N-Acetylneuraminic acid and 3′-sialyllactose supplementation unlock shared and unique atheroprotective mechanisms through the gut–liver–coronary axis in hypercholesterolemic LDLR^−/− mice

Abstract

Background: Atherosclerosis remains a major global health burden. Sialic acid (N-acetylneuraminic acid, Neu5Ac) and its derivative 3′-sialyllactose (3′-SL) show potential for prevention, but their mechanisms of action are unclear. This study investigated their prophylactic effects and underlying mechanisms in an atherosclerotic model. Methods: Forty male LDLR^−/− mice were randomly assigned to four groups (n = 10 per group): (1) high-cholesterol diet (HCD) control, (2) HCD + Neu5Ac (oral), (3) HCD + 3′-SL (oral) and (4) normal chow control. Ten wild-type C57BL/6J mice served as baseline controls. After 12 weeks of intervention, atherosclerotic plaque formation, serum lipids (total cholesterol (TC), triglycerides (TG), LDL and HDL), inflammatory markers (hs-CRP, IL-1β and TNF-α), hepatic lipid deposition, coronary chemokines (CXCL1 and CCL5), LOX-1, and cytokines (IL-18 and IFN-γ) were assessed. Multi-omics analyses (16S rRNA sequencing for the gut microbiota, hepatic transcriptomics, and metabolomics) were performed. Statistical analysis used one-way or two-way ANOVA followed by Tukey's post-hoc test, with P < 0.05 considered significant. Results: Both Neu5Ac and 3′-SL significantly attenuated atherosclerosis compared to HCD controls. Key findings included: reduced serum inflammation (hs-CRP, IL-1β and TNF-α) and improved lipid profiles (reduced TC, TG, LDL and increase HDL); decreased hepatic lipid deposition; lowered coronary chemokines (CXCL1 and CCL5), LOX-1, and pro-inflammatory cytokines (IL-18 and IFN-γ); and diminished atherosclerotic plaque formation. Multi-omics revealed synergistic protection involving: metabolic remodeling (coordinated regulation of cholesterol, fatty acid, and bile acid metabolism); anti-inflammatory/antioxidant effects (suppression of pro-inflammatory pathways and oxidative stress); and multi-organ protection (enhanced gut barrier integrity/pathogen clearance; upregulated hepatic detoxification/reduced steatosis; stabilized coronary extracellular matrix/endothelial function). It is noteworthy that 3′-SL primarily modulates the gut microbiota and its metabolites, thereby indirectly ameliorating liver and coronary lesions, whereas Neu5Ac could be absorbed directly to improve liver metabolism and exert anti-inflammatory effects. Conclusions: Neu5Ac and 3′-SL exert potent prophylactic effects against HCD-induced atherosclerosis in LDLR^−/− mice. These effects are mediated through synergistic modulation of metabolic, inflammatory, and microbial pathways along the gut–liver–coronary axis. This axis integrates lipid metabolism, inflammation, and oxidative stress responses, highlighting a novel multi-targeted mechanism for atheroprotection. These findings support the potential of sialylated compounds as dietary interventions for atherosclerosis prevention.

---

1. Introduction

Atherosclerosis, a dynamic and multifactorial disease characterized by lipid accumulation, chronic inflammation, and endothelial dysfunction, remains the primary contributor to cardiovascular diseases (CVDs), the leading global cause of mortality.^1,2 The pathogenesis initiates with endothelial dysfunction triggered by risk factors such as hypertension, hyperlipidemia, and diabetes, leading to increased vascular permeability and leukocyte adhesion.^3,4 Subsequent lipid deposition, particularly of oxidized low-density lipoprotein (LDL), drives a self-perpetuating inflammatory cascade.⁵ Macrophage recruitment to lesion sites facilitates foam cell formation and plaque progression, with dynamic phenotypic shifts between pro-inflammatory M1 and reparative M2 macrophages during disease stages.^6,7 Despite advances in lipid-lowering therapies, residual inflammatory risks underscore the need for novel strategies targeting both metabolic and immune pathways.^8,9

Emerging evidence highlights bioactive dietary components as promising candidates for atherosclerosis chemoprevention.¹⁰ Sialic acid (SA; N-acetylneuraminic acid), a nine-carbon sugar ubiquitously expressed on glycoproteins and glycolipids, exhibits multifaceted bioactivities, including antioxidant, anti-inflammatory, and immunomodulatory properties.^11–13 Its derivative, 3′-sialyllactose (3′-SL)—a predominant human milk oligosaccharide—has garnered attention for its roles in gut microbiota modulation, systemic inflammation suppression, and neurodevelopment.^13,14 Previous animal studies conducted by our research team have shown that SA, particularly 3′-SL, can prevent and ameliorate glucose and lipid metabolism disorders in high-fat diet-fed mice.^15,16 It not only reduces the LDL/triglyceride ratio but also increases high-density lipoprotein (HDL) levels. Meanwhile, other cell-based studies have revealed that both sialic acid and 3′-SL alleviate endothelial oxidative stress by scavenging free radicals and enhancing nitric oxide bioavailability, while also inhibiting the uptake of oxidized LDL by macrophages and reducing foam cell formation.^17,18 These findings suggest their potential therapeutic value against atherosclerotic dyslipidemia.

However, the precise mechanisms by which free SA and 3′-SL exert atheroprotection remain incompletely defined, particularly in the context of macrophage polarization, gut–liver–coronary axis crosstalk, and endothelial repair. It is noteworthy that free sialic acid, as a fundamental active unit, can be directly absorbed and exert systemic effects, whereas 3′-SL, by virtue of its specific structure, exhibits stronger microbial regulatory functions within the intestine and may indirectly modulate systemic inflammation and lipid metabolism through microbiota–host interactions. The two complement each other while having distinct emphases in terms of bioavailability pathways and mechanisms of action. Further distinction and mechanistic elucidation can be achieved through integrated multi-omics technologies. Animal models, notably LDLR^−/− and APOE^−/− mice, have been instrumental in deciphering disease mechanisms due to their genetic tractability and ability to replicate human-like atherosclerotic features under high-fat diets.^19,20 These models recapitulate key hallmarks, including endothelial dysfunction, lipid-laden plaque formation, and vascular remodeling, providing robust platforms for therapeutic evaluation. This study employs LDLR^−/− mice to systematically investigate the prophylactic effects of two sialic acid isoforms on atherosclerosis progression. By elucidating their impacts on lipid homeostasis, inflammatory signaling, and plaque stability, our findings aim to advance the development of sialic acid-based therapeutics for CVD management.

2. Materials and methods

2.1 Atherosclerosis-prone mouse models

Male C57BL/6J mice (n = 10) and LDLR^−/− mice (n = 40) on a C57BL/6J background were obtained from Cyagen Biosciences (Suzhou) Co., Ltd. Animals were housed under standardized conditions (22 ± 1 °C, 40–60% relative humidity) with 12/12-hour light–dark cycles and ad libitum access to water. Following a 1-week acclimation period, mice were stratified into two dietary groups: a control group (n = 10 LDLR^−/− C57BL/6J background mice and n = 10 C57BL/6J mice), maintained on standard chow [normal diet (ND), Beijing Keao Xieli Feed Co., Ltd; Beijing Feed Certificate (2018) 0673]; and an experimental group (n = 30 LDLR^−/− mice), fed a high-cholesterol diet (D12108C; Research Diets, Inc.) formulated with 40% fat-derived calories and 1.25% (w/w) cholesterol [Jiangsu Collaborative Pharmaceutical Bioengineering Co., Ltd, Su Feed Certificate (2019) 01008].

All experimental procedures were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of the Laboratory Animal Center of Xia-men University and the International Association of Veterinary Editors guidelines for the Care and Use of Laboratory Animals. The protocols for animal use were reviewed and approved by the Animal Ethical and Welfare Committee of the Laboratory Animal Center of Xiamen University (approval no. XMULAC20240172).

2.2 Supplementation by using two types of sialic acids

N-Acetylneuraminic acid (Neu5Ac; HPLC purity ≥ 98%) and 3′-sialyllactose (3′-SL; HPLC purity ≥ 95%) were obtained from CABIO Biotech (Wuhan) Co., Ltd (Wuhan, China). Both compounds were dissolved in 0.9% sterile saline and freshly prepared daily prior to administration. The solutions were administered via oral gavage (0.1 mL per 10 g body weight) to mice for 10 consecutive weeks, following our pre-established experimental protocol (see Table 1 for the detailed supplementation scheme). The dosage regimen was determined through allometric scaling based on interspecies metabolic rate differences, aligned with published pharmacokinetic parameters for Neu5Ac in murine models.²¹

Table 1 Animal experimental protocol

Mice	Group	N	Test substance	Supplementation dose in mice (mg d^−1 kg^−1)	Gavage concentration (mg mL^−1)
a The dose of the 3′-SL group was converted to the same N-acetylneuraminic acid content as the supplementation dose of N-acetylneuraminic acid based on its N-acetylneuraminic acid specific gravity in 3′-sialyllactose.
C56/BL6J	Normal diet (WT)	10	Normal saline	—	—
C56/BL6J	Normal diet (ND)	10	Normal saline	—	—
LDLR^−/−
C56/BL6J	High fat diet group (HCD)	10	Normal saline	—	—
LDLR^−/−
C56/BL6J	Neu5Ac	10	N-Acetylneuraminic acid	40	4.00
LDLR^−/−
C56/BL6J	3′-SL	10	3′-Sialyllactose	81^a	8.10
LDLR^−/−

---

2.3 Body measurements and blood glucose

Body weight and food intake were monitored weekly throughout the study. Fasting blood glucose (FBG) levels were measured at weeks 0, 4, 8, and 10 of the intervention. Following the 10-week experimental period, mice underwent a 12-hour fasting period and subsequently received an oral glucose tolerance test (OGTT) via gavage administration of 20% glucose solution at a dose of 10 μL per g body weight. Blood glucose concentrations were dynamically monitored using a glucometer (Sanuo Biological Sensing Co., Ltd, China) through tail vein sampling at 0, 30, 60, and 120 minutes post-glucose administration.

