Ability of short-chain fatty acids to reduce inflammation and attract leucocytes to the inflamed skin of gilthead seabream (Sparus aurata L.)

Abstract

The aim of the study was to investigate the potential preventive use of short-chain fatty acids (SCFAs) to modulate inflammatory responses in gilthead seabream (Sparus aurata) skin. Initially, in vitro experiments were conducted to evaluate the effects of various concentrations of butyric acid, acetic acid and propionic acid, as well as their combination, on the cytotoxicity and cell viability of three different cell lines. The results determined the safe concentration of SCFAs, which was then used for an in vivo study. Fish were allocated into six groups and administered different combinations of SCFAs via intramuscular injection, followed by an injection of carrageenan as an inflammatory agent. Skin samples were taken from the injection site three hours post-administration and used to analyse gene expression and immunohistochemistry. The results demonstrated that treatment with SCFAs resulted in increased expression of proinflammatory and anti-inflammatory genes and leucocyte markers in the inflamed skin of fish. The highest gene expression and recruitment of acidophilic granulocytes were observed in fish injected with propionic acid and carrageenan. It is concluded that acetic acid is the most effective anti-inflammatory SCFA tested in gilthead seabream exposed to acute inflammation induced by carrageenan injection. Acetic acid exhibited the most pronounced direct anti-inflammatory effect, although propionic acid appeared to play a significant role in several mechanisms contributing to the resolution of inflammation and recruitment of immune cells to the site of carrageenan-inflamed area in gilthead seabream skin.

Introduction

The microbiota comprises a diverse array of commensal microorganisms that colonize all surfaces and cavities of animal mucosa (skin, gills, intestine, nasal, oral, etc.), where they exert a remarkable role in the state of health of the hosts. The microbiota is involved in nutrient processing, vitamin synthesis and fermentation of carbohydrates, lipids and proteins, as well as the synthesis of amino acids (essential and non-essential). In addition, the microbiota can produce hundreds of proteins and metabolites that modulate key host functions, such as short-chain fatty acids (SCFAs), among others¹. However, the specific mechanisms underlying the complex interactions between the gut microbiota and host health, particularly their regulation of the hypothalamic-pituitary-adrenal (HPA) axis and the immune system, remain largely unknown². It is evident that the microbiota plays a crucial role in maintaining the structural integrity of the intestinal barrier, which, in turn, acts as a formidable defense against invasion and colonization of pathogenic organisms. Furthermore, it actively contributes to the development and modulation of the host immune system². Consequently, a comprehensive examination of microbiota-host interactions, along with an in-depth exploration of the various effects and underlying mechanisms attributed to individual microorganisms or their metabolic by-products, deserve further investigation.

In this regard, SCFAs, carboxylic acids with less than 6 carbon atoms such as acetate (AA), propionate (PA) and butyrate (BA), are produced by the microbiota in mammals through carbohydrate fermentation and have very interesting properties. They are not only an important source of energy, but also signalling molecules with beneficial effects on various physiological processes³. SCFA have been found to play a regulatory role in metabolic and immune system disorders^4,5. Acetate, propionate and butyrate are rapidly absorbed in the intestinal lumen and are involved in several phases of the inflammatory process⁴. SCFA have been associated with several cellular processes, such as gene expression, differentiation, chemotaxis, proliferation and apoptosis^4,6. Butyrate has been shown to improve epithelial barrier function and intestinal integrity by modulating tight junction proteins and mucin expression⁷. In addition, it has been observed to decrease inflammation in parasite-infected fish⁸. Both butyrate and propionate are known to regulate inflammatory responses in epithelial cells and leucocytes^9,10, modulating leucocyte recruitment, cytokine production, lymphocyte activation and phagocytosis, as well as oxygen radical production¹¹. Conversely, there are also several studies showing that SCFAs can promote inflammatory responses^10,12. In aquaculture, SCFAs and their salts have been used as growth promoters¹³ and as immune stimulators, which improve the overall health of aquatic organisms^13,14.

Skin inflammation is crucial for fish health, acting as both protection and an indicator of underlying problems. The skin serves as the first defence, providing a barrier against pathogens, environmental stressors, and toxins. Inflammation in fish skin typically signals an immune response to infection, injury, or harmful stimuli, aiding in mobilizing immune cells, increasing mucus production, and initiating tissue repair to maintain skin integrity and prevent pathogen spread¹⁵. However, chronic or excessive inflammation can cause tissue damage, compromising the skin’s barrier function and increasing infection susceptibility¹⁶. It may also impair osmoregulation, essential for water and salt balance, leading to reduced growth, poor feed conversion, and higher mortality in aquaculture¹⁵. Visible inflammation signs, like lesions or ulcers, can reduce the fish’s commercial value, leading to significant economic losses¹⁷.

