Larval digestive system ontogeny and early weaning in neon tetra Paracheirodon innesi

doi:10.21203/rs.3.rs-1682680/v1

Download PDF

Research Article

Larval digestive system ontogeny and early weaning in neon tetra Paracheirodon innesi

https://doi.org/10.21203/rs.3.rs-1682680/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 23 Oct, 2023

Read the published version in Fish Physiology and Biochemistry →

You are reading this latest preprint version

The intensive culture of characid teleosts for ornamental trade is highly dependent on live feed organisms, particularly Artemia nauplii, to provide nutrition through the larval stage. Live feeds have inherent disadvantages relative to prepared microparticulate diets (MDs), specifically availability, labor and cost. In this research, the dependence of larval Paracheirodon innesi on live Artemia was confirmed via a dietetics trial. Next, digestive system ontogeny was characterized from the onset of exogenous feeding through metamorphosis. P. innesi exhibited an agastric larval stage, as well as low digestive enzyme activity at the onset of exogenous feeding followed by abrupt increases in trypsin, lipase and pepsin activity. Differentiation of the stomach, including gastric gland formation and production of neutral mucopolysaccharides, as well as the onset of pepsin activity, did not occur until 20 days post hatch (dph). This shift from agastric to gastric digestive modes is indicative of a proliferation of digestive capacity and subsequent prey diversity in other fish species exhibiting similar altricial larval stages.

Based on this information, different schedules for weaning from Artemia to a MD were evaluated. For P. innesi fed until 32 dph, weaning beginning at 12 dph and 17 dph resulted in similar survival to live Artemia (mean: 22.0 ± 1.7%), and the MD resulted in the lowest survival (0.8 ± 0.3%). These results indicate that weaning is possible prior to gastric differentiation, potentially resulting in the reduction of Artemia use in the larval culture P. innesi.

Characidae

larval nutrition

weaning

gastrointestinal development

The neon tetra Paracheirodon innesi is a small, neotropical characid that is one of the most highly traded fish in the ornamental industry (Weitzman et al. 1983; Chapman et al. 1997, 1998). While an estimated 80% of the market demand, or 12-15 million individuals annually, is satisfied via import from wild collections originating in their native range, foreign and domestic aquaculture of P. innesi represents an economically significant endeavor with the potential for expansion (Chapman et al. 1998; Chao 2001; Kucharczyk et al. 2010). P. innesi, like many other species in the ornamental aquaculture trade, are characterized by having very small, poorly developed larvae at first feeding, which has led to the ubiquitous use of live zooplankters as first feed items (Lim et al. 2003; Sanaye et al. 2014). While rotifers (Brachionus spp.), Paramecium spp., and other live feed organisms have been used during the critical larval stage of P. innesi culture, newly hatched Artemia franciscana nauplii are the most widely used for this application (Chapman et al. 1998; Lim et al. 2003; Sanaye et al. 2014).

Commercial producers of freshwater ornamental fish rely heavily on newly-hatched 1-2 instar stage A. franciscana for early larval nutrition in intensive and semi-intensive aquaculture operations (Lim et al. 2003). Artemia cysts are collected from natural brine lakes, and ,in the USA, primarily from the Great Salt Lake in Utah, where cyst formation and subsequent collections are dependent on fluctuations in salinity following environmental perturbations (Sorgeloos and Kulasekarapandian 1984; Lavens and Sorgeloos 2000; Kolkovski 2001). Cysts are then packaged and distributed on a commercial scale to various aquaculture sectors, with penaeid shrimp culture comprising the bulk of the market (Kolkovski 2001). This wild caught feed item, while convenient and nutritious relative to some other live feeds, is characterized by increasing costs, limited availability, and variable quality, with periods of shortage being a primary concern to ornamental fish producers. Evaluation of appropriate alternatives, as well as development of culture methods aimed at reducing the duration of Artemia use during the larval production phase, are warranted.

The challenges associated with rearing larval fish, and particularly those associated with live feed production, have led to numerous studies evaluating potential alternatives (Lazo et al. 2000; Cahu and Zambonino Infante 2001; Kolkovski 2013; Thompson et al. 2019). Of particular interest to researchers and producers alike are commercially available microparticulate diets (MDs), which can provide more complete nutrition than live feeds, and are more convenient and reliable (Hamre et al. 2013; Kolkovski 2013; Lipscomb et al. 2020a). While total replacement of live feeds with MDs is ideal, this is not feasible for all species, particularly those with small, altricial larval stages (Rønnestad et al. 2003). In many marine fishes, as well as some freshwater species, first feeding larvae are physiologically incapable of utilizing complex nutrients found in MDs, therefore reduced reliance on live feeds requires optimizing the timing of weaning following critical developmental milestones (Cahu and Zambonino Infante 2001; Kolkovski 2001; Faulk and Holt 2009a).

The gastrointestinal (GI) tract in teleosts begins as a straight and undifferentiated tube that generally resembles the intestine of a more developed juvenile (Govoni et al. 1986). At the earliest stages, collective digestive enzyme activities are generally low, and proteolytic hydrolysis is primarily catalyzed in the intestinal lumen by trypsin and other trypsin-like enzymes of pancreatic origin (Govoni et al. 1986; Rønnestad et al. 2003; Holt 2011). In gastric fishes, developing larvae exhibit incremental ncreases in activities of individual enzymes, as well as the differentiation of the anterior GI into a functioning stomach. Gastric glands, a pyloric valve, and folding and elongation of the GI, in conjunction with the production of hydrochloric acid, pepsin, and protective neutral mucopolysaccharides all characterize this critical developmental milestone (Govoni et al. 1986; Faulk et al. 2007; Faulk and Holt 2009a; Holt 2011; Ghosh and Chakrabarti 2015). The development of the stomach affords the larvae a greatly increased aptitude for digesting complex nutrients, particularly those present in MDs (Rønnestad et al. 2003; Thompson et al. 2019). Previous authors have used the timing of the differentiation of the stomach, as well as peaks in digestive enzyme activities, to develop data-based weaning protocols for use in the production of fishes that exhibit this delayed mode of larval development (Faulk and Holt 2009b; Thompson et al. 2019).

