The succession of microbial community in the organic rich fish-farm sediment during bioremediation by introducing artificially mass-cultured colonies of a small polychaete, Capitella sp. I

Available online at www.sciencedirect.com

Marine Pollution Bulletin 57 (2008) 68–77 www.elsevier.com/locate/marpolbul

The succession of microbial community in the organic rich fish-farm sediment during bioremediation by introducing artificially mass-cultured colonies of a small polychaete, Capitella sp. I Tadao Kunihiro a,*, Tomoaki Miyazaki a, Yuuta Uramoto a, Kyoko Kinoshita a, Akihiro Inoue a, Sayaka Tamaki a, Daigo Hama b, Hiroaki Tsutsumi a, Kouichi Ohwada a

a

Faculty of Environmental and Symbiotic Sciences, Prefectural University of Kumamoto, 3-1-100 Tsukide, Kumamoto 862-8502, Japan b Keiten Co. Ltd., 2-14 Kusuura, Hondo, Kumamoto 863-0044, Japan

Abstract We monitored seasonal changes of the abundance and composition of microorganisms in the fish-farm sediment in Kusuura Bay, Amakusa, Japan, using the quinone profiling technique, during bioremediation by introducing cultured colonies of polychaete, Capitella sp. I. In November 2004, approximately 9.2 million cultured worms were transferred to the fish-farm sediment, which increased rapidly, and reached 458.5 gWW/m2 (528,000 indiv./m2) in March 2005. During this fast-increasing period of Capitella, the microbial quinone content of the surface sediment (0–2 cm) also increased markedly, and reached 237 lmol/m2 in January 2005, although the water temperature decreased to the lowest levels in the year. Particularly, the mole fraction of ubiquinone-10 in total quinones in the sediment, indicating the presence of a subclass of Proteobacteria, increased by 9.3%. These facts suggest that the bacterial growth was enhanced markedly by the biological activities of worms in the sediment, and the bacteria played an important role in the decomposition of the organic matter in the sediment.  2007 Elsevier Ltd. All rights reserved.

1. Introduction Fish farming with net pens following discharges of large amounts of food residues and fish faeces has a serious negative effect on the benthic environment (Tsutsumi and Kikuchi, 1983; Holmer and Kristensen, 1992; Findlay and Watling, 1995; Tsutsumi, 1995; Wu, 1995; Karakassis et al., 1998; Pawar et al., 2002; Yokoyama, 2002). The organic rich sediment just below the fish-farm tends to exhibit highly reduced conditions with the production of hydrogen sulfide due to accelerated activity of sulfatereducing bacteria during the warm seasons, and to cause deficiency of dissolved oxygen in the bottom water. Conse*

Corresponding author. Tel.: +81 96 383 2929x732766; fax: +81 96 384 6765. E-mail address: [email protected] (T. Kunihiro). 0025-326X/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2007.10.009

quently, the macrobenthic communities became extremely poor in both species diversity and biomass (Findlay and Watling, 1995; Tsutsumi, 1995; Wu, 1995; Karakassis et al., 1998; Pawar et al., 2002; Yokoyama, 2002; Tsutsumi et al., 2005). During the recovery process of fauna in such areas, following the seasonal or long-term improvement of the bottom conditions, a restricted number of small threadlike polychaetes including Capitella sp. and spionid species commonly recolonize rapidly (Rosenberg, 1973, 1976; Tsutsumi, 1987, 1990; Yokoyama et al., 1997; Karakassis and Hatziyanni, 2000). Tsutsumi and Montani (1993) focused on the ecological characteristics of the Capitella sp., and proposed a new method of bioremediation by introducing artificially mass-cultured colonies of Capitella into the eutrophicated sediment under net pens. They expected very rapid growth of the Capitella after releasing the seed population to the

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

sediment when the benthic conditions are recovering after summer, and further expected the accelerated decomposition of organic matter in the sediment, thereby improving the bottom condition in a relatively short period of time (Chareonpanich et al., 1993, 1994a, 1994b; Montani and Tsutsumi, 1996). This method has been tested in the laboratory (Chareonpanich et al., 1993, 1994a, 1994b), outside pools (Tsutsumi et al., 2002), and in a fish-farm (Tsutsumi et al., 2005). In these experiments, the organic matter in the sediment decreased markedly during the rapid growth of the worms. Such a high potential of decomposing activity, however, does not seem to be due solely to the feeding activities of the Capitella colonies. Chareonpanich et al. (1994a) confirmed an increase of microbial ATP content in the sediment with the growth of Capitella colonies. Wada et al. (2005) detected intensive microbial activities around the burrows of Capitella by using tetrazolium salt, 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride (INT). The interaction between the bacteria and Capitella seems to play an important role in the accelerated decomposition of the organic matter in the sediment. In this study, we monitored the microbial biomass and community structure in the bottom sediments inside and outside the farms of red sea bream, in Amakusa, western Japan, using a quinone profiling technique (cf. Hedrick and White, 1986; Hiraishi, 1999; Katayama et al., 2000; Hu et al., 2001), during a bioremediation experiment using artificially cultured Capitella colonies. We attempt to clarify how the microbial population changed in abundance and community structure during the rapid growth of Capitella, and discuss the interaction between microorganisms and Capitella colonies in the decomposition of organic matter in the sediment. 130°E 35°N

