Anti-Inflammatory Activity and ROS-Inhibitory Effect of Gomphrenin-Enriched Preparations

1. Introduction

The growing interest in plant-derived bioactive compounds has shifted from identifying molecules with antioxidant capacity toward elucidating their specific mechanisms of action in chronic inflammatory processes [1]. Inflammation is an essential component of host defense and tissue repair, however, persistent or dysregulated inflammatory responses can contribute to tissue damage and disease progression. In the intestine, inflammatory processes are particularly complex because epithelial cells and immune cells continuously interact with dietary components, microbial products, and endogenous danger signals. Excessive production of inflammatory mediators, increased oxidative stress, and impairment of epithelial barrier function are not independent events but form a closely interconnected network that can sustain and amplify intestinal inflammation [2,3,4]. Consequently, evaluation of potential anti-inflammatory agents requires consideration of multiple cellular processes rather than reliance on a single endpoint. While numerous phytochemicals have demonstrated anti-inflammatory potential in vitro and in vivo, it has become increasingly evident that structurally related metabolites may differ substantially in their biological activity [5,6]. Among naturally occurring nitrogen-containing pigments, betalains have emerged as promising modulators of inflammation. However, current knowledge is strongly biased toward betanin from red beet (Beta vulgaris), whereas other naturally occurring betalain structures remain largely unexplored [7,8]. This imbalance limits our understanding of how structural modifications within the betalain scaffold influence biological activity and whether individual betalain subclasses exhibit distinct anti-inflammatory properties. Particularly little is known about gomphrenins and their naturally occurring derivatives, despite their considerable structural diversity resulting from variations in glycosylation, acylation, and decarboxylation patterns [9,10]. These structural modifications may affect not only pigment stability and bioavailability but also interactions with molecular targets involved in inflammatory signaling. Betalains are a unique class of water-soluble nitrogen-containing pigments naturally occurring in plants of the order Caryophyllales [11,12]. Beyond their well-established role as natural colorants, these compounds have attracted considerable scientific interest due to their diverse biological activities, including antioxidant, anti-inflammatory, anticancer, and cardioprotective effects [13]. Among these properties, their anti-inflammatory potential has emerged as a particularly promising area of research, given the central role of chronic inflammation in the development of numerous non-communicable diseases. Current evidence suggests that betalains modulate inflammatory responses through multiple molecular mechanisms, including the regulation of oxidative stress, inhibition of pro-inflammatory signaling pathways such as NF-κB and MAPK, and suppression of the production of inflammatory mediators and cytokines [14,15]. While the majority of available studies have focused on betanin and related pigments derived from red beet, considerably less attention has been paid to betalains occurring in other plant species. Gomphrenins, a distinctive group of betacyanins predominantly found in species of the genus Gomphrena and several other representatives of the family Amaranthaceae, represent a structurally diverse subclass of betalains with potentially unique biological properties [16,17]. Despite increasing interest in their chemistry, stability, and biosynthesis, knowledge regarding their anti-inflammatory activity remains limited. Therefore, a comprehensive evaluation of the anti-inflammatory potential of gomphrenin derivatives is warranted to better understand their biological functions and assess their potential application as naturally derived bioactive promising candidates for future functional food and nutraceutical applications.

An important question is whether the biological effects of betalain-rich preparations can be predicted from the properties of individual pigments, or whether they depend on the composition of the preparation as a whole. Preparations differing in their gomphrenin composition may therefore exhibit distinct effects on inflammatory responses, and their biological activity cannot necessarily be extrapolated from the effects previously reported for betanin. To address this issue, the present study investigated preparations obtained from Basella alba f. rubra and Gomphrena globosa, including crude and purified fractions, with particular focus on their effects on cellular responses associated with inflammation and oxidative stress. The comparison of crude and purified B. alba preparations further provided an opportunity to examine how changes in chemical composition may influence biological activity and whether enrichment of specific constituents, removal of potentially antagonistic compounds, or altered interactions among co-occurring metabolites contribute to differences in cellular responses.

2. Materials and Methods

2.1. Plant Material

Basella alba L. forma rubra fruits were obtained from the greenhouse of the Agricultural University of Krakow (Faculty of Biotechnology and Horticulture) where sowing of the seeds was performed in a 3:1 ratio of soil and coconut pith mass and watered daily. The seedlings were transplanted to fertile soil with plenty of organic matter and a pH of 6.5-6.8. The plants were designed to support the fast climbing of the vines which were trellised so that they reached up to 3 m in length at controlled humidity and temperature, ensuring continuous flowering and fruiting as described in our previous article [18]. After harvesting, the fruits were stored at -18 °C until further use.

Dried purple flowers of Gomphrena globosa L. were purchased from a China market.

2.2. Sample Preparation

The fruits were extracted with demineralized water for 1 h at room temperature (0.6 L per 1 kg of raw material). During extraction, the fruits were gently crushed (without damaging the seeds) to enhance the extraction efficiency. The liquid extract was separated from the plant material by centrifugation at 1300 rpm, with the plant material contained in a nylon filter bag. The extraction was repeated three times. The combined extracts were frozen and freeze-dried, yielding B. alba fruit extract (BE) which was stored at -18 °C prior to further studies.

For the preparation of betacyanin-enriched B. alba fruit extract (BE₁), the raw material was extracted as described above, except that EDTA was added during extraction (0.4 mL of saturated EDTA solution per 1 L of water), after which the extract was stored as described above, then filtered under reduced pressure through a layer of silica gel with a particle size of 60-200 µm (VWR, USA), and diluted with water before further use. The prepared extract was purified using a weak anion-exchange resin (Sepra™ ZT WAX, Phenomenex, USA), which had been acetone-activated and water-conditioned. After pigment adsorption, the resin was washed with demineralized water. The pigments were eluted with an aqueous solution containing 50% (v/v) acetone and 5% (v/v) formic acid, followed by 50% (v/v) acetone and 10% (v/v) formic acid. The eluate was concentrated under reduced pressure at 22 °C using a rotary evaporator, diluted with demineralized water and applied to a hydrophobic ODS C18 sorbent (Chromabond, Macherey & Nagel, Germany), which had been acetone-activated and water-conditioned. After washing with demineralized water, the pigments were eluted with an aqueous solution containing 50% (v/v) acetone and 5% (v/v) acetic acid, followed by 50% (v/v) acetone and 10% (v/v) acetic acid. The eluate was concentrated, freeze-dried, and stored at -18 °C.

