Introduction

Cheese is a traditional fermented food with high nutritional value and functional potential. Proteolysis is one of the most important processes of cheese, where the enzymatic hydrolysis of milk proteins during fermentation and ripening produces bioactive peptides. During ripening, proteolysis leads to the release of bioactive peptides, which may exhibit antihypertensive, antioxidant, immunomodulatory, and antimicrobial activities (Chourasia et al., 2021; Daliri et al., 2017; Öztürk and Akın, 2021).

The proteolytic system of lactic acid bacteria (LAB) used as starter cultures play a key role in the formation of bioactive peptides primarily through the production of proteolytic enzymes, proteinases, and, notably, peptidases, which actively hydrolyze milk proteins (McSweeney, 2017). Mesophilic species, such as Lactococcus lactis, are active in the initial phase of proteolysis. In contrast, thermophilic strains, such as Streptococcus thermophilus, provide deeper cleavage of casein fractions, particularly β- and αS1-caseins. This promotes the formation of long-chain peptides with potential physiological activity. In particular, strains of Lactobacillus plantarum and Lactobacillus rhamnosus are reported to enhance the formation of peptides, such as Val-Pro-Pro (VPP) and Ile-Pro-Pro-Pro (IPP), which are well known for their angiotensin-converting enzyme (ACE) inhibitory and cardiovascular protective effects (Fan, et al., 2019; Helal et al., 2023; Baptista et al., 2020).

The novelty of this study dwells in the investigation of the effect of mesophilic and thermophilic cultures on the capability of bioactive peptides from camel milk cheese. Camel milk has traditionally been consumed in arid regions because of its nutritional and therapeutic properties. Owing to its unique casein profile and rich content of bioactive molecules, camel milk is a promising substrate for the development of functional dairy products (Konuspayeva et al., 2023). Despite its potential, the application of camel milk in making cheese historically has been limited by its low coagulation capacity and unique protein composition (Faye et al., 2020). However, recent advances in enzymatic coagulation systems and starter culture technology have facilitated the successful production of soft, semi-hard, and hard cheeses from camel milk (Faye et al., 2020; Mohamed and El Zubeir, 2021). Previous studies have investigated the fermentation of camel milk in fermented milk drinks using Lactobacillus rhamnosus, and reported ACE inhibitory and antioxidant activities of the released peptide fractions (Moslehishad et al., 2013). Yet, no specific bioactive peptides have been identified in camel milk cheese, which has served to fill this scientific gap.

The present study aimed to assess the effect of different starter cultures on the formation of biologically active peptides in camel milk cheese throughout the entire ripening period. This was achieved using a database of milk bioactive peptides and chemometrics, particularly principal component analysis (PCA), which not only simplifies data analysis but also enhances the presentation of results by making complex datasets more accessible and highlighting key insights (Tarapoulouzi et al., 2020).

Materials and Methods

Cheese-making

Raw milk of camel (Camelus dromedarius) was collected from a private herd in the Turkestan region (Kazakhstan). Two trials for producing cheese were conducted. In the first trial, cheese was prepared using 1.5% of commercial liquid mesophilic culture (Lactococcus lactis, subsp. Lactis; Lactococcus cremoris; Lactococcus lactis, subsp. lactis biovar diacetylactis [Biochem srl, Italy]), and in the second trial, 1.5% of commercial liquid thermophilic culture (Streptococcus thermophiulus; Lactobacillus debrueckii, subsp. lactis; and Lactobacillus helveticus [Biochem srl]) was used. Cheese samples were produced with the addition of rennet Chy-Max M 1000 (Ch. Hansen, Denmark). In all, 6 L of camel milk were divided into two portions of 3 L each. Protein, fat, total solids, lactose, pH, and non-fat milk solids of camel milk were analysed by Milkoscan FT120 (Foss, Denmark). Raw camel milk was clarified and 40% CaCl₂ (w/v) was added to it in a dose 0.1% to 1 L of milk to improve its rennet ability. The trial camel milk was inoculated with mesophilic culture and acidified to a pH of 5.8–6, and rennet Chy-Max M 1000 was added at 50 international milk clotting units per litre (IMCU/L) of milk. Fermentation of camel milk was realised at 32–34°C. After coagulum formation, it was cut into 4–6-mm cubes. The whey was separated and the cheese was self-pressed and moulded. The cheese samples were treated with hot whey (85–95°C) for 5 min and then cooled on a cheese table (for up to 30 min) and salted in abrine (10% NaCl, w/v) at 10°C. The cheese samples were ripened in moulds filled with brine at 10–12°C for 30 days. The samples were studied on the 1st, 10th, 20th, and 30th day of ripening. The same approach was used for producing the second trial, except for adding thermophilic starter culture, and fermentation was realised at 38–40°C.

