Rchit Penglee, Lei Go, Yjun Hung, Liping Liu, Sukkrit Nimitkul, Bolong Bo,＊

aKey Laboratory of Exploration and Utilization of Aquatic Genetic Resources (Shanghai Ocean University), Ministry of Education, International Research Center for Marine Biosciences at Shanghai Ocean University, Ministry of Science and Technology, National Demonstration Center for Experimental Fisheries Science Education,Shanghai Ocean University, Shanghai, 201306, China

bDepartment of Aquaculture, Faculty of Fisheries, Kasetsart University, Chatuchak, Bangkok, 10900, Thailand

ABSTRACT

Nitric oxide (NO)—the product of arginine metabolism catalyzed by nitric oxide synthases (NOS)—is a wellknown neurotransmitter which plays an important role in metabolism and amino acid transportation in the nervous system. In particular, it can inhibit monoamine neurotransmitter transportation which affects animal behavior, especially social behavior. Shoaling—is a one kind of social behavior. It is a behavior that individual fish choose to join with their group within two factors; food and predation risk. Shoaling fish has quickly responded to predator and increased the change in feeding competition. In addition, shoaling also effect to stress response on stock density of aquaculture system. The effect of NO molecular signaling on the dopamine pathway was investigated using zebrafish (Danio rerio) as a model organism. Our aim was to understand the role of NOS and NO in shoaling behavior, which is typical of zebrafish. The concentration of NO in the zebrafish brain was modulated using a knockout for the neuronal NOS gene, and NO production was induced through treatment with L-arginine. The existence of NO in the zebrafish brain was confirmed by using a fluorescent probe. Dopamine concentration in the brain was measured by UPLC tandem mass spectrometer. We measured shoaling cohesion of all individual fish of D. rerio, using average distance between all pairs offish (nearest neighbor distance) and analyzed tracking by Zebralab ViewPoint software. Collectively, our results suggest that a lower level of NO was associated with a higher level of dopamine, which in turn leads to the shoaling behavior.

ARTICLEINFO

Keywords:

Nitric oxide

Neuronal nitric oxide synthase

Shoaling

Dopamine transporter

Neurotransmitter

Abbreviations: CNS, central nervous system; CRISPR, clustered regularly interspaced short palindromic repeats; CuFL, copper (II) complex fluorescence; DAT,dopamine transporters; L-Arg, L-arginine; mRNA, messenger RNA; NO, nitric oxide; NOS, nitric oxide synthase; nos1 −/−, homozygous nos1 knockout; nNOS or nos1,neuronal NOS; iNOS or nos2, inducible NOS; eNOS or nos3, endothelial NOS; PCR, polymerase chain reaction; sgRNA, single guide RNA; STR, short tandem repeat;NADPH, nicotinamide adenine dinucleotide phosphate.

1.Introduction

Social behavior is defined as behavior among two or more organisms interacting such as in communication, allogrooming, aggression, mating behavior, and parental behavior (Gammie, 2010). Previous work identified the mesolimbic system (responsible for reward and social decision making) as a crucial neural network in social behavior (Schultz, 2015).In teleost fish, one kind of social behavior is shoaling (O’Connell &Hofmann, 2011), which is an antipredator adaptation in fish and is a robust tool for measuring social behavior in animals (Kim et al. 2017).Groups offish are commonly termed either shoals or schools. However,shoaling is defined as a social group offish that has no implications for structure or function, whereas schooling involves synchronized and polarized swimming. Schooling is therefore one of the behaviors exhibited by fish in shoals (Miller & Gerlai, 2012a). Although fish shoals are often large and conspicuous, swimming in a group protects fish from predators in several ways, and shoaling fish are able to detect and respond to predator quickly (Pitcher, 1993).

Many kinds of animals have shown shoaling-like behavior described,for example, as swarming or crowding. Shoaling can protect the member in shoal from predator and greater access to according to a phase transition study of locusts, the levels of messenger RNA (mRNA) of nitric oxide synthase (NOS), phosphorylated NOS, and nitric oxide (NO) were found to be different in solitary locusts compared with locusts in the gregarious phase (Hou et al. 2017). It is therefore possible that NO and NOS have a role in the molecular mechanisms underlying shoaling, but understanding of this potential role is limited. Zebrafish (Danio rerio)exhibit the shoaling cohesion in zebrafish increases when they age up. In addition, shoaling level in zebrafish is associated with concentration of certain neurotransmitters (e.g., dopamine; 3,4-dihydroxyphenylacetic acid; serotonin; and 5-hydroxyindoleacetic acid) (Buske and Gerlai 2011, 2012).