Carrageenan, a linear sulfated polysaccharide of D-galactose and 3, 6-anhydro-D-galactose, is obtained by extraction from specific red algae of the class Rhodophyceae¹⁸. Carrageenan has been widely used for decades as a model of acute inflammation in mammals^19–21 and has been recently applied in fish^22,23. Our research group has studied the inflammation model produced by an intramuscular injection of carrageenan in gilthead seabream (Sparus aurata L.), as well as its resolution within a few hours. The results showed that a subcutaneous injection of κ/λ-carrageenan triggers acute inflammation in seabream skin, with rapid recruitment of acidophilic granulocytes and release of humoral mediators in the cutaneous mucosa^22–24. Having this model of inflammation already available, our intention is to investigate the possible application of pro- and/or anti-inflammatory molecules. The aim of the present work is to study the anti-inflammatory capacity of SCFAs (acetic acid, butyric acid, propionic acid and their mixture). First, an in vitro assay was developed in order to study the effects of SCFAs on the viability of three established cell lines and then the in vivo effects on gilthead seabream skin in case of acute inflammation.

Materials and methods

in vitro assay

Cell lines

Three established cell lines were purchased and used in this study and seeded in 75 cm² plastic tissue culture flasks (Nunc): one from human (HeLa) and two from fish (PLHC-1 and SAF-1). HeLa cells (uterine human carcinoma; ATCC^® CCL-2) were cultured (37 ºC, 85% humidity and 5% CO₂) in Minimum Essential media (MEM, Gibco) supplemented with 10 IU mL^− 1 glutamine (Sigma-Aldrich), 100 IU mL^− 1 penicillin (Biochrom), 100 mg mL^− 1 streptomycin (Biochrom) and 10% foetal bovine serum (FBS; Gibco). PLHC-1 cells (Poeciliopsis lucida hepatocellular carcinoma; ATCC^® CRL-2406) were cultured (30 ºC, 85% humidity and 5% CO₂) in Minimum Essential Medium with Earle’s Balanced Salts (EMEM; Gibco) supplemented with 10 IU mL^− 1 glutamine, 100 IU mL^− 1 penicillin, 100 mg mL^− 1 streptomycin and 5% FBS, 0.1 mM non-essential amino acids and 1.0 mM sodium pyruvate. Finally, SAF-1 cells (S. aurata fibroblast; ECACC^® 00122301) were cultured (25 ºC; 85% humidity) in L-15 Leibowitz medium (Gibco) supplemented with 10 IU mL^− 1 glutamine, 100 IU mL^− 1 penicillin, 100 mg mL^− 1 streptomycin and 5% FBS.

Short chain fatty acids

Three SCFAs, butyric acid (Sigma-Aldrich, BA), acetic acid (Emsure^®, AA), and propionic acid (Sigma-Aldrich, PA) were used alone or together (SUM). SCFAs were diluted in the corresponding medium of the cells to be used in each assay, to make stock solutions. After the addition of the acids to the culture medium, the pH of the media was measured and it was observed that the SCFA did not influence this parameter.

Incubation of cell lines with short chain fatty acids

Once the cell lines were in exponential growing phase (80% confluence), they were detached from the culture flasks by 3 min exposure to trypsin (0.05% in phosphate buffered saline (PBS) for HeLa and 0.25% in PBS for PLHC-1 and SAF-1, pH 7.2–7.4), following standard methods. The detached cells were collected by centrifugation (200 g, 5 min, at the corresponding temperature for each cell line) and the cells were counted using trypan blue (Thermo Fisher Scientific) in a Neubauer chamber (Thermo Fisher Scientific).

For each cell line, 50,000 cells well^− 1 were seeded in 96-well flat bottom culture plates (Nunc) and incubated 24 h at the corresponding temperatures. The culture medium was replaced with the corresponding SCFAs treatment (0, Control, 10, 20, 20, 30, 40 to 50 µM, alone or three SCFAs together) and incubated for 24 h.

Cell viability assays and contrast phase light microscopy

Cell viability was assessed by colorimetric MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma-Aldrich) assay. After incubation with SCFAs, the supernatant was collected and analyzed for cell viability with trypan blue (Sigma-Aldrich). Following this, the cells were washed with PBS and 200 µL well^− 1 of MTT (1 mg mL^− 1) were added. After 4 h of incubation, the cells were washed again and the formazan crystals solubilized with 100 µL well^− 1 of DMSO (dimethyl sulfoxide; Sigma-Aldrich). Plates were shaken (5 min, 100 rpm, room temperature and dark) and absorbance was determined at 560 nm on a microplate reader (FLUOstar Omega). There was one negative control for each concentration of each acid without cells. Cells were also studied with a phase contrast microscope (Leica) and photographed before and 24 h after treatment.

in vivo assay

The experiment was carried out at the Marine Fish Facilities at the University of Murcia (Spain). A total of 30 specimens of the marine teleost S. aurata with a mean weight of 92.6 ± 1.3 g were randomly distributed in ten tanks (250 L, flow rate 900 L h^− 1). Water parameters were 28‰ salinity and 20 ºC temperature. The fish were maintained under artificial photoperiod (12 L: 12D) and quarantined for one month and were fed a commercial pellet diet (Skretting) at a rate of 1.5% body weight day^− 1. Prior to any manipulation, the fish were anesthetized with 50 mg L^− 1 of clove oil (Guinama^®). At the end of the experiment the fish remained alive.

