Reproductive management: conditioning, spawning and development of Peruvian grunt Anisotremus scapularis in southern Peru

Capture and transport of A. scapularis

The fishing area included the rocky beaches of Llostay in Tacna, Peru (18°10′26″S, 70°38′37″W) (Fig. 1). Three fishing campaigns were carried out during the morning and afternoon on March 11 and 17 and on April 6 of 2016, in which 15 A. scapularis adults were captured (Fig. 2). The campaigns were carried out with fishing rods with BaitHolder hooks at 50 m offshore from the low tide line, because this capture method minimizes stress through careful handling and rapid transport, using Emerita analoga (Pacific sand crab) as fishing bait. The fishes were moved to the beach shore, the hooks were removed by expert fishermen, and the specimens were stabilized in a 50 L oval tub using jets of whipped seawater using a 20 L bucket and 2 L jugs. The weight of the specimens was recorded in situ with a digital scale (0.1 precision, Makita, La Mirada, CA, USA) and the size was measured with an ichthyometer (1 mm precision). The transference of specimens to the aquaculture centre comprised less than seven individuals per tank or 3 kg of live weight. The transports were driven from Llostay Beach to the Morro Sama aquaculture center (17°59′39.7″S, 70°52′59.1″) in a 1 m³ fiberglass tank filled to 500 L with raw seawater installed in a 4 × 4 truck. The oxygen supplied through a silicone hose ending in an air stone maintained the O₂ concentration saturated at (90–98)% in the transport tank. The temperature during the transport was maintained between 19 °C and 20 °C by means of 1 kg ice bags immersed in the tank.

Conditioning and feeding

At the Morro Sama aquaculture center, a preventive treatment with 99% oxytetracycline was applied at 50 ppm for 60 min in the transport tank to prevent infections on putative injuries caused by fishing/handling. The conditioning stage lasted seven days during which preventive antiparasite treatments were carried out with 37% formalin, i.e., 1 ml of formalin per 6 L of water (23.6 ppm). The open culture system consisted of an Australian-type circular conditioning tank (Fig. 3) made of corrugated steel plates covered with a black PVC geomembrane. Its volumetric capacity was 9 m³ (4 m in diameter and 1 m in height) without a filter system, water inlet, and outlet by constant pumping at 0.61 L/seg. Tank cleaning was performed by emptying and sweeping the tank twice a week (Tuesday and Friday) in the morning (8 am). Physicochemical parameters were recorded daily at 2:00 h, 8:00 h, 14:00 h and 20:00 h. Water parameters such as temperature, dissolved oxygen concentration and pH were measured with a thermometer, an oximeter (YSI model 550) and a pH-meter (ECOSENSE-pH100A pH-meter; YSI, Yellow Springs, OH, USA), respectively. Air inflow to the fish tank was regulated to maintain the dissolved oxygen concentration at 5.0–7.0 mg/L. The concentration of ammoniacal nitrogen (N-NH₃), ammonia (NH₃) and ammonium (NH₄) were determined monthly with a DR-400 HACH spectrophotometric kit (range 0.000–2.500 mg/L). Salinity was measured once a week with an RHS-28ATC refractometer.

Feeding the broodstock consisted of progressive diets based on the copepod E. analoga, fish meal, pellets and processed feedstuff. During the first post-treatment stage (38 days from March 18 to April 25, 2016), the frozen Pacific sand crab E. analoga was supplied daily ad libitum at a rate of 5% of total biomass. The subsequent diet was applied for 34 days (April 26 to May 24, 2016) and consisted of coating the E. analoga in fish meal, at a rate of 4.06% of total biomass. Starting in June 2016, the definitive diet consisted on an artisanal-made semi-moist feed (i.e., its proximal analysis gave 37.13% protein, 6.57% fat, 9.45% ashes, 0.73% fiber, and 44.32 humidity, (Montes et al., 2019)) delivered twice a day at a rate of 2.24% of live biomass and combined with whole frozen E. analoga two times per week, to make it attractive and palatable to the Peruvian grunt specimens. On October 4, 2016, an intramuscular individual identification chip (FELIXCAN-FDX-B) was implanted in the dorsal area of the specimens to facilitate individual monitoring of growth and spawning. The values of FCA, SGR and the percentages of growth and survival are presented, together with other data obtained in the different samples (Table 1).

