The fat-soluble vitamin D, which comprises a group of secosterols found naturally in few foods, is an essential nutrient for animal growth and physiological metabolism(Reference Lu, Chen and Zhang1). Vitamin D consists of five different types: D1, D2, D3, D4 and D5. Among these, the two major forms are vitamin D2 (ergocalciferol) and vitamin D3 (cholecalciferol). Vitamin D2 is produced in plants and fungi, while vitamin D3 is produced in animals. Generally, dietary vitamin D3 supplementation in the diets can be used to maintain adequate health in aquatic animals. The classical action of vitamin D3 is to maintain Ca and P homoeostasis by enhancing the absorption ability from the intestine(Reference Wen, Liu and Tian2,Reference DeLuca3) . In recent years, additional diverse physiological function of vitamin D3, including its effects on immune response, antioxidant capacity and lipid metabolism, has gradually been explored(Reference Bikle4–Reference Zügel, Steinmeyer and May7).
Contrary to mammals, which are able to produce vitamin D3 in the skin after exposure to solar UV-B radiation, fish and crustacean cannot synthesise vitamin D de novo and the diet is the only vitamin D source(Reference Holick8,Reference Lambert9) . Currently, the research regarding the influence of vitamin D3 on the growth and physiological metabolism in aquatic animals is limited. Previous studies mainly focus on the optimum dietary vitamin D3 requirement for various fishes and found that vitamin D3 requirement varies from different species(Reference Lock, WaagbØ and Wendelaar Bonga10,Reference Ling-Hong, Xian-Ping and Jun11) . A few researches have demonstrated the influence of vitamin D3 on the antioxidant and immune capacity in pearl oyster Pinctada fucata martensii, Atlantic Salmon (Salmo salar) and yellow catfish (Pelteobagrus fulvidraco)(Reference Yang, Du and Hao12–Reference Soto-Davila, Valderrama and Inkpen14). It has been found that vitamin D3 is also closely connected with regulating lipid metabolism in visceral adipose tissue of zebrafish(Reference Peng, Shang and Wang6). In contrast, high dosage of vitamin D3 could cause metabolic disorders and induce chronic stress effects in aquatic animals(Reference Yang, Du and Hao12,Reference Miao, Xie and Ge15) . Therefore, optimum vitamin D3 supplementation level in the formulated diets should be carefully evaluated and determined, especially for unstudied aquaculture species. To the best of our knowledge, exploring vitamin D3 optimal requirement and physiological function in crustaceans was currently only reported in Penaeus monodon (Reference Shiau and Wang16), Eriocheir sinensis (Reference Liu, Wang and Bu17) and Litopenaeus vannamei (Reference Wen, Liu and Tian2).
Pacific white shrimp L. vannamei is one of the most important economic marine shrimp, which makes the shrimp aquaculture become the fastest-growing industry worldwide(Reference Amaya, Davis and Rouse18,Reference Alam19) . Currently, researches reporting the effects of dietary vitamin D3 inclusion on growth and physiological metabolism in L.vannamei remain limited. Therefore, this study aims to evaluate the dietary vitamin D3 requirement in L.vannamei based on growth performance, and further compare the specific effects of different vitamin D3 supplementation levels on the tissue Ca/P concentrations, antioxidant capacity, non-specific immune response and lipid metabolism. This study would contribute to a in-depth understanding of vitamin D3’s physiological function in L.vannamei.
Materials and methods
Ethics statement
All experimental procedures complied with the Standard Operation Procedures (SOP) of the Guide for Use of Experimental Animals of Ningbo University.
Animals and experimental procedure
Our experiment was carried out in the Ningbo Marine Fishery Science and Technology Innovation Base. Healthy shrimp post-larvae were raised in aerated semi-intensive pond at room temperature (28 ± 2°C), salinity (21·6–23·5), pH (7·6–7·9) and dissolved oxygen (4·29–5·8 mg/l).
