In shrimp culture pond, bottom soil is generally a sink of large amounts of
organic matters derived from uneaten feed, fertilizers and excretory products
of aquatic organisms (Kassila, 2003). It is known that
the major input of nitrogen waste comes from protein in feed pellets in which
approximately 15-30% of nitrogen in feed is converted into shrimp biomass, while
the rest is accumulated in high organic content sediment at the pond bottom
(Hargreaves, 1998). Release of toxic inorganic nitrogen
compounds especially ammonia is from the decomposition process of organic matters
and conversion of ammonia to other nitrogenous compounds of nitrogen cycle is
regulated by microorganisms in the sediment.
In general, aerobic decomposition of organic waste especially at the pond bottom
results in ammonia-nitrogen release. The major sources of organic matter for
ammonification are uneaten feed, animal excretion and microorganism debris.
In nature, toxic inorganic nitrogen compounds i.e., ammonia and nitrite are
converted to low toxic nitrate-nitrogen by nitrification process. High concentration
of nitrate in aquaculture pond can result in eutrophication, however, this is
rarely occurred since nitrate is eliminated from the pond by denitrification
process in sediment (Boyd, 1995; Burford
and Lorenzen, 2004).
To enhance denitrification process, various organic carbon compounds such as
cellulose, molasses, hydrolyzed corn straw, acetic acid, propanol, butanol,
glycol, methane, glucose, ethanol and methanol have been used as the substrates
for denitrifying bacteria (Sobieszuk and Szewczyk, 2006;
Shrimali and Singh, 2001; Ferguson,
1994; Mateju et al., 1992). Methanol is reported
as one the most effective and the most economical carbon source for denitrifying
bacteria especially in aquaculture systems (Hamazah and
Ghararah, 1996; Menasveta et al., 2001).
Based on the stoichiometric relationship of heterotrophic denitrification, the
methanol to nitrate ratio have been reported in term of C:N ratio. The C:N ratio
for complete denitrification process with methanol as a carbon source was between
3 to 10 (Mateju et al., 1992; Ebeling
et al., 2006).
Present preliminary study (Kutako et al., 2005)
showed that organic nitrogen waste in shrimp pond sediment was removed by a
series of ammonification, nitrification and denitrification processes in which
the accumulation of ammonia or nitrite to toxic levels could be found if there
was high loading of organic nitrogen. With our attempt to develop the Recirculating
Aquaculture System (RAS) in lining pond (pond without bottom soil) incorporated
with nitrification biofilters, accumulation of nitrate is commonly found after
long usage of the RAS. There are several techniques available to remove high
nitrate in aquaculture pond including plant uptake and several types of denitrification
reactors. However, these options are not suitable due to problems with cost
and operation complexity, thus, nitrate removal can be practically achieved
only by water discharge. Recently, enhance denitrification process in soil by
organic carbon addition i.e., acetate and ethanol was reported by Martin
et al. (2009). Another preliminary study by present group (Malapol
et al., 2008) found that denitrification rate of the sediment increased
from 380 mg-N/m2/day in control without methanol to 5400 mg-N/m2/day
in sediment tank with methanol addition. Therefore, high nitrate water from
RAS could be removed by tank or pond containing sediment layer. This leaded
to the development of nitrate removal technique using enhanced denitrification
in sediment tank system. However, carbon addition might affect the balance between
nitrogen removal processes in which the inhibitory effect of organic carbon
on nitrification process has been reported (Strauss and
Lamberti, 2000). The objective of this study was to obtain more information
on effect of methanol additions on nitrogen transformation through nitrification
process and changes of bacterial diversity in sediment from shrimp pond. A better
understanding of these topics will be useful in improving techniques to remove
nitrate from closed system aquaculture pond using enhanced denitrification pond
MATERIALS AND METHODS
Sediment sampling and sediment chambers set up: Shrimp pond in Pathum
Thani Province, Thailand, was chosen for sediment sampling. This shrimp pond
was closed-system farm culturing white shrimp (Penaeus vannamei). Shrimp
pond sediment with approximately 5 cm in depth was collected by grab sampler
during May-June 2007. Concentration of inorganic nitrogen compounds (ammonia,
nitrite and nitrate) in the water column, pore water and extracted soil are
shown in Table 1. Soil texture of sediment sample was 58%
clay, 36% slit and 6% sand. Sediment sample was immediately transferred to the
laboratory (Center of Excellence for Marine Biotechnology, Chulalongkorn University)
in dark and cold condition, where it was immediately packed into sediment chambers.
