- Email: [email protected]
Bacteria and organic matter dynamics during a bioremediation treatment of organic-rich harbour sediments
Marine Pollution Bulletin 46 (2003) 1164–1173 www.elsevier.com/locate/marpolbul
Bacteria and organic matter dynamics during a bioremediation treatment of organic-rich harbour sediments Mauro Fabiano *, Daniela Marrale, Cristina Misic Dipartimento per lo Studio del Territorio e delle sue Risorse (Dip.Te.Ris.), Universit a di Genova, C.so Europa 26, 16132 Genova, Italy
Abstract We studied the dynamics of bacteria and organic matter in the Ancient Port of Genoa (Italy) during a bioremediation treatment of sediment (during summer–autumn 1998) in an area characterised by continuous sewage discharge. A strong increase in total benthic bacterial density (TBN) was recorded at the end of the study, from 14 108 to 58–172 108 cell g1 in different parts of the treated area. The TBN increase was linked to organic matter depletion, from more than 40 to less than 20 mg g1 . In order to highlight the main ecological mechanisms involved in bioremediation, a laboratory experiment based on both water and sediment from the basin studied was carried out. We observed an increase in TBN during the first 20 days and a decrease in sediment organic matter (up to about 20%). Increases of organic matter (about 2-fold) and TBN (from 21 to 33 109 cell l1 ) occurred in the overlying water, suggesting a strong association between the sediments and water column processes. Hydrolytic activities, which double in the sediment and increase up to a 300-fold in the water, are consistent with the decrease in sediment organic matter and with the water fraction dynamics. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Harbour; Organic matter; Biochemical composition; Bacteria; Environmental response and recovery
1. Introduction Nowadays, great emphasis is placed on environmental biotechnology and attaining sustainable development: in particular, biological techniques can be applied effectively in the remediation of sediments contaminated by organic pollutants from a variety of sources (Wolfe et al., 1996). The use of different metabolic pathways and/or the increase in autochthonous degradation processes to destroy or, at least, reduce the concentration of undesirable and/or hazardous substances, is the strategy of bioremediation technologies. Since nearly all natural products and a large number of synthetic compounds are degraded by bacteria, irrespective of their molecular weight or structural complexity, these microorganisms have become the key factor in bioremediation (Wackett and Ellis, 1999). In situ bioremediation is accomplished by adding (seeding) an exogenous microbial population to a contaminated site, thus attempting to significantly
* Corresponding author. Tel.: +39-0103538066; 0103538140. E-mail address: [email protected] (M. Fabiano).
fax:
+39-
0025-326X/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0025-326X(03)00166-8
increase the rates of natural degradation, with the advantage of a relatively low environmental impact (Head, 1998). The bioremediation approach has mainly been applied for cleaning up localised polluting accidents, such as hydrocarbon spills or wastewater ponds. The application of this method to restore marine coastal areas characterised by high organic matter concentration is still experimental. Huge amounts of organic materials are known to accumulate in areas with a restricted water exchange, such as harbours or sheltered bays, generally connected with anthropogenic-derived inputs (Cardell et al., 1999; Meyer-Reil and K€ oster, 2000; DellÕAnno et al., 2002). These high levels of organic substrates accumulate in the sediment, generating one of the major stresses for the marine environment. Eutrophication processes change the trophic state of the ecosystem to hypereutrophy and, in extreme cases, to dystrophy. The shift in the biochemical composition of the organic substrates (Danovaro et al., 1993), the changes in the structure of the resident communities (Meyer-Reil and K€ oster, 2000) and oxygen depletion leading to mass mortality of marine organisms (Cloern, 2001) are some of the ecological consequences of concern. The changes
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
in the biochemical composition of organic substrates and the relationships with the decomposers (bacteria, for instance) during bioremediation may, thus, give clues about the pathways for recovering an organic matter contaminated site. In addition, the choice of bioremediation strategies strictly depends on the bioavailability of the contaminants (Murphy et al., 1999). This suggests the need for both quantitative and qualitative studies on the organic matter. In order to assess the environmental response during bioremediation, a study was carried out in a semienclosed basin of the ancient harbour of Genoa. The ecological processes were assessed by monitoring the microbiological communities (focusing attention on the benthic bacterial community) and the biochemical composition of the sedimentary organic matter. Moreover, to better define the functional relationship between bacteria and organic matter in the sediments and the overlying water, a laboratory experiment was carried out using samples collected from the harbour site and treated with the same protocol.
2. Materials and methods 2.1. The biotechnological product Biovase is a biotechnological product. Its main feature is the presence of natural bacteria bio-fixed to a porous aluminium-silicated support. The microbial community is formed of Bacillus subtilis, B. licheniformis, Pseudomonas putida, Lactobacillus helveticus, Lactococcus lactis, Trichoderma reesci, T. hazonium, Phanerochaete chrysosporium, Nitrosomonas spp., Acinetobacter genospecies and Arthrobacter spp. 2.2. In situ samplings The ancient port of Genoa was chosen for in situ bioremediation experiments, being a semi-enclosed basin (approximately 2 ha), and having a high economic and tourist value. This area has features distinctive from the rest of the Ligurian Gulf: it is still used for port activities, but the presence of the historic town centre and modern infrastructures add to the natural morphology of the coast, forming a shallow sea floor (5–7 m depth) (Fig. 1). The system is strongly affected by natural and human input, which, through its accumulation in the stratified mud, contributes to organic matter enrichment. On 2 June 1998, the site was first inspected, and sediment samples were collected at two stations inside the basin (stations 1 and 2). A third station (station 3), outside the basin but with similar environmental characteristics, was chosen as a control. On 3 June and 27 July two seedings were performed, by means of a special craft fitted for this purpose and equipped with a
1165
Fig. 1. In situ experiment: sampling area and station locations.
vibrating spreading system. Sampling was carried out on a monthly basis from 2 June to 26 November. Mud samples were manually collected by SCUBA divers in two replicate cores and immediately transported to the laboratory. Water column parameters (temperature, salinity, oxygen saturation) were collected from the sub-surface layer (about 2 m above the sediment) with a multiparametric probe. Wind presence or absence was also recorded. 2.3. Experimental set-up and samplings During the following autumn, sediment and water samples were collected at the same harbour site and immediately transported to the laboratory. The sediment was placed in six square-based boxes (almost 3 kg of mud in each box) and filled up with 12 l of marine water. The overlying water of the boxes was aerated (Penn-Plax silent-air, 5 atm) just above the sediment surface, by means of air tubes, thereby oxygenating and stirring the water. One day after installation, four boxes were inoculated with the bioremediation product: 9 g was added to two boxes (treatment A) and 150 g to the other two (treatment B), while the two remaining untreated boxes were used as a control. To assess organic matter and bacterial dynamics, water and sediment samples were collected on a fortnightly basis. 2.4. Microbial parameters The total number of bacteria (TBN) was estimated using epifluorescence microscopy (Zeiss Universal Microscope) after Acridine Orange staining as described by Danovaro and Fabiano (1995). The counts were performed assigning bacteria to different size classes
1166
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
according to Palumbo et al. (1984): small size bacteria (SSB, <0.065 lm3 ), medium size bacteria (MSB, 0.065– 0.320 lm3 ), and large size bacteria (LSB, 0.320–0.780 lm3 ). The bacteria biomass values were calculated following Lee and Fuhrman (1987). Data were normalised to dry weight after desiccation (60 °C, 24 h) for sediment, and per litre for water samples. All counts were replicated at least four times. The coefficient of variation of these analyses was never higher than 15%. The contribution of bacterial biomass in terms of nitrogen and carbon equivalents was calculated assuming a bacterial carbon/nitrogen ratio ¼ 4 and a protein/N-protein ratio ¼ 6.25 (Lee and Fuhrman, 1987). 2.5. Biochemical composition of the organic matter Approximately 1 g of mud was used for each sediment organic matter biochemical analysis, and 0.2–0.5 l of water, filtered on Whatman GF/F filters, for each suspended material biochemical analysis. All the analyses were replicated 4 times. Carbohydrate (CHO) concentrations were calculated according to Dubois et al. (1956). Data are expressed as glucose equivalents. Protein (PRT) analyses were carried out following Hartree (1972). Concentrations are presented as albumin equivalents. Lipids (LIP) were determined following Marsh and Weinstein (1966) and Bligh and Dyer (1959). Concentrations are expressed as tripalmitine equivalents. 2.6. Enzymatic activities Enzymatic (b-glucosidase and leucine aminopeptidase) extracellular activities were analysed immediately after retrieval as described by Chr ost and Velimirov (1991). Data were normalised to litre or dry weight (60 °C, 24 h) and reported as micromoles of substrate released per litre or per sediment gram per hour. Each analysis was replicated three times for sediment and water samples for each substrate concentration. 2.7. Dissolved organic matter (DOM) DOM was determined with the synchronous fluorescence method (Ferrari and Mingazzini, 1995), a simultaneous scanning of the excitation and emission monochromators, keeping a constant difference between them. Generally, the lower the difference, the higher the resolution. We utilised a difference of 25 nm. The spectra were corrected by comparison with the Milliq water spectrum. This technique gives a qualitative determination of two classes of dissolved compounds: protein-like (peak A, emission wavelength of about 300 nm) and humic-like (peak B, emission wavelength ranging between 420 and 450 nm).