Nitric oxide (NO) is a highly diffusible gas and crosses biological membranes without difficulty. Its half-life is only a few seconds, but even during this short period it can diffuse a few hundred micrometers(Gally et al. 1990). It is well known as a neurotransmitter that reflected on metabolism and transport of the amino acid in the nervous system(Wiesinger, 2001). In mammalian cells, NO is produced from an oxidation reaction between L-arginine (L-arg) and oxygen, that has nicotinamide adenine dinucleotide phosphate as a co-substrate and is catalyzed by NOS (Yarlagadda et al. 2017). NO can be generated by three different isoforms of NOS: neuronal NOS (nNOS or nos1), inducible NOS (iNOS or nos2), and endothelial NOS (iNOS or nos3). However,teleost fish only have nos1 and nos2 genes (Lepiller et al. 2009). Both forms can be found in the central nervous system (CNS), but specific actions on neurotransmission may be attributed primarily to NO produced by nos1. Previous work has suggested that NO produced at glutamatergic synapses may influence the function of a large number of neurons in a sphere around the synapse (Kiss, 2000), in particular, in the process that involves monoamine transporters. NO exerts an inhibitory effect on dopamine transporters (DAT) (Kiss et al., 2004; Motahari et al.,2016). Dopamine transporters are important monoamine transporters in the dopaminergic pathway that drive reuptake of extracellular dopamine neurotransmitters into presynaptic neurons, which can in turn affect behavior (Vaughan & Foster, 2013).

Here, we present NO and nos1 had potential role to shoaling behavior. Both NO and nos1 was found in zebrafish brain (Holmqvist et al. 2000), besides zebrafish are the best shoaling species model (Miller& Gerlai, 2012a; Zala et al., 2012). Thus, zebrafish were used for animal model study. We further hypothesized that NO would affect shoaling behavior though dopaminergic pathway.

2.Methods

2.1.Animals and housing

In total, 500 wild-type zebrafish (Danio rerio; WT, AB strain) and 500 zebrafish with nos1−/−mutations were used for all experiments. The zebrafish originated from progenitors obtained from the Biology and Life Science Laboratory of Shanghai Ocean University. All fish used in this study were bred, raised, and housed in the same environment and conditions. Gender could not visually be determined when the experiments commenced at 0- to 60-days post fertilization.

Zebrafish embryos were raised at 28C in groups of twenty in cell culture dishes (90 mm ×20 mm) with 1% methylene blue (1 mL: 1000 mL). After 7 days post fertilization, the animals were transferred to a square tank (15 cm × 25 cm × 12 cm). 60 days post fertilization later,the fish were transferred to a rearing system where the water was maintained at 28 ± 0.5C, and fish were kept in a 13-h light/12-h dark cycle, with lights on at 8:00 a.m. and off at 9:00 p.m. All fish were fed twice daily with paramecium until 14 days post fertilization, after which animals were fed twice daily with nauplii of brine shrimp (Artemia sp;FengNian, Tianjin, China). All experiments were performed in accordance with the animal welfare regulations for scientific research formulated by the Institutional Animal Care and Use Committee(IACUC) of Shanghai Ocean University.

2.2.Generation of the nos1 mutant

We used HiScribe™ T7 Quick High Yield RNA Synthesis Kit (New England Biolabs, Massachusetts, UK) and pT3TS-nCas9n to generate single guide RNA (sgRNA) and Cas9 mRNA, respectively. Then, the sgRNA were purified with the RNA Clean & Concentrator-25 kit (Zymo Research, Irvine, California, USA). Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 target site design was done online(http://genome.ucsc.edu) and exon 1 of the zebrafish gene nos1 was targeted (sgDNA: 5-AGGGCGAACTGAGATGATGTT-3). The purified sgRNAs (50 pg/μL) were co-injected with Cas9 protein (300 pg/μL) into single cell, fertilized, wild type zebrafish embryos. CRISPR injected embryos were raised to adulthood and self-crossed to produce the F1 generation. The F1 progeny were genotyped by sequence analysis of genomic DNA isolated by caudal fin clipping, and screened for the germ line mutation by short tandem repeat (STR) analysis. The primer sequences used to amplify the exon 1 sequences were 5-TTGACCAGAAAGCCCAGTCC-3’ (forward), and 5-ATCAGCGTTGACGCACAGTA-3’ (reverse). The identified heterozygous mutant zebrafish were selfcrossed to produce the F1 generation. The homozygous nos1 knockout(nos1−/−) zebrafish were screened from F3 progeny produced by inbreeding the F2 homozygous progeny.