The betanidin 6-O-β-glucoside (gomphrenin)-enriched fraction (BE₂) and the 6′-O-trans-4-coumaroyl-gomphrenin (globosin)-enriched fraction (BE₃) were obtained as follows. The fruit extract prepared as described for BE₁ was acidified with hydrochloric acid to 0.04 M and loaded onto a strong cation-exchange resin with sulfonic acid functional groups. The column was washed sequentially with 0.1 M hydrochloric acid, 5% (v/v) formic acid and demineralized water. The pigments were eluted with 50% (v/v) aqueous acetone, with EDTA added to the eluate during elution (0.1 mL of saturated EDTA solution per 1 L of eluate). The eluate was concentrated and purified using a hydrophobic ODS C18 sorbent as described for the BE₁ sample, then separated using preparative high-performance liquid chromatography (HPLC) (Shimadzu Corp., Japan) on a Varian Dynamax Microsorb C18 column (41 × 250 mm, particle size 8 µm; Varian, USA). Mobile phase A consisted of 1% (v/v) aqueous formic acid and phase B of acetone. The gradient of phase B was: 6% (v/v) at 0–15 min, 8% (v/v) at 15-25 min, 10% (v/v) at 25-35 min, 15% (v/v) at 35-45 min, 20% (v/v) at 45-55 min, 30% (v/v) at 55-63 min, and 70% (v/v) at 63-70 min, at a flow rate of 40 mL/min. Fractions containing the target pigments were selected based on monitoring by a liquid chromatography coupled with diode array detection and electrospray ionization mass spectrometry (LC-DAD-ESI-MS) system, concentrated, freeze-dried, and stored at -18 °C.

The enriched extract from dried purple flowers of G. globosa L. (GE₁) was obtained by extracting plant material with 1% (v/v) aqueous formic acid solution (30 L per 1 kg) supplemented with saturated EDTA solution (0.2 mL per 10 L). The obtained extract was subsequently filtered through silica gel and purified using an anion-exchange resin and a hydrophobic ODS C18 sorbent, as described for the BE₁ sample, except that strong anion-exchange resin (Sepra™ ZT-SAX, Phenomenex, USA) was used instead of a weak anion-exchange resin, as previously reported [9]. The eluate was concentrated, freeze-dried, and stored at -18 °C.

2.3. Chromatographic Analysis Using Liquid Chromatography-Diode Array Detection-Electrospray Ionization Mass Spectrometry (LC-DAD-ESI-MS)

For the analysis of extracts and fractions collected during purification, a low-resolution LC-MS-8030 mass spectrometric system (Shimadzu, Japan) coupled to LC-20ADXR HPLC pumps, an injector model SIL-20ACXR, and a PDA detector (photodiode array) model SPD-M20A, all controlled with LabSolutions software, version 5.60 SP1 (Shimadzu), was applied [9,18].

The samples were eluted through a 150 mm × 4.6 mm i.d., 5.0 μm, Kinetex C18 chromatographic column preceded by a guard column of the same material (Phenomenex, USA). The injection volume was 7 μL, and the flow rate was 0.5 mL/min. The column was thermostated at 40 °C. The separation of the analytes was performed with binary gradient elution. The mobile phases were A, 1% (v/v) formic acid in water and B, methanol. The gradient profile was (t (min), % B), (0, 10), (12, 55), (15, 85). The full-range PDA signal was recorded, and individual chromatograms were displayed at 538, 505, 490, and 440 nm. Positive ion electrospray mass spectra were recorded using the LC-MS system. The ionization electrospray source operated in positive mode (ESI+), with an electrospray voltage of 4.5 kV and a capillary temperature of 250 °C. N₂ gas was used for the spray. Argon was used as the collision gas for the collision-induced dissociation (CID) experiments. The relative collision energies for MS/MS analyses were set at -35 V on an arbitrary scale. The system recorded total ion chromatograms, mass spectra, and ion chromatograms in the selected ion monitoring mode (SIM). The normalized composition of detected betacyanins was calculated from the peak areas of the corresponding SIM chromatographic signals. The sum of the peak areas of the detected signals assigned to identified betacyanins was set to 100%.

2.4. Determination of Total Betacyanin Content

To determine the total betacyanin content, the samples were analyzed spectrophotometrically using an Infinite 200 PRO microplate reader (Tecan, Switzerland).

The total betacyanin concentration was calculated according to the Beer-Lambert law:

A = C ∙ ε ∙ l,

where: A is the mean absorbance, C is the molar concentration of the analyte in the solution (mol dm^-3), l is the optical path length (cm), and ε is the molar extinction coefficient (dm³ mol^-1 cm^-1).

A molar extinction coefficient (ε) of 65 000 dm³ mol^-1 cm^-1 and an optical path length (l) of 0.53 cm were used for the calculations. The absorbance values used in the calculations corresponded to the absorption maxima recorded within the wavelength range of 500-570 nm. The total betacyanin content was calculated using the molecular weight of the dominant pigment: 550 g mol^-1 for BE, BE₁ and BE₂; 696 g mol^-1 for BE₃; 726 g mol^-1 for GE₁.