Many members of the order Characiformes have been identified as species that exhibit extended agastric larval phases, including pacu Piaractus mesopotomicus (Tesser 2002), bocón Brycon amazonicus (Neumann 2008), and dorado Salminus brasiliensis (Mai and Zaniboni-Filho 2005), as reviewed by Portella et al. (2014). Additionally, it has been shown that live-feed dependence of the black tetra Gymnocorymbus ternetzi larvae at first feeding was due to an agastric larval period and weaning to MD is possible shortly before stomach differentiation (Lipscomb et al. 2020b) By categorizing the GI development and digestive enzyme ontogeny of P. innesi, a reduction in weaning time and minimization of reliance on Artemia can be achieved.

In addition to utilizing biochemical assays to quantify digestive enzyme activities, histological and histochemical analysis of GI structures provides further evidence for timing of stomach differentiation. Routine sectioning and staining with hematoxylin and eosin (H & E) reveal the formation of gastric glands, which house oxynticopeptic cells, responsible in teleosts for producing both pepsinogen and hydrochloric acid. Additionally, histochemical preparation with periodic acid Schiff (PAS) stain has been used to identify the presence of neutral pH mucopolysaccharides, which are produced by the epithelial cells of the stomach to protect the tissues from hydrochloric acid and autolytic hydrolysis catalyzed by pepsin (Ferraris et al. 1987; Albrecht et al. 2001; Çinar and Şenol 2006; Ghosh and Chakrabarti 2015; Faccioli et al. 2016). Together, these techniques can be used to accurately identify the critical milestone of stomach differentiation and subsequently optimize weaning schedules.

Here, we used an applied dietetics trial to evaluate the effectiveness of feeding protocols associated with three commercially available MDs relative to Artemia nauplii for first feeding in P. innesi. We also characterized the development of the gastrointestinal tract using histology (H & E stain) and histochemistry (PAS), as well as the ontogeny of three critical digestive enzymes (trypsin, bile-salt dependent lipase, and pepsin) through 32 days post hatch (dph), a time period inclusive of stomach differentiation. Subsequently, the appearance of gastric glands, PAS positive neutral mucopolysaccharides and initiation of pepsin activity, which are indicative of the appearance of a functioning stomach, were used to guide empirical evaluations of weaning from Artemia to a MD. Lastly, the impacts of different weaning regimes on survival, growth, and digestive enzyme activity were assessed. The results of this study can be used to optimize the larval culture of P. innesi, reducing the overall reliance of commercial producers on live feeds during this critical production stage.

2.1 General Methods

Larval P. innesi for each trial were acquired from a commercial ornamental producer (VW Fish Hatcheries Inc, Lakeland, FL). At the hatchery, brood fish were placed in pairs in individual spawning tanks with spawning media and were allowed to volitionally spawn. Following spawning, brood fish were removed, and embryos hatched in-situ. At 1 dph, larvae were pooled and transferred to the University of Florida’s Tropical Aquaculture Laboratory, Ruskin, FL where they were then counted by hand into experimental tanks. Dissolved oxygen (DO), pH, temperature, total ammonia nitrogen (TAN), and nitrite-N were monitored weekly for each tank in each trial. DO and temperature were measured using an optical dissolved oxygen meter (ProComm II, YSI Industries, Yellow Springs, OH). TAN, nitrite, and pH were measured using colorimetric assays according to manufacturer’s protocols (Hach Company, Loveland, CO).

2.2 Dietetics trial

Two dph larval P. innesi were stocked into twenty-four 3 L tanks at a density of 200 larvae per tank. Tanks were maintained on flow-through, degassed well water for the duration of the feeding trial. Beginning at 4 dph, concurrent with yolk-sac depletion, larvae were offered one of four feeding treatments (n = 6 per treatment) ad libitum to satiation, as indicated by the presence of feed in the tank following 15 minutes of feeding time. The treatments were as follows: newly hatched, 1^st – 2^nd instar stage A. franciscana nauplii twice per day, or one of three commercial MDs twice per day [MD1 (Larval AP-100, <100 µm, Ziegler Bros Inc., Gardners, PA; 50% protein, 12% fat), MD2 (Golden Pearls 50 -100 µm, Brine Shrimp Direct, Ogden, UT, USA; 60% protein, 8% fat), and MD3 (Otohime A1, 75 - 150 µm, Marubeni Nisshin Feed Co. LTD, Tokyo, Japan; 53% protein, 8% fat)]. Artemia cysts were decapsulated in 10% sodium hypochlorite and stored at 4°C prior to hatching in a 30 g/L saline solution.

For evaluation of larval feeding, two feeding techniques were employed, congruent with established methods for live feed delivery and MD delivery, respectively (Siccardi et al. 2009). For feeding Artemia, nauplii were harvested using a small brine net and transferred to a 100 ml sample cup filled with degassed well water. Nauplii were then introduced to the experimental tanks to reach a concentration of 100 nauplii/L per feeding. MDs were simply added directly to the tank at a rate of ~ 13.3 mg/L per day, spread across two feedings. Following 10 days of feeding (14 dph), all larvae remaining in each tank were harvested, enumerated and 10 randomly chosen larvae were photographed on a Sedgewick rafter cell using a trinocular dissecting microscope with a digital camera. Notochord length (mm) of larvae were digitally measured using Image J software (version 1.50i, National Institutes of Health, Bethesda, MD).

