论文部分内容阅读
Effects of Photosynthetic Bacteria-enhanced Biological Floc Replacement Diets on Tilapia Growth Water Environment and Water Microbial Diversity
Abstract With Oreochromis niloticus as the object of study， glucose was added as a carbon source to promote the formation of the biological flocs for replacing part of the feed， and three gradients were set up， namely Group A （all feed）， Group B （replacement of 10% feed） and Group C （replacement of 20% feed）， so as to explore the effects of photosynthetic bacteria-enhanced biological flocs on tilapia growth and water environment conditions. Meanwhile， the Biolog-ECO technology was applied to study the changes of microbial carbon metabolism diversity in aquaculture water. The results showed that the utilization of microbial carbon sources under different feed replacement gradients increased with the extension of the culture time. The overall performance was in order of 10% replacement>all feed>20% feed replacement. A suitable replacement rate could not only enhance the overall utilization of carbon sources by water microorganisms， but also save culture costs. The principal component analysis showed that the carbon source metabolism of the water microbial communities under different feed replacement gradients was significantly different. Specifically， polysaccharides， esters and amino acids were the preferred carbon sources of water microbes， while the utilization of amines and acids was low.
Key words Photosynthetic bacteria; Biological floc; Bait substitute; Microbial community; Tilapia
Tilapia （Oreochromis niloticus） belongs to Oreochromis in Cichlidae of Perciformes in Perciformes， which is native to inland waters of Africa， and is also called "African crucian carp". Most tilapia is omnivorous fish， and they often feed on plants and debris in water bodies. Genetic improved farmed tilapia is a new tilapia strain selected in the Philippines by research institutions such as the World Fish Center at the end of the 20th century. It has strong adaptability， stable genetic traits， short breeding cycle， and fast growth， and is currently one of the leading species in the tilapia aquaculture industry[1]， as well as one of the fine species promoted by the United Nations Food and Agriculture Organization to the world. With the development of the intensive and high-yield model of aquaculture， the residual feed and fecal accumulation caused by a large amount of feed input and the harmful substances produced thereby will not only cause certain harm to the aquaculture object itself， but also affect the aquaculture water environment. With the development of microbial technology， the research of microbes in aquaculture is also deepening. There are many kinds of microorganisms， widely distributed， sensitive to the environment， easy to mutate， and irreplaceable in biological systems[2-3]. Photosynthetic bacteria can decompose and utilize organic matter in aquaculture water bodies， and have the effects of purifying water quality and improving the micro-ecological environment of water bodies[4]. The Biolog-ECO plate method directly cultures microbial populations on microplates， measures the degree of utilization of different carbon sources by microorganisms， and monitors the changes in light absorbance caused by them in real time， so as to explore the physiological characteristics of microorganisms and their community structure[5-6]. In recent years， there have been many reports on the application of photosynthetic bacteria in aquaculture[7-8]. Studies have shown that using substances including glucose， honey and sucrose as an external carbon source can well form biological flocs [9-12]. Due to the low cost of glucose and its ability to form stable and efficient biological flocs， in this study， we used glucose as an additional organic carbon source and simultaneously added photosynthetic bacteria to the aquaculture water body for enhancement， aiming to explore the water quality control issues in aquaculture and the microbial community structure and carbon source utilization situation in the formation of biological flocs and provide reference data for the study of feed replacement by biological flocs and microbial community structure under pond culture.
Materials and Methods
Culture management
The aquaculture experiment was carried out in the Freshwater Fisheries Research Center of the Chinese Academy of Fishery Sciences， Wuxi City， Jiangsu Province from July to August 2018. The experiment lasted for 60 d. Nine round culture buckets with a volume of about 1 000 L were selected. In the early stage of the experiment， fresh surface soil was added to the culture buckets to provide a source of indigenous microorganisms， and then a string of composite filler composed of fully plastic clips and aldolized vinylon yarn was hung in each bucket to provide an attachment substance for microorganisms and promote the formation of biological flocs. Finally， about 700 L of water was added to each bucket. The culture system was then sun for full exposure. In order to ensure sufficient oxygen supply during the culture experiment， a microporous aerator tube was installed at the bottom of each culture bucket to increase the dissolved oxygen concentration of the water body while making the water body fully agitated and mixed. Twenty individuals of the same size were raised in each culture bucket， with a body weight of （6.50±0.50） g. Feeding was carried out at the same frequency during the culture process， and the initial feeding amount was about 10% of the tilapia body weight. As the water environment and weather changed， the feeding amount was adjusted to 2%-3% of the tilapia body weight. The experiment was set with 3 different feed replacement gradients， namely Group A （all feed）， Group B （replacement of 10% feed） and Group C （replacement of 20% feed）， each of which was set with 3 parallels. After the ammonia level in the aquaculture water met the test requirements， we started to add the carbon source and photosynthetic bacteria liquid.
Experimental methods
According to the freshwater pond biological floc ecological cultivation technology construction operating procedures introduced by the Freshwater Fisheries Research Center， Chinese Academy of Fishery Sciences， the carbon source addition amount was determined according to A（g）=H×S×（30×CTAN-N-38）， wherein H is the average water depth of the pond， m; S is the pond area， m2; and CTAN-N is the initial ammonia nitrogen concentration of the pond， mg/L. Glucose was added once every 2 d， and the time of addition was 30 min after feeding in the morning. The carbon source was dissolved in the culture water first， and then evenly sprinkled in the culture buckets. The photosynthetic bacteria broth was cultured in the laboratory and its concentration was 109 CFU/ml. According to related study by Chen et al.[13]， it was determined that the photosynthetic bacteria broth was used once every 1 week， with 100 ml each time， 6 times in total.
After the formation of biological flocs was stabilized， water samples were collected from 30 cm below the water surface every 1 week to determine the total phosphorus content， total nitrogen content， ammonia nitrogen content， nitrite content， pH value and other related indicators. The ammonium molybdate spectrophotometric method was used to determine the total phosphorus content; the Nessler's reagent spectrophotometric method was used to determine the ammonia nitrogen content; the N-（1-naphthyl）-ethylenediamine photometric method was used to determine the nitrate content; and the pH value was determined with a portable pH meter. After the biological flocs were stabilized， an appropriate amount of water sample was taken and diluted with 0.85% NaCl solution， and then added to a Biolog-ECO plate with an 8-well pipette by 150 μl per well. The water sample was cultured at 28 ℃ in dark place， and the absorbance was read at 590 and 750 nm wavelengths at 12， 24， 48， 60， 72， 84， 96， 108， 120， 132， 144， 156， 168， and 180 h with a microplate reader. Statistical methods
The calculation formula of the average well color development （AWCD）[14] was AWCD =Σ（C-R）/n， wherein C represents the absorbance difference between the two wavelengths （590， 750 nm） of each carbon source well; R represents the absorbance of the control （blank） well; and n is the number of carbon source types on Biolog-ECO （n=31 in this study）. In this study， AWCD data of the water bodies cultured for 72 h was selected to calculate the diversity Shannon index， Simpson index， McIntosh index and richness index， referring to the literature [15] for details. Meanwhile， the 72 h data was selected for principal component analysis （PCA） and analysis of variance （ANOVA） of the utilization of different carbon sources by microorganisms， to determine the differences in the metabolism of microbial communities between different groups of treatments. Statistical analysis was performed through SPSS25.0 statistical software， and data statistics and drawing were completed through Excel2016 and Origin 8.1.
Results and Analysis
Effects of different diet replacement gradients on tilapia culture
As shown in Table 1， after reducing the feed of tilapia， using glucose as an external carbon source for replacing the feed with different proportions， and adding photosynthetic bacteria for enhancement， the final body weight and weight gain showed significant differences between different treatment groups （P<0.05）， and the differences in feed coefficient between different treatment groups were not significant.