2.5. Biological Studies

2.5.1. Reagents

Dimethyl sulfoxide (DMSO, 113635509) was purchased from Chempur (Piekary Śląskie, Poland). Citric acid, sodium citrate tribasic dihydrate, and glucose for acid citrate dextrose solution (ACD) were purchased from Chempur (Piekary Śląskie, Poland). Dextran from Leuconostoc mesenteroides, propidium iodide, dexamethasone, luminol, formyl-met-leu-phenylalanine (f-MLP), sodium butyrate, and Triton X-100 were purchased from Sigma-Aldrich (Saint Louis, MO, USA). RPMI 1640 medium, amphotericin B, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) were purchased from Sigma-Aldrich (Saint Louis, MO, USA). Pancoll Human was purchased from PAN-Biotech (Aidenbach, Germany). Phosphate buffered saline (PBS) and Dulbecco’s modified Eagle medium (DMEM) enriched with 3.7 g/L NaHCO3 and 4.5 g/L glucose, with stable L-glutamine, low endotoxin, fetal bovine serum (FBS) as well as penicillin-streptomycin were purchased from Biowest (Nauillé, France). Hanks’ balanced salt solution (HBSS) and TrypLE™ Express Enzyme were purchased from Gibco Thermo Fisher (Waltham, MA, USA). Recombinant human IL-1β protein, LPS from Escherichia coli (0111:B4), and formic acid were obtained from Merck (Darmstadt, Germany). Recombinant tumor necrosis factor-α (TNF-α) as well as recombinant inteferon-γ were purchased from Pepro-Tech (Cranbury, NJ, USA). Sets of immunosorbent assays for human IL-8, TNF-α, and IL-1β were purchased from BD Biosciences (Erembodegem, Belgium). Pierce BCA Protein Assay Kit (23225) was obtained from Thermo Fisher Scientific (Rockford, IL, USA).

2.5.2. Cytotoxicity and IL-8 Assay in Caco-2 Cell Line

Caco-2 cells were cultured in 75 mL flasks at a seeding density of 10⁵ cells/mL in DMEM supplemented with 20% FBS, 2 mM L-alanyl-L-glutamine, 100 IU/mL penicillin, 100 µg/mL streptomycin, and 2.5 ng/mL amphotericin B. Cultures were maintained at 37 °C in a humidified atmosphere with 5% CO₂. Cells in passages ranging from 11 to 28 were used for all experiments.

Cytotoxicity was assessed via an MTT assay. Cells treated with extracts (1-100 µg/mL) for 48 h were washed, incubated with MTT (0.5 mg/mL), and dissolved in DMSO. Absorbance was measured at 560 nm and 620 nm using a microplate reader (Biotek Epoch Microplate Reader, USA). Triton X-100 (10%, v/v) served as the positive control.

For experiments, cells were seeded in 24-well plates (2 × 10⁵ cells/well) in 20% FBS medium. After 24 h, the medium was replaced with DMEM containing 5% FBS, 2 mM L-alanyl-L-glutamine, 100 IU/mL penicillin, 100 µg/mL streptomycin, 2.5 ng/mL amphotericin B, and 5 mM sodium butyrate to induce differentiation for 72 h. Differentiated cells were treated overnight with extracts (1, 10, 50, and 100 µg/mL) before stimulation with IL-1β, TNF-α, IFN-γ, and LPS to induce IL-8 secretion as it was previously described. After 24 h, supernatants were collected, and IL-8 levels were quantified using ELISA and normalized to protein content with the Pierce™ BCA Assay Kit. Dexamethasone (5-100 µM) served as a positive control [19].

2.5.3. The Cytotoxicity, ROS, and Cytokine Assays in PMN

Neutrophils were isolated from buffy coats obtained from healthy volunteers (<35 years old) at the Warsaw Blood Donation Centre, using anticoagulant citrate dextrose solution (ACD). Isolation involved dextran sedimentation, density gradient centrifugation (Pancoll Human, 1.077 g/mL), and erythrocyte removal via water-induced hemolysis. The cells were suspended in RPMI 1640 or HBSS and stored at 4 °C until use.

For ROS production assays, neutrophils (3.5 × 10⁵ cells/well) in HBSS were seeded in 96-well plates, while cytokine secretion assays used 2 × 10⁶ cells/well in RPMI 1640 with 10% FBS, 2 mM L-glutamine, and 10 mM HEPES. Cell cytotoxicity was assessed via flow cytometry, with cells treated with extracts (1, 10, 50, and 100 µg/mL) for 24 hours. Following treatment, cells were stained with propidium iodide (PI) and analyzed on a BD FACSVerse (BD Biosciences, San Jose, CA, USA). Cytotoxicity was expressed as a percentage of PI(+) cells, with Triton X-100 as the positive control.

ROS production by f-MLP-stimulated neutrophils was measured using chemiluminescence. Cells were treated with extracts (1-100 µg/mL), luminol (2.5 mM), and f-MLP (1.5 µg/mL). Luminescence was recorded over 40 minutes in a microplate reader (SYNERGY 4, Biotek, Winooski, VT, USA), and inhibition percentages were calculated against not-stimulated controls (n.st. control) [19]. Ascorbic acid (ASC) was used as a positive control at concentrations 1 (5.7 µM), 25 (142 µM), and 50 (268 µM) µg/mL.

Cytokine secretion (IL-8, TNF-α, IL-1β) was measured by ELISA after LPS stimulation (100 ng/mL). Neutrophils were pre-treated with extracts (1-100 µg/mL) for 30 minutes and cultured for 24 hours. Results were expressed relative to LPS-stimulated controls, with dexamethasone (Dex) as the positive control [19].

2.6. Statistical Analysis

Data on the normalized composition of detected betacyanins in the studied preparations determined by LC-DAD-ESI-MS analysis, and total betacyanin contents determined spectrophotometrically, are presented as mean ± standard deviation (SD) of three independent measurements.

Biological data are presented as mean ± standard error of the mean (S.E.M). Each sample was analyzed in triplicate across three independent experiments. Statistical differences between means were assessed by verifying variance homogeneity and normality of distribution, followed by one-way ANOVA. When necessary, non-parametric tests such as Kruskal–Wallis or U-Mann–Whitney were applied. Statistical significance was set at p < 0.05. Data analysis was performed using Statistica 13 software (TIBCO Software Inc., Palo Alto, CA, USA).

3. Results

3.1. Betacyanin Composition of the Purified Preparations

The LC-DAD-ESI-MS analysis demonstrated pronounced differences in both the qualitative and quantitative composition of betacyanins among the obtained preparations (Table 1 and Table 2), reflecting the efficiency of the applied purification procedures.

