Lepidocephalichthys thermalis, commonly known as Ayirai meen or the Indian spiny loach, is an omnivorous freshwater fish that naturally feeds on algae, detritus, insect larvae and small benthic invertebrates and represent a behaviourally active group of tropical species frequently associated with sandy or gravel-dominated substrates in lowland riverine and wetland habitats (Balaganesan et al., 2018). Currently, land-based cultivation of small indigenous fish species remains largely underexplored, despite their strong market demand and high economic value. Their naturally small body size enables production using minimal water resources and limited culture space (Velmurugan et al., 2024). However, the lack of standardised culture techniques and nutritionally optimised feed formulations remains a critical constraint for large-scale production of L. thermalis (Velmurugan et al., 2024).
       
To bridge this knowledge gap, it is necessary to evaluate suitable aquaculture production systems that can support the physiological health and stress tolerance of L. thermalis. In the present study, five culture systems, namely FRP tanks, raceway tanks, biofloc tanks, cement tanks and lined ponds, were selected because they represent commonly used land-based aquaculture systems with distinct operational and environmental conditions. Freshwater cultured fish, such as Labeo rohita, typically exhibit variable proportions of saturated, monounsaturated and polyunsaturated fatty acids, which are largely influenced by dietary inputs and environmental conditions (Relekar et al., 2025). Polyunsaturated fatty acids are crucial nutrients in aquaculture because they influence key physiological functions in fish, including metabolic regulation, improved resilience to environmental stress and proper reproductive performance (Thiruvasagam et al., 2024). Highly unsaturated fatty acids are recognized as fundamental dietary components that promote healthy growth, survival and normal physiological development across numerous marine and freshwater fish species (Tocher, 2015).
       
Digestive enzymatic activity represents an essential parameter for assessing an organism’s capacity to efficiently metabolize and assimilate nutrients (Manimaran et al., 2025). These enzymatic patterns in Oreochromis niloticus are shaped by culture system conditions, with observed variations associated with feeding behaviour, water quality and microbial community availability (Monroy-Dosta et al., 2025). Antioxidant biomarkers serve as critical determinants of piscine physiological status and exhibit marked variation in response to the aquaculture production framework, encompassing pond-based, recirculating and integrated culture technologies (Turlybek et al., 2025). Cortisol serves as a quantifiable biomarker of the primary neuroendocrine stress response. It functions as a central endocrine modulator in fish, where sustained elevation suppresses protein biosynthesis while enhancing proteolytic processes, ultimately constraining somatic growth (Schreck et al., 2016). Accordingly, due to its high consumer demand and significant economic value in the Indian local market, the objective of this study was to optimize aquaculture practices for L. thermalis by identifying an appropriate production system and evaluating its effects on the fatty acid profile, digestive efficiency, antioxidant capacity and cortisol levels.

Ethical statement
 
The present study was conducted following the guidelines issued for care and use of animals in scientific research by the committee for the purpose of control and supervision of experiments on animal (CPCSEA), ministry of environment and forests (Animal welfare division), Government of India. Further, this study was approved by ethical committee of Tamil Nadu Dr. J. Jayalalithaa Fisheries University, Nagapattinam (TNJFU, 2024), Tamil Nadu, India.
 
Experimental setup
 
The present study was conducted at the Theni Centre for Sustainable Aquaculture, Directorate of Sustainable Aquaculture, Theni, Tamil Nadu, India during 2025-26. Healthy loach seeds with a uniform mean body mass of 0.14±0.00 g per individual were selected from the facility’s breeding unit. Five different production systems were used for a 90-day study: FRP tanks (4.3 m²), raceway tanks (4.3 m²), biofloc tanks (4.3 m²), cement tanks (12.3 m²) and lined ponds (50 m²).
       
After acclimatization, the selected L. thermalis seeds were divided into five experimental treatments consisting of FRP tanks (T4), raceway tanks (T5), bio floc tanks (T6), cement tanks (T7) and lined ponds (T8), stocked at a density of 300 individuals/m². As no published information is currently available regarding the optimal stocking density for L. thermalis in these rearing systems, the adopted stocking ratio was chosen empirically and may represent a limitation of the study. Each group was reared in duplicate for twelve weeks and the loaches were fed a formulated diet.
 
