There were separations of the treatment groups from the PCA plots from plasma and fecal samples as expected. This corresponded with previous studies with the same diets that showed differences in standardized amino acid digestibility and selected fecal fermentative metabolites7,8. The random forest plots showed that the most determining compounds that contributed to treatment clustering were mostly from protein and lipid metabolism as well as some xenobiotics. This was not surprising as the main difference in the protein ingredients was the amino acid and lipid compositions. A previous study also demonstrated the modulatory effect of CLH on gut microbiota in dogs8; therefore, differences in xenobiotics from the microbial activity in the present study also were expected.
Some changes in protein metabolites in the plasma were in accordance with the diet composition. For example, there were higher glutamate and its metabolites in the CLHd plasma as well as in the diet composition. This could indicate that the amino acids were absorbed into the body for utilization. Glutamate is the fuel source of various cells, an excitatory neurotransmitter in the brain, and a precursor of antioxidant glutathione; low glutamate level was also linked to stress and higher susceptibility to infections9. Therefore, the high plasma glutamate concentration could provide more energy and antioxidation capacity for the cells. Similarly, phenol sulfate, a downstream metabolite from tyrosine that is linked to renal disease and could lead to oxidative stress, was lower in concentration in CLHd diet, plasma, and feces than in CONd10,11. There were also higher pipecolate concentrations in both the diet and plasma from CLHd than CONd. It was reported by previous research that pipecolate exhibited antioxidative effects and, therefore, could be another level of protection against oxidative stress12. On the other hand, the CLHd diet was high in histidine, phenylalanine, and tyrosine but the plasma from CLHd was low in histidine, phenylalanine, and tyrosine when compared with CONd. This could indicate that the metabolism of these amino acids was higher in the CLHd group than in the CONd. The higher plasma imidazole propionate, a downstream metabolite from histidine, in the CLHd group could further support this hypothesis. Histidine is an indispensable amino acid that serves as an active site of enzymes as well as a precursor of carnosine and histamine13. The metabolite imidazole propionate could interfere with insulin signaling through the mechanistic target of the rapamycin complex and patients with metabolic diseases were reported to have higher fecal concentrations than healthy individuals14. However, a normal or healthy range of plasma or fecal imidazole propionate in dogs has yet to be established. Phenylalanine is an indispensable amino acid and can be used to produce tyrosine in the body; both amino acids are categorized as aromatic. They serve as building blocks of proteins and also precursors of catecholamines, such as dopamine, norepinephrine, and epinephrine; dietary tyrosine, having an important role in catecholamine synthesis, was reported to help with reducing blood pressure in hypertensive rats and working memory in humans under stress15. Since the standardized amino acid digestibility of CLHd was comparable to CONd and higher than 80% for phenylalanine and tyrosine7, it could be suggested that phenylalanine and tyrosine were absorbed into the body and used for the production of needed catecholamines, resulting in a greater plasma concentration of dopamine 3-O-sulfate in the CLHd group. Some microbial fermentative product derivatives were also different between CLHd and CONd. Indole propionate, produced from the microbial metabolism of tryptophan, was low in the CLHd diet but higher in the CLHd plasma and feces. The higher tryptophan concentration from the CLHd diet could lead to this difference with the higher substrate supply to the microbes. Since indole propionate showed antioxidation and anti-inflammatory properties in previous studies, the higher concentration in the plasma and feces could indicate the hydrolyzed protein supported its production and, thus, providing a higher capability against oxidative stress and inflammation in healthy animals16.
Differences in the fecal protein metabolite concentrations in the current study were expected as a previous study showed higher fecal butyrate with lower isovalerate and total phenol/indole in dogs fed the CLHd than CONd, which also corresponded with the microbiota differences8. The lower fecal 5-aminovalerate, pipecolate, and 2-aminoadipate from lysine degradation in the CLHd group when compared with the CONd could be due to the less substrate entering the hindgut for fermentation17,18. This would support the finding that the plasma 5-aminovalerate level was lower in the CLHd group as well. The higher fecal spermidine, spermine, and N-aceylspermine from ornithine in the CLHd group could mean more protection from oxidation as they previously demonstrated antioxidative properties19,20. Previous studies have shown spermidine extended lifespan and health span in different species; therefore, the finding in the present study could indicate that hydrolyzed protein would support health in animals under oxidative stress21,22,23. Higher fecal dimethylglycine, betaine, cystathionine, and cysteine from methionine metabolism could indicate a sufficient supply of dietary methionine from the CLHd since excessive methionine not incorporated in protein converts to S-adenosylmethionine in the first step of catabolism and then to other products24. In addition, cysteine is known to be the precursor of antioxidant glutathione25. Therefore, the higher concentrations of downstream metabolites from methionine could be an indication of higher antioxidation capability from consuming the CLHd. In concert, lower fecal succinate concentration was found in the CLHd when compared with CONd. It has been reported that reducing the accumulation of succinate by altering the gut microbiota could reduce inflammation in animals26,27,28.