Considerable differences were observed in the total betacyanin content of the analyzed samples (Table 2). The total betacyanin content in freeze-dried fruit extract BE was 56 ± 2 mg g^-1.

Purification markedly increased the concentration of betacyanins in the preparations. The total betacyanin content increased to 235 ± 57 mg g^-1 in BE₁, 312 ± 83 mg g^-1 in the gomphrenin-enriched fraction BE₂ and 188 ± 12 mg g^-1 in the globosin-enriched fraction BE₃ (Table 2). The highest betacyanin concentration observed in BE₂ indicates the efficiency of the preparative HPLC procedure for isolating the major non-acylated betacyanin, gomphrenin and its isoform. In contrast, BE₃ contained a lower total pigment concentration because this fraction was enriched in naturally less abundant acylated derivatives, particularly globosin and other hydroxycinnamoyl conjugates.

A markedly lower total betacyanin content was determined for the purified G. globosa extract (GE₁), amounting to 12 ± 1 mg g^-1 (Table 2). The betacyanin profile was characterized by a high proportion of feruloyl-, coumaroyl- and sinapoyl-substituted betacyanins, which is consistent with the characteristic pigment profile previously reported for G. globosa [20].

Purification of the crude extract on a weak anion-exchange resin followed by ODS C18 chromatography yielded sample BE₁, which exhibited the highest structural diversity of betacyanins, including non-acylated, malonylated and hydroxycinnamoyl derivatives of gomphrenin and isogomphrenin (Figure 1 and Figure 2). Nevertheless, the betacyanin profile was strongly dominated by gomphrenin (betanidin 6-O-β-glucoside) 3, representing 69.7 ± 1.1%, whereas isogomphrenin 3′ accounted for 13.1 ± 1.5% (Table 2).

The purification strategy employing cation-exchange chromatography followed by preparative HPLC enabled selective enrichment of individual pigments in the betacyanin profile. Sample BE₂ represented a gomphrenin-enriched fraction, in which gomphrenin 3 constituted 51.9 ± 4.8% of the detected betacyanins and isogomphrenin 3′ 47.9 ± 4.8%, where malonylated derivatives 4 were detected at low levels (Figure 3). Consequently, more than 99% of the betacyanin composition consisted of the two glucosylated betanidin isomers (Table 2).

In contrast, sample BE₃ was enriched in acylated betacyanins (Figure 4). The predominant compound in the betacyanin profile was globosin (6′-O-E-4-coumaroyl-gomphrenin) 12, accounting for 48.3 ± 0.5%, followed by 6′-O-E-feruloyl-gomphrenin (basellin) 13 (12.8 ± 0.1%) and its isoform. Only a residual amount of gomphrenin 3 was detected in the prepared fraction (4.6 ± 0.3%) (Table 2). The considerable proportion of hydroxycinnamoyl derivatives demonstrates that preparative HPLC effectively separated the less hydrophilic acylated pigments from the parent glucosides.

The total betacyanin content of the purified G. globosa extract (GE₁) was lower than that of B. alba f. rubra, while the pigment composition differed substantially (Figure 5). Unlike BE₁, GE₁ contained relatively low amounts of non-acylated gomphrenin 3 (2.2 ± 0.1%) and isogomphrenin 3′ (5.6 ± 0.5%), whereas acylated derivatives collectively predominated in the betacyanin profile. The major betacyanin was basellin 13 (36.4 ± 1.1%) followed by its isoform, isobasellin 13′ (25.4 ± 0.9%), globosin 12 (5.6 ± 0.4%), and sinapoyl derivative gandolin 11 (4.3 ± 0.4%) (Table 2). This betacyanin profile agrees with previous reports indicating that G. globosa accumulates predominantly acylated gomphrenin derivatives [21], whereas B. alba fruits are characterized by the predominance of non-acylated gomphrenin [22].

The chromatographic profiles also reflect the influence of ion-exchange chemistry on pigment isolation (Figure 1, Figure 2, Figure 3, Figure 4 and Figure 5). Weak anion-exchange purification (BE₁) retained the complex profile of betacyanins, while the subsequent preparative HPLC enabled fractionation according to polarity, yielding fractions enriched either in the parent glucosides (BE₂) or in hydroxycinnamoyl derivatives (BE₃). The applied purification strategy provided an efficient approach for obtaining betacyanin preparations differing markedly in chemical composition for further studies on the physicochemical or biological properties.

3.2. Bioactivity

None of the tested samples affected the viability of Caco-2 or PMN cells over the investigated concentration range (Figure 6), indicating that the observed effects on cytokine secretion (Figure 7 and Figure 8) and ROS generation (Figure 7) were not attributable to cytotoxicity. In particular, the Caco-2 and PMN cells maintained viability above 85% and 90%, respectively, after betacyanin-enriched fraction treatment. The differences in biological responses may therefore reflect variations in betacyanin profiles as well as differences in the overall composition and abundance of other bioactive constituents present in the tested preparations.

Purification of the crude B. alba extract BE significantly altered its biological properties, as evidenced by a marked reduction in the secretion of cytokines/chemokines, particularly TNF-α and IL-8, and IL-1β to a lesser extent (Figure 8). This enhanced anti-cytokine activity may result from both the removal of potentially interfering matrix components and the enrichment of bioactive compounds in BE₁.

Comparison of the B. alba preparations further suggests that differences in betacyanin composition may contribute to their distinct biological effects. In particular, BE₃, enriched in acylated betacyanins, generally induced a more pronounced reduction in TNF-α, IL-8, and IL-1β secretion than the gomphrenin-enriched preparation BE₂ (Figure 8). Its effect was mostly significant compared with the (+) LPS control across the entire concentration range. The secretion of TNF-α, IL-8 (CXCL8), and IL-1β in BE3 (100 μg/mL) treatment was 39.17%, 44.61%, and 58.93% compared to control, whereas BE₂ exhibited only limited activity. In contrast, both preparations effectively suppressed ROS generation by activated PMNs (Figure 9). These findings indicate that the biological activity of the preparations may depend not only on the total betacyanin content but also on the relative abundance and structural characteristics of individual betacyanins. In addition, the anti-cytokine/chemokine effect may not be related only to the antioxidant activity of BE₂ and BE₃.