After the quarantine period, fish from each tank were injected intramuscularly, below the lateral line of the left flank of the fish, at the beginning of the dorsal fin tip, with 50 µl of PBS (control), or 50 µl of 20 µM SCFA (BA, AA, PA or the combination of the three acids, at the identical concentration of 20 µM as previously specified when utilized individually, designated as SUM) forming 5 groups, with two tank replications. One hour later, fish were reinjected into the same area of the first injection with 50 µL of λ-carrageenan (1% in PBS; Sigma-Aldrich; Carr), to establish the acute inflammation model previously determined²³. All fish were sampled 3 h after the second injection. Thus, fish were divided into the following groups: PBS + PBS (negative control), PBS + Carr (positive control), BA + Carr, AA + Carr, PA + Carr and SUM + Carr. Skin samples were obtained with a 4-mm diameter biopsy punch (Integra™ Miltex™) around the injection site. One part of this biopsy was preserved in TRIzol reagent (Invitrogen) and used for gene expression studies, and another part was fixed in 10% formalin (Sigma-Aldrich) and used for microscopic analysis.

Gene expression

Total RNA was extracted from the skin using TRIzol Reagent²⁵. The concentration and the purity were evaluated using nanodrop (Thermo Fisher Scientific) before treating with DNase I (Promega) to remove the contamination from genomic DNA. The Complementary DNA (cDNA) was synthesised²⁶.

The expression of genes relevant to the inflammatory response [interleukin 6 (il-6), interleukin 1β (il-1β), tumor necrosis factor α (tnfα), interleukin 10 (il-10), transforming growth factor 1 β (tgf-1β), and nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (nf-κB1) and 2 (nf-κB2)], and cell markers involved in inflammation [neutrophil cytosolic factor 4 (phox40) and colony-stimulating factor 1 receptor (csf1r)] genes was determined by real-time PCR (QuantStudio 5, Applied Biosystems) using SYBR Green qPCR Master Mix (Applied Biosystems)²⁶. The primers used are shown in Table 1. The relative expression of all genes was calculated by the 2^− ΔΔCT method²⁷, using S. aurata 18 S ribosomal RNA (rps18) and elongation factor 1-alpha (ef1α) as the endogenous reference. For normalization of gene expression, it is best to use more than 1 house-keeping gene, preferably from different cellular pathways, to avoid possible small changes in them due to treatments. Although these genes were selected because they have shown stability between treatments, the use of different endogenous genes significantly reduces this variation.

Table 1.

Primers used for RT-PCR analysis .

Gene	Forward primer (5’-3’)	Reverse primer (3’-5’)	GeneBank no.
ef1-α	TGTCATCAAGGCTGTTGAGC	GCACACTTCTTGTTGCTGGA	AF184170
rps18	CGAAAGCATTTGCCAAGAAT	AGTTGGCACCGTTTATGGTC	AM490061
il-6	AGGCAGGAGTTTGAAGCTGA	ATGCTGAAGTTGGTGGAAGG	AM749958
il-1β	GGGCTGAACAACAGCACTCTC	TTAACACTCTCCACCCTCCA	AJ277166
pcna	GAGCAGCTGGGTATTCCAGA	CTGTGGCGGAGAACTTGACT	P12004
tnf-α	TCGTTCAGAGTCTCCTGCAG	TCGCGCTACTCAGAGTCCATG	AJ413189
nf-κB1	CCGACAGACGTTCACAGACA	TCTTCAGCTGGACGAACACC	B005908
nf-κB2	ATCACAGCGCAGAGATCGAG	TGCGGGATGTAGGTGAACTG	B012900
il-10	CTCACATGCAGTCCATCCAG	TGTGATGTCAAACGGTTGCT	FG261948
tgf-1β	GCATGTGGCAGAGATGAAGA	TTCAGCATGATACGGCAGAG	AF424703
phox40	GCGGAGTTGAACCTGAAGAG	TCACCTTCTGTGTCGCTGTC	AM749961
csf1r	ACGTCTGGTCCTATGGCATC	AGTCTGGTTGGGACATCTGG	AM050293

Immunohistochemistry

Skin samples of approximately 2 mm² were fixed in fresh 10% neutral buffered formalin at room temperature for 24 h. Skin samples were decalcified by treatment with 1 mM EDTA (ethylenediaminetetraacetic acid, Sigma-Aldrich) for 72 h and transferred to 70% ethanol. After serial dehydration steps in ethanol, samples were embedded in hydrophilic Paraplast Plus (Thermo Fisher Scientific) following the manufacturer’s recommended methods. Sections were cut at 5 μm with a microtome, mounted on poly-L-lysine-coated glass slides (Sigma-Aldrich), and stained by immunohistochemistry.

