Preliminary evaluation of macroalgal protein content

Before LC-HRMS peptide annotation, the total extractable protein content of the investigated macroalgae was evaluated using the BCA assay (Table 1), since protein availability represents a key parameter influencing the generation of peptide hydrolysates. Overall, the protein content showed a marked species-dependent variability, ranging from 1.10 ± 0.04% in Codium fragile to 9 ± 2% in Chaetomorpha linum. Among green algae, Chaetomorpha linum displayed the highest protein content, whereas Codium fragile showed the lowest value. Brown algae exhibited intermediate-to-high protein levels, with Halopteris scoparia and Sargassum vulgare showing relatively high protein contents, while Rugulopteryx okamurae presented a lower value. Red algae generally displayed intermediate protein contents, ranging from 1.3 ± 0.2% in Chondrophycus tenerrimus to approximately 4% in Sphaerococcus coronopifolius and Jania rubens. This variability is consistent with the well-known dependence of seaweed biochemical composition on species, taxonomic group, environmental conditions, and physiological status [35]

It should also be noted that some species showed higher variability among biological replicates, as reflected by larger standard deviations. This was particularly evident for Gongolaria gibraltarica (4 ± 2%), and to a lesser extent for species such as Ulva rotundata and Laurencia obtusa. Such lower repeatability is likely attributable to the intrinsic heterogeneity of wild-collected macroalgal biomass, including differences in thallus morphology, developmental stage, epiphyte removal efficiency, local microenvironmental exposure, and extraction efficiency. Therefore, the observed variability should not be interpreted only as analytical variability of the BCA assay, but also as a reflection of the biological heterogeneity of the collected algal material. For this reason, four independent biological replicates were processed for each species, each undergoing independent extraction, hydrolysis, ultrafiltration, and LC-HRMS analysis, in order to retain biological variability while ensuring robustness of the comparative peptidomic analysis.

Short-chain peptides annotation

The proposed suspect screening pipeline enabled an extensive characterization of the short-chain peptide fraction across the investigated algal species. Short-chain peptides were annotated using a specialized processing workflow on Compound Discoverer 3.1, as previously described [32]. Overall, 332 short-chain peptides were putatively identified after manual interpretation of the MS/MS spectra, including 102 dipeptides, 174 tripeptides, and 56 tetrapeptides (Table S1). The Kinetex XB-C18 stationary phase was selected on the basis of previous short-chain peptidomics studies from our group [36, 37] in which this phase provided robust retention, reproducible chromatographic performance, and effective UHPLC-HRMS profiling of di-, tri-, and tetrapeptides. Nevertheless, due to the broad polarity range of short peptides, no single stationary phase can be considered fully comprehensive. Previous comparisons of C18 with porous graphitic carbon and HILIC separations demonstrated that these chromatographic mechanisms provide complementary selectivity and may increase peptide coverage. Therefore, the present C18-based method was intended as a robust comparative profiling platform for macroalgal hydrolysates, while orthogonal stationary phases could be explored in future studies to further expand the coverage of highly polar or poorly retained short peptides. As expected for very short peptide sequences, most annotated compounds eluted at early retention times, in agreement with their limited hydrophobic retention under reversed-phase conditions. However, the chromatographic method allowed the detection of both early-eluting polar peptides and more retained hydrophobic sequences, supporting its suitability for comparative short-chain peptide profiling. A representative full-scan base peak chromatogram is reported in Figure S1.

