1 Introduction
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder primarily characterized by cognitive impairment and is frequently accompanied by weight loss, reduced appetite, and metabolic dysregulation. These metabolic disturbances not only exacerbate disease progression but also substantially diminish patients’ quality of life [
1]. In recent years, olfactory dysfunction has been recognized as one of the earliest clinical hallmarks of AD, highlighting a close link between the olfactory system, AD pathology, and metabolic regulation [
1,
2]. AD-associated anorexia not only compromises nutritional intake and daily functioning but may further aggravate neurodegeneration through disrupted energy metabolism and heightened inflammatory responses [
3,
4]. Accordingly, restoring appetite and metabolic homeostasis in AD patients is of considerable importance for delaying disease progression and improving overall well-being.
Current interventions for AD-related appetite loss mainly rely on pharmacological treatments and nutritional supplementation. However, these approaches are often limited by modest efficacy, adverse side effects, and poor long-term compliance, and no disease-specific therapy has yet proven effective in correcting appetite dysfunction in AD [
5]. The olfactory system is anatomically and functionally interconnected with the hypothalamus and limbic structures, enabling odor stimuli to directly influence neurotransmitter release, emotional states, and appetite regulation [
6,
7]. Olfaction-based aromatherapy has emerged as an attractive non-invasive strategy owing to its favorable safety profile and suitability for application, attracting increasing attention in the context of neurodegenerative disorders [
5,
8–
10]. Accumulating evidence indicates that olfactory stimulation with aromatic compounds can alleviate depression, anxiety, and cognitive deficits, while also modulating energy metabolism and immune responses [
10,
11]. These findings underscore the innovative potential of targeting olfactory pathways to ameliorate anorexia associated with AD.
Eugenol, a major bioactive constituent of clove (
Syzygium aromaticum), exhibits well-documented neuroprotective, anti-inflammatory, antioxidant, and metabolic regulatory properties [
11–
13]. Previous studies have demonstrated that eugenol readily penetrates the central nervous system and can act directly on the olfactory bulb and hypothalamic circuits via inhalation, thereby influencing central appetite-regulatory networks [
12]. In animal models, oral or injectable administration of eugenol has been shown to enhance food intake in healthy mice, primarily by modulating the expression of orexigenic neuropeptides [agouti-related peptide (AGRP) and neuropeptide Y (NPY)] and anorexigenic neuropeptides (proopiomelanocortin, POMC) in the hypothalamus, potentially through activation of the TRPV1-CaMKK2/AMPK signaling pathway [
14]. Moreover, eugenol has been reported to attenuate neuronal loss [
11], suggesting that it may confer dual benefits in AD models by simultaneously improving metabolic function and providing neuroprotection.
The 5×FAD transgenic mouse is a widely used model for investigating early AD pathology and associated metabolic disturbances, characterized by pronounced appetite reduction, weight loss, cognitive impairment, and neuroinflammatory activation [
15–
17]. Notably, 5×FAD mice exhibit decreased body weight and negative energy balance as early as six months of age, accompanied by marked glial activation and elevated inflammatory cytokine levels [
17,
18]. Despite these insights, it remains unclear whether eugenol delivered via atomized inhalation can ameliorate hypothalamic neuroinflammation and rectify dysregulated appetite signaling pathways in 5×FAD mice, thereby promoting food intake and improving overall health status.
Therefore, the present study employed 5×FAD mice to systematically investigate the effects of eugenol atomization inhalation on body weight, food intake, and appetite-related hormones. By integrating serum multiplex analyses, hypothalamic transcriptomic profiling, and immunohistochemical approaches, we sought to elucidate the molecular mechanisms through which eugenol modulates hypothalamic appetite signaling and neuroinflammatory responses via the olfactory pathway. The findings of this study provide novel mechanistic insights into non-pharmacological interventions for AD-associated metabolic dysfunction and offer a scientific foundation for the development of aromatherapy-based adjunctive strategies in the management of AD.
2 Materials and Methods
2.1 Materials and reagents
Standards of eugenol (purity ≥ 98%) were obtained from Sigma-Aldrich (Shanghai, China). The solubility of eugenol in water is 2460 mg/L at 25 ℃. Enzyme-linked immunosorbent assay (ELISA) kits for insulin (INS), ghrelin, AGRP, POMC, α-melanocyte-stimulating hormone (α-MSH), and the neurotransmitters serotonin (5-HT) and dopamine (DA) were obtained from Beijing Huabodeyi Biology Science and Technology Co., Ltd. (Beijing, China). ELISA kits for NPY, leptin (LEP), glucagon-like peptide-1 (GLP-1), and ionized calcium-binding adaptor molecule 1 (IBA-1) were purchased from Beijing Mreda Technology Co., Ltd. (Beijing, China). ELISA kits for glial fibrillary acidic protein (GFAP) were obtained from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China). The NPY Rabbit Polyclonal Antibody (12833-1-AP) was purchased from Proteintech Group, Inc. (Wuhan, China). Anti-AGRP antibody (#ab254558) and rabbit recombinant monoclonal POMC antibody (#ab254257) were obtained from Abcam Shanghai Trading Co., Ltd. (Shanghai, China). Anti-MC4R polyclonal antibody (#K111045P) was provided by Solarbio Technology Co., Ltd. (Beijing, China). Horseradish peroxidase (HRP)-conjugated rabbit anti-goat IgG (#GB23204) was purchased from Servicebio Biotechnology Co., Ltd. (Wuhan, China). The Fos proto-oncogene, AP-1 transcription factor subunit (c-Fos) Antibody (#HY-P80081) was obtained from MedChemExpress (Shanghai, China).
