*2.6. Comparative Effects of T. lutea F&M-M36 Extract and FX on mir-146b and mir-223 Expression*

In RAW 264.7 macrophages stimulated with LPS, the expression of mir-146b was significantly enhanced compared to control cells (Figure 6 Panels B), whereas that of mir-223 was strongly reduced (Figure 6 Panels A); both of these effects were counteracted by Celecoxib 3 μM. *T. lutea* F&M-M36 extract, and FX showed similar effects in reducing the expression of mir-146b (*p* < 0.05). On the contrary, the expression of mir-223 was induced in cells treated with *T. lutea* F&M-M36 extract, but this difference did not reach statistical significance.

**Figure 4.** Effect of *T. lutea* F&M-M36 extract and FX on COX-2 protein expression in LPS-stimulated RAW 264.7 cells. Panels (**A**–**D**): COX-2 protein expression determined by immunocytochemistry with an anti-COX-2 antibody (red fluorescence). Nuclei were counterstained with DAPI (blue fluorescence); Magnification = 400×; Scale bar = 20 μm. Panel (**E**): Densitometric analysis of cells positive for COX-2. Panel (**F**): Densitometric analysis of dot blot results on COX-2 protein expression; above bars, representative dot blot images are shown. ### *p* < 0.001 vs. unstimulated RAW 264.7 macrophages (CTRL). \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. LPS. ˆˆ *p* < 0.01 and ˆˆˆ *p* < 0.001 vs. FX by ANOVA test and Dunnett's Multiple Comparison test. Data are expressed as means ± SEM of four replicates.

**Figure 5.** Gene expression profiles of unstimulated RAW 264.7 macrophages (CTRL), RAW 264.7 macrophages stimulated with LPS and those treated with LPS in the presence of *T. lutea* F&M-M36 extract at 100 μg/mL and FX at 470 ng/mL. Each column represents a different treatment and each row a different gene; the color code indicates down-regulation (green) or up-regulation (red) compared to LPS.

**Table 1.** Effect of *T. lutea* F&M-M36 extract at 100 μg/mL and FX at 470 ng/mL on COX-2, iNOS, SOD2, IL-1β, TNF-α, IL-10, IL-6, HO-1, Arg1, and NLRP3 mRNA expression in LPS-stimulated RAW 264.7 cells.


Data are expressed as means ± SEM of four replicates; for each target gene, the relative amount of mRNA was calculated as the ratio to RPLP-1 mRNA [19]; ### *p* < 0.001 vs. CTRL; \*\*\* *p* < 0.001 vs. LPS; ˆ *p* < 0.05 and ˆˆˆ *p* < 0.001 vs. FX by one-way ANOVA and Dunnett's multiple comparisons test.

**Figure 6.** Effect of T. lutea F&M-M36 extract and FX on mir-223 (Panel **A**) and mir-146b (Panel **B**) expression in LPS-stimulated RAW 264.7 cells. ### *p* < 0.001 vs. CTRL; \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 vs. LPS by one-way ANOVA and Dunnett's Multiple Comparisons test.

#### **3. Discussion**

Inflammation is a key component of several chronic human diseases such as inflammatory bowel diseases, diabetes, cardiovascular diseases, neurodegeneration, and cancer [23]. The identification of new anti-inflammatory compounds is a great challenge for the scientific community, and in this context, the microalga *T. lutea* may represent an interesting source for the discovery of novel strategies for the prevention, and even control, of inflammation.

