Combining the Salt‐Tolerant Plants Limonium algarvense Erben and Polygonum maritimum L. Promotes Synergistic Bioactivities for Enhanced Cosmetic Applications

ABSTRACT

The demand for natural and sustainable ingredients in dermo‐cosmetic formulations has promoted the use of coastal salt‐tolerant plants as novel bioresources. This study evaluated aqueous extracts of Limonium algarvense and Polygonum maritimum, two salt‐tolerant coastal plants and their mixtures (3:1, 1:1, 1:3, w/w) for antioxidant, metal‐chelating, enzyme‐inhibitory, antimicrobial and cytotoxic properties. Chemical profiling revealed diverse phenolic signatures dominated by catechin, myricetin and quercetin derivatives, which are known contributors to antioxidant and enzyme‐modulating activity. L. algarvense showed the strongest superoxide‐scavenging activity (EC₅₀ = 219 µg/mL), while P. maritimum was the most effective hyaluronidase inhibitor (EC₅₀ = 42.9 µg/mL). The 1:3 mixture displayed the most pronounced synergistic behaviour, notably in hydroxyl‐radical scavenging (EC₅₀ = 62.7 µg/mL; SE = 0.08) and antibacterial activity against Escherichia coli (MIC = 39.68 µg/mL; SE = 0.31). The 3:1 mixture showed enhanced elastase inhibition (EC₅₀ = 33.0 µg/mL; SE = 0.76) and increased copper‐chelating capacity (EC₅₀ = 533 µg/mL, SE = 0.52). All extracts maintained cell viability above 90%, supporting their safety for topical applications. Together, these results indicate that L. algarvense and P. maritimum deliver complementary and synergistic activities relevant to dermo‐cosmetic innovation, emphasising their potential as sustainable, multifunctional ingredients compatible with saline agriculture and low‐freshwater production systems.

1. Introduction

The demand for natural, sustainable and multifunctional ingredients in skincare formulations has grown steadily in recent years. The global skincare market was valued at $146.7 billion in 2021 and is projected to reach $273.3 billion by 2031 [1]. To remain competitive, dermo‐cosmetic companies are investing heavily in research and development, seeking innovative formulations and ingredients that combine efficacy with environmental and social responsibility [2, 3]. Consumers increasingly favour products that not only meet cosmetic needs but also incorporate natural and organic components aligned with eco‐friendly and ethical principles [3].

A major goal of skincare formulations is to mitigate or slow visible signs of skin ageing, which are closely linked to environmental stressors such as UV radiation and pollution [4]. These factors increase the production of reactive oxygen species (ROS) that oxidize cellular components (nucleic acids, proteins and lipids) causing damage and impairing biological function [5]. Consequently, antioxidants are widely included in dermo‐cosmetic products to reduce oxidative stress and delay ageing. In addition, specific bioactive compounds are employed to inhibit key enzymes associated with skin ageing and loss of elasticity, including tyrosinase, elastase, collagenase, hyaluronidase and lipase [6, 7, 8]. By suppressing these enzymes, such formulations can effectively reduce wrinkles, pigmentation, inflammation and other ageing manifestations, promoting healthier skin. Antimicrobial ingredients are also essential in dermo‐cosmetic formulations, serving both dermatological and preservative roles. Bacteria and fungi can disrupt the skin microbiome, leading to acne, dermatitis or infections [9], while microbial contamination may compromise product safety and stability. Natural antimicrobials thus contribute to product preservation and skin protection, complementing antioxidant and enzyme‐inhibitory effects [10].

In response to this shift toward natural and sustainable skincare, research has increasingly focused on plant‐derived bioactives that ensure both safety and efficacy. Salt‐tolerant coastal plants (halophytes) have emerged as particularly promising resources, as their adaptation to saline environments promotes the synthesis of protective metabolites such as phenolics and flavonoids with potent antioxidant and enzyme‐inhibitory properties. Among these, Limonium algarvense Erben (sea lavender) and Polygonum maritimum L. (sea knotgrass) are two halophytic species native to Atlantic–Mediterranean coastlines, extensively studied for their phytochemical richness and pharmacological potential [11, 12, 13, 14, 15, 16, 17]. Notably, L. algarvense (LA) has previously shown synergistic antioxidant and cholinesterase‐inhibitory effects when combined with Camellia sinensis (green tea) [18]. However, no studies have examined synergistic interactions between LA and P. maritimum (PM), nor evaluated combined salt‐tolerant plant extracts in a cosmetic context. In addition, both species can be cultivated under greenhouse or saline conditions, ensuring a sustainable and scalable supply of biomass while reducing harvesting pressure on natural populations [16, 19, 20, 21]. This sustainability further reinforces their suitability as environmentally responsible resources for dermo‐cosmetic applications.

Synergistic combinations of plant extracts have been increasingly explored in dermo‐cosmetic research. For example, Ismail et al. [22] demonstrated that palm‐based formulations enriched with Hibiscus sabdariffa extract enhanced skin hydration, while creams combining palm oleochemicals with Ganoderma lucidum extract produced skin‐whitening effects comparable to arbutin‐ or licorice‐based products through inhibition of tyrosinase activity. More recently, Kapini et al. [23] evaluated whole extracts of nine Australian native species and identified a ternary mixture of Citrus aurantium, Tasmannia lanceolata and Mentha australis (1:1:1, w/w) that exhibited synergistic behaviour, reducing nitric oxide and IL‐6 production in LPS‐stimulated RAW264.7 macrophages, improving intracellular antioxidant responses and accelerating wound closure in human dermal fibroblasts. These studies confirm that botanical combinations can potentiate antioxidant, anti‐inflammatory and skin‐repair mechanisms. However, no synergistic strategy has been investigated using Limonium, Polygonum or other salt‐tolerant coastal species. The evaluation of combined aqueous extracts of LA and PM therefore represents a previously unexplored approach in dermo‐cosmetic research.

Based on the distinct yet complementary phytochemical profiles of the two species, we hypothesised that combining LA and PM would produce stronger biological responses than the individual extracts, particularly in antioxidant, enzyme‐inhibitory and antimicrobial assays relevant to dermo‐cosmetic applications. This study was therefore designed to address the current lack of information on the synergistic potential of salt‐tolerant coastal plants. To test this hypothesis, aqueous extracts of LA, PM and their mixtures (3:1, 1:1 and 1:3, w/w) were analysed by ultrahigh‐performance liquid chromatography coupled with high‐resolution mass spectrometry (UHPLC–HRMSⁿ) for comprehensive metabolite profiling and evaluated for antioxidant, enzyme‐inhibitory, antimicrobial and cytotoxic activities. This approach allowed us to determine whether combining these species enhances biological functionality and to assess their suitability as natural multifunctional ingredients for sustainable cosmetic formulations.

2. Results and Discussion

2.1. Chemical Profiling

The UHPLC–HRMSⁿ analysis revealed that the metabolite profiles of LA, PM and their mixtures were dominated by three major chemical groups: flavonoids, condensed tannins, and hydroxycinnamic acid derivatives, with additional contributions from phenylpropanoid amides, coumarin glycosides, and amino acid derivatives (Table 1; Figure S1).

TABLE 1.

List of annotated metabolites in extracts of Limonium algarvense (LA; sea lavender), Polygonum maritimum (PM; sea knotgrass), and their mixtures.