Preparation of water-soluble extracts

The method described by Théolier et al. (2014), with some modifications, was used to prepare water-soluble extracts. In this method, 0.2 g of cheese was thoroughly rinsed with Milli-Q water (Millipore Corp., Bedford, MA, USA). The sample was homogenised by rigorous vortexing in 5× volume of Milli-Q water for 5 min at room temperature. The homogenate was incubated for 1 h at 40°C, with stirring at 300 rpm. The homogenate was centrifuged at 3,800 revolutions per minute (rpm) for 30 min and the pellet was discarded. The supernatant was stored at -20°C for mass spectrometry analysis.

Mass spectrometry analysis

The tandem mass spectrometry (MS/MS) parameters were based on Makhammajanov et al. (2024). The samples were reduced, alkylated, and then digested by trypsin (20 ng/μL; Sigma-Aldrich, Italy) at 37°C overnight. The obtained peptides were purified and concentrated using ZipTip-C18 (Millipore, Cork, Ireland). The eluted peptides were dried using a rotation-type evaporator (Eppendorf, Hamburg, Germany) and re-suspended in 10 μL of 0.1% (w/v) trifluoroacetic acid. Peptides were analysed using online nanoflow reversed-phase C18 liquid chromatography–tandem mass spectrometry (LC-MS/MS). A trapping column (Acclaim PepMap 100 C18 pre-column) and a Dionex HPLC pump were used for chromatography. Peptides were separated on an Acclaim Pep-Map RSLC column (Thermo, MA, USA) using a 75-min multistep acetonitrile gradient at a flow rate of 0.3 mL/min. An unmodified captive spray ion source (capillary 1,300 V, dry gas 3.0 L/min, dry temperature 150°C) was used to interface the LC system to the Impact II ESI-QUAD-TOF mass spectrometer (Bruker Daltonics, Germany). Full-scan MS spectra were acquired at a spectral rate of 2.0 Hz, followed by the acquisition of one MS/MS spectrum. The MS/MS peak lists in Mascot generic format were searched on a local server using the Mascot 2.6.1 software (Matrix Science, UK) against the Swiss-Prot protein database (release 2022_03 and NCBI). The studied parameters were oxidation of methionine, and carbamidomethylation of cysteine residues as variable modifications. Peptide and protein data were extracted using high peptide confidence. A mass error window of 100 ppm and 0.05 dalton (Da) was allowed for MS and MS/MS, respectively.

Chemometric analysis

Owing to potentially large diversity and unidentified origin of bioactive peptides and their precursors, chemometrics was first proposed as a method for an objective approach to assess peptide formation in camel milk cheese using different starter cultures during ripening, although this is still insufficiently studied. Chemometric analysis was carried out by the SIMCA software (version 17; Umetrics, Sweden). The PCA was applied. Ellipse in the plots defines Hotelling’s T2 confidence region, which is a multivariate generalisation of Student’s t-test and provides a 95% confidence interval (95% CI) for observations. The number of studied components chosen is given with the symbol A. R2X and R2Y represent the cumulative modelled variation, explaining the quality of the model, and Q2 is an estimate of the model’s predictive ability. R2X, R2Y, and Q2 values (not less than 0.5) show robust models with predictive reliability (Tarapoulouzi et al., 2020).

Identification of bioactive peptides

Peptide sequences identified by LC-MS/MS were screened for known biological activities using the milk bioactive peptide database (MBPDB; Nielsen et al., 2017). Each peptide was manually entered into the database’s sequence search tool (http://mbpdb.nws.oregonstate.edu) to determine its sequence identity to previously characterised bioactive peptides. The search results were filtered based on exact matches, and peptides with full homology to document functional sequences were considered. For each matched peptide, the reported biological activity (e.g., ACE inhibitory, antioxidant, antimicrobial, and immunomodulatory) was recorded according to the annotation provided in the database.