First spawning and incubation of eggs

Only adult fishes above 500 g in weight were captured to fund an aquaculture broodstock. The authors used the biometric manipulations to gauge their sex and maturity status throughout abdominal massage. The eggs were systematically collected upon laying in a cylindrical collector of 300 μm planktonic mesh, with a circular ring at the top that fit into a rigid vertical frame of 0.3 × 0.3 × 0.6 m, placed at the outlet of the tank overflow (Fig. 4). The eggs were transferred to a 20 L bucket and transported to the incubation room where they were washed on a 300 μm sieve with sterilized seawater filtered at 1 μm in a cartridge filter and treated with UV (AL-PVC -160W). The eggs were then transferred to a 1 L test tube in sterile water for 10–15 min to separate the dead eggs precipitated at the bottom. Finally, the volumetric count of the floating eggs (fertilized and viable) and the precipitates (non-viable) was carried out. The viable laying fraction was immersed for 5 min in 4 L of a 1.5% Aqua-Yodo dilution at a rate of 2 mL/L of seawater. Subsequently, they were kept for 4 days in a black fiberglass tank of 500 L capacity with 400 L seawater (salinity range 34.7–35.6 psu and temperature 17.36 ± 1.54 °C), with a light supply of air through stone diffusers (air stone model ASI-3) at a rate of 0.1 L/min. This system allowed the uniform distribution of the eggs in the water column at an incubation density of 50 egg/L. The partial replacement of 30% seawater was carried out daily (9:00 a.m.) in the incubation tank using a 300 μm sieve. This protocol for egg management was carried out after each spontaneous spawning.

Egg sampling and morphological characterization of the embryo

In each egg sampling, the temperature and hydrogen potential were recorded. From the first day of incubation until hatching, a per-hour subsampling of 130 mL water was randomly taken to characterize the embryonic development according to Montes et al. (2019). Eggs were observed and photographed under a stereoscopic microscope (Japan Optical Co. Ltd. model XTL-2310, Shiga, Japan) and their characteristics were documented.

Statistical analysis

Handling of specimens required the use of eugenol (106 mg/L) as anesthesia. From April 2016 to October 2016, monthly biometrics were carried out with a 1 mm precision ichthyometer to gauge size growth of the broodstock. The weight was recorded on a 5 kg scale (Coretto EC-30). The distribution of these parameters was normal, and correlations (Pearson) were carried out with the statistical package R-studio version 1.4.1106 (R-studio, Inc., Boston, MA, USA), the ggplot2 package and the Excel package. The specific growth rate SGR was calculated according to Cook et al. (2000) using the following equation, either for the weight or the body length in terms of gain percentage per day.

Specific growth rate (SGR),

(1) $$SGR=(LfWf-LiWi/t)\times 100$$where SGR is the specific growth rate being Wf and Wi, the final and the initial fresh body weight (in grams), respectively, Lf, Li, the final and the initial body length in centimeters, respectively, and t the period between the initial and the final feeding period measured in days.

The viability rate was calculated for each spawning as,

(2) $$\mathrm{V}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(\mathrm{\%})=(\mathrm{F}\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{g}\mathrm{g}\mathrm{s}/\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{N}\mathrm{o}.\mathrm{e}\mathrm{g}\mathrm{g}\mathrm{s})\times 100.$$

The hatching rate was determined after 48 h of incubation as,

(3) $$\mathrm{H}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(\mathrm{\%})=(\mathrm{H}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{d}\mathrm{L}\mathrm{a}\mathrm{r}\mathrm{v}\mathrm{a}\mathrm{e}/\mathrm{F}\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{g}\mathrm{g}\mathrm{s})\times 100.$$

Ninety-six hours after the start of egg incubation, the number of larvae with an absorbed yolk sac was counted to determine larval survival as,

(4) $$\mathrm{S}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(\mathrm{\%})=(\mathrm{L}\mathrm{a}\mathrm{r}\mathrm{v}\mathrm{a}\mathrm{e}\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{y}\mathrm{o}\mathrm{l}\mathrm{k}\mathrm{s}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{e}\mathrm{d}/\mathrm{H}\mathrm{a}\mathrm{t}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{d}\mathrm{l}\mathrm{a}\mathrm{r}\mathrm{v}\mathrm{a}\mathrm{e})\times 100.$$

Capture and transport of A. scapularis

The seawater temperature ranged from 18 °C to 18.5 °C during the Peruvian grunt fishing campaigns in March and April 2016. The global survival rate of the captured fish in the three campaigns ranged from 0% to 46.7%. Total survival amounted to 73.33% (11 specimens) and mortality reached 26.67% (four specimens). The latter specimens did not survive due to physical damage caused by the fishing gear, which consisted of fatal injuries in mouth and gills. Eight males and three females were identified with lengths ranging from 30.0 to 44.7, with weights ranging from 0.598 to 2.293 kg. The density of fish transported from Llostay Beach to the Morro Sama aquaculture center ranged 4–7 fish/m³ and a maximum of 3 kg of live weight per tank. No correlation was observed between the number of fish transported per tank and survival, i.e., transport densities of 4 fish/m³ resulted in a survival rate of 26.7% but 7 fish/m³ resulted in a 46.67% survival rate.

Conditioning and feeding

The mean temperature and mean dissolved oxygen in the conditioning tank between April 2016 and May 2017 were 17.36 ± 1.54 °C and 5.96 ± 1.03 mg/L, respectively (Fig. 5).

The chemical parameters showed concentrations of 17 ± 0.38 μg/L ammoniacal nitrogen (N-NH₃), 1.21 ± 0.02 μg/L ammonia (NH₃) and 1.29 ± 0.03 μg/L ammonium (NH₄). The salinity ranged 34.7–35.6 psu (Fig. 6) and the pH (± SD) was 7.29 ± 0.63.

