Collagen hydrolysates (CH) are produced through extraction and hydrolysis of collagen-rich raw materials such as bone, hide and skin from various animal species. The types of collagen present varies among tissues and the specific collagen sequence is determined by the animal source. Collagen is an evolutionary conserved protein across species, yet as a natural product, differences can be expected between CH from different raw material origins. The manufacturing process produces a complex mixture of peptides of varying lengths and identities. The degree of hydrolysis determines the average peptide length, quantifiable as the mean molecular weight. It is important to note that the molecular weight does not specify peptides’ identity but is determined by the sequence of amino acids (AAs). The oligopeptides generated during the CH manufacturing process are further cleaved during the gastrointestinal passage (1–3). Previously, it was assumed that CH are fully broken down into single AAs during absorption into the blood stream, serving primarily as building blocks for protein synthesis. However, recent studies confirm that both free AAs and bioactive di- and tripeptides appear in the human bloodstream after CH ingestion (4, 5). This is supported by the presence of a specific transport system, PepT1, which contributes to trans-cellular movement of di- and tripeptides across the enterocyte (6). Thus, CH are an important source of bioactive peptides, which, once cleaved from the parent collagen protein and absorbed into the bloodstream, can exert beneficial effects on body functions beyond their nutritional value (1, 7–10). CH-derived bioactive peptides support the strength, structure and elasticity of key extracellular matrix components in cartilage, joints, ligaments and tendons (1, 11–13), bone (14–16), skeletal muscle (17, 18), and skin (19–21). They achieve this by modulating cell functions linked to matrix protein synthesis (22, 23), growth, differentiation and morphogenesis. Additionally, anti-thrombotic (24), anti-hypertensive (23, 25, 26), and neuroprotective (27) functionalities of CH-derived peptides have been described, highlighting their broad potential as functional ingredients for food and dietary supplements.

Collagen is characterized by a distinct AAs composition, unusually high in glycine (Gly) and proline (Pro), and is one of the few animal proteins that contains hydroxyproline (Hyp) (28). The characteristic collagen motif consists of a repeating (Gly-Xaa-Yaa) n sequence, where Xaa and Yaa can be any AAs but are often Pro (28%) and its post-translational modification Hyp (38%) making Gly-Pro-Hyp the most common triplet (10.5%) in the collagen sequence (29). Due to its high abundance, Hyp can be used as a marker to trace collagen in food or blood samples (30). The unique ring structures of Pro and Hyp distinguish them from other AAs in terms of rigidity, chemical stability and biochemical reactions (31), allowing them to form bonds with adjacent AAs that resist hydrolysis by digestive enzymes. This is confirmed by the presence of Hyp-containing di- and tripeptides in the blood after collagen ingestion (4, 32–34). Kleinnijenhuis et al. described changes in size distribution and peptide patterns of CH during digestion and absorption, confirming that Hyp-Gly and especially Pro-Hyp significantly contribute to the total Hyp increase in blood after CH ingestion (5).

Despite the growing knowledge on collagen-derived signature peptides and their individual bioactivities, the identity of all peptides in the complex mix of CH products is not fully described, and the detailed changes these peptides undergo during digestion and absorption are generally unknown. This renders the attempts to elucidate the molecular basis of a CH’s biological activity extremely difficult. Consequently, whether the varying bioactivity of CH products is related to differences in peptide identity and bioavailability, differences in the raw material source, molecular weight (i.e., average peptide length before ingestion), or other elements of the product’s manufacturing process is not well understood. Recent studies have compared the bioavailability of Hyp-containing peptides from various collagen sources, focusing on fish, pork and chicken-derived CH (4, 35, 36). However, bovine-hide derived CH is the predominantly used product in much of the world, particularly the Western hemisphere. To our knowledge, there is no comprehensive study comparing the bioavailability of the most common skin- and hide-derived CH from fish, pork and bovine sources.

To investigate the impact of the raw material source and the molecular weight (2,000 and 5,000 Da, bovine) on the bioavailability, we compared fish skin, porcine skin and bovine hide-derived CH in a single study. This involved targeted analysis of highly abundant and collagen-typical AAs (Hyp) and peptides (Pro-Hyp, Hyp-Gly, Gly-Pro-Hyp, Pro-Gly, Gly-Pro) in human blood after CH ingestion. Additionally, systemically available free and total Hyp were measured to estimate the ratio of uptake in form of free AAs or peptides.

Bioavailability refers to the fraction of a nutrient or bioactive compound that can be absorbed into the bloodstream from food upon consumption and is available for use by the body. Factors such as food processing, food composition, genetic differences and individual variations in gastrointestinal digestion and absorption can affect nutrient bioavailability (38, 39). For CH products ongoing discussions focus on how elements of the manufacturing process, raw material origin and degree of hydrolysis (which determines the mean molecular weight) influence differences in bioavailability between CH products and whether these differences translate into varying bioactivity.