2.3 Characterization of gastrointestinal development and ontogeny

Two dph larvae were stocked into each of three 10 L tanks maintained on flow-through, degassed well water at a density of 600 per tank. Prior to the first feeding of the day at 3, 5, 7, 9, 11, 14, 17, 20, 23, 26, and 28 dph, 10 larvae from each tank were harvested and comingled in the same sampling cup (n = 30 larvae per sampling day). All 30 larvae were photographed on a Sedgewick rafter cell and measured for NL prior to hypural plate formation (20 dph) or SL thereafter (mm). After photographs were taken, 20 larvae were placed on a Kimwipe to absorb excess moisture, placed in 1.5 ml microcentrifuge tubes, and stored at -80°C for future processing for pancreatic and gastric digestive enzyme assays (10 larvae for each category of enzyme). The remaining 10 larvae were placed in a 1.5 ml microcentrifuge tube filled with 1 ml of Trump’s fixative (Electron Microscopy Sciences, Hatfield, PA) and stored at 4°C for future histological and histochemical processing.

All histological and histochemical processing was completed at the Fish and Wildlife Research Institute, operated by the Florida Fish and Wildlife Conservation Commission, St. Petersburg, FL. Samples were embedded in epoxy, longitudinally sectioned, and stained with either routine H & E (all sampling days), or periodic acid Schiff’s (PAS)/iron-hematoxylin with metanil yellow for contrast (14, 17, 20, and 23 dph Quintero-Hunter et al., 1991) to identify neutral mucopolysaccharides produced in the functioning, glandular stomach (Çinar and Şenol 2006; Ghosh and Chakrabarti 2015).

2.4 Weaning study

Following characterization of GI development and digestive enzyme ontogeny, a study was conducted to evaluate the effect of different weaning schedules on survival, growth and digestive enzyme activity. Twenty 10 L tanks maintained on flow-through, degassed well water were each stocked with 500 P. innesi larvae at 1 dph. Fish were fed beginning at 4 dph using one of five different feeding regimes: Artemia nauplii for the entirety of the trial, MD3 for the entirety of the trial, or one of three weaning schedules (n = 4, Figure 1). During periods of cofeeding, half rations of each MD3 and Artemia were given. MD3 was chosen for the weaning trial based on having the highest numerical survival following the dietetics trial. Weaning schedules consisted of a period of feeding exclusively Artemia from first feeding, followed by a 5-day period of cofeeding with Artemia and MD3, and finally a transition to exclusively MD3. Schedules were as follows: W1 (Artemia 4 – 12 dph; cofeed 13 – 17 dph; MD1 18 – 32 dph), W2 (Artemia 4 – 17 dph; cofeed 18 – 22 dph; MD3 23 – 32 dph), and W3 (Artemia 4 – 22 dph; cofeed 23 – 27 dph; MD3 28 – 32 dph). At 32 dph, each tank was harvested, fish were enumerated, and 20 fish were photographed, measured and frozen at -80°C for future processing for pancreatic and gastric digestive enzyme assays as previously described. Tanks fed exclusively MD3 exhibited low survival and were thus excluded from subsequent enzymatic assays.

2.5 Digestive enzyme activity analysis

Digestive enzyme activity assays were performed according to Zambonino Infante and Cahu (1994, pepsin), and Holt et al. (2007, trypsin and bilesalt-dependent lipase). On the morning of each assay, samples were thawed in an aluminum cooling chamber nested in ice. For measuring pepsin activity, samples were homogenized with a motorized micro-pestle in 400 µl reagent grade deionized water. Samples were then centrifuged for 10 minutes at 1700 G at 4, and 100 µl of each supernatant was transferred to a new microcentrifuge tube. In addition, a negative control with 100 µl of reagent grade deionized water, as well as a positive control with 100 µl of 50 µg/ml pepsin solution were aliquoted into microcentrifuge tubes. With tubes in a water bath at 37, 500 µl of hemoglobin substrate (60 mM HCl, 2% w/v equine hemoglobin) was added to each sample. After 20 minutes, 1 ml of 5% trichloroacetic acid was added to the tubes to quench the reaction. For each sample, a blank was created by adding 1 ml of 5% trichloroacetic acid prior to the addition of the hemoglobin substrate. Samples and blanks were centrifuged for 10 minutes at 1700 G and 20 µl of each supernatant was pipetted into a 96-well plate in triplicate, in addition to positive and negative controls. The concentration of tyrosine liberated from hydrolysis of the substrate was measured as absorbance at 280 nm using a BioTek spectrophotometer (Synergy HTX Multi-Mode Microplate Reader, BioTek Instruments, Inc., Winooski, VT, USA). Pepsin activity was expressed as µmol of tyrosine liberated per minute.

Samples of 10 fish for both trypsin and bile-salt dependent lipase were homogenized in a buffer solution (20 mM Tris-HCl, 1mM EDTA, 1 mM CaCl₂, 7.5 pH) in 1.5 ml microcentrifuge tubes and centrifuged for 10 minutes at 1700 G at 4. To measure trypsin activity, 20 µl of each sample supernatant, as well as 15 µg/ml porcine trypsin solution (positive control) and homogenization buffer (negative control) were pipetted in triplicate into a 96-well plate resting on an aluminum cooling plate in ice. Trypsin substrate solution (100 µL, 1mM N_α-benzoyl-DL-arginine p-nitroanilide dissolved in dimethyl sulfoxide) was quickly added to each well using a multichannel pipette. At 30 liberation of p-nitroaniline from hydrolysis of the substrate was measured as absorbance at 410 nm every 45 seconds for 30 minutes. Trypsin activity was expressed as µmol of p-nitroaniline liberated per minute (U).

