The Inﬂuence of Human Interference on Zooplankton and Fungal Diversity in Poyang Lake Watershed in China

: The Poyang water system in Jiangxi Province, China, is important for ﬂoodwater storage, diversity maintenance, and the economy of the Poyang Lake watershed. In recent years, pollution has destroyed the ecosystem and impacted human health and the related economy. The water quality of the Poyang Lake watershed and the impact of human interference must be assessed. Conventional analysis and high-throughput sequencing were used to evaluate the structure of both zooplankton and fungi in six sub-lakes of the Poyang Lake watershed under di ﬀ erent anthropogenic inﬂuences. The sub-lakes included were Dahuchi Lake (in natural preserve, DHC), Shahu Lake (in natural reserve, SH), Nanhu Lake (out of natural preserve, NH), Zhelinhu Lake (artiﬁcial reservoir, ZLH), Sixiahu Lake (agricultural lake artiﬁcially isolated from Poyang Lake, SXH), and Qianhu Lake (urban lake, QH). The densities and biomass of the zooplankton in DHC, SH, NH were higher compared with those in SXH, ZLH and QH ( p < 0.05). Zooplankton distribution of SXH was the most strongly associated with total nitrogen (TN), total phosphorus (TP) and chlorophyll a (Chl a), while QH was highly associated with pH, conductivity (Cond), and water temperature (WT). For fungal diversity, a large number of beneﬁcial fungi, Basidiomycota (phylum level) and Massarina (genus level) were obtained from DHC (55.3% and 27.5%, respectively), SH (54.4% and 28.9%, respectively), and NH (48.6% and 1.4%, respectively), while a large number of pathogenic Chytridiomycota (at phylum level) were identiﬁed from SXH (21.0%), ZLH (5.5%), and QH (7.5%). Manmade pollutants have impacted the natural hydrology and water quality and promoted variation between the zooplankton and fungi in the six sub-lakes, reducing the relative abundance of beneﬁcial fungi and increasing the number of pathogens in the environment, which threatens human health and economic production. Understanding the diversity among the zooplankton and fungi in the six sub-lakes of the Poyang Lake watershed may help guide future water management practices.


Introduction
Poyang Lake is the largest inland lake in China and is located along the main part of the low-to-mid section of the Yangtze River, which is the longest river in Asia. A total of 44 million people live on 166,900 km 2 of land in the Poyang Lake watershed [1]. The Poyang Lake ecosystem provides billions 1.
To analyze the assemblage of zooplankton and fungi to determine whether human activities have led to significant eco-environmental variance between six sub-lakes.

2.
To study whether the change of microbial assemblage will threaten human health and economic production.

Study Site Description and Water Sampling
Dahuchi Lake (DHC, 29.138358 N, 115.955195 E), Shahu Lake (SH, 29.168276 N, 115.933349 E), and Nanhu Lake (NH, 29.198585 N, 115.860815 E) are sub-basins of Poyang Lake (Table 1). DHC and SH lie in the Poyang Lake National Nature Reserve while NH Lake is outside of the reserve. These three sub-lakes connect to Poyang Lake. Zhelinhu (ZLH, 29.228336 N, 115.510613 E) is an artificial reservoir located upstream of Poyang Lake. Sixiahu Lake (SXH, 29.281464 N, 115.906471 E) was a sub-basin of Poyang Lake but is now separated from Poyang Lake by a dam that supports fishing. Qianhu Lake (QH, 28.655046 N, 115.820929 E) is an artificial urban lake surrounded by several colleges, the Administrative Center of Jiangxi Province, and Qianhu Hotel. Table 1 shows the sampling times, human interferences, and the number of sampling sites tested from the sub-lakes. Zooplankton and fungi were collected from DHC, SH, NH, ZLH, SXH, and QH lakes on July 12th and 13th, 2018 in the Poyang watershed, Jiangxi Province, China ( Figure 1). Samples were obtained from the lakeshore of each of the sub-lakes at three discrete sampling sites at 1-km intervals at a depth of 50 cm. Then, 10 L samples of zooplankton were obtained, 250 mL samples of fungi were obtained, and 250 mL samples were taken for physicochemical analysis. A Plexiglas water collector (WB-PM, Beijing Purity Instrument Co., Ltd., Beijing, China) was used for water sampling and the water collector was submerged and washed twice before water samples were collected. At sampling sites, the zooplankton samples were fixed by 4% formalin to avoid imprecise biomass and density measurements. In the laboratory, water samples for fungi were filtered with Whatman GF/F filters (pore size: 0.7 µm) to obtain sediments, which were stored in a refrigerator at 4°C for further DNA extraction and high-throughput sequencing analyses of fungi. Water samples for nutrient salts (total nitrogen and total phosphorus) and chlorophyll a concentration were stored in a 4 • C refrigerator in the lab, and measured for total nitrogen (TP), total phosphorus (TP), and chlorophyll a concentration within 24 h of collection.