2.3.DNA extraction for sequencing analy sis

DNA extraction from zebrafish caudal fins (embryos and adults) was achieved using a protocol based on the use of sodium hydroxide and trisaminomethane hydrochloride (modified from Dupret et al. 2018). The tips of the caudal fins of adult zebrafish (F0, F1, F2) and embryos (F3)were cut off and put into centrifuge tubes; 10 μL of 50 mM sodium hydroxide was then added to the tissue biopsy. Next, the samples were incubated at 95C for 20 min before the solution was neutralized by adding 1 μL of 1 M tris-aminomethane hydrochloride. The DNA sample was stored at 20C until used. Polymerase chain reaction (PCR)amplification of nos1 mutant and nos1 wild type genes (using forward and reverse primers) was done to distinguish wild type and nos1 knockout base pairs by restriction fragment length polymorphism and STR assays (Sangon Biotech, Shanghai, China). In the STR assay, DNA samples were analyzed by amplifying one or several of the STR regions.If there were differences in the size of the STRs between the samples, it was determined by size separation on the gel, while the restriction fragment length polymorphism assay showed any differences between the DNA sequences. Any sequence of interest was first amplified by PCR and the product was subjected to restriction enzyme digestion to identify whether the amplified base pairs were wild-type or mutant.

2.4.L-arginine treatment

L-arginine (L-arg) was used to produce NO in the zebrafish cells.Wild-type and nos1 −/− zebrafish were maintained in small plastic tanks (18 cm × 11 cm × 11 cm) containing 200 mL of 2 mM L-arg for 2 h(The suitable dose and time that the fish can be survival and showed the significantly statistical different from control group). Control groups were maintained in 200 mL plastic tanks with tap water without chlorine for 2 h. The behavioral and biochemical experiments were performed immediately after the 2 h exposure to L-arg or to tap water.

2.5.Shoaling assay

All tests were performed on the 42nd days post fertilization in a large, white, circular acrylic tank (Fig. 1A). After the initial 2 h administration of L-arg, treated fish were placed in fresh treatment solution and transferred to testing tank, whereas control fish were placed in fresh water before being transferred to the testing tank. Each group(shoal) consisted of ten fish. Fish were netted as a group and immediately released in the center of testing tank. The testing tank size of every tank were equal. The temperature in the testing room was kept at 27–28C. Animals were first habituated to the testing tank for 120 s before the testing session started. Still images were obtained every 1 s of the 15 min duration of the recorded trials. The trajectories of the fish in the first 5 min of each session were discarded. All videos were recorded with a digital camera (Sony DSC-W520) and the shoaling parameter of each group was monitored by Zebralab ViewPoint software (Viewpoint Life Sciences Inc., Shanghai, China). Shoaling parameters analyzed were average distance between all pairs offish (nearest neighbor distance),swimming speed, and polarization of shoal.

Fig. 1.Shoaling simulation model: (A) The behavior recording setup (modified from Miller & Gerlai, 2012b). The arena size was kept proportional to the average body length of the fish (4X the body length). (B) The distance was calculated as mean of all nearest neighbor distances for each focal zebrafish (red arrow). The red arrow means the closest distance between one fish and other individual (Buske, 2013). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

2.6.Fish euthanized for brain samples

Zebrafish from each treatment (wild-type with L-arg, nos1−/−,nos1−/−with L-arg and control) were sacrificed at the same time to obtain a total of 1000 brain samples. Fish were sacrificed at 42 days post fertilization. The fish were sacrificed corresponded to that used after the 2 h exposure to L-arg or to tap water by decapitation; the brains were quickly dissected on ice and placed in a microcentrifuge tube where they were kept at −80C until used.

2.7.Dopamine concentration measurement in whole brains

Chromatographic analysis was carried out on an Acquity™ UPLC HClass System (Waters, Shanghai, China) with a flow through needle sample manager. Mobile phase A (0.2% formic acid in water [v/v]) and mobile phase B (methanol-acetonitrile [v/v, 30 : 70]) were operated with a gradient elution at 0.4 mL/min as follows: 68% A (0–3.0 min),68% A → 63% A (3.0–7.5 min), 63% A → 5% A (7.5–7.6 min), 5% A(7.6–8.0 min), 5% A → 68% A (8.0–8.1 min), 68% A (8.1–8.5 min). The column temperature was set at 45C and the autosampler temperature was maintained at 15C. The multiple reaction monitoring transitions had a mass to charge ratio ranging from 322.0 → 136.9 for dopamine.The compound dependent parameters were dwell time of 200 ms,declustering potential of 100 V, entrance potentials of 10 V, collision cell exit potential of 10 V, and collision energy of 45 V for dopamine (modify from Zhang et al. 2016).

