Turnover Rates and Diet–Tissue Discrimination Factors of Nitrogen and Carbon Stable Isotopes in Seahorse Hippocampus reidi Juveniles Following a Laboratory Diet Shift

1. Introduction

Dietary studies in fishes provide useful information to assess predator–prey relationships, competition, food intake, and food-web dynamics. Traditionally, they were based on gut content analysis, but stable isotope analysis (SIA) is an analytical tool that can be applied in dietary traceability studies and long-term food utilization by organisms [1,2,3,4,5]. The basic assumption of SIA is that the isotopic composition of an organism’s tissues reflects that of its diet [6] offset by a trophic discrimination factor [7].

The two most commonly measured stable isotope ratios are ¹³C/¹²C and ¹⁵N/¹⁴N; both ratios are usually higher in the consumer’s tissues compared to its diet because the lighter isotope (¹²C and ¹⁴N) is preferred in metabolic processes [8,9] and also because heavier isotopes are preferentially retained. Specifically, the stable-carbon isotope ratio (¹³C/¹²C) of tissues reflects the sources of organic carbon consumed, with little tissue fractionation [8]. Stable-nitrogen isotope ratios (¹⁵N/¹⁴N) increase with successive trophic levels [9], allowing estimates of the consumer’s trophic position [7,10,11,12].

Discrimination represents the difference between isotope values for diet and fully equilibrated consumer tissue [11]. Discrimination factors for carbon and nitrogen usually average 0–1‰ and 3–4‰ per trophic level, respectively [13], depending on the tissues/species considered [7,12]. However, the magnitude of this per trophic-step isotope fractionation (Δ¹³C or Δ¹⁵N) can be affected by other many factors such as diet quality, feeding ratio, nutritional stress, body size, age, physiological status, and excretory mechanisms [14,15,16,17,18,19,20].

Generally, stable isotope values are fitted to growth or time-based models [14,15,16,17,18,19,20]. Most laboratory diet-switch experiments [21,22,23,24,25,26,27] show that growth is the primary factor causing stable isotopic changes in fish following a diet shift. In the case of fish larvae, experimental stable isotope studies investigating the effects of a diet shift on stable isotope incorporation are scarce. These studies are relevant in identifying the diet preferences of larvae and juveniles, understanding nutrition needs, improving rearing techniques, and interpreting field stable isotope studies [28]. In addition, knowledge of species turnover and discrimination factors are important for the accurate interpretation of isotopic data.

Hippocampus reidi (Teleostei: Family Syngnathidae) is a tropical euryhaline seahorse, inhabiting estuarine areas along the Western Atlantic coast, mainly in Brazil [29,30,31]. It is one of the most significant seahorse species in the marine aquarium trade [32,33]. Previous studies in seahorse cultivation have demonstrated the limited digestive capability in early developing juveniles and the low digestion of Artemia nauplii compared to copepods [34,35,36]. Consequently, the initial feeding enhances the growth and survival of copepods. The present study was performed in the early developmental stages of H. reidi by assessing the influence of diet shift on isotopic changes (δ¹³C and δ¹⁵N) in juveniles fed on two different feeding schedules including copepods and Artemia nauplii. This experimental study aimed to determine for the first time in seahorses: (1) δ¹³C and δ¹⁵N turnover rates in juveniles as functions of change in body mass and time, (2) contributions of metabolism and growth to those turnover rates, and (3) diet–tissue discrimination factors.