Additionally, Peptide Ranker (http://distilldeep.ucd.ie/PeptideRanker/ [accessed: 14 March 2024]) server for the prediction of bioactive peptides was used. For every peptide, Peptide Ranker predicts the bioactive probability (between 0 and 1) of that peptide. The closer the predicted probability of peptide to 1, the more confident is Peptide Ranker for the peptide to be bioactive (Mooney et al., 2012).

Data processing

To examine differences in chemical composition, ANOVA was applied. If variations were detected, Tukey test was used to compare mean values, with a significance level set at 5%. For each sample, three replicates were measured.

Results

The physicochemical characteristics of raw camel milk were as follows: protein content, 3.18 ± 0.20%; fat content, 4.03 ± 0.20%; and lactose content, 4.66 ± 0.10%. Total solids were measured at 12.73 ± 0.20%, while non-fat solids accounted for 8.44 ± 0.10%. The pH value of the milk samples was 6.42 ± 0.05. The relatively low standard deviations (SD) across samples indicated a homogeneous composition of milk, which was favourable for reproducible cheese production. These values were consistent with previously reported compositional ranges for raw camel milk and confirmed its technological suitability for further fermentation and curdling (Alhaj et al., 2023).

Mass spectrometry measurements

Proteins of camel milk cheese, produced using mesophilic and thermophilic cultures, had a complex peptide profile, which was analysed by LC-MS/MS. Peptides in camel milk cheeses found were β-CN- and α-CN-derived peptides (established in the current study).

Figure 1 shows the β-CN-derived peptide with the sequence GLHPVPQPLVPVIA. The observed molecular mass of the peptide was 1435.8551 Da. The MS/MS spectrum provided various b- and y-ions from the fragmentation of peptide GLHPVPQPLVPVIA. Ion b5 with a mass-to-charge ratio (m/z) of 583.2745 corresponds to the fragment GLHPV; ion b6 with m/z 698.3413 corresponds to the fragment GLHPVP; ion b7 with m/z 813.4079 corresponds to the fragment GLHPVPQ; ion b8 with m/z 928.4745 corresponds to the fragment GLHPVPQP; ion b9 with m/z 1043.5411 corresponds to the fragment GLHPVPQPL; and ion y4 with m/z 503.2785 corresponds to the fragment VPIA.

A Mascot score of 73 indicated a high-confidence match; however, it could not be directly referred from Figure 1. The Mascot software obtained Mascot score from the quality and several matched fragment ions of MS/MS spectrum. By analysing the peaks corresponding to b- and y-ions in the spectrum and their intensities, the score provided a quantitative measure of confidence in peptide identification. GLHPVPQPLVPVIA was identified as a high-confidence peptide, and the observed masses of the fragment ions matched well with theoretical values. The experimental mass (1435.8597 Da) was very close to the calculated mass (1435.8551 Da), with a deviation within acceptable limits (24.4 ppm), indicating high mass accuracy (Table 1). The b- and y-ions provided strong evidence supporting the peptide sequence. The detailed LC-MS/MS spectrum analysis confirmed the presence of the peptide GLHPVPQPLVPVIA by matching observed fragment ions (b5, b6, b7, b8, b9, and y4) to their theoretical m/z values. This high-confidence identification, supported by the Mascot search results, demonstrated the reliability of this particular analysis of β-CN and α-CN peptides. In this study, it was revealed that most of the identified peptides had common sequences. It was also observed that the largest number of peptides was detected by β-CN casein origin.

Figure 1. MS/MS spectrum of sequence GLHPVPQPLVPVIA. Following sequence interpretation and database searching (NCBI), the peptide was identified from β-CN casein at 20 days of cheese ripening. Observed mass: 1435.8551 Da; ions score: 73.

Chemometric analysis of peptides during cheese ripening

Changes in the peptide dynamics of camel milk cheese with mesophilic and thermophilic cultures during the 30 days of ripening are presented in Figure 2. The results obtained by chemometric analysis using the SIMCA software show that the PCA chemometric method successfully classified samples into two groups.

Figure 2. The principal component analysis (PCA) score scatter plot presents (A) the clustering of mesophilic and thermophilic data, and (B) the clustering of mesophilic and thermophilic samples on the 1st, 10th, 20th, and 30th day of ripening. A = 4.