For bile-salt dependent lipase, 10 µl of each sample supernatant, 7.5 µg/ml lipase enzyme standard solution (positive control) and homogenization buffer (negative control), were pipetted in triplicate into a 96-well plate resting on an aluminum cooling plate in ice. Quickly, 200 µl of lipase substrate solution (0.35 mM 4-nitrophenyl n-Caproate (4-NPC) dissolved in 70 µl ethanol, 0.5 M Tris-HCl, 6 mM sodium taurochlorate, 1 M NaCl, pH = 7.4) was added to each well. At 30°C, hydrolyzation of 4-NPC was measured at 400 nm absorbance and units were expressed as 1 µM 4-NPC hydrolyzed per minute (U).

2.6 Statistical analysis

Water quality parameters among treatments were compared using one-way ANOVA following verification of assumptions of normality using a Shapiro-Wilks test and homogeneity of variance using a Bartlett’s test. Growth rate recorded during the GI characterization study was determined using linear regression. Proportional survival data for each trial was evaluated using the glm function (family = binomial) in R to perform a logistic regression on the Bernoulli distributed survival data (Crawley 2005). Effect of treatment on survival was evaluated using a likelihood ratio test, followed by a post hoc pairwise Tukey HSD test on the log odds of survival if a significant effect of treatment was detected. All survival data are reported as mean percent survival ± standard error (SE) of the proportion. Standard lengths and digestive enzyme activities resulting from different treatment groups were compared using one-way ANOVA following confirmation of assumptions of normality and homogeneity of variance. If a significant effect of treatment on standard length or enzyme activity was detected, a post hoc pairwise Tukey HSD test was used to compare differences between individual treatments. Values for SL and digestive enzyme activity (U) are reported as mean ± standard error of the mean. If individual tanks had zero surviving larvae at the conclusion of the trial, these tanks were omitted from analysis of SL and an unbalanced design was employed. A priori significance levels were set at a = 0.05 and all analyses were completed in Program R (RStudio, V. 0.99.903 2015. RStudio: Integrated Development for R. RStudio, Inc., Boston, MA).

3.1 Dietetics trial

Survival varied significantly among treatments following 10 days of initial feeding with either Artemia, MD1, MD2, or MD3 (X²₄ = 1,187.300, P < 0.001, Figure 2a). Feeding behavior in each treatment group was confirmed visually. Survival of P. innesi larvae fed Artemia nauplii from first feeding (4 dph) until 14 dph (39.8 ± 3.3%) was significantly higher than the survival resulting from MDs (mean: 1.4 ± 0.6%, P < 0.001). Because survival resulting from MDs was zero in many of the tanks fed MDs, these tanks were omitted from analysis of differences in NL. There were significant differences in NL among treatments at 14 dph (F₃ = 9.342, P < 0.001), with Artemia resulting in higher NL (4.35 ± 0.04mm) than MD2 and MD3 (3.86 ± 0.08 mm and 4.06 ± 0.07mm, respectively, P £ 0.010), while NL resulting from MD1 did not vary from other treatments (4.11 ± 0.07mm, P ³ 0.078). Mean standard lengths ± SE of all remaining larvae in each MD treatment group, as well as 10 larvae from each of the tanks fed Artemia are presented in Figure 2b.

3.2 Characterization of gastrointestinal development and ontogeny

Larval P. innesi growth from 3 to 29 dph was linear (R² = 0.887) and NL/SL increased at a rate of 0.154 mm/day (Figure 3). Mean NL at 3 dph was 3.26 ± 0.03 mm and increased to a mean SL of 7.25 ± 0.26 mm at 29 dph. Yolk was present at 3 dph and was nearly completely absorbed by 5 dph. The GI tract persisted as a straight, uniform tube divided into esophagus, intestine (including stomach anlage), and rectum from 3 dph until 20 dph, when differentiation of the stomach was observed (Figure 4). At 20 dph, gastric glands were present in the stomach and a pyloric valve separated the stomach and the anterior intestine (Figures 4 and 5). The GI exhibited a fold initiating between the pyloric region of the stomach and the anterior intestine (Figure 4c). By 29 dph, further folding along the entirety of the GI occurred, as well as proliferation of gastric glands in the lamina propria and the appearance of sections of 6 distinct pyloric caeca (Figure 4d, 5d). Histochemical preparations with PAS stain revealed a distinct absence of neutral mucopolysaccharides in the stomach anlage prior to gastric gland formation (Figure 6a). Following the appearance of gastric glands and concurrent stomach differentiation, the apical surface of the stomach epithelium as well as much of the stomach lumen were strongly PAS positive, indicating the presence of neutral mucopolysaccharides in these regions (Figure 6c,d).

While trypsin activity was detectable from 3 dph, activity on a per fish basis remained relatively low (£ 144.95 U/fish) until 29 dph, when a 6.42-fold increase to 931.06 U/fish occurred (Figure 7a). Lipase exhibited a similar pattern, with enzyme activities of £ 12,655.71 U/fish occurring between 3 and 20 dph, followed by sharp increases at 23, 26, and 29 dph to 45,672.15 U/fish (Figure 7b). Pepsin activity was undetectable until 11 dph, when very slight activity (0.342 U/fish) was detected. Pepsin activity was low and sporadic (£ 1.58 U/fish) until 29 dph, when activity increased 3.78 times relative to the previous peak (5.97 U/fish, Figure 7c).