Overall, our results demonstrate that *T. lutea* F&M-M36 methanolic extract and FX, at equivalent concentrations, exert anti-inflammatory activities by regulating a number of pro-inflammatory mediators. It is interesting to highlight that the effects on the COX-2/PGE2 axis are concentration-dependent and therefore suggestive of a pharmacological mechanism of action of *T. lutea* F&M-M36 methanolic extract and FX; the prominent reduction of COX-2/PGE2 exerted by *T. lutea* F&M-M36 methanolic extract also suggests that compounds other than FX may exert additive or synergistic effects. This is also consistent with previous reports documenting the superior activities of botanical extracts compared to single components [24]. *T. lutea* F&M-M36 methanolic extract contains polyphenols equivalent to 6.22 mg of gallic acid/g dry weight, exhibiting a much lower content of total polyphenols compared to that reported for other species, as a polyphenolic content of 515 mg GAE per 100 g DW and of 13.4 mg GAE/g EW measured in an ethanolic extract from the closely related species *I. galbana* [14,24]. These differences may be ascribed to the extraction solvents used (methanol instead of ethanol), although differences in the analyzed species and in cultivation conditions may also have contributed [25].

Despite the presence of phenolic compounds in *T. lutea* being previously described, scarce information is available on their composition; our HPLC characterization showed that *T. lutea* F&M-M36 methanolic extract contains a number of simple phenolic acids which have characteristic UV spectra (maximum absorption in the 200–290 nm range [26,27].

Simple phenolic acids derivatives of hydroxybenzoic and gallic acids have been previously proved to exert anti-inflammatory activities; gallic acid exerted inhibitory effects on

LPS-stimulated PGE2 and IL-6 production and COX-2 expression in RAW 264.7 cells [27], and inhibited several NLRP3 inflammasome markers in an in vitro model of intestinal inflammation [28]. Moreover, we previously demonstrated that hydroxytyrosol, p-coumaric acid, or foods rich in simple phenols exhibited anti-inflammatory properties in in vitro and in vivo models of colon inflammation [18,20,29]. On the other hand, we cannot exclude the contribution of other, not characterized components of our methanolic extract. In particular, our *T. lutea* F&M-M36 biomass contains 4.1% of dry-weight polyunsaturated fatty acids (PUFAs) and 2.61% of total ω-3 [7] that are known to exert immunomodulatory and anti-inflammatory activities [30].

In addition, although FX is the main carotenoid found in *T. lutea*, other compounds such as diadinoxanthin, diatoxanthin, and β-carotene were found in an ethyl acetate extract from *T. lutea* containing a total amount of 132.8 mg of carotenoids/g of extract [31]. The antiinflammatory activities of carotenoids such as β-carotene at relatively high concentrations (50–100 μM) have been reported in LPS-induced RAW264.7, showing effects on IL-1β, IL-6, and TNF-α; [32]. In the same model, other authors found significant effects of β-carotene 5 μM on IL-12, p40, and IL-1β expression [33].MiRNAs are endogenous non-coding RNA molecules that silence target mRNA by binding to the 3 UTR of mRNA [34]. Several miRNAs are regulated during the inflammatory process [35]; mir-223 is emerging as an important regulator of the innate immune system, and its deficiency enhances proinflammatory macrophage activation [36,37]. mir-223 targets NLRP3 result in reduced inflammation [38,39]. Our results pointed out a peculiar superior effect of the *T. lutea* F&M-M36 methanolic extract toward the NLRP3/mir223 axis. For the first time, we showed that *T. lutea* F&M-M36 methanolic extract has the ability to enhance the secretion of mir-223 by LPS-stimulated RAW 264.7, although to a lesser extent than the selective COX-2 inhibitor Celecoxib, and that this effect may be attributable to the phenolic content of the extract, considering the negligible effects of FX alone.

The activity of *T. lutea* F&M-M36 methanolic extract was prominent over that of FX on the COX-2/PGE2 pathway and NLRP3/mir-223 axis, whereas similar effects were observed when other inflammatory mediators were investigated. The ability to simultaneously target different biological inflammatory networks certainly represents an added value of both the extract and FX.

Macrophages polarization between M1 and M2 phenotypes is an important regulatory mechanism for inflammation. M1 macrophages are classically activated by LPS and sustain inflammation, whereas M2 or M2-like phenotypes are associated with the resolution of inflammation [40]. M1 macrophages express pro-inflammatory cytokines such as TNF-α, COX-2, and IL-6, while M2 macrophages express IL-10 and Arg1, thus exhibiting antiinflammatory properties [41].

