In the last decades, the increase in the world human population, which is increasingly consuming more fish and fish products due to their role as part of a healthy diet, has been the main driver for the growth in total fisheries and aquaculture (1). Indeed, recent estimates set an expected world fish production of 201 million tons by 2030, comprising an overall increase of 12.8% from 2018 to 2020, with an increasing share from aquaculture (2). The observed increase in fish consumption is translated into an increase in fish waste, being the majority of lost or unused biomass (e.g., skin, bones, heads, viscera, and trimmings) found during filleting and preparation of products for the final consumer (3). However, these by-products have the potential to originate high-value products that can be reintroduced into food and feed chains due to their richness in nutrients and bioactive compounds, such as proteins, bioactive peptides, polyunsaturated fatty acids (PUFAs), minerals, and enzymes (4, 5). Taking advantage of fish by-products has not only economic benefits but also counteracts the negative environmental impact of improper fish waste disposal, namely through the release of organic compounds into aquatic environments (5, 6). Indeed, the valorization of fish by-products is commonly considered the preferred option to manage fish waste, whereas when important pollutants are present, waste management processes should determine the fate of fish by-products (7). The application of the circular economy concept to the fishery field increases the rate of reuse and decreases the underutilization of fish waste, thus contributing to a reduction of the pressure on natural resources (3, 8), which is aligned with the Circular Economy Action Plan (9) launched by the European Union as part of the European Green Deal. Protein hydrolysates and oils rich in omega-3 PUFA, such as eicosapentaenoic (C20:5 n-3, EPA) and docosahexaenoic (C22:6 n-3, DHA) acids, are among the most common functional feed ingredients obtained from fish by- and co-products (BCPs) with great interest, among others, in companion animal feeding due to the current trends observed in this sector (6). Indeed, along with a growing worldwide pet population (10), owners are increasingly favoring sustainable functional ingredients that not only ensure energy and nutrient requirements but also benefit the health and wellbeing of their animals, contributing to a decrease in the risk of chronic disease development and an increase in life expectancy (11).

In this context, in vitro and in vivo studies, namely with dogs, have shown that protein hydrolysates provide amino acids more easily absorbed in the small intestine (12), with numerous biological activities such as hypoallergenic (13), anti-stress (14, 15), satiety (16), antihypertensive, antithrombotic, immunomodulatory, and antioxidative effects (17), as well as acting as palatability enhancers (13, 18). However, despite the high interest in protein hydrolysates and their use in commercial diets, studies on companion animals are scarce, especially with fish hydrolysates.

Regarding the long-chain omega-3 PUFA, EPA and DHA are daily recommended for fully-grown (19), growing, and reproductive dogs (19–21), being also considered to play a role in disease prevention and used in several therapeutic diets (e.g., cardiovascular, renal, gastrointestinal, orthopedic, and dermatological) at higher doses (22–26). Fish oil is considered the primary source of EPA and DHA (27), but due to sustainability reasons and also taking into consideration the Sustainable Development Goal of the United Nations Agenda 2030 to prevent overfishing, alternative sources of fish oil obtained from farmed and wild fish are required. Although several marine sources [e.g., microalgae, macroalgae, krill, copepods, single-cell organisms; (28, 29)] and genetically modified plant oilseeds [e.g., rapeseed, thale cress, and camelina; (30–33)] have been evaluated as EPA and DHA sources, the extraction of oil from fish waste originated from the processing stage might constitute a sustainable approach that also has the advantage of being able to be applied locally, contributing to a lower environmental impact. Therefore, the present study aimed to evaluate the effects of the inclusion of locally produced fish hydrolysate and oil obtained from fish waste in the substitution of imported shrimp hydrolysate (mainly from Ecuador, the major producer of shrimp) and salmon oil (mainly from the North of Europe, Norway, and Scotland, the main producers of salmon), on palatability, apparent total tract digestibility, fecal characteristics and metabolites, blood fatty acid profile, and coat quality of adult dogs.

Trials were approved by the Animal Ethics Committee of the School of Medicine and Biomedical Sciences, University of Porto, and licensed by the Portuguese General Directorate of Food and Veterinary Medicine (Permit No 0421/000/000/2021). Trained scientists in laboratory animal science (FELASA, category C) conducted the experiments in line with good animal welfare practices (European Union Directive 2010/63/EU).