In addition, the two groups, mesophilic and thermophilic, were well separated, as shown in Figure 2a. Within these groups, clusters were obtained for the observations based on the ripening period of sample groups. In addition, as seen in Figure 2A, all thermophilic samples were grouped together; however, two mesophilic samples, both of 30 days of ripening, were significant outliers above Hotelling’s ellipse. Figure 2B presents mesophilic and thermophilic samples on the 1st, 10th, 20th, and 30th day of ripening for each category. Only one sample, particularly from the mesophilic group, on the 1st day of ripening was closer to mesophilic samples on the 10th day of ripening.

Figure 3 presents PCA score plots of the ripening results for 30 days. In Figure 3A, for mesophilic group samples, the 20- and 30-day measuring intervals had good separation and repeatability among the replicates of each interval (cluster). Besides, two mesophilic samples were significant outliers on the 1st and 10th day of ripening. In addition, there was only one sample from the 1st day of ripening which emerged in the group of 10 days of ripening, and this outcome was in agreement with the data given in Figure 2A. Furthermore, Figure 3B shows, for thermophilic samples, an excellent separation of samples in four time intervals; however, two outliers appeared from the 1st day and the 20th day of ripening, and this was not observed in Figure 2B. Overall, good separation and repeatability among the replicates of each group (mesophilic and thermophilic) were observed for the 30-day ripening interval.

Figure 3. The principal component analysis (PCA) score scatter plot over 30 days of ripening for (A) mesophilic, and (B) thermophilic samples. A = 3.

Identification of bioactive peptides

The biofunctionality of peptides was determined using an online database, and ACE inhibition and antioxidant properties were observed, comparable to other cheese studies (Nielsen et al., 2017). In Tables 1 and 2, the peptides identified from camel milk cheeses during 30 days of ripening are presented; these were collected using mesophilic and thermophilic cultures during 30 days of ripening.

The bioactive peptides obtained are fragments of β-CN-derived peptides, which are more sensitive to proteolytic enzymes than α-casein. In all, 21 peptides in cheeses using mesophilic cultures for its production were detected; five peptides of these were detected on the 1st day of ripening. Seven peptides were detected on the 10th day of ripening. Four peptides were detected on the 20th day, and five peptides on the 30th day of ripening. In camel milk cheeses using thermophilic culture for their production, 17 peptides were identified. Of these peptides, four peptides were detected on the 1st day, three peptides each on the 10th day and 20th day of ripening, and four peptides on the 30th day of ripening, which matched the fragments of β-CN-derived peptides.

Less number of peptides were also identified in α-CN origin. In camel cheeses prepared using mesophilic culture in fragments of α-CN origin, one peptide each was found on the 1st, 10th, and 30th day, but two peptides were identified on the 20th day of ripening.

However, slightly more peptides relevant to the α-CN origin were identified in cheeses using thermophilic cultures, compared to mesophilic cultures. There were three peptides on the 1st day, and two peptides each on the 10th, 20th, and 30th day of ripening.

For the first time, 508 novel peptides were identified in camel milk cheeses, of which 52 had potential biological activity, as confirmed by MBPDB.

The precursor MPVQA enclosed in the AMPVQAVLPFQEPVPDPVR sequence, released from β-CN-derived peptides also identified in samples with mesophilic and thermophilic cultures, was described as dipeptidyl-peptidase IV (DPP-IV) inhibitory activity with a maximum inhibitory concentration of 93.3 micromolars (µM) (IC50), had a role in the potential regulation of glycemia in humans (Nongonierma et al., 2018). In addition, the precursor QEPVPDPVR from β-CN-derived peptides matched the peptides derived in the study done by Parastouei et al. (2022) and characterised with ACE-inhibitory potential. Most of the identified peptides in our study had common sequences; for example, the GLHPVPQPLVPVIA peptide was identified in the production of cheese using mesophilic culture on the 1st day and 20th day of ripening, having QP precursor (Gómez-Ruiz et al. 2006), VP (Norris et al., 2014), and PVPQP (Gupta et al., 2009; Kohmura et al., 1989), which, as reported, had antioxidant, ACE-inhibiting, antimicrobial, and immunomodulatory properties. ETIIPK peptide was found in cheese ripened by thermophilic starter culture on the 1st and 30th day of ripening, and precursor AVVSPIQF was identified in α-CN derived in cheeses using mesophilic culture. Murali et al. (2021) while studying whey protein hydrolysates showed anticancer and anti-inflammatory properties of camel whey. El Hatmi et al. (2016) while studying fermented camel milk with S. thermophilus LMD-9 found the same peptide AMPVQAVLPFQEPVPDPVR. In our study, the peptide identified in cheese using thermophilic culture on the 1st day of ripening, AMPVQAVLPFQEPVPDPVR (β-CN-derived peptides), matched the data that demonstrated anti-inflammatory properties. The α-CN-derived ILELAVVSPIQFR peptide found in cheeses using thermophilic culture on the 1st day and 30th day of ripening, as well as using mesophilic culture on the 1st day of ripening, was attributed to DPP-IV inhibitory features (Nongonierma et al., 2018). In addition, ILELAVVSPIQFR peptide of α-CN origin showed ACE inhibitory and antioxidant properties (Erhardt et al., 2016). Many potential bioactive peptides, all from β-CN, are ACE inhibitors in vitro or have a hypothetical effect in vivo. Camel milk fermented with Lacticaseibacillus rhamnosus (Lb. rhamnosus) demonstrated ACE inhibitory properties higher than that of cow’s fermented milk (Moslehishad et al., 2013; Murali et al., 2021). ILELAVVSPIQFR, GLHPVPQPLVPVIA, and VLPVPQQMVPYPQR peptides also were identified in camel milk (Cebo and Saadaoui, 2015), but their biofunctionality was not detected. Peptide QENIDELKDTR found in α-CN origin using thermophilic culture on the 20th day of ripening was also detected in camel milk’s α-CN (Cebo and Saadaoui, 2015), but bioactivity was not determined.