Even though there were higher peptides and indispensable amino acids in the CLHd and CHd diets than in the CONd, the fecal concentrations of peptides and indispensable amino acids were lower. This indicated the higher digestibility of amino acids in the hydrolyzed proteins. The finding corresponded with the higher standardized amino acid digestibility of the test hydrolyzed proteins than CM from a previous study using a precision-fed rooster assay7. In addition, fecal metabolites from amino acid fermentation were also low in the CHd group. This could further support the hypothesis that fewer amino acids were available to enter the large intestine for microbial degradation from hydrolyzed proteins. The gut microbiota could also favor incorporating the available amino acids into microbial proteins over fermenting the substrates for energy and, thus, resulting in fewer degradation products. In addition, it was previously reported that amino acid composition could affect the rate of protein degradation29; therefore, the different compositions among the protein ingredients could also be a cause of this observation. Urea, one of the markers for protein degradation, could be found higher in concentration with high-protein diets and is toxic at high levels30,31. Therefore, a lower urea level from the CLHd could be beneficial to the dogs, especially those consuming higher levels of dietary protein.
Some differences in the plasma and fecal lipid metabolites between CLHd and CONd were in accordance with chemical composition of diets. Arachidonate, cholesterol, osbond acid, and sphingolipids were higher in concentration in both the CLHd diet and plasma samples than CONd. Moreover, cholesterol, osbond acid, and DHA were also higher in CLHd fecal samples than CONd. Mead acid and glycerol-3-phosphate were of greater concentrations in CLHd than CONd in diet compositions and fecal samples. On the other hand, erucate was low in both CLHd diet and plasma. The finding of differences in concentrations of arachidonate, cholesterol, osbond acid, mead acid, sphingolipids, glycerol-3-phosphate, and erucate in plasma and/or fecal samples could be the direct result of the amount of nutrients that were ingested.
Polyunsaturated fatty acids have been studied for their structural function on cell membranes and their effects on inflammation and diseases32,33,34. Both arachidonate and osbond acid are omega-6 fatty acids which are important fatty acids that support skin and coat health; however, the ratio between omega-6 and omega-3 fatty acid intake should be monitored as excessive omega-6 fatty acids could cause a rise in inflammatory responses33,35. Mead acid, an omega-9 polyunsaturated fatty acid, is usually a minor fatty acid in healthy individuals but could increase in concentration when essential fatty acids are deficient36. Nonetheless, excessive or deficient essential fatty acids should not be a problem in the current study since all diets were formulated to be complete and balanced. It was also reported that dietary supplementation of mead acid could decrease inflammation since it is an arachidonic acid analog37. Limited research has been done on the physiological effect of erucate, an omega-9 monounsaturated fatty acid, but some studies showed its neuroprotective potential as it was known as a component of Lorenzo’s Oil to treat adrenoleukodystrophy38. Glycerol-3-phosphate is at the crossroads of different macronutrient metabolism pathways and serves as the backbone of most glycerolipids39,40. The higher glycerol-3-phosphate concentration could indicate more abundant substrates for lipogenesis in dogs fed CLHd than CONd. Sphingolipids have been studied for their roles in the immune, cardiovascular, and nervous systems; though the mechanisms of sphingolipid regulation have yet to be deciphered, sphingolipids were reported to be crucial for normal brain function and inflammatory responses41.
Interestingly, the concentration of lanosterol, the precursor of cholesterol, was lower in CLHd feces. It could indicate that the precursor lanosterol was efficiently used to produce cholesterol in the dogs from the CLHd group and, thus, resulted in higher cholesterol and lower lanosterol in the feces. In agreement with the lanosterol and steroid concentrations, in the mevalonate pathway or hydroxymethylglutaryl coenzyme A (HMG-CoA) pathway, there was higher fecal mevalonate in the CLHd group in contrast to CONd. Mevalonate is the product of HMG-CoA reductase and the higher diet concentration of 3-hydroxy-3-methylglutarate (HMG), the substrate of HMG-CoA reductase, could result in a higher level of fecal mevalonate. Since the HMG-CoA pathway contributes to the synthesis of cholesterol, more substrates in the pathway could also lead to a higher cholesterol concentration in the CLHd group. Contradictory to the finding in the present study, several studies on hydrolyzed protein showed decreased blood cholesterol42,43,44,45,46. It was hypothesized by previous studies that the mechanism behind the hypocholesterolemic effect was the decrease in cholesterol solubility in micelles and cholesterol absorption. However, some studies also found no change or an increase in plasma cholesterol after consuming protein hydrolysates47,48. It was suggested by researchers that the effect on cholesterol metabolism could be related to the composition of amino acids in the protein; a low ratio of lysine-arginine would exhibit hypocholesterolemic effects while high ratios would have the opposite result49,50. In the current study, both protein hydrolysate ingredients CLH and CH had higher ratios of lysine-arginine than CM. Therefore, the amino acid composition could be another possible explanation for the higher plasma and fecal cholesterol in CLHd and CHd groups, besides the higher cholesterol level in the hydrolyzed protein diets.