The use of fish BCPs from the agrifood sector for the extraction of oil rich in omega-3 fatty acids and protein hydrolysates constitutes a sustainable solution to the increasing problem of fish waste disposal. If applied locally, this circular economy further contributes to lowering dependency on third countries and mitigating the environmental impact of the food sector. For the full exploitation of these potential high-value products, in vivo studies are needed. Therefore, this current study aimed to investigate the effects of fish hydrolysate and oil from fish wastes during the processing stage on palatability, apparent total tract digestibility, fecal characteristics and metabolites, blood fatty acid profile, and coat quality of adult dogs. The experimental diet containing fish hydrolysate and oil was compared to a commercial diet including shrimp hydrolysate, commonly used in fish diets (55, 56), but recently shown to have the potential to be included in dog diets (37), and salmon oil, a common oil source used in dog food. With shrimp hydrolysate and salmon oil mainly imported from third countries, the use of fish hydrolysate and oil locally produced might contribute to the increased sustainability of the pet food sector.

Both diets used in the current study ensured the energy and nutrient requirements of adult dogs, according to FEDIAF (19). The chemical composition of the diets reflected the composition of the hydrolysates and oils used, with the main differences being observed in ash, CP, and EE contents, and fatty acid profiles. The lower ash content of the experimental diet might be a favorable aspect of the selection of the studied sources for addition in high-digestibility diets, as a high ash content may compromise protein digestibility (57). Regarding the EE content and fatty acid profile, it is well known that these parameters can be affected by fish species, size, age, sex, fish parts, diet, extraction method, and commercial source. The contents of omega-3 long-chain PUFA, namely EPA, DPA, and DHA, are generally higher in oil from marine than freshwater fish species (58–60). In vitro and in vivo studies with dogs showed that these conditionally essential fatty acids (19) play important metabolic roles in cell membrane fluidity, neural development, cognitive status, visual acuity, and inflammatory and immunologic responses (61–63). Similarly, the CP content and amino acid profile of seafood and crustacean hydrolysates might differ depending on the primary source of protein (species and part of the animal), enzyme source, and hydrolysis conditions (17, 64). An earlier study (37) showed that in shrimp hydrolysate obtained from the enzymatic hydrolysis of heads and cephalothoraxes of Litopenaeus vannamei, as the one used in the current study, lysine and leucine were the essential amino acids found in the highest amounts, and glutamic acid plus glutamine were the main non-essential amino acids. For fish hydrolysates, the review from Chalamaiah, Kumar (65) reported aspartic acid and glutamic acid to be found in high amounts in most of the fish protein hydrolysates, and that muscle, head, skin, and visceral hydrolysates contain all the essential and non-essential amino acids, whereas in fish frame protein hydrolysates, aromatic amino acids were not reported.

In the present study, diet palatability was evaluated using the two-bowl test, considered effective when evaluating intentional product enhancements but recognized to be unable to detect eventual flavor fatigue, longer-term nutritional feedback on taste and satiety, and dependent on the reference diet (36). Additionally, the number of animals included in the test is an important consideration when interpreting the results. Although using 20 animals over 2 or 4 days is considered to increase the test resolution (36), the number of animals used in palatability tests greatly differs in the literature, with studies reporting the use of 8 (66), 10 (67), and 16 (68) animals. In the current study, despite the sample size (12 animals) being defined according to an earlier study that evaluated the dietary supplementation with shrimp hydrolysate (37), the number of animals used was lower than the generally recommended (20 animals), so some caution should be taken when interpreting the results obtained. Animals accepted well both the control and the experimental diet, with no differences in the first approach, taste, or intake ratio. It is well known that factors such as food flavor, texture, ingredients, chemical composition, temperature, previous experience of the animal, and size and shape of the kibble can affect palatability (69), and this constitutes an important aspect when formulating diets as palatability might affect both food intake and the provision of nutrients and strongly determine food repurchase potential. In the current study, both diets presented the same kibble size, shape, and texture. Although animal protein hydrolysates are among the most common palatability enhancers used in commercial pet food, scientific studies are scarce (70). Sensorial characteristics of hydrolyzed proteins are associated with the raw protein source and mixture of peptides (71), with short peptides and amino acids such as taurine, glycine, arginine, glutamic acid, and alanine acting as feeding stimulants for companion animals (13). Moreover, bitterness was earlier associated with the molecular weight of peptides, with reports of increased bitterness in soy protein hydrolysates when the molecular weight of the peptides ranged from 4 to 2 kDa and decreased bitterness with peptides <1 kDa (72). Despite these reports, several studies showed the high palatability of protein hydrolysates, such as fish (18) and chicken (73) hydrolysates for dogs.