2.4 Sample collecting and index testing

Following 10 weeks of experimental intervention, mice were subjected to a 12-hour fasting period prior to terminal procedures. Anesthesia was induced via inhalation of 4% isoflurane (Shenzhen Reward Life Technology Co., Ltd, China) followed by euthanasia through cervical dislocation. Cardiac blood samples were obtained through retro-orbital plexus puncture. Coronary artery tissues, hepatic specimens, and colonic mucosa were immediately excised and rinsed with ice-cold physiological saline (0.9% NaCl). The body fat (peritesticular and perirenal fat) and liver tissues were rapidly collected, rinsed with normal saline, drained, and weighed using analytical balances (±0.1 mg precision). Blood samples were centrifuged at 2000 rpm (382g, 4 °C) for 15 min to isolate the serum, which was subsequently aliquoted for biochemical analyses. All collected samples (tissues and serum) were snap-frozen in liquid nitrogen and stored at −80 °C until further processing.

Serum lipid profiles—including total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C)—were quantified using an automated biochemistry analyzer (Mindray BS-220) with manufacturer-calibrated reagent kits. Inflammatory biomarkers (hs-CRP, IL-1β, IL-6 and TNF-α) and oxidized LDL (ox-LDL) levels were determined in the serum, while coronary tissue homogenates were analyzed for chemokine/cytokine expression (CCL5, CXCL1, IL-10, IL-18 and IFN-γ) using species-specific ELISA kits (Wuhan Beinle Biotechnology Co., Ltd, China) following standardized protocols.

2.5 Oil Red O and H&E staining in coronary and liver tissues

(1) Gross Oil Red staining of coronary tissues: target blood vessels were isolated, peripheral adipose tissues were removed and fixed in a fixative (>24 h), and washed twice with PBS. Residual fat was removed, and vessels were longitudinally dissected using scissors. Vessels were rinsed briefly in tap water (5 s), immersed sequentially in 60% isopropyl alcohol (3 s) and Oil Red O staining solution (Solarbio Life Sciences, Beijing, China) and stained at 37 °C (in darkness, for 60 min). They were differentiated in 60% isopropyl alcohol until plaques appeared orange-red/red (other areas colorless), and were then rinsed with distilled water. Vessels were blotted dry, placed on a scaled background plate, and photographed under optimized focus/exposure with scale inclusion.

(2) The coronary and liver tissue samples were fixed using 4% paraformaldehyde, paraffin-embedded, and cut into sections. Oil Red Oxygen dyeing was performed using the modified Oil Red Oxygen staining kit (Solarbio Life Sciences, Beijing, China), following the instructions provided by the manufacturer. Hematoxylin–eosin (HE) staining was performed using an HE staining kit (Solarbio); stains were then assessed via microscopy (Leica-DM4B, Germany).

2.6 16S rDNA sequencing

The colon contents were frozen in liquid nitrogen immediately after sampling and stored at −80 °C. A FastPure Stool DNA Isolation Kit (MJYH, Shanghai, China) was used to extract the total DNA from these samples, after which the conserved 16S rDNA region (V3: 341F, CCTACGGGNGGCWGCAG; V4: 806F, GGACTACHVGGGTATCTAAT) was PCR amplified using appropriate primers and barcodes. The PCR product was extracted from 2% agarose gel and purified using the PCR Clean-Up Kit (YuHua, Shanghai, China) according to the manufacturer's instructions and quantified using Qubit 4.0 (Thermo Fisher Scientific, USA). After demultiplexing, sequences were quality-filtered (fastp v0.19.6) and merged (FLASH v1.2.11). High-quality reads were denoised into amplicon sequence variants (ASVs) using DADA2 in QIIME2 (2020.2). Sequences were rarefied to 20 000 per sample (Good's coverage: 97.90%). Taxonomic annotation was performed against SILVA v138 via QIIME2's classifier, while metagenomic functions were predicted using PICRUSt2 through HMMER alignment, EPA-NG/Gappa phylogenetic placement, castor copy normalization, and MinPath pathway mapping, following the standard workflows. Following OTU determination, we assessed the gut microbiota indices, including community composition, diversity and intestinal flora function, for which we used the real-time online Majorbio cloud platform (https://cloud.majorbio.com) for data analysis.

2.7 Transcriptomic profiling of liver, colonic, and coronary tissues

Total RNA was extracted from the tissue using TRIzol® reagent according to the manufacturer's instructions. Then, RNA quality was determined by using a 5300 Bioanalyser (Agilent) and quantified using an ND-2000 (Nano Drop Technologies). Only a high-quality RNA sample (OD_260/280 = 1.8–2.2, OD_260/230 ≥ 2.0, RQN ≥ 6.5, 28S : 18S ≥ 1.0, >1 μg) was used to construct the sequencing library. RNA purification, reverse transcription, library construction and sequencing were performed at Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd (Shanghai, China) according to the manufacturer's instructions. After quantification by Qubit4.0, sequencing libraries were constructed and subsequently sequenced on a DNBSEQ-T7 platform (PE150) with the DNBSEQ-T7RS High-Throughput Sequencing Kit (FCL PE150) v3.0. Clean reads were aligned to the reference genome using HISAT2, followed by a transcript assembly with StringTie. Differential expression analysis was performed with DESeq2/DEGseq based on RSEM-calculated TPM values (filtering criteria: |log2FC| ≥ 1 with FDR < 0.05 for DESeq2 or FDR < 0.001 for DEGseq). Significantly enriched KEGG pathways of DEGs were identified (Bonferroni-corrected P-value < 0.05).

2.8 Metabolomic profiling of the liver tissue and colonic luminal content

The preprocessing of the mouse tissue sample was performed according to a previous study.²² LC-MS/MS analysis of the sample was conducted on a Thermo UHPLC-Q Exactive HF-X system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Waters, USA) at Majorbio Bio-Pharm Technology Co. Ltd (Shanghai, China). The mobile phases consisted of 0.1% formic acid in water : acetonitrile (95 : 5, v/v) (solvent A) and 0.1% formic acid in acetonitrile : isopropanol : water (47.5 : 47.5, v/v) (solvent B). The flow rate was 0.40 mL min⁻¹ and the column temperature was 40 °C. The pretreatment of LC/MS raw data was performed by using Progenesis QI (Waters Corporation, Milford, USA) software, and a three-dimensional data matrix in CSV format was exported. The metabolites were identified by searching the database, and the main databases were the HMDB (https://www.hmdb.ca/), Metlin (https://metlin.scripps.edu/) and Majorbio Databases. The data matrix obtained by searching the database was uploaded to the Majorbio cloud platform (https://cloud.majorbio.com) for data analysis. The data matrix obtained by searching the database was uploaded to the Majorbio cloud platform (https://cloud.majorbio.com) for data analysis.

2.9 Quantitative real-time PCR

Total RNA was isolated from cells or tissues using the TRIzol method (Invitrogen TRIzol, cat#15596026). First-strand cDNA was synthesized using a cDNA synthesis kit (Thermo Fisher Scientific™, cat# K1641). Quantitative PCR was performed using the Fast SYBR™ Green Master Mix (KAPA Biosystems, cat# KM4101). The results were analyzed on an ABI StepOnePlus real-time PCR system (Bio-Rad, USA, CFX-Connect 96), using the 2^−ΔΔCt method as described previously.²³ Primers were designed using Primer Premier 5 and synthesized by Genecreate Biotech (Wuhan, China); they are listed in Table S1. GAPDH was used as a loading control, and mRNA levels were normalized relative to GAPDH levels.

2.10 Statistical analysis

All continuous variables underwent normality assessment using Shapiro–Wilk tests (α = 0.05) and variance homogeneity evaluation via Levene's test prior to analysis. Data meeting both normality (P > 0.05) and homoscedasticity assumptions were analyzed using one-way ANOVA with Fisher's LSD post hoc comparisons for intergroup differences. For normally distributed datasets exhibiting heteroscedasticity (Levene's P < 0.05), between-group comparisons were performed using Welch's ANOVA followed by Dunnett's T3 multiple comparison tests. Non-parametric datasets were subjected to Kruskal–Wallis H tests with Nemenyi–Damico–Wolfe–Dunn post hoc analysis for omnibus and pairwise comparisons, respectively.

Multivariate repeated measures analysis was conducted using linear mixed-effects models with Kenward–Roger approximation for degrees of freedom. All statistical inferences were made at α = 0.05 (two-tailed), with adjusted p-values reported for multiple comparisons. Analyses were performed using validated procedures in SPSS 22.0 (IBM Corp.).

3. Results

3.1 Weight, visceral fat and serum inflammatory cytokine levels

Compared with the HCD group, under the conditions of comparable dietary energy intake (no statistically significant difference), mice supplemented with 3′-SL showed a more pronounced weight gain trajectory during the intervention period, culminating in a marginally higher final body weight than the HCD group at the 12-week end point (29.95 ± 1.59 g vs. HCD 28.92 ± 1.50 g, P < 0.05). In contrast, the Neu5Ac-supplemented group exhibited a weight progression comparable to that of the HCD group, with no statistically significant difference in the terminal body weight relative to the HCD controls. Furthermore, no significant difference in the final body weight was observed between the 3′-SL and Neu5Ac supplementation groups (P > 0.05) (Fig. 1A and B). Neither perirenal fat nor peritesticular fat differed significantly among the 3′-SL, Neu5Ac, and HCD groups (P > 0.05 for all comparisons) (Fig. 1C and D).