A total of 720 shrimp (initial weight 0·50 ± 0·01 g) were randomly distributed into six treatments, each of which had three tanks (300-litre cylindrical fibreglass tanks filled with 200 litre of water) of forty shrimp per tank. We firstly counted twenty shrimp and weighed, and then selected remaining twenty shrimp to make the final total weight of about 20 g. The six diets (45·71 % crude protein and 7·36 % crude lipid) were formulated to contain different vitamin D3 levels: the control group was fed the basal diet without extra vitamin D3 supplementation, and the other five treatment groups were supplemented with 0·05, 0·10, 0·30, 0·40, 0·80 mg/kg of vitamin D3, respectively. Vitamin D3 was purchased from Guangzhou Chengyi Aquatic Products Technology Co., Ltd. The vitamin D3 content in each treatment group was measured in Guangzhou Analysis and Test Center in China (GB/T 17818-2010-1). The analysed vitamin D3 supplementations levels in each diet were 0·18, 0·23, 0·27, 0·48, 0·57 and 0·98 mg/kg, respectively. The ingredient and composition of the basal diet are shown in Supplementary Table S1. Shrimp were fed manually three times a day (8–10 % of body weight) at 06.00 h (35 % of the diets), 12.00 h (30 % of the diets) and 18.00 h (35 % of the diets). Daily amount of formulated feed was adjusted every 2 weeks according to the weight of shrimp in each treatment. Dead shrimp were removed, weighed and recorded immediately, and above 60 % of seawater in each tank was exchanged before the first feeding every morning.
Sample collection
Following the 8-week feeding period, all the shrimp in each tank were taken out, counted and weighed to obtain the final body weight (FBW). The data were further used to determine the specific growth rate (SGR), percent weight gain (PWG), survival rate and feed efficiency (FE).
PWG, % = (final body weight (g) − initial body weight (g))/initial body weight (g) × 100;
SGR, % day−1 = (Ln (final body weight) − Ln (initial body weight)) × 100/days;
FE = weight gain (g, wet weight)/feed consumed (g, dry weight).
Ten shrimp per tank were sampled and pooled to make one sample per tank, and therefore three pooled samples per treatment. Haemolymph samples were collected from the pericardial cavity and placed into 1·5-ml centrifuge tubes overnight at 4°C before centrifugation (1811 g, 10 min). The supernatant was collected for anayzing Ca/P concentrations, antioxidant and immune enzyme activity. Muscle and carapace were collected for measuring Ca/P concentrations. Hepatopancreas of the remaining shrimp was harvest for measuring Ca/P concentrations, antioxidant and immune enzyme activity and related gene expression.
Experimental parameters measured
Determination of tissue calcium and phosphorus concentrations
Tissue samples were weighed, freeze-dried and then digested in 70 % HNO3 solution at 80°C. Drops of HNO3 solution were added until the liquid became clear and bright. Ca/P concentrations in serum, carapace, hepatopancreas and muscle of L. vannamei were determined by diagnostic reagent kits (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer’s instructions.
Analysis of antioxidant and non-specific immune enzymes activities
Hepatopancreas samples were homogenised on ice in 0·9 % NaCl solution, and the supernatant was removed after centrifugation (1811 g for 10 min at 4°C). The supernatant of haemolymph and hepatopancreas was stored at −80°C until further analysis.
The activities of catalase (CAT), superoxide dismutase (SOD) and total antioxidative capacity (T-AOC), alkaline phosphatase (ALP), acid phosphatase (ACP), nitric oxide synthase (NOS), malondialdehyde (MDA), phenoloxidase (PO) in serum and hepatopancreas were measured using diagnostic reagent kits (Nanjing Jiancheng Bioengineering Institute) according to manufacturer’s instructions.
Gene expression analysis
Total RNA was extracted from samples using TRIzol reagent (Vazyme Biotech co., Ltd) and subsequently measured its quality using a Nano-Drop ND-2000 spectrophotometer (Nano-Drop Technologies). The cDNA was synthesised using the Primer-ScriptTM One Step RT-PCR Kit (TaKaRa Biotechnology). The primer sequences were designed using the Primer Premier 5.0 software and listed in the Supplemental Table S2. The target genes were as follows: glutathione (gsh), cat and sod were antioxidant-related genes. Lysozyme (lzm), alp, acp, tnf-α, apoptosis-inducing factor (aif) and member RAS oncogene family (rab6a) were immune-related genes. Sterol-regulatory element binding protein (srebp), acetyl-CoA carboxylase 1 (acc1), fatty acid synthetase gene (fas) fatty acid transport proteins (fatp), fatty acid binding protein (fabp), carnitine palmitoyltransferase1 (cpt) and acyl-CoA oxidase (aco) were lipid metabolism-related genes.