Diet formulation and administration
 
The formulated diet consisted of incorporated fish meal, tapioca flour, groundnut oil cake, wheat flour, soybean meal, fish oil, rice bran, mono calcium phosphate, salt and vitamin-mineral premixes, resulting in a final composition of 35.9% crude protein and 4.5% crude lipid. All components, excluding the vitamin-mineral premix, were homogenized, shaped into a cohesive mass and subjected to thermal processing in a pressure cooker for 10-15 min, with intensive kneading ensuring uniform integration. Following cooling and aeration, the vitamin-mineral premix was incorporated into the processed matrix. The resulting dough was subsequently pelletized using a manual pelletizer, with each batch handled separately. Pellets were sun-dried individually and preserved in airtight containers until use. Feeding was administered at 10% body weight during the initial month, followed by a reduced ration of 5% biomass for the remainder of the culture duration. Growth sampling was performed at 14-day intervals using 20 fish from each treatment.
 
Fatty acid conversion
 
Lipids isolated from each loach specimen were utilized for fatty acid profiling, with extraction and derivatization conducted via a modified protocol based on Sukhija and Palmquist (1988). The approach has been previously validated for reproducible recovery of fatty acids. Frozen lipid samples were allowed to thaw under refrigerated conditions (4°C) overnight and subsequently homogenized using a personal vortex mixer (V-1 plus, peQLab, UK). An aliquot of 0.1 g from each sample was transferred into pre-cleaned SOVIREL tubes (treated with DeCon 90 and air-dried).
       
A volume of 1.7 mL of methanol: toluene (4:1, v/v) was introduced, followed by mixing. Within a fume hood, 250 µL of acetyl chloride was dispensed gradually using a glass pipette. The mixture was agitated for 30 s and incubated using a heating block system at 100°C for 1 hour (Techne DriBlock^R BD-3D). After thermal treatment, samples were cooled for 20 min, after which 5 mL solution of potassium chloride (5% w/v in distilled water) was added. The contents were gently mixed by inversion and then centrifuged for 5 min at 1000 g. The supernatant was carefully aspirated in a pipette of Gilson and relocated into amber glass vials fitted with inserts (Chromacol Ltd., Hertfordshire), followed by storage at 4°C until gas chromatographic analysis.
 
Gas chromatography
 
The prepared extracts were subjected to analysis using the gas chromatograph of Shimadzu GC-2014, with helium as a carrier gas and isolation achieved on an SGE Forte BPX70 column. Peak identification was accomplished through comparison with an external standard Supelco FAME Mix C4-C24. The oven temperature program commenced at 50°C with a 1 min hold, followed by a gradual increase at 2°C min^-1 to 188°C, held for 600 s, then elevated at the same rate to 240°C, with a 240 s hold, before returning to baseline conditions. Quantification was performed by aligning sample peak areas with those of corresponding reference standards and individual fatty acids profile were reported as proportions of the quantified total fatty acids.
 
Tissue homogenate preparation for determination of digestive enzymes
 
Specimens were randomly sampled at the conclusion of each experimental trial. A 5% tissue homogenate was prepared from each treatment group using a mortar and pestle under maintained low-temperature conditions. The homogenates were then centrifuged at 10,000 rpm for 10 min at 4°C. The resulting supernatant was carefully collected into 5 mL tubes and preserved at -20°C for subsequent enzymatic assays. Prior to analysis, aliquots were diluted according to established digestion protocols (Skea et al., 2005).
 
Protease
 
Protease activity was quantified using a casein hydrolysis method (Drapeau, 1976; Joshna et al., 2024). The reaction mixture consisted of 1% casein prepared in 0.05 M Tris-phosphate buffer (pH 7.8) and pre-equilibrated for 5 min at 37°C, prior to addition of tissue homogenate to initiate the reaction. After incubation for 10 min, the reaction was terminated by adding 10% trichloroacetic acid, followed by filtration of the mixture. To account for baseline activity, a control was prepared without incubation by adding the tissue homogenate. Enzyme efficiency was expressed as the amount required to increase the absorbance by 0.001 per minute at 280 nm under the assay conditions (37°C, pH 7.8).
 