Sections were washed for 15 min with PBS-Tween (Sigma-Aldrich), blocked against nonspecific antibody binding with bovine serum albumin (BSA, Sigma-Aldrich) (5% in PBS, 30 min) and washed again (10 min). Samples were incubated (4 °C, 1 h) with a home-produced mouse monoclonal D2 antibody specific for acidophilic granulocytes from gilthead seabream²⁸. The antibody was sourced directly from the hybridoma cell line and utilized as the culture supernatant diluted in PBS-T (1:100), and washed again with PBS-T (15 min). The samples were then incubated (1 h) with α-mIgG-HRP (1:100 in PBS-T, Roche), and washed again with PBS-T (15 min) and Tris-HCl (10 min). Samples were incubated for 20 min with diaminobenzidine (DAB, Sigma-Aldrich), washed with Tris-HCl (10 min) and counterstained with diluted hematoxylin (1:50 in distilled water). Mounting was performed with DPX (Sigma-Aldrich) and the samples were examined with a bright-field optical microscope (Leica Q550IW).

Statistical analyses

To rigorously determine statistical differences across experimental groups (SCFA administrated), our analysis was initiated with a one-way ANOVA, aimed at uncovering significant mean disparities. Preceding this, the Shapiro-Wilk test validated the data’s adherence to normal distribution assumptions, while Levene’s test assessed variance homogeneity, crucial for selecting the appropriate post-hoc test: Tukey’s HSD for equal variances.

A stringent significance level of p < 0.05 was upheld, denoting a 95% confidence interval to ensure the robustness of our findings. The IC₅₀ model, instrumental in evaluating cell viability, demonstrated high fidelity to the experimental data, as evidenced by R² values exceeding 0.9. This comprehensive analysis employed SPSS for Windows^® (version 25.0) and GraphPad Prism for Windows^® (version 9.0), facilitating advanced statistical testing and graphical data representation, thereby underpinning the validity and precision of our conclusions.

Results

in vitro cytotoxicity of short chain fatty acids

The impact of SCFAs on cell culture viability was evaluated in vitro. HeLa and PLHC-1 cells showed a very similar response to SCFAs (Fig. 1a). SCFAs at 10 µM did not alter the viability of HeLa or PLHC-1 cells but higher concentrations were cytotoxic, with PA being the most cytotoxic (Fig. 1a). IC₅₀ values showed that toxicity was in the order of PA > AA > SUM > BA for HeLa cells and PA > BA ≈ SUM ≈ AA for PLHC-1 cells (Fig. 1b).

Fig. 1.

(a) Cell viability (%) of HeLa, PLHC-1 and SAF-1 cells after exposure to different concentrations of the short chain fatty acids butyric acid (BA), acetic acid (AA) or propionic acid (PA), alone or together (SUM), for 24 h. (b) Absolute IC₅₀ values of the cell lines after exposure to the acids for 24 h. Data were adjusted and the R² value was presented. Data are expressed as mean ± SEM (n = 8). Different letters denote significant differences between treatments (p < 0.05).

In marked contrast, the SAF-1 cell line was differentially impacted by culture with SCFAs. Indeed, when incubated with BA at concentrations of 10, 20 or 30 µM, a significant increase in cell viability was observed (p < 0.05), showing a bell-shaped curve. However, incubation of SAF-1 cells with 40–50 µM AA or PA, or 30–50 µM SUM, produced significant cytotoxicity (Fig. 1a). Among the tested SCFAs, SUM showed the lowest IC₅₀ value for SAF-1 cells, whereas PA had the highest IC₅₀ value (Fig. 1b).

These findings on the cytotoxic effects of SCFAs on cell lines were confirmed by direct phase contrast microscopy and trypan blue exclusion test. In the supernatants studied with trypan blue, viable cells were never observed, but only cellular debris. Cells treated with 20 µM SCFAs, such as HeLa and PLHC-1 cells, showed a significant reduction in cell number, which was correlated with less alive cells by trypan blue assay. These cells appeared rounded and detached from the substrate, with a smaller size compared to control cells (Fig. 2). However, no apparent changes were observed in the morphology of SAF-1 cells after incubation with 20 µM SCFAs, as they maintained a similar appearance to control cells (Fig. 2) and no statistically significant changes were obtained between control and SCFAs treated cells after performing the trypan blue assay.

Fig. 2.

Microscopic images of HeLa, PLHC-1 or SAF-1 cells after exposure to 20 µM of the short chain fatty acids butyric acid (BA), acetic acid (AA) or propionic acid (PA), alone or altogether (SUM), for 24 h (Scale bar = 50 μm).

in vitro experiments guide in vivo studies by identifying non-cytotoxic concentrations of BA, AA, and PA in different cell lines, assessing synergistic or antagonistic SCFA combinations, and revealing cell line-specific responses to predict tissue-specific outcomes. Establishing cytotoxicity thresholds ensures the safety of in vivo studies. Consequently, the in vitro results from this work informed the in vivo experiment conducted in this study.