Effects of biological flocs on the water quality of aquaculture water
As shown in Table 2， the contents of total phosphorus and ammonia nitrogen in group C were significantly lower than those in the control group （P<0.05）， and the contents of total nitrogen and nitrate were lower than those in the control group， but the differences were not significant. The ammonia nitrogen content of the treatment group B was significantly lower than that of the control group （P<0.05）， and while the total phosphorus， total nitrogen， and nitrate contents were not significantly different from those of the control group， and were 11.20%， 3.41% and 29.65% lower than those of the control group. With the increase of feed replacement gradient， except for pH value， the water quality indicators of different treatment groups showed a decreasing trend; there was a significant difference in total phosphorus content between group B and group C （P<0.05）; and the pH differences were not significant. Effects of biological flocs on the utilization of six types of carbon sources in aquaculture water
The average well color development （AWCD） can effectively reflect the overall activity of microorganisms in the water body， and reflect the ability and preference of microorganisms to utilize 31 carbon sources. It can be seen from Fig. 1 that the utilization intensity of all carbon sources by water microorganisms under different feed replacement gradients increased with the extension of the culture time. The microbial activity in the water within 12 h was low. In the period of 12-120 h， the AWCD varied with the extension of the culture time， it gradually increased. From 120 to 156 h， the utilization of carbon sources by microorganisms showed a trend of first decreasing and then increasing. In the period of 156-168 h， the AWCD gradually increased and was then stabilized. The order of AWCD under different feed replacement gradients was 10% replacement group>control group>20% replacement group， indicating that the 10% replacement group had the highest microbial carbon source metabolism intensity， while the ability of the 20% replacement group to use the carbon source decreased overall.
It can be seen from Table 3 that the Shannon index， Simpson index， McIntosh index and richness index of microorganisms in the water body of group B were higher than those in the control group， but the differences were not significant; and the treatment group C was lower than the control group.
Utilization of different carbon sources by microbial communities in tilapia culture water
The 31 single carbon sources contained in the Biolog-ECO plate are mainly divided into six categories， namely polysaccharides， amino acids， esters， alcohols， amines and acids. As shown in Fig. 2， for different diet replacement gradients， water microorganisms had different preferences for the utilization of carbon sources. The 10% replacement group had the highest utilization intensity in the utilization of amino acids， esters， amines， and acids as carbon sources， while the 20% replacement group showed the strongest utilization of alcohol after 60 h. With the passage of time， the utilization of carbon sources by different treatment groups showed an upward trend. In the utilization of amine carbon sources， the trend was first decreasing， then increasing and finally stable in the period of 120-156 h.
In order to study the effects of biological flocs and photosynthetic bacteria on the carbon metabolism of water microbial communities， 72 h of Biolog-ECO culture was selected as the analysis time point， and SPSS25.0 software was used for PCA analysis， during which three main components were extracted. The first principal component （PC1） was 37.649%， the second principal component （PC2） was 19.163%， and the third principal component （PC3） was 11.090%. The three-dimensional principal component combined 67.902% of the information of all 31 carbon sources. The first two principal components were extracted for mapping （Fig. 3） to characterize the carbon source metabolism characteristics of the microbial communities in each group of water body. The accumulative contribution rate of the first two principal components reached 56.812%. The distance between different treatment groups in principal component analysis indicated the similarity level between different treatment groups. The closer the distance， the higher the degree of similarity， and the more similar the utilization ability of different carbon sources. It can be seen from Fig. 3 that the control group A is mainly distributed in the first and second quadrants， treatment group B is mainly distributed in the third and fourth quadrants， and treatment group C is mainly distributed in the first， second and third quadrants， indicating that there were obvious differences in water microbial carbon metabolism characteristics between treatment group B and the control group， and the difference between treatment group C and the control group was not significant. Table 4 lists the correlation coefficients between the 31 carbon sources and the 3 principal components， reflecting the importance of each variable to the principal components. The higher the correlation of the carbon source， the greater the contribution rate of the carbon source to the differentiation of different treatments. However， a high correlation coefficient could not fully indicate that the actual utilization of the carbon source was also high， and might also be due to the high difference in the utilization of carbon sources between different treatments. Table 4 shows that there were 21 kinds of substrates having a correlation coefficient of 0.55 or more with principal component 1 for the three groups of water microbial metabolism substrates， and 7 kinds of substrates having a correlation coefficient of more than 0.75， including 2 kinds of polysaccharides， 2 kinds of amino acids， and one kind of ester， amine and acid; only 4 substrates had a correlation coefficient of more than 0.55 with principal component 2; and there were only 3 substrates having a correlation coefficient of 0.55 or more with principal component 3. It indicated that the differences in the community structure of the water microbial metabolic function between the different treatment groups and the control group were mainly reflected in the substrates having a higher correlation coefficient with principal component 1. The correlation coefficients with principal component 1 above 0.80 were α-D-lactose， methyl pyruvate， L-threonine and phenethylamine.
Discussion and Conclusions
Effects of photosynthetic bacteria-enhanced biological flocs on the growth of tilapia
Wang et al.[16] reported that genetic improved farmed tilapia had no obvious stress response during the construction of the biological floc system， and its growth was more favorable than the breeding effect of the circulating aquaculture system. Li et al.[17] found in the study of culturing small-size tilapia species by biological floc technology that biological flocs could be used as protein bait to be eaten by tilapia species to improve their growth performance. Liu et al.[18] found in the study of the tilapia culture system without carbon source addition that it was feasible to not add additional carbon source in the water recycling system， while their purpose of adding carbon source was to adjust the ratio of carbon to nitrogen in the water body and promote the formation of biological flocs. In this study， the feed was reduced， and glucose and photosynthetic bacteria were added to adjust the microbial structure of the water body. The results showed that with the increase in the diet replacement gradient， the weight gain of tilapia decreased， which might be due to the effect of adding photosynthetic bacteria. Furthermore， with the increase of the diet replacement gradient， the effect became greater， which might be related to changes in the organic carbon cycle. Crab et al.[19] reported that different types of external carbon sources had different effects on the nutritional composition of biological flocs. Specifically， compared with the group fed with bait only， the differences in the feed coefficient from the groups with different feed replacement gradients were not significant， and the 20% feed replacement group showed a feed coefficient decreased by 3.8%， and was clearly observed to have a growth rate of tilapia slowed down. Xu[20] found that in the biological floc culture system with sucrose as an external carbon source， the biological flocs could increase the activity of lipase， amylase and protease in shrimp stomach. Tang et al.[21] studied the effect of adding Bacillus subtilis and carbon source on the growth of tilapia under the condition of zero water exchange， and found that the addition of B. subtilis and carbon source improved the survival rate， growth rate and feed conversion efficiency of tilapia to a certain extent， without reaching the significant level. In this study， with the increase in the feed replacement gradient， the growth of tilapia showed a significant downward trend， which might be related to the type of external carbon source， or might be because the feed replacement gradient was too high， and the biological flocs could not fully meet the growth needs of tilapia， which restricted the growth of tilapia. Biological flocs play many different roles in the aquaculture ecosystem and play a key role[21-24]. In this study， whether the composition of biological flocs and the addition of photosynthetic bacteria affect the activity of certain digestive enzymes of tilapia， which in turn affects tilapia growth， needs to be verified by further experiments. Effects of photosynthetic bacteria-enhanced biological flocs on the water quality of aquaculture water