Lipase
 
Using the titrimetric method described by Cherry and Crandall (1932), lipase enzyme activity was quantified, by measuring fatty acids liberated in a stabilized olive oil emulsion through enzymatic hydrolysis of triglycerides. The crude enzyme extract was evaluated by titrating the released fatty acids against standardized sodium hydroxide. The assay system comprised 1.5 mL of 0.1 M Tris-HCl buffer (pH 8.0) and 1.5 mL of stabilized lipase substrate, followed by the addition of crude enzyme extract (1.0 mL). The reaction mixture was maintained at 27°C for 24 h, after which 3 mL of 95% ethanol was added to terminate the reaction. As an indicator, phenolphthalein (0.9%, w/v) was employed and titration was performed with 0.05 N NaOH until the appearance of persistent pink colour at endpoint.
 
Amylase
 
Based on Priyatharshni et al., (2024), amylase activity was evaluated using dinitro salicylic acid method by quantifying reducing sugars. The assay mixture comprised of tissue homogenate, phosphate buffer (pH 7.0) and starch substrate of 1% (w/v), followed by incubation for 30 min at 37°C. Following incubation, dinitrosalicylic acid reagent was added. For 5 min and the mixture was heated in a boiling water bath. After cooling to room temperature, the reaction mixture was diluted with distilled water and absorbance was measured at 540 nm. Maltose was used as the standard and enzymatic activity was expressed as the amount of maltose liberated from starch per minute at 37°C.
 
Preparation of tissue homogenate for the determination of antioxidant enzymes
 
Whole fish samples were homogenized on a wet weight basis using a handheld homogenizer with 0.25 M sucrose solution at pH 7.0 at a ratio of 1:20 weight per volume. The homogenized samples were maintained at 4 to 5°C in an icebox. The mixture was then centrifuged for 10 minutes at 4°C and 5000 rpm in a refrigerated centrifuge. Then, supernatant was carefully collected, transferred into sterile vial pipes and kept at minus 20°C for subsequent analysis (Oyedemi et al., 2010).
 
Total protein concentration in tissue
 
Protein concentration of the homogenate was quantified using Lowry et al. (1951), employing bovine serum albumin acting as the standard reference. The samples’ absorbance was measured at 660 nm and concentrations were determined from a standard calibration curve.
 
Glutathione peroxidase (GPx) activity
 
GPx efficiency was assessed using the method described by Paglia and Valentine (1967). The reaction mixture consisted of 200 µL tissue homogenate, 200 µL phosphate buffer, 200 µL EDTA (ethylenediaminetetraacetic acid), 100 µL sodium azide and 100 µL reduced glutathione and was incubated for 10 minutes at 37°C. Subsequently, 100 µL of hydrogen peroxide and 100 µL of nicotinamide adenine dinucleotide phosphate were added to initiate the reaction. After incubation, the reaction was terminated with 500 µL of 10 per cent trichloroacetic acid, followed by centrifugation for 5 minutes at 10,000 rpm. An aliquot of 1000 µL of supernatant was transferred to a test tube, mixed with 2000 µL of tris buffer and 0.05 mL of DTNB (5,5-dithiobis (2-nitrobenzoic acid)) and absorbance was recorded at 412 nm over 3 minutes. Enzyme activity was shown as units/min/mg protein, where one unit represents the amount of enzyme required to oxidize 0.001 mmol/L of reduced glutathione per minute.
 
Catalase activity
 
For the estimation of catalase activity, whole fish samples were collected and analyzed following the method explained by Takahara et al., (1960) and Kavitha et al., (2025). Phosphate buffer (2450 µL) was added to 0.05 mL of tissue homogenate and the reaction was initiated by adding hydrogen peroxide solution (1000 µL). The reduction in absorbance was monitored at 240 nm at 15 second intervals for 3 minutes. Enzyme and phosphate buffer blank were run simultaneously. Catalase activity was expressed as µmol of hydrogen peroxide decomposed per minute per milligram of protein, with one unit described as the amount of enzyme required to degrade 1 µmol of hydrogen peroxide per minute.
 