*T. lutea* F&M-M36 methanolic extract and FX promoted some morphological and molecular characteristics of the M2 anti-inflammatory phenotype in RAW macrophages, such as increased expression of IL-10 and Arg1 and decreased expression of IL-6. The extent of these effects is almost completely attributable to the FX content.

Previous findings indicate that FX (100 μg/mL) inhibited the secretion of IL-1β and TNF-α and promoted that of IL-10 and IFN-γ in Caco-2 cells stimulated with LPS [8]. In LPSinduced RAW 264.7, FX 15-60 μM (corresponding to about 10–40 μg/mL) significantly inhibited NO, TNF-α, and IL-6 production but slightly reduced PGE2 production [10] and inhibited NF-κB activation and MAPK phosphorylation at 12–50 μM [11]. In the same model, the half-maximal inhibitory concentration (IC50) for IL-6 production was 2.19 μM [12]. In a recent report, Kim et al. (2021) [42] found that the pretreatment of RAW 264.7 with FX 5 μM also significantly decreased LPS-induced expression of IL-6, IL-1β, and TNF-α by activating the NRF2/PI3K/AKT pathway. It is worth highlighting that all these studies were conducted with FX concentrations largely greater than ours (470 ng/mL). From a pharmacological point of view, the smaller is the concentration at which the molecule is active, the greater is its potential application. Recently, in a model of metabolic syndrome, a high-fat diet, supplemented with 12% (w/w) of freeze-dried *T. lutea*, significantly reduced plasma TNF-α levels and increased IL-10 in abdominal adipose tissue [43].

In addition, for the first time, we reported the ability of *T. lutea* F&M-M36 methanolic extract to reduce the secretion of mir-146b, and this effect was almost completely attributable to FX [44]. Increased levels of mir-146b are associated with inflammatory disease: in particular, mir-146b is increased in the serum of patients with inflammatory bowel disease and decreases after treatment with infliximab [45]; moreover, circulating mir-146b correlates with endoscopic disease activity in patients with inflammatory bowel disease [46].

*T. lutea* is not approved for human consumption, and its safety has been evaluated only in short-term studies in animal models [2,47,48]. However, *T. lutea* is currently used in aquaculture [1], and our data suggest that it could be added to animal feed not only for its high nutritional value, but also as an anti-inflammatory additive.

Overall, our results demonstrate that *T. lutea* F&M-M36 methanolic extract exerts promising anti-inflammatory activity, even more pronounced than that of FX alone, thus providing the background for conducting studies on its long-term safety and efficacy in inflammatory disease models.

## **4. Materials and Methods**

#### *4.1. Microalgal Biomass*

The biomass of *T. lutea* F&M-M36 strain belonging to the Fotosintetica & Microbiologica (F&M) S.r.l. Culture Collection (Florence, Italy) was produced at Archimede Ricerche S.r.l. (Camporosso, Imperia, Italy). *T. lutea* F&M-M36 was cultivated in F medium [49] in GWP®-II photobioreactors [50] in a semi-batch mode. The lyophilized biomass was stored at −20 ◦C until extraction.

#### *4.2. Microalgal Extract Preparation*

An aliquot of 250 mg of lyophilized *T. lutea* F&M-M36 biomass was extracted in 30 mL of methanol, overnight, at room temperature (RT). The mixture was then sonicated twice for 3 min at the maximum power. The solvent was separated from the biomass by filtration on paper. The residual biomass was extracted again with 15 mL of methanol at 37 ◦C for 4 h; then, the exhausted biomass was removed by filtration on paper, and the extract (30 + 15 = 45 mL) was evaporated under vacuum. The dry residue was solubilized in DMSO to obtain a final concentration of the extract of 65 mg/mL. Fucoxanthin (purity ≥ 95%) was purchased by Sigma Aldrich (Milan, Italy).