Fig. 1 Impact of Neu5Ac and 3′-SL supplementation for 12 weeks on metabolic parameters and serum inflammatory markers in LDLR^−/− mice fed a high-cholesterol diet (HCD). (A–D) Metabolic parameters: (A) changes in body weight from weeks 0 to 12 of the intervention. Note: * denotes statistically significant differences between groups; (B) average daily food energy intake per mouse; (C) peritesticular fat weight; (D) perirenal fat weight; (E–H) serum inflammatory markers: (E) high-sensitivity C-reactive protein (Hs-CRP); (F) interleukin-1β (IL-1β); (G) tumor necrosis factor-α (TNF-α); and (H) interleukin-6 (IL-6). The difference between values with completely different superscripts was statistically significant, P < 0.05. Data are means ± SD (n = 10).

Compared to the HCD group, 3′-SL supplementation and Neu5Ac supplementation could decrease serum inflammation (hs-CRP, IL-1β, IL-1 and TNF-α) at the end of the 12-week intervention, all P < 0.05. There were no significant differences between 3′-SL supplementation and Neu5Ac supplementation (Fig. 1E–H).

3.2 Gross Oil Red O staining, HE staining and key atherosclerotic metrics of coronary artery tissues

Histopathological evaluation via Oil Red O and H&E staining showed that both Neu5Ac and 3′-SL supplementation significantly suppressed coronary lipid deposition (P < 0.05) and mitigated atherosclerotic lesion severity compared to the high-cholesterol diet (HCD) group (decreased by 54.50–68.89% and 57.93–66.27%, respectively) (Fig. 2A–C). Mechanistically, the intervention groups exhibited marked downregulation of chemokines (CXCL1 and CCL5) and LOX-1 receptor protein expression in coronary tissues (P < 0.05). Notably, the protective effects were mediated through modulation of the local inflammatory microenvironment: pro-inflammatory cytokines (IL-18 and IFN-γ) were significantly reduced (P < 0.05), while anti-inflammatory IL-10 levels were elevated (P < 0.05). Collectively, these findings indicate that Neu5Ac and 3′-SL exerted synergistic atheroprotective effects by multi-target regulation of lipid metabolism and inflammatory responses, thereby attenuating atherosclerosis progression (Fig. 2D–I).

Fig. 2 Histological assessment of aortic atherosclerosis (Oil Red O and HE staining) and coronary artery atherosclerotic metrics after 12-weeks of Neu5Ac/3′-SL supplementation in LDLR^−/− mice fed a high-cholesterol diet (HCD). (A–C) Histological assessment of aortic atherosclerosis (n = 4): (A) gross Oil Red O staining of aortic atherosclerotic lesions; (B) Oil Red O-stained relative lesion area in coronary arteries; (C) histological analysis of aortic tissues by gross Oil Red O staining; (D–I) key atherosclerotic metrics in coronary artery tissues: (D) C–C motif chemokine ligand 5 (CCL5); (E) C–X–C motif chemokine ligand 1 (CXCL1); (F) interleukin-10 (IL-10); (G) interleukin-18 (IL-18); (H) interferon gamma (IFN-γ); and (I) lectin-type oxidized LDL receptor 1 (LOX1). The difference between values with completely different superscripts was statistically significant, P < 0.05. Data are means ± SD (n = 6).

3.3 Oil Red O staining, HE staining of liver tissues and glycolipid metabolism

Compared to the HCD group, neither Neu5Ac nor 3′-SL supplementation significantly reduced liver weight (P > 0.05). However, histopathological analysis via H&E and Oil Red O staining revealed a marked attenuation of hepatic lipid accumulation in both intervention groups (decreased by 35.03–47.91% and 51.47–53.09%, respectively) (P < 0.05), accompanied by preserved hepatic architecture and improved lipid homeostasis (Fig. 3A–D). Notably, the interventions significantly ameliorated systemic dyslipidemia, as evidenced by reduced serum TC, TG, and LDL-C levels (P < 0.05), alongside elevated HDL-C (P < 0.05) (Fig. 3G–J). In contrast, no significant effects on blood glucose regulation were observed (P > 0.05) (Fig. 3E and F).

Fig. 3 Histological assessment of liver tissue (Oil Red O and HE staining) and glycolipid metabolism after 12-weeks of Neu5Ac/3′-SL supplementation in LDLR^−/− mice fed a high-cholesterol diet (HCD). (A–C) Histological assessment of liver tissue (n = 4): (A) hepatic histopathology by HE staining; (B) hepatic lipid deposition visualized by Oil Red O staining; (C) histological analysis of liver tissues by Oil Red O staining; (D) Liver tissue weight; (E and F) glucose metabolism: (E) fasting blood glucose (FBG); (F) oral glucose tolerance tests (OGTT); (G–J) lipid metabolism: (G) low-density lipoprotein (LDL); (H) high-density lipoprotein (HDL); (I) total cholesterol (TC); and (J) triglycerides (TG). The difference between values with completely different superscripts was statistically significant, P < 0.05. Data are means ± SD (n = 10).

3.4 Gut microbiota

Compared to the HCD group, both Neu5Ac and 3′-SL supplementation could change the gut microbiota (P < 0.05) (Fig. 4A), although no statistically significant difference in β-diversity (ACE index) was observed (P > 0.05) (Fig. 4B). Neu5Ac supplementation revealed five signature microbes, while 3′-SL supplementation identified eight characteristic bacterial taxa (Fig. 4C). At the phylum level, both interventions increased the relative abundance of Verrucomicrobiota, Bacteroidota, and Desulfobacterota, while 3′-SL specifically reduced Firmicutes (P < 0.05). Functional profiling revealed divergent metabolic impacts: 3′-SL downregulated microbial pathways linked to energy metabolism, lipid metabolism, and lipopolysaccharide (LPS) biosynthesis, whereas Neu5Ac upregulated pathways associated with energy metabolism, sphingolipid metabolism, and glycolysis (P < 0.05) (Fig. 4F–H).

Fig. 4 Effects of Neu5Ac and 3′-SL supplementation for 12 weeks on the gut microbiota in LDLR^−/− mice (n = 6) fed a high-cholesterol diet (HCD). (A) PCoA analysis on the amplicon sequence variant (ASV) level of group sample relationships; (B–E) species composition and diversity: (B) the abundance-based coverage estimator (ACE) index of the AVS level; (C) Venn diagram showing shared and unique ASV among groups; (D) relative abundance at the phylum level; (E) relative abundance at the genus level; (F–H) functional analysis of flora: (F) B-class classification function of colonic flora based on the reference sequence of PICRUSt2; (G) MetaCyc pathway enrichment analysis based on PICRUSt2 inference; (H) KEGG Orthology (KO) profiles of colonic flora predicted by PICRUSt2; (I–N) the correlation of serum lipids, inflammation markers and intestinal flora: (I) Mantel-test network heatmap analysis of correlations between serum Hs-CRP, IL-10, IL-18, IL-6, and TNF-α with intestinal flora; (J) Mantel-test network heatmap analysis of correlations between serum TC, TG, LDL and HDL with intestinal flora; (K) Mantel-test network heatmap analysis of correlations between serum Hs-CRP, IL-10, IL-18, IL-6, TNF-α, TC, TG, LDL and HDL with the intestinal flora. (L) Pearson's correlation heatmap serum Hs-CRP, IL-10, IL-18, IL-6, TNF-α, TC, TG, LDL and HDL with the intestinal flora at the phylum level; (M) Pearson's correlation heatmap serum Hs-CRP, IL-10, IL-18, IL-6, TNF-α, TC, TG, LDL and HDL with the intestinal flora at the genus level; and (N) Kruskal–Wallis H test bar plot showing the difference in abundance at the genus level. Note: * denotes statistically significant differences between the groups, *0.01 < P ≤ 0.05, **0.001 < P ≤ 0.01, ***P ≤ 0.001.

Notably, correlation analysis identified Bacteroidota enrichment as positively associated with elevated HDL levels. At the genus level (Fig. 4E), both interventions increased Akkermansia and norank_f__Muribaculaceae while reducing unclassified_f__Atopobiaceae and Faecalibaculum (P < 0.05). Additionally, 3′-SL uniquely decreased Lactobacillus (P < 0.05), and Neu5Ac uniquely enriched Lactobacillus and *norank_o__Clostridia_UCG-014 (P < 0.05). Critically, decreased norank_f__Muribaculaceae abundance correlated negatively with anti-inflammatory IL-10, whereas enriched Lactobacillus and *norank_o__Clostridia_UCG-014* were negatively associated with serum TC, LDL, and pro-inflammatory markers (hs-CRP and IL-6) (P < 0.05), but positively correlated with IL-10 (P < 0.05) (Fig. 4L–N). Systemic inflammation markers (hs-CRP, IL-18, IL-6 and TNF-α) exhibited strong positive correlations with lipid profiles (TC, TG and LDL) (P < 0.05). These findings underscore that Neu5Ac and 3′-SL synergistically attenuate dyslipidemia and inflammation through differential modulation of the gut microbiota structure, metabolic function, and host–microbiota crosstalk (Fig. 4I–K).

3.5 Transcriptomic analyses of coronary artery, liver and colon tissues

3.5.1 Colonic tissue regulation.
Compared to the high-cholesterol diet (HCD) group. Neu5Ac supplementation significantly upregulated 56 genes and downregulated 41 genes (Fig. 5A–C). It enhanced anti-inflammatory and immunomodulatory responses (MAPK signaling, inflammatory mediator regulation of TRP channels and cytokine–cytokine receptor interactions) (Fig. 5D), which could contribute to elevating anti-inflammatory cytokines (e.g., IL-10) while reducing pro-inflammatory factors (e.g., IL-1β and IL-6). Additionally, it improved lipid metabolism (glycerophospholipid metabolism, lipid and atherosclerosis pathways) (Fig. 5D), which could contribute to increasing HDL levels and decreasing LDL-C and TG. Furthermore, Neu5Ac strengthened intestinal barrier integrity (Fc gamma R-mediated phagocytosis and protein processing in the endoplasmic reticulum) (Fig. 5D), preserving epithelial cell homeostasis.

