Obesity-induced Nerve Degeneration and Inflammation: Therapeutic Effects of β-Hydroxybutyrate and Melatonin on Pyroptosis, ER Stress, and Hippocampal Dysfunction in High-Fat Diet-Fed Rats

Abstract

Obesity-induced neurodegeneration is mediated by multiple pathological mechanisms, including inflammation, oxidative stress, pyroptosis, and endoplasmic reticulum (ER) stress, with pyroptosis and ER stress emerging as central contributors. This study investigated the neuroprotective efficacy of β-hydroxybutyrate (BHB) and melatonin, both individually and in combination, in mitigating these pathways and attenuating obesity-associated neurodegeneration. Thirty male Sprague-Dawley rats were allocated into five groups over a 7-week experimental period: (1) Control (standard diet); (2) High-fat diet (HFD), administered daily via oral gavage as an emulsion; (3) HFD + BHB (4% in drinking water); (4) HFD + melatonin (10 mg/kg/day in drinking water); and (5) HFD + BHB + melatonin. Post-treatment, blood and brain tissues were analyzed for biochemical markers, histopathological changes, and mRNA expression levels of pyroptosis-related (NLRP3, Caspase-1, Caspase-3, Caspase-11, GSDMD, NF-κB, IL-1β, IL-10) and ER stress-related (BIP, CHOP, sXBP1, TXNIP) genes via quantitative RT-PCR. Co-administration of BHB and melatonin significantly elevated serum ketone levels, ameliorated HFD-induced dyslipidemia, and attenuated weight gain (p < 0.001). Histopathological analysis revealed preserved neuronal density, volume, and architecture in the hippocampal CA1 and dentate gyrus (DG) regions, alongside reduced neurodegeneration (p < 0.001). The combined intervention suppressed neuroinflammatory markers and oxidative stress while elevating anti-inflammatory IL-10 levels. Furthermore, it downregulated pyroptosis-related (NLRP3, Caspase-1, Caspase-3, Caspase-11, GSDMD, NF-κB, IL-1β) and ER stress-related (BiP, CHOP, sXBP1, TXNIP) gene expressions (p < 0.001), suggesting a dual mechanism of action. BHB and melatonin demonstrate significant potential in counteracting obesity-driven neurodegeneration through synergistic anti-inflammatory, anti-pyroptotic, and ER stress-modulating effects. Their combined administration presents a multi-targeted therapeutic strategy to preserve neuronal integrity and mitigate brain damage in obesity. These findings corroborate prior evidence and underscore the translational relevance of BHB and melatonin in managing obesity-related neural pathologies.

Introduction

Obesity constitutes a significant global health challenge, impacting millions worldwide and increasingly recognized as a risk factor for neurodegenerative disorders, such as Alzheimer’s disease (AD) [1, 2]. Traditionally linked to metabolic conditions including diabetes and cardiovascular disease, obesity also profoundly affects brain health by promoting cognitive decline, memory impairment, and heightened vulnerability to neurodegeneration [3, 4]. This complex interplay involves systemic inflammation, oxidative stress, insulin resistance, and disrupted cellular functions, with programmed cell death processes, primarily pyroptosis and endoplasmic reticulum (ER) stress emerging as critical pathways mediating obesity-related neural damage [3, 5].

Pyroptosis is a distinct form of programmed cell death characterized by inflammatory responses, triggered by the activation of caspases such as caspase-1, caspase-3, and caspase-11 in response to cellular stress or infection [6]. Caspase-1 associates with the NLRP3 inflammasome, which, through cleavage of gasdermin D (GSDMD), induces pore formation in the cell membrane, resulting in loss of membrane integrity and release of pro-inflammatory cytokines such as interleukin-1β (IL-1β) [7]. Caspase-11 is activated by toxic molecules like lipopolysaccharides, while caspase-3 modulates both apoptotic and pyroptotic pathways [8, 9].

In parallel, ER stress arises when the protein-folding capacity of the ER is overwhelmed, leading to accumulation of unfolded or misfolded proteins and activation of the unfolded protein response (UPR) via sensors such as protein kinase R–like endoplasmic reticulum kinase (PERK), inositol-requiring enzyme 1 alpha (IRE1α), and activating transcription factor 6 (ATF6) [10]. Key molecular components including BiP (binding immunoglobulin protein), sXBP1 (spliced X-box binding protein 1), and the pro-apoptotic protein CHOP (C/EBP homologous protein) orchestrate attempts to restore homeostasis [11]; however, chronic ER stress, as observed in obesity, precipitates neuronal damage and accelerates neurodegenerative processes. Importantly, ER stress and pyroptosis are interrelated: ER stress can activate the NLRP3 inflammasome, thereby inducing pyroptosis, while pyroptosis exacerbates ER dysfunction through inflammatory mediator release [12], establishing a deleterious feedback loop particularly detrimental to neuronal survival [13].