2.2 Animals and treatments
All animal experimental procedures were conducted in accordance with the Regulations for the Administration of Laboratory Animals of the People’s Republic of China and were approved by the Ethics Committee of Beijing Technology and Business University (license No. BTBU2024135; Beijing, China; 07/2024–12/2027). All experimental animals were six-month-old female mice purchased from Cyagen Biosciences Inc. (Beijing, China). The mice were on a C57BL/6J genetic background and included homozygous 5×FAD transgenic mice and non-transgenic wild-type (WT) littermates. All mice were housed individually under specific pathogen-free conditions with standard bedding, in a temperature- and humidity-controlled environment (23 ± 2 ℃) under a 12 h light/dark cycle, with free access to food and water. After one week of acclimatization, mice were randomly assigned to four groups: water-atomized 5×FAD mice (CAD, n = 10), eugenol-atomized 5×FAD mice (EAD, n = 10), water-atomized WT mice (CWT, n = 10), and eugenol-atomized WT mice (EWT, n = 10). Eugenol atomization inhalation experiments were then performed.
Eugenol was administered using a small-animal nebulization system (PARI, Germany). An eugenol aqueous solution (20 mg/kg) was delivered by atomized inhalation, with ultrapure water used as the control. This value refers to the body weight-normalized eugenol exposure level prepared for each atomization session rather than the actual inhaled or systemically absorbed dose. The nebulization rate was 0.45 mL/L, the aerosol particle size was 2.2 μm, and the exposure chamber volume was 70 L (Fig. 1A). Prior to each nebulization session, mice were fasted for 12 h, after which food and water were provided ad libitum. Nebulization was performed once daily for 40 min per session (starting at 9:00 AM) over a period of four consecutive weeks. Food intake was recorded at 2, 4, and 12 h after each nebulization. Body weight was monitored throughout the intervention period to assess longitudinal changes associated with eugenol inhalation. On day 28, all mice were anesthetized, and blood samples were collected via orbital puncture. Blood samples were centrifuged at 2000 × g for 10 min at 4 ℃, and serum and plasma were collected and stored at −80 ℃ for subsequent biochemical analyses. The hypothalamus, olfactory bulb, and stomach tissues were then rapidly harvested, snap-frozen in liquid nitrogen, and stored at −80 ℃ until further analysis.
2.3 Analysis of serum appetite-related hormones, neurotransmitters, and glial-related markers GFAP and IBA-1
To elucidate the potential mechanisms by which eugenol modulates appetite regulation, serum levels of INS, ghrelin, NPY, AGRP, LEP, GLP-1, POMC, α-MSH, and the appetite-related neurotransmitters 5-HT and DA were quantified using ELISA. Serum GFAP and IBA-1 were measured as AD-related glial activation markers. All assays were performed strictly in accordance with the manufacturers’ instructions. Absorbance was measured at 450 nm using a multimode microplate reader (Spark, Tecan, Switzerland), and concentrations were calculated from standard curves generated for each assay.
2.4 Hematoxylin and eosin (H&E) staining
H&E staining was performed on mouse stomach and olfactory bulb tissues to evaluate histological morphology. Immediately after dissection, tissues were fixed in 4% paraformaldehyde (PFA) for 24 h and subsequently embedded in paraffin. Paraffin sections (4 μm) were deparaffinized and rehydrated through graded ethanol, followed by hematoxylin staining for nuclear visualization and eosin staining for cytoplasmic components. Sections were then dehydrated through graded ethanol, cleared in xylene, and mounted with neutral resin. Histological features were examined under a light microscope (SWE-CX63, Servicebio, China), with nuclei appearing blue and cytoplasm red, and evaluated based on tissue architecture and staining characteristics.
2.5 Immunohistochemistry (IHC) analysis
Paraffin-embedded hypothalamic sections were used for immunohistochemical detection of NPY, AGRP, POMC, and MC4R. Briefly, hypothalamic sections were deparaffinized, rehydrated, and subjected to heat-induced antigen retrieval in EDTA buffer. After blocking endogenous peroxidase activity and nonspecific binding, sections were incubated overnight at 4 ℃ with the corresponding primary antibodies, followed by HRP-conjugated secondary antibodies. Immunoreactivity was visualized using DAB substrate, counterstained with hematoxylin, and examined under a light microscope. Quantitative analysis was performed based on the positive cell ratio, average optical density, and H-score.
2.6 c-Fos immunohistochemical analysis
c-Fos immunohistochemistry was performed on paraffin-embedded olfactory bulb and hypothalamic sections using the same general HRP-DAB staining procedure described above. Sections were incubated with the anti-c-Fos primary antibody overnight at 4 ℃, followed by HRP-conjugated secondary antibody incubation and DAB visualization. c-Fos-positive signals were identified as brown nuclear staining. Neuronal activation was quantified using the positive cell ratio, average optical density, and H-score.
2.7 Bulk RNA-seq analysis
Total RNA was extracted from hypothalamic tissues, and RNA integrity and fragment size distribution were assessed using an Agilent 4150 TapeStation system. Messenger RNA (mRNA) was enriched using oligo(dT) magnetic beads to remove ribosomal and other non-coding RNAs. The enriched mRNA was fragmented and reverse-transcribed into complementary DNA (cDNA). Sequencing adapters were ligated to both ends of the cDNA fragments, followed by amplification to construct sequencing libraries. Libraries were purified using magnetic beads, quantified, normalized, and subsequently converted into DNA nanoballs (DNBs). The DNBs were loaded onto the DNBSEQ-T7RS platform for paired-end sequencing (PE150).