Fig. 5 Comparative transcriptomic analysis of coronary artery, colon, and liver tissues in LDLR^−/− mice fed a high-cholesterol diet (HCD) and supplemented with 3′-SL or Neu5Ac (n = 6 per group). (A–E) Transcriptome-wide profiling of differentially expressed genes (DEGs) and functional enrichment in colonic tissue: (A) bar chart of multiple group differences in transcript expression; (B) PCA analysis of group sample relationships; (C) Venn diagram showing shared and unique DEGs among the groups; (D) enrichment difference bubble plot: top 20 KEGG pathways enriched with differential transcript expression between the Neu5Ac supplementation group and the HCD group; and (E) enrichment difference bubble plot: top 20 KEGG pathways enriched with differential transcript expression between the 3′-SL supplementation group and the HCD group. (F–J) Transcriptome-wide profiling of differentially expressed genes (DEGs) and functional enrichment in liver tissue: (F) bar chart of multiple group differences in the transcript expression; (G) PCA analysis of group sample relationships; (H) Venn diagram showing shared and unique DEGs among the groups; (I) enrichment difference bubble plot: Top 20 KEGG pathways enriched with differential transcript expression between the Neu5Ac supplementation group and the HCD group; and (J) enrichment difference bubble plot: top 20 KEGG pathways enriched with differential transcript expression between the 3′-SL supplementation group and the HCD group. (K–O) Transcriptome-wide profiling of differentially expressed genes (DEGs) and functional enrichment in coronary artery tissue: (K) bar chart of multiple group differences in the transcript expression; (L) PCA analysis of group sample relationships; (M) Venn diagram showing shared and unique DEGs among the groups; (N) enrichment difference bubble plot: top 20 KEGG pathways enriched with differential transcript expression between the Neu5Ac supplementation group and the HCD group; and (O) enrichment difference bubble plot: top 20 KEGG pathways enriched with differential transcript expression between the 3′-SL supplementation group and the HCD group. Note#: CP means cellular processes, EIP means environmental information processing, GIP means genetic information processing, HD means human diseases, M means metabolism; and OS means organismal systems.

3′-SL supplementation upregulated 64 genes and downregulated 24 genes (Fig. 5D). It suppressed hyperactive immune responses (complement and coagulation cascades, IL-17 signaling, Th17 differentiation, and neutrophil extracellular trap formation) (Fig. 5E), which could contribute to reducing IL-6 and TNF-α release. It also ameliorated oxidative stress and metabolic dysregulation (selenocompound metabolism and lipid and atherosclerosis pathways) and enhanced pathogen clearance (antigen processing and presentation) (Fig. 5E).

Shared effects. PCA revealed close clustering between Neu5Ac and 3′-SL groups, with 13 overlapping differentially expressed genes (DEGs) (Fig. 5B and C), and had common KEGG pathways (selenocompound metabolism, lipid and atherosclerosis pathways, and protein processing in the endoplasmic reticulum) (Fig. 5D and E), indicating synergistic improvements in colonic lipid metabolism and oxidative stress.

3.5.2 Hepatic metabolic protection.
Compared to the high-cholesterol diet (HCD) group. Neu5Ac supplementation markedly regulated 164 upregulated and 1034 downregulated genes (Fig. 5F–H). It activated lipid catabolism (cholesterol metabolism, unsaturated fatty acid biosynthesis, fatty acid degradation and PPAR signaling) to promote β-oxidation, reducing hepatic triglyceride (TG) accumulation (Fig. 5I). Concurrently, it alleviated oxidative stress (ROS-related chemical carcinogenesis and oxidative phosphorylation) to protect mitochondrial function and modulated energy metabolism (thermogenesis and propanoate metabolism) (Fig. 5I), which could indirectly mitigate cardiac oxidative damage.

3′-SL supplementation modulated 14 upregulated and 52 downregulated genes (Fig. 5F–H). It improved lipid homeostasis (cholesterol metabolism, glycerophospholipid metabolism and taurine metabolism) (Fig. 5J), suppressed inflammatory and oxidative injury (arachidonic acid metabolism, MAPK signaling and complement cascades) (Fig. 5J), and reduced hepatocyte apoptosis and cholestasis.

Synergistic effects. PCA showed proximity between Neu5Ac and 3′-SL groups, sharing 68 core DEGs (Fig. 5G and H). Both interventions reduced systemic lipotoxicity (biosynthesis of unsaturated fatty acids, fatty acid degradation, PPAR signaling pathway, cholesterol metabolism and fatty acid elongation) (Fig. 5I and J), which could ameliorate hepatic steatosis and inflammation (see the histopathological analysis results).

3.5.3 Coronary protection against atherosclerosis.
Compared to the high-cholesterol diet (HCD) group. Neu5Ac supplementation upregulated 94 genes and downregulated 578 genes (Fig. 5K–M). It optimized lipid metabolism (cholesterol metabolism, primary bile acid biosynthesis and linoleic acid metabolism) to reduce plaque lipid cores (Fig. 5N), inhibited inflammatory cascades (complement activation and tryptophan metabolism), attenuated oxidative DNA damage (ascorbate metabolism and DNA adduct repair), and enhanced endothelial function (arginine biosynthesis and glycine metabolism) to promote NO-mediated vasodilation (Fig. 5N).

3′-SL supplementation upregulated 285 genes and downregulated 375 genes (Fig. 5K–M). It stabilized plaque structure (ECM–receptor interactions and focal adhesion pathways), suppressed inflammatory-oxidative crosstalk (PI3K–Akt signaling and retinol metabolism), and improved xenobiotic detoxification (cytochrome P450 pathways) (Fig. 5O).

Common mechanisms. Both groups shared 132 overlapping DEGs, with PCA showing tight clustering (Fig. 5L and M). These findings highlight their synergistic roles in attenuating atherosclerosis (AS) progression by inhibiting inflammatory infiltration, and restoring endothelial homeostasis.

Further analysis of the study results found that supplementation with Neu5Ac and 3′-SL shows significant synergistic regulation of signaling pathways in the colon, liver, and coronary arteries. Under both interventions, the top 20 significantly enriched pathways (TOP20) in the colon were consistently highly enriched in the liver. Furthermore, a higher degree of overlap was observed between the TOP20 pathways responsive in the liver and coronary arteries—a pattern particularly pronounced in the 3′-SL intervention group (Fig. S1). It is noteworthy that Neu5Ac supplementation enriches the inflammatory mediator-regulated TRP channel signaling pathway across colon, hepatic, and coronary artery tissues, while 3′-SL supplementation enriches complement and coagulation cascade signaling pathways across colon, hepatic, and coronary artery tissues. These findings indicate that Neu5Ac and 3′-SL may coordinately modulate core pathways through the gut–liver–coronary axis, thereby exerting anti-atherosclerotic (AS) effects.

3.6 Metabolomic analyses of the liver and colonic contexts

KEGG enrichment analysis revealed that compared to the HCD group, supplementation with Neu5Ac and 3′-SL improved colonic lipid metabolism, oxidative stress, and immune protection-related signaling pathways, as evidenced by the consistent enrichment patterns of both differential metabolites and differentially expressed genes (Fig. 6A and B). Regarding alterations in colonic metabolites (Fig. 6C), the intervention markedly modulated colonic metabolite profiles through multi-pathway coordination: (1) decreased levels of vitamins (thiamine), glycerophospholipids (PS, PE), and sterols (4A-methylzymosterol-4-carboxylic acid, 24-methylenecholesterol) potentially attenuated pathogen overgrowth, oxidative stress, and dysregulated lipid metabolism, thereby contributing to anti-inflammatory responses and epithelial barrier integrity; and (2) elevated levels of steroids/hormones (estrone and stigmasterol), lipid metabolites (diacylglycerol, lysophosphatidylcholine, phosphatidylglycerol and CDP-diacylglycerol), bile acid derivatives (taurochenodeoxycholic acid and withanolide B), antioxidants (α-tocopherol acetate), and energy metabolism intermediates (dihydroxyacetone phosphate acyl ester) synergistically enhanced immune regulation, antioxidant defense, and membrane stability, facilitating the gut microbiota equilibrium. This metabolic reprogramming suggests that the intervention may restore colonic microenvironmental homeostasis via the “microbiota–metabolite–host” axis.

Fig. 6 Comparative metabolomic analysis of colonic content and liver tissues in LDLR^−/− mice fed a high-cholesterol diet (HCD) and supplemented with 3′-SL or Neu5Ac (n = 6 per group). (A–C) Differential metabolites and functional enrichment in the colonic content: (A) differential abundance score: top 20 KEGG pathways enriched with differential metabolites between the Neu5Ac supplementation group and the HCD group; (B) differential abundance score: top 20 KEGG pathways enriched with differential metabolites between the 3′-SL supplementation group and the HCD group; (C) heatmap: differential metabolites of the top 20 KEGG pathways (Neu5Ac_Colon_vs_HCD_Colon and 3′-SL_Colon_vs_HCD_Colon); (D–F) differential metabolites and functional enrichment in liver tissue: (D) differential abundance score: top 20 KEGG pathways enriched with differential metabolites between the Neu5Ac supplementation group and the HCD group; (E) differential abundance score: top 20 KEGG pathways enriched with differential metabolites between the 3′-SL supplementation group and the HCD group; and (F) heatmap: differential metabolites of the top 20 KEGG pathways (Neu5Ac_Liver_vs_HCD_Liver and 3′-SL_Liver_vs_HCD_Liver).