69

2. Materials and methods 2.1. Study area Kusuura Bay is an enclosed bay (11.4 km2 area, 4 km · 3.5 km), located between Amakusa Kamishima and Amakusa Shimoshima Islands (130 13 0 E, 32 23 0 N) in Kyushu, western Japan (Fig. 1). The water depth at the center of the bay is approximately 16–20 m. In 1973, fish-farms were established in an area of approximately 400 m · 200 m at the center of the bay. In the present study, 102 net pens of 12 m square and 8 m in depth, are set in the fish-farms, and approximately 5–20 tons of red sea bream, Pagrus major, and yellowtail tuna, Seriola quinqueradiata, is cultured in the pens. Two sampling stations (St. F and St. C) were set beside a net pen in the fish-farm and outside the fish-farm, respectively approximately 400 m apart. The mud content of the sediment at St. F and St. C was approximately 10% and 60%, respectively. 2.2. Sampling of the sediment to assess the microbial biomass and community structure We monthly collected sediment samples to assess the microbial biomass and community at St. F and St. C from December 2003 to September 2005. In this period, approximately 1,690,000 and 9,200,000 individuals of mass-cultured Capitella colonies were transferred to the sea floor to treat the organic rich sediment just below the net pen at St. F on December 6, 2003 and November 5, 2004, respectively (Tsutsumi et al., 2005). At St. F, four sediment core samples were collected with acrylic hand core samplers (45 mm in diameter) by divers from five points at four corners and the center of the sea

135°E

Sea of Japan

33°N

Kusuura Bay Pacific Ocean

31°N

Amakusa KamishimaIs.

Kusuura Bay

St. F

St. C Fish net pens

Amakusa ShimoshimaIs.

0

100

200 m

Fig. 1. Location of sampling stations in Kusuura Bay, Amakusa, Kyushu, western Japan.

70

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

floor just below the net pen. Each of these four sediment core samples was used for determination of TOC, redox potential (ORP) and quinone content, sediment density and biomass of Capitella, respectively. At St. C, six sediment core samples were collected in the same manner as St. F. Three of them were used for determination of ORP and quinone content. The remaining three samples were used for determination of the density of sediment. We also collected a sediment core sample for determination of TOC with a K.K. type core sampler (40 mm in diameter) and three samples using an Eckman-Birge grab sampler from a boat. We subsampled ten sediment samples from the grab samples for the quantitative analysis of Capitella with core samplers (5 cm · 5 cm · 5 cm). 2.3. Physico-chemical analysis of the sediment The sediment samples for determination of TOC, redox potential (ORP) and quinone content and density of sediment were sliced into two layers, the surface layer (up to 2 cm in depth) and the subsurface layer (from 2 to 4 cm in depth). For determination of the density of sediment in each layer, the sliced sediment samples were dried at 110 C overnight. The density of sediment in each layer was calculated by dividing the dry weight by the volume (33 cm3). For determination of TOC of the sediment in each layer, the sliced sediment samples were first freezedried, rinsed with 2 N HCl to remove inorganic carbonate, and further vacuum-dried. TOC of the sediment was determined with an elemental analyzer (Fisons, NA-1500n). Here, we calculated the TOC content in the two different layers from not only TOC mg/g but also TOC g/m2 using the data of density of sediment. ORP of the sediment in each layer was determined with a platinum electrode (RM-20P, TOA DKK). The measurement was calibrated with temperature. The TOC and ORP values at St. F were expressed as the mean of five different sampling sites (four corners and the center of the net pens). Those at St. C were expressed as the mean of three samples. 2.4. Analysis of quinones The sediment samples used for determination of ORP values were all mixed in each layer of two different sampling stations (St. F and St. C), and used for quinone analysis. These samples were stored in the dark at 20 C until used for the analysis. Microbial quinone content in the sediment was determined by a modified method as previously described by Hu et al. (1999, 2001), after removing benthic animals. Quinones were first extracted from the sediment samples with a mixture of chloroform–methanol (2:1, v/v), and re-extracted into hexane. Menaquinones and ubiquinones contained in the crude extract were separated and purified using Sep-Pak Plus Silica (Waters corp.). The types and concentrations of these quinones were determined using a HPLC equipped with an ODS column (Zorbax-ODS, 4.6 (I.D.) ·250 mm, Shimadzu-Dupont) and a

photodiode array detector (SPD-M10Avp, Shimadzu Co.). A mixture of methanol and isopropyl ether (9:2, v/v) was used as the mobile phase at a flow rate of 1.0 mL/min. The temperature of the column oven was maintained at 35 C. The ubiquinone-10 was purchased from Sigma Chemical Co. and used as the quantitative standard. The wavelengths used to quantify quinones were 275 nm and 270 nm for ubiquinones and menaquinones, respectively. Plastoquinone-9 was quantified from its molar extinction coefficient. The type of quinones was identified according to the retention time on the column and the UV spectrum of each peak observed in the photodiode array detector. The linear relationship between the logarithm of the retention times of quinones and the number of their isoprene units were also used to identify the quinone types. The quinone content in two different layers (the surface and subsurface layers) of the sediment was calculated from the quinone content of the sediment and the sediment density. In this paper, we refer to the quinones according to the following abbreviations (ubiquinone: UQ, menaquinone: MK, plastoquinone: PQ, vitamin K1: VK1). The number indicates that of isoprene units in the side chain of the quinone. For example, UQ-10 represents a ubiquinone with 10 isoprenoid units, and MK-9(H2) represents a menaquinone with 9 isoprenoid units where one of the 9 units is hydrogenated with 2 hydrogen atoms. To quantitatively evaluate the changes in the microbial community during the period of the bioremediation experiment, a dissimilarity index (D) based on the quinone profiling data was calculated using the following equation (Hiraishi et al., 1991) Dði; jÞ ¼

n 1X jfki  fkj j 2 k¼1

where fki and fkj are the mole fractions of the k quinone component in the i and j samples, respectively. The distance matrix was used by multi-dimensional scaling (MDS) analysis and cluster analysis. The MDS analysis was performed using SPSS 13.0J (SPSS Inc., Japan) with the dissimilarity value. The between-groups linkage method was applied to form clustering with the aid of the StatPartner program (OHa Inc.).