Raw FASTQ files were subjected to quality control using Fastp (v0.23.2), including adapter trimming, quality filtering, and read pruning, to generate high-quality clean reads for downstream analyses. Reference genome sequences and gene annotation files were obtained from the appropriate databases. Clean paired-end reads were aligned to the reference genome using STAR (v2.7.8a). Gene expression levels were quantified using StringTie (v2.2.1) and normalized as transcripts per kilobase per million mapped reads (TPM) to account for sequencing depth and gene length. For samples with biological replicates, differential expression analysis was performed using the DESeq2 R package (v1.42.1), with P-values adjusted by the Benjamini-Hochberg method to control the false discovery rate. Genes with |log2 fold change| > 1 and adjusted P < 0.05 were defined as differentially expressed genes (DEGs). For samples without biological replicates, differential expression analysis was conducted using the edgeR package (v3.14.0) with the same significance thresholds. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses and visualization were performed using the clusterProfiler package (v4.10.1), and terms or pathways with adjusted P < 0.05 were considered significantly enriched.
2.8 Measurement of systemic inflammatory cytokines in serum
A Luminex bead-based multiplex assay based on xMAP® technology was employed to simultaneously quantify multiple cytokines in serum samples. The procedure was performed according to the manufacturer’s instructions for the Bio-Plex Pro™ Cytokine Panel (Bio-Rad; Cat. No. 12009159; Lot No. 64611953). Briefly, plasma supernatants were diluted twofold, and 50 μL of each sample was incubated with mixed fluorescently coded microspheres, serially diluted standards, and quality controls in a 96-well plate for 60 min at room temperature with shaking. After three magnetic washes with 200 μL wash buffer, 50 μL of biotinylated detection antibody was added and incubated for 30 min at room temperature with shaking. Following additional washes, 50 μL of streptavidin-phycoerythrin was added and incubated in the dark for 15 min. After a final wash, beads were resuspended in assay buffer, incubated for 1 min, and analyzed using a Luminex® 200TM system. Cytokine concentrations were calculated from standard curves with correlation coefficients (R2) > 0.99 and analyzed using Milliplex Analyst software (version 5.1).
2.9 Statistical analysis
All data are presented as mean ± standard deviation (SD). Statistical analyses were performed using IBM SPSS Statistics 27. Differences among groups were evaluated by one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test for multiple comparisons. Data visualization was conducted using GraphPad Prism 9.5.0 (GraphPad Software, Inc., San Diego, CA, USA). Statistical significance was defined as P < 0.05 (*, P < 0.05; **, P < 0.01; ***, P < 0.001).
3 Results
3.1 Eugenol atomization inhalation improves feeding behavior and body weight in 5×FAD mice
In this study, 6-month-old 5×FAD mice and WT mice matched for age and genetic background were subjected to atomized eugenol inhalation. Changes in food intake following atomization inhalation are shown in Figs. 1B–1D. Within the first 2 h after inhalation, food intake in the CAD group was significantly lower than that in the CWT group (
P < 0.01), confirming the presence of a pronounced anorexia phenotype in this AD mouse model [
17]. Eugenol inhalation rapidly stimulated feeding behavior in mice of both genotypes. At 2 h and 4 h post-intervention, food intake in the EAD group was markedly higher than that in the CAD group (
P < 0.001). At 2 h after inhalation, food intake increased from (1.25 ± 0.14) g in the CAD group to (1.70 ± 0.12) g in the EAD group, corresponding to an absolute increase of 0.45 g and a relative increase of 36.3%. At 4 h, cumulative food intake increased from (1.69 ± 0.12) g in CAD mice to (2.19 ± 0.26) g in EAD mice, representing an absolute increase of 0.50 g and a relative increase of 29.4%. Similarly, food intake in the EWT group showed an increasing trend, although the difference compared with the CWT group did not reach statistical significance. At 12 h, the EAD group still showed higher cumulative food intake than the CAD group, increasing from (3.07 ± 0.28) g to (3.43 ± 0.41) g, with an absolute increase of 0.36 g and a relative increase of 11.9%. However, the magnitude of the difference was substantially attenuated (
P < 0.05), indicating a clear time-dependent orexigenic effect of eugenol. In WT mice, the effect of eugenol on food intake was less pronounced. Compared with CWT mice, EWT mice showed only modest increases in cumulative food intake, with increases of 0.05 g at 2 h, 0.26 g at 4 h, and 0.16 g at 12 h, corresponding to relative increases of 3.0%, 13.4%, and 4.9%, respectively. This pattern suggests that the appetite-promoting effect of eugenol was more evident under AD-associated appetite-deficient conditions.
To evaluate the physiological consequences of long-term intervention, body weight changes were monitored throughout a 4-week atomization period (Figs. 1E and 1F). During the entire experimental course, baseline body weight in WT mice (CWT and EWT) remained consistently higher than that of AD model mice (CAD and EAD), in line with the metabolic abnormalities commonly observed in AD models [
17]. After 4 weeks of intervention, CAD mice gained (2.46 ± 1.11) g, whereas EAD mice gained (3.02 ± 2.42) g, corresponding to an additional increase of 0.56 g and a relative increase of 22.8% compared with CAD mice. Body weight gain in the EAD group was significantly greater than that in the CAD group (
P < 0.01). Although body weight gain in the EWT group exceeded that of the CWT group, the difference did not reach statistical significance. Overall, body weight increased in all groups over the 4-week period. Both CWT and EWT mice exhibited rapid weight gain during the early phase of intervention, followed by a plateau. At corresponding time points, body weight in the EAD group consistently exceeded that of the CAD group with the same genotype.