Brain tissue samples (5 brains per sample) were homogenized in 25 μL 1X phosphate-buffered saline (0.1 M phosphate buffer, 0.15 M sodium chloride; pH 7.4) with 475 μL 80% methanol. An aliquot of 25 μL of the sonicate from each sample was analyzed for protein content using the BCA Protein Assay Kit (CWBIO, Shanghai, China). To each tube, 1 μL of stabilizer (0.2 N perchloric acid and 1 N ascorbic acid) was added, and the tissue was centrifuged at 13,000 revolutions per min for 10 min at 4C. The supernatant was collected carefully into an autosampler vial and placed on the thermosttated autosampler (which was set for −80C)until used for dopamine analysis. Standard dopamine was used to quantify and identify the peaks on the chromatographs. The retention time for dopamine was approximately 23.7 min (Chatterjee & Gerlai,2009). The detection limit for dopamine was determined by running the known concentrations of dopamine in the UPLC tandem mass spectrometer under the set conditions. For this purpose, a standard solution of 1 mg/mL was made with dopamine hydrochloride and diluted accordingly to get the desired concentrations of the stock solution for running in the UPLC tandem mass spectrometer. The sensitivity was selected at the concentration at which there was a signal to noise ratio of 3:1. The data was analyzed by MassLynx 4.1 software (Waters,Shanghai, China).

2.8.NO fluorescent imaging

Because NO has a very short half-life, the fluorescence experiments were done on the same day as the shoaling recording. The solution used for NO detection was copper (II) complex (CuFL), followed by imaging by confocal microscopy (Leica Microsystems Inc, Shanghai, China).

A stock solution of 1 mM CuFL was prepared in dimethyl sulfoxide at room temperature in a 3 mL corning centrifuge tube covered in aluminum foil, then aliquoted in an Eppendorf tube (0.1 mL per tube) as stock solution. After that, CuFL was combined with copper (II) chloride(1 mM in double distilled water) in a ratio of 1:1 to make a 0.5 mM stock solution of CuFL at room temperature. Fresh CuFL solution was made on the day of analysis (Lim, 2007).

Five brains from fish subjected to each treatment were dissected on ice and placed in an Eppendorf tube. Then CuFL (0.5 mM) and 17 betaestradiol (0.1 mM) were added to the sample in a ratio of 2:1 and incubated at 37C for 40 min (modify from Nimittkul, 2014).

The fluorescent signal was read using confocal laser scanning microscopy (LesicaSP8, Leica Microsystems Inc, Shanghai, China). Excitation and emission are 492 nm and 520 nm respectively. All pictures were taken with the same microscope setting. The experiments were repeated three times.

2.9.Total RNA extraction and quantitative PCR

Real-time PCR for DAT expression was performed using zebrafish brains dissected 42 days post fertilization. RNA samples were extracted using RNAiso Plus (Takara Bio Inc, Shiga, Japan). The concentration of total RNA was measured by complementary DNA which was synthesized from 1 μg/500 ng of total RNA with a HiScript III RT SuperMix for qPCR(+gDNA Wiper) kit (Vazyme, Nanjing, China, R323-01). Quantitative PCR was performed using the ChamQ Universal SYBR qPCR Master Mix kit (Vazyme, Nanjing, China) on a 7500 Fast RealTime PCR System(Applied Biosystems, Singapore).

2.10.Data analysis

All statistical data were analyzed using RStudio (Version 1.1.463,Boston, Massachusetts, USA) and expressed as mean ±standard error of mean. The data from each experiment was tested the distribution by Shapiro-Wilk test. Nearest neighbor distance, speed and polarization data were analyzed by the student’s t-tests. For DAT gene expression and dopamine concentration in the zebrafish brain, data were analyzed using the Mann-Whitney test. Results were considered to be statistically significant when the p-value was less than 0.05 for pairwise t-test and Mann-Whitney test.

3.Results

3.1.Identification of nos1 mutant

The nos1−/−zebrafish were screened from the F2 progeny, and the mutation of the nos1 gene was confirmed by STR analysis (Fig. 2B). DNA sequencing results confirmed the targeted single nucleotide insertion mutations. These mutations led to a truncation of the nos1 protein,resulting in a product that lacked a catalytic domain (Fig. 2D). The phenotype between wild type and nos1−/−zebrafish was not different(Fig. 2C), whereas body length was significantly different, but only at 15 days post fertilization (p <0.001, n =30; Fig. 3).