ID	RT (min)	m/z value	Molecular formula	Annotated compound	Adduct	MS2 fragments	LA	LA + PM, 3:1	LA + PM, 1:1	LA + PM, 1:3	PM
1	3.29	166.09	C9H11NO2	Phenylalanine	[M+H]+	120 (100), 131 (2), 149 (3)	++	++	++	++	++
2	8.23	595.14	C30H26O13	Procyanidin	[M+H]+	443 (100), 425 (78), 291 (74), 287 (43), 317 (35), 427 (31), 305 (29), 275 (20)	−	−	−	−	+
3	9.04	307.08	C15H14O7	Epigallocatechin	[M+H]+	139 (100), 289 (70), 151 (53), 169 (8), 181 (6)	−	+	+	+	+
4	10.03	579.15	C30H26O12	Procyanidin B2	[M+H]+	427 (100), 409 (89), 291 (54), 289 (24), 301 (21), 247 (17), 411 (11)	−	−	−	−	+
5	10.09	147.04	C9H8O3	Coumarate	[M−H2O+H]+	119 (100)	−	−	+	+	+
6	10.36	291.09	C15H14O6	Catechin	[M+H]+	139 (100), 123 (88), 273 (22), 147 (8)	−	+	+	−	+
7	10.53	763.15	C37H30O18	Epigallocatechin‐(4β→8)‐epigallocatechin‐3‐O‐gallate	[M+H]+	425 (100), 443 (87), 459 (85), 595 (74), 317 (71), 289 (51), 287 (40)	+	+	+	+	+
8	11.3	761.13	C37H28O18	Epigallocatechin‐(2β→7,4β→8)‐epigallocatechin‐3‐O‐gallate	[M+H]+	407 (100), 425 (89), 743 (86), 617 (52), 455 (47)	++	+	++	++	+
9	11.35	225.08	C11H12O5	Sinapate	[M+H]+	207 (100), 193 (6), 165 (2), 183 (1)	−	−	−	−	+
10	11.37	207.06	C11H12O5	Sinapate	[M−H2O+H]+	175 (100), 192 (7), 147 (5), 119 (2)	−	+	+	−	+
11	11.48	409.11	C17H22O10	1‐Sinapoyl‐d‐glucose	[M+Na]+	247 (100), 185 (13), 379 (2)	−	+	−	−	+
12	11.49	404.15	C17H22O10	1‐Sinapoyl‐d‐glucose	[M+H3N+H]+	207 (50), 242 (3), 175 (2)	−	−	−	−	+
13	11.79	747.15	C37H30O17	Epigallocatechin‐(4β→8)‐epicatechin‐3‐O‐gallate ester	[M+H]+	443 (100), 425 (65), 579 (50), 287 (43), 317 (17)	−	+	+	−	+
14	12.1	459.09	C22H18O11	Epigallocatechin gallate	[M+H]+	289 (100), 139 (26), 151 (20)	+	+	+	+	+
15	12.81	633.11	C28H24O17	Myricetin 3‐(6″‐galloylgalactoside)	[M+H]+	319 (100), 315 (32), 615 (16), 361 (14)	+	+	+	+	+
16	13.2	481.1	C21H20O13	Myricetin 7‐O‐β‐d‐glucopyranoside	[M+H]+	319 (100), 463 (3), 385 (2), 361 (2)	+	+	+	+	+
17	13.22	329.09	C14H16O9	Bergenin	[M+H]+	167 (100), 311 (34)	+	−	−	+	−
18	13.62	317.07	C16H12O7	3‐O‐Methylquercetin	[M+H]+	302 (100)	+	+	+	+	+
19	13.62	625.18	C28H32O16	Narcissin	[M+H]+	317 (100), 479 (17), 302 (5)	+	−	+	+	−
20	13.96	599.2	C27H34O15	Phloretin 3′,5′‐di‐C‐glucoside	[M+H]+	563 (100), 581 (49), 461 (35), 479 (7)	+	−	−	+	−
21	13.99	465.1	C21H20O12	Myricetrin	[M+H]+	319 (100), 447 (8), 429 (6), 361 (1)	+	++	++	++	++
22	14	319.04	C15H10O8	Myricetin	[M+H]+	273 (100), 165 (45), 301 (39), 245 (35), 153 (34)	++	++	++	++	++
23	14.83	551.1	C24H22O15	Quercetin 3‐O‐malonylglucoside	[M+H]+	319 (100), 429 (10), 489 (8), 329 (6), 533 (6)	+	+	+	+	+
24	14.93	573.08	C24H22O15	Quercetin 3‐O‐malonylglucoside	[M+Na]+	255 (100), 341 (48), 529 (34), 511 (5)	+	−	+	+	−
25	15.11	449.11	C21H20O11	Quercitrin	[M+H]+	303 (100), 287 (88), 413 (5), 431 (5)	+	+	+	+	+
26	15.52	303.05	C15H10O7	Quercetin	[M+H]+	257 (100), 229 (89), 285 (53), 165 (51), 247 (38)	++	++	++	++	++
27	16.08	535.11	C24H22O14	Luteolin 7‐O‐(6″‐malonylglucoside)	[M+H]+	287 (100), 517 (2), 329 (2)	+	+	+	+	+
28	16.19	314.14	C18H19NO4	Feruloyltyramine	[M+H]+	177 (100), 145 (8), 121 (1)	+	−	−	+	−
29	16.49	344.15	C19H21NO5	N‐trans‐feruloylmethoxytyramine	[M+H]+	177 (100), 145 (6)	+	−	−	−	−
30	16.51	287.05	C15H10O6	Kaempferol	[M+H]+	287 (100), 153 (21), 241 (6)	+	+	+	+	+

Note: +, presence; ++, presence in higher abundance; −, absence.

Abbreviations: m/z, mass‐to‐charge ratio; RT, retention time.

The metabolite composition of these extracts varied among different species and mixtures. Particularly, several metabolites were uniquely found in either LA or PM, while others were consistently detected across all extracts. PM showed a characteristic enrichment in condensed tannins, particularly procyanidin and procyanidin B2, together with several sinapate derivatives (e.g., sinapate and 1‐sinapoyl‐d‐glucose). These compounds were absent from LA and represent markers of the PM profile. In contrast, LA was distinguished by the presence of phenylpropanoid amides (feruloyltyramine and N‐trans‐feruloylmethoxytyramine), bergenin, and methylthioadenosine sulfoxide, none of which were detected in PM or in the mixtures.

Across all extracts, flavonoids, especially myricetin derivatives (myricetin, myricetin glycosides, narcissin), quercetin derivatives (quercetin, quercitrin, quercetin 3‐O‐malonylglucoside), catechin/epicatechin‐type compounds, and galloylated epigallocatechin dimers, formed the core shared chemical profile. These metabolites were consistently detected in both species and in all mixture ratios.

The mixtures generally retained the major constituents of each plant, combining LA‐specific and PM‐specific metabolites. Among them, the 1:1 mixture displayed the greatest chemical diversity, incorporating nearly all compounds present in the individual extracts. Some metabolites that were undetected in LA or PM alone, such as coumarate and certain epigallocatechin derivatives, appeared in the mixtures, suggesting matrix‐dependent extraction or ionisation effects.

Taken together, these patterns indicate that each species contributes distinct chemical classes to the mixtures, LA providing phenylpropanoid amides and myricetin‐rich flavonoids, and PM contributing procyanidins and sinapate derivatives, resulting in chemically broader and compositionally complementary extracts.

Comparison with previous phytochemical analyses of these species predominantly report simple phenolic acids and common flavonoids [13, 14, 23, 24, 25]. In contrast, the present study identified more complex phenolic structures, including phenylpropanoid amides (e.g., feruloyltyramine) several sinapate derivatives, and higher order flavanol oligomers. Although earlier studies acknowledge phenolic acids and flavonoids as major constituents [11, 15, 17, 26], none reported the presence of phenylalanine, epigallocatechin‐(4β→8)‐epigallocatechin‐3‐O‐gallate, epigallocatechin‐(2β→7,4β→8)‐epigallocatechin‐3‐O‐gallate, sinapate derivatives, bergenin, phloretin 3′,5′‐di‐C‐glucoside, feruloyltyramine, nor N‐trans‐feruloylmethoxytyramine. It is therefore, to the best of our knowledge and based on the available phytochemical literature for both PM and LA, this is the report documenting the occurrence of these metabolites in these species.