Densityand Biomass of Zooplankton
Zooplankton samples were filtered on site using a 25# zooplankton net (200-µm mesh size, net opening 20 cm diameter, Beijing Purity Instrument Company) and the sample volume was concentrated to 40 mL. The concentrated sample used in density calculations was well-stirred and a 5-mL subsample was observed under a dissecting microscope at 10 × 10 magnification to count the species and number of Rotifera, Cladocera, and Copepoda. The lengths of crustaceans were measured for biomass calculations to determine their weight according to the length-weight regression curve. Rotifera were measured by volume to determine their weight [26].

Physiochemical Analysis of Water
Conductivity (Cond), pH, water temperature (WT), dissolved oxygen (DO), and turbidity (Turb) were measured in situ at the sampling site using the Multi-function Water Quality Monitor (YSI 6600 V2, YSI Inc., Yellow Springs, OH, USA). Chlorophyll a (Chl-a) was extracted from the samples using 45% acetone for 24 h and was analyzed and calculated using the fluorometric method (Turner Designs, San Jose, CA, USA) [27]. Total nitrogen and total phosphorus were calculated using the colorimetry method [28].

Extraction, Polymerase Chain Reaction (PCR) Amplification and High-Throughput Sequencing of Aquatic Fungal Genome
Lake water was filtered through a Whatman GF/F filter (Whatman, Maidstone, UK; Pore size, 0.7 µm), and then the fungal DNA was extracted from fungi in sediments filtered by GF/F filter using DNA genomic kit (Tiangen Biotech Co., Ltd., Beijing, China). The concentration and purity of each DNA Diversity 2020, 12, 296 4 of 16 sample were determined by UV spectrophotometer to prepare the samples for amplification (Nano Drop, Thermo Scientific Inc., Waltham, MA, USA). PCR amplification relies on the V4 region, which is a preserved region of fungal 18S ribosomal DNA. This site was amplified by primer 547F/V4R (547F, 5 -CCAGCASCYGCGGTAATTCC-3 ; V4R, 5 -ACTTTCGTTCTTGATYRA-3 ). PCR products were analyzed with sequence reads using Illumina HiSeq 2000 (BioProject accession number PRJNA560147).