Lu et al.[25] studied grass carp using biological floc technology with different C/N levels and found that the C/N ratio of 15 was the best， at which a significant effect was achieved in reducing the levels of ammonia nitrogen and nitrite in the water. Li et al.[26] used wheat starch as a carbon source in aquaculture water to reduce the content of tri-state nitrogen， improve water quality， and generate edible flocs for fish， thereby reducing feed coefficient[26]. Shang[27] proposed in the study on the application of cassava residues in the aquaculture wastewater treatment technology using biological flocs that the addition of sucrose could effectively control ammonia nitrogen pollution. Tang et al.[28] found in the study of the effect of carbon sources on the quality of Anguilla marmorata culture water that the biological floc group was significantly lower than the non-biological floc group in terms of total phosphorus， total nitrogen and tri-state nitrogen. In this study， the total phosphorus and ammonia nitrogen levels of the 20% feed replacement group were significantly lower than those of the control group （P<0.05）; and the total phosphorus level of the 10% feed replacement group was reduced by 11.20%， but the difference was not significant， and the ammonia nitrogen level was significantly lower than the control group （P<0.05）. The nitrate contents of the 20% feed replacement group and the 10% feed replacement group were reduced by 44.25% and 29.65% on average， respectively， which was the same as the trend of the above research results. Both autotrophic nitrification and heterotrophic nitrification have good degradation effects on inorganic nitrogen. Wang et al.[29-30] found that heterotrophic nitrification in biological flocs is beneficial to the removal of inorganic nitrogen in aquaculture water and reduces the eutrophication of the aquaculture water， and the function of adding carbon source to cultivate biological flocs for the rapid heterotrophic transformation of ammonia nitrogen is of great significance in aquaculture. In tilapia culture waters， using glucose as an additional carbon source and indigenous microorganisms as a bacterial source to construct a culture system while using photosynthetic bacteria for enhancement can further regulate water quality and reduce the levels of total nitrogen， total phosphorus and ammonia nitrogen in the water， and simultaneously enhance the utilization rate of the carbon sources of polysaccharides， esters and amino acids by water microorganisms and reduce the utilization rate of the carbon sources of amines and acids. Effects of photosynthetic bacteria-enhanced biological flocs on the diversity of microbial communities in aquaculture water
The AWCD value in the Biolog-ECO plate reflects the overall ability and preference of the microbial community to use different single carbon sources. Under the same type of carbon source， the comparison of the utilization of carbon sources by different microorganisms can reflect that there are certain differences in the utilization of different types of carbon sources by water microorganisms enhanced by photosynthetic bacteria under different feed replacement gradients[23]. The average AWCD value of the feed that was replaced by 10% was the highest， followed by the control group， and the feed that was replaced by 20% was the lowest. Therefore， it can be seen that appropriately reducing a certain amount of feed is beneficial to enhancing the utilization of carbon sources by the water microbial communities. The use of different diversity indices can reflect the changes in the functional diversity of different microbial communities[31]. In this study， with the increase of the feed replacement gradient， there were no significant changes in water microbial diversity and abundance， indicating that the addition of carbon source and photosynthetic bacteria had no significant impact on microorganisms in the water environment. It is in contrast to the result of Li et al. [32] that biological flocs can increase the Shannon index and richness index of water microbes [32]， and is consistent with the results of the study by Shi et al. [33] on the effect of combined fillers on the microbial diversity of tilapia culture environments， which might be related to the attachment substance of microorganisms and the addition of photosynthetic bacteria. In recent years， with the development of microecological preparations， their advantages of improving the breeding environment， maintaining the microecological balance of aquaculture water and promoting animal growth have been continuously expanded in aquaculture.
The Biolog-ECO method reflects the functional diversity of microbial communities through the utilization of different types of carbon sources by microorganisms. Its principal component analysis can effectively reflect the functional structure characteristics of microbial communities[34]. The 31 carbon sources on the plate are mainly divided into six categories： polysaccharides， amino acids， esters， alcohols， amines， and acids. The analysis results of PC1， PC2， and PC3 showed that the main types of carbon sources that played a role in the metabolic function of the water microbial communities under different feed replacement gradients were polysaccharides， esters and amino acids. The differences under different treatments were mainly reflected in the utilization of amino acids and ester carbon sources， most prominent in amino acids. Combining the overall utilization of different carbon sources by different treatment groups， it could be concluded that there were obvious differences in the utilization characteristics of carbon sources by microorganisms in different treatments， which might have a certain relationship with the addition of glucose as a carbon source in the water or the addition of photosynthetic bacteria. With the increase of the feed replacement rate， the microbial diversity in the water body changed. Therefore， the composition and content of the residual feed and feces as the main input carbon source underwent significant changes， which in turn affected the metabolism of different types of carbon sources by microorganisms in the water body， and ultimately affected the type and intensity of carbon source utilization. Studies have shown that the extracellular polymers produced by different carbon sources are quite different[35]， and algae in water bodies can produce a large amount of esters and hydrocarbons under conditions of sufficient light and carbon dioxide[36]， which further influence and participate in the composition of biological flocs and the ways in which microorganisms benefit from different carbon sources[37]. The research by Zhang et al.[38] found that the structure of microbial communities in the system would change with different carbon sources. Ballester et al.[39] reported that microorganisms in biological flocs played an important role in maintaining water quality and providing essential nutrients. In this study， the microbes in tilapia culture waters under different feed replacement rates were distributed in different areas， indicating that there were certain differences in the characteristics of carbon source metabolism. The principal component analysis method can well reflect the changes in the comprehensive state of water quality， and then extract more useful information[40]. Due to the limitations of the Biolog-ECO micro-plate technology， it is difficult to completely reflect the changing laws of the functional diversity of the microbial communities in different treatments using this technology alone. The research on the types of microbes in biological flocs requires further exploration. Appropriate reduction of feed input could enhance the carbon source metabolism capacity of water microorganisms， and reducing feed by 10% could not only save culture costs， but also enhance the overall utilization of different carbon sources by the culture water microorganisms. This study was completed under the experimental conditions， and there is still a certain gap with the actual production of pond culture. It is also necessary to consider the specific feed replacement ratio as appropriate， and meanwhile， attention should be paid to the selection of bacteria in the process of choosing probiotics for enhancement. References
[1] DEY MM， GUPTA MV. Socioeconomics of disseminating genetically improved Niletilapiain Asia： an introduction[J]. Aquaculture Economics&Management， 2000， 4（1/2）： 5-11.
[2] FIERER N， BRADFORD MA， JACKSON RB. Toward an ecological classification of soil bacteria[J]. Ecology， 2007， 88（6）： 1354-1364.
[3] KONOPKA A， OLIVER L， TURCO RF. The use of carbon substrate utilization patterns in environmental and ecological microbiology[J].Microbial Ecology， 1998， 35（2）： 103-115.
[4] ZHENG YT， HU KH， GAO SF， et al. Separate and screen purple nonsulfur photosynthetic bacteria to efficiently purify aquiculture water quality[J]. Journal of Fujian Agriculture and Forestry University： Natural Science Edition， 1998， 27（3）： 257-260. （in Chinese）
[5] GARLAND JL， MILLS AL. Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization[J]. Applied and Environmental Microbiology， 1991， 57（8）： 2351-2359.
[6] GARLAND JL. Analysisand interpretation of community-level physiological profiles in microbial ecology[J]. FEMS Microbiology Ecology， 1997， 24（4）： 289-300.
[7] SHEN JY， YIN WL， LIU W， et al. Improvement of fisheries water quality and fish growth of photosynthetic bacteria HZPSB[J]. Bulletin of Science and Technology， 2004， 20（6）： 481-484. （in Chinese）
[8] VRATI S. Single cell protein production by photosynthetic bacteria grown on clarified effluents of biogas plants[J]. Applied Microbiology and Biotechnology， 1984， 19（3）： 199-202.