3. Results The in situ sampling period covered two seasons, summer and autumn, characterised by different environmental features (Table 1), which led to strong changes in the organic matter of the harbour sediment. Besides the regular trends of temperature and salinity, determined by the seasonal period and by rainfall, we noted an absence of wind during the summer (from June to September), and an increase of wind strength and water circulation during autumn. Nevertheless, the low hydrodynamism was highlighted by the scarce oxygen concentrations (Table 1), which never reached full saturation (maximum of 84%) although the water column is very shallow. In the laboratory experiment the oxygenation was maintained by continuous aeration. The values ranged from 7.7 to 8.3 mg l1 at the beginning to 4.6–6.4 mg l1 at the end of the experiment. The saturation ranged from 100% to 68%. 3.1. In situ experiment 3.1.1. Bacterial dynamics Bacterial density (TBN, Fig. 2) in the topmost centimetre of the sediments showed different dynamics in the two stations undergoing treatment. TBN increased regularly and significantly at station 2 during the course of the treatment (up to 9-fold), while it had lower changes at station 1. On 2 June, TBN was 19 108 cell g1 at station 2, while at the end of September (two months Table 1 In situ experiment: environmental parameters related to the water column in the three stations Date
Station
T (°C)
S (PSU)
Oxygen (%)
Wind (presence/absence)
02.06 16.06 27.07 26.08 24.09 03.11 26.11
1
20.16 19.89 25.09 25.22 22.50 18.00 13.00
35.63 37.09 37.27 37.44 37.48 37.20 36.32
79 69 72 59 65 62 57
) ) ) ) + + +
02.06 16.06 27.07 26.08 24.09 03.11 26.11
2
20.17 20.00 25.70 25.28 22.70 18.07 13.10
35.71 36.87 37.2 37.35 37.47 37.18 36.57
77 70 79 67 65 62 57
) ) ) ) + + +
02.06 16.06 27.07 26.08 24.09 03.11 26.11
3
20.07 19.58 25.25 25.20 22.56 17.89 13.08
35.27 36.47 37.3 37.00 37.18 37.00 36.32
84 68 61 70 62 60 54
) ) ) ) + + +
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173 250
TBN
200 150 100 50 0 2 Jn
16 Jn
27 Jl 26 Au 24 Se S1 S2 S3
03 No 26 No
Fig. 2. In situ experiment: total bacterial number (TBN, cells 108 g1 ) in the three stations during the sampling time. Arrows denote seeding.
after the second seed with the product) it amounted to 106 108 cell g1 and reached its maximum value (173 108 cell g1 ) on 3 November. 3.1.2. Changes in sediment organic matter load and composition Fig. 3 shows the concentration of carbohydrates, proteins and lipids in the top 1 cm of the harbour sediments. Carbohydrates were the dominant biochemical
carbohydrate
40 30 20 10 0 2 Jn
16 Jn
27 Jl
2 Jn
16 Jn
27 Jl
2 Jn
16 Jn
27 Jl
26 Au 24 Se
03 No 26 No
40
protein
30 20 10 0 26 Au 24 Se 03 No 26 No
4
lipid
3 2 1
1167
component of the organic material in the natural sediments. After treatment, significant variations occurred in the sediment parameters, with a clear shift in the biochemical composition of the organic load. For example, carbohydrates had high values at station 2 at the beginning of June (28.5 mg g1 ), and they decreased sharply (about 96%) in July (0.9 mg g1 ). Significant differences (t-test, p < 0:05 or less) were detected between the control site (station 3) and station 1 until August, confirming that the concentrations of carbohydrates were actually higher during the first months of bioremediation. From September the control and station 1 were similar, suggesting that the bioremediation treatment was effective starting from autumn. Instead, the differences between the control site and station 2 were significant (t-test, p < 0:02) only at the beginning of the study period, before the treatment. Then the two sites became very similar in concentration, suggesting a prompt response of the system to the treatment. Proteins showed an irregular downward trend at station 2, from 2 June (15.3 mg g1 ) to the beginning of November (6.3 mg g1 ), with two peaks in August and at the end of November (8.5 and 8.7 mg g1 , respectively). Instead, proteins showed a depletion of approximately 50% at station 1 only from July to September and then returned to the starting concentrations. Proteins prevailed over carbohydrates after treatment. The differences for protein concentrations between the control site and the two treated stations were similar to those previously described for carbohydrates. At station 2 the data suggested a quicker decrease in the sedimentary organic matter to values similar to those of the control site. The bacterial-nitrogen contribution to the N protein pool of sediments––estimated by converting bacterial biomass into nitrogen content––was negligible before treatment (0.8% and 1.2% for stations 1 and 2, respectively), but increased considerably, reaching 7.4% and 20.7% of the organic nitrogen pool (at stations 1 and 2, respectively) on 26 November. Lipids did not show strong variations, ranging between 0.4 and 2.0 mg g1 at all the stations. The first and the last samplings were, in fact, not significantly different (t-test, p > 0:05). Despite their relatively low concentrations at the beginning of the study, if compared to the other biochemical fractions, lipids generally became more concentrated than carbohydrates starting from August. No significant differences (t-test, p > 0:1 or more) were detected between the control site and the other two stations.
0
S1
26 Au
S2
24 Se
03 No 26 No
3.2. Laboratory experiment
S3
Fig. 3. In situ experiment: sedimentary organic matter (carbohydrates, proteins and lipids, mg g1 ) during the sampling time. Arrows denote seeding: (H) concentration significantly different from the control station one (t-test, p < 0:05).
3.2.1. Bacterial dynamics Changes in benthic bacteria density are shown in Table 2. Before treatment, TBN values amounted to 187 108 cell g1 . The maximum value was reached
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
Table 2 Laboratory experiment: total abundance of bacteria (TBN) and abundance of the three size classes (TBN1, TBN2 and TBN3) in the sediment and water Days
TBN
S.D.
TBN1
TBN2
Sediment Control 10 20 43 60
185 293 82 213
A treatment 10 20 43 60 B treatment 10 20 43 60
8 48 14 63
78 60 23 114
65 201 36 63
42 33 23 36
265 350 81 152
15 67 12 1
135 183 24 52
94 122 35 62
36 45 22 38
319 534 369 142
50 69 87 31
110 186 163 39
111 248 123 68
98 100 83 35
19 43 74 61
2 4 2 17
14 14 18 6
5 14 23 34
0 16 32 20
141 152 174 390
8 5 8 59
47 33 37 23
40 93 2 40
54 26 135 326
B treatment 10 221 20 133 43 816 60 3604
96 4 288 1083
126 27 250 462
32 64 223 648
62 42 343 2494
Water Control 10 20 43 60 A treatment 10 20 43 60
TBN3
contribution of size classes
1168
large 75%
medium small
50% 25% 0%
before treatment
cont.
B
A
Fig. 4. Laboratory experiment: contribution (%) of different size classes to total bacterial number (sediment) before treatment and in the three experimental systems (control, A treatment and B treatment) at the end of the experiment (60 days).
3.2.2. Sediment, particulate and dissolved organic matter Concentrations of carbohydrates and proteins in the sediment of the experimental microcosms are reported in Fig. 5. Carbohydrates, which were the main biochemical class of organic matter in the untreated sediments, decreased to a low level in treatment A (from 14.2 to 12.4 mg g1 ; about 13% of sediment carbohydrates were decomposed). Treatment B displayed a higher concentration decrease (from 19.3 to 14.1 mg g1 ). However, especially the B treatment was actually not very effective. In fact we observed no significant differences between the concentrations of the control and the two treatments during the experiment (t-test, p > 0:1), except after 60 days of treatment A, when the carbohydrate concentration of the A boxes was significantly lower than the control (t-test, p < 0:03). The treatment was particularly effective for PRT concentration in A treated boxes, which decreased from
Data are reported in cell 108 g1 (sediment) and cell 108 l1 (water) with standard deviations. carbohydrate
40 30 20 10 0 10
20
43
60
10
20
43
60
40 30 protein
20 days after the beginning of the experiment (350 and 534 108 cell g1 in A and B treatments, respectively). Then, the two treated systems displayed decreasing values. Contributions of small, medium and large-size bacteria to sedimentary TBN are summarised in Fig. 4. Small and medium-size benthic bacteria were the most abundant (44% and 42% of TBN, respectively) before treatment. Changes in the community structure occurred after two months, with an increase in medium and large-size bacteria and a reduction in the percentage of the smaller component in the treated systems. Bacterial abundance in the water (Table 2) showed irregular increases until the end of the experiment, ranging from 141 to 390 and from 221 to 3604 108 cell l1 for the A and B treatments, respectively.
100%
20 10 0
A treatment
B treatment
control
Fig. 5. Laboratory experiment: sedimentary organic matter (carbohydrates and proteins, mg g1 ) during the sampling time: (H) concentration significantly different from the control one (t-test, p < 0:05).
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
5.4 to 3.5 mg g1 (about 35%). We did not observe significant differences between the control and the A boxes in the first two samplings (t-test, p > 0:2), while during the third sampling the concentration in treatment A was significantly lower ðp < 0:02Þ and, during the fourth sampling, the difference was still slightly significant ðp < 0:06Þ. The significant differences between the control and the treatment B boxes highlighted a strong increase in proteins in the treated sediment despite the bioremediation treatment. In the water of the A boxes, where sedimentary organic matter was depleted in the course of the experiment, particulate organic matter showed irregular trends (Fig. 6). Particulate proteins and particulate carbohydrates were affected differently by the treatment, since particulate carbohydrates decreased regularly from 0.9 to 0.3 mg l1 , whereas particulate proteins first increased to 5.0 mg l1 after 20 days of incubation, and dropped to 2.4 mg l1 during the third sampling. Dissolved organic matter generally recorded a progressive increase with time as reported in Table 3. The first peak indicates the most bioavailable fraction (protein-like), while the second peak corresponds to the refractory and structural material (humic-like). In particular, an increase in the protein-like fraction was observed in the A and B systems (from 171 to 319 fu and from 144 to 278 fu, for A and B, respectively), while lower difference was observed for the control boxes (from 154 to 231 fu). The B peak showed increases (from two to three fold) in all the systems, in particular in the B treatment boxes (from 63 to 174 fu).
carbohydrate
Table 3 Laboratory experiment: values of the peaks A and B related to dissolved organic matter (DOM) Days
Peak A
S.D.
Peak B
S.D.