Quantitative real-time PCR (qPCR) was carried out on a StepOne Plus Real-Time PCR system (Applied Biosystems) using a SYBR Premix Ex Taq II kit (Vazyme Biotech co., Ltd). The amplifications were carried out in a 20-μl reaction mixture, which consisted of 10 μl of 2 × SYBR Green I Master Mix, 1·0 μl each of the forward and reverse gene-specific primers (10 μM), 2 μl of cDNA, and 6 μl of DEPC water. The PCR program was conducted at an initial denaturation step at 95°C for 2 min, followed by forty-five cycles of 95°C for 10 s, 58°C for 10 s and 72°C for 20 s. The relative mRNA expression was calculated by the 2−ΔΔCt method. β-actin is used as an internal control. Three independent biological replicates were performed for each sample.
Statistical analysis
Statistical analysis was conducted using IBM SPSS statistics 23.0 (SPSS Inc.). Data presented are mean and standard error. The values of growth performance, tissue Ca/P concentrations, antioxidant enzyme activity, non-specific immune enzymes activity and gene expression among different vitamin D3 levels were analysed with normality and homoscedasticity tests. Subsequently, one-way ANOVA was applied when the statistical assumptions were fulfilled. Duncan’s multiple range test was conducted to compare the differences. Otherwise, Kruskal–Wallis test was applied. All tests were carried out at α = 0·05 confidence level. Taking PWG as the evaluation index, quadratic regression analysis was used to determine the optimal vitamin D3 requirement. Effects were considered significant at P < 0·05.
Results
Growth performance and feed utilisation
As shown in Table 1, FW, PWG and SGR showed a trend of increasing first and then decreasing. 0·48 mg/kg of vitamin D3 supplementation group achieved the highest value (P < 0·05) of FW, PWG, FE and SGR. Survival rate was not affected (P > 0·05) by different vitamin D3 supplementation levels in L. vannamei. Quadratic regression analysis showed that the optimum dietary vitamin D3 requirement based on PWG was determined to be 0·49 mg/kg (Fig. 1).
IW, initial weight; FBW, final body weight; PWG, percent weight gain; SGR, special growth rate; FE, feed efficiency.
Mean in the same row with different superscripts are significantly different (P < 0·05).
*PWG, % = (final body weight (g) − initial body weight (g))/initial body weight (g) × 100.
† SGR, % day−1 = (Ln (final body weight) − Ln (initial body weight)) × 100/d.
‡ FE = weight gain (g, wet weight)/feed consumed (g, dry weight).
Tissue calcium and phosphorus concentrations
As shown in Fig. 2, when dietary vitamin D3 supplemental level exceeded 0·48 mg/kg, P concentration was significantly increased (P < 0·05) in hepatopancreas and carapace of L. vannamei, while Ca concentration was significantly increased (P < 0·05) in carapace of L. vannamei. Ca concentration was not affected (P > 0·05) by different vitamin D3 supplementation levels in serum, hepatopancreas and muscle of L. vannamei.
Antioxidant enzyme activity
As shown in Fig. 3, the activity of T-AOC and CAT in serum and hepatopancreas showed a trend of increasing first and then decreasing while hepatopancreas SOD activity had the same trend. The above enzyme activity reached the highest value (P < 0·05) in the 0·48 mg/kg of vitamin D3 supplementation group. MDA content was not affected (P > 0·05) by different vitamin D3 supplementation levels in serum and hepatopancreas of L. vannamei.
Non-specific immune enzymes activity
As shown in Fig. 4, the activity of ALP, ACP in serum and hepatopancreas, and NOS in hepatopancreas showed a trend of increasing first and then decreasing. The above enzyme activity reached the highest value (P < 0·05) in the 0·48 mg/kg vitamin D3 supplementation group. Serum NOS activity reached the highest value (P < 0·05) in the 0·98 mg/kg vitamin D3 supplementation group.
Expression of antioxidant- and immune-related genes
As shown in Fig. 5(a), antioxidant-related genes (sod, gsh and cat) showed a trend of increasing first and then decreasing. The above antioxidant-related gene expression reached the highest value (P < 0·05) in the 0·48 mg/kg of vitamin D3 supplementation group.