3.2 Eugenol inhalation alleviates histopathological alterations in the olfactory bulb
To determine whether eugenol affects feeding behavior by altering peripheral organ structure, and to assess its impact on peripheral tissues in the AD model, gastric tissues from all groups were examined by H&E staining (Figs. 1G–1J). In the CAD group, the overall gastric mucosal architecture was intact, with well-organized epithelial cells and clearly defined gastric glands. No obvious signs of atrophy, hyperplasia, abnormal differentiation, or structural disorganization were observed. The mucosal, submucosal, and muscular layers were clearly delineated, and no significant inflammatory cell infiltration, edema, or fibrosis was detected in the interstitium. Compared with the CWT group, the CAD group did not exhibit statistically significant morphological alterations in gastric mucosa. Following eugenol atomization, gastric mucosal morphology in both the EAD and EWT groups remained comparable to that of their respective water-treated controls (CAD and CWT), with no evident structural changes or improvements. Overall, the mucosal architecture remained regular, and no new pathological features were observed.
To assess the impact of AD pathology on olfactory input pathways and the potential protective effects of eugenol, morphological analyses of the olfactory bulb were performed (Fig. S1A). In the CAD group, olfactory bulb tissues displayed pronounced AD-associated pathological alterations. Specifically, apoptotic cell levels were markedly elevated, indicating severe disruption of cellular survival homeostasis. In addition, neuronal atrophy was evident, accompanied by compromised tissue structural integrity. In contrast, eugenol-treated EAD mice exhibited a substantial attenuation of olfactory bulb pathology. Compared with the CAD group, the number of abnormally apoptotic cells was significantly reduced, neuronal atrophy was markedly alleviated, and both cellular morphology and structural integrity of the olfactory bulb were notably improved. In WT mice, olfactory bulb morphology remained normal in both the CWT and EWT groups, with no detectable apoptosis or neuronal atrophy. These findings indicate that eugenol inhalation exerts no detectable toxicity on the olfactory bulb under physiological conditions and that its protective effects are specific to AD-related pathological damage.
3.3 Eugenol modulates peripheral appetite-related hormones and neurotransmitters
Serum levels of appetite-related hormones and neurotransmitters in each group are presented in Figs. 2A–2I. Under control conditions, baseline serum ghrelin and NPY levels in the CAD group were significantly lower than those in the CWT group (P < 0.01). Following eugenol inhalation, serum ghrelin and NPY levels were robustly and synergistically upregulated in mice of both genotypes (P < 0.001). Specifically, NPY levels increased by approximately 10.4% in the EAD group relative to the CAD group and by 11% in the EWT group relative to the CWT group. Serum ghrelin levels increased to 1.17-fold and 1.11-fold of their respective control levels in the EAD and EWT groups, respectively. The effects of eugenol on AGRP exhibited genotype-dependent differences. Serum AGRP levels were significantly elevated in the EWT group compared with the CWT group (P < 0.001). In contrast, although AGRP levels in the EAD group showed a modest increase relative to the CAD group, the difference did not reach statistical significance.
At baseline, serum POMC and α-MSH levels were significantly higher in CAD mice than in CWT mice (P < 0.05). Eugenol inhalation markedly reduced POMC and α-MSH levels in both the EAD and EWT groups compared with their respective controls (P < 0.001), indicating a pronounced suppression of anorexigenic peptide expression. Serum GLP-1 levels were significantly elevated in the CAD group compared with the CWT group, whereas LEP levels did not differ significantly between the two groups (Fig. S1B). After eugenol intervention, both LEP and GLP-1 levels were significantly decreased in mice of both genotypes (P < 0.01). The reduction in LEP was more pronounced in the EWT group (45.2%) than in the EAD group (22.7%), while the decrease in GLP-1 was greater in the EAD group (16.2%) than in the EWT group (5.9%). Under basal conditions, serum DA levels were significantly lower in the CAD group than in the CWT group (P < 0.05), whereas serum 5-HT levels were significantly elevated (P < 0.001). Eugenol inhalation significantly increased DA levels while concomitantly suppressing 5-HT levels in both genotypes (P < 0.001). The reduction in 5-HT was particularly pronounced in the EAD group. In addition, eugenol intervention significantly decreased serum INS levels in both WT and 5×FAD mice. Collectively, these results demonstrate that eugenol inhalation shifts the peripheral hormonal and neurotransmitter milieu toward an orexigenic profile, characterized by reduced anorexigenic factors (POMC, α-MSH, LEP, GLP-1, 5-HT, and INS) and enhanced orexigenic signals (ghrelin, NPY, and DA).