Control 10 20 43 60
154 228 nd 231
12 3 nd 15
54 92 nd 123
5 11 nd 6
A treatment 10 20 43 60
171 307 223 319
9 66 11 58
68 102 78 137
9 8 4 3
B treatment 10 20 43 60
144 198 221 278
10 3 7 8
63 87 126 174
6 4 7 5
Data are reported in fluorescence intensity units.
3.2.3. Water and sediment enzymatic activities The values (V max) of the two hydrolytic enzyme activities, in sediment and water, 10 and 60 days after beginning of the laboratory experiment, are presented in Table 4. In sediments, both activities showed a clear (up to 2- or 3-fold) increase at the end of the experiment. The trends are the same for the treated boxes and control, however, values about 50% higher were recorded for treatments A and B than for untreated sediments. Enhanced degradation is confirmed by the results obtained in the water compartment, where the values increased by 10 (leucine aminopeptidase) and as much as 300 times (b-glucosidase).
4. Discussion
6
4.1. Microbial community
4 2 0 10
20
43
60
6
protein
1169
4 2 0 10
20
A treatment
43
B treatment
60
control
Fig. 6. Laboratory experiment: organic matter in the water (carbohydrates and proteins, mg l1 ).
Bacterial abundance in the harbour sediments not only largely exceeded the values reported in the literature for nearby coastal areas (Albertelli et al., 1999; Danovaro and Fabiano, 1995; Danovaro et al., 1996), but also data related to areas rich in organic matter (Bottcher et al., 2000). Benthic bacteria have been proved to be highly effective ‘‘markers’’ in monitoring the performance of bioremediation by microbial inoculants (Wackett and Ellis, 1999). During the in situ experiment, at the beginning of treatment the three stations showed similar TBN (ANOVA, p ¼ 0.64), while at the end of the sampling period station 2 showed a remarkable increase (on average 8-fold) and in station 1 the increase was only 3fold on average. The poor water renewal and the continuous organic matter input characterising station 1
1170
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
Table 4 Laboratory experiment: enzymatic activity (V max) for leucine aminopeptidase (LA) and b-glucosidase (BG) in the sediment (lmol g1 h1 ) and water (lmol l1 h1 ) Days
Sediment
Water
LA
S.D.
BG
S.D.
LA
Control 10 60
25.9 41.4
6.0 2.2
1.7 6.1
0.1 0.6
1754.3 nd
A treatment 10 60
43.3 86.0
9.9 11.1
3.2 8.2
0.5 0.8
B treatment 10 60
38.3 87.8
0.7 6.8
6.4 nd
0.9 nd
S.D.
BG
S.D.
41.7 nd
3.0 173.7
0.1 5.2
289.0 2980.7
27.8 52.5
1.9 236.0
0.1 2.7
254.1 3927.1
12.6 140.0
1.0 350.4
0.1 9.5
Data are reported for the first (10 days) and last (60 days) samplings.
seem to be limiting factors affecting the increase in bacterial density if compared to station 2. In the laboratory experiment, an increase in bacterial abundance in the water similarly to that occurred in the sediment of station 2 was observed, while the increase in the sediment was related only to the first 20 days of treatment. These differences may be due to several reasons, such as the competition between autochthonous and allochthonous microorganisms and the need, for the latter, of acclimation to the new environmental (physical and chemical) conditions. However, a change in cell size may be one of the microbial community responses to the new conditions dependent on bioremediation treatments. Literature data about bacteria cell size are often contradictory. Sometimes the authors report an increase in cell size in hypereutrophic environments if compared to meso- and oligotrophic areas (Sommaruga, 1995), while other authors report that the differences between small and large bacteria abundances are not significant (Tumber et al., 1993). Our sampling area data are in agreement with the first observation. At the beginning of the experiment, when no bioremediation was active, large bacteria composed about 14% of the total number. The large size fraction in oligotrophic Ligurian coastal sediments as well as in mesotrophic seagrass sites is generally smaller if compared to the other size classes (about 1% of the total number; Danovaro et al., 1996). A change in bacterial size occurred during the laboratory experiment with an increase in medium and large size fractions. Such changes, which occurred when bioremediation was in progress, are likely to be the result of adaptive mechanisms (Sommaruga and Robarts, 1997). These changes also suggest that microbial community dynamics may be strongly affected by the competition between autochthonous and allochthonous bacteria, by changes in the quality of bacteria food (due to the increase in DOM, for instance, as suggested by Robarts and Wicks, 1990) and, probably, by the grazing pressure
exerted by protozoans and meiofauna (Lebaron et al., 2001).
4.2. Microbial activity Functional parameters, such as enzymatic activity, were investigated to evaluate microbial community efficiency and organic matter exploitation. Hydrolytic enzymes are known to be the rate-limiting step of degradation (Chr ost and Velimirov, 1991) and the enzymatic activity is substrate specific. Due to these characteristics, these parameters may give clues to what fraction of organic matter is mainly going to be consumed and, therefore, what may be the preferential trophic development pathway followed by the heterotrophic microbial community. The laboratory experiment provided interesting information, although this information gives only a potential response of the system under optimal conditions. In agreement with other studies on marine coastal areas (Meyer-Reil, 1986; Lopez et al., 1995), aminopeptidase (the proteasic enzyme) showed the highest activity, both in the sediment and in the water, but it showed only a slight increase during the experiment. Conversely, b-glucosidase activity (linked to the carbohydrate fraction) in the water increased more than 300 times, whereas it increased only 3–6 times in the sediments, suggesting that a bioremediation process involves the whole sediment–water system. This is also in agreement with the increasing dissolved organic matter concentration, composed of labile (peak A) and refractory (peak B) compounds, which were not metabolised by bacteria and accumulated in the overlying water. Therefore, in situ and laboratory experiments have shown that bioremediation treatment enhance natural processes which involve both the sediment and the water column. The transfer of organic materials from the sediment to the water makes
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
it easier to remove organic loads, which are often linked to inorganic or refractory organic compounds. Other authors (Karner and Herndl, 1992; Meyer-Reil and K€ oster, 2000) have observed that the bacteria community development in areas with increasing organic matter concentrations may generally follow two phases. During the first phase, a coupling between bacteria activity (both in terms of hydrolytic activity and biomass production) and organic matter supply is observed. Conversely, during the second phase, some environmental forcing (such as temperature and light; Meyer-Reil and K€ oster, 2000) or internal processes (such as enzyme hyperproduction by bacteria; Karner and Herndl, 1992) may affect this balance, thus leading to a loose of the hydrolysis-uptake coupling. However, the likely presence of hydrolytic enzymes not connected with bacteria but persisting in the environment (Misic et al., 1998) may lead to misleading interpretations of bacterial activity. The uncoupled solubilisation-bacteria consumption process is, therefore, an important mechanism by which bacteria exert biochemical pressure on the large-size phase of organic matter (both sedimentary and particulate) to regulate the flux to dissolved organic matter (Karner and Herndl, 1992; Smith et al., 1992). These processes may explain the different rates of increase of bacteria number and enzymatic activity. Since enzyme activities are mainly stimulated by organic matter availability, the relatively low aminopeptidase/glucosidase ratios found in our study may be related to the higher availability (in terms of easily degradable glucides) of the carbohydrate fraction in the Genoa harbour sediments. Moreover, K€ oster et al. (1997) observed that the proteolytic/glycolytic enzyme ratio increased from mesotrophic to eutrophic and hypertrophic conditions. Therefore, the decrease in this ratio in our experiment is a signal of recovering. 4.3. Organic matter and its consumption The sediments in the Genoa harbour are characterised by an extremely high organic matter content, in agreement with scientific literature on highly eutrophic coastal areas (Fabiano et al., 2001; Meyer-Reil and K€ oster, 2000). In particular, protein and carbohydrate concentrations were much higher than those reported for the adjacent coastal areas of the Ligurian Sea (Danovaro and Fabiano, 1995). Organic matter and microbial biomass are generally directly correlated. However, with increasing eutrophication, this relationship is not simply linear. Meyer-Reil and K€ oster (2000) observed that in the Southern Baltic Sea the correlation was linear up to carbon concentrations of 10 mg cm3 , and increasing carbon concentrations were reflected by a decrease of microbial biomass and potential decomposition. The existence of a concentration threshold, which divides the balanced systems
1171