2.2. Antioxidant Activity

The antioxidant activity of LA, PM, and their mixtures in various ratios (3:1, 1:1 and 1:3) was evaluated using the DPPH^•, ABTS^•+, O₂ ^•−, and HO^• assays, as well as copper‐chelating activity (CCA) (Table 2). In the DPPH^• assay, LA was the most active single extract (EC₅₀ = 155 µg/mL). Among the mixtures, the LA + PM (1:3) combination exhibited the best synergistic effect (SE = 0.78). The ABTS^•+ assay revealed that PM was the most active extract (EC₅₀ = 560 µg/mL), while the LA + PM (3:1) mixture demonstrated enhanced activity with a SE of 0.72. For the superoxide (O₂ ^•−) assay, LA displayed the lowest EC₅₀ value (219 µg/mL), though none of the mixtures showed synergistic interactions. Regarding hydroxyl radical (HO^•) scavenging activity, PM emerged as the most active extract (EC₅₀ = 139 µg/mL), whereas the LA + PM (1:3) mixture showed the strongest synergistic effect (SE = 0.08). Finally, in the CCA assay, the LA + PM (3:1) mixture showed the best activity (EC₅₀ = 533 µg/mL) with a notable synergistic effect (SE = 0.52).

TABLE 2.

Antioxidant activity (EC₅₀, µg/mL) and synergistic effect (SE) of Limonium algarvense (sea lavender), Polygonum maritimum (sea knotgrass) and their mixtures (3:1, 1:1 and 1:3 ratio).

Extract	DPPH•		ABTS•+		O2 •−		HO•		CCA
	EC50	SE	EC50	SE	EC50	SE	EC50	SE	EC50	SE
LA	155 ± 21b	—	989 ± 116c	—	219 ± 9a	—	2426 ± 373b	—	893 ± 115c	—
LA + PM, 3:1	319 ± 43c	1.3	640 ± 165b	0.72	402 ± 10b	1.5	4497 ± 762c	2.4	533 ± 45b	0.52
LA + PM, 1:1	427 ± 59d	1.3	700 ± 136bc	0.90	385 ± 22b	1.2	1046 ± 622ab	0.82	1740 ± 211e	1.5
LA + PM, 1:3	331 ± 10c	0.78	999 ± 120c	1.5	461 ± 26c	1.3	62.7 ± 28a	0.08	1286 ± 109d	1.0
PM	508 ± 28d	—	560 ± 103b	—	409 ± 10b	—	139 ± 8a	—	1435 ± 52de	—
Gallic acid *	51 ± 9.5a		20.9 ± 2.0a
Catechin *					613 ± 15d		425 ± 0ab
EDTA *									176 ± 5a

Note: EC₅₀ values represent the mean ± standard error of the mean (SEM) of at least three experiments. Values followed by different letters (a–e) are significantly different (p < 0.05) (Tukey HSD test). SE represents antagonistic (SE > 1), additive (SE = 1) or synergistic (SE < 1) effects.

^*Positive controls.

The antioxidant properties of LA and PM are closely linked to their rich phenolic profiles, including flavonoids such as myricetin derivatives, catechins, and tannins, which have been extensively documented as potent free radical scavengers [11, 12, 13, 14, 15, 16, 18]. Compounds like gallocatechol, epicatechin and quercitrin identified in these extracts are known to neutralize ROS and prevent oxidative damage by donating electrons or stabilizing reactive species [24]. Interestingly, the synergistic antioxidant effects observed between LA and PM extracts can be attributed to multiple mechanisms, including the regeneration of potent antioxidants by weaker compounds, formation of stable intermolecular complexes, and potential creation of new bioactive molecules with enhanced activity. These interactions, previously reported in other plant‐based systems, suggest a complex interplay between bioactive compounds that amplifies the overall antioxidant capacity [25]. In addition, combinations of LA with green tea (C. sinensis) have previously been reported to have synergistic effects on antioxidant and neuroprotective properties [11].

From a cosmetic perspective, the observed antioxidant properties are highly relevant, as oxidative stress induced by environmental factors such as UV radiation and pollution is a major contributor to skin aging. The ability of these extracts to scavenge diverse free radicals (DPPH^•, ABTS^•+, O₂ ^•−, HO^•) and chelate metals highlights their potential as effective natural ingredients in formulations aimed at protecting the skin from oxidative damage, enhancing cellular defence mechanisms and preventing signs of premature aging [26, 27]. Notably, the CCA observed, particularly in the 3:1 mixture, is of interest for mitigating metal‐catalysed oxidative reactions, which are known to exacerbate skin damage and instability in formulations [28].

In conclusion, the combination of LA and PM extracts demonstrates significant antioxidant potential through synergistic interactions and a rich phytochemical composition. These findings highlight their value as innovative, sustainable and effective ingredients for cosmetic applications, with the added benefit of reducing the environmental impact of product development. Further studies on the precise mechanisms of synergy and long‐term stability in formulations are warranted to optimize their application in the dermo‐cosmetic industry.

2.3. Enzymatic Inhibitory Activity

The enzymatic inhibitory activity of LA, PM, and their mixtures in different ratios (3:1, 1:1 and 1:3) was evaluated for tyrosinase, collagenase, hyaluronidase and elastase inhibition (Table 3). In the tyrosinase assay, PM was the most active single extract (EC₅₀ = 9.05 µg/mL), outperforming the positive control, kojic acid (EC₅₀ = 15.3 µg/mL). Among the mixtures, LA + PM (1:3) showed the highest synergistic effect (SE = 0.72) with a notable EC₅₀ value of 11.1 µg/mL, just followed by LA + PM (1:1) (EC₅₀ = 17.0 µg/mL; SE = 0.78).

TABLE 3.

Enzyme inhibitory activity (EC₅₀, µg/mL) and synergistic effect (SE) of Limonium algarvense (LA, sea lavender), Polygonum maritimum (PM, sea knotgrass) and their mixtures in different ratios, over the tested enzymes.

Extract	Tyrosinase		Collagenase		Hyaluronidase		Elastase
	EC50	SE	EC50	SE	EC50	SE	EC50	SE
LA	34.4 ± 1.4d	—	31.2 ± 4.0d	—	55.1 ± 5.3cd	—	46.4 ± 2.0e	—
LA + PM, 3:1	31.0 ± 1.7c	1.1	30.5 ± 4.3d	1.1	50.6 ± 5.2c	0.97	45.0 ± 3.0de	1.0
LA + PM, 1:1	17.0 ± 2.8b	0.78	25.8 ± 3.1c	1.1	58.0 ± 2.1d	1.2	38.7 ± 3.6cd	0.87
LA + PM, 1:3	11.1 ± 1.6ab	0.72	19.1 ± 2.0b	0.96	44.7 ± 2.2bc	0.97	33.0 ± 5.0bc	0.76
PM	9.05 ± 1.83a	—	16.1 ± 1.1ab	—	42.9 ± 2.0b	—	42.5 ± 3.1d	—
Kojic acid *	15.3 ± 1.3b
Epigallocatechin gallate *			8.71 ± 2.95a				7.04 ± 1.05a
Tannic acid *					7.73 ± 1.06a

Note: EC₅₀ values represent the mean ± standard error of the mean (SEM) of at least three duplicated experiments. Values followed by different letters (a–f) are significantly different (p < 0.05) (Tukey HSD test). Values followed by different letters are significantly different (p < 0.05) (Tukey HSD test). SE represents antagonistic (SE > 1), additive (SE = 1) or synergistic (SE < 1) effects.

^*Positive controls.