Superoxide dismutase (SOD) activity
 
Superoxide dismutase activity was evaluated following the method described by Kono (1978). A sample volume of 70 µL was used for the assay in combination with the respective reagents. Sodium carbonate solution was obtained by dissolving 0.529 g in 100 mL of distilled water. Nitro blue tetrazolium solution (0.096 M) was obtained by dissolving 0.004 g in 50 mL distilled water, while Triton X 100 solution was obtained by dissolving 0.06 g in 100 mL distilled water. Hydroxylamine hydrochloride (20 mM) was obtained by dissolving 0.138 g in 100 mL of distilled water and the pH was adjusted to 6.0. Absorbance was recorded at 540 nm for 5 minutes at room temperature.
 
Determination of cortisol hormone
 
Cortisol levels were quantified using an enzyme-linked immunosorbent assay (ELISA) following the method described by Zhang et al., (2020). Tissue samples were diluted in chilled phosphate-buffered saline and centrifuged for 20 minutes at 3000 rpm under refrigerated conditions. The supernatant was collected and analysed for cortisol using Sigma ELISA kits according to the manufacturer’s instructions.
 
Statistical analysis
 
Statistical analyses were performed using SPSS version 20.0 for Windows (SPSS Inc., Chicago, IL, USA). All experiments were conducted in duplicate (n=2), the data are presented as Mean±Standard deviation (SD). Differences among treatments were analysed using one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post hoc test at a significance level of P<0.05. Statistical outputs (Table 1) including F-values, degrees of freedom (df) and effect sizes (η²) were calculated to determine the magnitude of treatment effects.

Fatty acid profile
 
Significant differences in fatty acid profile (Fig 1) were observed among different production groups (P<0.001). In terms of arachidonic acid, the T5 and T6 groups were not significantly different. L. thermalis produced under lined pond groups had significantly elevated levels of fatty acid profile (µg g-1 crude lipid) such as alpha-linolenic acid (APA) (14.42±0.02), eicosapentaenoic acid (EPA) (14.10±0.02), docosahexaenoic acid (DHA) (20.04±0.02), linoleic acid (LA) (67.63±0.02) and arachidonic acid (AA) (6.01±0.02).


 
Digestive efficiency
 
The digestive enzyme efficiency (Fig 2) differed significantly across the various production groups (P<0.001). However, the T4 and T5 groups did not differ significantly from one another in lipase and amylase activity. The lined pond groups produced L. thermalis had significantly elevated levels of digestive enzymes (U mg^-1 protein) such as protease (59.12±0.010), lipase (6.42±0.010) and amylase (1.72±0.010).


         
Antioxidant capacity
 
Analysis revealed significant differences in antioxidant enzyme capacity (Fig 3) among the different production groups (P<0.001). The lined pond groups produced L. thermalis had significantly higher levels of antioxidant capacity (U mg^-1 protein) such as GPx (5.4±0.03), catalase (5.9±0.01) and SOD (5.8±0.01). 


 
Cortisol hormone regulation
 
Significant differences in cortisol hormone level (Fig 4) were observed across the different production groups (P<0.001). The lined pond groups of L. thermalis exhibited a reduced cortisol level (3.240±0.02 µg dL^-1).


 
Influence of culture systems on fatty acid composition
 
The fatty acid profile of muscle tissues in species such as Salmo salar, Oncorhynchus mykiss and Sparus aurata is primarily governed by the lipid sources included in formulated aquaculture diets (Tocher, 2015) and in this context, feed was formulated to attain a final dietary lipid level of 4.5% as verified through lipid analysis, thereby providing a basis for evaluating variations in muscle fatty acid composition across different production groups. Elevated concentrations of ALA in cultured fish enhance the nutritional value of fish flesh. Because, this fatty acid function serves as a precursor of metabolic for long chain fatty acids of omega 3 such as EPA and DHA (Calder, 2018). The fatty acid profile of cultured fish, including DHA and EPA, is largely determined by feed formulation and aquaculture management practices rather than by species characteristics alone (Glencross et al., 2007).
       