Some compounds were comparable in the diets but different in plasma concentrations. Myristate and palmitate concentrations were similar between the CLHd and CONd diets but lower in the plasma of CLHd. Previous research has shown that dietary palmitate caused an increase in low-density lipoprotein receptor activity51. Therefore, the lower plasma palmitate could also be related to the high plasma cholesterol in CLHd. Phospholipid GPC was slightly lower in the CLHd diet than the CONd diet but higher in CLHd plasma and feces. Since palmitate is used for fatty acid synthesis, the lower plasma palmitate could be a result of a higher phospholipid synthesis, and, thus, the products were excreted in the feces at a higher level52. Generally, phospholipids are crucial for cell membrane integrity and signaling53. Glycerophosphocholine, in particular, could stimulate growth hormone secretion and fat oxidation54. This could indicate a more efficient lipid metabolism in the dogs fed CLHd. Furthermore, EPA and DPA concentrations were greater in the CLHd diet but lower in the plasma than in CONd. This was interesting as previous research showed a positive correlation between dietary polyunsaturated fatty acids and their plasma concentrations55,56,57. However, this may not be physiologically significant as the DHA concentrations were higher in both plasma and fecal samples in the CLHd group than in the control.
The most notable difference in lipid metabolites between CHd and CONd groups was the bile acid metabolism. The main function of bile acids is to help with lipid digestion and absorption in the intestine58. Animals can produce primary bile acids in the liver and microbes can use the primary bile acids to produce secondary bile acids. Dogs consuming CHd had lower plasma primary and secondary bile acids than the CONd. This was interesting because plasma cholesterol, the precursor of bile acids, was higher in CHd than in CONd. Therefore, it was not the lack of substrates that resulted in lower bile acid levels. Even though high or low circulating bile acid concentrations that are out of the normal range are indications for diseases59,60, the dogs in the present study were all healthy and, thus, diseases should not be a concern. The lower plasma bile acid concentrations could mean a more efficient hepatic uptake in the CHd group, resulting in fewer bile acids entering systemic circulation. In the fecal samples, CHd showed higher concentrations of conjugated bile acids and lower unconjugated bile acids than CONd. In the body, most bile acids are conjugated to increase water solubility and become impermeable to cell membranes, allowing their high concentrations in the bile and small intestine58. However, when bile acids enter the large intestine, they undergo bacterial deconjugation and the unconjugated bile acids will be readily reabsorbed into the enterocytes and then the liver61. The lower unconjugated bile acids in CHd fecal samples could be an indication of more efficient reabsorption of bile acids in the hindgut. On the other hand, the higher conjugated bile acids could mean a less efficient deconjugation from the bacteria and a potential for more amino acid loss. However, since the plasma levels of taurine were not different between CHd and CONd, there should not be a concern for excessive taurine loss. The difference in conjugated and unconjugated bile acids excretion from the current study could be a result of modified gut microbiota from consuming the hydrolyzed protein.
The plasma and fecal metabolites jointly showed the anti-inflammatory and antioxidative potentials of the test protein hydrolysate in healthy adult dogs. Previous studies also showed protein hydrolysates from chicken origin exhibited anti-inflammatory properties44,62,63 and antioxidation effects64,65. Several canine studies also used metabolomics to observe anti-inflammatory and antioxidation functions from dietary treatments through indicators in blood and feces, such as methionine, glycine, polyamines, succinate, chenodeoxycholic acid, adenosine, linolenic acid26,66,67,68,69. Some studies included healthy dogs to examine if administered supplements, such as black ginseng extracts, hemp oil, and gallnut tannic acid, helped with inflammation and oxidative stress levels26,67,69. A study by Lyu et al. (2022) aimed to determine the metabolomic profiles of healthy dogs fed a high-starch or high-fat diet. Ambrosini et al. (2020) included dogs with inflammatory bowel disease and observed some serum biomarkers for the disease improved after administering a hydrolyzed diet66,68. Their findings in the correlation of metabolites and inflammation/oxidation showed corresponding results with the present study.