Gene expression of the in vivo inflammatory response

In the in vivo assay, skin samples were obtained directly from the injection site to examine the expression of several genes associated with pro- or anti-inflammatory functions. The expression levels of three proinflammatory genes (il-1β, il-6, and tnfα) were examined. In skin samples from fish injected with PBS + Carr, a significant increase in the expression of il-1β and il-6 genes (p < 0.05) was observed compared to the PBS + PBS group (Fig. 3).

Fig. 3.

Relative expression level of proinflammatory genes [interleukin 1β (il-1β), interleukin 6 (il-6) and tumor necrosis factor α (tnfα)] in the skin of gilthead seabream that were intramuscularly with 50 µl of PBS, butyric acid (20 µM; BA), acetic acid (20 µM; AA), propionic acid (20 µM; PA), or the acid mixture (20 µM; SUM), and one hour later were intramuscularly injected with 50 µl of PBS or carrageenan (1%; Carr), to be sampled 3 h later. The error bars in the columns denote standard error of the means (n = 6). Different letters denote significant differences between treatments (p < 0.05).

However, il-1β gene expression was significantly decreased in skin samples from fish injected with AA + Carr or BA + Carr, whereas it was significantly increased in samples from fish injected with PA + Carr, compared with expression in control samples (p < 0.05) (Fig. 3). As for il-6 expression, significant decreases (p < 0.05) were observed in skin samples from fish injected with AA + Carr or SUM + Carr when compared with those injected with PBS + Carr (Fig. 3). In addition, tnf-α expression in fish skin was significantly increased (p < 0.05) only in fish samples injected with PA + Carr, compared with expression in control samples.

The expression levels of two selected anti-inflammatory genes (il-10 and tgf-β1) were also examined, and their expression increased in PBS + Carr injected fish, although the increases were significant only for il-10 (p < 0.05), respect to the expression observed in PBS + PBS injected fish (Fig. 4). Expression of the il-10 gene in the skin decreased significantly in the BA + Carr group and increased significantly in the PA + Carr group compared with PBS + Carr group (Fig. 4). Comparing the effects of SCFAs with the PBS + Carr injected group, the expression of tgf-β1 was significantly increased (p < 0.05) in samples from PA + Carr or SUM + Carr injected fish respectively, whereas no significant changes (p > 0.05) in the expression of this gene were detected in the AA- or BA- and Carr injected fish, respect to the PBS + Carr group (Fig. 4).

Fig. 4.

Relative expression level of anti-inflammatory genes [interleukin 10 (il-10) and transforming growth factor 1 β (tgf-1β)] in the skin of gilthead seabream that were intramuscularly injected with 50 µl of PBS, butyric acid (20 µM; BA), acetic acid (20 µM; AA), propionic acid (20 µM; PA), or the mixture of the acids (20 µM; SUM), and one hour later were intramuscularly injected with 50 µl of PBS or carrageenan (1%; Carr) to be sampled 3 h later. The error bars in the columns denote the standard error of the means (n = 6). Different letters denote significant differences between diets and asterisks indicate differences between treatments (p < 0.05).

We also included the expression of two transcription factors relevant to the inflammatory response, the nf-κB1 and nf-κB2 genes. Carr injection did not significantly modify their expression (Fig. 5). Compared with the PBS + Carr group, the expression of nf-κB1 was significantly increased (p < 0.05) in PA + Carr and SUM + Carr. However, the expression of nf-κB2, significantly increased in fish injected with BA + Carr, PA + Carr or SUM + Carr (Fig. 5).

Fig. 5.

Relative expression level of transcription factors [nuclear factor kappa B subunit 1 (nf-κB1) and nuclear factor kappa B subunit 2 (nf-κB2)] in the skin of gilthead seabream injected intramuscularly with 50 µl of PBS, butyric acid (20 µM; BA), acetic acid (20 µM; AA), propionic acid (20 µM; PA), or the mixture of the acids (20 µM; SUM), and one hour later were intramuscularly injected with 50 µl of PBS or carrageenan (1%; Carr) to be sampled 3 h later. The error bars in the columns denote standard error of the means (n = 6). Different letters denote significant differences between diets and asterisks indicate differences between treatments (p < 0.05).

Cellular recruitment

The expression of phox40 and csf1r, two genes used as cellular markers of acidophilic granulocytes and macrophages, respectively, was examined. Compared with the control group (PBS + PBS), the expression of both genes was significantly increased (p < 0.05) in the skin of fish injected with PBS + Carr (Fig. 6a). Significant increases (p < 0.05) in the expression of both genes were observed in fish injected with PA + Carr or SUM + Carr (Fig. 6a) compared with fish injected with PBS + Carr, whereas AA or BA did not.

Fig. 6.