As shown in Fig. 5(b), expression levels of alp, acp and lzm genes showed a trend of increasing first and then decreasing. Expression of aif gene showed a trend of decreasing first and then increasing. The expression levels of alp, acp and lzm genes reached the highest value (P < 0·05), while aif expression reached the lowest value (P < 0·05) in the 0·48 mg/kg of vitamin D3 supplementation group. Gene expression of tnf-α and rab 6 a was not affected (P > 0·05) by different vitamin D3 supplementation levels in hepatopancreas of L. vannamei.
Expression of lipid metabolism-related genes in hepatopancreas
As shown in Fig. 6, expression levels of srebp, acc1 and fas genes showed a trend of increasing first and then decreasing. The above gene expression reached the highest value (P < 0·05) in the 0·48 mg/kg of vitamin D3 supplementation group. Compared with the control group, supplementing 0·98 mg/kg of vitamin D3 significantly increased (P < 0·05) the expression of cpt1, aco, fabp and fatp genes in the hepatopancreas of L. vannamei, while supplementing 0·23 mg/kg, 0·27 mg/kg, 0·48 mg/kg and 0·57 mg/kg of vitamin D3 had no effects (P > 0·05) on the gene expression of srebp, acc1 and fas in hepatopancreas of L. vannamei.
Discussion
The present study clearly demonstrated that the growth performance of L. vannamei was significantly influenced by the dietary vitamin D3 supplementation levels, and shrimp feeding 0·48 mg/kg of vitamin D3 diet achieved the best growth performance. Further quadratic regression analysis showed that the optimum dietary vitamin D3 requirement based on PWG was determined to be about 0·49 mg/kg. In a series of early work on vitamin D3 nutrition showed that dietary essentiality of vitamin D3 to maintain normal growth and a diet supplied with appropriate amount of vitamin D3 could exert a positive effect on the growth of aquatic animals. In general, feed nutrition levels (purified diet or commercial feed), a consequence of a faulty managed feed, species and environmental condition might be partially responsible for the difference in optimal vitamin D3 requirement in aquatic animals. Lock et al. (2010) summarised the minimum dietary requirement for vitamin D3 in fish, including Monopterus albus (0·125 mg/kg), Salmo gairdneri (0·04 mg/kg), Ictalurus punctatus (0·006 mg/kg), Salmo salar (0·06 mg/kg) and Oreochromis niloticus × O. aureus (0·009 mg/kg)(Reference Lock, WaagbØ and Wendelaar Bonga10). As for crustacean species, the dietary vitamin D3 requirement of juvenile grass shrimp (Penaeus monodon) was 0·1 mg/kg (purified diets)(Reference Shiau and Wang16). The optimal vitamin D3 requirement for larval crabs (Eriocheir sinensis) was 4825–5918 μg/kg (0·12 mg/kg−0·15 mg/kg), and crabs fed 9000 μg/kg (0·225 mg/kg) showed the highest survival rate after 120-h salinity stress(Reference Liu, Wang and Bu17). Moreover, growth performance was not affected by the supplementation of vitamin D3 (0·017 mg–0·081 mg/kg) for juvenile L. vannamei at low salinity rearing conditions (10–15 g/l)(Reference Wen, Liu and Tian2). Interestingly, our result demonstrated that optimal vitamin D3 requirement for L. vannamei was higher than other aquatic animals. The reason for the differences in vitamin D3 requirements may be partially related to species, feed nutrition levels (protein level, lipid level, etc), initial weight and breeding density. However, the specific molecular mechanism is unknown at present, which needs to be further explored.