Despite its considerably lower total betacyanin content, GE₁ also exhibited anti-cytokine/chemokine activity (Figure 8). Because this preparation was obtained from a different botanical source and was naturally enriched in acylated betacyanins, direct comparison with the B. alba-derived preparations should be interpreted with caution. Moreover, extracts of G. globosa flowers may contain a range of other bioactive phenolic constituents, including flavonoids such as quercetin, kaempferol and isorhamnetin derivatives, as well as hydroxycinnamic acids, which could contribute to the observed biological effects [23]. Nevertheless, the similar biological responses observed for GE₁ and BE₃ support the possibility that hydroxycinnamoyl-substituted betacyanins contribute at least in part, to the modulation of inflammatory responses. It is worth noting that the potency of GE₁ antioxidant activity is much weaker than BE₂ or BE₃. Due to the lower gomphrenin 3 and isogomphrenin 3′ content in GE₁, compared to BE₂ and BE₃, it might be suggested that these phytochemicals are mainly responsible for the antioxidant activity in the tested fractions.

All tested preparations significantly reduced IL-8/CXCL8 secretion in the Caco-2 model.. All fractions at the highest concentrations decreased the chemokine secretion by more than 50%. CXCL8 is the most potent human neutrophil-attracting chemokine and plays crucial roles in the response to infection and tissue injury. The reduction of IL-8/CXCL8 secretion by gastrointestinal-related cells might lead to limitation of leukocyte recruitment to the site of inflammation and support termination of the inflammatory state.

4. Discussion

The LC-DAD-ESI-MS analysis demonstrated that the purification procedures resulted in preparations with markedly different betacyanin profiles, both in terms of total pigment concentration and, more importantly, in the relative abundance and structural diversity of individual betacyanins (Table 1). This chemical differentiation provides an important basis for interpreting the biological effects observed in the cellular models. Rather than representing preparations differing only in the amount of betacyanins, BE₁, BE₂, BE₃, and GE₁ constituted chemically distinct pigment mixtures, characterized by different proportions of non-acylated, malonylated, and hydroxycinnamoyl-substituted betacyanins (Table 2). Consequently, the biological differences observed among these preparations should be considered in the context of both their quantitative pigment content and their qualitative composition.

The crude freeze-dried B. alba fruit extract (BE) contained 56 ± 2 mg of total betacyanins per g dry weight, whereas purification substantially increased the pigment concentration in all B. alba-derived preparations (Table 2). Thus, the purification procedures increased the total betacyanin concentration approximately 4.2-, 5.5-, and 3.4-fold relative to the crude extract, respectively. This enrichment confirms the effectiveness of the chromatographic procedures but also demonstrates that purification resulted in preparations with substantially different pigment concentrations. A comparable enrichment effect was reported by Xu et al. (2024) [24], who observed an increase in betacyanin content from 9.28 ± 0.04 mg/g in a crude red dragon fruit peel extract to 229.45 ± 10.12 mg/g after solid-phase extraction (SPE), corresponding to an approximately 25-fold increase. These findings support the effectiveness of purification in concentrating betacyanins, while highlighting that the magnitude of enrichment may vary considerably depending on the plant source and purification strategy [24].

The chemical composition of BE₁ indicated that the initial purification step retained a broad spectrum of B. alba-derived betacyanins. Although BE₁ contained numerous compounds, including non-acylated, malonylated, and hydroxycinnamoyl-substituted derivatives, its profile was strongly dominated by gomphrenin 3 (betanidin 6-O-β-glucoside, Table 2), which accounted for 69.7 ± 1.1% of the detected betacyanins. Isogomphrenin (compound 3’, Table 2) represented a further 13.1 ± 1.5%. Thus, approximately 83% of the betacyanin profile of BE₁ consisted of these two non-acylated glucosides. The remaining fraction comprised a diverse set of minor pigments, including malonylated and hydroxycinnamoyl derivatives. BE₁ can therefore be considered a broadly enriched B. alba betacyanin preparation dominated by non-acylated gomphrenin-type pigments rather than a chemically homogeneous fraction. The predominance of gomphrenin and isogomphrenin in BE₁ is consistent with previous reports on the betacyanin composition of B. alba fruits. Sutor-Świeży et al. (2022) [22] reported gomphrenin and isogomphrenin as major pigments in B. alba fruits, accounting for 39.7% and 13.7% of the total betacyanin content, respectively, while also identifying a diverse range of malonylated and hydroxycinnamoyl-acylated gomphrenin derivatives. Similarly, Lin et al. (2010) [25] identified gomphrenin I as the principal purple pigment of B. alba fruits, confirming the characteristic predominance of gomphrenin-type betacyanins in this species [22,25].

In contrast, the preparative HPLC procedure used to obtain BE₂ resulted in selective enrichment of the two non-acylated glucosides. Gomphrenin and isogomphrenin represented 51.9 ± 4.8% and 47.9 ± 4.8% of the detected betacyanins, respectively, while only trace amounts of malonylated gomphrenin remained. Consequently, more than 99% of the quantified betacyanin profile of BE₂ consisted of these two structural isomers. This is an important feature of BE₂ because it makes this preparation considerably more chemically defined than BE₁ and BE₃. At the same time, the approximately 1:1 ratio of gomphrenin to isogomphrenin means that the biological effect of BE₂ cannot be assigned specifically to gomphrenin itself. Any activity of this fraction may reflect the combined effects of both isomers and their possible interactions. The selective enrichment of gomphrenin-type pigments by preparative HPLC is consistent with previous studies demonstrating that chromatographic fractionation can yield chemically defined B. alba betacyanin preparations enriched in specific gomphrenin derivatives. Sutor-Świeży et al. (2024) isolated highly purified gomphrenin and several acylated gomphrenin derivatives from B. alba extracts and evaluated their bioactivities individually, highlighting the importance of distinguishing the effects of individual pigments from those of complex pigment mixtures. This is particularly relevant for BE₂, where the nearly equimolar presence of gomphrenin and isogomphrenin prevents the observed biological activity from being unequivocally attributed to gomphrenin alone [18].