from the others where the potential for consuming and recycling the organic matter is affected, is a very important ecological feature. Bioremediation may be effective when the proportion of the organic matter to the allochthonous and autochthonous bacteria is kept under the maximum limit for decoupling. Moreover, from an ecological point of view, the water–sediment system of the ancient port of Genoa has similar features to those of hyper-eutrophic systems characterised by strong fluctuations in the abundance of many of the components of the food web. The exponential growth of several species, in the form of pulses, and their sudden collapse are good examples of low ecosystem stability and loss of homeostatic mechanisms. However, at the same time, they may be required in order to allow such ecosystems to exist. These features are somehow enhanced when bioremediation is carried out. The treatment, in a first phase, introduces huge amounts of allochthonous microbes, better balancing the bacteria and organic matter, which, after the acclimation, facilitates the assessment of recovery. However, although the quantity of organic substrates may highly influence the development of the microbial community and its recycling activity, the quality of organic materials is equally important. At station 2, the constant depletion of carbohydrates from the first seeding suggests that the resident carbohydrates were mainly labile, and the remnants were, instead, composed of the more refractory glucidic substances. Another conclusion is that the input of carbohydrates from the coast was not continuous, but that the concentrations were the result of a long-term accumulation. Conversely, the proteins, generally thought to be an essential and largely bioavailable food source for benthic metabolism (Mayer et al., 1995), were not degraded at the same rate as the carbohydrates. The decomposition of specific organic substrates is unlikely to follow the same patterns, but rather to depend on the capabilities of microorganisms (as a consequence of a specific population structure––Boschker et al., 2001––and of the expression of specific tools for degradation––hydrolytic enzymes––Misic and Fabiano, 1996) and on the biochemical features of the substrate. Thus, the high protein content detected throughout the entire sampling period may be related to the refractory nature of the sedimentary protein pool. Proteins may decrease their availability by means of sorption by sediment (Mayer et al., 1995). Moreover, especially in coastal areas, freshwater input supplies the system with complex molecules (humic acids for instance; Hubberten et al., 1994), which embed proteins in highly refractory micro and macro aggregates. Moreover, direct protein glycosilation may decrease their degradation rates (Smith et al., 1992). Thus, an increase in protein refractivity due to environmental conditions has to be taken into
1172
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
account before applying bioremediation protocols. Moreover, the input of proteins to the system was, probably, more continuous, leading to high protein: carbohydrate ratios (on average more than 5 during the whole study period of the in situ experiment), typical of eutrophic and eutrophicated systems (Fabiano et al., 2001). In the laboratory, the protein trends of the A system decreased, instead, showing that a more effective recovery of the sedimentary system can be achieved by stopping organic matter enrichment. The increase in proteins during the B experiment was probably due to high bacterial growth and enhanced hydrolytic activity (especially in the overlying water) likely to lead to a generally higher release of proteins embedded in the refractory component of the organic matter. This, in turn, did not lead to a net protein increase, but to their enhanced detectability by our analytical method. However, the high protein content of the sedimentary organic matter can also be due to increased (up to 11fold) bacterial abundance and biomass. Actually, the bacterial-nitrogen contribution to the N protein pool of sediments increased considerably. Bacteria are present in high numbers, also in natural conditions, and their biomass is greater than the biomass of the other benthic organisms (Meyer-Reil and K€ oster, 2000). Protein liberation from aggregates and bacteria increase are processes that, despite increasing the protein content of the sediment and the overlying water, change the availability of this high food-value compounds. Free or living proteins are more labile and need lower time and energy resources to be recycled, thus increasing the rates and fluxes of energy and materials in the system. Acknowledgement We would like to thank Idrabel Italia S.r.L. for kindly providing the bioremediation product. References Albertelli, G., Covazzi-Harriague, A., Danovaro, R., Fabiano, M., Fraschetti, S., Pusceddu, A., 1999. Differential responses of bacteria, meiofauna and macrofauna in a shelf area (Ligurian Sea, NW Mediterranean): role of food availability. Journal of Sea Research 42, 11–26. Bligh, E.G., Dyer, W., 1959. A rapid method for total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37, 911–917. Boschker, H.T.S., de Graaf, W., K€ oster, M., Meyer-Reil, L.A., Cappenberg, T.E., 2001. Bacterial populations and processes involved in acetate and propionate consumption in anoxic brackish sediment. FEMS Microbiology Ecology 35, 97–103. Bottcher, M.E., Hespenheide, B., Llobet-Brossa, E., Beardsley, C., Larsen, O., Schramm, A., Wieland, A., Bottcher, G., Berninger, U.-G., Amann, R., 2000. The biogeochemistry, stable isotope geochemistry and microbial community structure of a temperate intertidal mudflat: an integrated study. Continental and Shelf Research 20, 1749–1769.
Cardell, M.J., Sarda, R., Romero, J., 1999. Spatial changes in sublittoral soft-bottom polychaete assemblages due to river inputs and sewage discharges. Acta Oecologica 20 (4), 343–351. Cloern, J.E., 2001. Our evolving conceptual model of the coastal eutrophication problem. Marine Ecology Progress Series 210, 223–253. Chr ost, R.J., Velimirov, B., 1991. Measurement of enzyme kinetics in water samples: effect of freezing and soluble stabilizer. Marine Ecology Progress Series 70, 93–100. Danovaro, R., Fabiano, M., 1995. Seasonal and interannual variation of benthic bacteria in a seagrass bed of the Mediterranean Sea: relationship with labile organic compounds and other environmental factors. Aquatic Microbiology Ecology 9, 17–26. Danovaro, R., Fabiano, M., Della Croce, N., 1993. Labile organic matter and microbial biomasses in deep-sea sediments (Eastern Mediterranean Sea). Deep-Sea Research 40, 953–965. Danovaro, R., Della Croce, N., Fabiano, M., 1996. Microbial response to oil disturbance in the coastal sediments of the Ligurian Sea (Western Mediterranean). Chemistry and Ecology 12, 187– 198. DellÕAnno, A., Mei, M.L., Pusceddu, A., Danovaro, R., 2002. Assessing the trophic state and eutrophication of coastal marine systems: a new approach based on the biochemical composition of sediment organic matter. Marine Pollution Bulletin 44, 611–622. Dubois, M., Gilles, K.A., Hamilton, P.A., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Analytical Chemistry 28, 350–356. Fabiano, M., Misic, C., Manini, E., Danovaro, R., Povero, P., Chiatti, S., 2001. Distribution and biochemical composition of suspended and sedimentary organic matter in the northern Adriatic. In: Faranda, F.M., Guglielmo, L., Spezie, G. (Eds.), Structure and Processes in the Mediterranean Ecosystems. Springer Verlag, Heidelberg, pp. 447–454. Ferrari, G.M., Mingazzini, M., 1995. Synchronous fluorescence spectra of dissolved organic matter (DOM) of algal origin in marine coastal waters. Marine Ecology Progress Series 125, 305–315. Hartree, E.F., 1972. Determination of proteins: a modification of the Lowry method that gives a linear photometric response. Analytical Biochemistry 48, 422–427. Head, I., 1998. Bioremediation: towards a credible technology. Microbiology 144, 599–608. Hubberten, U., Lara, R.J., Kattner, G., 1994. Amino acid composition of seawater and dissolved humic substances in the Greenland Sea. Marine Chemistry 45, 121–128. Karner, M., Herndl, G.J., 1992. Extracellular enzymatic activity and secondary production in free-living and marine-snow associated bacteria. Marine Biology 113, 341–347. K€ oster, M., Dahlke, S., Meyer-Reil, L.A., 1997. Microbiological studies along a gradient of eutrophication in a shallow coastal inlet in the Southern Baltic Sea (Nordrugensche Bodden). Marine Ecology Progress Series 152, 27–39. Lebaron, P., Servais, P., Troussellier, M., Courties, C., Muyzer, G., Bernard, L., Sch€afer, H., Pukall, R., Stackebrandt, E., Guindulain, T., Vives-Rego, J., 2001. Microbial community dynamics in Mediterranean nutrient-enriched seawater mesocosms: changes in abundances, activity and composition. FEMS Microbiology Ecology 34, 255–266. Lee, S., Fuhrman, J.A., 1987. Relationship between biovolume and biomass of naturally derived marine bacterioplankton. Applied Environmental Microbiology 53, 1298–1303. Lopez, N.I., Borras, G., Vallespin os, F., 1995. Effect of heavy metals on enzymatic degradation of organic matter in sediments off Catalonia (Northeastern Spain). Scientia Marina 59, 149–154. Marsh, J.B., Weinstein, W.J., 1966. A simple charring method for determination of lipids. Journal of Lipid Research 7, 574–576.
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173 Mayer, L.M., Schick, L.L., Sawyer, T., Plante, C.J., 1995. Bioavailable amino acids in sediments: a biomimetic, kinetic-based approach. Limnology and Oceanography 40, 511–520. Meyer-Reil, L.A., 1986. Measurement of hydrolytic activity and incorporation of dissolved organic substrate by microorganisms in marine sediments. Marine Ecology Progress Series 31, 143–149. Meyer-Reil, L.A., K€ oster, M., 2000. Eutrophication of marine waters: effects of benthic microbial communities. Marine Pollution Bulletin 41, 255–263. Misic, C., Fabiano, M., 1996. A functional approach to the assessment of the nutritional value of the particulate organic matter. Chemistry and Ecology 13, 51–63. Misic, C., Povero, P., Fabiano, M., 1998. Relationship between ectoenzymatic activity and availability of organic substrates (Ross Sea, Antarctica): an experimental approach. Polar Biology 20, 367–376. Murphy, T.P., Lawson, A., Kumagai, M., Babin, J., 1999. Review of emerging issues in sediment treatment. Aquatic Ecosystem Health and Management 2, 419–434. Palumbo, R., Ferguson, J.E., Rublee, P.A., 1984. Size of suspended bacterial cells and association of heterotrophic activity with size fraction of particles in estuarine and coastal waters. Applied Environmental Microbiology 48 (1), 157–164.
1173
Robarts, R., Wicks, R.J., 1990. Heterotrophic bacterial production and its dependence on autotrophic production in a hypertrophic African reservoir. Canadian Journal of Fisheries and Aquatic Science 47, 1027–1037. Smith, D.C., Simon, M., Alldredge, A.L., Azam, F., 1992. Intense hydrolytic enzyme activity on marine aggregates and implications for rapid particle dissolution. Nature 359, 139–142. Sommaruga, R., 1995. Microbial and classical food webs: a visit to a hypertrophic lake. FEMS Microbiology Ecology 17, 257–270. Sommaruga, R., Robarts, R.D., 1997. The significance of autotrophic and heterotrophic picoplancton in hypertrophic ecosystems. FEMS Microbiology Ecology 24, 187–200. Tumber, V.P., Robarts, R.D., Arts, M.T., Evans, M.S., Caldwell, D.E., 1993. The influence of environmental factors on seasonal changes in bacterial cell volume in two saline lakes. Microbiology Ecology 26, 9–20. Wackett, L.P., Ellis, L.B.M., 1999. Predicting biodegradation. Environmental Microbiology 1 (2), 119–124. Wolfe, D.A., Krahn, M.M., Casillas, E., Sol, S., Thompson, T.A., Lunz, J., Scott, K.J., 1996. Toxicity of intertidal and subtidal sediments contamined by the Exxon Valdez spill. In: Rice, S.D., Spies, R.B., Wolfe, D.A., Wright, B.W. (Eds.), Proceedings of the Symposium Exxon Valdez Oil Spill, pp. 121–139.
