per amino acid. *