2. Materials and Methods

2.1. Broodstock

Adult seahorses Hippocampus reidi Ginsburg, 1933 were maintained in ad hoc aquaria [37] at the Instituto de Investigaciones Marinas (IIM-CSIC) in Vigo (Spain) and fed twice daily on a diet consisting of long-time enriched adult Artemia sp. (EG, AF, MC450; Iberfrost, Tomiño, Spain; 40–70 Artemia seahorse⁻¹ dose⁻¹) [38] and frozen Mysidaceans Neomysis sp. (Ocean Nutrition, Essen, Belgium). When available, a single daily dose of wild-captured Mysidacea (15–20 Leptomysis sp. and/or Siriella sp.) was also provided. Seawater was maintained at a constant temperature within an annual temperature regime of 26 ± 0.5 °C. A natural-like photoperiod regime for the species was applied (16L:8D). Pumped seawater was filtered (5 µm), UV treated, and 10–15% was exchanged daily. Water quality was checked periodically for NO₂, NO₃, and NH₄/NH₃ content (0 mg L⁻¹) using Sera Test Kits (Sera GmbH, Heinsberg, Germany). Salinity and pH levels were constantly maintained 38 ± 1 and 8.1 ± 0.1, respectively. Wastes and uneaten food were removed daily by siphoning the bottom of the aquaria.

2.2. Experimental Design

Seahorse juveniles were obtained from a batch released by one male held in captivity for 3 years. Immediately after the male’s pouch release, the juveniles were randomly transferred (3.3 juveniles L⁻¹) into 4 (2 aquaria per treatment) 30 L aquaria [39]. The rearing system was illuminated by 20 W Power Glo lamps (Hagen, Montreal, Canada) under a 14 L:10 D photoperiod regime. Water temperature was adjusted to 26 °C. Total seawater volumes in the rearing system were replaced twice per hour by means of an external inflow (24 L h−1) of 20 μm filtered and UV-treated seawater. Aquaria were gently aerated in the upper part of the water column at a continuous flow rate of 700 mL min⁻¹. Twice daily, wastes and feces were siphoned out, and dead seahorses were removed and counted.

Two feeding schedules (diets A6 and A11) were compared considering the following feeding conditions:

Diet A6

-

First feeding (0 to 5 DAR): Two daily doses of cultivated copepods Acartia tonsa (1 copepod mL⁻¹);

-

Artemia feeding (6 to 60 DAR): Two daily doses of Artemia nauplii (1–2 Artemia mL⁻¹).

-

Diet A11

-

First feeding (0 to 10 DAR): Two daily doses of cultivated copepods Acartia tonsa (1 copepod mL⁻¹);

-

Artemia feeding (11 to 60 DAR): Two daily doses of Artemia nauplii (1–2 Artemia mL⁻¹).

The experimental feeding schedules were established considering previous results on the effect of diet in the early rearing of H. reidi juveniles [36]. In that study, we demonstrated an enhancement of seahorse juveniles with longer periods of feeding on copepods.

The copepods were cultivated in 700 L tanks at 26–27 °C and 38 salinity and fed every two days on microalgae Rhodomonas lens (10³ cells mL⁻¹). Only copepods retained by a 125 μm mesh were offered to the seahorses. Artemia nauplii were obtained from cysts (AF, Inve, Spain) hatched at 28 °C for 20 h in 20 L units. After hatching, the nauplii were gently rinsed with tap-water, collected on a 125 μm mesh, rinsed, and offered to seahorse juveniles. The biochemical compositions of the prey are provided in [36]. The mean dry weight and length of the copepods were 1.67 ± 0.01 µg and 614 ± 140 µm, respectively, and 1.67 ± 0.04 µg and 576 ± 82 µm in Artemia nauplii, respectively.

The rearing was maintained until 60 DAR (days after the male’s pouch release), and the isotopic study was carried out on the periods of feeding on Artemia nauplii (6–60 DAR in diet A6; 11–60 DAR in diet A11).

2.3. Sampling and Analyses

Samples of juveniles were regularly collected to determine carbon (δ¹³C) and nitrogen (δ¹⁵N) isotope values, the elemental concentration of C and N, wet weight, and standard length. Samples of juveniles for stable isotope analysis (SIA) and weight and length measurements were randomly collected (n = 4 per diet) at 6, 11, 18, 25, 32, 46, and 60 DAR from each aquarium before the first daily feed administration. Samples of copepods and Artemia nauplii were also collected (n = 5) at different times along the experimental period, rinsed with distilled water, and kept frozen at −80 °C until further analysis.