The sediment chambers were made of acrylic plastic tube with 11 cm in diameter
and 40 cm height (Fig. 1).
||Concentration of ammonia, nitrite and nitrate in the water
column, pore water and extracted soil from shrimp pond
|Mean±SD, n = 3
||Schematic diagram of the sediment chamber used in this study
Sediment was packed to the thickness
of 10 cm in the chamber then the chambers were filled with 1.4 L of 20 g L-1 seawater and incubated under dark condition with continuous aeration through
Experimental procedures: To test the effects of organic carbon addition on inorganic nitrogen conversion and microbial community, the experimental units were separated into three groups i.e., (1) control chambers without organic carbon addition, (2) treatment-1 in which chambers were supplemented with methanol at C:N ratio of 2:1 (1.85 mL methanol L-1) and (3) treatment 2 that methanol concentration was increased to C:N ratio of 4:1 (3.7 mL methanol L-1). All treatments were performed with 3 replicates. The experiment was conducted in dark condition to eliminate the activity of photosynthetic microorganisms. During the experimental period, water samples were collected and analyzed in order to monitor the changes of inorganic nitrogen compounds. At the initial day (day 0) and final day (day 48), sediment samples were collected and changes of sediment bacterial communities were evaluated using PCR-DGGE technique.
Water samples in the experimental chambers were collected
at 10 cm below the water surface following by filtered through a GF/C filter and
kept in refrigerator prior to analysis. Ammonia (mg NH4
and nitrite (mg NO2
) were analyzed using Phenate method
and Diazotization method, respectively (Strickland and Parson,
). Nitrate (mg NO3
) was measured by the ultraviolet
spectrophotometric method at 220 and 275 nm (APHA, 1992
Other water quality parameters included alkalinity (titration method: APHA
), dissolved oxygen
(digital DO-meter model HI91410 (Hanna Instrument,
Portugal)), temperature and pH (pH meter model pH 10 (YSI Inc., USA)).
Sediment analysis: During the experiment, Oxidation-Reduction Potential
(ORP) at 2 cm below soil surface was continuously monitored using HI98240 ORP
meter (Hanna Instrument, Portugal). Organic matter in sediment was determined
using combustion method (Kim et al., 1997). On
weekly basis, sediment in the chamber was collected using 1 cm diameter core
made of clear acrylic tube. The core was gently penetrated into the sediment
layer and only top 2 cm of sediment layer was collected for further analysis.
Sediment samples were placed in a micro-centrifuge tube and spin at 3000 rpm
for 20 min to separate pore water. Consequently, 1 g of wet sediment was extracted
with 10 mL of 2 M KCl solution at 25°C and shaken for 30 min at 120 rpm
(Schinner et al., 1996). The soil mixture was then
centrifuged and filtered with GF/C filter in order to obtained extracted sediment
solution. Finally, both pore water and extracted sediment were analyzed for
inorganic nitrogen compounds.
Bacterial diversity analysis
Genomic DNA extraction and PCR-DGGE analysis: Sediment samples were persevered
at -20°C for bacterial diversity analysis. Bacterial genomic DNA was extracted
directly from 0.25 g of the sediment using a protocol of the Fast DNA spin kit
for soil (Qbiogene, USA). Total DNA was subsequently verified by electrophoresis
in 1% agarose gel for 20 min at 100 V. Gel was stained with ethidium bromide
and visualized by gel documentation (Dolphin-Doc Plus, USA). The 16S rDNA was
amplified using PRBA338f primer attached with GC clamp (40 bp) at 5' end (5'
CgC CCg CCg CgC gCg gCg ggC ggg gCg ggg gCA Cgg ggg gAC TCC TAC ggg Agg CAg
CAg 3') and PRUN518r primer (ATT ACC gCg gCT gCT gg) (Muyzer
et al., 1993). This primer set is specific to all bacterial groups.
Polymerase chain reaction mixture (30 μL total volume) was composed of
5 μL of genomic DNA solution, 7 μL of dH2O, 1.5 μL
of each primers (10 μM) and 15 μL of Taq PCR Master Mix kit (QIAGEN,
Valencia, USA). The PCR protocol was initial denaturation at 92°C for 2
min followed by 30 cycles of denaturation at 92°C for 1 min, annealing at
55°C for 30 sec and extension at 72°C for 1 min. PCR cycling was completed
by a final extension at 72°C for 6 min. The quality of PCR product was examined
by gel electrophoresis with 2% (w/v) of agarose in 1xTAE buffer.