According to an adapted confidence-level classification based on Schymanski/MSI criteria, these annotations should be considered confidence level 2 identifications since they were supported by accurate mass measurements and diagnostic MS/MS fragmentation patterns but were not confirmed using authentic analytical standards [38]. Table S1 reports the identification data for the tentatively annotated sequences. Since leucine (Leu) and isoleucine (Ile) are isobaric residues that cannot be unequivocally discriminated using conventional MS2 fragmentation data alone, peptide annotations containing these amino acids were classified as confidence level 3 identifications according to adapted Schymanski/MSI criteria. Therefore, the three-letter notation “Xle” and the one-letter code “J” were employed to indicate either Leu or Ile residues. In particular, the putatively identified sequences spanned di-, tri-, and tetrapeptides, with a predominance of dipeptides, consistent with the applied ultrafiltration cutoff and enzymatic hydrolysis conditions. Indeed, enzymatic hydrolysis with Alcalase® is considered a straightforward process that can be easily controlled and optimized to produce a hydrolysate rich in short-chain bioactive peptides, as smaller peptides are more likely to reach the bloodstream and target organs intact [17]. The putatively identified short-chain peptides covered a broad compositional and physicochemical space, as reflected by differences in amino acid composition, sequence arrangement, peptide length, and GRAVY index values reported in Table S1. In this context, structural diversity refers to the variability in residue combinations and sequence order among di-, tri-, and tetrapeptides, rather than to large conformational or higher-order structural features. This molecular diversity highlights the remarkable biochemical complexity of algal-derived hydrolysates and supports their potential as reservoirs of structurally diverse peptide sequences associated with predicted bioactivities. From a compositional perspective, the dataset revealed the recurrent presence of small hydrophobic and neutral amino acids, especially Val, Ala, Gly, Xle (Leu/Ile), Pro, Phe, Tyr, and Met, indicating that the peptide pool is largely hydrophobic to amphipathic in character. The predominance of hydrophobic and amphipathic peptide motifs is particularly relevant from a nutraceutical perspective, since these structural features are frequently associated with antioxidant, antihypertensive, and multifunctional bioactive properties [39, 40].

Comparative statistical analysis of aacroalgal peptidomes

Principal Component Analysis (PCA) performed on the normalized peak areas of all 332 annotated peptide sequences revealed a certain degree of clustering trend among the different macroalgal groups (brown, green, and red algae) (Fig. 2).

Fig. 2

Principal component analysis (PCA) score plot (left) and corresponding loadings plot (right) obtained from the normalized peak areas of all identified peptide sequences in brown, green, and red macroalgae samples

It should be noted that PCA was used here as an exploratory and unsupervised tool to visualize the main sources of variance in the peptide dataset, rather than to develop a predictive classification model. Therefore, the observed clustering trends were interpreted together with univariate statistical outputs, avoiding supervised modeling approaches that could overfit the data given the exploratory nature of the study and the limited sample size relative to the number of peptide variables.

The first two principal components explained a substantial proportion of the total variance (PC1 44.7%, PC2 13.4%), highlighting differences in peptide composition among species. Brown algae showed a more distinct distribution pattern compared to red and green algae. Inspection of the corresponding loadings plot indicated that many peptides, especially dipeptides, were strongly associated with the brown algal samples. Since short dipeptides are widely recognized for their potential bioactive properties, these observations suggested that brown algae may represent a particularly rich source of structurally diverse short-chain peptides. To further investigate the distribution of short-chain peptides, we focused specifically on the dipeptide fraction by extracting a second data matrix containing only identified dipeptides. Multivariate PCA resulted in a higher proportion of the total variance explained by the first two components (PC1 55.9%, PC2 11.8%) and a clearer clustering of brown algae (Figure S2). Later, one-way ANOVA was performed to evaluate differences among brown, green, and red macroalgae. As shown in the significance plot (Fig. 3A), several dipeptides exhibited statistically significant differences in abundance across the algal groups, with many of the most significant features associated with brown algae. The overall distribution of normalized dipeptide peak areas (Fig. 3B) indicated that brown algae generally contained higher dipeptide levels compared to green and red species.

Fig. 3

Statistical analysis of dipeptide abundance in brown, green, and red macroalgae. A Volcano-style plot showing the significance distribution of identified dipeptides obtained from one-way ANOVA performed on the dipeptide-only data matrix; selected significantly discriminant dipeptides are highlighted. B Box-and-whisker plot representing the overall distribution of normalized dipeptide peak areas among algal groups, showing a higher dipeptide content in brown algae. C–F Representative box-and-whisker plots of significantly different dipeptide sequences (KQ, FY, PV, and FV) identified by ANOVA. Statistical significance is indicated by the following: *p < 0.1; **p < 0.01; ***p < 0.001; ****p < 0.0001

To illustrate these trends, selected examples of significantly different dipeptides are reported in the box-and-whisker plots (Fig. 3C–-F), including KQ, FY, TY, and FV.