Dietary fat content is also known to affect palatability, with dogs preferring high-fat diets (74, 75). Being unsaturated fatty acids, particularly EPA and DHA, highly susceptible to deterioration (76), the free radical formation and breakdown of hydroperoxides release organic substances such as short-chain aldehydes, ketones, and alcohols, seen as off-flavors, which might promote the rejection of the food (77, 78). Although not measured in the current study, the stability of the oils used might not have constituted an issue, as no differences were observed on the palatability test, and both diets were well accepted during the feeding trial.

The daily food allowance was determined to ensure energy requirements and adjusted according to the ideal body weight and body condition score, and thus these parameters remained constant during the study. As diets presented similar energy content and digestibility, the daily amount of food required to ensure the requirements did not vary between diets. The differences found in the intake of CP, EE, and some fatty acids reflect the different chemical compositions of the diets. These differences were not reflected in the fecal output and stool quality that were adequate for dogs fed both diets and constitute important parameters for pet owners. An earlier study with dogs found higher fecal volume and lower fecal DM content with the dietary inclusion of 258 g/kg of hydrolyzed chicken liver when compared to a control diet with poultry byproduct, bovine meat, and bone meal, that might be explained by the high osmolarity of that attracts water to the intestinal lumen, but fecal DM output and fecal scores were not affected (79).

Digestibility of DM, nutrients, and energy was similar between diets except for NDF digestibility, which was higher in the control diet but did not affect the intake of digestible NDF, fecal pH, total VFA production, acetate concentration, or ammonia-N concentration. The differences in the dietary EE content and fatty acid profile might contribute to explaining the observed effects on NDF digestibility among diets, namely through the modulation of fecal microbial populations (80, 81). Although no significant effects were observed on CP digestibility, the significantly higher intake of digestible CP observed with the control diet might have contributed to the higher fecal ammonia-N and valerate concentrations (and a tendency for a higher iso-valerate concentration), suggesting that a higher amount of undigested protein has reached the intestine with this diet. Indeed, ammonia-N is generated by the deamination of amino acids, being toxic at high concentrations, and branched-chain fatty acids originate from the fermentation of branched-chain amino acids (82, 83). Although not herein measured, the results obtained might have affected fecal odor that results from the formation of several putrefactive substances from the colonic fermentation of endogenous and undigested amino acids, such as ammonia, aliphatic amines, branched-chain fatty acids, indoles, phenols, and volatile sulfur-containing compounds, that may also exert adverse effects on colonic health (84, 85). While large individual variation inhibited the significance of differences in H₂S emissions, the higher values obtained with the control diet might suggest a higher availability of sulfur-amino acids in the colon of dogs fed the control diet. Using diets not differing on CP digestibility (70.1 to 73.7%), but differing on Zn source (Zn proteinate and Zn sulfate), and without or with the addition of a solid-state fermentation product of Aspergillus niger with residual enzymatic activity, Pereira, Guedes (50) also found a high individual variation on the number of episodes (3.3–10.3), average (11.0–17.0 ppm), and maximum intensity (20.4–33.8 ppm) of adult Beagle dogs. Further studies are needed to confirm the effect observed as H₂S inhibits the utilization of butyrate by colonocytes caused by mitochondrial metalloproteins and is associated with ulcerative colitis (86) as well as the severity of flatus malodor (51). Additionally, despite the scarcity of information available, the study performed by Zhao and McCamish (87) with intact and hydrolyzed soy protein intestinally perfused in increasing amounts in dogs with duodenal and midintestinal fistulas has shown that intestinal transit is slowed in a load-dependent fashion regardless of the source of protein, but it is more pronounced with intact protein, and that a greater amount is absorbed in the proximal half of the small intestine when protein is delivered in a hydrolyzed form, also concluding that as digestion is the rate-limiting step of assimilation in the small intestine, a longer residence time might be needed with a higher protein load. Despite these observations, no effects were observed on fecal output or characteristics.