3.4 Eugenol inhalation modulates neuronal activity in the olfactory bulb and hypothalamus
Immunohistochemical staining revealed c-Fos-positive signals as brown granular deposits (Figs. 3A and 3B). In the water-treated control groups, the number of c-Fos-positive neurons in the hypothalamus was significantly higher in CAD mice than in CWT mice, as evidenced by denser brown staining (Fig. 3A). Quantitative analyses based on the positive cell ratio, average optical density (AOD), and histological score (H-score) confirmed a markedly elevated level of neuronal activation in the CAD group compared with the CWT group (P < 0.001; Figs. 3C–3E). Following 4 weeks of eugenol atomized inhalation, the number of c-Fos-positive neurons in the hypothalamus of EAD mice was significantly reduced, accompanied by weaker staining intensity. Both AOD and H-score values were significantly lower than those of the CAD group (P < 0.01), indicating that eugenol effectively attenuated AD-associated hypothalamic neuronal hyperexcitability. In WT mice, hypothalamic c-Fos expression was also reduced in the EWT group relative to the CWT group, with significantly decreased AOD and H-score values (P < 0.001), suggesting a general calming effect of eugenol on hypothalamic neuronal activity.
As the primary relay station for olfactory sensory input, neuronal activation in the olfactory bulb was also assessed (Fig. 3B). In the water-treated controls, c-Fos expression in the olfactory bulb was lower in CAD mice than in CWT mice, indicating a hypoactive neuronal state. Following eugenol inhalation, c-Fos expression in the olfactory bulb was significantly enhanced in both AD model and WT mice. AOD and H-score values were markedly increased in the EAD and EWT groups compared with their respective controls (P < 0.01; Figs. 3F–3H), demonstrating that eugenol effectively augments olfactory bulb neuronal activity.
3.5 Eugenol reshapes the hypothalamic neuropeptide signaling pathways governing energy intake regulation
The expression of key orexigenic and anorexigenic neuropeptides in the hypothalamus was examined and quantitatively analyzed by IHC (Figs. 4A–4D). In water-treated control mice, hypothalamic neuropeptide expression in the CAD group exhibited a pronounced and functionally opposing imbalance compared with the CWT group. Specifically, the positive cell ratio and AOD of the orexigenic neuropeptides NPY and AGRP were markedly lower in CAD mice than in CWT mice (P < 0.01). Quantitative analysis showed that baseline AGRP expression in the CWT group was 2.064-fold higher than that in the CAD group (Fig. 4E). In contrast, expression levels of the anorexigenic peptide POMC and its receptor MC4R were significantly elevated in the CAD group relative to the CWT group (P < 0.01), with baseline POMC and MC4R levels being 3.421-fold and 3.139-fold higher, respectively (Fig. 4E).
After 4 weeks of eugenol inhalation, orexigenic signaling in the hypothalamus was markedly enhanced in both genotypes. In the EAD group, the positive cell ratio, AOD, and H-score values of NPY and AGRP were all significantly increased compared with the CAD group, with NPY and AGRP expression rising by approximately 7.11-fold and 2.49-fold, respectively (Figs. 4E–4G). In the EWT group, AGRP expression was also significantly elevated (P < 0.01). Concurrently, eugenol inhalation robustly suppressed anorexigenic pathways. Hypothalamic POMC and MC4R expression levels in the EAD group were significantly lower than those in the CAD group, decreasing by approximately 60.98% and 68.14%, respectively (P < 0.01). POMC expression was likewise significantly reduced in the EWT group (P < 0.01; Figs. 4E–4G). Immunohistochemical images visually corroborated these findings, showing increased numbers and intensified staining of NPY/AGRP-positive neurons, alongside diminished POMC/MC4R-positive signals following eugenol treatment (Figs. 4A–4D).
3.6 Eugenol attenuates systemic and AD-related inflammatory responses in 5×FAD mice
Alterations in the serum inflammatory cytokine profile and AD-associated neuroinflammatory markers are summarized in Figs. 5A and 5B. Compared with their respective water-treated controls, eugenol inhalation markedly reduced serum levels of multiple core pro-inflammatory mediators in both EWT and EAD mice. These included SDF-1α/CXCL12, IFN-γ, TARC/CCL17, KC/CXCL1, MIP-1β/CCL4, IL-16, MCP-1/CCL2, SCYB16/CXCL16, IP-10/CXCL10, BCA-1/CXCL13, I-TAC/CXCL11, IL-4, Fractalkine/CX3CL1, I-309/CCL1, GM-CSF, MCP-5/CCL12, TNF-α, MIP-3β/CCL19, CTACK/CCL27, MIP-1α/CCL3, IL-1β, IL-6, ENA-78/CXCL5, IL-2, and RANTES/CCL5 (Figs. 5A and 5B). Among these, I-TAC/CXCL11, IL-6, and CTACK/CCL27 were significantly downregulated (P < 0.05; Fig. 5B). This anti-inflammatory effect was consistently observed in both genotypes, indicating that eugenol exerts broad-spectrum anti-inflammatory activity independent of AD pathology. Notably, reductions in TNF-α, SCYB16/CXCL16, RANTES/CCL5, and IL-16 were observed exclusively in the EAD group (P < 0.05; Fig. 5B), whereas decreases in SDF-1α/CXCL12 and Eotaxin-2/CCL24 were specific to the EWT group (P < 0.001; Fig. 5B), suggesting pathology-dependent modulation of selected chemokines by eugenol. In addition to suppressing pro-inflammatory mediators, eugenol inhalation significantly increased the anti-inflammatory cytokine IL-10, with a pronounced upregulation observed in the EWT group (P < 0.01; Fig. S1C).