3.3 Weaning study

Survival at 32 dph varied significantly among weaning treatments (X²₅ = 712.64, P < 0.001), with exclusively feeding MD1 from first feeding resulting in the lowest survival (0.75 ± 0.33%). Fish fed exclusively Artemia and those receiving weaning treatments W1 and W2 resulted in similar survival (22.20 ± 1.62%, 21.50 ± 1.72%, and 22.20 ± 1.85%, respectively, P ³ 0.988). Treatment W3 resulted in higher survival (16.20 ± 1.30%) than MD1, but lower survival than the other treatments (P < 0.001, Figure 8a). Because survival was exceedingly low for fish fed exclusively MD1, SL and digestive enzyme activities for this treatment group were not included in those analyses. Growth among the remaining treatments was similar (F₃ = 1.830, P = 0.142), with an overall final mean standard length of 7.38 ± 0.08 mm (Figure 8b).

Trypsin activity varied significantly among treatments (F₃ = 3.755, P = 0.049), however, post hoc pairwise comparison did not reveal significant relationships between individual treatments, due to the conservative nature of Tukey HSD for protection against type I error (P ³ 0.066, Figure 9a). Mean trypsin activity at 32 dph for all treatments was 1,281.91 ± 124.91 U/fish. Activities for both lipase and pepsin did not vary significantly among treatments (F₃ = 0.520, P = 0.678, and X²₃ = 1.752, P = 0.625, respectively). Mean lipase activity for all treatments at 32 dph was 28,316.35 ± 2,698.40 U/fish and mean pepsin activity was 18.87 ± 1.23 U/fish (Figure 9b,c).

Evaluation of larval dietetics confirmed that, at least for the three MDs evaluated, P. innesi was incapable of thriving on an MD from first feeding, particularly when compared to larvae fed Artemia (Figure 2). This pattern was confirmed in the weaning trial, where larvae fed exclusively MD3 from first feeding survived very poorly (0.75 ± 0.33%) relative to all other treatments (Figure 8a). When taken with the evidence for an extended, agastric larval phase revealed through the physiological characterization of developmental ontogeny (Figures 4-7), it can be assumed that limited digestive capacity due to low digestive enzyme activity and the absence of acidic digestive function led to this result.

During early development of larval teleosts there exist three modes of digestive system ontogeny: agastric, precocial, and altricial (Ronnestad et al. 2013). In agastric fishes (i.e. Cyprinidae, Gobiidae), there is a complete reliance on alkaline, pancreatic proteases for hydrolysis of proteins over the entirety of the fishes life, and digestive capacity increases as a function of gut length as the larvae develops, as well as the development of other structures like pharyngeal teeth (Dabrowski 1984; Dabrowski and Poczyczyński 1988; Hur et al. 2016). In precocial fishes (i.e. Salmonidae, Bagridae), larvae develop a fully functioning stomach with concomitant acid protease activity prior to the onset of exogenous feeding, vastly increasing their digestive capacity relative to agastric fishes and allowing for introduction of MDs from first feeding (Dabrowski 1984; Yang et al. 2010). Finally, altricial fishes (i.e. Sparidae, Paralichthyidae) exhibit an agastric larval phase, where protease activity is due to alkaline proteases at first feeding until which time the stomach differentiates and gastric digestion begins (Santamaría et al. 2004; Faulk and Holt 2009b). In altricial fishes, the successful introduction of MDs in an aquaculture setting largely hinges on the differentiation of a functional stomach, as established by previous authors (Faulk and Holt 2009b; Rønnestad et al. 2013; Thompson et al. 2019).

Complete stomach differentiation in P. innesi appeared to occur by 20 dph, evidenced by appearance of a pyloric valve, gastric glands, and PAS positive neutral mucopolysaccharide production (Figures 4-6). The ontogeny of pepsin activity did not exhibit as clear of a pattern, potentially indicating some age specific variation in stomach development on an individual scale. Slight peaks at 11 and 17 dph (figure 7c) may be the result of accelerated development of a few fish that were present in the pooled sample of ten fish for the pepsin assays, introducing variation in the sample result. Alternatively, previous authors who observed a similar phenomenon in leopard grouper Mycteroperca rosacea ascribed these early peaks in acid protease activity to other acid enzymes in other regions of the body of the larvae (Martínez-Lagos et al. 2014). Thompson et al. (2019) performed micro-dissections of pigfish Orthopristis chrysoptera larvae, effectively eliminating the potential for this type of false positive, which could have clarified the results in this study. However, given that feeding began at 4 dph and yolk absorption was nearly complete by 5 dph (Figure 4a), and that there was a complete lack of pepsin activity until 11 dph, a protracted period of agastric, completely alkaline digestion is apparent.

The presence of PAS positive neutral mucopolysaccharides beginning at 20 dph provides further evidence for the initiation of acidic digestive function beginning at this age. Neutral mucins containing PAS positive mucopolysaccharides have been theorized to be produced by epithelial cells located in the mucosa of the stomach to protect this layer from the activity of the concurrently produced hydrochloric acid and pepsin by the oxynticopeptic cells of the gastric glands (Scocco et al. 1996; Pradhan et al. 2013; Ghosh and Chakrabarti 2015). The mucins in question effectively create a boundary between the acidic environment of the stomach lumen and the active pepsin, protecting the apical surface of the stomach epithelium from autolysis. The production of these compounds signals a functional shift, where acid and pepsin are actively being produced (Scocco et al. 1996).

Other studies have implicated a similar pattern of agastric larval digestive function as causative in low survival following first feeding with MDs. Hamlin et al. (2000) described the larval digestive system ontogeny in haddock Melanogrammus aeglefinus, which begin exogenous feeding at 2 dph, but do not exhibit stomach differentiation and gastric gland formation until 33 dph. The authors of this study indicated that weaning M. aeglefinus from live feeds to an MD should be conducted following this transition (Hamlin et al. 2000). Likewise, a study that described digestive enzyme ontogeny and evaluated weaning from live feeds to an MD in larval O. chrysoptera showed that pepsin activity remained low until 30 dph and weaning at 32 dph yielded significantly higher survival than earlier weaning times (Thompson et al. 2019). Additionally, we showed a similar pattern of development in G. ternetzi, a characid relative of P. innesi (Lipscomb et al. 2020b). In the case of G. ternetzi, pepsin activity, gastric gland formation, stomach differentiation and neutral mucopolysaccharide production all occurred at 22 dph, although weaning with cofeeding initiated at 13 dph resulted in similar survival to later weaning times (Lipscomb et al. 2020b).