Thioredoxin Interacting Protein (TXNIP) also serves as a pivotal regulator linking ER stress and inflammation by binding to thioredoxin, thereby modulating oxidative stress and triggering inflammatory cascades that culminate in cellular dysfunction and neuronal apoptosis [14]. Given the detrimental impact of pyroptosis and ER stress on obesity-associated brain injury, therapeutic strategies targeting these pathways are under active investigation. Beta-hydroxybutyrate (BHB), an endogenous ketone body produced during fasting or ketogenic diets, has demonstrated neuroprotective effects by serving as an alternative brain energy substrate and exerting anti-inflammatory actions [15].

Similarly, melatonin (MEL), widely recognized for its role in circadian regulation, functions as a potent antioxidant and anti-inflammatory agent in the brain, enhancing mitochondrial function and modulating inflammatory pathways [16].

Moreover, the balance between pro-inflammatory and anti-inflammatory cytokines plays a critical role in regulating neuroinflammation; interleukin-10 (IL-10), a potent anti-inflammatory cytokine, serves as a key modulator by suppressing excessive immune responses and inhibiting inflammasome activation. Reduced IL-10 levels, as observed in obesity, may exacerbate pyroptosis and ER stress–mediated neuronal damage. Conversely, enhancing IL-10 expression can counteract the effects of pro-inflammatory mediators such as IL-1β, thereby preserving neuronal integrity and promoting a protective environment within the brain [17].

Despite evidence supporting the individual neuroprotective roles of BHB and MEL, their combined effects on obesity-induced brain damage remain underexplored. The current study thereby aimed to elucidate how BHB and MEL attenuate pyroptosis and ER stress in the brains of obese rats by examining key molecular markers including caspases, GSDMD, NLRP3, IL-1β, IL-10, BiP, TXNIP, sXBP1, and CHOP. Understanding these interactions may inform novel therapeutic approaches to prevent or delay obesity-related neurodegenerative disorders.

Materials and Methods

Experimental Design

In this experiment, thirty male Sprague-Dawley rats (RRID: MGI:5651135) were utilized, each with an average weight of 160 ± 10 g. Parenthetically, male rats were selected due to documented sex-specific differences in metabolic responses to high-fat diets (HFD) and compensatory energy expenditure mechanisms, which are more consistent in male rodents. The animals were randomly assigned into five groups, each containing six rats, and housed in standard laboratory conditions, included a controlled temperature of 21 °C ± 2 °C, relative humidity of 60% ± 10%, and a 12-hour light/dark cycle. Throughout the study, all animals had ad libitum access to food and water. The dietary and treatment regimens for each group were as follows: (1) Control Group: Rats were fed a control diet for seven weeks. This diet adhered to the AIN-93 M standards, comprising 10% energy from lipids, 14% from proteins, and 76% from carbohydrates, with an energy density of 3.58 kcal/g. (2) HFD Group: Rats received an HFD providing 50% of its energy from lipids (10% from soybean oil and 40% from sheep’s tail fat), 14% from proteins, and 36% from carbohydrates, with an energy density of 5.01 kcal/g (Table 1) [18]. (3) HFD-BHB Group: Rats were fed an HFD supplemented with BHB for seven weeks. The BHB was administered as a 4.2% (w/v) solution in their drinking water [19]. This isotonic solution-maintained sodium and potassium concentrations at 0.45% each. The rats had unrestricted access to this drinking solution. (4) HFD-MEL Group: HFD-fed rats received MEL supplementation at a dose of 10 mg/kg/day [20]. Melatonin was dissolved in a minimal volume of absolute ethanol, then diluted in drinking water to achieve the required dosage, corresponding to a final ethanol concentration of 0.066% (w/v). (5) HFD-BHB + MEL Group: Rats were fed an HFD and simultaneously received both BHB (4.2% (w/v)) and MEL (10 mg/kg/day) in their drinking water. Fresh solutions of BHB and melatonin were prepared biweekly, and melatonin dosages were adjusted according to the animals’ body weight. Water bottles for this group were shielded from light using aluminium foil to protect melatonin from degradation.

Table 1.