Bacteria and organic matter dynamics during a bioremediation treatment of organic-rich harbour sediments Mauro Fabiano *, Daniela Marrale, Cristina Misic Dipartimento per lo Studio del Territorio e delle sue Risorse (Dip.Te.Ris.), Universit a di Genova, C.so Europa 26, 16132 Genova, Italy
Abstract We studied the dynamics of bacteria and organic matter in the Ancient Port of Genoa (Italy) during a bioremediation treatment of sediment (during summer–autumn 1998) in an area characterised by continuous sewage discharge. A strong increase in total benthic bacterial density (TBN) was recorded at the end of the study, from 14 108 to 58–172 108 cell g1 in different parts of the treated area. The TBN increase was linked to organic matter depletion, from more than 40 to less than 20 mg g1 . In order to highlight the main ecological mechanisms involved in bioremediation, a laboratory experiment based on both water and sediment from the basin studied was carried out. We observed an increase in TBN during the first 20 days and a decrease in sediment organic matter (up to about 20%). Increases of organic matter (about 2-fold) and TBN (from 21 to 33 109 cell l1 ) occurred in the overlying water, suggesting a strong association between the sediments and water column processes. Hydrolytic activities, which double in the sediment and increase up to a 300-fold in the water, are consistent with the decrease in sediment organic matter and with the water fraction dynamics. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Harbour; Organic matter; Biochemical composition; Bacteria; Environmental response and recovery
1. Introduction Nowadays, great emphasis is placed on environmental biotechnology and attaining sustainable development: in particular, biological techniques can be applied effectively in the remediation of sediments contaminated by organic pollutants from a variety of sources (Wolfe et al., 1996). The use of different metabolic pathways and/or the increase in autochthonous degradation processes to destroy or, at least, reduce the concentration of undesirable and/or hazardous substances, is the strategy of bioremediation technologies. Since nearly all natural products and a large number of synthetic compounds are degraded by bacteria, irrespective of their molecular weight or structural complexity, these microorganisms have become the key factor in bioremediation (Wackett and Ellis, 1999). In situ bioremediation is accomplished by adding (seeding) an exogenous microbial population to a contaminated site, thus attempting to significantly
* Corresponding author. Tel.: +39-0103538066; 0103538140. E-mail address: [email protected] (M. Fabiano).
fax:
+39-
0025-326X/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0025-326X(03)00166-8
increase the rates of natural degradation, with the advantage of a relatively low environmental impact (Head, 1998). The bioremediation approach has mainly been applied for cleaning up localised polluting accidents, such as hydrocarbon spills or wastewater ponds. The application of this method to restore marine coastal areas characterised by high organic matter concentration is still experimental. Huge amounts of organic materials are known to accumulate in areas with a restricted water exchange, such as harbours or sheltered bays, generally connected with anthropogenic-derived inputs (Cardell et al., 1999; Meyer-Reil and K€ oster, 2000; DellÕAnno et al., 2002). These high levels of organic substrates accumulate in the sediment, generating one of the major stresses for the marine environment. Eutrophication processes change the trophic state of the ecosystem to hypereutrophy and, in extreme cases, to dystrophy. The shift in the biochemical composition of the organic substrates (Danovaro et al., 1993), the changes in the structure of the resident communities (Meyer-Reil and K€ oster, 2000) and oxygen depletion leading to mass mortality of marine organisms (Cloern, 2001) are some of the ecological consequences of concern. The changes
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
in the biochemical composition of organic substrates and the relationships with the decomposers (bacteria, for instance) during bioremediation may, thus, give clues about the pathways for recovering an organic matter contaminated site. In addition, the choice of bioremediation strategies strictly depends on the bioavailability of the contaminants (Murphy et al., 1999). This suggests the need for both quantitative and qualitative studies on the organic matter. In order to assess the environmental response during bioremediation, a study was carried out in a semienclosed basin of the ancient harbour of Genoa. The ecological processes were assessed by monitoring the microbiological communities (focusing attention on the benthic bacterial community) and the biochemical composition of the sedimentary organic matter. Moreover, to better define the functional relationship between bacteria and organic matter in the sediments and the overlying water, a laboratory experiment was carried out using samples collected from the harbour site and treated with the same protocol.
2. Materials and methods 2.1. The biotechnological product Biovase is a biotechnological product. Its main feature is the presence of natural bacteria bio-fixed to a porous aluminium-silicated support. The microbial community is formed of Bacillus subtilis, B. licheniformis, Pseudomonas putida, Lactobacillus helveticus, Lactococcus lactis, Trichoderma reesci, T. hazonium, Phanerochaete chrysosporium, Nitrosomonas spp., Acinetobacter genospecies and Arthrobacter spp. 2.2. In situ samplings The ancient port of Genoa was chosen for in situ bioremediation experiments, being a semi-enclosed basin (approximately 2 ha), and having a high economic and tourist value. This area has features distinctive from the rest of the Ligurian Gulf: it is still used for port activities, but the presence of the historic town centre and modern infrastructures add to the natural morphology of the coast, forming a shallow sea floor (5–7 m depth) (Fig. 1). The system is strongly affected by natural and human input, which, through its accumulation in the stratified mud, contributes to organic matter enrichment. On 2 June 1998, the site was first inspected, and sediment samples were collected at two stations inside the basin (stations 1 and 2). A third station (station 3), outside the basin but with similar environmental characteristics, was chosen as a control. On 3 June and 27 July two seedings were performed, by means of a special craft fitted for this purpose and equipped with a
1165
Fig. 1. In situ experiment: sampling area and station locations.
vibrating spreading system. Sampling was carried out on a monthly basis from 2 June to 26 November. Mud samples were manually collected by SCUBA divers in two replicate cores and immediately transported to the laboratory. Water column parameters (temperature, salinity, oxygen saturation) were collected from the sub-surface layer (about 2 m above the sediment) with a multiparametric probe. Wind presence or absence was also recorded. 2.3. Experimental set-up and samplings During the following autumn, sediment and water samples were collected at the same harbour site and immediately transported to the laboratory. The sediment was placed in six square-based boxes (almost 3 kg of mud in each box) and filled up with 12 l of marine water. The overlying water of the boxes was aerated (Penn-Plax silent-air, 5 atm) just above the sediment surface, by means of air tubes, thereby oxygenating and stirring the water. One day after installation, four boxes were inoculated with the bioremediation product: 9 g was added to two boxes (treatment A) and 150 g to the other two (treatment B), while the two remaining untreated boxes were used as a control. To assess organic matter and bacterial dynamics, water and sediment samples were collected on a fortnightly basis. 2.4. Microbial parameters The total number of bacteria (TBN) was estimated using epifluorescence microscopy (Zeiss Universal Microscope) after Acridine Orange staining as described by Danovaro and Fabiano (1995). The counts were performed assigning bacteria to different size classes
1166
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
according to Palumbo et al. (1984): small size bacteria (SSB, <0.065 lm3 ), medium size bacteria (MSB, 0.065– 0.320 lm3 ), and large size bacteria (LSB, 0.320–0.780 lm3 ). The bacteria biomass values were calculated following Lee and Fuhrman (1987). Data were normalised to dry weight after desiccation (60 °C, 24 h) for sediment, and per litre for water samples. All counts were replicated at least four times. The coefficient of variation of these analyses was never higher than 15%. The contribution of bacterial biomass in terms of nitrogen and carbon equivalents was calculated assuming a bacterial carbon/nitrogen ratio ¼ 4 and a protein/N-protein ratio ¼ 6.25 (Lee and Fuhrman, 1987). 2.5. Biochemical composition of the organic matter Approximately 1 g of mud was used for each sediment organic matter biochemical analysis, and 0.2–0.5 l of water, filtered on Whatman GF/F filters, for each suspended material biochemical analysis. All the analyses were replicated 4 times. Carbohydrate (CHO) concentrations were calculated according to Dubois et al. (1956). Data are expressed as glucose equivalents. Protein (PRT) analyses were carried out following Hartree (1972). Concentrations are presented as albumin equivalents. Lipids (LIP) were determined following Marsh and Weinstein (1966) and Bligh and Dyer (1959). Concentrations are expressed as tripalmitine equivalents. 2.6. Enzymatic activities Enzymatic (b-glucosidase and leucine aminopeptidase) extracellular activities were analysed immediately after retrieval as described by Chr ost and Velimirov (1991). Data were normalised to litre or dry weight (60 °C, 24 h) and reported as micromoles of substrate released per litre or per sediment gram per hour. Each analysis was replicated three times for sediment and water samples for each substrate concentration. 2.7. Dissolved organic matter (DOM) DOM was determined with the synchronous fluorescence method (Ferrari and Mingazzini, 1995), a simultaneous scanning of the excitation and emission monochromators, keeping a constant difference between them. Generally, the lower the difference, the higher the resolution. We utilised a difference of 25 nm. The spectra were corrected by comparison with the Milliq water spectrum. This technique gives a qualitative determination of two classes of dissolved compounds: protein-like (peak A, emission wavelength of about 300 nm) and humic-like (peak B, emission wavelength ranging between 420 and 450 nm).