While P. innesi exhibit histological and histochemical stomach differentiation at 20 dph, cofeeding beginning at 13 dph resulted in similar survival and growth to fish fed exclusively Artemia from first feeding through 32 dph, as well as cofeeding initiated at 17 dph. Given that cofeeding ceased and exclusive feeding with MD1 occurred at 17dph for weaning treatment W1, it appears that weaning to an MD prior to completion of stomach formation is possible for P. innesi (Figure 8). This may be explained by the increased activity of lipase and a slight increase in pepsin that were concurrent with this weaning period (Figure 7).

Alternatively, authors have suggested that initiating larval nutrition with live feeds may infer digestive benefits from the exogenous digestive enzymes provided by the prey animals, both as active enzymes and through autolysis of live feed tissues (Kolkovski 2001). Additionally, research has shown that live feeds may increase the production of endogenous enzymes in larval fish (Hjelmeland et al. 1988). Hjelmeland et al. (1988) observed that ingestion of either copepods or Artemia in larval herring Clupea harengus elicited higher levels of trypsin in the intestine than inert particles, although the inert particles did elicit some response. Either of these interactions between live prey and larval digestive function could help explain the observed results of successful weaning prior to complete stomach differentiation in P. innesi.

Timing of weaning in P. innesi did not significantly impact pepsin or lipase activities, and although a significant effect on trypsin activity was detected (F₃ = 3.755, P = 0.049), pairwise comparisons did not reveal differences between individual treatments (Figure 9). Given that samples were taken after a period of overnight fasting, positive feedback interactions between ingested feed items and enzyme activity were likely limited. This means that observed digestive enzyme activities can be interpreted as basal zymogen production, and not acutely modulated by ingesta. This observation agrees with the previous assertion by Zambonino, Infante, and Cahu (2001) that diet may modulate peaks in enzyme activities following ingestion, but timing of increases and decreases in activity is largely genetically determined.

The results of this study show that live feed dependence of P. innesi at first feeding is likely due to an altricial pattern of GI development, where the stomach and associated functions are absent during the initiation of exogenous feeding, and do not appear until between 11 and 20 dph. Weaning prior to the differentiation of the stomach organ and appearance of gastric glands, pyloric valve, and PAS positive neutral mucins is possible, with weaning beginning at 13 dph inferring similar survival to exclusively feeding Artemia. This result is consistent with previous work with another characid, G. ternetzi, which potentially reveals a pattern of digestive function common to small-bodied characids, although further research into more species is warranted (Lipscomb et al. 2020b). In conclusion, commercial production of P. innesi can be accomplished with minimal live feed inputs if the early weaning protocols established herein are implemented. This potentially increases the resilience of this aquaculture endeavor in periods of Artemia shortage, as well as provides an overall economic advantage for the critical larval culture stage in this ornamental fish.

Acknowledgements

The authors acknowledge Noretta Perry of the Fish and Wildlife Research Institute, Florida Fish and Wildlife Conservation Commission, for histological and histochemical processing. In addition, we acknowledge VW Fish Hatcheries Inc, Lakeland, FL, for providing fish for this research.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Conflicts of interest/Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Ethics approval/declarations

All research herein was conducted under approval by and in compliance with standards developed by the IACUC of the University of Florida.

Consent to participate

Not applicable

Consent for publication

Not applicable

Data Availability

The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Code availability

The statistical analyses and figures herein were created in Program R (RStudio, V. 0.99.903 2015. RStudio: Integrated Development for R. RStudio, Inc., Boston, MA). The custom code is available upon reasonable request.

Authors' contributions

TL conceptualized the research, designed and conducted the experiments, conducted associated assays, analyzed resulting data and wrote the manuscript. RY facilitated histological processing, interpreted histological slides, and reviewed/edited the manuscript. SR assisted in conducting the experiments, conducted associated assays, and reviewed/edited the manuscript. MD conceptualized the research, assisted in conducting the experiments, and reviewed/edited the manuscript.