2.5. Biomass of Capitella population For estimation of the biomass of the Capitella population, the sediment samples were fixed in a formalin solution with a dye, Rose Bengal. After sieving the samples through a 0.125 mm mesh, the worms were sorted from the residues on the sieve. Then, their numbers were counted, their body sizes were measured with a microscope image analyzer, and their wet and dry weights were estimated from their body size (Tsutsumi et al., 2005). The Capitella biomass was expressed as the mean of five different sampling sites (four

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

corners and the center of the net pens) at St. F and the mean of three samples at St. C. 3. Results 3.1. Seasonal fluctuations in total amount of food expended for fish culture In the fish-farms, two types of food, soft extruder pellet (SEP, Mercian) and moist pellet (MP), were used for the culture of fish. Water content was 22.2 and 50.7%, and TOC content 467 and 449 g-C/kg-dry, respectively. The feeding amount was adjusted to the physiological conditions of the reared fish, and the total amount of reared fish in the net pens changed markedly from month to month (approximately 10–20 tons per net pen), because of their growth and shipping. Fig. 2 shows the total amount of organic matter expended at St. F as food for reared fish per month. It fluctuated between 0.52 and 1.03 ton-C/ month from December 2003 to August 2004, increased rapidly in the autumn and reached a peak in November 2004 (2.48 ton-C/month), decreasing to 0.17 ton-C/month in April 2005, because of a decrease in the number of reared fish and decline of feeding activities of the fish during the cold seasons. 3.2. Seasonal fluctuations of TOC and ORP in the sediment Fig. 3 shows the seasonal fluctuations of TOC and ORP of the sediment in the surface layer (up to 2 cm in depth) and the subsurface layer (2–4 cm in depth) at St. F and St. C. TOC amounts were calculated as mg/g and g/m2 as mentioned earlier. In the discussion that follows, we express the amount as g/m2. At St. F, the TOC levels in the surface and subsurface layers were 406 g-C/m2 and 543 g-C/m2 in December 2003, respectively. TOC of these two layers decreased rapidly to 185 g-C/m2 and 76 g-C/ m2 in January 2004, and fluctuated in a low range between 133 g-C/m2 and 282 g-C/m2 until July 2004. Accompanied

with the increase of feeding for reared fish in the net pen (Fig. 2), the TOC level only in the surface layer increased markedly from August 2004 again, and reached a peak, 470 g-C/m2, in November 2004. It gradually decreased during the winter, and 213 g-C/m2 was noted in May 2005. The TOC level of the subsurface layer decreased to 76 g-C/m2 in January 2004 and fluctuated in a low range of less than 268 g-C/m2 except September 2004 (380 g-C/m2) until August 2005. At St. C, outside the fish-farm, the TOC levels of both the surface and subsurface layers fluctuated in a low range between 118 and 207 g-C/m2 throughout the period of the present study. The ORP levels of the sediment showed a clear seasonal fluctuation pattern in both the surface and subsurface layers at St. F. The ORP levels of these two layers remained positive during the cold seasons (between 101 and 409 mV from January to March 2004, and between 33 and 104 mV from November 2004 to April 2005), while they remained negative during the warm seasons (between 99 and 1 mV from July to September 2004 and between 170 and 44 mV from May 2005 to October 2005). At St. C, the seasonal fluctuation pattern of the ORP level of the surface layer was similar to that of St. F, but it was positive throughout the present study. In January 2004 and March 2005, the ORP levels reached peaks of 399 and 340 mV, respectively. In the subsurface layer, the ORP levels remained in a relatively low range between 52 and 155 mV throughout the present study. 3.3. Biomass of Capitella population

1

Fig. 4 shows the seasonal fluctuations of biomass of Capitella at St. F and St. C. The Capitella population increases during the cold seasons in the present study areas (Tsutsumi et al., 1990, 1991, 2005). In this study, the most rapid growth of Capitella was observed at St. F just after the release of the artificially cultured colonies (approximately 1.69 million individuals in December 2003 and 9.2 million individuals in November 2004). Consequently, the worm biomass reached 120.4 gWW/m2 (approximately 134,000 indiv./m2) in February 2004 and 458.5 gWW/m2 (approximately 317,000 indiv./m2) in March 2005, respectively. The dense patches of Capitella then decreased rapidly in the spring of 2004 and of 2005. The extremely low density patches of less than 500 indiv./m2 remained in the summer during the present study. At St. C, the worms were found in extremely low densities of less than 0.2 gWW/m2 (320 indiv./m2) only during the winter. In other seasons, no worms were found.