(a) Relative expression level of cell marker genes [(neutrophil cytosolic factor 4 (phox40) and colony-stimulating factor 1 receptor (csf1r)] in the skin of gilthead seabream intramuscularly injected with 50 µl of PBS, butyric acid (20 µM; BA), acetic acid (20 µM; AA), propionic acid (20 µM; PA), or the mixture of the acids (20 µM; SUM), and one hour later were intramuscularly injected with 50 µl of PBS or carrageenan (1%; Carr) to be sampled 3 h later. The error bars in the columns denote standard error of the means (n = 6). Different letters denote significant differences between diets and asterisks indicate differences between treatments (p < 0.05). (b) Representative histological Sect. (5 μm) of gilthead seabream skin intramuscularly injected with PBS + PBS, PBS + Carr, AA + Carr, AB + Carr, PA + Carr and SUM + Carr. Staining: immunohistochemical technique with an antibody against membrane protein D2 and Mayer’s haematoxylin. (Scale bar = 10 μm).

Immunohistochemistry was performed on skin samples to detect the presence of acidophilic granulocytes at the injection site. No stained cells were observed in samples from control fish (PBS + PBS injection). However, immunostained cells were present in the samples from fish injected with PBS + Carr, and a large number of immunostained cells were observed in the samples from fish injected with PA + Carr and SUM + Carr (Fig. 6b).

Discussion

SCFAs are metabolic by-products produced by the intestinal microbiota, although they have also been recognized as regulators of the immune system in mammalian skin²⁹. In the absence of studies on the effects of SCFAs in fish, acetic, butyric and propionic acids were selected for this study, as they constitute 95% of the SCFAs present in the human intestine³⁰. These SCFAs exhibit diverse and sometimes opposing functions, such as anti-inflammatory^31,32, pro-inflammatory³³, apoptosis-promoting³⁴, and cell proliferation activating³⁵. Specific functions are hypothesized to be influenced not only by the type of SCFA utilized, but also by their concentration³⁶. At the molecular level, SCFAs exert their functions by activating specific receptors¹³ or transcription factors³⁷. The aim of this study was to investigate the potential preventive use of SCFAs in the regulation of inflammatory responses in the skin of gilthead seabream, a fish species extensively cultivated on the Mediterranean and Atlantic coasts³⁸, To the best of our knowledge, only one in vitro study has reported the effects of different combined SCFAs on selected immune functions of human microglia-like cells, focusing on neuroinflammation³⁹. Consequently, this study aims to contribute to the understanding of the effects of SCFAs on immune responses in fish skin, providing valuable information on their potential applications.

The three selected SCFAs in the present study (butyric, acetic, and propionic acid) can interact in various ways within biological systems, potentially influencing each other’s effects. These interactions can occur at both the metabolic and functional levels. SCFAs can act synergistically to enhance beneficial biological outcomes. Their balance and relative concentrations are key to determining whether their interactions are synergistic, additive, or antagonistic. For example, the combined presence of butyric, acetic, and propionic acids can promote gut health by supporting intestinal barrier function and anti-inflammatory responses more effectively than each SCFA individually. This synergy can lead to improved production of regulatory T cells (Tregs), modulation of immune responses, and enhanced intestinal homeostasis⁴⁰. Furthermore, while SCFAs often work in concert, they could have antagonistic or additive effects depending on the specific biological context. For instance, propionic acid modulates glucose metabolism, while butyric acid is more involved in anti-inflammatory pathways. Besides this, depending on the ratios, one SCFA could dominate in specific pathways, leading to additive or antagonistic interactions in areas like energy metabolism, immune function, or microbiome regulation⁴¹.