These representative sequences showed significantly higher abundances in brown algae, supporting the hypothesis that brown macroalgae are enriched in potentially bioactive dipeptides relative to the other algal groups. Based on these findings, the subsequent analyses were focused exclusively on brown algae, still using a data matrix composed only of 102 dipeptides, to compare native and invasive species. The PCA score plot shows a substantial overlap between edible and invasive brown algae samples, indicating the absence of a strong statistical separation between the two groups based on their dipeptide composition (Fig. 4).

Fig. 4

Volcano plot highlighting differential dipeptide abundance between invasive and edible brown algae (left). Positive log2(FC) values indicate dipeptides more abundant in invasive species, while negative values indicate higher abundance in edible species. Multivariate and differential analysis of dipeptide profiles in edible and invasive brown algae. PCA score plot based exclusively on the dipeptide data matrix (right), showing the distribution of edible and invasive brown algae samples according to their dipeptide composition

The partial overlap between the two groups indicates the absence of a strong statistical separation, suggesting a highly comparable dipeptide profile between edible and invasive species. PC1 and PC2 explain 59.3% and 17.0% of the total variance, respectively.

This result supports the hypothesis that invasive brown algae may represent a valuable and sustainable source of bioactive dipeptides. Therefore, based on our results, biomass derived from algal blooms could be effectively valorized through recovery strategies aimed at producing formulations enriched in bioactive compounds, contributing both to environmental management and to the development of candidate peptide-enriched hydrolysates for future functional investigations. The volcano plot (Fig. 4) further supports this interpretation. Since the analysis was performed exclusively on dipeptides, the observed differences specifically reflect variations in dipeptide abundance between invasive and edible brown algae. Compounds such as JY, JV, VJ, and TJ appear markedly upregulated in invasive samples, whereas only a limited number of dipeptides are more abundant in edible algae. Overall, the volcano plot reveals a general trend toward higher abundance of specific dipeptides in invasive brown algae, reinforcing their potential as an alternative and sustainable source of bioactive compounds for nutraceutical applications.

Distribution of essential amino acid-containing short-chain peptide sequences in red, green, and brown algae