0

3.4. Change in the microbial quinone content and the mole fractions of quinone

3 Feeding (ton-C/month)

71

SEP MP 2

D J F M A M J J A S O N D J F MA M J J A S 2004

2005

Fig. 2. Monthly feeding amount in the net pen at St. F. In this fish-farm, two types of food, soft extruder pellet (SEP) and moist pellet (MP) (Mercian, Kumamoto, Japan), were used.

Microbial quinones, which are one of the coenzymes in the electron transport chain of microbial cells, are divided into two groups, respiratory quinone (including ubiquinone and menaquinone) and photosynthetic quinone

72

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

a

b

600

600

St. F

St. F

0~2 cm 2~4 cm

St. C

0~2 cm 2~4 cm

ORP (mV)

TOC (g-C/m2)

400 400

200

200

0

0~2 0~ cmm 2~4 cm 2~4 cm

0

-200

600

600

St. C

0~2 cm 2~4 cm

ORP (mV)

TOC (g-C/m2)

400 400

200

200

0

0 D

F

A

J

A

O

D

2004

F

A

J

A O

2005

-200

D

F

A

J

A

O

D

2004

F

A

J

A

O

2005

Fig. 3. Seasonal fluctuations of (a) TOC and (b) ORP of the two different layers of the sediment, the surface layer (from the surface to 2 cm in depth) and the subsurface layer (from 2 cm to 4 cm in depth) at St. F and St. C.

300

400

St. F

St. F St. C

Quinone content (μmol/m2)

Biomass of Capitella (gWW/m2)

500

300 200 100 0

0~2 cm 2~4 cm

200

St. C 0~2 cm 2~4 cm

100

0 D F A 2004

J

A

O

D

F A 2005

J

A

O

D F 2004

A

J

A

O

D

F A 2005

J

A

O

Fig. 4. Seasonal fluctuations of the biomass of Capitella at St. F and St. C. Arrows indicate the release of the cultured colonies of Capitella sp. I at St. F.

Fig. 5. Microbial quinone content in the two layers of the sediment, the surface layer (from the surface to 2 cm in depth) and the subsurface layer (from 2 cm to 4 cm in depth) at St. F and St. C.

(including plastoquinone and vitamin K1). In general, one species or genus of bacteria has only one dominant species of respiratory quinone. Quinone profile, which is usually defined as the mole fraction of each quinone species, expresses the microbial community structure. The amount of total quinones has a linear relationship with microbial biomass (Saitou et al., 1999; Hiraishi et al., 2003). Fig. 5 shows the changes of quinone content in the surface and subsurface layers of the sediment at St. F and St. C. At St. F inside the fish -farm, the quinone content of the surface layer was in a relatively low range between 67 and

84 lmol/m2 (mean value 78 lmol/m2) from May to August 2004, increased markedly with the increase of the Capitella population from November 2004, and reached a peak, 237 lmol/m2, in January 2005. This peak value was approximately three times higher than the levels in the summer in 2004. The quinone content decreased rapidly from March 2005 (229 lmol/m2) to July 2005 (90 lmol/ m2). In the subsurface layer, the quinone content fluctuated between 41 and 143 lmol/m2 throughout the period. It reached a peak of 99 and 143 lmol/m2 in March 2004 and April 2005, respectively. At St. C, the quinone content

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77 MK-10 MK-9 MK-8 MK-7 MK-6

St. F (0~2 cm) 100

PQ-9 +VK1

PQ-9

80

MK

60 UQ-10

40 UQ

20 UQ-8

0

MAMJ J A S N D J MA MJ J S

2004

2005 St. F (2~4 cm)

100

PQ-9 +VK1

80

MK

60 40

UQ-10 UQ

20 UQ-8

0

MAMJ J A S N D J MA MJ J S

2004

2005

Mole fraction of quinones (mol%)

UQ-others UQ-11(H2) UQ-10(H2) UQ-9(H2) UQ-8(H2)

Mole fraction of quinones (mol%)

Mole fraction of quinones (mol%)

Mole fraction of quinones (mol%)

UQ-11 UQ-10 UQ-9 UQ-8 UQ-7

MK-10(H8) MK-10(H4) MK-9(H4) MK-10(H2) MK-9(H2)

73

PQ-9 VK1 MK-others MK-9(H10)

St. C (0~2 cm) 100

PQ-9

PQ-9 +VK1

80 60

MK

40 20

UQ

UQ-8

0

MAMJ J A S N D J MA MJ J S

2004

2005 St. C (2~4 cm)

100 PQ-9

PQ-9 +VK1

80 60

MK

40 20

UQ

UQ-8

0

MAMJ J A S N D J MA MJ J S

2004

2005

Fig. 6. Seasonal changes in the mole fractions of quinone at the two layers of the sediment, the surface layer (from the surface to 2 cm in depth) and the subsurface layer (from 2 cm to 4 cm in depth).