[9] YANG ZW， ZHANG Z， GE H， et al. The effects of different carbon source on larval rearing of Litopenaeus vannamei by biofloc technology[J]. Journal of Fujian Fisheries， 2015， 37（5）： 347-352. （in Chinese）
[10] LIU KM， YOU HZ， MA L， et al. Effects of dietary fermented soybean meal on the growth and apparent digestibility of the spotted halibut Verasper variegatus[J]. Hebei Yuye， 2019（4）： 28-30. （in Chinese）
[11] WANG CH. Study on glucose addition on water environmental control for the zero-water exchange shrimp culture system[D]. Shanghai： Shanghai Ocean University， 2016. （in Chinese）
[12] WANG J， MA GN， DENG YG， et al. Effects of glucose and salinity on the growth of artemia， the formation of bioflocs and microbial diversity in culture system[J]. Oceanologia Et Limnologia Sinica， 2015， 46（2）： 372-380. （in Chinese）
[13] CHEN JZ， HU GD， WU W， et al. Studies on the implications of the beneficial bacteria in Eriocheir sinensis farming[J]. Journal of Shanghai Fisheries University， 2003， 12（3）： 271-273. （in Chinese） [14] YAN FJ， TIAN XL， DONG SL， et al. Seasonal variation of functional diversity of aquatic microbial community in Apostichopus japonicus culture pond[J]. Chinese Journal of Applied Ecology， 2014， 25（5）： 1499-1505. （in Chinese）
[15] LI ZF， WANG GJ， WANG GJ， XIE J， et al. Microbial carbon metabolic characteristics of biofilm communities in the grass carp culture pond based on Biolog-ECO plates[J]. Journal of Fisheries of China， 2014， 38（12）： 1985-1995. （in Chinese）
[16] WANG CH， GAO Q， TAN HX， et al. Effects of initial establishment of a biofloc technology system on the growth and immune enzyme activity of GIFT tilapia （Oreochromis niloticus）[J]. Journal of Fishery Sciences of China， 2015， 22（4）： 707-715. （in Chinese）
[17] LI CB， LI ZF， HAN LQ， et al. The effect of biofloc technology on indoor cultivation of small tilapia species[J]. Journal of Aquaculture， 2015， 36（7）： 29-35. （in Chinese）
[18] LIU WC， LUO GZ， TAN HX. The effect of no carbon source addition on water quality， growth performance and microbial community in a tilapia biofloc culture system that is reused in water[C]∥Abstracts of the 2017 Annual Conference of the Chinese Fisheries Society. Beijing： Chinese Fisheries Society， 2017. （in Chinese）
[19] CRAB R， CHIELENS B， MATHIEU W， et al. The effect of different carbon sources on the nutritional value of bioflocs， a feed for Macrobrachium rosenbergii post larvae[J].Aquaculture Research， 2010， 41（5）： 559-567.
[20] XU WJ. Research and application of functional effects of biological flocs in shrimp zero-water exchange culture system[D]. Qingdao： Ocean University of China， 2014. （in Chinese）
[21] TANG HJ， ZHANG XH， KONG CM， et al. Effect of Bacillus subtilis and molasses on water quality and growth of tilapia under zero-water exchange[J]. Journal of Huazhong Agricultural University， 2018， 37（3）： 82-86. （in Chinese）
[22] MCINTOSH D， SAMOCHA TM， JONES ER， et al. The effect of a commercial bacterial supplement on the high-density culturing of Litopenaeus vannamei with a low-protein diet in an outdoor tank system and no water exchange[J]. Aquacultural Engineering， 2000， 21（3）： 215-227.
[23] CHOI KH， DOBBS FC. Comparison of two kinds of Biolog microplates （GN and ECO）in the irability to distinguish among aquatic microbial communities[J]. Journal of Microbiological Methods， 1999， 36（3）： 203-213.
[24] MARTNEZ-CRDOVA LR， MARTNEZ-PORCHAS M， EMERENCIANO MGC， et al. From microbes to fish the next revolution in food production[J].Critical Reviews in Biotechnology， 2017， 37（3）： 287-295. [25] LU BG， WANG HY， XIE J， et al. The effect of different C/N levels on the formation of grass carp pond biofloc and its water quality[J]. Journal of Fisheries of China， 2013， 37（8）： 1220-1228. （in Chinese）
[26] LI Y， LIU LP， ZHAO GX， et al. The effect of adding carbon source in the aquaculture water on water quality and the growth of tilapia[J]. Journal of Dalian Fisheries University， 2013， 28（1）： 55-60. （in Chinese）
[27] SHANG Q. Study on the application of cassava residue in the treatment of aquaculture wastewater by biological flocs[D]. Nanning： Guangxi University， 2017. （in Chinese）
[28] TANG XF， LIU LP， SHUAI D， et al. The effect of carbon source on water quality and production performance of Anguilla marmorata culture system[J]. Journal of Dalian Fisheries University， 2019， 34（1）： 70-79. （in Chinese）
[29] WANG T， LIU QS， DUAN YF， et al. Study on the autotrophic and heterotrophic nitrification performance of low C/N domesticated biological flocs[J]. Marine Fishery， 2018， 40（5）： 614-624. （in Chinese）
[30] SCHNEIDER O， SERETI V， EDING EH， et al. Analysis of nutrient flows in integrated intensive aquaculture systems[J]. Aquacultural Engineering， 2005， 32（3/4）： 379-401.
[31] GARLANDJL. Analytical approaches to the characterization of samples of microbial communities using patterns of potential C source utilization[J].Soil Biology & Biochemistry， 1996， 28（2）： 213-221.
[32] LI ZF， WANG GJ， YU DG， et al. Effect of bioflocs on water quality and metabolic functions of microbial community in zero-water exchange tanks[J]. Journal of Shanghai Ocean University， 2015， 24（4）： 503-512. （in Chinese）
[33] SHI LL， FAN LM， CHEN JC， et al. Effects of combined fillers on the water quality， growth of tilapia， and functional diversity of the microbial community in tilapia aquaculture water [J]. Journal of Agro-Environment Science， 2017， 36（8）： 1618-1626. （in Chinese）
[34] LI J. Soil microbial characteristics and seasonal changes in long-term different fertilization systems[D]. Beijing： Chinese Academy of Agricultural Sciences， 2008. （in Chinese）
[35] WILN BM， ONUKI M， HERMANSSON M， et al. Microbial community structure in activated sludge floc analysed by fluorescence in situ hybridization and its relation to floc stability[J]. Water Research， 2008， 42（8/9）： 2300-2308.
[36] VERMA NM， MEHROTRA S， SHUKLA A， et al. Prospective of biodiesel production utilizing microalgae as the cell factories： A comprehensive discussion[J].AfricanJournal of Biotechnology， 2010， 9（10）： 1402-1411. [37] BODK I， BLSTKOV A， SEDLCEK S， et al. Biodiesel waste as source of organic carbon for municipal WWTP denitrification[J].Bioresource Technology， 2009， 100（8）： 2452-2456.
[38] ZHANG Z， YANG ZW， GE H， et al. The effect of different carbon sources on the structure， nutrient composition， bacterial community and water quality of the biological flocs in the water body for raising standard-diameter Litopenaeus vannamei[J]. Journal of Fisheries of China， 2019， 43（3）： 639-649. （in Chinese）
[39] BALLESTER E， ABREU PC， CAVALLI RO， et al. Effect of practicaldiets with different protein levels on the performance of Farfantepenaeus paulensisjuveniles nursed in a zero exchange suspended microbial flocs intensive system[J]. Aquaculture Nutrition， 2010， 16（2）： 163-172.