At baseline, serum GFAP levels were higher in the CAD group than in the CWT group. Following eugenol intervention, GFAP levels were markedly reduced in both the EAD and EWT groups (P < 0.001), with decreases of 39.3% and 24.7%, respectively (Fig. 5B). Consistent with this pattern, baseline serum IBA-1 levels were significantly elevated in CAD mice compared with CWT mice (P < 0.001). Eugenol inhalation resulted in a pronounced reduction in serum IBA-1 levels in both EAD and EWT mice (P < 0.001), with decreases of 14.2% and 9.3%, respectively (Fig. 5B).
3.7 Transcriptomic profiling reveals that eugenol modulates hypothalamic neuro-immune-metabolic pathways
To systematically investigate the impact of eugenol inhalation on the hypothalamic transcriptome, differential gene expression analyses were performed on hypothalamic tissues. As shown in Figs. 6A, 6B, S2A and S2B, eugenol inhalation induced distinct transcriptional alterations in WT and 5×FAD mice. KEGG pathway analysis of the EAD vs. CAD comparison revealed significant enrichment of immune-related pathways (Fig. 6D), including allograft rejection, graft-versus-host disease, and Th1/Th2 cell differentiation. GO functional enrichment further demonstrated that eugenol primarily activated immune-associated processes in 5×FAD mice, such as antigen presentation, T-cell differentiation, and cytokine-mediated signaling, while also influencing cytoskeletal organization and nucleic acid metabolic processes (Fig. 6E). These findings suggest that eugenol may exert neuroprotective effects in the hypothalamus of 5×FAD mice through modulation of immune-related gene networks. In contrast, transcriptomic changes induced by eugenol in WT mice differed markedly from those observed in the AD model. Pathway analyses indicated that eugenol predominantly regulated genes involved in retinol/isoprenoid metabolism, cytoskeletal organization, and muscle-related functions, with comparatively weaker effects on immune pathways (Figs. S2C and S2D). This pattern suggests that, under physiological conditions, eugenol primarily targets metabolic homeostasis, cellular structure, and basal immune functions.
To further delineate the specificity of eugenol-induced transcriptional regulation under physiological and pathological conditions, Venn analyses were performed on DEGs identified in the CWT vs. CAD, EWT vs. CWT, and EAD vs. CAD comparisons (Fig. 6C). Only a single gene was shared across all three comparisons, indicating minimal overlap between genes jointly regulated by AD pathology and eugenol treatment. A limited number of DEGs overlapped exclusively between the EWT vs. CWT and EAD vs. CAD comparisons, underscoring the pronounced context dependence of eugenol’s transcriptional effects in physiological versus pathological states. It is worth noting that except for one gene shared by the three groups of DEGs, 17 candidate DEGs were shared between the CWT vs. CAD and EAD vs. CAD comparisons (Fig. 6C). These genes were associated with AD pathology yet were also selectively modulated by eugenol, representing core molecular links between disease progression and therapeutic intervention. Importantly, these 17 shared candidate DEGs exhibited opposing expression patterns during disease progression and eugenol treatment (Figs. 6F and 6G), demonstrating that eugenol inhalation selectively reverses a subset of AD-associated transcriptional alterations in the hypothalamus rather than inducing nonspecific gene expression changes.
In addition, 146 DEGs were uniquely identified in the EWT vs. CWT comparison, 225 DEGs were specific to the CWT vs. CAD comparison, and 64 DEGs were uniquely detected in the EAD vs. CAD comparison. These genes were not altered by either AD pathology or eugenol treatment alone in WT mice, highlighting a disease-specific transcriptional response to eugenol intervention.
4 Discussion
Extensive clinical and preclinical evidence indicates that patients with AD and corresponding animal models exhibit early-onset body weight loss, reduced food intake, and disrupted energy homeostasis [
15–
17]. These non-cognitive manifestations are closely associated with disease progression and prognosis, yet the underlying mechanisms and safe, effective intervention strategies remain poorly defined. Current therapeutic options for AD-associated nutritional and metabolic impairments are notably limited. In this study, we systematically demonstrate that eugenol atomized inhalation improves feeding behavior, modulates neuroendocrine signaling, and reshapes hypothalamic transcriptional networks in an AD mouse model, thereby providing compelling experimental support for an olfaction-mediated, non-invasive intervention strategy.
Based on previous pharmacological evidence, the eugenol aqueous solution used in this study was prepared at a dose of 20 mg/kg. Huang et al. demonstrated that dietary supplementation with 0.01% and 0.02% eugenol promoted food intake in mice, with the highest dietary level corresponding to approximately 27 mg/kg per day [
14]. In addition, Jung et al. reported that eugenol administered orally at 10 or 30 mg/kg/day for 2 months improved cognitive impairment and reduced AD-related pathological manifestations in 5×FAD mice, including neuronal loss, Aβ deposition, necroptosis activation, and neuroinflammatory responses [
11]. Recent work by Wang et al. provided additional inhalation-route evidence by showing that inhaled eugenol is a biologically active intervention capable of producing systemic metabolic and inflammatory effects [
12]. Therefore, the 20 mg/kg dose used in the present study was selected as a moderate dose within the previously reported biologically active range.
However, because atomized inhalation differs from oral administration in aerosol deposition, respiratory absorption, and systemic bioavailability, the body weight-normalized dose prepared for atomization cannot be directly equated with the actual inhaled or systemically absorbed dose. Although recent evidence suggests that repeated inhalation of eugenol can exert systemic metabolic effects in mice [
12], future studies should include dose–response experiments, aerosol exposure quantification, and pharmacokinetic analysis to determine the minimal effective dose, optimal exposure range, and potential toxicity threshold of eugenol atomized inhalation.