of both the surface and subsurface layers fluctuated at low levels between 11 and 36 lmol/m2 (mean value 18 lmol/m2) throughout the period of the present study. Fig. 6 shows the seasonal changes in the mole fraction of quinones in the sediment at St. F and St. C. At St. F, in both the surface and subsurface layers, arranging the quinone types in the order of mole fraction we get UQs  MKs > PQ-9+VK1. The mole fraction of UQ-10 was largest among all these quinone species. It fluctuated in a range between 30.1 and 47.0% in the surface layer and between 31.5 and 40.8% in the subsurface layer throughout the period of the present study. In the second dominant quinone type, MK, MK-7 occupied the largest mole fraction in MKs. Its mole fraction fluctuated between 6.2 and 8.8% in the surface layer and between 5.7 and 11.1% in the subsurface layer. At St. C, in both the surface and subsurface layers, arranging the quinone types in the order of mole fraction we get MKs P UQs > PQ-9+VK1. However, PQ-9 had the largest mole fraction among all these quinone species. It fluctuated between 9.1 and 36.5% in the surface layer and between 17.6 and 33.7% in the subsurface layer. The second dominant quinone species was UQ-8. It occupied 13.4–21.6% of the mole fraction in the surface layer and 10.7–21.2% in the subsurface layer. At St. F, the mole fraction of UQ-10 increased from the autumn and the winter in the surface layer. It ranged

between 36.9 and 47.0% (mean 41.1%) in total quinones from November 2004 to May 2005. This period coincided with the high biomass period of Capitella between 45.9 and 458.5 gWW/m2. In contrast, Capitella had low biomass periods between 0 and 3.1 gWW/m2 and between 0 and 8.3 gWW/m2 during the warm seasons from May 2004 to September 2004 and from June 2005 to September 2005, respectively, while the mole fraction of UQ-10 in these two periods was in the lowest ranges between 30.1 and 33.7% (mean 32.0%) and between 31.3 and 35.3% (mean 33.2%), respectively. The difference of the mean mole fraction of UQ-10 in the surface layer between the high biomass period and low biomass periods of Capitella was 9.3% (n = 8, p < 0.01, ANOVA). Thus, the difference in the mole fraction of UQ-10 in these two periods was statistically significant. 3.5. Succession of microbial community To analyze the interaction between Capitella and the heterotrophic bacteria, MDS and cluster analyses, based on the dissimilarity value matrix data calculated from the mole fraction of respiratory quinone in the surface layer of the sediment, were performed (Figs. 7 and 8). Here, we focus on the respiratory quinone including ubiquinones and menaquinones, which are contained mainly in

74

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

a

1.5

St. F Low Capitella biomass St. F High Capitella biomass St. C

Dimension 2

1.0 0.5 0 -0.5 -1.0 -1.5 -2.0

Group I -1.0

Group II 0

1.0

2.0

4. Discussion 3.0 3

Dimension 1

b

1.0 05.1

Dimension 2

0.5

05.6

05.4 05.5

05.7

0

04.8 04.6

04.12 04.11 05.3 05.9 04.9 04.3 04.4

04.7

-0.5 04.5

-1.0 0.5

1.0

1.5 Dimension 1

2.0

gram further indicates that the respiratory quinone profiles in Group II were divided into three groups (Group II-A, in May 2004; Group II-B, from December 2004 to April 2005; Group II-C, others). Group II-B was characterized by the high biomass period of Capitella population 121.2– 458.5 gWW/m2, of which the most noteworthy characteristics in respiratory quinone profile were the highest range of UQ-10 mole fraction in total quinones (41.2–47.0%).

2.5

Fig. 7. MDS (Multi-Dimensional Scaling) map of the respiratory quinone (ubiquinones and menaquinones) based on the dissimilarity value matrix data of the surface layer in this study (Stress and squared correlation value: 0.99, Stress value: 0.06). (a) St. F and St. C, (b) Detail MDS map at St. F. The figures indicate the sampling year and month (e.g. 04.7 indicates July 2004).

heterotrophic bacteria. Each plot in the MDS map in Fig. 7 represents the respiratory quinone profile at each sampling occasion at St. F and St. C, and the distance between the two plots expresses the degree of dissimilarity in the structure of the heterotrophic bacterial communities. The dendrogram in Fig. 8 classifies heterotrophic bacterial community structure at all sampling occasions at St. F and St. C in dissimilarity of UQ and MK profiles. As shown in Fig. 7, the respiratory quinone profiles can be divided clearly into two different groups, Group I (St. C, the outside of the fish-farm) and Group II (St. F, the inside of the fish-farm). These results suggest that the heterotrophic bacterial community structure in the sediment outside and inside the fish-farm was clearly different. One of the noteworthy points in this MDS map is that the plots during the high biomass period of Capitella were concentrated on the areas larger than 1.3 in Dimension 1, which were located on the furthest side from the plots of Group I. In Fig. 8, values of less than 0.1 of dissimilarity are not recognized as different quinone profiles (97% statistical reliability, cf. Hu et al., 2001). According to this standard, mainly two different groups of respiratory quinone profile were found in the dendrogram in Fig. 8 as in the MDS map in Fig. 7, Group I at St. C and Group II at St. F. This dendro-