Sampled juveniles were anaesthetized with tricaine methane-sulfonate (MS222; 0.1 g L⁻¹) (Sigma-Aldrich, Darmstadt, Germany), transferred to Petri dishes, photographed, and weighed individually on a Sartorius microbalance (± 0.01 mg). Standard lengths (SL) were measured (SL = head + trunk + curved tail) from digital photographs using image processing software (NIS Elements, Nikon Tokyo, Japan).

For SIA, whole seahorses were rinsed with distilled water, frozen at −20 °C, freeze-dried, dried in oven for 24 h at 60 °C, and homogenized (Mini Beadbeater-6018 homogenizer, BioSpec, Bartlesville, USA). The analyses were made in bulk seahorses on sub-samples of about 1 mg dry weight biomass. Due to potential alterations in δ¹³C and, to a lesser extent, δ¹⁵N values, it is recommended that samples with a high lipid content (commonly >5% weight or C:N > 3.56) [40] be defatted for SIA [41]. Since C:N values were higher than 5% in some samples (i.e., especially in the prey), we applied specific internal conversion factors for lipid normalization (0.940, 0.922 and 0.903 for δ¹³C in copepods, Artemia nauplii, and H. reidi juveniles, respectively; 1.370, 1.059, and 1.019 for δ¹⁵N in copepods, Artemia nauplii, and H. reidi juveniles, respectively)

δ¹³C and δ¹⁵N values and elemental composition (total C and N percentage) were analyzed at Servizos de Apoio á Investigación (SAI) of the University of A Coruña (Spain). Samples were measured by continuous-flow isotope ratio mass spectrometry using a FlashEA1112 elemental analyser (Thermo Finnigan, Monza, Italy) coupled to a Delta Plus mass spectrometer (FinniganMat, Bremen, Germany) through a Conflo II interface. Carbon and nitrogen stable isotope abundance was expressed as per mil (‰) relative to VPDB (Vienna Pee Dee Belemnite) and Atmospheric Air, according to the following equation:

δX = (R_sample/R_reference) − 1,

where X is ¹³C or ¹⁵N and R is the corresponding ratio of ¹³C/¹²C or ¹⁵N/¹⁴N. As part of an analytical batch run, a set of international reference materials for δ¹⁵N values (IAEA-N-1, IAEA-N-2, USGS25) and δ¹³C values (NBS 22, IAEA-CH-6, USGS24) were analyzed. The range of C:N ratios in sampled tissues (2.8–5.7) were within the range (0.4–6.9) of the reference materials used. The precision (standard deviation) for the analysis of δ¹³C and δ¹⁵N of the laboratory standard (acetanilide) was ± 0.15‰ (1-sigma, n = 10). Standards were run every 10 biological samples. The SIA procedure fulfils the requirements of the ISO 9001 standard. The laboratory is submitted to annual intercalibration exercises (e.g., Forensic isotope ratio mass spectrometry scheme–FIRMS, LGC Standards, Lancashire, UK).

2.4. Data Treatment

For comparative purposes among treatments, changes in δ¹³C and δ¹⁵N were studied by modelling the period comprising days 6 (diet A6) or 11 (diet A11) and 60, when only Artemia nauplii were offered. Isotopic data from seahorse juveniles were described applying two first-order one-compartment models [42] as functions of growth (relative dry weight increase) or development progress (days) [43,44]:

-

Growth-based model G [21]

The empirical equation that describes the isotopic changes occurring during growth is as follows:

δ = δ_eq + a W_R^c

where δ_eq is the model-fitted δ¹⁵N or δ¹³C isotopic ratio in equilibrium with the diet. W_R is the ratio between the weight attained at sampling times (W_t) and the weight when the food was switched (W_i), c is the metabolic decay constant indicative of the relative contribution of metabolic turnover to changes in isotopic ratios, and a is a constant provided by model-fitting. The value of a is the difference between the initial isotopic value (when the food was switched) and the equilibrium isotopic value (a = δ_i − δ_eq). δ corresponds to the isotopic (δ¹³C or δ¹⁵N) value at weight W_t.