Here, eugenol inhalation rapidly increased food intake in both WT and 5×FAD mice and selectively promoted long-term body weight gain in AD mice. Notably, these effects were substantially more pronounced in the AD model than in WT controls, highlighting a strong pathology-dependent efficacy. This observation suggests that eugenol is not a nonspecific appetite stimulant but instead preferentially targets dysregulated feeding circuits under AD-related pathological conditions. Consistent with this interpretation, histological analysis of gastric tissue revealed no overt morphological alterations in the gastric mucosa attributable to either AD pathology or eugenol treatment, indicating that the improvement in feeding behavior is unlikely to arise from peripheral digestive system remodeling. Rather, these findings support the prevailing view that AD-associated anorexia is primarily driven by central regulatory dysfunction [
19]. Further histological evidence reinforced this central mechanism. Odor information originates in the olfactory epithelium, where millions of olfactory sensory neurons transduce chemical cues into electrical signals that are relayed to higher brain regions. The olfactory bulb serves as a critical gateway for food perception and the generation of feeding motivation [
20,
21]. In the present study, 5×FAD mice exhibited marked apoptosis and neuronal atrophy in the olfactory bulb, whereas eugenol inhalation significantly ameliorated these pathological features. These findings imply that eugenol may preserve olfactory bulb neuronal integrity, thereby restoring sensory input related to feeding behavior at its origin.
The hypothalamus constitutes a central hub that integrates peripheral hormones, neurotransmitters, and sensory signals to regulate energy homeostasis [
7]. Within the arcuate nucleus, POMC neurons exert anorexigenic effects through cleavage into α-MSH, which suppresses food intake primarily via activation of MC4R [
22]. In contrast, AGRP neurons are activated during energy deficiency and represent one of the most potent orexigenic drivers of feeding behavior [
23,
24], while NPY rapidly but transiently stimulates food intake [
25,
26]. Previous studies have shown that eugenol can modulate this axis via the TRPV1-mediated CaMKK2/AMPK signaling pathway [
14]. In line with these reports, our serum analyses and immunohistochemical data consistently revealed a characteristic imbalance in 5×FAD mice, marked by attenuated orexigenic signaling (reduced NPY and AGRP) and enhanced anorexigenic signaling (elevated POMC and MC4R). Eugenol inhalation reduced multiple anorexigenic hormones and neurotransmitters while augmenting orexigenic factors in both genotypes, with substantially greater effects under AD pathological conditions. In particular, eugenol robustly upregulated hypothalamic NPY and AGRP while concomitantly suppressing POMC and MC4R in 5×FAD mice, indicating a coordinated correction of AD-associated energy dysregulation at the neuropeptide network level. When integrated with peripheral hormonal changes, this central-peripheral synergy provides a coherent biological framework for the observed improvements in feeding behavior.
Analysis of c-Fos expression further revealed region-specific imbalances in neuronal activity under AD pathology. The hypothalamus of 5×FAD mice exhibited sustained neuronal hyperactivation, whereas the olfactory bulb displayed pronounced hypoactivity. This combination of hypothalamic overexcitation and olfactory hyporesponsiveness may reflect aberrant energy sensing and impaired feeding drive in AD. Eugenol inhalation exerted bidirectional yet functionally complementary effects, suppressing excessive hypothalamic activation while enhancing olfactory bulb neuronal responsiveness. Such rebalancing of neuronal activity suggests that eugenol does not act merely as a generalized neuronal depressant or stimulant; rather, it restores physiological activation dynamics by reconfiguring functional connectivity along the sensory-hypothalamic axis.
Chronic neuroinflammation is a hallmark of AD and is increasingly recognized as a contributor to hypothalamic dysfunction and metabolic disturbance [
17]. Pro-inflammatory cytokines can directly inhibit AGRP/NPY neuronal activity while potentiating POMC signaling, thereby inducing anorexia-like phenotypes [
17,
27]. Eugenol inhalation exerted broad anti-inflammatory effects, reducing circulating levels of multiple pro-inflammatory cytokines and chemokines [
14]. Notably, the upregulation of the anti-inflammatory cytokine IL-10 was more prominent in WT mice, consistent with previous reports of eugenol’s immunomodulatory properties [
28–
31]. Moreover, the marked reductions in GFAP and IBA-1 levels indicate attenuation of astrocytic and microglial activation [
32,
33]. The anti-inflammatory and neuroprotective effects observed in the present study are consistent with previous reports on eugenol. Previous studies have shown that eugenol improves cognitive function, reduces Aβ deposition, inhibits necroptosis, and modulates microglial phagocytosis and inflammation in 5×FAD mice [
11]. In addition, Huang et al. demonstrated that eugenol promotes appetite through the TRPV1–CaMKK2/AMPK signaling pathway [
14]. Wang et al. further reported that inhaled eugenol can exert systemic metabolic and anti-inflammatory effects [
12]. Compared with these previous studies, the present work provides complementary evidence that atomized eugenol inhalation improves AD-associated appetite loss in parallel with reduced inflammatory activation and altered olfactory–hypothalamic activity. These findings suggest that eugenol exerts multi-level regulatory effects involving neuroprotection, inflammation modulation, metabolic regulation, and appetite control.