For collagenase inhibition, PM again demonstrated superior activity among the extracts (EC₅₀ = 16.1 µg/mL). The LA + PM (1:3) mixture exhibited synergistic effects (SE = 0.96), with EC₅₀ values of 19.1 µg/mL. In the hyaluronidase assay, PM retained its position as the most active extract (EC₅₀ = 42.9 µg/mL). Among the mixtures, the LA + PM (1:3) ratio exhibited the strongest synergistic effect (SE = 0.97), with an EC₅₀ (44.7 µg/mL). For elastase inhibition, PM had a notable EC₅₀ value of 42.5 µg/mL, while the LA + PM (1:3) and LA + PM (1:1) mixtures showed strongest synergistic interaction (SE = 0.76 and 0.87, respectively), presenting EC₅₀ values lower than the solo extracts (33.0 and 38.7 µg/mL, respectively). Overall, PM was consistently the most active individual extract across all assays, while the LA + PM (1:3) mixture showed the highest degree of synergistic effects.

Both species have previously been reported to exhibit enzymatic inhibitory properties. For instance, the acetone extracts from the aerial parts of PM have demonstrated strong tyrosinase inhibition, as well as the suppression of melanin production in melanoma cells [14]. Similarly, the seeds of LA have been shown to inhibit elastase, hyaluronidase, tyrosinase and collagenase, with EC₅₀ values below 20 µg/mL [16]. Moreover, the enzymatic inhibition profiles of LA and PM reflect the diverse bioactivities of their phenolic compounds, such as flavonoids (e.g., myricetin, catechins) and phenolic acids, which are known to interact with enzymes like tyrosinase, hyaluronidase, elastase and collagenase [29, 30]. For instance, catechins and their derivatives, which are abundant in PM, have been shown to inhibit tyrosinase by binding to its active site and reducing melanin formation [31]. Similarly, quercitrin and myricetin derivatives identified in LA have been associated with collagenase and hyaluronidase inhibition, contributing to the preservation of the skin's structural integrity and hydration [32].

In addition to the available literature on these extracts, the present results are consistent with established structure–activity relationships for the enzymes studied. For tyrosinase, inhibition is strongly associated with the ability of phenolic compounds to form coordination complexes with the copper ion located in the enzyme's active site [32]. Flavonoids readily form such complexes, particularly those containing catecholic moieties, which enhance their affinity for copper [32]. Several annotated metabolites in the extracts—such as catechin, epigallocatechin, epigallocatechin gallate (EGCG) and myricetin derivatives—possess these structural features, supporting their observed inhibitory effects. Similarly, collagenase activity depends on both its catalytic residues and a Zn(II)‐binding site [33]. Many of the identified metabolites contain carboxylate groups or related functionalities capable of chelating zinc, a mechanism widely associated with collagenase inhibition [33]. In contrast, hyaluronidase inhibition is typically enhanced by molecules with multiple hydroxyl substitutions, which can form electrostatic interactions with residues such as Asp129, Glu131, Gln288 and Asp292 in the enzyme's active pocket [32]. Several metabolites reported here fulfil these structural criteria. Regarding elastase, its catalytic function relies on a classical His57–Ser195–Asp102 triad, which can interact favourably with flavonoid‐type structures [32]. Compounds such as myricetin, myricetrin, and quercitrin—detected in the extracts—exhibit these characteristic interactions, supporting their relevance to elastase inhibition.

The unique phytochemical composition of each species likely drives their distinct inhibitory profiles, while the synergistic effects in mixtures suggest potential complementary interactions. These findings align with previous studies on these species, reinforcing their utility as sources of enzyme inhibitors for cosmetic applications. Thus, the enzymatic inhibition profiles of LA, PM and their mixtures highlight their potential as natural alternatives to synthetic inhibitors commonly used in cosmetics. Natural inhibitors offer advantages such as lower toxicity and better consumer acceptance, aligning with the growing demand for eco‐friendly and sustainable ingredients. The observed synergistic effects in specific mixtures suggest complex interactions between bioactive compounds, possibly altering enzyme‐binding dynamics or stabilizing inhibitory molecules, as proposed in other plant‐based systems [34]. Such effects could be exploited by designing formulations targeting specific enzymatic pathways.

The inhibition of collagenase and hyaluronidase is particularly relevant for maintaining skin integrity and hydration, as these enzymes degrade collagen and glycosaminoglycans, accelerating the aging process. The bioactivity observed aligns with the presence of phenolic compounds like flavonoids and tannins, which have been reported as potent inhibitors of these enzymes in other halophytes [30, 32]. These findings underscore the potential for LA and PM extracts to serve as multifunctional anti‐aging agents. Tyrosinase inhibition, a key mechanism for managing hyperpigmentation and uneven skin tone, further emphasizes the applicability of these extracts in skin‐whitening formulations. However, unlike synthetic inhibitors such as kojic acid, natural compounds may offer improved stability and reduced adverse effects, making them safer for long‐term use [35]. Interestingly, elastase inhibition was limited to PM, reflecting the distinct phytochemical profiles of the two species. This highlights the potential for PM to address loss of skin elasticity, an area where further optimization could enhance its efficacy.

Overall, the enzyme inhibitory profiles of LA and PM, together with their sustainable cultivation potential, support their use as bioactive halophyte sources for topical cosmetic applications. The observed synergistic effects in selected mixtures suggest complementary interactions between phytochemicals, offering a promising foundation for the development of multi‐target formulations targeting skin‐aging mechanisms. Future work should focus on elucidating the molecular mechanisms underlying these effects, as well as on optimizing formulation stability, bioavailability, and efficacy for potential dermo‐cosmetic use.

2.4. Antimicrobial Activity

The minimal inhibitory concentrations (MICs) of LA, PM and their mixtures in different ratios were assessed against four Candida yeast strains (Table 4).

TABLE 4.

Minimal inhibitory concentrations (MIC; µg/mL) of Limonium algarvense (LA, sea lavender), Polygonum maritimum (PM, sea knotgrass) and their mixtures in different ratios, against yeast isolates.

Samples/strain	Candida tropicalis (YEPGA 6184)		Candida albicans (YEPGA 6379)	Candida parapsilosis (YEPGA 6551)		C. albicans (YEPGA 6183)
	MIC	SE	MIC	MIC	SE	MIC
LA	158.74 (100–200)	—	—	—	—	—
LA + PM, 3:1	—	> 1.26	—	158.74 (100–200)	< 0.94	—
LA + PM, 1:1	—	> 1.26	—	158.74 (100–200)	< 0.89	—
LA + PM, 1:3	—	> 1.26	—	—	—	—
PM	158.74 (100–200)	—	—	158.74 (100–200)	—	—
Fluconazole a	2		1	4		2

Note: MIC values are reported as the geometric means of three independent replicates (n = 3). MIC range concentrations are reported within the brackets.

^a Positive control.

For Candida tropicalis, both LA and PM exhibited comparable activity with MIC values of 158.74 µg/mL. However, none of the mixtures showed improved activity, displaying an antagonistic effect (SE > 1.26). For Candida parapsilosis, both PM and LA + PM mixtures (1:1 and 3:1) demonstrated activity, with MIC values of 158.74 µg/mL, with SE of < 0.89 and < 0.94, respectively. In turn, neither the individual extracts nor the mixtures showed inhibitory activity against Candida albicans (YEPGA 6379 and YEPGA 6183).

The antibacterial activity of LA, PM, and their mixtures were tested against various bacterial strains (Table 5). For Escherichia coli (ATCC 10536), LA and PM demonstrated similar MICs (158.74 µg/mL), as well as the LA + PM (1:3) and (3:1) combinations, showing to have additive effects (SE = 1). Towards the E. coli PeruMycA 2 strain, PM demonstrated a MIC of 39.68 µg/mL, while the mixtures showed synergistic effects, particularly LA + PM (3:1) that showed the strongest activity (MIC = 19.84 µg/mL) with a synergistic SE of 0.32.

TABLE 5.