4.1. Composition of the benthic microbial community inside and outside the fish-farms Ubiquinone, menaquinone and photosynthetic quinones are key components of the electron transfer reactions used for aerobic/nitrate, anaerobic/aerobic respirations, and photosynthesis, respectively (Collins and Jones, 1981; Hedrick and White, 1986; Hiraishi, 1999). Eukaryotic organisms possess only ubiquinones (Hutson and Threlfall, 1980; WFCC-MIRCEN World Data Centre for Microorganisms, 1995). The microbial communities in the sediment inside (St. F) and outside of the fish-farms (St. C) were dominated by ubiquinones and photosynthetic quinones (PQ and VK1), respectively (Fig. 6). These facts suggest that the benthic environment inside the fish-farms was dominated by aerobic bacteria and eukaryotic microorganisms, such as protozoa, metabolizing the organic matter, such as food residues and fish faeces sinking from the net pens, while sediment outside the fish-farms was dominated by photosynthetic microorganisms, such as micro-algae. Inside the fish-farm, the most dominant quinone species in the sediment was UQ-10 (Fig. 6), which is found in the a subclass of Proteobacteria and some eukaryotic microorganisms (Hutson and Threlfall, 1980; Collins and Jones, 1981; Yokota, et al. 1992; WFCC-MIRCEN World Data Centre for Microorganisms, 1995). The a-subclasses of Proteobacteria in the marine sediment are mainly Azospirillum sp., Beijerinckia sp., Bradyrhizobium sp., Rhodomicrobium sp., Rhodospirillum sp., Roseobacter sp. and Sphingomonas sp. (James and Russell, 1996; Urakawa et al., 1999). Therefore, these bacteria are considered to be dominant in the sediment inside the fish-farm. Sulfate-reducing bacteria are commonly observed in marine sediments, which possess mainly MK-6 or MK-7 (WFCC-MIRCEN World Data Centre for Microorganisms, 1995). Inside the fish-farms, MK was the second dominant quinone type, and MK-7 occupied the largest mole fraction in MKs (Fig. 6). It indicates that sulfatereducing bacteria were second dominant in the sediment inside the fish-farms. 4.2. Microbial population changes in abundance and community structure during the rapid growth of Capitella The sediment outside the fish-farms (St. C) was well oxygenated throughout the period of the present study, while

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

75

0.5

Dissimilarity of UQ and MK

0.4

0.3

Group I

Group II

0.2

0

C05.5 C04.11 C04.4 C04.12 C05.3 C04.9 C05.4 C05.6 C05.1 C04.8 C05.9 C05.7 C04.3 C04.6 C04.7 C04.5 F04.5 F05.4 F05.3 F05.1 F04.12 F05.6 F05.7 F04.8 F04.7 F04.6 F04.11 F05.9 F04.9 F04.3 F05.5 F04.4

0.1

Group II-B Group II-A

Group II-C

Fig. 8. Classification of the respiratory quinone profiles (only ubiquinones and menaquinones) based on the dissimilarity value matrix data of the surface layer in this study. Abbreviation in each sample indicates the sampling site and occasion. For example, F04.3 represents May 2004 at St. F.

in the sediment inside the fish-farm during the high biomass period of Capitella in the cold seasons (Pearson’s correlation coefficient, p < 0.05) (Fig. 9). Wada et al. (2005) indicated in laboratory microcosms that the worms promoted the formation of oxidative conditions in the deeper layers of the sediment by introducing the dissolved oxygen rich water through their burrowing activity. In the marine benthic ecosystem, in general, the interactions between benthic bacteria and eukaryotic fauna (including protozoan, meiobenthos and macrobenthos) in

300

R2=0.59 Quinone content (μmol/m2)

the fish-farm sediment (St. F) was highly enriched due to organic discharge becoming highly reduced condition during the warm seasons (Figs. 2 and 3b). From the end of summer, TOC content in the surface sediment inside the fish-farms increased further along with the increase of the feeding amount to reared fish (Figs. 2 and 3a). However, DO levels of the bottom water began to recover from the early autumn through vertical mixing of the water column due to the decrease of water temperature in the present study areas (Srithongouthai et al., 2006), and consequently the ORP values of the sediment inside the fish-farms changed to an oxidized conditions from November 2004 (Fig. 3b). The content of respiratory quinone (including ubiquinone and menaquinone) of the sediment also increased rapidly from the autumn to the winter (Fig. 5), although the water temperature decreased to the lowest levels in the year. These results indicate that the microorganisms, especially heterotrophic microorganisms, increased rapidly in the sediment inside the fish-farms during the cold seasons. The high growth activities of heterotrophic microorganisms during the cold seasons can be explained partly by the increase of available organic matter in well oxygenated conditions in the sediment. However, we focus on the influence of reworking activities by the rapidly increasing Capitella population on the abundance and community structure of the microorganisms in this season. Alongi (1985) and Wu et al. (2003) reported that bacterial activities were enhanced in the tubes, burrow-linings and burrow water of Capitella. We found a significantly positive correlation between Capitella biomass and the quinone content

200

100

High biomass period Low biomass period 0 0

100

200

300

400

500

Capitella biomass (gWW/m2) Fig. 9. Relationship between Capitella biomass and quinone content of the surface sediment up to a depth of 2 cm at St. F.