Nucleotide fragments of the PCR products were separated by DGGE (Dcode system, Bio-Rad, USA). A duplicate analysis of DGGE pattern was performed within the gel. Twenty microliter of the PCR products was run on polyacrylamide gel (8% of a 37.5:1 acrylamide-bisacrylamide mixture solution in 1xTAE buffer). Denaturing gradients of 30 to 60% (100% denaturant contained 7M urea and 40% formamide) was used. The electrophoresis was operated at 130 V for 6 h in 1xTAE buffer at a constant temperature of 60°C. Thereafter, DNA bands were stained with silverstar® staining kit (Bioneer, Korea) following the manufacturers protocol.
Bacterial diversity analysis of DGGE
Shannon-Weaver diversity index: Number of DNA bands and their relative intensity
within each lane in the gel were counted. This data was then used to calculate
the diversity indices including species richness (R), evenness (E) and the Shannon-Weaver
index (H) (Shannon and Weaver, 1949). Species richness was
determined from the bands number in each lane. The Shannon-Weaver index was
used as a diversity index and it was calculated as follows:
where, Pi is the importance probability of the bands in each lane,
calculated from ni/N, where ni is the peak height of a
band and N is the sum of all peak heights in each lane. Species evenness was
calculated using the equation (Eichner et al., 1999):
Cluster analysis: Similarity between the DNA banding pattern generated
by PCR-DGGE of the sediment samples were analyzed using Nei and Lis coefficients.
The UPGMA (unweighted pair-group method with arithmetic averages: Nei
and Li (1979)) with BIO-1D software (Vilber Lourmat, France) was used for
cluster analysis and dendrogram generation.
DNA sequencing and phylogenetic tree analysis: The interested DNA bands
were cut from DGGE gel, immersed in 50 μL sterilized water and kept at
4°C for 16 h. Eluting DNA solution was re-amplified, purified using a QIAquick
purification kit protocol (QIAGEN, Valencia, USA) and sequenced (Tech Dragon,
Hongkong). DNA sequences were analyzed using Seqmatch program of the Ribosomal
Database Project II (RDP II) (http://rdp.cme.msu.edu/).
Distance matrix analyses were performed with a Jukes-Cantor model. With RDP
II database, 16S rRNA sequences were aligned with the nearest neighbors and
a phylogenetic tree was then constructed by neighbor-joining method.
Nucleotide sequence accession numbers: The 16S rRNA gene partial sequences obtained in this study have been submitted to the GenBank under accession numbers, FJ386435 through FJ386437 and FJ665406 through FJ665419.
Inorganic nitrogen compounds conversion: During 48 days experiment,
DO in water column of all experimental units was almost constant at 6.5±1.8
mg O2 L-1 due to continuous aeration. Change in inorganic
nitrogen compounds in the water column is shown in Fig. 2.
It was found that, nitrification process with both ammonia oxidation and nitrite
oxidation steps was somewhat inhibited by methanol addition in which ammonia
and nitrite took longer time to be eliminated. For more detail, between initial
day to day 9th, ammonia in control chambers (without methanol) was gradually
released to the highest concentration of 8.74±2.16 mg N L-1
(Fig. 2A) and following by subsequently decreased to 0.29±0.05
mg N L-1 at day 18th.
Changes in (A) ammonia, (B) nitrite and (C) nitrate in water
column of the control (without methanol) and treatment (with methanol addition
at C:N = 2:1 or 4:1) chambers during 48 days incubation.
With treatment 1, after adding methanol to the C:N ratio of 2:1, ammonia was
continuously released to the highest concentration of 9.73±2.17 mg N
L-1 at day 9th and it was completely removed from the water in day
27th. A similar result was found in treatment-2 which supplemented with higher
methanol (C:N ratio of 4:1) (Fig. 2B). Increases of nitrite
in all experimental units were found after ammonia oxidation (Fig.
2B). At day 46, nitrite was finally removed by the nitrite oxidation process
that converts nitrite to nitrate (Fig. 2C).
Results in Fig. 2 also indicated that inhibition level of
nitrification process was according to the concentration of methanol. This was
due to the progression of ammonia to nitrite was quickest in control chambers.
Figure 3A-C show that, ammonia in pore water of all chambers
was high at the initial day then it was decreased to lower than 15 mg N g-1
Concentrations of (A) ammonia, (B) nitrite and (C) nitrate
in pore water from sediment chambers during 48 days of incubation
The results indicated that decomposition of organic matter in the sediment
rapidly released ammonia to the water column. Over 48 days incubation, concentration
of nitrite in pore water of all experimental units were constantly low concentration
(Fig. 3B) and nitrate in pore water was also constant (Fig.