BE₃ exhibited a fundamentally different chemical profile. In contrast to BE₂, it was strongly enriched in acylated betacyanins, particularly hydroxycinnamoyl derivatives (Table 2). Globosin (6’-O-E-4-coumaroyl-gomphrenin, Table 2) represented 48.3 ± 0.5% of the detected betacyanins, while isoglobosin (compound 12’, Table 2) accounted for 19.8 ± 0.4%. Basellin (6’-O-E-feruloyl-gomphrenin, compound 13 Table 1) and isobasellin (compound 13’) represented 12.8 ± 0.1% and 4.4 ± 0.1%, respectively. Additional acylated derivatives, including malabarin/isomalabarin and other conjugates, were also present, whereas gomphrenin accounted for only 4.6 ± 0.3 mg/g DW. Thus, BE₃ represents a chemically distinct preparation in which hydroxycinnamoyl-substituted betacyanins predominate over the parent non-acylated glucosides. The lower total betacyanin concentration of BE₃ compared with BE₂ is therefore accompanied by a major qualitative shift in pigment composition.

This difference between BE₂ and BE₃ is particularly relevant to the biological results. BE₂ contained the highest total concentration of betacyanins (312 ± 83 mg/g DW), whereas BE₃ contained significantly less (188 ± 12 mg/g DW). Nevertheless, BE₃ generally produced a stronger reduction in TNF-α, IL-8, and IL-1β secretion than BE₂ (Figure 8). The preparations were applied to cells at identical mass concentrations, thus observation cannot be explained simply by the total amount of betacyanin delivered to the cells. BE₂ would, on a mass basis, provide considerably more total betacyanin than BE₃, yet BE₃ possessed a more complex betacyanin profile and displayed greater activity against several inflammatory endpoints. This finding therefore suggests that differences in the qualitative composition of the preparations, including their betacyanin profile, may be more important than total betacyanin concentration for some biological response. This interpretation is consistent with previous studies indicating that the biological activity of betalain-containing preparations cannot be explained solely by their total pigment content. Smeriglio et al. (2021) [26] demonstrated that betalain-rich extracts could exert stronger anti-inflammatory effects than their major isolated betalains when compared at equivalent concentrations, suggesting that interactions among different constituents may contribute to the overall biological response. Similarly, Fernando et al. (2023) [27] showed that individual betalain pigments differed in their effects on inflammatory and oxidative-stress markers, indicating that the biological response depends on the structural characteristics of the pigments. Furthermore, Martinez et al. (2015) demonstrated that a betalain-rich preparation from Beta vulgaris significantly reduced TNF-α and IL-1β production in an experimental model of inflammation, confirming that betalain-rich preparations can directly modulate key inflammatory mediators. Taken together, these findings support the hypothesis that the qualitative composition of BE₂ and BE₃ may contribute to their different anti-inflammatory effects. Thus, the higher total betacyanin concentration of BE₂ does not necessarily translate into stronger biological activity, and the relative abundance and structural diversity of individual betacyanins should be considered when interpreting the biological effects of B. alba-derived preparations [26,27,28].

Although BE₁ contained a high total betacyanin concentration (235 ± 57 mg/g DW) and was dominated by gomphrenin, it also retained a significantly greater structural diversity than BE₂. Its biological properties were therefore likely to reflect the combined effects of major non-acylated pigments and a range of minor derivatives. The fact that purification from BE to BE₁ significantly altered the inflammatory response indicates that the biological properties of the crude extract cannot be predicted solely from its total pigment content. The crude extract contained only 56 ± 2 mg/g DW total betacyanins, but it also contained a complex matrix of non-pigment constituents that were largely removed or altered during purification. The increased activity of purified preparations may therefore result from pigment enrichment, removal of inhibitory or competing matrix components, improved accessibility of pigments, or changes in pigment stability. Similar observations have been reported for other betalain-containing systems, in which the surrounding matrix influenced pigment stability, accessibility, and biological behavior. Herbach et al. (2006) [29] demonstrated that the juice matrix substantially affected the stability and structural transformations of betacyanins, including betanin isomerization and phyllocactin deacylation, highlighting the importance of non-pigment components in determining pigment reactivity. More recently, Roman-Maldonado et al. (2025) [30] showed that the composition and complexity of the food matrix significantly affected betalain stability and bioaccessibility, demonstrating that a higher initial pigment concentration does not necessarily translate into greater availability of the compounds at their biological site of action. In addition, Smeriglio et al. (2021) [26] reported that betalain-rich extracts could exhibit stronger anti-inflammatory effects than individual major betalains, suggesting that interactions among constituents may contribute to the overall biological response. Therefore, the increased activity observed after purification of BE may reflect not only the approximately fourfold enrichment of betacyanins but also changes in the surrounding chemical environment of the pigments. Removal of non-pigment constituents may have reduced compounds capable of promoting pigment degradation or interfering with their activity, while simultaneously increasing pigment accessibility and stability. However, because these matrix components were not individually characterized in the present study, their contribution to the observed difference between BE and BE₁ should be regarded as a probable explanation rather than a demonstrated mechanism [26,29,30].