Due to its scarce presence in animal protein, Hyp is considered as the signature AAs for collagen and is used as a marker to track the uptake of CH-derived metabolites into the bloodstream (30). In the present study, investigated CH from all sources (bovine, porcine and fish) and mean molecular weights (2,000 Da vs. 5,000 Da) were absorbed, as indicated by a substantial increase in free Hyp, reaching maximum plasma concentrations 6–10 times higher than baseline within 100–130 min after ingestion. Correspondingly, the uptake kinetic patterns measured for total Hyp were very similar for all CH, however with a ΔC_max approximately twice as high compared to free Hyp. This finding clearly indicates that a considerable fraction of Hyp remains in peptide-bound form in the circulation—between 36 and 47% depending on the CH investigated –, confirming that CH are digested and absorbed both as free AAs and peptide-bound forms.

Notably, there were minimal variations in the total Hyp content of the CH products before ingestion. After oral intake, statistically significant differences between CH were observed for some, but not all, pharmacokinetic endpoints. Descriptively, porcine CH showed a higher iAUC for total Hyp compared to fish CH. The iAUC of total Hyp, but not free Hyp, or any of the investigated peptides, was significantly lower for the HMW compared to the LMW bovine CH. Remarkably, inter-individual differences in uptake kinetics seemed to be pronounced as indicated by the spread of the individual data points for the iAUC of some parameters (e.g., total Hyp, Figure 3, or Hyp-Gly, Figure 10), which may partially contribute to the observed differences.

The cross-over design of this study controlled for confounding factors such as genetic constitution, gut motility, digestive enzyme activity, gut mucosa composition, and gut microbiome composition and functionality to the maximum extent possible. However, the pilot scale sample size limits the generalization of the observed differences. Further studies with a larger sample size are warranted to confirm the findings.

The observed levels of free and total Hyp align with information currently reported in the scientific literature (36, 40). In a study with five volunteers, Ohara et al. compared the uptake of three sources of CH from fish scale, fish skin, or porcine skin with the same mean molecular weight and reported a contribution of Hyp-containing peptides to total Hyp uptake of about 30% (36). In the current study, this fraction was estimated to be even higher between 36 and 47% corresponding to a substantial part of all absorbed Hyp. In situ intestinal perfusion analysis suggested that CH was absorbed predominantly as peptides, showing that the peptides could pass through the liver and reach the systemic circulation (41). Similar to findings in this study, Ohara et al. reported a slightly higher AUC for porcine skin CH compared to fish skin CH (1.1-fold total Hyp), which was significant for free Hyp (1.5-fold). Additionally, fish scale CH resulted in a different uptake profile with significantly higher AUC for free and total Hyp compared to fish skin CH (36). Despite the same animal source, it is known that fish scale has a different collagen composition than fish skin potentially explaining the differences in absorption (42).

Targeted analysis of the selected peptides in plasma revealed that Pro-Hyp was the most abundant peptide detected in this study after intake of the different CH sources. Intake of 10 g yielded high plasma concentrations (e.g., porcine CH ΔC_max 3.8 μg/mL) despite Pro-Hyp being the peptide with the lowest average abundance in the CH products prior to ingestion. These results clearly indicate that this peptide is formed and released during digestion and absorption in the gastrointestinal tract by cleavage from larger oligopeptides. In contrast, the peptide Gly-Pro had the highest abundance in the CH products prior to ingestion but resulted in a ΔC_max 100 times lower than that of Pro-Hyp.

Thus, the initial peptide content of the CH products is not a direct indicator for the expected blood concentrations of a given peptide. Literature supports this, showing substantial increases of peptides in plasma after single-dose ingestion as reported by Kleinnijenhuis et al. (5). In addition, CH products with a different mean molecular weight, bovine HMW (~5,000 Da) and LMW (~2,000 Da), led to highly comparable peptide and AAs concentrations in blood. These observations indicate that di- and tripeptides are further released during digestion and absorption.

This is also supported by Iwasaki et al., who showed that a CH (89% components <3,000 Da) significantly enhanced plasma concentrations of Pro-Hyp and Hyp-Gly compared to gelatin, which has a high molecular weight (100,000 Da). The absorption of free Hyp following gelatin ingestion (94.4 ± 16.4 nmol/mL) was significantly lower compared to CH (169.1 ± 32.5 nmol/mL) (35), indicating that the dipeptides were cleaved from the parent collagen protein during gastrointestinal passage, but that a very high initial molecular weight of gelatin and the complex physicochemical properties of gelling collagen reduces the peptide’s bioavailability.