[40] FAN LM， WU W， HU GD， et al. Effect of restoration of intensive ponds by multilevel bio-system with the method of principal component analysis[J]. Chinese Agricultural Science Bulletin， 2010， 26（23）： 392-396. （in Chinese）
Abstract With Oreochromis niloticus as the object of study， glucose was added as a carbon source to promote the formation of the biological flocs for replacing part of the feed， and three gradients were set up， namely Group A （all feed）， Group B （replacement of 10% feed） and Group C （replacement of 20% feed）， so as to explore the effects of photosynthetic bacteria-enhanced biological flocs on tilapia growth and water environment conditions. Meanwhile， the Biolog-ECO technology was applied to study the changes of microbial carbon metabolism diversity in aquaculture water. The results showed that the utilization of microbial carbon sources under different feed replacement gradients increased with the extension of the culture time. The overall performance was in order of 10% replacement>all feed>20% feed replacement. A suitable replacement rate could not only enhance the overall utilization of carbon sources by water microorganisms， but also save culture costs. The principal component analysis showed that the carbon source metabolism of the water microbial communities under different feed replacement gradients was significantly different. Specifically， polysaccharides， esters and amino acids were the preferred carbon sources of water microbes， while the utilization of amines and acids was low.
Key words Photosynthetic bacteria; Biological floc; Bait substitute; Microbial community; Tilapia
Tilapia （Oreochromis niloticus） belongs to Oreochromis in Cichlidae of Perciformes in Perciformes， which is native to inland waters of Africa， and is also called "African crucian carp". Most tilapia is omnivorous fish， and they often feed on plants and debris in water bodies. Genetic improved farmed tilapia is a new tilapia strain selected in the Philippines by research institutions such as the World Fish Center at the end of the 20th century. It has strong adaptability， stable genetic traits， short breeding cycle， and fast growth， and is currently one of the leading species in the tilapia aquaculture industry[1]， as well as one of the fine species promoted by the United Nations Food and Agriculture Organization to the world. With the development of the intensive and high-yield model of aquaculture， the residual feed and fecal accumulation caused by a large amount of feed input and the harmful substances produced thereby will not only cause certain harm to the aquaculture object itself， but also affect the aquaculture water environment. With the development of microbial technology， the research of microbes in aquaculture is also deepening. There are many kinds of microorganisms， widely distributed， sensitive to the environment， easy to mutate， and irreplaceable in biological systems[2-3]. Photosynthetic bacteria can decompose and utilize organic matter in aquaculture water bodies， and have the effects of purifying water quality and improving the micro-ecological environment of water bodies[4]. The Biolog-ECO plate method directly cultures microbial populations on microplates， measures the degree of utilization of different carbon sources by microorganisms， and monitors the changes in light absorbance caused by them in real time， so as to explore the physiological characteristics of microorganisms and their community structure[5-6]. In recent years， there have been many reports on the application of photosynthetic bacteria in aquaculture[7-8]. Studies have shown that using substances including glucose， honey and sucrose as an external carbon source can well form biological flocs [9-12]. Due to the low cost of glucose and its ability to form stable and efficient biological flocs， in this study， we used glucose as an additional organic carbon source and simultaneously added photosynthetic bacteria to the aquaculture water body for enhancement， aiming to explore the water quality control issues in aquaculture and the microbial community structure and carbon source utilization situation in the formation of biological flocs and provide reference data for the study of feed replacement by biological flocs and microbial community structure under pond culture.
Materials and Methods
Culture management
The aquaculture experiment was carried out in the Freshwater Fisheries Research Center of the Chinese Academy of Fishery Sciences， Wuxi City， Jiangsu Province from July to August 2018. The experiment lasted for 60 d. Nine round culture buckets with a volume of about 1 000 L were selected. In the early stage of the experiment， fresh surface soil was added to the culture buckets to provide a source of indigenous microorganisms， and then a string of composite filler composed of fully plastic clips and aldolized vinylon yarn was hung in each bucket to provide an attachment substance for microorganisms and promote the formation of biological flocs. Finally， about 700 L of water was added to each bucket. The culture system was then sun for full exposure. In order to ensure sufficient oxygen supply during the culture experiment， a microporous aerator tube was installed at the bottom of each culture bucket to increase the dissolved oxygen concentration of the water body while making the water body fully agitated and mixed. Twenty individuals of the same size were raised in each culture bucket， with a body weight of （6.50±0.50） g. Feeding was carried out at the same frequency during the culture process， and the initial feeding amount was about 10% of the tilapia body weight. As the water environment and weather changed， the feeding amount was adjusted to 2%-3% of the tilapia body weight. The experiment was set with 3 different feed replacement gradients， namely Group A （all feed）， Group B （replacement of 10% feed） and Group C （replacement of 20% feed）， each of which was set with 3 parallels. After the ammonia level in the aquaculture water met the test requirements， we started to add the carbon source and photosynthetic bacteria liquid.
Experimental methods
According to the freshwater pond biological floc ecological cultivation technology construction operating procedures introduced by the Freshwater Fisheries Research Center， Chinese Academy of Fishery Sciences， the carbon source addition amount was determined according to A（g）=H×S×（30×CTAN-N-38）， wherein H is the average water depth of the pond， m; S is the pond area， m2; and CTAN-N is the initial ammonia nitrogen concentration of the pond， mg/L. Glucose was added once every 2 d， and the time of addition was 30 min after feeding in the morning. The carbon source was dissolved in the culture water first， and then evenly sprinkled in the culture buckets. The photosynthetic bacteria broth was cultured in the laboratory and its concentration was 109 CFU/ml. According to related study by Chen et al.[13]， it was determined that the photosynthetic bacteria broth was used once every 1 week， with 100 ml each time， 6 times in total.
After the formation of biological flocs was stabilized， water samples were collected from 30 cm below the water surface every 1 week to determine the total phosphorus content， total nitrogen content， ammonia nitrogen content， nitrite content， pH value and other related indicators. The ammonium molybdate spectrophotometric method was used to determine the total phosphorus content; the Nessler's reagent spectrophotometric method was used to determine the ammonia nitrogen content; the N-（1-naphthyl）-ethylenediamine photometric method was used to determine the nitrate content; and the pH value was determined with a portable pH meter. After the biological flocs were stabilized， an appropriate amount of water sample was taken and diluted with 0.85% NaCl solution， and then added to a Biolog-ECO plate with an 8-well pipette by 150 μl per well. The water sample was cultured at 28 ℃ in dark place， and the absorbance was read at 590 and 750 nm wavelengths at 12， 24， 48， 60， 72， 84， 96， 108， 120， 132， 144， 156， 168， and 180 h with a microplate reader. Statistical methods
The calculation formula of the average well color development （AWCD）[14] was AWCD =Σ（C-R）/n， wherein C represents the absorbance difference between the two wavelengths （590， 750 nm） of each carbon source well; R represents the absorbance of the control （blank） well; and n is the number of carbon source types on Biolog-ECO （n=31 in this study）. In this study， AWCD data of the water bodies cultured for 72 h was selected to calculate the diversity Shannon index， Simpson index， McIntosh index and richness index， referring to the literature [15] for details. Meanwhile， the 72 h data was selected for principal component analysis （PCA） and analysis of variance （ANOVA） of the utilization of different carbon sources by microorganisms， to determine the differences in the metabolism of microbial communities between different groups of treatments. Statistical analysis was performed through SPSS25.0 statistical software， and data statistics and drawing were completed through Excel2016 and Origin 8.1.
Results and Analysis
Effects of different diet replacement gradients on tilapia culture
As shown in Table 1， after reducing the feed of tilapia， using glucose as an external carbon source for replacing the feed with different proportions， and adding photosynthetic bacteria for enhancement， the final body weight and weight gain showed significant differences between different treatment groups （P<0.05）， and the differences in feed coefficient between different treatment groups were not significant.