From a nutraceutical point of view, the amino acid (AA) composition of protein hydrolysates is commonly evaluated as it is considered one of the most important aspects influencing the formulations’s nutritional value, as peptides rich in essential amino acids (EAAs) contribute to meeting the daily dietary requirements of humans [41]. The focus on EAA-containing short-chain peptides is particularly relevant in the nutraceutical field, as bioactive peptides enriched in essential amino acids have been associated with several health-promoting properties, including antioxidant, antihypertensive, antimicrobial, anti-inflammatory, and metabolic regulatory activities [42, 43]. In fact, short-chain peptides intrinsic enhanced gastrointestinal stability, absorption efficiency, and bioavailability do not only rely on sequence length, but they strongly depend on AA composition and sequence. Consequently, identifying specific EAA-containing peptide signatures may provide valuable insight into the nutraceutical potential and functional properties of algae-derived protein hydrolysates. We subsequently filtered the data matrix to retain only short-chain peptide sequences containing EAAs together with their relative abundances across all samples. Specifically, sequences containing methionine (M), phenylalanine (F), tyrosine (Y), tryptophan (W), threonine (T), and lysine (K) were selected for further analysis. In contrast, isoleucine (I) and leucine (L) were excluded for two main reasons: first, these isomeric amino acids cannot be reliably distinguished using conventional MS/MS approaches; and second, they were highly abundant across most detected sequences, potentially limiting their specificity and biological relevance. The multivariate and univariate analyses of EAA-containing short-chain peptides revealed distinct patterns depending on the amino acid class considered. PCA performed on the data matrix comprising M- and K-containing peptides showed extensive overlap among the brown, green, and red algal groups, with no statistically significant discrimination observed (Figure S3A–D). Although minor clustering tendencies were visible along PC1 and PC2, the broad overlap of confidence ellipses and the non-significant ANOVA results indicate that M- and K-containing peptide profiles were relatively conserved across the investigated algae species. This suggests that these peptide classes may represent common structural or metabolic features shared among the taxa rather than species-specific bioactive signatures. In contrast, the analysis of F-, Y-, W-. and T-containing peptides revealed a markedly different pattern and ANOVA identified significant differences among algal groups, indicating that aromatic EAA-containing peptides contributed strongly to species discrimination. The PCA and subsequent statistical analyses demonstrated that several F-, Y-, W- and T-containing peptide sequences were differentially abundant across brown, green, and red algae, supporting the hypothesis that aromatic amino acid-enriched peptides are more taxonomically and functionally informative (Figure S4A–D). This observation is particularly relevant from a nutraceutical perspective because peptides containing aromatic residues such as F, Y, and W are frequently associated with antioxidant and radical-scavenging activities due to their electron-donating and resonance-stabilizing properties [44], while T-containing peptides may contribute to immune modulation and metabolic regulation [45]. The univariate analyses further highlighted specific peptide sequences exhibiting strong statistical significance, as reflected by elevated − log10(p-value) scores and significant ANOVA outcomes. In the F-containing peptide dataset, sequences such as FT, DJF, JF, and VVF were among the representative peptides showing significant differential abundance, with several aromatic peptide sequences exhibiting higher relative abundances in green algae, suggesting a potential enrichment of antioxidant-related peptide motifs in this group (Figure S5A–F). Green algae species were also characterized by statistically significant production of W-containing short-chain sequences, with WN, AW, JW, and VW exhibiting significant differential abundance (Figure S6A–F). Similarly, within the Y-containing peptide dataset, sequences including FSY, AY, YA, and YPJE showed highly significant intergroup variation, with several peptides markedly enriched in brown algae (Figure S7A–F). Finally, T-containing short-chain peptide sequences appeared to be particularly enriched in red algal species, as evidenced by their higher relative abundances within this phylum (Figure S8A–F). Overall, these findings suggest that while M- and K-containing peptides do not provide strong discriminatory power among algal species, aromatic amino acid-containing peptides, particularly those enriched in F and Y residues, represent more distinctive biochemical signatures within the generated protein hydrolysates. Their differential abundance across algae groups likely reflects differences in the original protein composition and susceptibility to enzymatic hydrolysis among the algal species. From a nutraceutical perspective, these results are particularly relevant because they may support the development of plant-based peptide-enriched formulations with targeted functional properties. Identifying algal species that preferentially generate hydrolysates enriched in specific essential amino acid-containing peptides could therefore facilitate the selection of optimal raw materials to produce tailored nutraceutical ingredients with enhanced antioxidant, metabolic, or health-promoting potential.

In silico assessment of predicted bioactivities, sensory properties, antifungal potential, and antioxidant capacity of short-chain peptides from the invasive alien brown species

The following analyses should be interpreted as exploratory prioritization strategies aimed at identifying candidate peptide sequences potentially associated with bioactive and sensory properties.

Table S2 presents the in silico bioactivity and sensory characterization of the 30 short-chain peptide sequences showing the highest mean normalized peak areas within invasive brown algal samples. This subset was not selected as a statistical marker panel, but rather as an exploratory prioritization of the most abundant peptides in the invasive brown biomass for subsequent bioactivity and sensory prediction. These peptides were selected following multivariate analysis, where brown algae exhibited a more distinct peptide distribution pattern compared with red and green algae (Fig. 2). Since no statistically significant differences were observed between invasive and non-invasive brown algae, the invasive biomass was prioritized because of its potential for environmental reuse and valorization. The selected peptides were screened using several open-access peptide databases and prediction tools, including BIOPEP-UWM [46], AHTPDB [47], Bioactive Peptides, TastePeptidesDB, Umami_YYDS, Umami_IP [48], AntiFungiPept Predictor [49], and AnOxPePred [50].