3. Results The in situ sampling period covered two seasons, summer and autumn, characterised by different environmental features (Table 1), which led to strong changes in the organic matter of the harbour sediment. Besides the regular trends of temperature and salinity, determined by the seasonal period and by rainfall, we noted an absence of wind during the summer (from June to September), and an increase of wind strength and water circulation during autumn. Nevertheless, the low hydrodynamism was highlighted by the scarce oxygen concentrations (Table 1), which never reached full saturation (maximum of 84%) although the water column is very shallow. In the laboratory experiment the oxygenation was maintained by continuous aeration. The values ranged from 7.7 to 8.3 mg l1 at the beginning to 4.6–6.4 mg l1 at the end of the experiment. The saturation ranged from 100% to 68%. 3.1. In situ experiment 3.1.1. Bacterial dynamics Bacterial density (TBN, Fig. 2) in the topmost centimetre of the sediments showed different dynamics in the two stations undergoing treatment. TBN increased regularly and significantly at station 2 during the course of the treatment (up to 9-fold), while it had lower changes at station 1. On 2 June, TBN was 19 108 cell g1 at station 2, while at the end of September (two months Table 1 In situ experiment: environmental parameters related to the water column in the three stations Date
Station
T (°C)
S (PSU)
Oxygen (%)
Wind (presence/absence)
02.06 16.06 27.07 26.08 24.09 03.11 26.11
1
20.16 19.89 25.09 25.22 22.50 18.00 13.00
35.63 37.09 37.27 37.44 37.48 37.20 36.32
79 69 72 59 65 62 57
) ) ) ) + + +
02.06 16.06 27.07 26.08 24.09 03.11 26.11
2
20.17 20.00 25.70 25.28 22.70 18.07 13.10
35.71 36.87 37.2 37.35 37.47 37.18 36.57
77 70 79 67 65 62 57
) ) ) ) + + +
02.06 16.06 27.07 26.08 24.09 03.11 26.11
3
20.07 19.58 25.25 25.20 22.56 17.89 13.08
35.27 36.47 37.3 37.00 37.18 37.00 36.32
84 68 61 70 62 60 54
) ) ) ) + + +
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173 250
TBN
200 150 100 50 0 2 Jn
16 Jn
27 Jl 26 Au 24 Se S1 S2 S3
03 No 26 No
Fig. 2. In situ experiment: total bacterial number (TBN, cells 108 g1 ) in the three stations during the sampling time. Arrows denote seeding.
after the second seed with the product) it amounted to 106 108 cell g1 and reached its maximum value (173 108 cell g1 ) on 3 November. 3.1.2. Changes in sediment organic matter load and composition Fig. 3 shows the concentration of carbohydrates, proteins and lipids in the top 1 cm of the harbour sediments. Carbohydrates were the dominant biochemical
carbohydrate
40 30 20 10 0 2 Jn
16 Jn
27 Jl
2 Jn
16 Jn
27 Jl
2 Jn
16 Jn
27 Jl
26 Au 24 Se
03 No 26 No
40
protein
30 20 10 0 26 Au 24 Se 03 No 26 No
4
lipid
3 2 1
1167
component of the organic material in the natural sediments. After treatment, significant variations occurred in the sediment parameters, with a clear shift in the biochemical composition of the organic load. For example, carbohydrates had high values at station 2 at the beginning of June (28.5 mg g1 ), and they decreased sharply (about 96%) in July (0.9 mg g1 ). Significant differences (t-test, p < 0:05 or less) were detected between the control site (station 3) and station 1 until August, confirming that the concentrations of carbohydrates were actually higher during the first months of bioremediation. From September the control and station 1 were similar, suggesting that the bioremediation treatment was effective starting from autumn. Instead, the differences between the control site and station 2 were significant (t-test, p < 0:02) only at the beginning of the study period, before the treatment. Then the two sites became very similar in concentration, suggesting a prompt response of the system to the treatment. Proteins showed an irregular downward trend at station 2, from 2 June (15.3 mg g1 ) to the beginning of November (6.3 mg g1 ), with two peaks in August and at the end of November (8.5 and 8.7 mg g1 , respectively). Instead, proteins showed a depletion of approximately 50% at station 1 only from July to September and then returned to the starting concentrations. Proteins prevailed over carbohydrates after treatment. The differences for protein concentrations between the control site and the two treated stations were similar to those previously described for carbohydrates. At station 2 the data suggested a quicker decrease in the sedimentary organic matter to values similar to those of the control site. The bacterial-nitrogen contribution to the N protein pool of sediments––estimated by converting bacterial biomass into nitrogen content––was negligible before treatment (0.8% and 1.2% for stations 1 and 2, respectively), but increased considerably, reaching 7.4% and 20.7% of the organic nitrogen pool (at stations 1 and 2, respectively) on 26 November. Lipids did not show strong variations, ranging between 0.4 and 2.0 mg g1 at all the stations. The first and the last samplings were, in fact, not significantly different (t-test, p > 0:05). Despite their relatively low concentrations at the beginning of the study, if compared to the other biochemical fractions, lipids generally became more concentrated than carbohydrates starting from August. No significant differences (t-test, p > 0:1 or more) were detected between the control site and the other two stations.
0
S1
26 Au
S2
24 Se
03 No 26 No
3.2. Laboratory experiment
S3
Fig. 3. In situ experiment: sedimentary organic matter (carbohydrates, proteins and lipids, mg g1 ) during the sampling time. Arrows denote seeding: (H) concentration significantly different from the control station one (t-test, p < 0:05).
3.2.1. Bacterial dynamics Changes in benthic bacteria density are shown in Table 2. Before treatment, TBN values amounted to 187 108 cell g1 . The maximum value was reached
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
Table 2 Laboratory experiment: total abundance of bacteria (TBN) and abundance of the three size classes (TBN1, TBN2 and TBN3) in the sediment and water Days
TBN
S.D.
TBN1
TBN2
Sediment Control 10 20 43 60
185 293 82 213
A treatment 10 20 43 60 B treatment 10 20 43 60
8 48 14 63
78 60 23 114
65 201 36 63
42 33 23 36
265 350 81 152
15 67 12 1
135 183 24 52
94 122 35 62
36 45 22 38
319 534 369 142
50 69 87 31
110 186 163 39
111 248 123 68
98 100 83 35
19 43 74 61
2 4 2 17
14 14 18 6
5 14 23 34
0 16 32 20
141 152 174 390
8 5 8 59
47 33 37 23
40 93 2 40
54 26 135 326
B treatment 10 221 20 133 43 816 60 3604
96 4 288 1083
126 27 250 462
32 64 223 648
62 42 343 2494
Water Control 10 20 43 60 A treatment 10 20 43 60
TBN3
contribution of size classes
1168
large 75%
medium small
50% 25% 0%
before treatment
cont.
B
A
Fig. 4. Laboratory experiment: contribution (%) of different size classes to total bacterial number (sediment) before treatment and in the three experimental systems (control, A treatment and B treatment) at the end of the experiment (60 days).
3.2.2. Sediment, particulate and dissolved organic matter Concentrations of carbohydrates and proteins in the sediment of the experimental microcosms are reported in Fig. 5. Carbohydrates, which were the main biochemical class of organic matter in the untreated sediments, decreased to a low level in treatment A (from 14.2 to 12.4 mg g1 ; about 13% of sediment carbohydrates were decomposed). Treatment B displayed a higher concentration decrease (from 19.3 to 14.1 mg g1 ). However, especially the B treatment was actually not very effective. In fact we observed no significant differences between the concentrations of the control and the two treatments during the experiment (t-test, p > 0:1), except after 60 days of treatment A, when the carbohydrate concentration of the A boxes was significantly lower than the control (t-test, p < 0:03). The treatment was particularly effective for PRT concentration in A treated boxes, which decreased from
Data are reported in cell 108 g1 (sediment) and cell 108 l1 (water) with standard deviations. carbohydrate
40 30 20 10 0 10
20
43
60
10
20
43
60
40 30 protein
20 days after the beginning of the experiment (350 and 534 108 cell g1 in A and B treatments, respectively). Then, the two treated systems displayed decreasing values. Contributions of small, medium and large-size bacteria to sedimentary TBN are summarised in Fig. 4. Small and medium-size benthic bacteria were the most abundant (44% and 42% of TBN, respectively) before treatment. Changes in the community structure occurred after two months, with an increase in medium and large-size bacteria and a reduction in the percentage of the smaller component in the treated systems. Bacterial abundance in the water (Table 2) showed irregular increases until the end of the experiment, ranging from 141 to 390 and from 221 to 3604 108 cell l1 for the A and B treatments, respectively.
100%
20 10 0
A treatment
B treatment
control
Fig. 5. Laboratory experiment: sedimentary organic matter (carbohydrates and proteins, mg g1 ) during the sampling time: (H) concentration significantly different from the control one (t-test, p < 0:05).
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
5.4 to 3.5 mg g1 (about 35%). We did not observe significant differences between the control and the A boxes in the first two samplings (t-test, p > 0:2), while during the third sampling the concentration in treatment A was significantly lower ðp < 0:02Þ and, during the fourth sampling, the difference was still slightly significant ðp < 0:06Þ. The significant differences between the control and the treatment B boxes highlighted a strong increase in proteins in the treated sediment despite the bioremediation treatment. In the water of the A boxes, where sedimentary organic matter was depleted in the course of the experiment, particulate organic matter showed irregular trends (Fig. 6). Particulate proteins and particulate carbohydrates were affected differently by the treatment, since particulate carbohydrates decreased regularly from 0.9 to 0.3 mg l1 , whereas particulate proteins first increased to 5.0 mg l1 after 20 days of incubation, and dropped to 2.4 mg l1 during the third sampling. Dissolved organic matter generally recorded a progressive increase with time as reported in Table 3. The first peak indicates the most bioavailable fraction (protein-like), while the second peak corresponds to the refractory and structural material (humic-like). In particular, an increase in the protein-like fraction was observed in the A and B systems (from 171 to 319 fu and from 144 to 278 fu, for A and B, respectively), while lower difference was observed for the control boxes (from 154 to 231 fu). The B peak showed increases (from two to three fold) in all the systems, in particular in the B treatment boxes (from 63 to 174 fu).
carbohydrate
Table 3 Laboratory experiment: values of the peaks A and B related to dissolved organic matter (DOM) Days
Peak A
S.D.
Peak B
S.D.
Control 10 20 43 60
154 228 nd 231
12 3 nd 15
54 92 nd 123
5 11 nd 6
A treatment 10 20 43 60
171 307 223 319
9 66 11 58
68 102 78 137
9 8 4 3
B treatment 10 20 43 60
144 198 221 278
10 3 7 8
63 87 126 174
6 4 7 5
Data are reported in fluorescence intensity units.