KEGG enrichment analysis showed that compared to the HCD group, supplementation with Neu5Ac and 3′-SL ameliorated hepatic lipid metabolism, oxidative stress, inflammation and immunity, as well as energy metabolism homeostasis-related signaling pathways, with consistent co-enrichment patterns of both differential metabolites and genes in these pathways (Fig. 6D and E). Regarding the hepatic metabolic dynamics (Fig. 6F), hepatic metabolomic profiling revealed multi-dimensional functional improvements: (1) increased levels of carbohydrate metabolism intermediates (fructose 6-phosphate, glucose 1-phosphate and trehalose), amino acid derivatives (L-asparagine, L-proline, L-tryptophan and L-glutamine), bile acid-related metabolites (glycocholic acid and glycerophosphoric acid), and nucleotide derivatives (orotidine and deoxyinosine) supported hepatocyte energy storage, protein synthesis, and detoxification; (2) reduced levels of inflammatory mediators (leukotriene C4), oxidative stress markers (2-ketoglutaric acid, NADP⁺ and kynurenic acid), and dysregulated lipid metabolites (folic acid and taurine) likely enhanced immune cell activity by suppressing mitochondrial oxidative damage; and (3) notably, the 3′-SL intervention increased levels of oligosaccharides and polysaccharides (e.g., galactinol, stachyose, melibiose and pimelibiose). Galactose, a structural component of these oligosaccharides may accumulate in the liver due to 3′-SL metabolism, thereby promoting the synthesis of galactinol, stachyose, melibiose, and pimelibiose. These oligosaccharides are transported to the liver via the portal vein, where they exert beneficial effects by enhancing antioxidant defenses, suppressing inflammation, and regulating energy metabolism. The systemic remodeling of hepatic metabolic networks underscores the intervention's capacity to synergistically promote cellular repair, membrane stabilization, and redox balance.

3.7 Integrated analysis reveals correlations between differential genes and metabolites, with key findings validated by QT-PCR

Integrated analysis of differentially expressed genes (DEGs) and metabolites revealed that metabolites significantly correlated with DEGs were mainly categorized into four functional clusters: carbohydrate metabolism intermediates, bile acids and lipid-related metabolites, amino acids and derivatives, and vitamins and derivatives (Fig. S2). Notably, the downregulation of genes linked to lipid metabolism dysregulation, inflammation, and immune dysfunction (e.g., pro-inflammatory cytokines) was strongly correlated with dynamic alterations in metabolite levels (e.g., increased anti-inflammatory lipids and reduced oxidative stress markers) (Fig. 7A–G). These findings suggest that metabolic reprogramming may epigenetically or transcriptionally modulate gene expression, thereby synergistically ameliorating metabolic imbalance and immune dyshomeostasis.

Fig. 7 From multi-omics to mechanism: integrated transcriptomic and metabolomic profiling reveals key genes validated by QT-PCR. (A–D) Correlation between differential metabolites and differentially expressed genes in the colon content: (A) correlation between top 50 differential metabolites (Neu5Ac_Colon_vs_HCD_Colon) and metabolism-associated differentially expressed genes (top 20 KEGG pathways – Neu5Ac_Colon_vs_HCD_Colon) in the colon content; (B) correlation between top 50 differential metabolites (Neu5Ac_Colon_vs_HCD_Colon) and nerve regulation-associated differentially expressed genes (top 20 KEGG pathways – Neu5Ac_Colon_vs_HCD_Colon) in the colon content; (C) correlation between top 50 differential metabolites (Neu5Ac_Colon_vs_HCD_Colon) and immunity and inflammation-associated differentially expressed genes (top 20 KEGG pathways – Neu5Ac_Colon_vs_HCD_Colon) in the colon content; (D) correlation between top 50 differential metabolites (3′-SL_Colon_vs_HCD_Colon) and inflammation and thrombosis-associated differentially expressed genes (top 20 KEGG pathways – 3′-SL_Colon_vs_HCD_Colon) in the colon content; (E–G) correlation between differential metabolites and differentially expressed genes in the liver tissue: (E) correlation between top 50 differential metabolites (Neu5Ac_Liver_vs_HCD_Liver) and genes associated with key hepatic processes (lipid metabolism, fatty acid β-oxidation, detoxification, and oxidative stress) from the top 20 KEGG pathways in the liver (Neu5Ac_Liver vs. HCD_Liver); (F) correlation between the top 50 differential metabolites (3′-SL_Liver_vs_HCD_Liver) and the genes associated with the key hepatic processes (lipid metabolism, detoxification, and oxidative stress) the top 20 KEGG pathways in the liver (3′-SL_Liver_vs_HCD_Liver); (G) correlation between the top 50 differential metabolites (3′-SL_Liver_vs_HCD_Liver) and the genes associated with the key hepatic processes (fatty acid β-oxidation, immunity and inflammation) from the top 20 KEGG pathways in the liver (3′-SL_Liver_vs_HCD_Liver); and (H) QT-PCR analysis confirming the expression of key genes in the colon, liver, and coronary tissues following intervention. The difference between values with completely different superscripts was statistically significant, P < 0.05. Data are means ± SD (n = 6). * denotes statistically significant differences between the groups, *0.01 < P ≤ 0.05, **0.001 < P ≤ 0.01, ***P ≤ 0.001.

Based on the results of differential gene expression analysis and gene–metabolite correlation analysis in the liver, colon, and coronary tissues after intervention with the test substance, we selected four key genes from each of the three tissues for QT-PCR validation. As shown in Fig. 7(H), the verification results were consistent with the transcriptomic findings from each tissue, as detailed below.

In the colon tissue, both Neu5Ac and 3′-SL interventions upregulated the expression of Acsm3 (vs. HCD, P < 0.05), promoting lipid metabolism and reducing lipotoxicity, while downregulating Hspa1a to alleviate lipotoxicity-induced stress and improve liver metabolism. Both also upregulated Kng1 (vs. HCD, P < 0.05), contributing to improved intestinal microcirculation and barrier function. Additionally, 3′-SL significantly upregulated Ccl12 expression (vs. HCD, P < 0.05), promoting a reparative inflammatory response.

Both Neu5Ac and 3′-SL significantly upregulated Cyp7a1 (vs. HCD, P < 0.05), facilitating cholesterol metabolism, and downregulated Apoc3, thereby enhancing LPL activity and reducing triglyceride levels. Furthermore, both enhanced the expression of Ppara and Plin2 (vs. HCD, P < 0.05), increased lipid storage capacity, promoted fatty acid oxidation, and alleviated cellular stress, ultimately contributing to improved insulin sensitivity and energy metabolism balance.

In the coronary tissue, both Neu5Ac and 3′-SL downregulated Aopb (vs. HCD, P < 0.05), reducing lipid deposition, and decreased L1cam and Cxcr4 (vs. HCD, P < 0.05), inhibiting inflammatory cell infiltration into plaques and the release of pro-inflammatory factors. Moreover, 3′-SL upregulation of Postn expression (vs. HCD, P < 0.05) helps enhance plaque stability and prevent rupture.

These results further show that Neu5Ac and 3′-SL, through coordinated multi-tissue and multi-gene regulation, exert comprehensive effects in improving lipid metabolism, alleviating inflammatory responses, and enhancing tissue repair and stability.

4. Discussion

The present study elucidates the atheroprotective mechanisms of Neu5Ac and 3′-SL in LDLR^−/− mice fed a high-cholesterol diet (HCD); Fig. 8 shows their differential effects on the gut microbiota, colon, liver, and coronary arteries. By integrating multi-omics data, we have revealed their systemic effects on metabolic remodeling, inflammatory responses, oxidative stress, and gut microbiota–host crosstalk. Below, we contextualize these findings within the existing literature and propose mechanistic pathways underlying their protective roles.

Fig. 8 A summary of the differential results of Neu5Ac and 3′-SL in the gut microbiota, colonic transcription and metabolism, hepatic transcription and metabolism, and coronary transcriptional profiles.

4.1 Weight modulation and metabolic paradox

Despite comparable energy intake, 3′-SL supplementation induced a modest but significant increase in body weight compared to the HCD group, while Neu5Ac showed no such effect. This weight gain contrasts with the observed reductions in hepatic and coronary lipid deposition, suggesting that 3′-SL may influence energy partitioning rather than adiposity. Notably, neither group exhibited changes in visceral fat mass, implying that the weight difference likely reflects altered lean mass or fluid retention rather than fat accumulation.²⁴ This paradoxical dissociation between weight gain and improved metabolic health underscores the complexity of energy metabolism regulation by sialylated oligosaccharides, potentially mediated via gut microbiota-dependent mechanisms.^24–26

4.2 Anti-atherogenic effects: lipid metabolism and plaque stabilization

Both Neu5Ac and 3′-SL significantly attenuated coronary lipid deposition and atherosclerotic lesion severity, aligning with their ability to downregulate LOX-1 (a key ox-LDL receptor) and chemokines (CXCL1 and CCL5), reducing foam cell formation and inflammatory cell infiltration. These effects likely stem from coordinated transcriptional regulation of lipid metabolism pathways (e.g., PPAR signaling and cholesterol efflux)^27–29 and stabilized plaque structure via the extracellular matrix (ECM) receptor interaction pathway³⁰ and suppression of pro-inflammatory cytokines (IL-18 and IFN-γ) in coronary tissues.^31,32 The upregulation of IL-10 further highlights their immunomodulatory capacity, which stabilizes plaques by reducing inflammatory infiltration and oxidative damage.^33,34 These findings align with previous studies, indicating that sialic acid derivatives synergistically improve endothelial function and delay atherosclerosis progression through multi-target mechanisms.