Figure 2.Knockout of the neuronal nitric oxide synthase gene (nos1) using CRISPR/Cas9. (A) Schematic illustration of the nos1 locus; exon 1 of the nos1 gene was targeted.The CRISPR target site and PAM motif are indicated. (B) Short tandem repeat confirmation and sequence confirmation of single nucleotide insertion mutations. (C) Images of wild type (WT) and homozygous nos1 knockout (nos1 −/−) zebrafish at 42 days post fertilization. The fish do not show related morphology at this stage. (D) Predicted truncation of the nos1 protein.

Fig. 3.Body length of zebrafish. Total length measurements of wild type (WT)and homozygous nos1 knockout (nos1 −/−) zebrafish at 7, 15, 30, and 42 days post fertilization. N =30 for each time point. ＊＊ =p <0.01 via Student’s t-Test.Error bars indicate 95% Confidence Interval.

3.2.Fish without neuronal nitric oxide synthase deficiency display shoaling behavior

We generated the independent alleles of nos1−/−using CRISPR/Cas9 technology. It causes a predicted frameshift and early termination of protein translation by carrying 1-bp insertion in exon 1, causing of nitric oxide reduction in fish brain (Fig. 4). In addition, nos1−/−also has markedly increased DAT mRNA expression and dopamine concentration in fish brain (Figs. 6 and 7 respectively). The shoaling was tested at 42 days post fertilization. nos1−/−fish showed the average nearest neighbor distance inside their shoal were reduced compared with wild type (Fig. 5; t=4.17, p <0.001). They often swim near their fellow (Supplementary Video 3). We screened neurochemical compounds, especially those known to act at least in part by affecting nitric oxide signaling, upon the shoaling behavior of nos1−/−fish. 2 mM of L-arg was chosen for achieve a high nitric oxide situation in assessed shoaling behaviors. L-arg has the effect on shoaling behavior. It takes at least 2 h to manifest. Our results showed the fish that was incubated in L-arg swims separated from other members in their shoal compared with wild type that untreated with L-arg (Fig. 5; t=−7.66, p <0.001). In addition, L-arg exposure for 2 h causes the nos1−/−fish to resume swimming dispersed from shoal, as shown in the attached videos (Supplementary Video 2). Hence, L-arginine effectively restores the nos1−/−behaviors to wild type (Fig. 5; t=−4.65, p <0.001).

Fig. 4.Z-series from confocal laser scanning images of brain tissue. Wild-type (WT; AB strain) and homozygous nos1 knockout (nos1 −/−) zebrafish were first incubated wit (II) complex (CuFL; 1 μMfinal concentration) was added to all samples and incubated for another 40 min. The fluorescent signal was measured immediately after incubation. All pictures were taken with the same microscope setting using a confocal microscope from Leica Microsystems Inc (Shanghai, China).

The average swimming speed of each group was significant difference (F =64.04, p <0.001). The swimming speed offish shoal in L-arg treated group is significantly slower than control and nos1−/−group(Fig. 8; W=1441, p <0.001 and W=1600, p <0.001). Polarization distributions were parameter, that groups of zebrafish often form either highly polarized shoals (high shoaling cohesion) or weakly polarized shoals (low shoaling cohesion) (Miller &Gerlai, 2012a). Nitric oxide changes in zebrafish brain with NO substrate(L-arg) becoming decreasingly polarization distributions offish in shoal(Fig. 9). An effect of NO substrate (L-arg) and losing function of nos1(nos1−/−) to polarization distributions were observed. The result showed wild type fish that treated with L-arg spent less time to shoaling.The polarization of them was significantly lower than control and nos1−/−group (Wwt vs L-arg =1390, p <0.001; Wnos1−/−vs L-arg =1596, p <0.001). While, nos1−/−shoals spent more time shoaling and got high polarization. However, the polarization of nos1−/−shoal was no significantly higher than wild type shoal (W wt vs nos1−/−=738, p>0.05).

3.3.Nitric oxide distribution in brain issue of nos1−/−fish

nos1 is expressed in central and peripheral neurons in the brain. Its functions are synaptic plasticity in the central nervous system (CNS)(Forstermann & Sessa, 2011). The distribution of nitric oxide in nos1−/−fish is decreased compared with wild-type (Fig. 4). We concerned if this decline in nitric oxide might reflect increase the dopamine concentration by promote DAT mRNA expression in brain. As an inhibitory of monoamine transporter in synaptic shelf (Kiss et al., 2004).We used fluorescent probe to analyze the nitric oxide signaling throughout the brain tissue (Lim, 2007). We found that nitric oxide distribution in brain tissue are reduced (Fig. 4 nos1−/−).