Discussion

This study aimed to investigate the effect of different starter cultures on the formation of bioactive peptides in camel milk cheese. Camel milk cheese, studied during 30 days of ripening was produced by different starter cultures, and showed peptide profiles throughout the entire ripening period. It should be noted that while the cheeses were briefly treated in hot whey (85–95°C for 5 min), which could reduce the activity of some microbial cultures, spore-forming bacteria and heat-resistant enzymes still could contribute to proteolysis during storage.

Principal component analysis revealed clear clustering patterns depending on both ripening time and type of starter cultures. Samples aged 20–30 days formed well-separated clusters, indicating that prolonged ripening significantly altered peptide profile in both mesophilic and thermophilic cheeses. This confirmed the progressive quality of proteolysis and the accumulation of specific peptides over time. In contrast, samples obtained on the 1st day and 10th day overlapped with each other, suggesting that early proteolytic changes were less pronounced and that peptide differentiation became more distinct after prolonged ripening.

Moreover, PCA plot for thermophilic cheeses (Figure 3B) showed a clear differentiation of samples across four ripening stages, indicating a consistent progression of proteolysis over time in this group. Compared to mesophilic PCA results (Figure 2B), thermophilic samples showed clearer temporal differentiation, which reflected the faster and more intense proteolytic activity typically associated with thermophilic cultures.

However, two outliers were found in the cheese with thermophilic cultures, namely, one sample from day 1 and another from day 20 were located outside the expected clusters. These deviations, absent in mesophilic PCA, could be the result of minor inconsistencies in enzyme–substrate interactions, differences in starter quality, or technical peculiarities. Despite these anomalies, the overall PCA structure remained stable and confirmed the dynamic, culture-dependent evolution of peptide profile during cheese ripening.

Identified peptides had ACE inhibitory (Tagliazucchi et al., 2016), antimicrobial, and DPP-IV inhibitory properties (Nongonierma et al., 2018). Such peptides appeared to be related to each other, which could be the result of a diverse and complex set of proteolytic enzymes acting on the cheese matrix (Gupta et al., 2009; Korhonen and Pihlanto, 2006). The ILELAVVSPIQFR sequence found in α-CN origin in cheeses produced using mesophilic and thermophilic cultures had precursors of shorter sequence peptides such as ILELA, which showed DPP-IV inhibitory activity (Nongonierma et al., 2018). The precursor peptide EL had antioxidant properties (Suetsuna et al., 2000), whereas the precursor peptide VSP had ACE inhibitory properties (Weimann et al., 2009). Based on the above-mentioned results, camel milk cheeses using mesophilic and thermophilic cultures for production contain peptides that have biological activity and beneficial to human health in various ways.