Sediment bacterial community: Denaturing gradient gel electrophoresis
profile of 16S rRNA gene showed that bacterial community in the sediment was
distinctly changed during 48 days study period (Fig. 4). At
least 29 major bacteria species indicated by intense DGGE bands were found at
the initial day. After 48 days incubation, at least 4 new bacteria species were
found while at least 6 species were disappeared from control chamber.
16S rDNA-based Denaturing Gradient Gel Electrophoresis (DGGE)
profiles with cluster analysis
of the sediment bacterial community in sediment
chambers. Lane B denotes sediment sample at day 0 and lanes C to E are at
day 48. Lane C represents control (without methanol addition), lane D is
treatment-1 (C:N ratio of 2:1) and lane E is treatment-2 (C:N ratio of 4:1).
Arrows indicate DNA band for further sequencing
At day 48th, different DNA banding patterns were observed in control chamber
and all treatment chambers.
With total number of DNA bands and their relative intensities, bacterial diversity indexes were calculated and the results are in Table 2. It was found that, after 48 days incubation, the Shannon-Weaver index and species richness were higher in control chambers than treatments, while the diversity index was slightly lower. The lowest diversity index was found in high organic carbon addition (treatment 2).
Phylogenetic tree generated from the alignment of partial
16S rRNA gene sequences from sediment bacteria. DGGE bands analyzed in this
study were assigned as ORG-1 to ORG-16. The scale bar corresponds to 10
nucleotide substitution per100 nucleotide position
||Bacterial diversity in sediment chambers calculated from DGGE
To identify bacterial species in the DGGE gel, a total of 16 interested DNA
bands (Fig. 4) were excised, re-amplified and sequenced. It
was found that DNA bands ORG-10, ORG-12 and ORG-15 was appeared in all sediment
chambers during the incubation period. With RDP II database, these were closely
related to Thermobispora, Pseudomonas sp. and bacteria clone
env. OPS17. In Fig. 4, bands ORG-1, OGR-2, ORG-11, ORG-14
and ORG-15 appeared only day 0. Since ammonia and nitrite was removed in day
48, it was found the succession of new bacterial species, as represented in
DNA bands ORG-3, ORG-4, ORG-6 to ORG-9. Intensity of ORG-5 was increased in
sediment chamber which methanol supplemented to the C:N ratio of 4:1. This bacterium
was identified as Thioalcalovibrio denitrificans (Gammaproteobacteria)
which is a denitrifying bacterium that coexist with nitrifiers due to the oxygen
gradient in sediment chamber and organic carbon source (methanol). Hence, it
could be assumed that methanol could promote growth and activity of denitrifying
bacteria in the sediment chamber (Fig. 2C).
Phylogenetic distance tree for DNA bands ORG-1 to ORG-16 is shown in Fig. 5. Six lineages were evaluated in sediment reactor including phylum Proteobacteria, Firmicutes, Actinomycetales, Unclassified bacteria, Chloroflexi and Bacteriodetes. DNA bands ORG-3, ORG-5, ORG-7, ORG-8, ORG-2 and ORG-12 were grouped into Proteobacteria while all of them were most closely related to class Gammaproteobacteria which comprised of facultative anaerobic and fermentative gram-negative bacteria.
As the aim of this research was to enhance nitrogen removal by sediment from
shrimp pond, effect of methanol addition on ammonification, nitrification and
denitrification processes was testified. However, since the detection of each
process is rather complicate, this study hence evaluated the nitrogen conversion
processes by monitoring the changes of inorganic nitrogen compounds in the water.
With this assumption, ammonia in water column is continuously released from
the decomposition of organic matter in sediment under aerobic condition. This
process is so called the ammonification or aerobic decomposition. If the nitrification
and denitrification processes occurred without any limitation, low concentration
of ammonia, nitrite and nitrate would have been detected in the water because
the removal rates are higher than emerging rate. On the other hand, accumulation
of ammonia, nitrite or nitrate illustrated as peak concentrations in the water
during the experiment was assumed as an inhibition of the conversion process.
For example, a peak of nitrite found in water column of the sediment chamber
was due to an inhibition of nitrite oxidizing process in which ammonia could
be converted to nitrite but nitrite could not be converted to nitrate. Results
in Fig. 2 showed that release of ammonia in the sediment chamber
was due to the aerobic decomposition process (ammonification) of high organic
matter sediment (Avnimelech and Ritvo, 2003; Moriarty,
1997). Visscher and Duerr (1991) reported that decomposition
of organic matter usually encounters in the flocculent sediment layer, approximately
2 cm above the sediment surface. In general, nitrification can be found in both
water column and at sediment surface (Boyd, 1995; Kim
et al., 1997). As the activity of nitrifying bacteria were mostly
found in biofilm, nitrification process was hence targeted mostly at sediment-water
interactive zone with a minor portion in the suspended particles.