When c = −1, turnover is due to growth only (simple dilution model), whereas c-values < −1 indicate greater proportional contributions of metabolic turnover to overall isotopic shift, with more negative values representing greater contributions of metabolic turnover [21].

-

Development-based model D (adapted from [23])

Changes in stable isotope ratios were modelled as an exponential function of development progress. The model is represented as follows:

δ = δ_eq + a e^{−(m+k) t}

where δ, δ_eq, and a are as previously defined in Model G and t is the time (days) of feeding on the experimental diet, m is the model-fitted metabolic constant, and k is the growth rate parameter calculated for each duplicate considering dry weight changes from the day of diet shift (days 6 or 11) to the final experimental day (day 60). The growth rates k were calculated at each sampling day as:

k = ln(W_R)/t

As in most studies using the model-fitting from [23] equation (model D), we assumed that growth and metabolism interact independently even though it is known that body size and metabolism covariate [45,46].

To determine half-life (G₅₀ or D₅₀) or equilibrium (G₉₅ or D₉₅) tissue turnover, the equations were solved for α = 50% or 95%, respectively. The x-fold increase in dry weight (G_α) and the days (D_α) required to attain a given percentage tissue turnover were calculated as:

G_α = e^{ln (1−α/100)/c} (Model G),

see [47]

D_α = ln (1 − α/100)/(m + k) (Model D),

see [48].

k, G_α, and D_α values were calculated for each experimental group.

The relative contribution of tissue turnover derived from growth (P_g) and metabolism (P_m) was calculated as follows:

P_g = 2(G₅₀ − 1)/G₅₀

P_m = (2 − G₅₀)/G₅₀

Diet–tissue discrimination factors (Δδ) for δ¹³C and δ¹⁵N were estimated as the difference between the fish tissue in equilibrium and diet (Δδ = δX_eq − δX_diet) [42].

Values are provided as mean ± standard deviation. A Shapiro–Wilk test was used to test for the normality of variables. Analysis of variance (ANOVA Univariate General Linear Model) was applied to estimate the effects of diet on survival, growth parameters, and isotope data. When significant differences were found at an alpha value of 0.05, Tukey’s HSD post-hoc test was applied to determine the significance of pairwise differences. Statistica 8.0 (StatSoft, Tulsa, OK, USA) software was used to perform statistical analyses and model-fitting.

2.5. Bioethics

Animal maintenance and manipulation practices were conducted in compliance with all bioethics standards on animal experimentation of the Spanish Government (Real Decreto 1201/2005, 10th October) and the Regional Government Xunta de Galicia (REGA ES360570202001/15/FUN/BIOL.AN/MPO01 and ES360570202001/16/EDU-FOR07/MPO01).

3. Results

3.1. Survival and Growth

Mortalities in juveniles occurred from DAR onwards in both treatments and a high mortality (21.7%) was noticed from 32 DAR to 46 DAR in diet A6 (

Table 1. Survival, dry weight (DW), standard length (SL), and C:N ratios during the experimental period in Hippocampus reidi juveniles submitted to dietary treatments A6 and A11. Data are provided as means (two batches per diet) and standard deviations (sd).

Diet	Days	Survival	DW (mg)		SL (mm)		C:N
	(DAR)	(%)	Mean	sd	Mean	sd	Mean	sd
A6	6	100	0.80	0.02	12.33	0.32	5.75	0.00
	11	98.8	1.30	0.39	15.95	2.52	3.03	0.06
	18	90.4	3.33	0.07	20.19	0.10	3.20	0.03
	25	75.7	9.28	2.30	26.41	4.12	3.07	0.37
	32	70.9	10.50	1.66	29.32	2.34	2.88	0.04
	46	49.2	15.61	4.17	31.81	5.07	2.80	0.01
	60	47.9	22.78	3.11	35.01	1.80	2.85	0.04
A11	11	98.8	2.20	0.14	18.15	0.35	3.14	0.02
	18	95.2	3.53	0.85	22.05	2.39	3.07	0.06
	25	85.7	10.63	0.04	26.74	0.40	3.29	0.10
	32	85.4	17.38	6.29	33.82	4.66	3.14	0.20
	46	84.5	22.85	2.58	36.94	3.08	3.05	0.17
	60	77.4	65.85	7.41	51.21	5.87	2.91	0.08

; Figure 1). The highest final survival was attained with diet A11 (77.4 ± 28.6%), but it did not differ significantly from that in diet A6 (47.9 ± 20.6%) due to the large standard deviations (Tukey HSD test; p = 0.723).