Minimal inhibitory concentrations (MIC; µg/mL) of Limonium algarvense (LA, sea lavender), Polygonum maritimum (PM, sea knotgrass) and their mixtures in different ratios, against bacterial isolates.

Samples/bacterial strain	Escherichia coli (ATCC 10536)		E. coli (PeruMycA 2)		E. coli (PeruMycA 3)	Bacillus cereus (ATCC 12826)		Pseudomonas aeruginosa (ATCC 15442)		Bacillus subtilis (PeruMycA 6)	Salmonella typhi (PeruMycA 7)	Staphylococcus aureus (ATCC 6538)
	MIC	SE	MIC	SE	MIC	MIC	SE	MIC	SE	MIC	MIC	MIC
LA	158.74 (100–200)	—	125.99 (100–200)	—	—	—	—	79.37 (50–100)	—	—	—	158.74 (100–200)
LA + PM, 3:1	158.74 (100–200)	1	19.84 (12.5–25)	0.32	—	125.99 (100–200)	< 0.63	158.74 (100–200)	< 0.76	—	—	—
LA + PM, 1:1	—	—	79.37 (50–100)	0.96	—	158.74 (100–<200)	< 0.79	79.37 (50–100)	< 0.57	—	—	—
LA + PM, 1:3	158.74 (100–200)	1	39.68 (25–50)	0.38	—	—	—	79.37 (50–100)	< 0.72	—	—	—
PM	158.74 (100–200)	—	39.68 (25–50)	—	—		—	—	—	—	—	—
Ciprofloxacin a	31.49 (25–50)		9.92 (6.25–122.5)		79.37 (50–100)	125.99 (100–200)		125.99 (100–200)		125.99 (100–200)	79.37 (50–100)	—

Note: MIC values are reported as the geometric means of three independent replicates (n = 3). MIC range concentrations are reported within the brackets.

^a Positive control.

Against Bacillus cereus, none of the solo extracts showed activity, but the LA + PM (1:1) and LA + PM (3:1) mixtures achieved a MIC of 158.74 and 125.99 µg/mL, with SE of < 0.72 and < 0.63, respectively. For Pseudomonas aeruginosa, LA individual extract demonstrated a MIC of 79.37 µg/mL, while the mixtures LA + PM 1:3, 1:1 and 3:1 exhibited moderate activities (MIC = 79.37–158.74 µg/mL), with a synergistic effect lower than 0.76. For Staphylococcus aureus, LA exhibited moderate activity (MIC = 158.74 µg/mL), whereas PM and the mixtures were not effective. Finally, none of the extracts were active towards E. coli PeruMycA 3, Bacillus subtilis and Salmonella typhi. PM generally exhibited stronger activity against both yeast and bacterial strains compared to LA. Synergistic effects were observed in specific mixtures, particularly LA + PM (3:1) and (1:1).

When compared to benchmark antibiotics, the extracts were markedly less potent. For example, ceftriaxone—a first‐line medication for E. coli infections—has an MIC of 0.125 mg/mL [36]. Relative to this reference, the extracts exhibited approximately 1300‐fold lower potency against E. coli ATCC 10436. For E. coli PeruMyc2, the potency varied substantially among extracts, with the LA + PM (3:1) mixture performing best, yet still around 160‐fold less potent than ceftriaxone. For B. cereus, both mixtures (LA + PM 3:1 and 1:1) showed activity that was approximately 120‐ and 160‐fold less potent, respectively, when compared with vancomycin (MIC = 1 µg/mL) [37]. Similarly, ceftazidime—a first‐line treatment for P. aeruginosa—has an MIC of 2 µg/mL, corresponding to a potency around 80‐fold higher than LA + PM (3:1) and 40‐fold higher than the other active extracts. In the case of S. aureus, only the LA extract showed measurable activity, but it remained approximately 850‐fold less potent than oxacillin (MIC = 0.1235–0.25 µg/mL) [37]. Although the extracts displayed considerably lower potency than isolated antibiotics or pure phenolic standards, this outcome is expected for complex natural matrices. Extracts contain dozens of metabolites occurring at relatively low concentrations and in the presence of other constituents that may dilute, compete with, or modulate antimicrobial effects, thereby reducing apparent activity when compared with purified reference compounds. Importantly, lower potency does not diminish their potential relevance. Natural extracts often present more favourable safety profiles and can exert multimodal biological effects—such as anti‐inflammatory, antioxidant or anti‐biofilm actions—that are not captured by MIC measurements. These properties may be particularly advantageous for applications in topical formulations or natural preservative systems, where extreme antimicrobial potency is neither required nor always desirable. Furthermore, synergistic interactions within the metabolite mixture may contribute to functional effects beyond direct growth inhibition, suggesting that their utility may extend beyond conventional antimicrobial endpoints.

Both species have been previously reported to exhibit antimicrobial properties. For instance, PM has been described as inhibiting the growth of E. coli, S. aureus, P. aeruginosa and C. albicans [14], similar to other Polygonum species, such as P. hydropiper, P. spectabile, P. chinense and P. aviculare [38, 39, 40, 41]. Moreover, although this is the first time LA has been reported to possess antimicrobial properties, existing literature on other Limonium species corroborates the observed bioactivity, namely, in L. avei, L. brasiliense and L. lopadusanum [42, 43, 44, 45], highlighting their potential as natural preservatives in cosmetic formulations.

Moreover, the antimicrobial activity of LA and PM can be attributed to their phenolic constituents, such as flavonoids and tannins, which disrupt microbial membranes and inhibit enzymatic processes essential for microbial survival [46]. For instance, catechins derivatives, highly present in PM, have been reported to exhibit strong antimicrobial properties by interfering with bacterial cell walls and membranes [47]. Moreover, sinapic acid derivatives, such as synapoylhexoside, identified in PM have also been described with antibacterial properties against Minimum inhibitory concentrations of the hydrolysed extract against B. subtilis, E. coli, L. monocytogenes, Pseudomonas fluorescens, and S. aureus [48]. Quercetin, as well as its derivatives, present in LA further supports their antimicrobial potential, as these compounds are known to inhibit microbial growth through ROS generation and membrane destabilization [49].

Therefore, the antimicrobial activity of LA, PM and their mixtures against various bacterial strains highlights their potential as natural preservatives and skin microbiome modulators in cosmetic formulations. The bioactivity observed, particularly the synergistic effects in certain mixtures, underscores the complexity of interactions between bioactive compounds. Such combinations may enhance membrane disruption, interfere with microbial enzymatic pathways, or improve compound stability, as suggested in similar studies on common antimicrobial drugs [50]. The greater activity against Gram‐positive bacteria compared to Gram‐negative strains is consistent with the structural differences in bacterial cell walls. Gram‐positive bacteria possess a simpler cell wall structure, making them more susceptible to phenolic compounds and flavonoids, which are abundant in these extracts. This selectivity is advantageous in cosmetic applications, where maintaining a balance in the skin microbiome is crucial for preventing opportunistic infections without disrupting commensal microbial populations [9]. The limited activity against yeast strains, particularly C. albicans, highlights the specificity of these extracts toward bacterial pathogens. However, the activity against C. tropicalis and C. parapsilosis suggests potential applications in treating or preventing fungal infections associated with compromised skin barriers, such as those caused by excessive dryness or abrasion.

From a formulation perspective, the ability of these extracts to inhibit microbial growth at relatively low concentrations makes them promising natural preservatives. Synthetic preservatives, such as parabens, have faced increasing scrutiny due to their potential health risks, driving demand for safer, natural alternatives [9]. The antimicrobial activity of LA and PM, particularly in mixtures, provides an opportunity to develop multifunctional cosmetic products that combine antimicrobial efficacy with other skin‐protective benefits, such as antioxidant and enzyme‐inhibitory properties.