Ingredients and energy contents of standard diet and high-fat diet

Ingredients	Unit	Standard Diet	High-fat Diet
Dietary casein	g per kg of diet	140.0	175.0
Corn-derived starch	g/kg of diet	620.7	347.7
Sucrose	g/kg of diet	100.0	100.0
Rendered animal fat	g/kg of diet	—	338.0
Soy oil	g/kg of diet	40.0	40.0
Dietary fiber	g/kg of diet	50.0	50.0
Vitamin supplement	g/kg of diet	10.0	10.0
Mineral supplement	g/kg of diet	35.0	35.0
Cystine	g of diet	1.8	1.8
Choline	g/kg of diet	2.5	2.5
Antioxidant additive	g/kg of diet	0.008	0.008
Total formulation weight	g	1000	1000
Total caloric value	kcal	3802.8	5000
Energy from carbohydrates	%	76	36
Energy from proteins	%	14	14
Energy from fats	%	10	50

Following BHB and BHB + MEL treatments, blood ketone levels increased by approximately 33%, from a baseline of 0.6 mM to 0.9 mM. After seven weeks of treatment, the animals were sedated using ketamine/xylazine anesthesia (100 mg/kg ketamine and 10 mg/kg xylazine, Merck, Germany) and euthanized. Then, the animals were prepared for sampling.

Sample Preparation

Blood samples were collected via cardiac puncture for subsequent biochemical analyses. Total mRNA was extracted from the hippocampus tissue. The remaining tissues were preserved in 10% formalin buffer for histopathological examinations. All experimental procedures were approved by the Medical Ethics Committee of Shiraz University of Medical Sciences and strictly adhered to the established ethical guidelines for the care and use of animals (IR.SUMS.ACE.1403.056).

Measurement of Biochemical Indices

Blood samples were centrifuged at 3000 rpm for 15 min to isolate serum for biochemical analysis. Serum levels of beta-hydroxybutyrate BHB, fasting blood glucose (FBG), and a complete lipid profile, including triglycerides (TG), total cholesterol (Chol), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) were measured using a Pars Azmoon diagnostic kit (Tehran, Iran). These analyses were conducted according to the manufacturer’s protocols on a Perestig24i biochemical auto-analyser.

qRT-PCR Gene Expression Analyses

The mRNA expression levels of genes related to specific physiological functions, such as caspase 1 (CASP1), CASP3, CASP11, GSDMD, NLRP3, IL-1β, IL-10, BIP, CHOP, sXBP1, and TXNIP, were analyzed using qRT-PCR on an ABI 7500 Real-time PCR system (RRID: SCR_019334). Total RNA was extracted using the TRIzol (YTzol) reagent kit (Yekta Tajhiz Azma, Tehran, Iran). Complementary DNA (cDNA) was synthesized from the extracted RNA using a cDNA synthesis kit (Yekta Tajhiz Azma, Tehran, Iran). Amplification was carried out using SYBR Green master mix (Amplicon, Denmark). Each sample was processed in triplicate, and average values were used for statistical analysis. Negative controls with no template (NTC) were prepared using deionized distilled water (ddH2O). Gene expression was quantified using the comparative Ct method (2^−ΔΔCt), with β-actin serving as the internal housekeeping gene for normalization. Primer sequences for target genes were specifically designed using the Allele ID software (version 7.73) and are detailed in Table 2.

Table 2.

Primers’ sequences and accession numbers

Gene	Forward primer	Reverse primer
GSDMD	CCAGACAGCTTGAAGAGACCCT	GTTCTGGTTCTGGAGCACTGG
TXNIP	CTAGTGATTGGCAGCAGGTCG	TCTGGGGTGTCTGGGATGTT
IL-1β	CCTCTGTGACTCGTGGGATGA	GTCGTTGCTTGTCTCTCCTTGT
CASP1	GCTTCAGTCAGGTCCATCA	GACGTGTACGAGTGGGTGTT
CASP3	GGAACGCGAAGAAAAGTGACC	GTACCATTGCGAGCTGACATTC
CASP11	GGTGCGAAAGAACTGAGGC	GCATTTTGCCCACTTGGTTG
NLRP3	AGTGTGTTTTCCCAGACCC	GGTTGGTGCTGAGACTTGAG
NFκB	CCAGCACCAAGACCGAAGCAA	TTCACATCTCCCGTAACCGC
IL-10	CCCAGAAATCAAGGAGCA	CTGCTCCACTGCCTTGCTT
BIP	CAGCCCACCGTAACAATCAAG	TCCTGTCCCTTTGTCTTCAGC
CHOP	TACACCACCACACCTGAAAGC	GCAGGGTCAAGAGTAGTGAAG
SXBIP	GGACACGCTTGGGGATGAATG	CTGCACCTGCTGCGGACT
β-actin	CACACCCGCCACCAGTTCG	ACCCATTCCCACCATCACACC