Growth rates (G_R; mg day⁻¹) in seahorse juveniles were estimated from diet shifting (11 DAR in diet A1; 6 DAR in diet A6) to the end of the experimental period (60 DAR) (Figure 1). Final dry weights and final standard lengths for diet A11 (65.85 ± 7.41 mg; 51.21 ± 5.87%, respectively) were significantly higher than in diet A6 (22.78 ± 3.11 mg; 35.01 ± 1.80%, respectively) (Tukey HSD test; p < 0.0001 and p < 0.001, respectively) (Table 1; Figure 1).

C:N ratios declined significantly from 6 DAR to 11 DAR (ANOVA; F_(6,14) = 254.7, p < 0.001) and subsequently decreased slightly but not significantly until 60 DAR (2.85 ± 0.04 in A6; 2.91 ± 0.08 in A11) (Table 1). The ratios in A11 were significantly higher than in A6 (ANOVA; F_(1,14) = 4.95, p = 0.043).

3.2. Isotopic Variation with Ontogeny

Mean δ¹³C and δ¹⁵N values in copepods were −21.1 ± 0.7‰ and 5.2 ± 0.4‰, respectively. Mean δ¹³C and δ¹⁵N values in Artemia nauplii were −18.0 ± 1.9‰ δ¹³C and 12.3 ± 0.8‰ δ¹⁵N, respectively.

Isotopic values (δ¹³C and δ¹⁵N) increased progressively from first feeding on Artemia (11 DAR in A11; 6 DAR in A6) until the end of the experiment (60 DAR) (Figure 2). The shift from copepods to Artemia was characterised by sharp and fast enrichments in δ¹³C and δ¹⁵N during the first 5 days in diet A11 (1.3 and 2.7‰ increase, respectively), and especially in diet A6 (4.4 and 3.2‰, respectively). Afterwards, the enrichment was gradually reduced until reaching δ¹³C and δ¹⁵N final values of −15.05 and 14.63‰ for diet A11 and −14.59 and 14.98‰ for diet A6 (Figure 2). Final isotopic values in both treatments were not significantly different for δ¹³C (Tukey HSD test; p = 0.832) or for δ¹⁵N (Tukey HSD test; p = 0.994).

3.3. Turnover and Discrimination Factors

Fitting of isotopic values to growth (G) and time-based models (D) provided very high goodness-of-fit (R²) (

Table 2. Parameter estimates (mean ± SE) for best-fit models for growth and time based-models predicting δ¹³C and δ¹⁵N values in Hippocampus reidi juveniles in the periods 6–60 DAR (Diet A6) and 11–60 DAR (Diet A11) (Artemia nauplii feeding). Turnover rates: G₅₀, G₉₅ (x-fold increase in biomass) and D₅₀, D₉₅ (Days). Δδ: Trophic discrimination factor (Δδ = δX_eq − δX_diet).