SCFAs induce either cell apoptosis or proliferation in cell lines

Numerous assays have been conducted to investigate the influence of SCFAs on cell viability in various cell lines, both cancerous and non-cancerous. SCFA modulate cell viability by regulating the expression of several genes involved in cell survival and cell death pathways^42,43. SCFAs induce cell death through activation of the caspases cascade and mitochondrial apoptotic pathways^34,37,44, whereas they may protect cells from apoptosis through activation of anti-apoptotic pathways such as NF-κB, PI3K/Akt, and ERK1/2^43,45. The effect of SCFAs on cancer cells (such as human lung cancer cell line or HT26 colon cancer cell line) was the induction of apoptosis and cell cycle arrest^12,46. Although the mechanisms involved remain incompletely elucidated, it was demonstrated that SCFAs induce overproduction of reactive oxygen species, leading to mitochondrial membrane dysfunction, and also inhibit NF-κB and AKT/mTOR signaling pathways and induce LC3B protein levels, resulting in autophagy⁴⁷. Conversely, SCFAs could increase cell viability by promoting cell proliferation, migration, and adhesion³⁵, provided that the concentration of the fatty acid does not exceed the energetic needs of the cells, as demonstrated in a study with human colon cells³⁶. In the in vitro part of the present study, the results show that all the studied SCFAs could be cytotoxic, with important differences depending on the cell line. SCFAs induced comparable levels of cytotoxicity in HeLa and PLHC-1 cells, whereas effects on cell proliferation were noted in SAF-1 cells. Furthermore, the most cytotoxic SCFAs for HeLa and PHLC-1 cells was PA, whereas the least cytotoxic was BA. Although HeLa and PLHC-1 cell lines are more deeply characterized than SAF-1, a possible explanation could be that both cell lines present an overexpression of genes such as p53 and genes involved in the MAPK/ERK pathway, with respect to SAF-1, while for this cell line an overexpression of genes such as Myc-associated zinc finger (MAZ) protein has been documented, according to the Harmonizome database⁴⁸. However, many other factors may be involved in the effects of SCFA on cell viability. In the present study, the three cell lines utilized in the in vitro assays possess distinct origins and correspond to different cell types, factors that may contribute to the divergent results obtained. While HeLa cells are from human cervical cancer, PLHC-1 and SAF-1 are both originated from fish. Besides this, the PLHC-1 and SAF-1 cell lines originate from distinct fish species and exhibit disparate characteristics. Briefly, PLHC-1 cells are frequently utilized for investigating xenobiotic metabolism, oxidative stress, and responses to environmental contaminants. These cells demonstrate high metabolic activity and are noted for expressing enzymes such as cytochrome P₄₅₀, rendering them valuable for studying liver-specific responses⁴⁹. Conversely, SAF-1 cells are employed to investigate general cellular processes, including growth, stress responses, and fish immunology⁵⁰. On the other hand, butyrate is known to function as a histone deacetylase inhibitor (HDACi) at high concentrations (> 0.5 mM) in mammalian cancer cells⁵¹. Furthermore, at physiological concentrations, it can induce cell cycle arrest and apoptosis with or without the involvement of the tumour protein p53^51,52. These observations demonstrated in mammals have not yet been established in fish. Notably, it was also BA that induced cell proliferation in the SAF-1 cell line, as BA initiates a survival signal through increased phosphorylation of ERK1/2 in non-cancerous cells, whereas cancer cells exhibit a decrease in p-ERK1/2⁵³.

SCFAs as anti-inflammatory agents

For the in vivo assay, fish were intramuscularly injected with SCFAs. Intramuscular injection offers several advantages over administering a substance via the diet. It allows faster and more controlled absorption, ensuring a precise dose and avoiding variability caused by digestion and gut microbiota, which can alter or degrade the substance. In addition, the intramuscular route reduces the influence of individual factors, such as nutritional status or diet composition, and allows for a more sustained release, ensuring more consistent results. It also eliminates potential dietary interferences, making it easier to assess the direct effects of the administered substance. SCFAs are known to modulate the activation of immune/inflammatory responses⁵⁴. To test the possible anti-inflammatory properties of SCFAs, carrageenan was used to trigger a local acute inflammatory response²³. The high degree of variability in gene expression observed in fish is well-documented, as evidenced by the results obtained for certain genes in the present study. Some research proposed utilizing gene expression variability as a method for investigating gene regulation in mammals⁵⁵. Conducting analogous studies in fish species could potentially elucidate whether similar regulatory mechanisms are present in these organisms.

In this work we found that AA has the highest anti-inflammatory activity because it decreases the expression of il-1b and il-6 genes induced by carrageenan in the skin whilst PA causes the highest increments. This indicates that AA might reduce the inflammation, although PA would provide a more rapid resolution of inflammation. It should be noted that both processes of inflammation and anti-inflammation, a priori contradictory, overlap in time and space. Simultaneous activation of pro- and anti-inflammatory genes highlights the immune system’s delicate balance in regulating inflammation for homeostasis. Pro-inflammatory genes are crucial for swiftly responding to infections or tissue damage by mobilizing defences, while anti-inflammatory genes modulate excessive inflammation to prevent harm to healthy tissues and chronic inflammatory or autoimmune disorders. This balance ensures appropriate threat responses without prolonged damage. Additionally, the immune system often triggers both pathways in response to pathogens or damage signals, allowing adaptive responses suited to the cellular environment and stimulus nature⁵⁶. Further studies would be necessary to demonstrate the causes that provide this anti-inflammatory effect of SCFAs, although there are indications that the mode of action of SCFAs in modulating inflammation is very complicated and may be the result of many actions. For example, in humans, it has been hypothesized that SCFAs produced by skin commensal bacteria may activate anti-inflammatory pathways in the skin, thereby maintaining homeostasis³⁵. Due to the good results obtained, these same authors also suggested that SCFAs could be used as therapeutic agents to reduce inflammatory skin reactions³⁵. Our study fully supports this hypothesis in gilthead seabream skin though further characterization is still deserved.