As the digestibility of EE did not differ between diets, the higher intake of digestible EE reflected the higher EE intake. Differences in EPA and DHA intake were further reflected in the EPA and DHA of RBC and the omega-3 index. Indeed, the RBC omega-3 index has been reported as a biomarker of long-term EPA and DHA intake in humans (88, 89), being inversely correlated with neutrophil–lymphocyte ratio (a systemic inflammation biomarker, 90), cardiovascular disease risk, and acute death (90). Recently, Harris, Jackson (91) suggested its use as a health status biomarker in dogs. Although no range has yet been established for optimal health status in dogs, a 0.3 to 7.0% omega-3 index was determined in routine veterinary dog blood samples, with higher values being associated with improved health (91). An increased omega-3 index was also reported in the RBC of dogs fed a diet supplemented with krill meal or fish meal plus fish oil for 4 weeks (92) and krill oil for 6 weeks (93). The highest values were reported for krill oil (2.70%) and krill meal (2.36%), in the range of those reported for the control diet (2.58%). Moreover, the experimental diet promoted an increase in the omega-3 index by approximately 1.5-fold compared with the control diet, suggesting a potential health-promoting effect of fish BCP’s inclusion in pet food, namely on promoting cardiovascular function and anti-inflammatory activity.

The increased absorption and metabolization of EPA and DHA by dogs fed the experimental diet were hypothesized to promote the coat quality of dogs, as suggested by Combarros and Castilla-Castano (94), but no differences were observed by a näive panel in the present study. The lack of effects may be due to the similarity of the oils used, both rich in omega-3 PUFA, and the use of healthy dogs. Indeed, Logas and Kunkle (95) reported an improvement in the coat quality of dogs with skin diseases and poor coat quality after 6 weeks of supplementation with EPA-rich marine oil (180 mg EPA/120 mg DHA) compared with corn oil supplement; no differences were observed after 3 weeks. More recently, Combarros, Castilla-Castano (94) investigated the effect of a fish oil supplement (110 mg EPA/68 mg DHA) for 3 months on dogs with poor hair coats and observed a reduced clinical score after 2 months of EPA and DHA supplementation when compared to placebo-supplemented dogs, but no differences after 4 weeks. These results support the need for longer studies to fully unveil the potential of fish by-product hydrolysate and oil supplementation on the coat quality of healthy dogs.

The present study shows that a diet with the inclusion of locally produced fish oil and hydrolysate from the agrifood industry in place of imported salmon oil and shrimp hydrolysate was well accepted by dogs, not affecting food intake, digestibility, or fecal characteristics. The lower fecal ammonia-N and valerate concentrations suggest a lower amount of undigested protein reaching the small intestine or a higher protein absorption. Conversely, fish hydrolysate and oil promoted blood EPA, DHA, and omega-3 index, suggesting a potential health-promoting effect. Overall, the results support the potential of fish hydrolysate and oil from the agrifood industry to be used in dog food, contributing to increasing the sustainability of the pet food sector. Further research is needed to evaluate the functional role of these fish by-products.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by Animal Ethics Committee of School of Medicine and Biomedical Sciences, University of Porto, and licensed by the Portuguese General Directorate of Food and Veterinary Medicine (Permit No 0421/000/000/2021). The study was conducted in accordance with the local legislation and institutional requirements.

AC: Supervision, Writing – original draft, Resources, Project administration, Investigation, Funding acquisition, Formal analysis, Conceptualization. MM: Writing – review & editing, Investigation, Formal analysis, Conceptualization. APA: Writing – review & editing, Investigation. TA: Writing – review & editing, Resources. AR: Writing – review & editing, Resources. AA: Writing – review & editing, Resources. RM: Writing – review & editing, Resources. AF: Writing – review & editing, Resources, Investigation, Funding acquisition, Formal analysis, Conceptualization.