δ13C	Model G	δYeq	SE	c	SE	R2	G50	G95	D50	D95	Δδ
	A6	−16.3	0.5	−1.474	0.489	0.831	1.6	7.6	8.4	36.3	1.7
	A11	−15.7	0.4	−1.037	0.103	0.776	2.0	43.7	10.1	74.4	3.4
	Model D	δYeq	SE	m	SE	R2	G50	G95	D50	D95	Δδ
	A6	−16.2	0.9	0.040	0.070	0.816	2.5	8.8	6.8	29.4	1.8
	A11	−15.6	0.8	0.035	0.047	0.828	2.0	8.5	6.6	28.6	2.4
δ15N	Model G	δYeq	SE	c	SE	R2	G50	G95	D50	D95	Δδ
	A6	14.1	0.3	−1.326	0.269	0.934	1.7	9.6	9.3	40.3	1.8
	A11	14.9	0.5	−0.844	0.176	0.959	2.3	34.8	12.4	57.3	2.6
	Model D	δYeq	SE	m	SE	R2	G50	G95	D50	D95	Δδ
	A6	14.2	0.5	0.041	0.018	0.916	2.5	8.6	6.7	28.9	1.9
	A11	14.6	0.1	0.030	0.008	0.944	2.0	9.4	6.9	30.1	2.3

; Figure 3 and Figure 4). The isotopic equilibrium in challenged juveniles was reached with higher δ¹³C (less negative) and δ¹⁵N values in diet A11 than in diet A6. Both fitting models performed similarly.

For comparative purposes among fitting models, G₅₀ and G₉₅ turnover estimates from Model G were transformed into chronological turnover rates (D₅₀ and D₉₅; days) (Table 2). A similar transformation was applied to convert D₅₀ and D₉₅ values in G₅₀ and G₉₅ estimates. For that, the relationship between chronological time of development and juvenile weight for each experimental batch was considered.

Regarding seahorse weight, models G and D provided similar half-life (G₅₀) and near-complete turnover (G₉₅) estimates (Table 2; Figure 3). Half-life G₅₀ ranged from 1.6 to 2.5-fold increase in weight for δ¹³C and from 1.7 to 2.5- fold increase in weight for δ¹⁵N. Chronologically (model D), isotopic turnovers D₅₀ and D₉₅ were lower than those from model G (Table 2) for both isotopes, especially in diet A11 (Table 2; Figure 4). Half-life D₅₀ ranged from 6.6 to 10.1 days for δ¹³C and from 6.7 to 12.4 days for δ¹⁵N.

According to model G, the isotopic equilibrium after the dietary shift in diets A6 and A11 would be reached after 36.3 and 74.4 days for δ¹³C and after 40.3 and 57.3 days for δ¹⁵N, respectively (Table 2). For model D, the equilibrium would be reached in about one week (W_R range: 8.5–9.4), independently of the isotope and diet considered.

Except for δ¹³C in model G–diet A11, the discrimination factors resulting from models D and G were rather similar for each isotope, ranging from 1.7‰ to 2.4‰ for δ¹³C and from 1.8‰ to 2.3‰ for δ¹⁵N (Table 2). The discrimination factor estimated for δ¹³C in model G–diet A11 (3.4‰) was overestimated compared to the other estimates (2.3–2.6‰). Average discrimination factors considering both fitting models were 1.8 ± 0.1‰ for δ¹³C and 1.9 ± 0.1‰ for δ¹⁵N in diet A6, and 2.9 ± 0.7‰ for δ¹³C and 2.5 ± 0.2‰ for δ¹⁵N in diet A11 (Table 2; Figure 5).

Considering the estimates from model D of the metabolic (P_m) and growth (P_g) contributions, the model revealed that there was not any contribution of δ¹³C and δ¹⁵N to metabolism both in diet A6 and A11 (Figure 6). These results partially agreed with those in model G, but the simple dilution model revealed moderate metabolic contributions of δ¹³C (25%) and δ¹⁵N (19%) in diet A6.

4. Discussion

An extended feeding on copepods before the introduction of Artemia nauplii enhanced the performance in early developing seahorse H. reidi juveniles. Growth rates were notably affected and even though survivals did not differ significantly, they were systematically higher in juveniles fed on copepods for 10 days (diet A11) compared to those fed on that prey for only 5 days (diet A6). Those findings were partially related to both the nutritional characteristics of the experimental prey used in this study and the limited digestive capabilities in early developing juveniles.

Even though Artemia nauplii is one of the most used prey in the rearing of seahorse juveniles [34], it is well known that growth and survival are enhanced when copepods are included in the initial feeding [36,49,50,51]. Numerous publications refer to the high nutritional quality (e.g., n-3 HUFA and others) of copepods and their wide range in size. The characteristics of the prey used in the present study are provided in [36].