4.3 Hepatic protection: lipid homeostasis and oxidative stress mitigation

While liver weight remained unchanged, both interventions markedly reduced hepatic lipid accumulation and improved systemic dyslipidemia (lower TC, TG and LDL-C; higher HDL-C). Omics-based analyses revealed distinct mechanisms underlying their hepatoprotective and lipid-regulatory effects. Transcriptomic and metabolomic profiling showed that Neu5Ac enhanced lipid catabolism (β-oxidation³⁵ and cholesterol metabolism^36,37) and mitochondrial function,³⁸ thereby improving hepatic lipid metabolism and preserving hepatocyte integrity. In contrast, 3′-SL primarily targeted glycerophospholipid metabolism and detoxification pathways.³⁹ Notably, the hepatic enrichment of oligosaccharides (e.g., galactinol, stachyose and melibiose) in the liver suggests gut microbiota-mediated conversion of 3′-SL into bioactive metabolites.^40,41 3′-SL and stachyose had been shown to inhibit Toll-like receptor (TLR) 4-induced low-grade inflammation in macrophages and endothelial cells.^42,43 These hepatic improvements likely contributed to systemic lipid clearance, reducing lipotoxicity and coronary disease burden.

4.4 Gut microbiota reprogramming: a central mediator of systemic effects

Both supplements induced distinct yet complementary shifts in the gut microbiota composition. The enrichment of Akkermansia⁴⁴—and genera associated with anti-inflammatory and barrier-protective effects—correlated with elevated HDL and IL-10 levels. Conversely, enrichment in Lactobacillus and *Clostridia_UCG-014* aligned with decreased pro-inflammatory cytokines (IL-6 and TNF-α) and lipids (TC and LDL). Functional profiling highlighted 3′-SL's suppression of LPS biosynthesis, whereas Neu5Ac enhanced metabolism of sphingolipid, which regulates intestinal immunity.^45–47 These findings support the “gut–liver–coronary axis” hypothesis, wherein microbiota-derived metabolites (e.g., bile acids,⁴⁸ thiamine⁴⁹ and intermediates of lipid metabolism^49,50) systemically modulate host lipid and immune homeostasis.

4.5 Multi-omics integration: metabolic–immune crosstalk

Integrated analysis of serum lipid profiles, inflammatory cytokines, histopathological evaluation, and multi-omics data (e.g., gut microbiota/transcriptomic/metabolomic) collectively indicate that Neu5Ac and 3′-SL supplementation inhibited AS development in LDLR^−/− mice fed a high-cholesterol diet through the following mechanisms: (1) metabolic remodeling – coordinated regulation of cholesterol, fatty acid, and bile acid metabolism to reduce lipid accumulation; (2) anti-inflammatory and antioxidant effects – suppression of pro-inflammatory pathways (complementary cascades and IL-17/MAPK signaling) and mitigation of ROS/DNA damage; (3) multi-organ protection – gut, enhanced barrier integrity and pathogen clearance; colon, downregulation of pro-inflammatory pathways (IL-17 and complementary cascades) and enhanced barrier integrity via Fc gamma R-mediated phagocytosis; liver, synergistic upregulation of detoxification (cytochrome P450) and antioxidant (selenocompound metabolism) pathways, improved steatosis and systemic lipotoxicity; and coronary arteries, stabilization of the extracellular matrix (ECM–receptor interactions) and endothelial function (arginine biosynthesis); and (4) cross-organ crosstalk – systemic protection via the gut–liver–coronary axis, integrating lipid metabolism, inflammation, and oxidative stress responses. Notably, the strong correlation between serum inflammatory markers (hs-CRP and IL-6) and lipid profiles (LDL-C and TG) reinforces the interplay between metabolic dysfunction and immune activation in atherosclerosis progression.

Notably, based on the metabolic pathways of 3′-SL and Neu5Ac and the differential results observed in the colonic, hepatic, and coronary transcriptomes and metabolomes in this study, it can be inferred that the long-term mechanisms of action of 3′-SL and Neu5Ac may differ. 3′-SL, which is poorly absorbed into the bloodstream,⁵¹ primarily exerts its effects by modulating the gut microbiota structure, thereby influencing colonic luminal metabolites and transcriptional signaling pathways. These microbiota-derived metabolites subsequently enter systemic circulation to affect transcriptional programs in the liver and coronary arteries. This gut-centric mechanism is supported by the more pronounced alterations in the gut microbiota structure and function compared to Neu5Ac, along with consistent functional enrichment patterns observed in differential metabolites from both colon and liver.

In contrast, Neu5Ac is efficiently absorbed in the small intestine and rapidly enters the bloodstream,⁵² where it directly modulates hepatic and coronary transcription and metabolism. Subsequently, both Neu5Ac and liver-derived metabolites influence colonic signaling pathways via enterohepatic circulation. This systemic mechanism is evidenced by concurrent improvements in anti-inflammatory signaling across liver, colon, and coronary tissues, as well as consistent pathway enrichment observed among hepatic differential metabolites, colonic differential metabolites, and colonic transcriptomic changes.

4.6 Pleiotropic effects of sialic acid intervention: potential health benefits beyond atherosclerosis

This study primarily investigated the anti-atherosclerotic effects of sialic acids (Neu5Ac and 3′-SL). However, transcriptomic analyses further suggested that these substances may possess broader pleiotropic potential in other disease areas. Notably, Neu5Ac significantly modulated pathways in the liver associated with Alzheimer's disease and Parkinson's disease (such as oxidative phosphorylation and apoptosis),^53,54 suggesting potential neuroprotective functions. While previous research has predominantly focused on the role of sialic acids in promoting neurodevelopment during infancy,⁵⁵ our findings reveal their potential value in mitigating neurodegenerative processes in aging.

Furthermore, both sialic acids, particularly 3′-SL, showed substantial immunomodulatory capabilities in colon tissue. By inhibiting key inflammatory pathways including the IL-17 signaling pathway and complementary and coagulation cascades,^56,57 our results suggest their potential therapeutic value for inflammatory bowel disease and autoimmune disorders. Additionally, 3′-SL was found to regulate pathways related to transcriptional misregulation in cancer, gastric cancer, and longevity, indicating its potential exploratory value in cancer chemoprevention and healthy aging.

In summary, this study expands our understanding of the physiological functions of sialic acids and provides a novel theoretical foundation for their potential applications in neurodegenerative diseases, immune regulation, and aging-related conditions.

4.7 Clinical application and translational potential of sialic acids

The findings of this study provide new insights into the prevention and treatment of atherosclerosis by supplementing with Neu5Ac and 3′-SL to target the gut–liver–coronary axis, highlighting their significant translational potential. A crucial foundation for this potential lies in their established safety profile and regulatory approval status. For instance, Neu5Ac was approved as a novel food ingredient by China's National Health Commission in 2017 and is also used in food products in the European Union, Japan, and other regions.⁵⁸ Similarly, an application for 3′-SL as a novel food ingredient has been submitted to the European Commission, proposing its use in various food categories and formulas for special medical purposes.⁵⁹ These regulatory milestones pave the way for developing Neu5Ac and 3′-SL as dietary supplements or functional food ingredients for the broader population. Based on this, our results suggest that supplementing with sialic acid and its derivatives can not only aid in weight management but also exert preventive effects by modulating the gut microbiota and suppressing systemic inflammation. Consequently, these compounds hold promise as an innovative nutritional strategy, potentially used synergistically with existing lipid-lowering drugs to enhance the comprehensive management of coronary atherosclerosis, or combined with probiotics to offer a sustainable long-term intervention for patients with chronic metabolic diseases by regulating the gut microbiome.

4.8 Limitations

While this study provides mechanistic insights, several limitations warrant consideration.

First, the correlation analyses between transcriptomic and metabolomic data, while informative, cannot establish causal relationships between the observed molecular changes. Future research would benefit from incorporating longitudinal multi-omics data and functional validation experiments to better infer causality.

Second, although our study explores the role of the gut–liver axis, the evidence specifically supporting its extension to a gut–liver–coronary axis in the current experimental design remains indirect and insufficient. The mechanisms by which gut-derived metabolites or microbial signals might directly influence coronary atherosclerosis are not fully elucidated here, and more targeted studies are needed to solidify this concept.

Third, LDLR^−/− mice model familial hypercholesterolemia but may not fully represent human atherosclerosis. The pharmacokinetics of Neu5Ac and 3′-SL is unclear, including their bioavailability and tissue-specific distribution. Future studies should explore dose–response relationships, validate findings in human cohorts, and investigate whether combined supplementation enhances therapeutic efficacy.

5. Conclusion

Neu5Ac and 3′-SL exert multi-targeted protection against atherosclerosis through synergistic modulation of lipid metabolism, inflammation, and gut microbiota ecology. Their ability to attenuate systemic lipotoxicity, oxidative stress, and plaque instability highlights their potential as dietary adjuvants for cardiovascular disease prevention. This work underscores the importance of a holistic, multi-omics approach for unraveling the complex interactions between diet, microbiota, and host pathophysiology.

Author contributions

All authors contributed to the study conception and design. WZ and HWL conceived and designed the experiments; WZ, LLZ, XNZ and XYH performed the experiments; WZ, LLZ and XYH analyzed the data; XNZ, WQB, DBG, XXC and LLP contributed reagents/materials/analysis tools; and WZ, LLZ and HWL wrote the paper. All authors have read and approved the final manuscript.

Conflicts of interest

The authors declare that they have no competing interests.

Ethics approval

All experimental procedures were conducted in accordance with the guidelines set by the Institutional Animal Care and Use Committee (IACUC) of the Laboratory Animal Center at Xiamen University, and in compliance with the International Association of Veterinary Editors’ guidelines for the Care and Use of Laboratory Animals. The animal use protocol, as detailed below, has been reviewed and approved by the Animal Ethical and Welfare Committee of the Laboratory Animal Center at Xiamen University (approval no. XMULAC20240172).

Data availability

The raw sequencing data from this study have been deposited in the Genome Sequence Archive⁶⁰ in the BIG Data Center (https://bigd.big.ac.cn), Beijing Institute of Genomics (BIG), Chinese Academy of Science, under the accession numbers: CRA026431 and CRA026618. The data can be accessed at the following links: https://ngdc.cncb.ac.cn/gsa/s/XLuh2ULU and https://ngdc.cncb.ac.cn/gsa/s/RMscZPp0.