Albrecht MP, Ferreira MFN, Caramaschi EP (2001) Anatomical features and histology of the digestive tract of two related neotropical omnivorous fishes (Characiformes; Anostomidae). J Fish Biol 58:419–430. doi: 10.1006/jfbi.2000.1462
Cahu C, Zambonino Infante J (2001) Substitution of live food by formulated diets in marine fish larvae. Aquaculture 200:161–180. doi: 10.1016/S0044-8486(01)00699-8
Chao NL (2001) The Fishery, Diversity, and Conservation of Ornamental Fishes in the Rio Negro Basin, Brazil - A review of Project Piaba (1989-99). Conserv Manag Ornam fish Resour Rio Negro Basin, Amaz Brazil – Proj Piaba 161–204
Chapman FA, Colle DE, Rottmann RW, Shireman J V. (1998) Controlled Spawning of the Neon Tetra. Progress Fish-Culturist 60:32–37. doi: 10.1577/1548-8640(1998)060<0032:CSOTNT>2.0.CO;2
Chapman FA, Fitz-Coy SA, Thunberg EM, Adams CM (1997) United States of America trade in ornamental fish. J World Aquac Soc 28:1–10. doi: 10.1111/j.1749-7345.1997.tb00955.x
Çinar K, Şenol N (2006) Histological and histochemical characterization of the mucosa of the digestive tract in flower fish (Pseudophoxinus antalyae). J Vet Med Ser C Anat Histol Embryol 35:147–151. doi: 10.1111/j.1439-0264.2005.00629.x
Crawley MJ (2005) Statistics: an introduction using R. John Wiley & Sons, West Sussex, England
Dabrowski K, Poczyczyński P (1988) Comparative experiments on starter diets for grass carp and common carp. Aquaculture 69:317–332. doi: 10.1016/0044-8486(88)90339-0
Dabrowski KR (1984) The feeding of fish larvae: present (state of the art) and prespective. Reprod Nutr Développement 24:807–833. doi: 10.1051/rnd:19840701
Faccioli CK, Chedid RA, Mori RH, et al (2016) Organogenesis of the digestive system in Neotropical carnivorous freshwater catfish Hemisorubim platyrhynchos (Siluriformes: Pimelodidae). Aquaculture 451:205–212. doi: 10.1016/j.aquaculture.2015.09.009
Faulk CK, Benninghoff AD, Holt GJ (2007) Ontogeny of the gastrointestinal tract and selected digestive enzymes in cobia Rachycentron canadum (L.). J Fish Biol 70:567–583. doi: 10.1111/j.1095-8649.2007.01330.x
Faulk CK, Holt GJ (2009a) Early weaning of southern flounder, Paralichthys lethostigma, larvae and ontogeny of selected digestive enzymes. Aquaculture 296:213–218. doi: 10.1016/j.aquaculture.2009.08.013
Faulk CK, Holt GJ (2009b) Early weaning of southern flounder, Paralichthys lethostigma, larvae and ontogeny of selected digestive enzymes. Aquaculture 296:213–218. doi: 10.1016/j.aquaculture.2009.08.013
Ferraris RP, Tan JD, De La Cruz MC (1987) Development of the digestive tract of milkfish, Chanos chanos (Forsskal): Histology and histochemistry. Aquaculture 61:241–257. doi: 10.1016/0044-8486(87)90153-0
Ghosh SK, Chakrabarti P (2015) Histological and histochemical characterization on stomach of Mystus cavasius (Hamilton), Oreochromis niloticus (Linnaeus) and Gudusia chapra (Hamilton): Comparative study. J Basic Appl Zool 70:16–24. doi: 10.1016/j.jobaz.2015.04.002
Govoni JJ, Boehlert GW, Watanabe Y (1986) The physiology of digestion in fish larvae. Environ Biol Fishes 16:59–77. doi: 10.1007/BF00005160
Hamlin HJ, Hunt Von Herbing I, Kling LJ (2000) Histological and morphological evaluations of the digestive tract and associated organs of haddock throughout post-hatching ontogeny. J Fish Biol 57:716–732. doi: 10.1006/jfbi.2000.1347
Hamre K, Yúfera M, Rønnestad I, et al (2013) Fish larval nutrition and feed formulation: Knowledge gaps and bottlenecks for advances in larval rearing. Rev Aquac 5:1–87. doi: 10.1111/j.1753-5131.2012.01086.x
Hjelmeland K, Pedersen BH, Nilssen EM (1988) Trypsin content in intestines of herring larvae, Clupea harengus, ingesting inert polystyrene spheres or live crustacea prey. Mar Biol 98:331–335. doi: 10.1007/BF00391108
Holt GJ (2011) Larval Fish Nutrition. John Wiley & Sons, Hoboken, NJ, USA
Hur S-W, Kim S-K, Kim D-J, et al (2016) Digestive Physiological Characteristics of the Gobiidae: - Characteristics of CCK-producing Cells and Mucus-secreting Goblet Cells of Stomach Fish and Stomachless Fish. Dev Reprod 20:207–217. doi: 10.12717/DR.2016.20.3.207
Kolkovski S (2013) Microdiets as alternatives to live feeds for fish larvae in aquaculture: Improving the efficiency of feed particle utilization. In: Allan G, Burnell G (eds) Advances in Aquaculture Hatchery Technology. Elsevier Science & Technology, Philadelphia, PA, pp 203–222
Kolkovski S (2001) Digestive enzymes in fish larvae and juveniles—implications and applications to formulated diets. Aquaculture 200:181–201. doi: 10.1016/S0044-8486(01)00700-1
Kucharczyk D, Targońska K, Zarski D, et al (2010) The reproduction of neon tetra, paracheirodon innesi (Myers, 1936), under controlled conditions. Polish J Nat Sci 25:81–92. doi: 10.2478/v10020-010-0006-x
Lavens P, Sorgeloos P (2000) The history, present status and prospects of the availability of Artemia cysts for aquaculture. Aquaculture 181:397–403. doi: 10.1016/S0044-8486(99)00233-1
Lazo JP, Dinis MT, Holt GJ, et al (2000) Co-feeding microparticulate diets with algae: Toward eliminating the need of zooplankton at first feeding in larval red drum (Sciaenops ocellatus). Aquaculture 188:339–351. doi: 10.1016/S0044-8486(00)00339-2
Lim LC, Dhert P, Sorgeloos P (2003) Recent developments in the application of live feeds in the freshwater ornamental fish culture. Aquaculture 227:319–331. doi: 10.1016/S0044-8486(03)00512-X