Due to the almost complete exhaustion of yolk reserves in newborn H. reidi [52] and the rapid adaptation to exogenous feeding, the juveniles undergo a fast initial change towards dietary isotopic values (e.g., copepods) [53]. Previous studies have shown that when organisms are provided with a new diet isotopically different from the previous diet, their tissues will eventually reflect the isotopic signature of the new diet [24,43,54]. From 6 DAR in diet A6 and from 11 DAR in diet A11 until the end of the experimental period (60 DAR), seahorse juveniles were fed on Artemia nauplii. The results revealed that both feeding schedules led to similar isotopic patterns with fast and continuous progression towards the dietary isotope values. In H. reidi, significant changes occur in gut morphology and physiology in 12–15 DAR juveniles [52] before the transition from planktonic to benthonic lifestyle (about 20 DAR) [55]. This finding agrees with those in other seahorse species, in which the development of the first intestinal loop and mucosal folding occurs at that age [56,57]. From that age, the intestinal absorption surface progressively increases [56]. Those changes should lead to better digestive efficiencies and significant enhancements in assimilation capabilities from that age onwards.

Isotope turnover rates were estimated from data fitting to models as functions of both body mass and time. The results from model D indicate that both dietary treatments performed similarly but not identically, especially for δ¹³C turnover (see fitted curves in Figure 4). However, inter-treatment differences were revealed by model G (body mass gain). A positive relationship between stable isotope turnover rates and body weight has previously been highlighted [44]. Moreover, the role of body size or weight change in the isotopic turnover rates of fish tissues is arguably more important than time or age. This is because fish growth rates are indeterminate and highly variable, being influenced by a range of abiotic (e.g., water temperature) and biotic (e.g., food availability) factors [44,58]. That statement was confirmed by our results (model G vs. model D), especially considering inter-treatment differences in G₉₅ or D₉₅ for δ¹³C.

The isotopic equilibrium was reached before the end of the experiment period in all cases, except for δ¹³C in diet A11–model G (about 85.4 days). Compared to diet A6, the equilibrium estimates (δ_eq) were lower in juveniles from diet A11, likely due to differences in initial diet–tissue isotopic values at diet shift (see Figure 2). Even though there is a broad variation of turnovers rates in fish larvae and juveniles, it has been suggested that tissues are in equilibrium with the diet after four to five half-lives [22,59,60], which agrees with our results (see D₅₀ vs. D₉₅ in Table 2).

The present study demonstrated the influence of initial isotopic values (i.e., previous feeding history) on discrimination factors. Even though average discrimination estimates in H. reidi juveniles from diet A11 (2.9‰ for δ¹³C and 2.5‰ for δ¹³N) were higher than in those from diet A6 (1.8‰ for δ¹³C and 1.9‰ for δ¹³N), the estimates agree with those in other fish larvae. However, differences in discrimination factors between both dietary treatments could rely on differences in juvenile condition at the onset of the experimental periods. In adult fish, trophic discrimination factors are commonly around 1.7‰ for δ¹³C and 2.5‰ for δ¹⁵N [61]. In marine larvae and early post-larvae, the discrimination ranges are highly variable, ranging broadly from 0.4 to 4.1‰ for δ¹³C and from 0.1 to 5.3‰ for δ¹³N (see review [62]). In Sciaenops ocellatus larvae, factor estimates were 1‰ for δ¹³C and 1.6‰ for δ¹⁵N [24]. The discrimination factor for δ¹⁵N in reared post-flexion larvae of Thunnus thynnus was 0.4‰ [63], much lower than in wild-caught adults maintained in captivity (1.6‰) [64]. In Pleuronectes americanus larvae, discrimination factor estimates were 0–2.5‰ for δ¹³C and –0.3‰ at 18°C and 2.2‰ at 13°C for δ¹⁵N [26].