In conclusion, the antimicrobial properties of LA and PM, particularly in synergistic combinations, support their potential as natural alternatives to synthetic antimicrobial agents in topical cosmetic formulations. Their selective activity against pathogenic microorganisms, combined with low cytotoxicity and sustainable sourcing, highlights their relevance for the development of multi‐functional dermo‐cosmetic products. Future research should focus on evaluating their chemical stability, compatibility with formulation excipients, and long‐term antimicrobial efficacy under realistic application conditions.

2.5. Cytotoxicity

The cell viability of LA, PM and their mixtures in various ratios (3:1, 1:1, 1:3) was evaluated in three mammalian cell lines: B16 A45 (melanoma), RAW264.7 (macrophages) and HEK293 (human embryonic kidney cells) (Figure 1). For B16 A45 and HEK293 cells, all extracts exhibited high cell viability, with values ranging from 96% to 127%. In RAW264.7 cells, LA and all mixture ratios have maintained cell viability above 81%, except for PM alone, which demonstrated lower cell viability (68.2%), indicating mild cytotoxicity. These results suggest that while LA is non‐toxic, PM shows some cytotoxicity, particularly toward RAW264.7 cells, but the mixtures effectively mitigate PM's cytotoxicity across all tested cell lines.

FIGURE 1.

Cytotoxicity assessment of Limonium algarvense and Polygonum maritimum extracts, and their mixtures, against murine melanoma (B16 4A5), murine monocyte/macrophage (RAW264.7), and human embryonic kidney (HEK293) cell lines. Results are expressed as cellular viability (%) at 100 µg/mL after 72 h of incubation. Values represent the mean ± standard error of the mean (SEM) of six independent experiments (n = 6). For each cell line, columns marked with different letters (a–c) denote significant differences by Tukey's multiple comparisons tests (p < 0.05).

The cytotoxicity findings align with previous studies on LA and PM, which have consistently been reported as non‐cytotoxic. Extracts from LA, including those obtained from flowers, leaves and seeds, have shown high cell viability (> 90%) across various mammalian cell lines, including murine melanoma (B16 4A5) and human embryonic kidney (HEK293) cells, up to 100 µg/mL [15, 16]. Similarly, extracts of PM have been demonstrated to exhibit low toxicity in studies focusing on their anti‐inflammatory and antioxidant potential, with cell viabilities exceeding 80% in RAW264.7 macrophages and other mammalian cell lines [14].

In our study, all extracts and mixtures likewise maintained high viability, indicating low cytotoxicity under the tested in vitro conditions. These results provide preliminary support for the relevance of these salt‐tolerant plants in dermo‐cosmetic research. To avoid overinterpretation, we emphasise that these findings represent only initial indications of safety and do not substitute for dermatological evaluations. Interestingly, the ability of the mixtures to attenuate the mild cytotoxicity observed for PM suggests that combining the extracts may improve not only their bioactivity profile but also their in vitro tolerability. This trend is consistent with the growing interest in synergistic interactions within botanical formulations, where the combined action of phytochemicals can optimise functional performance while minimising undesirable effects [51, 52].

3. Conclusions

This study demonstrates the potential of LA and PM extracts, particularly the 1:3 ratio mixture, as safe and multifunctional candidates for topical cosmetic applications. The synergistic effects observed in this combination enhanced antioxidant capacity, inhibition of skin‐aging‐related enzymes, and antimicrobial activity, while maintaining minimal cytotoxicity. These bioactivities, linked to phenolic compounds such as catechin, myricetin, and quercetin derivatives, provide a strong basis for developing innovative multi‐target dermo‐cosmetic formulations. Beyond their functional benefits, the use of halophyte species offers significant sustainability advantages: these plants thrive in saline soils and can be irrigated with brackish or seawater, reducing freshwater consumption and enabling cultivation in marginal lands. This approach promotes biodiversity, valorises underutilized plant resources, and aligns with eco‐friendly industrial practices, opening opportunities for the cosmetic industry to integrate sustainable raw materials into high‐performance formulations. Future research should prioritize formulation optimization to ensure stability, sensory properties, and efficacy in real cosmetic products, as well as advanced in vitro and ex vivo models to confirm safety and performance under conditions that mimic human skin. Clinical studies with volunteers will also be essential to validate the observed synergistic effects and support regulatory compliance, bridging the gap between laboratory findings and industrial implementation.

4. Experimental

4.1. Chemicals

2,2‐Diphenyl‐1‐picrylhydrazyl (DPPH), 2,2′‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulfonic acid) (ABTS), gallic acid, sodium acetate, pyrocatechol violet (PV), copper sulphate pentahydrate (CuSO₄∙5H₂O), ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA), sodium hydroxide (NaOH), catechin hydrate, trichloroacetic acid (TCA), sodium phosphate monobasic (KH₂PO₄), ascorbic acid, and iron(III) chloride hexahydrate (FeCl₃∙6H₂O), 2‐deoxy‐d‐ribose, hydrogen peroxide (H₂O₂), lipase (EC 3.1.1.3) from porcine pancreas Type II 100–400 U/mg, tyrosinase (EC 1.14.18.1) from mushroom ≥ 1000 unit/mg, elastase (EC 3.4.21.35) from porcine pancreas Type I, ≥ 4.0 units/mg protein, 4‐nitrophenyl dodecanoate (PNP), orlistat, l‐tyrosine, arbutin, potassium phosphate dibasic, potassium phosphate monobasic and N‐(methoxysuccinyl)‐Ala‐Ala‐Pro‐Val‐4‐Nitroanilide (MAAPVN) were purchased from Sigma‐Aldrich (Germany). Any other reagent used was purchased to VWR International (Leuven, Belgium).

4.2. Plant Material and Growth Conditions

LA and PM were cultivated in controlled greenhouse conditions, using a substrate mix of peat and perlite (3:1, v/v), as detailed previously [14, 15, 19]. After 14 weeks, flowers of LA and leaves of PM were harvested, frozen, lyophilized, powdered and stored at −20°C.

The plants originated from seed banks derived from previously identified wild populations, taxonomically identified by Prof. Ana Caperta (LA) and Prof. Manuela David (PM), according to standard diagnostic keys. Voucher specimens of both species are deposited in the XtremeBio Herbarium (CCMAR), with accession XBH01.2 for LA and XBH22.1 for PM.

4.3. Extraction

The mixtures of the two species were prepared by mixing plant powders in ratios of 3:1, 1:1 and 1:3 prior to extraction. Then, extractions were performed by mixing 1 g of plant powder with 40 mL of distilled water (1:40 w/v) and extracting it in an ultrasonic bath for 30 min. The resulting extracts were filtered, freeze‐dried, resuspended to a stock solution of 10 mg/mL, and stored at −20°C.

4.4. Phytochemical Composition

4.4.1. Sample Preparation

All extracts were vortex‐homogenized and further diluted in distilled water to a final concentration of 2 mg/mL and filtered with a 0.2 nm syringe adapted filter. A quality control (QC) sample was prepared by mixing an equal volume of each extract.