76

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77

the sediment lead to decomposition of organic matter in the sediment (Karakassis and Hatziyanni, 2000; Mazzola et al., 1999; Vezzulli et al., 2002). In the initial stage of the decomposition of organic matter, the heterotrophic bacteria play a key role by transferring organic matter into the bacterial cells (Meyer-Reil, 1991; Hoppe et al., 2002). Several studies have attempted to elucidate the potential role of benthic bacteria as a source of carbon and energy for microbial grazers (Fenchel, 1975; Fry, 1982; Danovaro, 1996; Hoppe et al., 2002). Alongi (1985) indicated that protozoa grazing on bacteria increased on the wall of Capitella tubes, and Chareonpanich et al. (1994b) estimated that 100 g wet-weight of Capitella per m2 of the sediment could decompose approximately 3.1 g-C of organic matter per day due to the interaction of worms and microorganisms. Thus, the findings in the present study suggest that an increase in the Capitella population strongly promotes the growth of the bacterial group, especially belonging to a subclass of Proteobacteria, and that the rapidly increasing bacteria associated with the worms play an important role in the decomposition of the organic matter in the organically enriched sediment. Acknowledgements We would like to express our thanks to Richard Lavin for his critical reading of the manuscript. This study was supported by the Research and Development Program for New Bio-industry Initiatives of the Bio-oriented Technology Research Advancement Institution (BRAIN) of Japan. References Alongi, D.M., 1985. Microbes, meiofauna, and bacterial productivity on tubes constructed by the polychaete Capitella capitata. Mar. Ecol. Prog. Ser. 23, 207–208. Chareonpanich, C., Montani, S., Tsutsumi, H., Matsuoka, S., 1993. Modification of chemical characteristics of organically enriched sediment by Capitella sp. I. Mar. Pollut. Bull. 26, 375–379. Chareonpanich, C., Montani, S., Tsutsumi, H., 1994a. Roles of a depositfeeding polychaete, Capitella sp. I, on the biological and chemical changes of the experimental marine sediment systems. Tech. Bull. Fac. Agr. Kagawa Univ. 46, 21–26. Chareonpanich, C., Tsutsumi, H., Montani, S., 1994b. Efficiency of the decomposition of organic matter, loaded on the sediment, as a result of the biological activity of Capitella sp. I. Mar. Pollut. Bull. 28, 314–318. Collins, M.D., Jones, D., 1981. Distribution of isoprenoid quinones structural types in bacteria and their taxonomic implications. Microbiol. Rev. 45, 316–354. Danovaro, R., 1996. Detritus-bacteria-meiofauna interactions in a seagrass bed (Posidonia oceanica) of the NM Mediterranean. Mar. Biol. 127, 1–13. Fenchel, T., 1975. The quantitative importance of the benthic microfauna of an arctic tundra pond. Hydrobiologia 46, 445–464. Findlay, R.H., Watling, L., 1995. Environmental impact of salmon netpen culture on marine benthic communities in marine: a case study. Estuar 18, 145–179. Fry, J.C., 1982. Interactions between bacteria and benthic invertebrates. In: Nedwell, D.B., Brown, C.M. (Eds.), Sediment Microbiology. Academic Press, New York, pp. 171–201.

Hedrick, D.B., White, D.C., 1986. Microbial respiratory quinones in the environment. J. Microbiol. Meth. 5, 243–254. Hiraishi, A., Iwasaki, M., Kawagishi, T., Yoshida, N., Narihiro, T., Kato, K., 2003. Significance of lipoquinones as quantitative biomarkers of bacterial populations in the environment. Microbes Environ. 18, 89– 93. Hiraishi, A., 1999. Isoprenoid quinones as biomarkers of microbial populations in the environment. J. Biosci. Bioeng. 88, 449–460. Hiraishi, A., Morishima, Y., Takeuchi, J., 1991. Numerical analysis of lipoquinone patterns in monitoring bacterial community dynamics in wastewater treatment systems. J. Gen. Appl. Microbiol. 37, 57–70. Holmer, M., Kristensen, E., 1992. Impact of marine fish cage farming on metabolism and sulphate reduction of under-lying sediments. Mar. Ecol. Prog. Ser. 80, 191–201. Hoppe, H.-G., Arnosti, C., Herndl, G.F., 2002. Ecological significance of bacterial enzymes in the marine environment. In: Burns, R.G., Dick, R.P. (Eds.), Enzymes in the Environment. Marcel Dekker, New York, pp. 73–107. Hu, H.-Y., Fujie, K., Nakagome, H., Urano, K., Katayama, A., 1999. Quantitative analyses of the change in microbial diversity in a bioreactor for wastewater treatment based on respiratory quinines. Water Res. 33, 3263–3270. Hu, H.-Y., Lim, B.-R., Goto, N., Fujie, K., 2001. Analytical precision and repeatability of respiratory quinones for quantitative study of microbial community structure in environmental samples. J. Microbiol. Meth. 47, 17–24. Hutson, K.G., Threlfall, D.R., 1980. Biochemistry of terpenoid quinones and chromanols. In: Levandowsky, M., Hutner, S.H. (Eds.), Biochemistry and Physiology of Protozoa. Academic Press, New York, pp. 255–286. James, P.G., Russell, P.H., 1996. Phylogenetic analysis of the bacterial communities in marine sediments. Appl. Environ. Microbiol. 62, 4049– 4059. Karakassis, I., Hatziyanni, E., 2000. Benthic disturbance due to fish farming analyzed under different levels of taxonomic resolution. Mar. Ecol. Prog. Ser. 203, 247–253. Karakassis, I., Tsapakis, M., Hatziyanni, E., 1998. Seasonal variability in sediment profiles beneath fish-farm cages in the Mediterranean. Mar. Ecol. Prog. Ser. 162, 243–252. Katayama, A., Kai, K., Fujie, K., 2000. Characterization of soil microbiota with quinone profile. In: Bollag, J.M., Storzky, G. (Eds.), Soil Biochemistry, vol. 10. Marcel Dekker, New York, pp. 303–347. Mazzola, A., Mirto, S., Danovaro, R., 1999. Initial fish-farm impact on meiofaunal assemblages in coastal sediments of the Western Mediterranean. Mar. Pollut. Bull. 38, 1126–1133. Meyer-Reil, L.-A., 1991. Ecological aspects of enzymatic activity in marine sediments. In: Chro´st, R.J. (Ed.), Microbial Enzymes in Aquatic Environments. Springer-Verlag, Berlin, pp. 84–95. Montani, S., Tsutsumi, H., 1996. Environmental bioremediation in bottom mud by benthos. In: Ichida, Y., Hino, A. (Eds.), Environmental Bioremediation by Biological Processes – The Possibility of Bioremediation in Fisheries Environments. Kouseisha-Kouseikaku, Tokyo, pp. 65–78 (Japanese). Pawar, V., Matsuda, O., Fujioka, N., 2002. Relationship between feed input and sediment quality of the fish cage farms. Fish Sci. 68, 894– 903. Rosenberg, R., 1973. Succession in benthic macrofauna in a Swedish fjord subsequent to the closure of a sulphide pulp mill. Oikos 24, 255–258. Rosenberg, R., 1976. Benthic faunal dynamics during succession following pollution abatement in a Swedish estuary. Oikos 27, 414–427. Saitou, K., Nagasaki, K., Yamakawa, H., Hu, H.-Y., Fujie, K., 1999. Linear relation between the amount of respiratory quinones and the microbial biomass in soil. Soil Sci. Plant Nutr. 45, 775–778. Srithongouthai, S., Endo, A., Inoue, A., Kinoshita, K., Yoshioka, M., Sato, A., Iwasaki, T., Teshiba, I., Nashiki, H., Hama, D., Tsutsumi, H., 2006. Control of dissolved oxygen levels of the water in the net