Extraction, Polymerase Chain Reaction (PCR) Amplification and High-Throughput Sequencing of Aquatic Fungal Genome
Lake water was filtered through a Whatman GF/F filter (Whatman, Maidstone, UK; Pore size, 0.7 μm), and then the fungal DNA was extracted from fungi in sediments filtered by GF/F filter using DNA genomic kit (Tiangen Biotech Co., Ltd., Beijing, China). The concentration and purity of each DNA sample were determined by UV spectrophotometer to prepare the samples for amplification (Nano Drop, Thermo Scientific Inc., Waltham, MA, USA). PCR amplification relies on the V4 region, which is a preserved region of fungal 18S ribosomal DNA. This site was amplified by primer 547F/V4R (547F, 5′-CCAGCASCYGCGGTAATTCC-3′; V4R, 5′-ACTTTCGTTCTTGATYRA-3′). PCR products were analyzed with sequence reads using Illumina HiSeq 2000 (BioProject accession number PRJNA560147).
The Shannon-Wiener diversity index (H'), Pielou's evenness index (J) and Margalef richness index (D) of zooplankton were calculated using the formulas: where N is the total number of specimens; S is the total number of zooplankton at species level (Table A1); P i is the ratio of number of the i species to total number of zooplankton (n i /N). The evaluation standard of H' and D: 0 <H'(D) < 1 is heavy pollution; 1 <H'(D) < 2 is α-medium pollution; 2 <H'(D) < 3 is β-medium pollution; H'(D) > 3 is light pollution [32].
Redundancy analysis (RDA) used the zooplankton communities from different lakes as dependent variables based on the physicochemical factors of water (environmental variables). The Monte Carlo Permutation Test calculated the distribution rate of each factor to determine the significant physicochemical factors. The rare communities were ruled out using forward selection [18] and the collinear environmental variables were discarded to avoid type I error [16]. CANOCO for Windows (version 4.5, Biometrics-Plant Research International, Wageningen, the Netherlands) was used to perform the Monte Carlo test and RDA. PRIMER 5.0 helped form the non-metric multidimensional scaling ordination (NMDS) image to determine the Euclidean distance between samples and Bray-Curtis similarity was used in community clustering [33].
SPSS version 17.0 (SPSS Inc., Chicago, IL, USA) was used for one-way ANOVA analysis of the biomass and diversity of Rotifera, Cladocera, Copepoda, and total zooplankton species between different sub-lakes. Data were presented as mean ± SD or mean ± SEM; p < 0.05 indicated a significant result.

Zooplankton and Environment Data from Different Lakes
A total of 35 species of zooplankton were collected (Table A1). As shown in Figure 2 and Table 2, higher values for total density and biomass were obtained from DHC, SH, NH, and QH. The values of total density and biomass were lower in SXH and ZLH. For specific zooplankton, Rotifera, specifically, had higher values of density and biomass in DHC, SH, NH, and SXH versus in ZLH and QH (p < 0.05). Cladocera values were high in QH.  On X-axis, DHC, SH and NH were sub-lakes connected to Poyang Lakes, SXH was closed sub-lake, ZLH was tourism type reservoir and QH was artificial urban lake. The significant differences (* p < 0.05, ** p < 0.01, *** p < 0.001) have been presented in group comparisons. On X-axis, DHC, SH and NH were sub-lakes connected to Poyang Lakes, SXH was closed sub-lake, ZLH was tourism type reservoir and QH was artificial urban lake. The significant differences (* p < 0.05, ** p < 0.01, *** p < 0.001) have been presented in group comparisons.
The diversity indexes (D and H') ( Table 3) showed that DHC and SH had medium pollution levels, and NH, SXH, ZLH and QH were heavily polluted. One-way ANOSIM revealed that the sub-lakes had a significant impact on zooplankton communities (Global test: R = 0.849, p = 0.001). Non-metric multidimensional scaling analysis (NMDS) based on the number of individual zooplankton varied between zooplankton community structures among the six sub-lakes. The zooplankton from the six sub-lakes were divided into the following communities: DHC, SH, and NH, SXH, ZLH, and QH ( Figure 3A). Redundancy analysis (RDA) results showed that the total variance contribution of the physicochemical factors of pH, Turb, Chl-a, Cond, and DO to the changes of zooplankton communities is 67%. The characteristic value of the first ranking axis is 0.335 and the variance contribution rate is 46.7%. The eigenvalue of the second ranking axis is 0.146 and the variance contribution rate is 20.3%. Among the six sub-lakes, QH had highest Cond and pH and SXH had high concentrations of TN, TP, and Chl-a. A high pH was identified in SXH and ZLH ( Table 4). The Monte Carlo test found that pH and Chl-a concentrations were significantly related to the changes of zooplankton communities (F = 3.79, p = 0.004; F = 3.03, p = 0.004), and Turb had a significant effect on the zooplankton community structure (F = 2.9, p = 0.004, Figure 3B). pH, Turb, and Chl-a concentrations were the main environmental factors causing spatial differences in zooplankton communities. However, nutrient concentrations were not important environmental factors causing spatial differences among the zooplankton communities ( Figure 3C). Table 4. Physicochemical parameters in the six studied sub-lakes. Total nitrogen (TN), total phosphorus (TP), chlorophyll a (Chl-a), water temperature (WT), water depth (WD), turbidity (Turb), pH, dissolved oxygen (DO), conductivity (Cond) of six sub-lakes were detected and calculated. Mean ± SE.