Current literature shows that Pro-Hyp appears at very high concentrations after oral intake of CH (porcine skin, chicken feet and cartilage origin) reaching maximum concentrations after 60–120 min, consistent with our findings (4). Others confirmed Pro-Hyp as the most abundant peptide following ingestion of CH from porcine skin, fish skin and scale (36, 43). Shigemura et al. reported extensive variation between volunteers in the ratio of the dipeptides Pro-Hyp and Hyp-Gly, but only fish scale derived CH was investigated (43). In the present study, the ratio of Pro-Hyp to Hyp-Gly averaged 12.6 for the bovine CH LMW, 14.6 for the bovine CH HMW but only 6.6 for the fish CH and 10.5 for the porcine CH.

For the other investigated peptides, Hyp-Gly had much lower plasma concentrations than Pro-Hyp for all products (ΔC_max range: 0.29–0.76 μg/mL). For this dipeptide, there was a clear difference between products with fish CH presenting a significantly higher C_max compared to bovine CH LMW (p = 0.0112), and porcine CH (p = 0.0412). The other investigated peptides Gly-Pro-Hyp, Pro-Gly, and Gly-Pro were less abundant in plasma.

Despite differences in individual peptides between some of the CH investigated in the present study, the results indicate that the absorption of fish, porcine and bovine CH is highly comparable. Some peptides have been linked to specific bioactivities, but due to the complexity of the product (oligopeptide mix) and metabolites in the blood it is not possible to establish a simple cause-effect relationship to explain the biological benefits of CH. Relating these findings to biological activity was beyond the scope of this investigation, but general biological effects have been shown for diverse CH products, suggesting they can be used interchangeably for certain health applications. Benefits for skin health have been reported for CH of fish, porcine and bovine sources (19, 21, 44). The efficacy of CH from fish skin, porcine skin, or bovine hide was confirmed by Wauquier et al. regarding bone cell activity using CH-metabolite containing human serum in vitro (15). Furthermore, in a clinical study with osteoarthritis patients, both CH from porcine and bovine origin confirmed the ability to improve osteoarthritis symptoms with no reported differences between sources (12).

The present study compared the uptake profiles of different CH products in response to a single dose application. The complexity of CH-derived metabolite profiles in the blood might increase during long-term CH supplementation. According to Shigemura et al., daily ingestion of CH for a longer period, i.e., 4 weeks, changed the profile of Hyp-containing peptides in human blood (45). The compositional rate of Hyp-Gly tended to increase, while Pro-Hyp tended to decrease after daily ingestion. It was speculated that long-term CH ingestion might change exo- or endo-type protease activity in the digestive tract, potentially promoting beneficial effects. Further investigation is warranted to determine how the bioavailability of total and peptide-bound Hyp predicts the uptake kinetics and bioavailability of specific bioactive peptides, and how this is linked to CH’s bioactivity.

In the present study, selected CH-derived peptides described to be among the most abundant, were investigated for their bioavailability. Larger scale studies are needed to improve the understanding of how inter-individual physiological variability and differences in CH product characteristics may impact the bioavailability of bioactive components relevant to CH bioactivity. Further investigations are also warranted to extend the knowledge of the mechanisms underlying CH’s health benefits.

Substantial knowledge gaps exist on how different CH products compare regarding their digestion, bioavailability, and bioactivity, with bovine-derived CH being underrepresented in current research in particular.

In the present study the uptake of free Hyp was not affected by the source or molecular weight of the CH. However, the bioavailability of the peptides Hyp-Gly, Pro-Hyp, Gly-Pro-Hyp, and Pro-Gly from CH of different sources showed significant differences, indicating an impact of the animal species and potential inter-individual variation. On the other hand, the molecular weight of the CH did not affect the bioavailability of the selected peptides.

It is noteworthy that a growing body of literature reports health benefits for different CH products, regardless of their source or molecular weight. This suggests that certain products may be used interchangeably. However, further research is needed to understand the complex matrix connecting (1) the CH product origin (animal source, molecular weight, enzymatic hydrolysis process), (2) the metabolite patterns that appear in the circulation after CH ingestion, to (3) the bioactivity in the target tissues and the potential overall health benefit. For this purpose, defining the threshold concentrations of individual circulating metabolites and their potential interaction needed to trigger distinct physiological reactions in different target tissues are a crucial step. This would allow a better comparison of different CH products and help to predict their efficacy. In addition, investigating the impact of the food matrix on the CH efficacy is of importance, considering that CH are ingredients in functional foods and supplements.

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

The studies involving humans were approved by Landesärztekammer Baden-Württemberg (F-2019-075). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

NV: Writing – original draft, Conceptualization, Validation, Writing – review & editing. CS: Investigation, Software, Validation, Writing – review & editing, Conceptualization, Formal analysis, Project administration, Visualization, Writing – original draft. YM: Investigation, Software, Validation, Writing – review & editing. BS: Writing – review & editing, Data curation. SV: Writing – original draft. FH: Methodology, Writing – review & editing. AK: Methodology, Writing – review & editing. CIFS: Writing – original draft. JP: Writing – original draft, Funding acquisition, Resources, Supervision.