3.2.3. Water and sediment enzymatic activities The values (V max) of the two hydrolytic enzyme activities, in sediment and water, 10 and 60 days after beginning of the laboratory experiment, are presented in Table 4. In sediments, both activities showed a clear (up to 2- or 3-fold) increase at the end of the experiment. The trends are the same for the treated boxes and control, however, values about 50% higher were recorded for treatments A and B than for untreated sediments. Enhanced degradation is confirmed by the results obtained in the water compartment, where the values increased by 10 (leucine aminopeptidase) and as much as 300 times (b-glucosidase).
4. Discussion
6
4.1. Microbial community
4 2 0 10
20
43
60
6
protein
1169
4 2 0 10
20
A treatment
43
B treatment
60
control
Fig. 6. Laboratory experiment: organic matter in the water (carbohydrates and proteins, mg l1 ).
Bacterial abundance in the harbour sediments not only largely exceeded the values reported in the literature for nearby coastal areas (Albertelli et al., 1999; Danovaro and Fabiano, 1995; Danovaro et al., 1996), but also data related to areas rich in organic matter (Bottcher et al., 2000). Benthic bacteria have been proved to be highly effective ‘‘markers’’ in monitoring the performance of bioremediation by microbial inoculants (Wackett and Ellis, 1999). During the in situ experiment, at the beginning of treatment the three stations showed similar TBN (ANOVA, p ¼ 0.64), while at the end of the sampling period station 2 showed a remarkable increase (on average 8-fold) and in station 1 the increase was only 3fold on average. The poor water renewal and the continuous organic matter input characterising station 1
1170
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
Table 4 Laboratory experiment: enzymatic activity (V max) for leucine aminopeptidase (LA) and b-glucosidase (BG) in the sediment (lmol g1 h1 ) and water (lmol l1 h1 ) Days
Sediment
Water
LA
S.D.
BG
S.D.
LA
Control 10 60
25.9 41.4
6.0 2.2
1.7 6.1
0.1 0.6
1754.3 nd
A treatment 10 60
43.3 86.0
9.9 11.1
3.2 8.2
0.5 0.8
B treatment 10 60
38.3 87.8
0.7 6.8
6.4 nd
0.9 nd
S.D.
BG
S.D.
41.7 nd
3.0 173.7
0.1 5.2
289.0 2980.7
27.8 52.5
1.9 236.0
0.1 2.7
254.1 3927.1
12.6 140.0
1.0 350.4
0.1 9.5
Data are reported for the first (10 days) and last (60 days) samplings.
seem to be limiting factors affecting the increase in bacterial density if compared to station 2. In the laboratory experiment, an increase in bacterial abundance in the water similarly to that occurred in the sediment of station 2 was observed, while the increase in the sediment was related only to the first 20 days of treatment. These differences may be due to several reasons, such as the competition between autochthonous and allochthonous microorganisms and the need, for the latter, of acclimation to the new environmental (physical and chemical) conditions. However, a change in cell size may be one of the microbial community responses to the new conditions dependent on bioremediation treatments. Literature data about bacteria cell size are often contradictory. Sometimes the authors report an increase in cell size in hypereutrophic environments if compared to meso- and oligotrophic areas (Sommaruga, 1995), while other authors report that the differences between small and large bacteria abundances are not significant (Tumber et al., 1993). Our sampling area data are in agreement with the first observation. At the beginning of the experiment, when no bioremediation was active, large bacteria composed about 14% of the total number. The large size fraction in oligotrophic Ligurian coastal sediments as well as in mesotrophic seagrass sites is generally smaller if compared to the other size classes (about 1% of the total number; Danovaro et al., 1996). A change in bacterial size occurred during the laboratory experiment with an increase in medium and large size fractions. Such changes, which occurred when bioremediation was in progress, are likely to be the result of adaptive mechanisms (Sommaruga and Robarts, 1997). These changes also suggest that microbial community dynamics may be strongly affected by the competition between autochthonous and allochthonous bacteria, by changes in the quality of bacteria food (due to the increase in DOM, for instance, as suggested by Robarts and Wicks, 1990) and, probably, by the grazing pressure
exerted by protozoans and meiofauna (Lebaron et al., 2001).
4.2. Microbial activity Functional parameters, such as enzymatic activity, were investigated to evaluate microbial community efficiency and organic matter exploitation. Hydrolytic enzymes are known to be the rate-limiting step of degradation (Chr ost and Velimirov, 1991) and the enzymatic activity is substrate specific. Due to these characteristics, these parameters may give clues to what fraction of organic matter is mainly going to be consumed and, therefore, what may be the preferential trophic development pathway followed by the heterotrophic microbial community. The laboratory experiment provided interesting information, although this information gives only a potential response of the system under optimal conditions. In agreement with other studies on marine coastal areas (Meyer-Reil, 1986; Lopez et al., 1995), aminopeptidase (the proteasic enzyme) showed the highest activity, both in the sediment and in the water, but it showed only a slight increase during the experiment. Conversely, b-glucosidase activity (linked to the carbohydrate fraction) in the water increased more than 300 times, whereas it increased only 3–6 times in the sediments, suggesting that a bioremediation process involves the whole sediment–water system. This is also in agreement with the increasing dissolved organic matter concentration, composed of labile (peak A) and refractory (peak B) compounds, which were not metabolised by bacteria and accumulated in the overlying water. Therefore, in situ and laboratory experiments have shown that bioremediation treatment enhance natural processes which involve both the sediment and the water column. The transfer of organic materials from the sediment to the water makes
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
it easier to remove organic loads, which are often linked to inorganic or refractory organic compounds. Other authors (Karner and Herndl, 1992; Meyer-Reil and K€ oster, 2000) have observed that the bacteria community development in areas with increasing organic matter concentrations may generally follow two phases. During the first phase, a coupling between bacteria activity (both in terms of hydrolytic activity and biomass production) and organic matter supply is observed. Conversely, during the second phase, some environmental forcing (such as temperature and light; Meyer-Reil and K€ oster, 2000) or internal processes (such as enzyme hyperproduction by bacteria; Karner and Herndl, 1992) may affect this balance, thus leading to a loose of the hydrolysis-uptake coupling. However, the likely presence of hydrolytic enzymes not connected with bacteria but persisting in the environment (Misic et al., 1998) may lead to misleading interpretations of bacterial activity. The uncoupled solubilisation-bacteria consumption process is, therefore, an important mechanism by which bacteria exert biochemical pressure on the large-size phase of organic matter (both sedimentary and particulate) to regulate the flux to dissolved organic matter (Karner and Herndl, 1992; Smith et al., 1992). These processes may explain the different rates of increase of bacteria number and enzymatic activity. Since enzyme activities are mainly stimulated by organic matter availability, the relatively low aminopeptidase/glucosidase ratios found in our study may be related to the higher availability (in terms of easily degradable glucides) of the carbohydrate fraction in the Genoa harbour sediments. Moreover, K€ oster et al. (1997) observed that the proteolytic/glycolytic enzyme ratio increased from mesotrophic to eutrophic and hypertrophic conditions. Therefore, the decrease in this ratio in our experiment is a signal of recovering. 4.3. Organic matter and its consumption The sediments in the Genoa harbour are characterised by an extremely high organic matter content, in agreement with scientific literature on highly eutrophic coastal areas (Fabiano et al., 2001; Meyer-Reil and K€ oster, 2000). In particular, protein and carbohydrate concentrations were much higher than those reported for the adjacent coastal areas of the Ligurian Sea (Danovaro and Fabiano, 1995). Organic matter and microbial biomass are generally directly correlated. However, with increasing eutrophication, this relationship is not simply linear. Meyer-Reil and K€ oster (2000) observed that in the Southern Baltic Sea the correlation was linear up to carbon concentrations of 10 mg cm3 , and increasing carbon concentrations were reflected by a decrease of microbial biomass and potential decomposition. The existence of a concentration threshold, which divides the balanced systems
1171
from the others where the potential for consuming and recycling the organic matter is affected, is a very important ecological feature. Bioremediation may be effective when the proportion of the organic matter to the allochthonous and autochthonous bacteria is kept under the maximum limit for decoupling. Moreover, from an ecological point of view, the water–sediment system of the ancient port of Genoa has similar features to those of hyper-eutrophic systems characterised by strong fluctuations in the abundance of many of the components of the food web. The exponential growth of several species, in the form of pulses, and their sudden collapse are good examples of low ecosystem stability and loss of homeostatic mechanisms. However, at the same time, they may be required in order to allow such ecosystems to exist. These features are somehow enhanced when bioremediation is carried out. The treatment, in a first phase, introduces huge amounts of allochthonous microbes, better balancing the bacteria and organic matter, which, after the acclimation, facilitates the assessment of recovery. However, although the quantity of organic substrates may highly influence the development of the microbial community and its recycling activity, the quality of organic materials is equally important. At station 2, the constant depletion of carbohydrates from the first seeding suggests that the resident carbohydrates were mainly labile, and the remnants were, instead, composed of the more refractory glucidic substances. Another conclusion is that the input of carbohydrates from the coast was not continuous, but that the concentrations were the result of a long-term accumulation. Conversely, the proteins, generally thought to be an essential and largely bioavailable food source for benthic metabolism (Mayer et al., 1995), were not degraded at the same rate as the carbohydrates. The decomposition of specific organic substrates is unlikely to follow the same patterns, but rather to depend on the capabilities of microorganisms (as a consequence of a specific population structure––Boschker et al., 2001––and of the expression of specific tools for degradation––hydrolytic enzymes––Misic and Fabiano, 1996) and on the biochemical features of the substrate. Thus, the high protein content detected throughout the entire sampling period may be related to the refractory nature of the sedimentary protein pool. Proteins may decrease their availability by means of sorption by sediment (Mayer et al., 1995). Moreover, especially in coastal areas, freshwater input supplies the system with complex molecules (humic acids for instance; Hubberten et al., 1994), which embed proteins in highly refractory micro and macro aggregates. Moreover, direct protein glycosilation may decrease their degradation rates (Smith et al., 1992). Thus, an increase in protein refractivity due to environmental conditions has to be taken into
1172
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173
account before applying bioremediation protocols. Moreover, the input of proteins to the system was, probably, more continuous, leading to high protein: carbohydrate ratios (on average more than 5 during the whole study period of the in situ experiment), typical of eutrophic and eutrophicated systems (Fabiano et al., 2001). In the laboratory, the protein trends of the A system decreased, instead, showing that a more effective recovery of the sedimentary system can be achieved by stopping organic matter enrichment. The increase in proteins during the B experiment was probably due to high bacterial growth and enhanced hydrolytic activity (especially in the overlying water) likely to lead to a generally higher release of proteins embedded in the refractory component of the organic matter. This, in turn, did not lead to a net protein increase, but to their enhanced detectability by our analytical method. However, the high protein content of the sedimentary organic matter can also be due to increased (up to 11fold) bacterial abundance and biomass. Actually, the bacterial-nitrogen contribution to the N protein pool of sediments increased considerably. Bacteria are present in high numbers, also in natural conditions, and their biomass is greater than the biomass of the other benthic organisms (Meyer-Reil and K€ oster, 2000). Protein liberation from aggregates and bacteria increase are processes that, despite increasing the protein content of the sediment and the overlying water, change the availability of this high food-value compounds. Free or living proteins are more labile and need lower time and energy resources to be recycled, thus increasing the rates and fluxes of energy and materials in the system. Acknowledgement We would like to thank Idrabel Italia S.r.L. for kindly providing the bioremediation product. References Albertelli, G., Covazzi-Harriague, A., Danovaro, R., Fabiano, M., Fraschetti, S., Pusceddu, A., 1999. Differential responses of bacteria, meiofauna and macrofauna in a shelf area (Ligurian Sea, NW Mediterranean): role of food availability. Journal of Sea Research 42, 11–26. Bligh, E.G., Dyer, W., 1959. A rapid method for total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37, 911–917. Boschker, H.T.S., de Graaf, W., K€ oster, M., Meyer-Reil, L.A., Cappenberg, T.E., 2001. Bacterial populations and processes involved in acetate and propionate consumption in anoxic brackish sediment. FEMS Microbiology Ecology 35, 97–103. Bottcher, M.E., Hespenheide, B., Llobet-Brossa, E., Beardsley, C., Larsen, O., Schramm, A., Wieland, A., Bottcher, G., Berninger, U.-G., Amann, R., 2000. The biogeochemistry, stable isotope geochemistry and microbial community structure of a temperate intertidal mudflat: an integrated study. Continental and Shelf Research 20, 1749–1769.