The comparison of BE₂ and BE₃ is particularly informative because these fractions represent relatively distinct chemical profiles. BE₂ consisted almost exclusively of the non-acylated gomphrenin/isogomphrenin pair, whereas BE₃ was dominated by acylated derivatives, principally globosin, isoglobosin, basellin, and isobasellin. The stronger inhibition of TNF-α, IL-8, and IL-1β secretion by BE₃ therefore raises the possibility that hydroxycinnamoyl substitution contributes to the modulation of inflammatory signaling. This interpretation is supported by previous work on B. alba gomphrenins, in which the acylated pigments globosin and malabarin exhibited distinct anti-inflammatory activities and inhibited the secretion of IL-1β and TNF-α in LPS-activated human macrophages [18]. The present results and previous evidence suggest that the higher anti-inflammatory activity of BE₃ may be related not simply to the total amount of betacyanins, but to the structural characteristics of its predominant acylated gomphrenins. However, because BE₃ contains several acylated pigments and their isoforms simultaneously, the present data do not allow the activity to be attributed to a specific compound or to hydroxycinnamoylation alone. Acylation, particularly with hydroxycinnamoyl residues, can substantially modify the physicochemical properties of betalains and may consequently influence their stability and biological behavior. In B. alba, acylated gomphrenins such as globosin and malabarin have been shown to exhibit greater stability than non-acylated gomphrenin, while also displaying distinct antioxidant and anti-inflammatory activities, indicating that hydroxycinnamoyl substitution can affect both pigment persistence and biological function [18]. This enhanced stability has been attributed, at least in part, to structural effects associated with acylation and intramolecular interactions that may protect the betalain chromophore from degradation [11]. More broadly, the bioavailability and biological effects of betalains depend on their absorption, metabolism, and interactions with biological systems, although these processes have been studied much more extensively for non-acylated pigments than for hydroxycinnamoyl-substituted betacyanins. Therefore, the possibility that acylation influences membrane interactions, cellular uptake, metabolism, or the modulation of specific signaling pathways cannot be excluded, but these mechanisms require direct experimental verification for individual acylated B. alba betacyanins.

BE₂ and BE₃ differ in several structural types of acyl substituent, the relative abundance of individual acylated derivatives, and the total pigment concentration. Consequently, differences in minor pigments, phenolic constituents, degradation products, or other matrix components may account for the observed effects. Therefore, the relationship between acylation and biological activity should be regarded as an association requiring further experimental verification. Additionally, both preparations effectively reduced ROS generation in activated neutrophils. This difference is potentially important because it suggests that modulation of inflammatory mediator release and suppression of oxidative burst may be partially separable biological activities.

GE₁ contained the lowest total betacyanin concentration among all analyzed preparations (Table 2). Nevertheless, it exhibited measurable biological activity in the cellular models (Figure 8). Its pigment profile was fundamentally different from that of the B. alba-derived preparations. Non-acylated gomphrenin and isogomphrenin accounted for only 2.2 ± 0.1% and 5.6 ± 0.5%, respectively, whereas acylated pigments predominated (Table 2). Particularly interesting was the high abundance of feruloylated derivatives: basellin and isobasellin represented 36.4 ± 1.1% and 25.4 ± 0.9%, respectively, together accounting for more than 60% of the detected betacyanin profile. Additional acylated pigments included globosin/isoglobosin and sinapoylated derivatives such as gandolin and isogandolin (Table 2). The predominance of acylated gomphrenin derivatives is consistent with previous studies of G. globosa, in which feruloylated and coumaroylated gomphrenins were identified as characteristic pigments of the species [31]. Moreover, a comparative survey of Amaranthaceae species reported one of the highest proportions of acylated betacyanins in G. globosa, further supporting the characteristic enrichment of this species in hydroxycinnamoylated gomphrenin derivatives [20]. Importantly, Silva et al. (2012) demonstrated that G. globosa inflorescence extracts contain both diverse betacyanins and numerous phenolic compounds and exhibit antioxidant and anti-inflammatory activities, including inhibition of nitric oxide production in LPS-stimulated RAW 264.7 macrophages [32]. These findings are particularly relevant to the interpretation of GE₁ because its biological activity may reflect the combined contribution of betacyanins and non-pigment phytochemicals rather than the activity of betacyanins alone. GE₁ originated from a different species and therefore represents a distinct phytochemical matrix, in which phenolic acids, flavonoids, and other secondary metabolites may contribute to the observed cellular response. The relatively low total betacyanin content of GE₁, together with its measurable biological activity, therefore provides further evidence that total betacyanin concentration alone is insufficient to predict the biological activity of complex plant preparations.

The comparison between BE₃ and GE₁ is nevertheless informative from a structure-activity perspective. Both preparations were characterized by a high relative abundance of acylated betacyanins, although their profiles were not identical. BE₃ was dominated by coumaroylated derivatives, globosin and isoglobosin, whereas GE₁ was dominated by feruloylated derivatives, basellin and isobasellin. Previous studies have shown that acylation can influence the biological properties of betacyanins. For example, Spórna-Kucab et al. (2018) reported that naturally occurring acylated gomphrenin derivatives from G. globosa exhibited stronger antimicrobial activity than non-acylated gomphrenin/isogomphrenin, suggesting that the acyl substituent may contribute to the biological activity of the pigment [33]. This finding does not directly establish an anti-inflammatory effect of hydroxycinnamoylated betacyanins, but it supports the broader concept of a structure-activity relationship associated with acylation. In addition, the chemically distinct coumaroylated and feruloylated gomphrenins identified in B. alba fruits represent structurally related but non-identical pigments, providing a basis for considering the nature of the hydroxycinnamoyl substituent as a potential determinant of biological activity [22]. Consistent with this interpretation, different betalains have been shown to exhibit distinct patterns of cellular uptake, redox activity, and modulation of inflammatory markers, indicating that biological responses may depend on the individual pigment structure rather than on betalain concentration alone [34]. Therefore, if BE₃ and GE₁ exhibited similar biological effects despite their different dominant acyl substituents, this could suggest that activity is associated more broadly with hydroxycinnamoylated betacyanins, whereas differences in their responses would support a contribution of the specific acyl substituent or pigment structure. Additionally, the comparison of BE₂ and BE₃ provides a complementary perspective. Both preparations effectively suppressed ROS generation by activated neutrophils despite their markedly different betacyanin profiles (Table 2, Figure 9). This observation suggests that modulation of neutrophil oxidative responses may be a relatively common property of betacyanin-rich preparations and may not require the predominance of either non-acylated or acylated pigments. Previous work demonstrated that purified betanin significantly reduced ROS production in PMA-stimulated human neutrophils, confirming that betalain exposure can modulate neutrophil oxidative metabolism [35]. Moreover, different betalains have been shown to produce distinct effects on both oxidative and inflammatory cellular responses, suggesting that these endpoints may not necessarily be governed by identical structural determinants [34]. Thus, the similar ROS-inhibitory activity of chemically distinct BE₂ and BE₃ fractions may indicate that several structurally different betacyanins are capable of modulating cellular redox responses, whereas the stronger differences observed between BE₂ and BE₃ in cytokine secretion suggest that inflammatory signaling may be more sensitive to specific pigment structures or to interactions among pigments and other constituents.