The primary biological function of vitamin D3 is to maintain normal Ca and P homoeostasis by enhancing the absorption ability from the intestine(Reference Darias, Mazurais and Koumoundouros20,Reference Darias, Mazurais and Koumoundouros21) . Currently, the research regarding the influence of vitamin D3 on the tissue Ca and P deposition in crustaceans is limited. The body Ca and P deposition was not affected by dietary vitamin D supplementation (0·1–62·5 mg/kg ergocalciferol or cholecalciferol) in juvenile grass shrimp (Penaeus monodon)(Reference Shiau and Wang16). In our experiment, carapace Ca and P concentrations increased with increasing vitamin D3 supplementation levels (0·48–0·98 mg/kg), which confirmed vitamin D3 function in Ca and P homoeostasis of L. vannamei. Cal and P absorption is mediated by both an active transcellular pathway and a passive paracellular pathway through tight junctions. 1,25(OH)2D, the hormonally active form of vitamin D, increases intestinal transcellular Ca and P absorption at least in part by enhancing expression of cotransporter(Reference Christakos, Dhawan and Porta22,Reference Fukumoto23) . Similar to our study, Coloso et al. (2001) reported that the dietary combination of low P (0·3 %) of 10 000 μg/kg (0·25 mg/kg) vitamin D3 decreased soluble and faecal P levels in the effluent of rainbow trout Oncorhynchus mykiss (Reference Coloso, Basantes and King24). Sundell et al. (1990) suggested that 25(OH)D3 and 24,25(OH)2D3 that might be active regulators of Ca2+ transport across the intestinal mucosa were detected in Atlantic cod (Reference Sundell and Björnsson25). Miao et al. (2015) also found that whole body Ca content was increased with different dietary vitamin D3 supplementation levels (0–0·2 mg/kg) in Wuchang bream (Megalobrama amblycephala)(Reference Miao, Xie and Ge15).
It is now clear that vitamin D3 has additional physiological function beyond its classic effect on Ca and P absorption. It has been suggested that the active form of vitamin D3 may be shown as a membrane antioxidant. Our study showed that supplementing 0·48 mg/kg of vitamin D3 significantly increased the activities of T-AOC and CAT in serum and hepatopancreas of L. vannamei. Meanwhile, the expression levels of antioxidant-related genes (sod, gsh and cat) were increased in hepatopancreas of L. vannamei. Antioxidant enzymes such as SOD, GSH and CAT participated in delaying or preventing the oxidation of cellular oxidisable substrates, and therefore protected cells against oxidative damage(Reference Mates26,Reference Zhong, Gu and Gu27) . This finding suggested that dietary supplementing 0·48 mg/kg of vitamin D3 could improve antioxidant capacity in L. vannamei. It has been reported that vitamin D3 protected cell membranes against free radical-induced lipid peroxidation through interaction with phospholipid fatty acid side chains, then increased stabilisation of the membrane structure(Reference Boakye, Jansen and Schöttker28,Reference Wolden-Kirk, Gysemans and Verstuyf29) . Mechanistic study demonstrated that the antioxidant role of vitamin D3 was vitamin D receptor (VDR)-mediated transcriptional down-regulation of NOX2, a major isoform of NADPH oxidase, and up-regulation of Nrf2-keap1-mediated antioxidant pathways(Reference Kim, Perrelli and Ragni30). Similarly, some reports have demonstrated that dietary vitamin D3 could improve antioxidant capacity in sea cucumber and pearl oyster(Reference Yang, Du and Hao12,Reference Wang, Baoshan and Wang31) . Liu et al. (2021) reported that AOC and GSH activities were higher in the hepatopancreas of Chinese mitten crab (Eriocheir sinensis) fed with 6000 μg/kg (0·15 mg/kg) vitamin D3 when compared with the control group (without extra vitamin D3 supplementation)(Reference Liu, Wang and Bu17).
Increasing evidence suggested that vitamin D3 could modulate the innate and adaptive immune responses and has been used to treat various infections before the advent of effective antibiotics(Reference White32). It has been reported that vitamin D3 could also affect the immune function in some fish species(Reference Cerezuela, Cuesta and Meseguer33). Like other invertebrates, non-specific immunity is shrimp’s main defence against pathogens(Reference Li and Xiang34,Reference Jia, Wang and Lu35) . LZM, ALP, NOS, ACP and PO are identified as important immune indices of shrimp. NO, which is produced from l-arginine catalysed by the enzyme NOS, has be shown to be beneficial for defensing against pathogens(Reference Chakravortty and Hensel36). LZM has the capacity to hydrolyse bacterial cell walls that simultaneously regulate the synthesis and secretion of other immune factors, while ACP and ALP are important parts of lysosomal enzymes of crustaceans(Reference Liu, Zhang and Feng37). The major enzyme produced during proPO system activation is phenoloxidase, which oxidises phenolic compounds to produce quinones and help to kill pathogens(Reference Cerenius and Soderhall38). In the present study, we observed that both the activity of ALP, ACP, NOS and immune-related genes (alp, acp and lzm) expression in hepatopancreas increased and then decreased with increasing vitamin D3 supplementation levels, while numeric value reached the highest in the 0·48 vitmain D3 mg/kg supplementation level. This result seems to show that dietary vitmain D3 at 0·48 mg/kg supplementation level exerted the best immune function in L. vannamei. Recent research has demonstrated that vitamin D3 could exert its immunomodulatory actions via down-regulating inflammation-mediated signalling pathway (ROCK/NF-κB, JAK and STAT) and acitivating an anti-inflammatory action (up-regulation of the anti-inflammatory toll-like receptor)(Reference Kim, Perrelli and Ragni30). Similarly, Dioguardi et al. (2017) confirmed that a stimulation of phagocytosis and peroxidase activity of serum was observed in European sea bass (Dicentrarchus labrax L.) fed with vitamin D3 (0·09 mg/kg–0·9 mg/kg) diets(Reference Dioguardi, Guardiola and Vazzana39). Liu et al. (2021) observed that the mRNA expression of LZM was upregulated with 9000 μg/kg (0·225 mg/kg) vitamin D3 after 23-d feeding trial, suggesting that dietary vitamin D3 could enhance immunity in E. sinensis (Reference Liu, Wang and Bu17). Shiau and Hwang (1994) reported that adding 0·2 mg/kg of vitamin D3 significantly increased ALP activity in juvenile grass shrimp (Penaeus monodon)(Reference Shiau and Wang16).