Cardell, M.J., Sarda, R., Romero, J., 1999. Spatial changes in sublittoral soft-bottom polychaete assemblages due to river inputs and sewage discharges. Acta Oecologica 20 (4), 343–351. Cloern, J.E., 2001. Our evolving conceptual model of the coastal eutrophication problem. Marine Ecology Progress Series 210, 223–253. Chr ost, R.J., Velimirov, B., 1991. Measurement of enzyme kinetics in water samples: effect of freezing and soluble stabilizer. Marine Ecology Progress Series 70, 93–100. Danovaro, R., Fabiano, M., 1995. Seasonal and interannual variation of benthic bacteria in a seagrass bed of the Mediterranean Sea: relationship with labile organic compounds and other environmental factors. Aquatic Microbiology Ecology 9, 17–26. Danovaro, R., Fabiano, M., Della Croce, N., 1993. Labile organic matter and microbial biomasses in deep-sea sediments (Eastern Mediterranean Sea). Deep-Sea Research 40, 953–965. Danovaro, R., Della Croce, N., Fabiano, M., 1996. Microbial response to oil disturbance in the coastal sediments of the Ligurian Sea (Western Mediterranean). Chemistry and Ecology 12, 187– 198. DellÕAnno, A., Mei, M.L., Pusceddu, A., Danovaro, R., 2002. Assessing the trophic state and eutrophication of coastal marine systems: a new approach based on the biochemical composition of sediment organic matter. Marine Pollution Bulletin 44, 611–622. Dubois, M., Gilles, K.A., Hamilton, P.A., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Analytical Chemistry 28, 350–356. Fabiano, M., Misic, C., Manini, E., Danovaro, R., Povero, P., Chiatti, S., 2001. Distribution and biochemical composition of suspended and sedimentary organic matter in the northern Adriatic. In: Faranda, F.M., Guglielmo, L., Spezie, G. (Eds.), Structure and Processes in the Mediterranean Ecosystems. Springer Verlag, Heidelberg, pp. 447–454. Ferrari, G.M., Mingazzini, M., 1995. Synchronous fluorescence spectra of dissolved organic matter (DOM) of algal origin in marine coastal waters. Marine Ecology Progress Series 125, 305–315. Hartree, E.F., 1972. Determination of proteins: a modification of the Lowry method that gives a linear photometric response. Analytical Biochemistry 48, 422–427. Head, I., 1998. Bioremediation: towards a credible technology. Microbiology 144, 599–608. Hubberten, U., Lara, R.J., Kattner, G., 1994. Amino acid composition of seawater and dissolved humic substances in the Greenland Sea. Marine Chemistry 45, 121–128. Karner, M., Herndl, G.J., 1992. Extracellular enzymatic activity and secondary production in free-living and marine-snow associated bacteria. Marine Biology 113, 341–347. K€ oster, M., Dahlke, S., Meyer-Reil, L.A., 1997. Microbiological studies along a gradient of eutrophication in a shallow coastal inlet in the Southern Baltic Sea (Nordrugensche Bodden). Marine Ecology Progress Series 152, 27–39. Lebaron, P., Servais, P., Troussellier, M., Courties, C., Muyzer, G., Bernard, L., Sch€afer, H., Pukall, R., Stackebrandt, E., Guindulain, T., Vives-Rego, J., 2001. Microbial community dynamics in Mediterranean nutrient-enriched seawater mesocosms: changes in abundances, activity and composition. FEMS Microbiology Ecology 34, 255–266. Lee, S., Fuhrman, J.A., 1987. Relationship between biovolume and biomass of naturally derived marine bacterioplankton. Applied Environmental Microbiology 53, 1298–1303. Lopez, N.I., Borras, G., Vallespin os, F., 1995. Effect of heavy metals on enzymatic degradation of organic matter in sediments off Catalonia (Northeastern Spain). Scientia Marina 59, 149–154. Marsh, J.B., Weinstein, W.J., 1966. A simple charring method for determination of lipids. Journal of Lipid Research 7, 574–576.
M. Fabiano et al. / Marine Pollution Bulletin 46 (2003) 1164–1173 Mayer, L.M., Schick, L.L., Sawyer, T., Plante, C.J., 1995. Bioavailable amino acids in sediments: a biomimetic, kinetic-based approach. Limnology and Oceanography 40, 511–520. Meyer-Reil, L.A., 1986. Measurement of hydrolytic activity and incorporation of dissolved organic substrate by microorganisms in marine sediments. Marine Ecology Progress Series 31, 143–149. Meyer-Reil, L.A., K€ oster, M., 2000. Eutrophication of marine waters: effects of benthic microbial communities. Marine Pollution Bulletin 41, 255–263. Misic, C., Fabiano, M., 1996. A functional approach to the assessment of the nutritional value of the particulate organic matter. Chemistry and Ecology 13, 51–63. Misic, C., Povero, P., Fabiano, M., 1998. Relationship between ectoenzymatic activity and availability of organic substrates (Ross Sea, Antarctica): an experimental approach. Polar Biology 20, 367–376. Murphy, T.P., Lawson, A., Kumagai, M., Babin, J., 1999. Review of emerging issues in sediment treatment. Aquatic Ecosystem Health and Management 2, 419–434. Palumbo, R., Ferguson, J.E., Rublee, P.A., 1984. Size of suspended bacterial cells and association of heterotrophic activity with size fraction of particles in estuarine and coastal waters. Applied Environmental Microbiology 48 (1), 157–164.
1173
Robarts, R., Wicks, R.J., 1990. Heterotrophic bacterial production and its dependence on autotrophic production in a hypertrophic African reservoir. Canadian Journal of Fisheries and Aquatic Science 47, 1027–1037. Smith, D.C., Simon, M., Alldredge, A.L., Azam, F., 1992. Intense hydrolytic enzyme activity on marine aggregates and implications for rapid particle dissolution. Nature 359, 139–142. Sommaruga, R., 1995. Microbial and classical food webs: a visit to a hypertrophic lake. FEMS Microbiology Ecology 17, 257–270. Sommaruga, R., Robarts, R.D., 1997. The significance of autotrophic and heterotrophic picoplancton in hypertrophic ecosystems. FEMS Microbiology Ecology 24, 187–200. Tumber, V.P., Robarts, R.D., Arts, M.T., Evans, M.S., Caldwell, D.E., 1993. The influence of environmental factors on seasonal changes in bacterial cell volume in two saline lakes. Microbiology Ecology 26, 9–20. Wackett, L.P., Ellis, L.B.M., 1999. Predicting biodegradation. Environmental Microbiology 1 (2), 119–124. Wolfe, D.A., Krahn, M.M., Casillas, E., Sol, S., Thompson, T.A., Lunz, J., Scott, K.J., 1996. Toxicity of intertidal and subtidal sediments contamined by the Exxon Valdez spill. In: Rice, S.D., Spies, R.B., Wolfe, D.A., Wright, B.W. (Eds.), Proceedings of the Symposium Exxon Valdez Oil Spill, pp. 121–139.