Importantly, suppression of cellular ROS generation should not be equated directly with radical-scavenging activity. In activated neutrophils, ROS production is a regulated cellular process involving NADPH oxidase, and reduction of the measured oxidative burst may result from several mechanisms, including direct scavenging of reactive species, interference with NADPH oxidase activation, or modulation of upstream intracellular signaling. Indeed, Zielińska-Przyjemska et al. (2012) [35] suggested that the inhibitory effect of betanin on neutrophil chemiluminescence could involve interference with NADPH oxidase activation rather than representing simple chemical scavenging alone. Neutrophil ROS production is also an essential component of antimicrobial host defense. Activation of NADPH oxidase 2 (NOX2) generates superoxide and downstream oxidants that contribute to microbial killing, although excessive or dysregulated ROS production can also promote tissue injury and inflammation [36]. Consequently, the biological significance of the ROS-inhibitory effects observed here depends on the extent and context of the response. Excessive suppression of neutrophil oxidative activity could theoretically compromise antimicrobial defense, and therefore the present findings should be interpreted as modulation of cellular oxidative activity rather than as unequivocally beneficial antioxidant action. Additional mechanistic studies, including assessment of NADPH oxidase activation and downstream redox signaling, would be required to determine the mechanism underlying the observed inhibition.

The absence of significant effects on cell viability is an important observation when interpreting the biological results. None of the tested preparations significantly compromised Caco-2 or neutrophil viability over the investigated concentration range. This supports the interpretation that the observed reductions in cytokine secretion and ROS generation were not primarily caused by overt cytotoxicity. Similar observations have been reported for individual betalains in Caco-2 cells, where betanin, vulgaxanthin I, and indicaxanthin maintained cell viability above 90% at the concentrations tested while simultaneously modulating inflammatory and redox-related markers [34]. Nevertheless, maintenance of overall viability does not exclude minor effects on cellular metabolism, membrane function, or signaling pathways, and additional endpoints would be required to establish cellular tolerance more comprehensively.

The present chemical characterization also provides a more appropriate context for interpreting the previously reported biological activity of gomphrenin-rich preparations. Previous studies have demonstrated antioxidant and anti-inflammatory effects of B. alba preparations and purified or enriched gomphrenin derivatives, including modulation of cytokine secretion and cellular inflammatory signaling [18]. The present findings are generally consistent with these observations, but they extend them by showing that preparations with substantially different distributions of non-acylated and hydroxycinnamoylated gomphrenins can induce distinct cellular responses. At the same time, studies comparing betalain-rich plant extracts with individual purified pigments indicate that the activity of a complex preparation may also reflect interactions among several constituents rather than the action of a single dominant pigment [26]. Direct comparison with previous studies remains limited by differences in extraction procedures, pigment composition, experimental concentrations, inflammatory stimuli, and cell models. Overall, the present data support a model in which the biological activity of betalain-rich preparations is determined by the combined effects of total pigment abundance, individual betacyanin structures, acylation pattern, and the broader phytochemical matrix, rather than by total betacyanin concentration alone.

5. Conclusions

The present study demonstrates that chromatographic purification of B. alba and G. globosa extracts led to obtaining betacyanin-rich preparations with markedly different quantitative and qualitative pigment profiles. LC-DAD-ESI-MS analysis confirmed efficient enrichment of distinct structural classes of betacyanins, ranging from non-acylated gomphrenin/isogomphrenin-rich preparations to fractions dominated by hydroxycinnamoyl-substituted derivatives. The biological results indicated that the activity of the investigated preparations cannot be explained solely by their total betacyanin content. BE₂ contained the highest total concentration of betacyanins and was characterized by a betacyanin profile consisting almost exclusively of gomphrenin and isogomphrenin, whereas BE₃ contained substantially less total pigment but was enriched in acylated betacyanins and showed stronger inhibition of several pro-inflammatory cytokines. Similarly, GE₁ exhibited biological activity despite its considerably lower total betacyanin content and its distinct, predominantly acylated betacyanin profile. These findings indicated that the qualitative composition and structural characteristics of betacyanins may be important determinants of their cellular activity. It should be emphasized, however, that the investigated preparations were not composed exclusively of betacyanins but represented chemically complex plant-derived matrices containing additional constituents that may also have contributed to the observed biological effects. The strong inhibition of ROS generation observed for the B. alba preparations further demonstrates their capacity to modulate oxidative responses in activated neutrophils. However, differences in cytokine inhibition between the preparations indicated that modulation of oxidative burst and inflammatory mediator secretion may not be governed by identical mechanisms. The association between acylated-betacyanin enrichment and enhanced modulation of selected inflammatory endpoints is particularly noteworthy and provides a rationale for further investigation of hydroxycinnamoyl-substituted gomphrenins and related pigments. Nevertheless, because the tested fractions remain chemically complex and differ in both pigment composition and botanical origin, the present results do not establish a direct causal relationship between betacyanin acylation and anti-inflammatory activity. Overall, the findings identify B. alba and G. globosa as promising sources of structurally diverse betalains with potential to modulate inflammatory and oxidative cellular responses. More importantly, the study highlights that, in addition to total betacyanin content, the qualitative composition of betalain-rich preparations should be considered when evaluating their biological properties. Further studies using purified individual betacyanins, equimolar comparisons, mechanistic assays, and physiologically relevant intestinal models are required to establish definitive structure-activity relationships and to determine the translational relevance of these findings.