Although the present study focused primarily on appetite regulation and metabolic homeostasis, the potential relevance of eugenol inhalation to cognitive function should also be considered. Cognitive impairment is the core clinical manifestation of AD, and previous studies have reported that eugenol can alleviate AD-related cognitive deficits in 5×FAD mice [
11]. In the present study, eugenol atomized inhalation reduced systemic inflammatory mediators and AD-related glial activation markers, including GFAP and IBA-1, while restoring hypothalamic appetite-regulatory neuropeptide balance and improving neuronal activity patterns in the olfactory bulb and hypothalamus. These effects may be relevant to cognitive function because chronic neuroinflammation, negative energy balance, and metabolic dysfunction are recognized contributors to AD progression [
3,
4]. In this context, the improvement of appetite and body weight may not only reflect metabolic recovery but may also help reduce the vulnerability of the AD brain to inflammatory and energetic stress. Moreover, the olfactory bulb is an early-affected structure in AD and is functionally connected with limbic and hypothalamic circuits; therefore, the restoration of olfactory bulb activity after eugenol inhalation may have broader implications for neural circuit function beyond feeding behavior. These findings also suggest that eugenol inhalation may have broader relevance for AD-related neurobehavioral dysfunction, although its effects on cognitive performance require direct validation in future studies.
Hypothalamic transcriptomic profiling provided molecular-level clues supporting the physiological and histological observations. Eugenol induced distinct gene expression programs in WT and 5×FAD mice. In WT animals, DEGs were predominantly associated with metabolic regulation, cytoskeletal organization, and homeostatic maintenance, consistent with adaptive modulation rather than therapeutic reprogramming. In contrast, eugenol markedly influenced immune-, inflammatory-, and stress-related gene networks in the AD model. Importantly, Venn analysis identified 17 shared candidate DEGs that exhibited opposite expression trends during AD progression (CWT vs. CAD) and eugenol intervention (EAD vs. CAD). Such inverse regulation is widely regarded as a molecular hallmark of effective therapeutic intervention, indicating that eugenol selectively reverses specific AD-associated transcriptional abnormalities rather than inducing nonspecific perturbations. To strengthen the biological interpretation of these candidates, we further summarized their functional annotations and literature-supported relevance to AD, neuroinflammation, appetite regulation, and metabolic homeostasis in Supplementary Table S1.
Among the 17 reversed candidate genes,
Gdf15 and
Il5 are particularly relevant to the neuro-immune-metabolic framework of this study. GDF15 is a stress-responsive cytokine associated with mitochondrial stress, inflammation, aging, and neurodegenerative processes [
34,
35]. Functionally, the GDF15–GFRAL signaling axis has been implicated in appetite suppression, body weight regulation, and aversion-related responses [
36,
37], suggesting that AD-associated dysregulation of
Gdf15 may be relevant to anorexia and negative energy balance [
38,
39]. Therefore, the eugenol-induced reversal of
Gdf15 expression is consistent with the observed restoration of appetite-related neuropeptide balance and attenuation of inflammatory activation. IL-5, in contrast, is a type 2 immune cytokine involved in Th2/ILC2-associated immune regulation. Emerging evidence suggests that IL-5-related immune responses may influence neuroinflammation and neurological outcomes in aging or AD-related contexts [
40]. Thus, the reversal of
Il5 expression may reflect an immunoregulatory component of eugenol’s effect on the hypothalamic microenvironment.
Nevertheless, the transcriptomic findings in this study remain exploratory. The expression changes of the 17 reversed AD-associated candidate genes, including Gdf15 and Il5, require future validation by qPCR, Western blotting, immunohistochemistry, or spatial/cell type-specific approaches. In addition, causal relationships between individual genes and feeding behavior remain to be established. Future studies employing targeted molecular validation, cell type-specific manipulation, and functional behavioral assays will be necessary to delineate the precise roles of these candidate genes and related neuronal or immune cell populations in mediating the effects of eugenol inhalation.
Another limitation is that comprehensive long-term inhalation safety was not systematically evaluated. In the present study, H&E staining did not reveal overt pathological alterations in the stomach or olfactory bulb after eugenol atomization, and no obvious treatment-related mortality or severe behavioral abnormality was observed during the 4-week intervention. However, serum markers of hepatorenal function, including ALT, AST, BUN, and creatinine, as well as histological assessment of the liver, kidney, lung, nasal mucosa, and trachea, were not included. Therefore, future studies should evaluate respiratory tract irritation, pulmonary inflammation, nasal epithelial integrity, and hepatorenal toxicity after repeated eugenol inhalation.
5 Conclusions
In conclusion, this study demonstrates that eugenol atomized inhalation effectively improves feeding behavior and mitigates weight loss in a 5×FAD mouse model of Alzheimer’s disease. These effects are not attributable to peripheral gastrointestinal structural changes. Instead, they are driven by central mechanisms. These include preservation of olfactory bulb integrity and normalization of hypothalamic neuronal activity. They also involve restoration of orexigenic and anorexigenic neuropeptide signaling and suppression of systemic and AD-related neuroinflammation. Importantly, hypothalamic transcriptomic analyses show that eugenol selectively reverses a subset of AD-associated gene expression changes. This finding highlights a disease-context-dependent mode of action rather than nonspecific stimulation. Together, these findings establish a mechanistic framework in which eugenol inhalation modulates the olfactory-hypothalamic-immune axis to rebalance energy homeostasis under AD pathology. This work provides experimental evidence supporting olfaction-based aromatherapy as a safe, non-invasive, and mechanistically grounded adjunct strategy for alleviating metabolic disturbances associated with AD and lays a foundation for future translational and clinical investigations.
The Author(s) 2026. This article is published by Higher Education Press.
PDF
(16602KB)