Supplementary information (SI): Fig. S1:  Top 20 shared signaling pathways enriched in transcriptomic differences of liver, coronary artery, and colon tissues after Neu5Ac or 3′-SL intervention. (A–C) Common pathway enrichment of transcriptomic differences in the liver, coronary artery, and colon tissues between the Neu5Ac intervention and high-cholesterol diet (HCD) groups: (A) top 20 shared signaling pathways enriched in the coronary artery tissue; (B) top 20 shared signaling pathways enriched in the liver tissue; and (C) top 20 shared signaling pathways enriched in the colon tissue. (D–F) Common pathway enrichment of transcriptomic differences in the liver, coronary artery, and colon tissues between the 3′-SL intervention and high-cholesterol diet (HCD) groups: (D) top 20 shared signaling pathways enriched in the coronary artery tissue; (E) top 20 shared signaling pathways enriched in the liver tissue; and (F) top 20 shared signaling pathways enriched in the colon tissue. Fig. S2: Top 50 differential metabolites and differentially expressed genes (top 20 KEGG pathways). (A and B) Top 50 differential metabolites (Neu5Ac_Colon_vs_HCD_Colon) and differentially expressed genes (top 20 KEGG pathways Neu5Ac_Colon_vs_HCD_Colon) in the colon content: (A) heatmap: differentially expressed genes; (B) heatmap: differential metabolites; (C and D) top 50 differential metabolites (3′-SL_Colon_vs_HCD_Colon) and differentially expressed genes (top 20 KEGG pathways 3′-SL_Colon_vs_HCD_Colon) in the colon content: (C) heatmap: differentially expressed genes; (D) heatmap: differential metabolites; (E and F) top 50 differential metabolites (Neu5Ac_Liver_vs_HCD_Liver) and differentially expressed genes (top 20 KEGG pathways Neu5Ac_Liver_vs_HCD_Liver) in the liver: (E) heatmap: differentially expressed genes; (F) heatmap: differential metabolites; (G and H) top 50 differential metabolites 3′-SL_Liver_vs_HCD_Liver and differentially expressed genes (top 20 KEGG pathways – 3′-SL_Liver_vs_HCD_Liver) in the liver: (G) heatmap: differentially expressed genes; and (H) heatmap: differential metabolites. Table S1: Primer sequences used for real-time PCR. See DOI: https://doi.org/10.1039/d5fo04031k.