The results suggest that the peptide fraction from invasive brown algae is enriched in multifunctional short peptides with predicted antihypertensive, antidiabetic, antioxidant, sensory, and antifungal activities. Several peptides, including VY, YJ, AY, JY, VV, and SF, were particularly notable because they combined multiple predicted activities, since they were identified as known bioactive sequences in BIOPEP-UWM and AHTPDB, particularly as angiotensin-converting enzyme (ACE) inhibitors and dipeptidyl peptidase-IV (DPP-IV) inhibitors. These activities are commonly associated with antihypertensive and antidiabetic effects, respectively, suggesting that invasive brown algae may represent a promising source of health-promoting peptides. These peptides also exhibited some of the highest free radical scavenging (FRS) scores, with JY (0.502), YJ (0.494), AY (0.492), and VY (0.489) showing the strongest predicted antioxidant potential, aligned with the enhanced radical scavenging capacity of aromatic amino acids-containing peptides. The sensory analysis revealed that bitterness was the dominant predicted taste attribute among the identified peptides. Many hydrophobic short-chain peptides, particularly those enriched in V, Y, A, or F residues, were classified as bitter by both BIOPEP and TastePeptidesDB. This observation is consistent with previously reported relationships between peptide hydrophobicity and bitter taste perception [51]. In contrast, several peptides, including JGVS, GVJ, GJE, GTVA, ATJ, and SF, were predicted to possess umami characteristics, with VV and GVJ showing the highest umami intensity scores. These findings suggest that invasive brown algae may also contain flavor-enhancing peptides with potential applications in functional foods or food formulation. The simultaneous occurrence of predicted bioactive and flavor-related peptides further supports the potential exploitation of invasive brown algae as multifunctional ingredients for future food.

Although these analyses remain predictive and require experimental validation, they provide preliminary insights into the multifunctional potential of algae-derived short peptides.

The antifungal prediction analysis further highlighted the biofunctional potential of these peptides. Several sequences displayed high probabilities of antifungal activity, particularly JY, AJJJ, VJ, GVJ, VY, and YJ, with prediction probabilities exceeding 99% in some cases. In addition, some peptides demonstrated elevated predicted activity against specific fungal species, especially Candida neoformans and Candida albicans. The differences between antifungal prediction probabilities and AFI values arise because they capture distinct aspects of peptide behavior. The antifungal probability typically reflects how similar a sequence is to known antifungal peptides based on classification models, while AFI is a physicochemical index estimating potential functional potency based on properties such as charge, hydrophobicity, and amphipathicity. As a result, some peptides show high antifungal probability but low AFI because they resemble known antifungal motifs but lack strong membrane-active characteristics, whereas others display high AFI but low probability because they possess strong physicochemical features associated with antifungal activity but are not similar to sequences in the training dataset. Therefore, reliable identification of potential antifungal peptides should rely on a combined approach that considers both measures, prioritizing sequences where both metrics agree while also allowing high-AFI outliers for possible novel candidates, alongside basic physicochemical constraints. Although these predictions require experimental validation, they suggest that peptides derived from invasive brown algae may possess antimicrobial properties of potential interest for nutraceutical, pharmaceutical, or food preservation applications. Considering that Rugulopteryx okamurae is currently causing severe ecological and socioeconomic impacts along the southern Spanish coastline, including biodiversity loss, alteration of native benthic communities, and major accumulation events affecting coastal ecosystems and human activities [52], any additional knowledge regarding its molecular composition and functional properties contributes added value toward the sustainable valorization of this invasive biomass. Overall, these findings support the potential valorization of invasive species as sustainable and underexplored sources of multifunctional short-chain bioactive peptides. The enrichment of small peptides, especially dipeptides, together with their predicted antioxidant, antihypertensive, antidiabetic, sensory, and antifungal properties, supports the potential valorization of nuisance algal biomass within sustainable biorefinery and circular bioeconomy approaches.

The present study remains exploratory in nature and relies on putative peptide annotation supported by manual MS/MS interpretation and predictive in silico tools. Therefore, orthogonal structural confirmation and experimental validation of the proposed bioactivities will be required in future studies to confirm the functional relevance of the identified sequences.