Effects of biological flocs on the water quality of aquaculture water
As shown in Table 2， the contents of total phosphorus and ammonia nitrogen in group C were significantly lower than those in the control group （P<0.05）， and the contents of total nitrogen and nitrate were lower than those in the control group， but the differences were not significant. The ammonia nitrogen content of the treatment group B was significantly lower than that of the control group （P<0.05）， and while the total phosphorus， total nitrogen， and nitrate contents were not significantly different from those of the control group， and were 11.20%， 3.41% and 29.65% lower than those of the control group. With the increase of feed replacement gradient， except for pH value， the water quality indicators of different treatment groups showed a decreasing trend; there was a significant difference in total phosphorus content between group B and group C （P<0.05）; and the pH differences were not significant. Effects of biological flocs on the utilization of six types of carbon sources in aquaculture water
The average well color development （AWCD） can effectively reflect the overall activity of microorganisms in the water body， and reflect the ability and preference of microorganisms to utilize 31 carbon sources. It can be seen from Fig. 1 that the utilization intensity of all carbon sources by water microorganisms under different feed replacement gradients increased with the extension of the culture time. The microbial activity in the water within 12 h was low. In the period of 12-120 h， the AWCD varied with the extension of the culture time， it gradually increased. From 120 to 156 h， the utilization of carbon sources by microorganisms showed a trend of first decreasing and then increasing. In the period of 156-168 h， the AWCD gradually increased and was then stabilized. The order of AWCD under different feed replacement gradients was 10% replacement group>control group>20% replacement group， indicating that the 10% replacement group had the highest microbial carbon source metabolism intensity， while the ability of the 20% replacement group to use the carbon source decreased overall.
It can be seen from Table 3 that the Shannon index， Simpson index， McIntosh index and richness index of microorganisms in the water body of group B were higher than those in the control group， but the differences were not significant; and the treatment group C was lower than the control group.
Utilization of different carbon sources by microbial communities in tilapia culture water
The 31 single carbon sources contained in the Biolog-ECO plate are mainly divided into six categories， namely polysaccharides， amino acids， esters， alcohols， amines and acids. As shown in Fig. 2， for different diet replacement gradients， water microorganisms had different preferences for the utilization of carbon sources. The 10% replacement group had the highest utilization intensity in the utilization of amino acids， esters， amines， and acids as carbon sources， while the 20% replacement group showed the strongest utilization of alcohol after 60 h. With the passage of time， the utilization of carbon sources by different treatment groups showed an upward trend. In the utilization of amine carbon sources， the trend was first decreasing， then increasing and finally stable in the period of 120-156 h.
In order to study the effects of biological flocs and photosynthetic bacteria on the carbon metabolism of water microbial communities， 72 h of Biolog-ECO culture was selected as the analysis time point， and SPSS25.0 software was used for PCA analysis， during which three main components were extracted. The first principal component （PC1） was 37.649%， the second principal component （PC2） was 19.163%， and the third principal component （PC3） was 11.090%. The three-dimensional principal component combined 67.902% of the information of all 31 carbon sources. The first two principal components were extracted for mapping （Fig. 3） to characterize the carbon source metabolism characteristics of the microbial communities in each group of water body. The accumulative contribution rate of the first two principal components reached 56.812%. The distance between different treatment groups in principal component analysis indicated the similarity level between different treatment groups. The closer the distance， the higher the degree of similarity， and the more similar the utilization ability of different carbon sources. It can be seen from Fig. 3 that the control group A is mainly distributed in the first and second quadrants， treatment group B is mainly distributed in the third and fourth quadrants， and treatment group C is mainly distributed in the first， second and third quadrants， indicating that there were obvious differences in water microbial carbon metabolism characteristics between treatment group B and the control group， and the difference between treatment group C and the control group was not significant. Table 4 lists the correlation coefficients between the 31 carbon sources and the 3 principal components， reflecting the importance of each variable to the principal components. The higher the correlation of the carbon source， the greater the contribution rate of the carbon source to the differentiation of different treatments. However， a high correlation coefficient could not fully indicate that the actual utilization of the carbon source was also high， and might also be due to the high difference in the utilization of carbon sources between different treatments. Table 4 shows that there were 21 kinds of substrates having a correlation coefficient of 0.55 or more with principal component 1 for the three groups of water microbial metabolism substrates， and 7 kinds of substrates having a correlation coefficient of more than 0.75， including 2 kinds of polysaccharides， 2 kinds of amino acids， and one kind of ester， amine and acid; only 4 substrates had a correlation coefficient of more than 0.55 with principal component 2; and there were only 3 substrates having a correlation coefficient of 0.55 or more with principal component 3. It indicated that the differences in the community structure of the water microbial metabolic function between the different treatment groups and the control group were mainly reflected in the substrates having a higher correlation coefficient with principal component 1. The correlation coefficients with principal component 1 above 0.80 were α-D-lactose， methyl pyruvate， L-threonine and phenethylamine.
Discussion and Conclusions
Effects of photosynthetic bacteria-enhanced biological flocs on the growth of tilapia
Wang et al.[16] reported that genetic improved farmed tilapia had no obvious stress response during the construction of the biological floc system， and its growth was more favorable than the breeding effect of the circulating aquaculture system. Li et al.[17] found in the study of culturing small-size tilapia species by biological floc technology that biological flocs could be used as protein bait to be eaten by tilapia species to improve their growth performance. Liu et al.[18] found in the study of the tilapia culture system without carbon source addition that it was feasible to not add additional carbon source in the water recycling system， while their purpose of adding carbon source was to adjust the ratio of carbon to nitrogen in the water body and promote the formation of biological flocs. In this study， the feed was reduced， and glucose and photosynthetic bacteria were added to adjust the microbial structure of the water body. The results showed that with the increase in the diet replacement gradient， the weight gain of tilapia decreased， which might be due to the effect of adding photosynthetic bacteria. Furthermore， with the increase of the diet replacement gradient， the effect became greater， which might be related to changes in the organic carbon cycle. Crab et al.[19] reported that different types of external carbon sources had different effects on the nutritional composition of biological flocs. Specifically， compared with the group fed with bait only， the differences in the feed coefficient from the groups with different feed replacement gradients were not significant， and the 20% feed replacement group showed a feed coefficient decreased by 3.8%， and was clearly observed to have a growth rate of tilapia slowed down. Xu[20] found that in the biological floc culture system with sucrose as an external carbon source， the biological flocs could increase the activity of lipase， amylase and protease in shrimp stomach. Tang et al.[21] studied the effect of adding Bacillus subtilis and carbon source on the growth of tilapia under the condition of zero water exchange， and found that the addition of B. subtilis and carbon source improved the survival rate， growth rate and feed conversion efficiency of tilapia to a certain extent， without reaching the significant level. In this study， with the increase in the feed replacement gradient， the growth of tilapia showed a significant downward trend， which might be related to the type of external carbon source， or might be because the feed replacement gradient was too high， and the biological flocs could not fully meet the growth needs of tilapia， which restricted the growth of tilapia. Biological flocs play many different roles in the aquaculture ecosystem and play a key role[21-24]. In this study， whether the composition of biological flocs and the addition of photosynthetic bacteria affect the activity of certain digestive enzymes of tilapia， which in turn affects tilapia growth， needs to be verified by further experiments. Effects of photosynthetic bacteria-enhanced biological flocs on the water quality of aquaculture water