Fig. 5.Measurement of shoaling behavior and cohesion. Box plots of the average nearest neighbor distance between all pairs offish. The vertical line within the boxes represents the mean, and box whiskers represents the standard deviation. Nearest neighbor distance within the shoal significantly decreased with the concentration of nitric oxide in the fish. The control group is wild type fish with tap water. Mean ±standard deviation is shown. N =10 shoals (each shoal consisted of 10fish, the experiments were repeated four times). Zebrafish were tested at 42 days post fertilization. N.S. =not significant, ＊ =p <0.05, ＊＊=p <0.01, ＊＊＊p <0.001, via student’s t-test.

Fig. 6.Expression of the gene for dopamine transporter in the brain. Real time PCR indicated an increase of DAT truncated mRNA in homozygous nos1 knockout (nos1 −/−) zebrafish when compared with the wild-type (control).The error bar is standard error of the mean (SEM). N.S. =not significant, ＊＊p <0.01, ＊＊＊p <0.001 via Mann-Whitney test.

Fig. 7.Dopamine levels in the brain. Box plots of mass spectrometry of the dopamine level in the brain (normalized to total brain protein). The vertical line within the boxes represents the median, and box whiskers represents the standard deviation. Dopamine levels were significantly increased in the homozygous nos1 knockout (nos1 −/−) zebrafish compared with wild type. N =25 for each group. N.S. =not significant, ＊＊ =p <0.01, ＊＊＊ =p <0.001 via Mann-Whitney test.

Fig. 8.Shoaling speed. Comparisons of swimming speed of each treatment at shoaling period. The error bar is standard error of the mean (SEM). (N.S. =not significant, ＊＊p <0.01, ＊＊＊p <0.001 via Mann-Whitney test.).

3.4.Dopaminergic pathway as a nitric oxide target

The most significant changes in expression are in dopamine transporter (DAT or slc6a3), the transporter that response the function of reuptake the dopamine neurotransmitter from synaptic cleft at synapses(Blackstone, 2009). Real time-PCR of this gene confirmed inhibitor by nitric oxide. The dopaminergic pathway was focused because nitric oxide has been previously reported to have inhibited effect on monoamine transporter (Motahari et al., 2016), besides dopamine levels have relate with shoaling behavior in fish (Buske, 2013). We found the nos1−/− zebrafish, which nitric oxide loss, induced DAT expression compared with control treatment (wild-type) (Fig. 6; W =45, p <0.01).In addition, the dopamine concentration is increased in nos1−/−fish(Fig. 7). However, we did not find the significant different of L-arg treatment of wild-type fish in DAT expression and dopamine level compared with control group (wild-type with tap water).

Fig. 9.Polarization in each shoal. Comparisons of polarization of each treatment at shoaling period. The error bar is standard error of the mean (SEM). (N.S. =not significant, ＊＊p <0.01, ＊＊＊p <0.001 via Mann-Whitney test.).

4.Discussions

In this paper, we demonstrate that modulation of NO levels via knockout of the nos1 gene has a strong effect on grouping behavior in zebrafish, where low levels of NO were associated with decreased nearest neighbor distances within a shoal. Although social behaviors are of scientific interest, the complex mechanisms underlying their neuronal basis are not well understood. For this reason, we aimed to investigate the relatively simple social behavior of shoaling in zebrafish, a well characterized organism. As it has been previously reported that NO influences swarming behavior in insects, this was our starting point to examine the neuronal basis of shoaling in zebrafish. Although some functional roles of NO and nos1 have been reported in many organisms such as mice, insects, and fish (Hou et al. 2017; Guti´errez et al. 2017.),there have been no reported behavioral studies of the potential relationship between NOS and NO on shoaling in fish.

Shoaling or shoal cohesion can be measured in different ways, such as inter-individual distance, nearest neighbor distance, distance from the closest wall or center, distance from a corner, speed, distance travelled, or time spent in a perimeter or particular zone (Buske, 2013).However, nearest neighbor distance—the distance between a focal fish and the shoal member closest to it (Fig. 1B)—is the measure most frequently reported for the quantification of shoal cohesion in fish, and is commonly used to index shoal cohesion because (Miller & Gerlai,2007). In addition, the average polarization distributions offish in shoal also was used to measure the shoaling cohesion. The highly polarization means fish spent more time shoaling and the weakly polarization means fish spent less time shoaling (Miller & Gerlai, 2012a).