4.4.2. Chemical Profiling of the Extracts by UHPLC–HRMSⁿ Analyses

Prior to injection, all extracts were vortex‐homogenized, diluted in LC–MS‐grade water to a final concentration of 2 mg/mL and filtered with a 0.2 nm syringe adapted filter. A QC sample was prepared by mixing an equal volume of each extract. The chemical composition of the extracts was determined by a UHPLC–HRMSⁿ analysis. The chromatography was performed on a Thermo Scientific ultimate 3000 UHPLC with a Thermo Scientific Accucore RP‐18 (2.1 × 100 mm, 2.6 µm) column. The binary mobile phase consisted of water (A) and acetonitrile (B), both with 0.1% of formic acid. The gradient (in v/v %) started with 100% of A for 2 min. Subsequently, B was linearly increased up to 30% for 13 min. Then, B was linearly increased for a second time, up to 100% and for 16 min. The 100% concentration of B was kept for 4 min. Right after, the mobile phase was brought back to a 100% of A for 1 min and maintained for 4 min at it. The flowrate was 0.3 mL/min and the injection volume was 5 µL. Mass analysis was performed on an Orbitrap Elite (Thermo Scientific) mass spectrometer with a Heated ElectroSpray Ionization source (HESI‐II). HR‐MSn data were acquired under the following ionization parameters: spray voltages, 3.7 kV (positive polarity) and 4.0 kV (negative polarity); sheath gas, 35 arbitray units; auxiliary gas, 10 arbitrary units; heater temperature, 300°C; capillary temperature, 350°C; S‐Lenses RF level, 64.9%. The data was analysed in data dependent mode, using dynamic exclusion in both negative and positive polarities, and had a mass range from 100 to 1000 m/z. LC–MS profiles were analysed using Xcalibur 4.1. Thermo .raw data files were converted to .mzXML format in centroid mode using Proteowizard [53] and imported in MZMine 3 for feature finding, alignment and extraction. A blank was used for background feature removal and the final feature list was exported as ‘.mgf’ and ‘.csv’ files. These output files and a metadata table were used to create a Feature Based Molecular Network (FBMN) [54] using the online workflow [55] from the Global Natural Products Social (GNPS) ecosystem (https://gnps.ucsd.edu/ accessed on October 20, 2024) [54, 55]. The data was filtered by removing all MS/MS fragment ions within ±17 Da of the precursor m/z. MS/MS spectra were window filtered by choosing only the top six fragment ions in the ±50 Da window throughout the spectrum. The precursor ion mass tolerance was set to 0.05 Da and the MS/MS fragment ion tolerance to 0.05 Da. A molecular network was then created where edges were filtered to have a cosine score above 0.70 and more than four matched peaks. Further, edges between two nodes were kept in the network if and only if each of the nodes appeared in each other's respective top 10 most similar nodes. Finally, the maximum size of a molecular family was set to 100, and the lowest scoring edges were removed from molecular families until the molecular family size was below this threshold. The spectra in the network were then searched against GNPS spectral libraries [54, 55]. The library spectra were filtered in the same manner as the input data. All matches kept between network spectra and library spectra were required to have a cosine score above 0.7 and at least 4 matched peaks. Metabolite annotations obtained through GNPS library matching were manually verified and cross‐referenced with published literature to ensure accurate final annotation.

4.5. In Vitro Antioxidant Activity

4.5.1. Radical Scavenging Activity (RSA) on DPPH Radical

Extracts (22 µL) were mixed with 200 µL of a DPPH solution (120 µM in ethanol) and incubated at room temperature for 30 min. The absorbance was measured at 517 nm, and the results were expressed as a percentage relative to the negative control, which contained distilled water, and as half‐maximal effective concentration (EC₅₀) values (µg/mL) [56]. Gallic acid was used as the positive control.

4.5.2. RSA on ABTS Radical

Extracts (10 µL) were mixed with 190 µL of the ABTS^•+ solution and incubated in the dark for 6 min at room temperature (RT). The absorbance was measured at 734 nm, and the results were expressed as a percentage relative to the negative control and as EC₅₀ values (µg/mL) [56]. Gallic acid was used as the standard.

4.5.3. RSA on Superoxide (O₂ ⁻) Radicals

Extracts (100 µL) were mixed with 50 µL of Tris‐HCl buffer (16 mM, pH 8.0), 50 µL of nitrotetrazolium blue (NBT) (0.3 mM, diluted in the previous buffer), 50 µL of nicotinamide adenine dinucleotide (NADH) (0.936 mM, in 5 mM NaOH, pH 11), and 50 µL of phenazine methosulfate (PMS) (12 mM). It was incubated for 5 min at RT, and then the absorbance was read at 560 nm [11]. Catechin was used as the positive control. The results were expressed as a percentage relative to the negative control and as EC₅₀ values (µg/mL).

4.5.4. RSA on Hydroxyl (HO^•) Radical

Extracts (200 µL) were mixed with 40 µL of 2 deoxy‐d‐ribose (28 mM), 40 µL of H₂O₂ (1.0 mM), 40 µL of ascorbic acid (4.0 mM), 80 µL of EDTA/FeCl₃ (1:1, v/v) (EDTA: 1.04 mM; FeCl₃: 200 µM). All solutions were diluted in KH₂PO₄–KOH buffer (20 mM, pH 7.4). The mixture was incubated at 37°C for 1 h, and then 400 µL of thiobarbituric acid (TBA; 1%), and 400 µL of trichloroacetic acid (TCA; 2.8%) were added and taken to a boiling water bath (90°C–100°C) for 20 min. Once cooled at RT, 300 µL aliquots were taken and read at 532 nm [11]. Catechin was used as the standard. The results were expressed as a percentage relative to the negative control and as EC₅₀ values (µg/mL).

4.5.5. CCA

Extracts (30 µL) were mixed with 200 µL of sodium acetate buffer (50 mM; pH 6), 100 µL of CuSO₄ (50 mg/mL, in water), 6 µL of PV (4 mM, in water). The absorbance was measured at 632 nm [56], and the results were expressed as a percentage relative to the negative control and as EC₅₀ values (µg/mL). EDTA was used as the standard.

4.6. Enzymatic Inhibition

4.6.1. Tyrosinase

Extracts (10 µL) were mixed with an enzyme solution (333 U/mL) and incubated at RT for 10 min. Subsequently, 20 µL of 1.5 M l‐tyrosine and 110 µL of 0.1 M sodium phosphate buffer were added, followed by an additional incubation for 20 min at 25°C. The absorbance was measured at 490 nm [16]. Kojic acid was used as the standard, and the results were calculated as percentage inhibition relative to the negative control and as EC₅₀ values (µg/mL).

4.6.2. Elastase

Extracts (50 µL) were mixed with 1 µg/mL elastase enzyme (3.33 mg/mL) and incubated at 37°C for 15 min. Subsequently, 1.6 mM N‐succinyl‐Ala‐Ala‐Ala‐p‐nitroanilide (AAAPVN) in 0.2 mM Tris‐HCl buffer was added, followed by an additional incubation for 20 min. The absorbance was then measured at 410 nm [16]. EGCG was used as the positive control, and the results were expressed as percentage inhibition relative to the negative control and as EC₅₀ values (µg/mL).

4.6.3. Hyaluronidase

Extracts (50 µL) were combined with a hyaluronidase enzyme solution (0.02 M phosphate buffer containing NaCl and bovine serum albumin) and 0.1 M acetate buffer (pH 3.5), followed by incubation for 20 min. Subsequently, 10 µL of hyaluronic acid was added, and the mixture was incubated for an additional 20 min at 37°C. The absorbance was then measured at 600 nm [16]. Tannic acid was used as the standard, and the results were expressed as percentage inhibition relative to the negative control and as EC₅₀ values (µg/mL).

4.6.4. Collagenase

Extracts (50 µL) were combined with collagenase from Clostridium histolyticum (0.8 unit/mL in 50 mM tricine buffer) and incubated for 15 min. Subsequently, 150 µL of tricine buffer and 0.8 mM N‐[3‐(2‐furyl)acryloyl]‐Leu‐Gly‐Pro‐Ala (FALGPA) were added, followed by an additional incubation for 20 min. The absorbance was then measured at 340 nm [16]. EGCG was used as the standard, and the results were expressed as percentage inhibition relative to the negative control and as EC₅₀ values (µg/mL).