Taxa Abundance and KEGG Analysis
Several taxa derived from high-throughput sequencing data were analyzed to further investigate the fungal structure of different lakes. Atractospora, Scleroderma, and Velutarina were higher in ZLH compared with the same taxa in other lakes ( Figure 5). DCH had the highest relative abundance of Chalaropsis, Ephebe and Raffaelea, Sympodiomycopsis, and Thielaviopsis. High relative abundances of Ceratocystis, Massarina, Ophiodiaporthe, and Sympodiomycopsis were observed in SH and DCH. NH had the highest relative abundance of Catenochytridium and Phlyctochytrium. Basidiobolus was highly abundant in SXH. NH had generally low relative abundance in most taxa.
KEGG provided interpreted sequences to present the relative abundance of the six lakes and their impact on different human diseases ( Figure 6). All of the lakes were highly associated with infectious diseases. Degenerative diseases ranked as the second highest disease correlated with the six lakes, among which ZLH ranked the highest.
guidance for the management of the water quality to benefit people living on the Poyang Lake watershed. However, this study was limited by its short study period, making it difficult to separate natural variability from manmade changes, therefore, further studies should be conducted.
Author Contributions: H.Q. and T.C. designed the experiments; H.Q., L.C. and Q.L. made Data curation; H.Q., X.C. and T.C. analyzed the data and wrote the manuscript. All authors performed the experiments and discussed the results and commented on the final manuscript.

Conflicts of Interest:
The authors declare that they have no competing interests.

Discussion
The Poyang water system provides a habitat for numerous species and supports the economy and livelihoods of people residing in the Poyang Lake watershed [4]. However, the recent human destruction of the natural ecosystem has created unexpected consequences for human health and economic production [5,7,8]. Zooplankton are sensitive to changes in physicochemical factors [19] and fungi play a vital role in both purifying and toxifying the water [24,25]. Analyzing zooplankton