Lu et al.[25] studied grass carp using biological floc technology with different C/N levels and found that the C/N ratio of 15 was the best， at which a significant effect was achieved in reducing the levels of ammonia nitrogen and nitrite in the water. Li et al.[26] used wheat starch as a carbon source in aquaculture water to reduce the content of tri-state nitrogen， improve water quality， and generate edible flocs for fish， thereby reducing feed coefficient[26]. Shang[27] proposed in the study on the application of cassava residues in the aquaculture wastewater treatment technology using biological flocs that the addition of sucrose could effectively control ammonia nitrogen pollution. Tang et al.[28] found in the study of the effect of carbon sources on the quality of Anguilla marmorata culture water that the biological floc group was significantly lower than the non-biological floc group in terms of total phosphorus， total nitrogen and tri-state nitrogen. In this study， the total phosphorus and ammonia nitrogen levels of the 20% feed replacement group were significantly lower than those of the control group （P<0.05）; and the total phosphorus level of the 10% feed replacement group was reduced by 11.20%， but the difference was not significant， and the ammonia nitrogen level was significantly lower than the control group （P<0.05）. The nitrate contents of the 20% feed replacement group and the 10% feed replacement group were reduced by 44.25% and 29.65% on average， respectively， which was the same as the trend of the above research results. Both autotrophic nitrification and heterotrophic nitrification have good degradation effects on inorganic nitrogen. Wang et al.[29-30] found that heterotrophic nitrification in biological flocs is beneficial to the removal of inorganic nitrogen in aquaculture water and reduces the eutrophication of the aquaculture water， and the function of adding carbon source to cultivate biological flocs for the rapid heterotrophic transformation of ammonia nitrogen is of great significance in aquaculture. In tilapia culture waters， using glucose as an additional carbon source and indigenous microorganisms as a bacterial source to construct a culture system while using photosynthetic bacteria for enhancement can further regulate water quality and reduce the levels of total nitrogen， total phosphorus and ammonia nitrogen in the water， and simultaneously enhance the utilization rate of the carbon sources of polysaccharides， esters and amino acids by water microorganisms and reduce the utilization rate of the carbon sources of amines and acids. Effects of photosynthetic bacteria-enhanced biological flocs on the diversity of microbial communities in aquaculture water
The AWCD value in the Biolog-ECO plate reflects the overall ability and preference of the microbial community to use different single carbon sources. Under the same type of carbon source， the comparison of the utilization of carbon sources by different microorganisms can reflect that there are certain differences in the utilization of different types of carbon sources by water microorganisms enhanced by photosynthetic bacteria under different feed replacement gradients[23]. The average AWCD value of the feed that was replaced by 10% was the highest， followed by the control group， and the feed that was replaced by 20% was the lowest. Therefore， it can be seen that appropriately reducing a certain amount of feed is beneficial to enhancing the utilization of carbon sources by the water microbial communities. The use of different diversity indices can reflect the changes in the functional diversity of different microbial communities[31]. In this study， with the increase of the feed replacement gradient， there were no significant changes in water microbial diversity and abundance， indicating that the addition of carbon source and photosynthetic bacteria had no significant impact on microorganisms in the water environment. It is in contrast to the result of Li et al. [32] that biological flocs can increase the Shannon index and richness index of water microbes [32]， and is consistent with the results of the study by Shi et al. [33] on the effect of combined fillers on the microbial diversity of tilapia culture environments， which might be related to the attachment substance of microorganisms and the addition of photosynthetic bacteria. In recent years， with the development of microecological preparations， their advantages of improving the breeding environment， maintaining the microecological balance of aquaculture water and promoting animal growth have been continuously expanded in aquaculture.
The Biolog-ECO method reflects the functional diversity of microbial communities through the utilization of different types of carbon sources by microorganisms. Its principal component analysis can effectively reflect the functional structure characteristics of microbial communities[34]. The 31 carbon sources on the plate are mainly divided into six categories： polysaccharides， amino acids， esters， alcohols， amines， and acids. The analysis results of PC1， PC2， and PC3 showed that the main types of carbon sources that played a role in the metabolic function of the water microbial communities under different feed replacement gradients were polysaccharides， esters and amino acids. The differences under different treatments were mainly reflected in the utilization of amino acids and ester carbon sources， most prominent in amino acids. Combining the overall utilization of different carbon sources by different treatment groups， it could be concluded that there were obvious differences in the utilization characteristics of carbon sources by microorganisms in different treatments， which might have a certain relationship with the addition of glucose as a carbon source in the water or the addition of photosynthetic bacteria. With the increase of the feed replacement rate， the microbial diversity in the water body changed. Therefore， the composition and content of the residual feed and feces as the main input carbon source underwent significant changes， which in turn affected the metabolism of different types of carbon sources by microorganisms in the water body， and ultimately affected the type and intensity of carbon source utilization. Studies have shown that the extracellular polymers produced by different carbon sources are quite different[35]， and algae in water bodies can produce a large amount of esters and hydrocarbons under conditions of sufficient light and carbon dioxide[36]， which further influence and participate in the composition of biological flocs and the ways in which microorganisms benefit from different carbon sources[37]. The research by Zhang et al.[38] found that the structure of microbial communities in the system would change with different carbon sources. Ballester et al.[39] reported that microorganisms in biological flocs played an important role in maintaining water quality and providing essential nutrients. In this study， the microbes in tilapia culture waters under different feed replacement rates were distributed in different areas， indicating that there were certain differences in the characteristics of carbon source metabolism. The principal component analysis method can well reflect the changes in the comprehensive state of water quality， and then extract more useful information[40]. Due to the limitations of the Biolog-ECO micro-plate technology， it is difficult to completely reflect the changing laws of the functional diversity of the microbial communities in different treatments using this technology alone. The research on the types of microbes in biological flocs requires further exploration. Appropriate reduction of feed input could enhance the carbon source metabolism capacity of water microorganisms， and reducing feed by 10% could not only save culture costs， but also enhance the overall utilization of different carbon sources by the culture water microorganisms. This study was completed under the experimental conditions， and there is still a certain gap with the actual production of pond culture. It is also necessary to consider the specific feed replacement ratio as appropriate， and meanwhile， attention should be paid to the selection of bacteria in the process of choosing probiotics for enhancement. References
[1] DEY MM， GUPTA MV. Socioeconomics of disseminating genetically improved Niletilapiain Asia： an introduction[J]. Aquaculture Economics&Management， 2000， 4（1/2）： 5-11.
[2] FIERER N， BRADFORD MA， JACKSON RB. Toward an ecological classification of soil bacteria[J]. Ecology， 2007， 88（6）： 1354-1364.
[3] KONOPKA A， OLIVER L， TURCO RF. The use of carbon substrate utilization patterns in environmental and ecological microbiology[J].Microbial Ecology， 1998， 35（2）： 103-115.
[4] ZHENG YT， HU KH， GAO SF， et al. Separate and screen purple nonsulfur photosynthetic bacteria to efficiently purify aquiculture water quality[J]. Journal of Fujian Agriculture and Forestry University： Natural Science Edition， 1998， 27（3）： 257-260. （in Chinese）
[5] GARLAND JL， MILLS AL. Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization[J]. Applied and Environmental Microbiology， 1991， 57（8）： 2351-2359.
[6] GARLAND JL. Analysisand interpretation of community-level physiological profiles in microbial ecology[J]. FEMS Microbiology Ecology， 1997， 24（4）： 289-300.
[7] SHEN JY， YIN WL， LIU W， et al. Improvement of fisheries water quality and fish growth of photosynthetic bacteria HZPSB[J]. Bulletin of Science and Technology， 2004， 20（6）： 481-484. （in Chinese）
[8] VRATI S. Single cell protein production by photosynthetic bacteria grown on clarified effluents of biogas plants[J]. Applied Microbiology and Biotechnology， 1984， 19（3）： 199-202.