4.7. Antimicrobial Susceptibility Test

The in vitro antimicrobial activity of the extracts (LA; LA + PM, 1:3; LA + PM, 1:1; LA + PM, 3:1; PM) was assessed against six bacterial strains, namely E. coli (ATCC 10536), E. coli (PeryMycA 2), E. coli (PeruMyc 3), B. cereus (ATCC 12826), P. aeruginosa (ATCC 15442), B. subtilis (PeruMycA 6), S. typhi (clinical isolate), S. aureus (ATCC 6538) and four yeasts, namely, C. tropicalis (YEPGA 6184), C. albicans (YEPGA 6379), C. parapsilosis (YEPGA 6551) and C. albicans (YEPGA 6183). The C. parapsilosis (ATCC 22019) and Candida krusei (ATCC 6258) strains were used as QCs in the broth dilution antifungal test [57]. Multiple strains of the same bacterial or yeast species were included because they exhibit distinct phenotypic and antimicrobial susceptibility profiles, including differences in virulence, membrane composition and resistance mechanisms. Testing several strains per species ensures a broader and more robust assessment of antimicrobial activity and prevents strain‐specific artefacts.

4.7.1. Antibacterial Activity

The MICs of the samples were determined by a microdilution method according to the Clinical and Laboratory Standards Institute, M07‐A10 [58]. MICs were evaluated using extract concentrations in the range of 1.562–200 µg/mL, derived from serial twofold dilutions in Mueller–Hinton broth (MHB). Ciprofloxacin was used in the range of 0.12–1251 µg/mL as a control antibacterial agent [59]. For the preparation of the bacterial suspensions (inocula), three to five colonies of the bacterial strains used for the test were chosen from 24 h cultures on tryptic soy agar plates (TSA) and pre‐grown overnight in MHB to reach a cell density of approximately 1–2 × 10⁸ CFU/mL in each tube. This was confirmed with the plating of serial dilutions of the inoculum suspensions on Mueller–Hinton agar (MHA). The setup included bacterial growth controls in wells containing 10 µL of the test inoculum and negative controls without a bacterial inoculum. MIC end points were determined after 18–20 h incubation at room temperature (35°C).

4.7.2. Antifungal Activity

MICs of the marine organism samples against the yeasts were performed according to the CLSI M27‐A4 [57]. A RPMI‐1640 medium (Sigma) with l‐glutamine and without sodium bicarbonate, supplemented with 2% glucose w/v, buffered with 0.165 mol/L morpholinepropanesulphonic acid, pH 7.0, was used throughout the study. The inoculum suspensions were prepared in a saline solution containing 0.05% Tween 80 from 7‐day‐old cultures growing on Sabouraud dextrose agar (SDA) at 25°C and adjusted spectrophotometrically to optical densities that ranged from 0.09 to 0.11 (MacFarland standard). Yeast inoculum suspensions were diluted to a ratio of 1:50 in RPMI‐1640 to obtain twice the inoculum size, ranging from 0.2 to 0.4 × 104‐5 CFU/mL. This was further confirmed by plating the serial dilutions of the inoculum suspensions on SDA. Sample extracts had an MIC range of 1.56–200 µg/mL, fluconazole had an MIC range of 0.03–16 µg/mL. MIC end points (µg/mL) were determined after 24 h (for yeasts). For the sample extracts, the MIC end points were defined as the lowest concentration that showed total growth inhibition. MIC end points for fluconazole was defined as the lowest concentration that inhibited 50% of the growth when compared with the growth control [57]. Geometric means and MIC ranges were determined from the three biological replicates to allow for comparisons between the activities of the samples.

4.8. Cytotoxicity Evaluation

Murine melanoma cells (B16 4A5, RRID: CVCL_4612) were purchased from Sigma‐Aldrich (St. Louis, MO, USA). Murine monocyte/macrophage cells (RAW264.7, RRID: CVCL_0493) were obtained from Quimigen (Lisbon, Portugal). Human embryonic kidney cells (HEK293, RRID: CVCL_0045) were kindly provided by the research group of Prof. Dr. Dina Simes (CCMAR, University of Algarve, Portugal). All cell lines were grown as previously detailed [14, 60]. Briefly, after seeding and reattachment, the samples were tested at 100 µg/mL for 72 h of incubation. Then, the cell viability was assessed by the 3‐(4, 5‐dimethylthiazol‐2‐yl)‐2, 5‐diphenyltetrazolium bromide (MTT) colorimetric method [60]. The absorbance was recorded at 595 nm, and the results were calculated as the percentage of cell viability relative to a control sample containing distilled water.

4.9. Synergistic Effect Determination

The synergistic effect (SE) of the mixtures (LA + PM 3:1, 1:1 and 1:3) was evaluated by comparing the observed EC₅₀ of each mixture with the theoretical EC₅₀ expected under simple additivity. For each bioassay, the theoretical EC₅₀ (EC_50,teor) for a given mixture was calculated as the weighted arithmetic mean of the EC₅₀ values of the individual extracts, according to their proportions in the mixture:

$$\mathrm{E}{\mathrm{C}}_{50,\mathrm{teor}}=p_{\mathrm{LA}}\times \mathrm{E}{\mathrm{C}}_{50,\mathrm{LA}}+p_{\mathrm{PM}}\times \mathrm{E}{\mathrm{C}}_{50,\mathrm{PM}}$$

where p_LA and p_PM are the mass fractions of LA and PM in the mixture, respectively, and EC_{50_LA} and EC_{50_PM} are the EC₅₀ values of each extract tested alone.

The synergistic effect (SE) was then calculated as the ratio between the observed EC₅₀ of the mixture (EC_50,obs) and the theoretical EC₅₀:

$$\mathrm{SE}=\frac{\mathrm{E}{\mathrm{C}}_{50,\mathrm{obs}}}{\mathrm{E}{\mathrm{C}}_{50,\mathrm{teor}}}$$

Values of SE < 1 were interpreted as synergistic effects (the mixture requiring a lower EC₅₀ than expected), SE ≈ 1 as additive behaviour, and SE > 1 as antagonistic effects.

4.10. Statistical Analysis

The experiments were performed in three independent replicates (n = 3). EC₅₀ values were calculated as the mean of these three independent experiments, using sigmoidal fitting of the data in GraphPad Prism v8.0.1. Results are expressed as mean ± standard error of the mean (SEM). Statistical differences (p < 0.05) were assessed by one‐way ANOVA followed by Tukey's post hoc test, using XLSTAT trial version for Windows (Addinsoft 2023, New York, USA).

Author Contributions

Héctor D. Romero‐Cantú: methodology, investigation, visualisation, data curation, writing – original draft. Riccardo Trentin: data curation, methodology, validation, supervision. Eliana Fernandes: methodology, investigation, Inci Kurt‐Celep: investigation, resources, formal analysis, writing – original draft. Gökhan Zengin: investigation, formal analysis, resources, writing – original draft. Paola Angelini: investigation, resources, formal analysis, writing – original draft. Giancarlo Angeles Flores: investigation, resources, formal analysis, writing – original draft. Luísa Custódio: project administration, funding acquisition. Maria João Rodrigues: conceptualization, methodology, investigation, visualisation, validation, data curation, supervision, writing – review and editing, funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

This study received Portuguese national funds from FCT—Foundation for Science and Technology through contracts UID/04326/2025, UID/PRR/04326/2025 and LA/P/0101/2020 (DOI:10.54499/LA/P/0101/2020), and from the operational programmes CRESC Algarve 2020 and COMPETE 2020 through contract EMBRC.PT ALG‐01‐0145‐FEDER‐022121. M.J.R. was supported by an FCT program contract (UIDP/04326/2020), and E.F. acknowledges FCT for the PhD fellowship (UI/BD/151301/2021).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: cbdv70868‐sup‐0001‐SuppMat.docx

Romero‐Cantú H. D., Trentin R., Fernandes E., et al. “Combining the Salt‐Tolerant Plants Limonium algarvense Erben and Polygonum maritimum L. Promotes Synergistic Bioactivities for Enhanced Cosmetic Applications.” Chemistry & Biodiversity 23, no. 1 (2026): e03370. 10.1002/cbdv.202503370

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.