The use of thermophilic cultures in cheese production resulted in the earlier and more intense formation of key peptides, such as VLPVPPQQQMVPYPQR and GLHPVPQPLVPVIA, which exhibit ACE inhibitory and antioxidant activities. These peptides were detected on day 0 and persisted throughout the maturation period, indicating the high proteolytic activity of thermophilic bacteria at elevated temperatures (Fernández-Esplá and Rul, 1999). In contrast, cheeses made with mesophilic cultures had a broader peptide profile and contained longer sequences, such as IYTFPFPQPQSLVYSHTEPIPYPPYPILPQQNFLPPLQPAVM. This indicates slower but more sustained proteolysis. Some peptides with potential immunomodulatory functions, such as EMPLLQSPVVVPFTESQSLTLTDLTDLENLHLPLPLPLLQSL, were detected only in cheeses prepared using mesophilic cultures. This is probably due to specific enzyme–substrate interactions inherent to the mesophilic microbiota.

Although there was some overlap in the identities of peptides between two culture types, there were significant differences in the timing, number, and diversity of bioactive sequences. These results suggested that thermophilic cultures could be more suitable for the accelerated production of specific short-chain bioactive peptides, whereas mesophilic cultures had greater peptide complexity and a longer generation time. It is obvious from the results that the vast majority of peptides originate from β-CN, while the number of peptides that originate from α-CN is significantly less, as shown in Table 2. Considering the overall peptide profile, the greater presence of β-CN peptides is mainly explained by the fact that this casein fraction is more susceptible to hydrolysis. This is due to its open conformation, mainly the high content of proline in its primary chain (Bhat et al., 2016).

One of the methodological limitations of this study was reliance on conventional LC-MS/MS techniques, which typically employed data-dependent acquisition (DDA) and were limited to targeted peptide identification. This approach could overlook low-abundance or unexpected bioactive peptides, especially in complex cheese matrices. In contrast, recent advances in untargeted proteomics using high-resolution mass spectrometry (HRMS), such as Orbitrap or Q-TOF instruments, allow for deeper and more comprehensive peptide profiling with improved sensitivity and resolution (Toldrá et al., 2021). In spite of this, research in the field of biopeptides offers promising opportunities, as various strains are used in the food industry, including cheese production, which show significant utility. Biopeptides have the potential to replace some pharmaceuticals, thus embodying the idea of “health through nutrition” (Kurbanova et al., 2022). From a broader food science perspective, identifying peptides with potential biological activity (e.g. ACE inhibitors, antioxidants, and DPP-IV inhibitors) highlights the potential of camel milk cheese as a functional food product. The data obtained in this study would contribute to improve cheese production technology, as it provided insights into the peptide profiles generated during the fermentation of camel milk using specific cultures.

In addition, camel milk cheeses were studied previously as well for their biopeptide profile. While these properties require further in vitro and in vivo validation, our findings enrich databases on peptides from lesser-studied milk sources and pave the way for developing bioactive ingredients for dairy-based health products. The future studies must focus on determining the bioactivity of peptides in camel milk cheeses in vivo and in vitro to fill this gap in knowledge about their functionality and potential health effects, and to provide producers and consumers officially substantiated information on functionality and potential health outcomes of locally produced cheeses.

Conclusions

In this study, camel milk was fermented by mesophilic and thermophilic cultures. The peptides were analysed using LC-MS/MS through the Mascot 2.6.1 software. The biological activity of peptides and their precursors was identified in novel, 52 identified peptides using the MBPDB and Peptide Ranker database. Of the identified peptides, 21 were derived from β-CN in cheese samples using mesophilic culture, while five peptides were derived from α-CN. On the other hand, 17 peptides were identified in cheese using thermophilic culture, with nine peptides derived from α-CN. β-CN- and α-CN-derived peptides differ using mesophilic and thermophilic cultures in producing camel cheese during 30 days of ripening. According to the MBPDB database, they have ACE inhibitory, antioxidant, antimicrobial, and DPP-IV inhibitory properties. Furthermore, this study demonstrated that thermophilic cultures promoted earlier and more intense release of specific bioactive peptides, which persisted throughout the ripening and were associated with ACE inhibitory and antioxidant activity. In contrast, mesophilic cultures resulted in a broader and more diverse peptide profile, such as longer sequences with potential immunomodulatory effects. In spite of some overlap, marked differences were observed in peptide diversity, timing, and functioning. PCA confirmed distinct culture- and time-dependent evolution of peptide profiles, with thermophilic samples exhibiting more defined temporal separation. These findings suggested that culture type must be selected strategically to modulate the biofunctional potential of camel milk cheese during ripening.