We used Zebralab ViewPoint video recording to detect shoal cohesion and observed that shoal cohesion increases as nearest neighbor distance decreases. Our experiment applies to tracking fish in 2 dimensions. We test fish in shallow tanks because ecologically relevant of zebrafish which lives in small streams and shallow water area, besides shallow tank also makes biases the distribution offish in shoal to observable two dimensions. In addition, Miller and Gerlai reported the result from 3D tracking of shoals are similar to the results obtained with 2D tracking (Miller & Gerlai, 2012b).

Because L-arg can alter in vivo production of NO in the cell, we induced a high concentration of NO in the brains of zebrafish by incubating them with an L-arg solution. We also generated a nos1 knockout zebrafish line to create a population where NO production was inhibited. Wild type zebrafish (with a normal level of NO production) incubated in tap water without chlorine was the control group. NOS is the family of enzymes that can synthesis NO from arginine and plays a critical role in the brain. When the nos1 gene was knocked out in our zebrafish, NO signaling in the brain was found to be lower than in wild type fish (AB strain). Furthermore, we also found that functional loss of NOS in nos1 −/−fish can increase the expression of the DAT gene,which is involved in the dopaminergic pathway. When we measured dopamine concentration, we found that the nos1 −/−zebrafish that had high expression of the DAT gene also had a higher concentration of dopamine compared with wild type zebrafish. Finally, we found that zebrafish with a high level of dopamine were observed to also have high shoal cohesion. On the basis of these results, we suggest a revision of the hypothetical NO signaling pathway where NOS1 catalyzes arginine metabolism in the zebrafish brain to produce NO, which then inhibits DAT. This in turn modulates the dopaminergic pathway, which affects shoaling behavior in zebrafish.

NO is a highly diffusible gas that crosses biological membranes without difficulty, and has numerous molecular targets including aspects of neurotransmission (Forstermann & Sessa, 2011). It has been reported that L-arg can release dopamine and increase the concentration of NO (Motahari et al., 2016; Prast & Philippu, 2001; Wiesinger, 2001).Conversely, in our study, the dopamine concentration in zebrafish that were incubated with L-arg did not significantly differ from that of the control group. Additionally, treatment with L-arg did not affect DAT expression in the brain. However, behavioral testing did find that the distance between all pairs of all fish was significantly different between the group treated with L-arg and the control group. Together with another study that found that L-arg exposure promoted and increased ATP, ADP, and AMP hydrolysis (Capiotti et al. 2013), our findings suggest that L-arg has a direct effect on behavior processes.

NO can be synthesized from arginine by a family of enzymes known as nNOS or NOS1 that are expressed in neurons of the CNS. Genetic evidence indicates a role for NOS1 in synaptic transmission, immunohistochemistry in the spinal cord, and the process of neurogenesis.Moreover, NO produced by NOS1 can act as a neurotransmitter in effector cells and synaptic neurons (Forstermann & Sessa, 2011; Gally et al. 1990; Zhou & Zhu, 2009). It has previously been reported that NOS and NO were found to play a role in downstream mediation of neuropeptide F, and were able to regulate locomotor plasticity underlying locust phase transition where NO and NOS activity increased after the reduction of neuropeptide F during locust crowding. Additionally,increasing levels of NO were associated with high locomotor activity in locusts (Hou et al. 2017). However, it should be noted that NO signaling in phase transitions is species specific and is controlled by several other neurotransmitters. Hence, it is possible that modulation of NOS and NO is just one aspect of the neuronal control of the mechanism underlying crowding behavior. Conversely, our results show that nos1 −/−zebrafish, who had a low concentration of NO in the brain, exhibited higher shoal cohesion than wild type fish that had a high concentration of NO.Furthermore, it has been reported that complete loss of function of nos1 in both mice and zebrafish is associated with decreased aggression and hyperactivity; additionally, the reduction of NO signaling also decreased monoamine oxidase activity (Guti´errez et al. 2017). It has also been suggested that NO exerts an inhibitory effect on dopamine transporters(Kiss et al., 2004). It has previously been reported that NO is an important neuronal messenger that mediates the release of dopamine,and that inhibition of NOS can lead to a decrease in locomotor activity(Motahari et al., 2016; Snozek & Langman, 2019). Dopamine is well-known monoamine neurotransmitter that plays a central role in social behavior, pleasurable reward behavior, and learning (Ayano,2016). Therefore, in this study we compared the impact of altered NO signaling on social behavior and changes to neurobiology though the dopaminergic pathway. It has been reported that NO exerts an inhibitory effect on dopamine transporter (Kiss et al., 2004). Previous research has suggested that a genetic element plays a role in modulating the dopaminergic pathway that controls levels of dopamine, which in turn affects social behavior. In a previous report, when the DAT gene in zebrafish(slc6a3) was knocked out, the dopamine level decreased and was associated with altered swimming behavior (Wang et al., 2019). We therefore aimed to investigate the effect of molecular changes in nos1 −/−zebrafish using quantitative PCR. Our experiments were done using zebrafish 42 days post fertilization (the best shoaling age in zebrafish(Buske & Gerlai, 2011)), and found that altered concentration of NO in nos1 −/−zebrafish also affected expression of the DAT gene in the brain,while treatment with L-arg did not alter DAT expression. Thus, we suggest that the nos1 gene might play a role in the dopaminergic pathway, and that this pathway might have an effect on shoaling behavior. Moreover, this research can be effectively applied to study the shoaling mechanism and downstream function of nos1 in fish. Together,these data demonstrate that NO and NOS1 not only affect the neurochemical in zebrafish, but are also able to induce changes to shoaling behavior in fish.