Discussion
The Poyang water system provides a habitat for numerous species and supports the economy and livelihoods of people residing in the Poyang Lake watershed [4]. However, the recent human destruction of the natural ecosystem has created unexpected consequences for human health and economic production [5,7,8]. Zooplankton are sensitive to changes in physicochemical factors [19] and fungi play a vital role in both purifying and toxifying the water [24,25]. Analyzing zooplankton and fungi diversity can improve our understanding of the factors governing water quality of the lakes and assess how human activity change the structure of zooplankton and fungus.
The biomass and density of zooplankton reflect the function and eutrophic state of the ecosystem. The biomass is positively correlated with biodiversity to a certain extent [34,35]. The total biomass and density of the zooplankton in the DHC, SH, NH, and QH lakes were higher than those in the ZLH and SXH lakes ( Figure 2, Table 1). The low diversity index showed heavy pollution in ZLH, SXH, and QH. Natural lakes have a self-purification system and healthy zooplankton structure that polluted lakes lack, due to frequent water fluctuation (good fluidity) [36,37]. The high total biomass and density of zooplankton in DHC and SH may be explained by the strict protection of the waters by local governments and good natural fluidity (DHC and SH connect to the main lake when the water level reaches 17.3 m and NH connects to the main lake at a higher water level). In SXH, extensive aquaculture production degraded the water quality, produced a large amount of waste, and accumulated excessive amounts of nitrogen and phosphorus so that the habitat, niche spaces, and zooplankton diversity were reduced. A high degree of aquaculture and species input limitation imposed by the dam can contribute its low diversity of zooplankton [38,39]. Zooplankton is vital to supporting nutrition for fishes, so reduced amount and species of zooplankton will negatively impact the fish predation pressure and aquaculture economy can be impaired. ZLH is the only upstream reservoir with functions including water conservation, tourism, aquaculture, and flood storage, and it has been subjected to industrial emissions, mismanaged aquaculture practices, and tourism construction [40]. The overuse of chemical detergents and poor aquaculture practices killed zooplankton, reduced their food resources, polluted their habitat, and reduced the overall diversity of zooplankton of ZLH [39]. Interestingly, a contradiction of low diversity index and high total density and biomass of Cladocera was observed in QH, which may suggest the overgrowth of more tolerant zooplankton species and more simple structure of dominant species in this polluted and eutrophic water body [41], such as Bosmina longispina in Taihu Lake, Jiangsu Province, and Brachionus forficula in Yueliang Lake, Nanchang [42].
We further analyzed the association between the physicochemical factors and sub-lakes based on zooplankton distribution, indicating that the dispersal and growth of zooplankton can be explained by TP, TN, Chl-a, DO, Cond, pH, and WT. TP is strongly correlated with the biomass of algae (containing Chl-a), resulting in an increase of zooplankton production, while TN has a general effect on the production of aquatic organisms [28]. Therefore, the higher content of TN, TP, and Chl-a together indicates the eutrophication and acquaculture degree of SXH, and eutrophication has impacted the zooplankton assemblage. The samples taken from SXH verify the results of previous study, which showed that Copepoda and Cladocera did not grow well in a eutrophic state, while Rotifera had a higher rate of reproduction as TP and Chl-a increased [43]. A high Cond indicates high levels of metal and chemical pollution and challenges the growth of zooplankton, but the density of specific tolerant species can increase under extreme conditions [44]. Zooplankton in QH was positively correlated with pH and Cond; correspondingly, QH had a low density and biomass of zooplankton with the exception of Cladocera, which may suggest the bad effect of heavy industrial pollution in QH [45]. High Cond reflected high salt loads and the biodiversity of zooplankton decreased under high alkaline and salt conditions, with only a few dominant species prevailing. The high concentrations of chemical gradients from QH can pollute groundwater, the surrounding farmlands, and impact human health [9].
The analysis of zooplankton and physicochemical factors revealed that the water quality of DHC and SH was relatively normal, further indicating that DHC and SH varied from other sub-lakes. RDA image of zooplankton analysis and PLS-DA image of fungal sequencing indicated the differences between NH, ZLH, SXH, and QH. Ascomycota and Chytridiomycota are the most common aquatic fungi at the phylum level [46] and Basidiomycota play significant role in the carbon cycle by decomposing organics [47]. Certain species of Chytridiomycota cause chytridiomycosis, which is known to kill huge populations of amphibians [48]. A higher relative abundance of Basidiomycota in DHC, SH, and NH and lower relative abundance of Chytridiomycota in DHC and SH suggest that DHC and SH have healthier zooplankton structure than other lakes. DHC and SH were noted for their beneficial fungal structure and lower risk for spreading disease. At the genus level, Massarina belongs to the predominant phylum Ascomycota and plays a vital role in decomposing wood and providing energy [49]. Several species of Basidiobolus lead to cutaneous zygomycosis in humans [50]. Massarina relative abundance was highest in DHC and SH, and the relative abundance of Basidiobolus was highest in SXH, which indicate the lakes of frequent human activities have worse condition of zooplankton than that of DHC and SH. The carbon cycle in NH, SXH, QH and ZLH is weaker and SXH is associated with more health concerns due to greater amounts of pathogenic fungal genus. KEGG analysis indicated that ZLH was highly associated with neurodegenerative disease and cancers, but it is still a primary water supply for the surrounding population. Certain pathogenic fungi are tolerant to alkaline conditions and are predominant in water [51]. We believe that the high pH values of QH, SHX, and ZLH indicate higher concentrations of pathogenic species and an increased risk for disease in humans and aquatic species.

Conclusions
We analyzed the three taxa (Rotifera, Copepoda, Cladocera) of zooplankton, diversity of fungi, and relevant environmental factors, finding that human activities destroyed the natural hydrology of sub-lakes, reducing the diversity of zooplankton and fungi, increasing noxious environmental factors and pathogens, and worsening the water quality to potentially harm human life and economic production. We analyzed the detailed microbial structure of the sub-lakes of the Poyang Lake watershed to present the influence of anthropogenic factors on the sub-lakes, providing guidance for the management of the water quality to benefit people living on the Poyang Lake watershed. However, this study was limited by its short study period, making it difficult to separate natural variability from manmade changes, therefore, further studies should be conducted.

Conflicts of Interest:
The authors declare that they have no competing interests.