[9] YANG ZW， ZHANG Z， GE H， et al. The effects of different carbon source on larval rearing of Litopenaeus vannamei by biofloc technology[J]. Journal of Fujian Fisheries， 2015， 37（5）： 347-352. （in Chinese）
[10] LIU KM， YOU HZ， MA L， et al. Effects of dietary fermented soybean meal on the growth and apparent digestibility of the spotted halibut Verasper variegatus[J]. Hebei Yuye， 2019（4）： 28-30. （in Chinese）
[11] WANG CH. Study on glucose addition on water environmental control for the zero-water exchange shrimp culture system[D]. Shanghai： Shanghai Ocean University， 2016. （in Chinese）
[12] WANG J， MA GN， DENG YG， et al. Effects of glucose and salinity on the growth of artemia， the formation of bioflocs and microbial diversity in culture system[J]. Oceanologia Et Limnologia Sinica， 2015， 46（2）： 372-380. （in Chinese）
[13] CHEN JZ， HU GD， WU W， et al. Studies on the implications of the beneficial bacteria in Eriocheir sinensis farming[J]. Journal of Shanghai Fisheries University， 2003， 12（3）： 271-273. （in Chinese） [14] YAN FJ， TIAN XL， DONG SL， et al. Seasonal variation of functional diversity of aquatic microbial community in Apostichopus japonicus culture pond[J]. Chinese Journal of Applied Ecology， 2014， 25（5）： 1499-1505. （in Chinese）
[15] LI ZF， WANG GJ， WANG GJ， XIE J， et al. Microbial carbon metabolic characteristics of biofilm communities in the grass carp culture pond based on Biolog-ECO plates[J]. Journal of Fisheries of China， 2014， 38（12）： 1985-1995. （in Chinese）
[16] WANG CH， GAO Q， TAN HX， et al. Effects of initial establishment of a biofloc technology system on the growth and immune enzyme activity of GIFT tilapia （Oreochromis niloticus）[J]. Journal of Fishery Sciences of China， 2015， 22（4）： 707-715. （in Chinese）
[17] LI CB， LI ZF， HAN LQ， et al. The effect of biofloc technology on indoor cultivation of small tilapia species[J]. Journal of Aquaculture， 2015， 36（7）： 29-35. （in Chinese）
[18] LIU WC， LUO GZ， TAN HX. The effect of no carbon source addition on water quality， growth performance and microbial community in a tilapia biofloc culture system that is reused in water[C]∥Abstracts of the 2017 Annual Conference of the Chinese Fisheries Society. Beijing： Chinese Fisheries Society， 2017. （in Chinese）
[19] CRAB R， CHIELENS B， MATHIEU W， et al. The effect of different carbon sources on the nutritional value of bioflocs， a feed for Macrobrachium rosenbergii post larvae[J].Aquaculture Research， 2010， 41（5）： 559-567.
[20] XU WJ. Research and application of functional effects of biological flocs in shrimp zero-water exchange culture system[D]. Qingdao： Ocean University of China， 2014. （in Chinese）
[21] TANG HJ， ZHANG XH， KONG CM， et al. Effect of Bacillus subtilis and molasses on water quality and growth of tilapia under zero-water exchange[J]. Journal of Huazhong Agricultural University， 2018， 37（3）： 82-86. （in Chinese）
[22] MCINTOSH D， SAMOCHA TM， JONES ER， et al. The effect of a commercial bacterial supplement on the high-density culturing of Litopenaeus vannamei with a low-protein diet in an outdoor tank system and no water exchange[J]. Aquacultural Engineering， 2000， 21（3）： 215-227.
[23] CHOI KH， DOBBS FC. Comparison of two kinds of Biolog microplates （GN and ECO）in the irability to distinguish among aquatic microbial communities[J]. Journal of Microbiological Methods， 1999， 36（3）： 203-213.
[24] MARTNEZ-CRDOVA LR， MARTNEZ-PORCHAS M， EMERENCIANO MGC， et al. From microbes to fish the next revolution in food production[J].Critical Reviews in Biotechnology， 2017， 37（3）： 287-295. [25] LU BG， WANG HY， XIE J， et al. The effect of different C/N levels on the formation of grass carp pond biofloc and its water quality[J]. Journal of Fisheries of China， 2013， 37（8）： 1220-1228. （in Chinese）
[26] LI Y， LIU LP， ZHAO GX， et al. The effect of adding carbon source in the aquaculture water on water quality and the growth of tilapia[J]. Journal of Dalian Fisheries University， 2013， 28（1）： 55-60. （in Chinese）
[27] SHANG Q. Study on the application of cassava residue in the treatment of aquaculture wastewater by biological flocs[D]. Nanning： Guangxi University， 2017. （in Chinese）
[28] TANG XF， LIU LP， SHUAI D， et al. The effect of carbon source on water quality and production performance of Anguilla marmorata culture system[J]. Journal of Dalian Fisheries University， 2019， 34（1）： 70-79. （in Chinese）
[29] WANG T， LIU QS， DUAN YF， et al. Study on the autotrophic and heterotrophic nitrification performance of low C/N domesticated biological flocs[J]. Marine Fishery， 2018， 40（5）： 614-624. （in Chinese）
[30] SCHNEIDER O， SERETI V， EDING EH， et al. Analysis of nutrient flows in integrated intensive aquaculture systems[J]. Aquacultural Engineering， 2005， 32（3/4）： 379-401.
[31] GARLANDJL. Analytical approaches to the characterization of samples of microbial communities using patterns of potential C source utilization[J].Soil Biology & Biochemistry， 1996， 28（2）： 213-221.
[32] LI ZF， WANG GJ， YU DG， et al. Effect of bioflocs on water quality and metabolic functions of microbial community in zero-water exchange tanks[J]. Journal of Shanghai Ocean University， 2015， 24（4）： 503-512. （in Chinese）
[33] SHI LL， FAN LM， CHEN JC， et al. Effects of combined fillers on the water quality， growth of tilapia， and functional diversity of the microbial community in tilapia aquaculture water [J]. Journal of Agro-Environment Science， 2017， 36（8）： 1618-1626. （in Chinese）
[34] LI J. Soil microbial characteristics and seasonal changes in long-term different fertilization systems[D]. Beijing： Chinese Academy of Agricultural Sciences， 2008. （in Chinese）
[35] WILN BM， ONUKI M， HERMANSSON M， et al. Microbial community structure in activated sludge floc analysed by fluorescence in situ hybridization and its relation to floc stability[J]. Water Research， 2008， 42（8/9）： 2300-2308.
[36] VERMA NM， MEHROTRA S， SHUKLA A， et al. Prospective of biodiesel production utilizing microalgae as the cell factories： A comprehensive discussion[J].AfricanJournal of Biotechnology， 2010， 9（10）： 1402-1411. [37] BODK I， BLSTKOV A， SEDLCEK S， et al. Biodiesel waste as source of organic carbon for municipal WWTP denitrification[J].Bioresource Technology， 2009， 100（8）： 2452-2456.
[38] ZHANG Z， YANG ZW， GE H， et al. The effect of different carbon sources on the structure， nutrient composition， bacterial community and water quality of the biological flocs in the water body for raising standard-diameter Litopenaeus vannamei[J]. Journal of Fisheries of China， 2019， 43（3）： 639-649. （in Chinese）
[39] BALLESTER E， ABREU PC， CAVALLI RO， et al. Effect of practicaldiets with different protein levels on the performance of Farfantepenaeus paulensisjuveniles nursed in a zero exchange suspended microbial flocs intensive system[J]. Aquaculture Nutrition， 2010， 16（2）： 163-172.
[40] FAN LM， WU W， HU GD， et al. Effect of restoration of intensive ponds by multilevel bio-system with the method of principal component analysis[J]. Chinese Agricultural Science Bulletin， 2010， 26（23）： 392-396. （in Chinese）