In aquaculture, NO also induces the behavior that effect to stress in fish intensive condition such as aggressive and shoaling. Guti´errez reported that fish with low concentration of NO and nos1 loss can be reduce aggressive behavior (Guti´errez et al., 2017). Shelton found high shoal cohesion in shoaling effect on high density and group size offish(Shelton, Price, Ocasio, & Martins, 2015). From these two reasons, they cause to increase the numerous offish in intensive system. In addition,several researches reported the fish that living together in high degree of shoaling and schooling, has high ability and success in feeding more than individual living (Pitcher, 1993, Godin, 1986) Furthermore, Smith and Warburton reported that the trend of increasing shoal cohesion were consistent with a postulated increase in antipredator vigilance with declining feeding motivation (Smith & Warburton, 1992).

Further research will need to investigate expression localization of NOS in the brain because, in addition to nos1 which is found in central and peripheral neurons (including involvement with synaptic plasticity in the CNS), we also identified nos2 (another isoform of the enzyme NOS) expression in the zebrafish brain (Lepiller et al. 2009). In our study, transgenic DAT expression increased shoaling behavior in nos1−/−fish. However, we did not find an effect of L-arg treatment on transgenic slc6a3 expression and dopamine levels, suggesting that we may need to target NO concentrations more precisely. Alternatively, the NO substrate we used (L-arg) might only contribute partially to NO levels, so that NO donors (such as Sodium nitroprusside, PAPA- NONOate) should be used directly increase concentration of NO. It would be of interest to mutate slc6a3 to investigate if such mutation generates behavior akin to that seen in nos1−/−fish. In addition, there are other genes regulated by NO and nos1 that could be important to social behavior and should be investigated.

In conclusion, our findings suggest that shoaling behavior due to nos1 and NO signaling may be mediated through the dopaminergic pathway by inhibiting DAT because we found a lower level of NO was associated with a higher level of dopamine, which in turn leads to the shoaling behavior.

CRediT authorship contribution statement

Rachit Penglee: Conceptualization, Methodology, Software, Validation, Formal analysis, Resources, Data curation, Writing - original draft, Writing - review & editing, Visualization. Lei Gao: Conceptualization, Methodology, Software, Validation. Yajuan Huang: Validation,Resources. Liping Liu: Conceptualization, Investigation, Funding acquisition. Sukkrit Nimitkul: Conceptualization, Methodology. Baolong Bao: Conceptualization, Methodology, Supervision.

Acknowledgements

We thank Hao Wu for helping with a Waters Acquity™ UPLC H-Class System. We thank Pingfan Zhou and Dongkang Zhao for helping with confocal laser scanning microscope. We thank Sangon Biotech for sequencing. We thank Chunxin Fan and Xiaojie Wang for advice in behavior detection.

Appendix A.Supplementary data

Supplementary data related to this article can be found at htt ps://doi.org/10.1016/j.aaf.2020.08.007.

Declarations of interests

None.

Funding

This work was supported funding by The China-ASEAN Maritime Cooperation Fund through the project “China-ASEAN Center for Joint Research and Promotion of Marine Aquaculture Technology” (grant number: DF) and the Key Laboratory of Exploration and Utilization of Aquatic Genetic Resources of Shanghai Ocean University (grant number:31872546 and 31472262).

Wild-type (WT; AB strain) and homozygous nos1 knockout (nos1−/−) zebrafish were first incubated with (II) complex (CuFL; 1 μMfinal concentration) was added to all samples and incubated for another 40 min. The fluorescent signal was measured immediately after incubation.All pictures were taken with the same microscope setting using a confocal microscope from Leica Microsystems Inc (Shanghai, China).