Other reasons for the anti-inflammatory and leucocyte recruitment effects of SCFAs are related to their ability to interact with leucocytes and endothelial cells through multiple mechanisms, including activation of G protein-coupled receptors (GPCRs) such as GPR41 and GPR43, as well as inhibition of histone deacetylase (HDAC). SCFAs primarily exerts its effects through several specific receptors and signaling pathways that influence various biological processes, particularly in the gut and immune system. Among the main receptors and pathways involved are the GPCRs, which mediate anti-inflammatory and metabolic responses. appetite regulation, gut motility and health, and energy metabolism. SCFAs are absorbed through transporters like monocarboxylate transporters (MCTs). These SCFAs may compete for these transporters and receptors, potentially affecting their individual bioavailability and signaling pathways. For example, higher concentrations of one SCFA might reduce the uptake or effectiveness of the others by saturating these pathways. On the other hand, BA is a potent HDAC inhibitor, while PA exhibits weaker inhibitory effects; both significantly influence gene expression. AA, although less potent, still exerts an impact on epigenetic regulation via HDAC inhibition. BA and PA demonstrate the capacity to inhibit the NF-κB signaling pathway, a crucial regulator of inflammation, whereas AA modulates the NF-κB pathway, thereby attenuating pro-inflammatory cytokine production. Furthermore, BA, AA and PA exhibit the ability to activate Peroxisome Proliferator-Activated Receptor Gamma (PPAR-γ), which plays a role in regulating inflammation. These acids also activate AMP-activated protein kinase (AMPK), which is essential for cellular energy homeostasis and possesses anti-inflammatory properties, thus contributing to the protective effects of butyrate in various tissues^57–59. While these results have been documented in mammals, they have not been confirmed in fish. Additional research is needed to establish whether the mechanisms and receptors associated with SCFA functions are similar across both vertebrate groups.

In this regard, to investigate the leucocytes that might be most involved in the resolution of carrageenan-mediated inflammation in seabream, the expression of several genes that are considered markers of macrophages and granulocytes was also studied. Carr injection induced the expression of acidophil and macrophage markers phox40 and csf1r, respectively in the skin, suggesting their recruitment, but they do not appear to be the maximally responsible for inflammation as nf-κBs did not increase. This fact also points to the involvement of other cells such as dendritic cells among others. The effects of the injection of SCFAs followed by Carr was different depending on the SCFAs tested. Again, PA and SUM produced the largest increases of cells in the inflamed area. In fact, recruitment of acidophilic granulocytes was also confirmed by immunohistochemistry, mainly elicited by PA + Carr. Our team has previously shown that acidophilic granulocytes in gilthead seabream are functionally equivalent to mammalian neutrophils⁶⁰ and that these cells are the first to be recruited to the site of inflammation, followed by macrophages^24,61. In an in vitro study, no effect of sodium butyrate was observed in unstimulated murine macrophages (cell line RAW264), but inhibited LPS-induced macrophage migration, as determined by chemotaxis, using a modified Boyden chamber⁶². SCFAs, such as propionate and butyrate, stimulate monocyte/macrophages and neutrophils recruitment, suggesting a proinflammatory action, although just the opposite effects have been described in mammals⁶³. The present results seem to suggest that SCFAs are not only important in attracting acidophils and macrophages to the inflamed area, but also in promoting their survival, proliferation and differentiation in gilthead seabream. This is also consistent with previous in vitro findings that demonstrated that SCFAs induce directional migration (chemotaxis) of neutrophils⁶⁴. The same authors suggested that this effect might depend on the activation of the G protein-coupled receptor GPR43⁶⁴, which couples to Gi/o and Gq proteins and is expressed on leucocytes, particularly in neutrophils and monocytes⁶⁵.

These results indicated the complexity of the SCFAs’ effects on skin inflammation and highlighted the necessity for further studies to elucidate the effects of SCFAs in fish skin and gut. A limitation of our study is that carrageenan, while a useful model for acute inflammation studies, has constraints in representing chronic, systemic, or complex inflammatory conditions. Its specificity for certain pathways may not fully capture the range of in vivo inflammatory responses.

Conclusions

The present results demonstrate that SCFAs can exhibit cytotoxic or induce proliferation depending on the cell type or line, concentration and specific SCFA. Utilizing an inflamed skin model, AA has demonstrated the most significant direct anti-inflammatory effect, although PA has been shown to play a crucial role in several mechanisms contributing to the resolution of inflammation and recruitment of immune cells to the site of carrageenan-mediated inflammation in gilthead seabream skin. This study elucidates the beneficial effects of SCFAs in promoting the resolution of inflammation and immune cell recruitment in fish skin, which may have implications for the development of novel therapeutic strategies for fish health.

Author contributions

N.A.R.: Methodology, Investigation, Writing – original draft, Writing – review & editing, Visualization and prepare the figures. M.E.Q.: Writing – original draft, Writing – review & editing. A.C.: Validation, Writing – review & editing, Visualization. M.A.E.: Term, Conceptualization, Resources, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition.

Funding

This work was funding by the project MINECO co-funded by the European Regional Development Funds (ERDF/FEDER) (grant no. AGL2017-88370-C3-1-R) and proyecto PID2020- 113637RB-C21 de investigación financiado por MCIN/AEI/10.13039/ 501100011033.

Data and model availability

The data are available on the DIGITUM institutional repository from the University of Murcia: http://hdl.handle.net/10201/132143 (accessed the 20 of June 2023).

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

All experimental protocols were approved by the Ethical Committee of the University of Murcia (protocol code A13150104) following the European Union guidelines for animal handling (2010/63/EU) and the ARRIVE guidelines.