Lipscomb TN, Patterson JT, Wood AL, et al (2020a) Larval growth, survival and partial budget analysis related to replacing Artemia in larval culture of six freshwater ornamental fishes. J World Aquac Soc 1132–1144
Lipscomb TN, Yanong RP, Ramee SW, DiMaggio MA (2020b) Histological, histochemical and biochemical characterization of larval digestive system ontogeny in black tetra Gymnocorymbus ternetzi to inform aquaculture weaning protocols. Aquaculture 520:734957. doi: 10.1016/j.aquaculture.2020.734957
Mai MG, Zaniboni-Filho E (2005) Efeito da idade de estocagem em tanques externos no desempenho da larvicultura do dourado <em>Salminus brasiliensis</em> (Osteichthyes, Characidae). Acta Sci Anim Sci 27:287–296. doi: 10.4025/actascianimsci.v27i2.1252
Martínez-Lagos R, Tovar-Ramírez D, Gracia-López V, Lazo JP (2014) Changes in digestive enzyme activities during larval development of leopard grouper (Mycteroperca rosacea). Fish Physiol Biochem 40:773–785. doi: 10.1007/s10695-013-9884-5
Neumann E (2008) Desenvolvimento inicial da jatuarana, Brycon amazonicus (Teleostei, Characidae). Sao Paulo State University
Portella MC, Jomori RK, Leitão NJ, et al (2014) Larval development of indigenous South American freshwater fish species, with particular reference to pacu (Piaractus mesopotamicus): A review. Aquaculture 432:402–417. doi: 10.1016/j.aquaculture.2014.04.032
Pradhan PK, Jena J, Mitra G, et al (2013) Ontogeny of the digestive enzymes in butter catfish Ompok bimaculatus (Bloch) larvae. Aquaculture 372–375:62–69. doi: 10.1016/j.aquaculture.2012.10.024
Quintero-Hunter I, Grier H, Muscato M (1991) Enhancement of histological detail using metanil yellow as counterstain in periodic acid Schiff’s hematoxylin staining of glycol methacrylate tissue sections. Biotech Histochem 66:169–172
Rønnestad I, Tonheim SK, Fyhn HJ, et al (2003) The supply of amino acids during early feeding stages of marine fish larvae: A review of recent findings. Aquaculture 227:147–164. doi: 10.1016/S0044-8486(03)00500-3
Ronnestad I, Yúfera M, Ueberschär B, et al (2013) Feeding behaviour and digestive physiology in larval fish: current knowledge, and gaps and bottlenecks in research. Rev Aquac 5:S59–S98. doi: 10.1111/raq.12010
Rønnestad I, Yúfera M, Ueberschär B, et al (2013) Feeding behaviour and digestive physiology in larval fish: Current knowledge, and gaps and bottlenecks in research. Rev Aquac 5:. doi: 10.1111/raq.12010
Sanaye S V, Dhaker HS, Tibile RM, Mhatre VD (2014) Effect of Green Water and Mixed Zooplankton on Growth and Survival in Neon Tetra, Paracheirodon innesi (Myers, 1936) during Larval and Early Fry Rearing. Int J Biol Biomol Agric Food Biotechnol Eng 8:159–163
Santamaría CA, Marín De Mateo M, Traveset R, et al (2004) Larval organogenesis in common dentex Dentex dentex L. (Sparidae): Histological and histochemical aspects. Aquaculture 237:207–228. doi: 10.1016/j.aquaculture.2004.03.020
Scocco P, Ceccarelli P, Menghi G (1996) Glycohistochemistry of the Tilapia spp. stomach. J Fish Biol 49:584–593. doi: 10.1006/jfbi.1996.0188
Siccardi AJ, Garris HW, Jones WT, et al (2009) Growth and survival of Zebrafish (Danio rerio) fed different commercial and laboratory diets. Zebrafish 6:275–280. doi: 10.1089/zeb.2008.0553
Sorgeloos P, Kulasekarapandian S (1984) Production and use of Artemia in aquaculture. C Spec Publ 1–59
Tesser MB (2002) Desenvolvimento do trato digestório e crescimento de larvas de pacu, Piaractus mesopotamicus (Holmberg, 1887) em sistemas de co-alimentação com náuplios de Artemia e dieta microencapsulada. Universidade Estadual Paulista
Thompson KL, Faulk CK, Fuiman LA (2019) Applying the ontogeny of digestive enzyme activity to guide early weaning of pigfish, Orthopristis chrysoptera (L.). Aquac Res 50:1404–1410. doi: 10.1111/are.14015
Weitzman SH, Weitzman SH, Fink WL (1983) Relationships of the neon tetras, a group of South American freshwater fishes (Teleostei, Characidae), with comments on the phylogeny of New World characiforms. Bull Museum Comp Zool Harvard Coll 150:339–395
Yang R, Xie C, Fan Q, et al (2010) Ontogeny of the digestive tract in yellow catfish Pelteobagrus fulvidraco larvae. Aquaculture 302:112–123. doi: 10.1016/j.aquaculture.2010.02.020
Zambonino Infante JL, Cahu CL (1994) Influence of diet on pepsin and some pancreatic enzymes in sea bass (Dicentrarchus labrax) larvae. Comp Biochem Physiol 109A:209–212. doi: 10.1016/0300-9629(94)90122-8
Zambonino Infante JL, Cahu CL (2001) Ontogeny of the gastrointestinal tract of marine fish larvae. Comp Biochem Physiol Part C Toxicol Pharmacol December 130:477–487. doi: 10.1016/S1532-0456(01)00274-5

Download PDF

Journal Publication

published 23 Oct, 2023

Read the published version in Fish Physiology and Biochemistry →

Reviewers agreed at journal
13 Jun, 2022
Reviewers invited by journal
13 Jun, 2022
Editor invited by journal
06 Jun, 2022
Editor assigned by journal
24 May, 2022
First submitted to journal
23 May, 2022

You are reading this latest preprint version

Larval digestive system ontogeny and early weaning in neon tetra Paracheirodon innesi

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods

3. Results

4. Discussion

Declarations

References

Status:

Journal Publication

Version 1