Epidemiologically, the function of vitamin D is also related to lipid metabolism. Certain studies have suggested the vitamin D3 negatively regulated the expression of various lipogenic genes in both the adipose tissue and liver in mammals(Reference Kang, Lee and An40,Reference Lee, Lee and Lim41) . Until now, the effects of vitamin D on lipid metabolism have not been studied much in aquatic animals. One recent paper demonstrated that vitamin D3 deficiency induced retarded growth and excessive visceral adipose tissue in zebrafish using Cyp2r1 gene knockout model(Reference Peng, Shang and Wang6). In our experiment, we found that supplementing different vitamin D3 levels could influence hepatopancreas lipid metabolism. Expression of adipogenesis-related genes (srebp, acc1 and fas) increased and then decreased in hepatopancreas of L. vannamei. Supplementing 0·48 mg/kg of vitamin D3 achieved highest expression of lipid synthesis-related genes, which might benefit the improved growth performance. In addition, the expression of various lipolysis genes (cpt1, fabp, fatp and aco) showed an upward trend with the increased dietary vitamin D3 supplemental levels, and supplementing 0·98 mg/kg of vitamin D3 significantly increased the expression of lipolysis genes in hepatopancreas of L. vannamei. This result indicated that vitamin D3 might modulate lipolysis with increased supplementation levels. Previous experimental studies in mammals might support our guess. Vitamin D can control energy metabolism in adipose tissue by affecting fatty acid oxidation, expression of uncoupling proteins, insulin resistance and adipokine production(Reference Chan and Han42). This function of vitamin D3 has be supported by previously published experiments on VDR knockout and overexpression models. It has been reported that 0·375 mg/kg of vitamin D3 could protect against diet-induced obesity possibly by up-regulating genes involved in fatty acid oxidation and mitochondrial metabolism which led to increased energy expenditure(Reference Marcotorchino, Tourniaire and Astier43). Similarly, intraperitoneal injection of vitamin D3 (1–5 μg/kg) could prevent high-fat diet-induced hepatic steatosis through the inhibition of lipogenesis and the promotion of fat acid oxidation in liver(Reference Yin, Yu and Xia44).
Conclusion
The results of present study indicated that dietary supplementation of 0·48 mg/kg of vitamin D3 could increase Ca and P concentraions, improve antioxidant capacity and immune response, and influence lipid metabolism in L.vannamei.
Acknowledgement
This research was supported by National Key R & D Program of China (2018YFD0900400), Ningbo Public Welfare Science and Technology Project (202002N3041), Industrial Chain Collaborative Innovation Project of the Demonstration Work on Innovative Development of the Marine Economy of the State Oceanic Administration (NBHY-2017-S2), K. C. Wong Magna Fund in Ningbo University.
TM. Dai conducted feeding experiments and drafted the article; LF. J designed this experiment, supervised project administration and revised the article; XY. T and JJ. L assisted in sampling and data analysis. M. J and P. S helped reviewing. QC. Z acquisited funding, supervised project administration, and gave constructive advice and valuable help during paper review.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114521004931