While models G and D differed partially in the proportions of isotopic tissue turnovers attributed to growth and metabolism, both models revealed that tissue turnovers in juveniles from diet A11 were fully related to weight gain. On the contrary, the turnover rates in juveniles from diet A6 partially directed the energy to the production of new tissues, reflecting a worse nutritional condition than those in diet A11. This statement would be the consequence of limited digestive capabilities and lower nutritional conditions in the former as a result of the advanced feeding on a non-optimal prey such as Artemia. The dilution of an initial carbon or nitrogen pool by the addition of newly deposited biomass would not be necessary for juveniles from diet A11.

Percentage estimates for growth turnover in juvenile seahorses were lower than those reported in the larvae/juveniles of many fish species [62] but agreed with the low isotopic contribution to metabolism reported for early developing fishes. Hesslein et al. [23] examined isotopic patterns in cultured broad whitefish (Coregonus nasus) juveniles in response to a dietary shift and attributed 90% of the observed isotopic changes to growth. Fry and Arnold [21] investigated δ¹³C shifts in juvenile brown shrimp (Penaeus aztecus) and reported that biomass gain was the primary cause of change in isotopic composition, although a low added effect of metabolic turnover was also detected. Reported estimates for growth contribution were 90% in red drum Sciaenops ocellatus [24], 61–79% in chub Squalius cephalus, 56–71% in roach Rutilus rutilus, and 42–51% in the muscle of grass carp Ctenopharyngodon idella juveniles [60]. However, other fish species (e.g., Fundulus heteroclitus juveniles) showed a high metabolic contribution to isotopic turnover [65]. A whole energy budget study carried out in the larvae of the flatfish Scophtalmus maximus reported a progressive increasing food absorption efficiency, accompanied by a significant decrease in the energy channelled to the metabolism (from 71 to 36%) with growth [66]. Our findings are consistent with the higher contribution of metabolism to isotopic change with a decreasing growth rate (e.g., diet A6) [21,44,67,68].

An important question arises regarding the applicability of our results to ex situ production systems: How long should copepods be offered to seahorse juveniles before switching to another prey (e.g., Artemia nauplii)? The availability of appropriate prey in the early life stages of seahorses is of paramount importance since one of the most critical factors in developing seahorses is the low digestion capability in early feeding juveniles [35,57], especially when fed on Artemia in rearing systems [47,48]. That limitation has been supported visually by direct observations of feces (occurrence of undigested Artemia nauplii) and physiologically by the absence of supranuclear vesicles in the intestine in juveniles fed exclusively on Artemia [36]. Furthermore, copepods are highly preferred to Artemia nauplii during the first two weeks of development [51] and can be mechanically broken into smaller pieces, improving the action of digestive enzymes [36]. The results achieved on biological indicators such as growth and mortality as well as on the contribution of growth and metabolism to tissue turnover demonstrate that extending the feeding on copepods from 6 DAR to 11 DAR enhanced the overall condition and welfare of H. reidi juveniles. Due to the high cost of copepod production, a co-feeding regime including copepods–Artemia has been proposed for large-scale ex situ production [69].

In summary, the present study provided new insights for the understanding of food assimilation and growth in early developing Hippocampus reidi juveniles in response to dietary shift. Our results highlighted the importance of copepods as the first prey, confirming that their feeding by seahorse juveniles should be extended as long as possible before the inclusion of Artemia nauplii on the feeding schedule. Longer periods of initial feeding on copepods would result in higher growth rates and survival by promoting juvenile welfare and a better nutritional condition. It can be concluded that the diet is an important factor contributing to daily variations in carbon and nitrogen stable isotopes profiles in juveniles and that the feeding history has implications on discrimination factors and to a lesser extent on isotopic turnover rates. Meanwhile, the elucidation of the role of growth and metabolism on stable isotope turnovers deserves further investigation to characterize the trophic dynamics more precisely in the species. To our knowledge, this is the first study on stable isotope turnover in seahorses, providing specific diet–tissue discrimination factors for δ¹³C and δ¹³N in seahorse juveniles.