In aquaculture environments where natural food availability is restricted, particularly in lined pond systems, fish depend largely on formulated feeds, which facilitates greater deposition of dietary DHA, EPA and LA in muscle tissues (Tacon and Metian, 2015), a pattern that corresponds with the elevated DHA, EPA and LA levels observed in the muscle of L. thermalis cultured under the lined pond groups in this investigation. Aquaculture environments with greater production control, such as lined pond systems, rely predominantly on formulated feeds that facilitate the effective incorporation of dietary AA into fish tissues (Tacon and Metian, 2015), corresponding with the elevated levels of AA observed in the L. thermalis reared under lined pond groups. In summary, fatty acid profiles in cultured L. thermalis are predominantly shaped by lined pond groups facilitating enhanced flesh quality through more efficient assimilation of dietary lipids into muscle tissue.
 
Influence of culture systems on digestive efficiency 
             
Indian major carps such as Labeo rohita are known to exhibit elevated intestinal protease activity when reared in semi intensive and intensive pond systems with protein rich diets (Das and Tripathi, 1991) and similarly, Cyprinus carpio displays a dynamic protease enzyme system that responds to dietary protein variations under pond based aquaculture conditions (Hidalgo et al., 1999). In lined pond biofloc culture systems, species such as Liza ramada demonstrate markedly higher intestinal lipase activity than those maintained in clear water environments, which is attributed to enhanced microbial biomass and improved nutrient recycling (Zaki et al., 2025). Amylase activity in Oreochromis niloticus reflects an efficient capacity for carbohydrate digestion in pond-based aquaculture systems, largely attributed to its omnivorous feeding behaviour (Santos et al., 2016). In agreement with these observations, the lined pond groups exhibited elevated digestive efficiency such as protease, lipase and amylase, thereby facilitating superior nutrient assimilation in L. thermalis.
 
Influence of culture systems on antioxidant capacity
 
Increased activity of GPx, catalase and SOD indicates enhanced antioxidative capacity and superior physiological condition in Oreochromis niloticus (Yu et al., 2023). Aquaculture systems that restrict sediment interaction and preserve stable water quality, in conjunction with strategic nutritional enhancement, can synergistically elevate GPx activity in cultured species such as Labeo rohita and Oreochromis niloticus (Dawood et al., 2018). Increased catalase activity in Oreochromis niloticus reared under intensive pond conditions reflects an adaptive antioxidant response to hydrogen peroxide arising from variable water quality (Abdel-Tawwab et al., 2019). According to Lushchak (2011), elevated SOD activity in intensive aquaculture systems, including lined ponds, promotes enhanced growth performance by alleviating oxidative damage linked to heightened metabolic demand in freshwater species. Consistent with these observations, a pronounced increase in antioxidant capacity, including GPx, catalase and SOD levels was observed in L. thermalis reared within lined pond groups, reflecting a strengthened defense against oxidative stress and an improved physiological status in the present study.
 
Influence of culture systems on cortisol level
 
In Oncorhynchus mykiss, elevated cortisol concentrations are indicative of prolonged stress exposure, leading to growth suppression (Wendelaar, 1997). Findings from Fjelldal et al., (2011) and El-Sayed (2006) demonstrate that refined culture management attenuate cortisol secretion in Salmo salar and Oreochromis niloticus; accordingly, comparative evaluation of distinct production systems identifies those that promote low cortisol level as conducive to stress-resilient growth; within this framework, the lined-pond groups are associated with reduced cortisol concentrations in L. thermalis relative to other culture settings. Overall, this pattern indicates that such regulated culture systems promote improved physiological stability by limiting energy allocation toward stress responses.

The present study demonstrates the significant role of the production system in L. thermalis culture. However, the present study was conducted under specific culture conditions and seasonal variability was not assessed. Although economic feasibility was calculated, detailed evaluation in relation to water requirements was beyond the scope of the present investigation and will be addressed in a separate study. Overall, lined pond system exhibited significantly higher fatty acid profiles together with elevated activities of digestive and antioxidant enzymes and reduced levels of cortisol levels, thereby enhancing the physiological and nutritional health of L. thermalis.

The authors sincerely thank Tamil Nadu Dr. J. Jayalalithaa Fisheries University, Directorate of Sustainable Aquaculture, Nagapattinam, Tamil Nadu, India, for the grants and facilities to conduct the experiment. We sincerely thankful to the Tamil Nadu State Government’s Loach Project, for the financial support.

The authors declare that they have no conflict of interest.