T. Kunihiro et al. / Marine Pollution Bulletin 57 (2008) 68–77 pens for fish farming with a microscopic bubble generating system. Fish Sci. 72, 485–493. Tsutsumi, H., Kikuchi, T., 1983. Benthic ecology of a small cove with seasonal oxygen depletion caused by organic pollution. Publ. Amakusa Mar. Biol. Lab. 7, 17–40. Tsutsumi, H., 1987. Population dynamics of Capitella capiteta (Polychaeta; Capitellidae) in an organically polluted cove. Mar. Ecol. Prog. Ser. 36, 139–149. Tsutsumi, H., 1990. Population persistence of Capitella sp. (Polychaeta; Capitellidae) on a mud flat subject to environmental disturbance by organic enrichment. Mar. Ecol. Prog. Ser. 63, 147–156. Tsutsumi, H., Kikuchi, T., Tanaka, M., Higashi, T., Imasaka, K., Miyazaki, M., 1991. Benthic faunal succession in a cove organically polluted by fish farming. Mar. Pollut. Bull. 23, 233–238. Tsutsumi, H., Kinoshita, K., Srithongouthai, S., Sato, A., Nagata, S., Inoue, A., Yoshioka, M., Ohwada, K., Hama, D., 2005. Treatment of the organically enriched sediment below the fish-farm with the biological activities of the artificially mass-cultured colonies of a small deposit feeding polychaete, Capitella sp. I. Benthos Res. 60, 17–24. Tsutsumi, H., 1995. Impact of fish net pen culture on the benthic environment of a cove in south Japan. Estuar 18, 108–115. Tsutsumi, H., Montani, S., 1993. Utilization of biological activities of capitellied polychaete for treatment of ‘‘Hedoro’’ (organically enriched sediment) deposited on the marine bottom below fish net pen culture. Nippon Suisan Gakkaishi 59, 1343–1347 (Japanese). Tsutsumi, H., Montani, S., Kobe, H., 2002. Bioremediation of organic matter loaded on the sediment in outdoor pools with a polychaete, Capitella sp. 1. Fish Sci 68, 613–616.

77

Urakawa, H., Kita-Tsukamoto, K., Ohwada, K., 1999. Microbial diversity in marine sediments from Sagami Bay and Tokyo Bay, Japan, as determined by 16S rRNA gene analysis. Microbiology 145, 3305–3315. Vezzulli, L., Chelossi, E., Riccardi, G., Fabiano, M., 2002. Bacterial community structure and activity in fish-farm sediments of the Ligrian sea (Western Mediterranean). Aquacul. Int. 10, 123–141. Wada, M., Wu, S.-S., Kogure, K., Tsutsumi, H., 2005. Short-term impact of biological activities of a burrowing polychaete, Capitella sp. I, on bacterial abundance and the chemical characteristics in organically enriched sediment. Benthos Res. 60, 59–66. WFCC-MIRCEN World Data Centre for Microorganisms (WDCM) database. Available from . Wu, R.S.S., 1995. The environmental impact of marine fish culture: towards a sustainable future. Mar. Pollut. Bull. 31, 159–166. Wu, S.S., Tsutsumi, H., Kita-Tsukamoto, K., Kogure, K., Ohwada, K., Wada, M., 2003. Visualization of the respiring bacteria in sediments inhabited by Capitella sp. I. Fish Sci. 69, 170–175. Yokota, A., Akagawa-Mathusita, M., Hiraishi, A., Katayama, Y., Urakami, T., Yamasato, K., 1992. Distribution of quinone systems in microorganisms: Gram-negative eubacteria. Bull JFCC 8, 136–171. Yokoyama, H., 2002. Impact of fish and pearl farming on the benthic environments in Gokasho Bay: Evaluation from seasonal fluctuation of the macrobenthos. Fish Sci. 68, 258–268. Yokoyama, H., Abo, K., Toyokawa, M., Toda, S., Yamamoto, S., 1997. Impact of mariculture on the spatial and temporal patterns of the macrobenthos in Gokasyo Bay. Bull. Natl. Res. Inst. Aquacul. 3, 7–16.