The 11 live specimens of Peruvian grunt conditioned as broodstock accepted the preventive treatment as well as the nutrition program ending by semi-moist balanced food with 37% protein at a feeding rate of 2.24% combined with the copepod E. analoga. The increase in average weight of the spawners biomass in the period between April and September 2016 was 539 ± 411 g/month, their average individual weight gain was 0.049 ± 0.037 g/month, and the average individual length increase was 0.444 ± 0.202 cm/month (Table 2). The correlation of weight-length increments per month was significant and positive (Pearson, r = 0.968, two-sided p = 0.001) (Fig. 7).

First spawning of A. scapularis broodstock

After 7 months under indoor conditioning (from October 2016 to May 2017), the broodstock spawned daily between 5:00 p.m. and 7:00 p.m. in a temperature range of 18–20 °C, natural light intensity of 335.85 lux and water flow rate of 0.61 L/s. Sex determination was affordable during biometric controls when the abdominal massage allowed to verify the sex of the expelled gametes. Eight and three out of 11 grunt specimens were males and females, respectively, and the sex ratio in the reproduction tank was ~3:1 male:female.

The dynamics of spontaneous spawning of A. scapularis in captivity from October 2016 to May 2017 showed two main peaks, in December–January and in March (Fig. 8).

Metrics for the spawning activity are given in Table 3. The survival rate indicates the percentage of viable eggs collected within 24 h post spawning (hps). Egg hatching (larvae with yolk sac) begun at 48 hps, the hatching rate was confirmed at 72 hps and larvae survival (larvae with or without the yolk sac consumed) was scored at 96 hps. All hatching parameters were significantly lower in October–November 2016 (beginning of spawning) and in February (middle spawning season). Hatching rate and larvae survival rate did not differ significantly throughout the spawning season (Table 3).

Embryological development of A. scapularis

The average diameter of Peruvian grunt oocytes was 735 ± 0.113 µm, that of fertile eggs was 800 ± 0.207 µm and the hatched larvae had 1.87 ± 0.05 mm in length. The main phases of the embryo development were completed at 42 hpf (hours post-fertilization) (Table 4).

The viable eggs were spherical, non-adherent and telolecithal type, i.e., they exhibited a large amount of translucent yolk at the vegetal pole and a discoidal meroblastic division during the first cleavages (Fig. 9). At 0.35–0.45 hpf the zygote was translucent, contained oily droplets in its center and presented a conspicuous accumulation of cytoplasm towards the animal pole to form the blastodisc (Fig. 9A). Between 1.00–2.32 hpf, the blastodisc giving rise to the first blastomere had already formed. The division of the blastomeres was symmetrical, the daughter cells were like each other in size (second cleavage). The development continued with the loss of symmetry in the blastomeres and their progressive decrease in size with further divisions (third cleavage). As in the previous step, there was no difference in cell shape and size at the 16 blastomeres stage (fourth and fifth cleavages; Fig. 9B). Between 4.55–5.60 hpf distal cells were smaller than their adjacent ones. A morula was formed with two cell layers. There was a large association of small cells from irregular to circular shape and flattened dorso-ventrally, concentrated in a single cell pole (Fig. 9C).

Between 13.0–14.0 hpf, the blastopore as origin of the blastocele was observed. The cell movement (epiboly) of blastomeres and the extension of the cell mass through the egg in the direction of the equatorial pole begun. The peripheral dendriform cells adopted the shape of a translucent rim (Fig. 9D). Between 15.55–18.50 hpf the blastoderm had covered approximately 1/4 to 1/3 of the yolk sac, the area where the neural groove appears was already identifiable and the germ ring begun to form (Fig. 9E). Between 20.26–40.0 hpf, the cephalic region was differentiated from the caudal region, and was still flattened dorso-ventrally due to the presence of the ocular buds. The cephalic region expanded laterally, and the optic lenses and optic vesicles were distinguished. A two-chambered beating heart was observed, and the closure of the blastopore gave rise to the Kuffper vesicle. The growth of the tail was observed in the perivitelline space, and the size of the embryo exceeded the amplitude of the yolk sac. The tip of the tail was levelled with the eyes and the thickening of the body was observed in a dorso-ventral direction. The somites were well defined, and the embryo begun moving and a duct was observed between the notochord and the heart, which corresponded to the digestive tract. In the caudal terminal region, a narrowness corresponding to the caudal pedunculus was observed and there was a marked development of the face (Fig. 9F). Organogenesis was completed between 41.50–42.15 hpf, the chorion began to tear, and an incipient ocular pigmentation was observed (Fig. 9G). Between 42.15–42.20 hpf, the embryo performed a series of spiral contortions and the chorionic membrane dissolved in less than 5 min (Fig. 9H). About 42.25 hpf, a newly hatched larvae was observed with a single embryonic fin which extended from the cephalic region to the rostral region. The yolk sac covered half the total length of the larva, and a closed, non-functional mouth was observed (Fig. 9I).