References

1. L. Nedkoff, T. Briffa, D. Zemedikun, S. Herrington and F. L. Wright, Global trends in atherosclerotic cardiovascular disease, Clin. Ther., 2023, 45, 1087–1091.
2. W. Chen, Z. Li, Y. Zhao, Y. Chen and R. Huang, Global and national burden of atherosclerosis from 1990 to 2019: trend analysis based on the Global Burden of Disease Study 2019, China Med. J., 2023, 136, 2442–2450.
3. E. Gusev and A. Sarapultsev, Atherosclerosis and inflammation: Insights from the theory of general pathological processes, Int. J. Mol. Sci., 2023, 24, 7910.
4. S. R. Pvln and V. Kiranmayi, Biochemical mechanisms underlying atherogenesis, J. Clin. Sci. Res., 2012, 01, 24–34.
5. J. P. Cooke, Mechanisms of atherosclerosis: new insights and novel therapeutic approaches, Methodist Debakey Cardiovasc. J., 2015, 11, 154–155.
6. L. Shen, W. Chen, J. Ding, G. Shu, M. Chen, Z. Zhao, S. Xia and J. Ji, The role of metabolic reprogramming of oxygen-induced macrophages in the dynamic changes of atherosclerotic plaques, FASEB J., 2023, 37, e22791.
7. Q. Xiao, R. Hou, L. Xie, M. Niu, X. Pan and X. Zhu, Macrophage metabolic reprogramming and atherosclerotic plaque microenvironment: Fostering each other?, Clin. Transl. Med., 2023, 13, e1257.
8. Y. Liu, K. Lu, R. Zhang, D. Hu, Z. Yang, J. Zeng and W. Cai, Advancements in the treatment of atherosclerosis: From conventional therapies to cutting-edge innovations, ACS Pharmacol. Transl. Sci., 2024, 7, 3804–3826.
9. A. C. Aprotosoaie, A. D. Costache and I. I. Costache, Therapeutic strategies and chemoprevention of atherosclerosis: What do we know and where do we go?, Pharmaceutics, 2022, 14, 722.
10. R. V. Giglio, A. P. Stoian, K. Al-Rasadi, M. Banach, A. M. Patti, M. Ciaccio, A. A. Rizvi and M. Rizzo, Novel therapeutical approaches to managing atherosclerotic risk, Int. J. Mol. Sci., 2021, 22, 4633.
11. M. Z. Zhu, W. Zhang, K. Dekyi, L. X. Zheng, Y. C. Zhang, Y. P. Lv and H. W. Li, Potential effects of sialic acid and 3′-sialyllactose on intestinal health and anti-cardiovascular disease in mice fed with a high-fat diet, J. Funct. Foods, 2024, 116, 106215.
12. T. Klaudia and N. Jozef, “Lost sugars”—reality of their biological and medical applications, Open Life Sci., 2012, 7, 777–793.
13. W. Jianrong, L. Zhang, Q. Mengdan, Z. Hongtao and Z. Xiaobei, Metabolic fate of dietary sialic acid and its influence on gut and oral bacteria, Syst. Microbiol. Biomanuf., 2021, 2, 41–49.
14. W. Haiyan, Z. Xiaoying, Y. Yu, H. Zhenquan, C. Xiuxiu, L. Mengjia, Z. Lili and G. Wupeng, Oligosaccharide profiles as potential biomarkers for detecting adulteration of caprine dairy products with bovine dairy products, Food Chem., 2024, 443, 138551.
15. W. Zhang, M. Zhu, K. Dekyi, L. Zheng, Y. Zhang, Y. Lü, D. Guo, X. Chen, L. Pan, X. Wang and H. Li, 3′-Sialyllactose regulates glucose and lipid metabolic disorders via the gut–liver axis in mice fed a high-fat diet, Food Sci. Hum. Wellness, 2025, 14, 9250185.
16. W. Zhang, M. Zhu, K. Dekyi, L. Zheng, Y. Zhang, Y. Lv, D. Guo, X. Chen, L. Pan, X. Wang and H. Li, The effect of anti–imbalance on glycolipid metabolic homeostasis via the liver–gut axis intervened by 3′-sialyllactose in obese mice, Food Front., 2024, 6, 402–417.
17. A. V. Poznyak, D. Zhang, A. V. Grechko, W. K. Wu and A. N. Orekhov, The role of sialic acids in the initiation of atherosclerosis, Minerva Cardioangiol., 2020, 68, 359–364.
18. L. A. Cuniberti, V. Martinez, J. Schachter, G. Magariños, P. C. Meckert, R. P. Laguens, J. Levenson and J. P. Werba, Sialic acid as a protective barrier against neointima development, Atherosclerosis, 2005, 181, 225–231.
19. A. V. Poznyak, V. A. Khotina, A. A. Melnichenko, V. N. Sukhorukov, I. A. Sobenin and A. N. Orekhov, Overview of the most popular currently used murine models of atherosclerosis, OnLine J. Biol. Sci., 2024, 24, 208–218.
20. A. V. Poznyak, Y. Y. Silaeva, A. N. Orekhov and A. V. Deykin, Animal models of human atherosclerosis: Current progress, Braz. J. Med. Biol. Res., 2020, 53, e9557.
21. D. Bian, X. Wang, J. Huang, X. Chen and H. Li, Maternal Neu5Ac supplementation during pregnancy improves offspring learning and memory ability in rats, Front. Nutr., 2021, 8, 641027.
22. E. J. Want, P. Masson, F. Michopoulos, I. D. Wilson, G. Theodoridis, R. S. Plumb, J. Shockcor, N. Loftus, E. Holmes and J. K. Nicholson, Global metabolic profiling of animal and human tissues via UPLC-MS, Nat. Protoc., 2013, 8, 17–32.
23. C. H. Zhu, Y. X. Li, Y. C. Xu, N. N. Wang, Q. J. Yan and Z. Q. Jiang, Tamarind xyloglucan oligosaccharides attenuate metabolic disorders via the gut–liver axis in mice with high-fat-diet-induced obesity, Foods, 2023, 12, 1382.
24. Z. Wei, Z. Meizhen, D. Kelsang, Z. Linxi, Z. Yichen, L. Youping, G. Dongbei, C. Xiaoxuan, P. Lili, W. Xinyue and L. Hongwei, The effect of anti–imbalance on glycolipid metabolic homeostasis via the liver–gut axis intervened by 3′-sialyllactose in obese mice, Food Front., 2024, 6, 402–417.
25. M. Yuan, Z. Zhang, T. Liu, H. Feng, Y. Liu and K. Chen, The role of nondigestible oligosaccharides in alleviating human chronic diseases by regulating the gut microbiota: A review, Foods, 2024, 13, 2157.
26. Y. Mingzhi, J. Zipeng, Z. Lutong, C. Nana, H. Huan, L. Wentao, Y. Zhixin, J. Siming, S. Deguang, W. Yizhen, J. Mingliang and L. Zeqing, 3′-Sialyllactose and B. infantis synergistically alleviate gut inflammation and barrier dysfunction by enriching cross-feeding bacteria for short-chain fatty acid biosynthesis, Gut Microbes, 2025, 17, 2486512.
27. A. J. Kattoor, G. Akshay and L. M. Jawahar, LOX-1: Regulation, signaling and its role in atherosclerosis, Antioxidants, 2019, 8, 218.
28. L. Chih-Hao, C. Ajay, U. Ned, L. Debbie, A. B. William and M. E. Ronald, Transcriptional repression of atherogenic inflammation: Modulation by PPARδ, Science, 2003, 302, 453–457.
29. K. Jaemi, K. Ji-Yun, B. Hye-Eun, K. Ji-Won, K. Hyoun-Ah, S. Chang-Hee, C. Sangdun, F. L. MacRae and J. Ju-Yang, Inhibition of Toll-like receptors alters macrophage cholesterol efflux and foam cell formation, Int. J. Mol. Sci., 2024, 25, 6808.
30. Z. Lu, W. Xin and H. Liang, Endothelial reprogramming in atherosclerosis, Bioengineering, 2024, 11, 325.
31. D. Yannis, A. Constantina, T. Panagiotis, T. Panagiotis, O. Evangelos, C. Christos, T. Konstantinos and T. Dimitris, Novel anti-inflammatory therapies in coronary artery disease and acute coronary syndromes, Life, 2023, 13, 1669.
32. E. M. James, R. M. Daryn, G. A. Tim and P. R. Dipak, Cytokines, macrophage lipid metabolism and foam cells: Implications for cardiovascular disease therapy, Prog. Lipid Res., 2011, 50, 331–347.
33. R. Tina and L. Reinhard L, Downregulated CD36 and oxLDL uptake and stimulated ABCA1/G1 and cholesterol efflux as anti-atherosclerotic mechanisms of interleukin-10, Cardiovasc. Res., 2006, 69, 527–535.
34. W. Liao, Y. Li, H. Zhao and S. Lu, The Lian-Dou-Qing-Mai formula activates the PPARγ-LXRα-ABCA1/ABCG1 pathway by regulating IL-10, leading to the promotion of cholesterol efflux and a reduction in atherosclerotic plaques, Histol. Histopathol., 2025, 40, 585–596.
35. H. Bhatia, B. R. Pattnaik and M. Datta, Inhibition of mitochondrial β-oxidation by miR-107 promotes hepatic lipid accumulation and impairs glucose tolerance in vivo, Int. J. Obes., 2016, 40, 861–869.
36. S. Zhang, F. Hong, C. Ma and S. Yang, Hepatic lipid metabolism disorder and atherosclerosis, Endocr., Metab. Immune Disord.: Drug Targets, 2022, 22, 590–600.
37. J. Y. L. Chiang, J. M. Ferrell, Y. Wu and S. Boehme, Bile acid and cholesterol metabolism in atherosclerotic cardiovascular disease and therapy, Cardiol. Plus, 2020, 5, 159–170.
38. R. G. Bardallo, I. Company-Marin, E. Folch-Puy, J. Roselló-Catafau, A. Panisello-Rosello and T. Carbonell, PEG35 and glutathione improve mitochondrial function and reduce oxidative stress in cold fatty liver graft preservation, Antioxidants, 2022, 11, 158.
39. H. Chen, Z. Wang, M. Qin, B. Zhang, L. Lin, Q. Ma, C. Liu, X. Chen, H. Li, W. Lai and S. Zhong, Comprehensive metabolomics identified the prominent role of glycerophospholipid metabolism in coronary artery disease progression, Front. Mol. Biosci., 2021, 8, 632950.
40. S. Liu, J. Hu, Y. Zhong, X. Hu, J. Yin, T. Xiong, S. Nie and M. Xie, A review: Effects of microbial fermentation on the structure and bioactivity of polysaccharides in plant-based foods, Food Chem., 2024, 440, 137453.
41. C. Federica, B. Nicole van, C. V. Nicol and J. L. Dirk, Galactose in human metabolism, glycosylation and congenital metabolic diseases: Time for a closer look, Biochim. Biophys. Acta, Gen. Subj., 2021, 1865, 129898.
42. A. R. Pessentheiner, N. J. Spann, C. A. Autran, T. G. Oh, K. V. Grunddal, J. K. Coker, C. D. Painter, B. Ramms, A. W. Chiang, C. Y. Wang, J. Hsiao, Y. Wang, A. Quach, L. M. Booshehri, A. Hammond, C. Tognaccini, J. Latasiewicz, L. Willemsen, K. Zengler, M. P. de Winther, H. M. Hoffman, M. Philpott, A. P. Cribbs, U. Oppermann, N. E. Lewis, J. L. Witztum, R. Yu, A. R. Atkins, M. Downes, R. M. Evans, C. K. Glass, L. Bode and P. L. Gordts, The human milk oligosaccharide 3′sialyllactose reduces low-grade inflammation and atherosclerosis development in mice, JCI Insight, 2024, 9, e181329.
43. S. Jiang, Q. Li, S. Han, H. Wang, X. Tang, T. Wang and X. Meng, Study on the anti-inflammatory effect of stachyose by inhibiting TLR4/NF-κB signalling pathway in vitro and in vivo, Clin. Exp. Pharmacol. Physiol., 2023, 50, 573–582.
44. E. Raden Mohamad Rendy Ariezal, A. Muhammad, K. Handono, D. Reiva Farah, S. Oki, M. P. Luba, E. C. N. Tamar and T. Hok Bing, Akkermansia muciniphila and Faecalibacterium prausnitzii in immune-related diseases, Microorganisms, 2022, 10, 2382.
45. E. L. Johnson, S. L. Heaver, J. L. Waters, B. I. Kim, A. Bretin, A. L. Goodman, A. T. Gewirtz, T. S. Worgall and R. E. Ley, Sphingolipids produced by gut bacteria enter host metabolic pathways impacting ceramide levels, Nat. Commun., 2020, 11, 2471.
46. J. Kunisawa and H. Kiyono, Sphingolipids and epoxidized lipid metabolites in the control of gut immunosurveillance and allergy, Front. Nutr., 2016, 3, 3.
47. B. M. Janelsins, M. Lu and S. K. Datta, Altered inactivation of commensal LPS due to acyloxyacyl hydrolase deficiency in colonic dendritic cells impairs mucosal Th17 immunity, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 373–378.
48. Q. Zhao, J. Wu, Y. Ding, Y. Pang and C. Jiang, Gut microbiota, immunity, and bile acid metabolism: Decoding metabolic disease interactions, Life Metab., 2023, 2, load032.
49. J. Wang, N. Zhu, X. Su, Y. Gao and R. Yang, Gut-microbiota-derived metabolites maintain gut and systemic immune homeostasis, Cells, 2023, 12, 793.
50. P. He, L. Yu, F. Tian, H. Zhang, W. Chen and Q. Zhai, Dietary patterns and gut microbiota: The crucial actors in inflammatory bowel disease, Adv. Nutr., 2022, 13, 1628–1651.
51. E. Vazquez, A. Santos-Fandila, R. Buck, R. Rueda and M. Ramirez, Major human milk oligosaccharides are absorbed into the systemic circulation after oral administration in rats, Br. J. Nutr., 2017, 117, 237–247.
52. M. Zhao, Y. Zhu, H. Wang, W. Zhang and W. Mu, Recent advances on N-acetylneuraminic acid: Physiological roles, applications, and biosynthesis, Synth. Syst. Biotechnol., 2023, 8, 509–519.
53. D. Caccamo, L. R. Pisani, P. Mazzocchetti, R. Ientile, P. Calabresi, F. Pisani and C. Costa, Neuroprotection as a potential therapeutic perspective in neurodegenerative diseases: Focus on antiepileptic drugs, Neurochem. Res., 2015, 41, 340–352.
54. L. Zuo, E. R. Prather, M. Stetskiv, D. E. Garrison, J. R. Meade, T. I. Peace and T. Zhou, Inflammaging and oxidative stress in human diseases: From molecular mechanisms to novel treatments, Int. J. Mol. Sci., 2019, 20, 4472.
55. B. Wang, Sialic acid is an essential nutrient for brain development and cognition, Annu. Rev. Nutr., 2009, 29, 177–222.
56. B. Verstockt, G. Van Assche, S. Vermeire and M. Ferrante, Biological therapy targeting the IL-23/IL-17 axis in inflammatory bowel disease, Expert Opin. Biol. Ther., 2016, 17, 31–47.
57. A. Ioannou, L. Kannan and G. C. Tsokos, Platelets, complement and tissue inflammation, Autoimmunity, 2012, 46, 1–5.
58. D. Turck, J. L. Bresson, B. Burlingame, T. Dean, S. Fairweather–Tait, M. Heinonen, K. I. Hirsch–Ernst, I. Mangelsdorf, H. J. McArdle, A. Naska, M. Neuhäuser–Berthold, G. Nowicka, K. Pentieva, Y. Sanz, A. Siani, A. Sjödin, M. Stern, D. Tomé, M. Vinceti, P. Willatts, K. H. Engel, R. Marchelli, A. Pöting, M. Poulsen, J. R. Schlatter, E. Turla and H. van Loveren, Safety of synthetic N–acetyl–d–neuraminic acid as a novel food pursuant to Regulation (EC) No 258/97, EFSA J., 2017, 15(7), e04918.
59. European Food Safety Authority (EFSA), T. Dominique, C. Paolo, F. Estefanía Noriega, F. Pablo Rodríguez and K. Helle Katrine, Scientific and technical assistance report on the evaluation of human–identical milk oligosaccharides (HiMOs) as novel foods, EFSA Supporting Publ., 2024, 21, 33.
60. T. Chen, X. Chen, S. Zhang, J. Zhu, B. Tang, A. Wang, L. Dong, Z. Zhang, C. Yu, Y. Sun, L. Chi, H. Chen, S. Zhai, Y. Sun, L. Lan, X. Zhang, J. Xiao, Y. Bao, Y. Wang, Z. Zhang and W. Zhao, The genome sequence archive family: toward explosive data growth and diverse data types, Genomics, Proteomics Bioinf., 2021, 19, 578–583.

---