As shown in Fig. 2C
, nitrate concentration in the chamber was
almost constant throughout the experiment. Here, it was probably due to the coupling
nitrification-denitrification processes in which nitrate from nitrification was
further reduced by denitrification process in sediment layer. Other possibility
was the anaerobic ammonia oxidation (anammox) process that ammonia and nitrite
are converted to nitrogen gas in anoxic sediment (Rich et
). However, anammox activity in natural condition with much lower
ammonia and nitrite concentration than that found in wastewater treatment plant
was rarely reported. With this study, denitrification occurred in both control
and treatment chambers throughout the study period. This was indicated by ORP
values between -265 to -364 mV in the sediment which was the potential levels
of nitrate reduction (Li and Irvin, 2007
). At an anoxic
sediment layer, nitrate was reduced to nitrogen gas through a series of intermediate
gaseous nitrogen oxide products (Ebeling et al., 2006
Several studies suggested that addition of organic carbon could simulate denitrifying
bacteria and other heterotrophic bacterial activities due to the fact that heterotrophic
bacteria grew faster than autotrophic bacteria (Moriarty, 1997
Avnimelech and Ritvo, 2003
; Cho and
; Ebeling et al., 2006
methanol is one of the most economical carbon source
s which effective in supporting
denitrification process, higher concentration of nitrite in treatment chambers
hence suggested that an addition of methanol into the sediment chamber promoted
the ammonification process rather than the denitrification process.
With present concept of the sediment tank system for nitrogen treatment in the recirculating aquaculture systems, the results suggested that methanol could be applied to enhance the efficiency of denitrification process for nitrate removal. However, since it was found that methanol addition also inhibited nitrification process, sediment tank should be separated from animal culture tank and methanol residue in the sediment tank must be avoided. Moreover, this technique could also be applied for nitrogen removal in outdoor earthen aquaculture pond. In normal practice, water in the aquaculture pond is drained out after animal harvest. Adding methanol at appropriate concentration to the pond after water drainage possibly enhances ammonification and denitrification processes hence organic nitrogen in the sediment can be eliminated. Nevertheless, further intensive studies with organic carbon addition for nitrogen removal in outdoor aquaculture pond are therefore required.
Results from DGGE analysis illustrated that number of dominant bacterial species
in bottom soil was reduced after incubation with high methanol additions since
methanol addition could stimulate growth of specific heterotrophic bacteria
group (Moriarty, 1997). Hence, bacterial diversity in
treatment chambers were differ from that found in control chambers. This was
probably due to the effect of methanol on the progression of bacterial community
(Lucas and Hollibaugh, 2001).
Cluster analysis of the DGGE patterns with UPGMA method is shown in Fig. 4 and three major clusters were illustrated. The results suggested that bacterial community in sediment chamber strongly differed after 48 days of incubation. In control chamber, similarity within the bacterial community was decreased from 100% at day 0 to 71% at day 48. On the other hand, bacterial community in treatment 1 and 2 was rather similar.
Methanol addition at C:N ratio of 2:1 and 4:1 could enhance heterotrophic decomposition (ammonification) and denitrification in sediment chambers under laboratory conditions. However, methanol addition was also inhibiting nitrification process since treatment chambers took longer time to reduce ammonia and nitrite concentrations. Methanol addition also promoted growth of some heterotrophic bacteria. With PCR-DGGE analysis of 16S rDNA, bacterial diversity indicated by Shannon-Weaver index, species richness, species evenness and cluster analysis was clearly changed after adding methanol and incubated for 48 days. Results from DNA sequencing analysis suggested that high C:N ratio stimulated growth of denitrifying bacteria i.e., Vibrio sp., Psuedomonas sp., Planococcus sp. and Thioalcalovibrio sp. The implementation of methanol-enhanced denitrification in sediment tank must be used with special attention since high concentration of methanol could affect natural nitrification process in the tank.
Maliwan Kutako acknowledges grant from the Program Strategic Scholarships for Frontier Research Network for the Ph.D. Program Thai Doctoral degree from the Commission on Higher Education, Thailand. We would like to thank the National Center of Excellence for Environmental and Hazardous Waste and Management (NCE-EHWM), Chulalongkorn University for the use of some molecular biology facilities. This study was partially supported by the National Center for Genetic Engineering and Biotechnology, National Science and Technology Development Agency, with a compliment from Graduate School, Chulalongkorn University.