Histopathological Processing and Evaluations

Upon completion of the experimental period, animals were deeply anesthetized and subjected to transcardial perfusion with phosphate-buffered saline (PBS), followed by fixation with 4% paraformaldehyde. Following euthanasia and transcardial perfusion, the brains were carefully removed and post-fixed in paraformaldehyde overnight to ensure optimal tissue preservation. The hippocampal tissue was then rapidly dissected under ice-cold conditions within approximately five minutes to minimize RNA degradation. Dissection was performed under a stereomicroscope using fine microspatulas: the cortex was gently peeled laterally to expose the hippocampus, which was carefully lifted out to avoid mechanical damage. Strict time control during perfusion and tissue handling was maintained to preserve both histological integrity and molecular quality. Coronal sections encompassing the hippocampal formation were subsequently prepared using a microtome, with section thickness standardized according to the established protocol. Brain sections were processed using routine hematoxylin and eosin (H&E) staining procedures to facilitate the evaluation of general histopathological features. Following staining, sections were dehydrated, cleared, and mounted on slides for subsequent microscopic (Nikon Eclipse E200) examination. Histological assessments were concentrated on the hippocampal formation, specifically targeting the dentate gyrus (DG) and CA1 subfield. Sections were examined under light microscopy at magnifications of X100 and X400. Neuronal density, presence of pyknotic nuclei, laminar organization, neuronal ballooning and vacuolization, and general cytoarchitectural integrity were then systematically evaluated. Representative photomicrographs were acquired for each experimental group. Scale bars were set at 100 μm for images captured at X100 magnification and 20 μm for those at X400 magnification.

Statistical Analyses

The GraphPad Prism (RRID: SCR_002798) (version 6.0; GraphPad Software, San Diego, CA, USA) was utilized for statistical evaluations. Prior to employing inferential statistics, the datasets underwent initial scrutiny for adherence to normal distribution and homogeneity of variances, ensuring the suitability of parametric tests. Subsequently, one-way analysis of variance (ANOVA) was performed to determine if significant differences existed among the various experimental groups. When a significant F-statistic was observed from the ANOVA, specific pairwise distinctions between groups were further elucidated via Tukey’s post-hoc multiple comparison test. All quantitative measurements are consistently reported as the mean ± standard deviation (SD). Statistical significance was pre-established at a probability threshold of p < 0.05. To maintain rigorous statistical interpretation, all analyses were conducted using a two-tailed approach, thereby minimizing the potential for Type I errors.

Results

Improving Effects of β-Hydroxybutyrate and Melatonin on Rats’ Body Weight

Figure 1 illustrates the weekly changes in body weight of the experimental rats throughout the study period. Statistical analyses revealed a significant interaction between the treatment type and time on body weight (p < 0.0001). Rats in the HFD group showed a marked increase in weight starting from the second week compared to the control group (p < 0.001). In contrast, weight gain was slower in the groups treated with either BHB or MEL. From the fourth week onward, both the HFD-BHB and HFD-MEL groups exhibited a significant reduction in body weight relative to the HFD group (p < 0.01 and p < 0.05, respectively). Notably, the combinatory treated group (HFD-BHB + MEL) demonstrated a significant decrease in weight as early as the second week (p < 0.001). By the end of the experiment, the average body weight of the HFD-BHB + MEL group was significantly lower than that of the HFD-BHB and HFD-MEL groups (p < 0.01 for both).

Fig. 1.

Weekly changes in rats’ body weight over the seven-week experimental period. A two-way repeated-measures ANOVA, followed by Tukey’s post-hoc test, revealed a significant interaction between treatment type and time on body weight (F(24,150) = 9.59, p < 0.0001, n = 6 per group). Symbols represent statistical significance as follows: (*) indicates a significant difference between the HFD group and the control group, while (#) represents a significant difference between the HFD group and the HFD-BHB group. (✝) shows a significant difference between the HFD-BHB group and the HFD-MEL group, and (§) highlights a significant difference between the HFD-BHB + MEL group and the HFD group. Levels of significance are denoted as follows: * or # or § or ✝ for p < 0.05, ** or ## or §§ or ✝✝ for p < 0.01, *** or ### or §§§ or ✝✝✝ for p < 0.001, and **** or #### or §§§§ or ✝✝✝✝ for p < 0.0001. Group abbreviations: HFD – rats on a high-fat diet; HFD-BHB – HFD rats supplemented with β-hydroxybutyrate for 7 weeks; HFD-MEL – HFD rats supplemented with melatonin for 7 weeks; HFD-BHB + MEL – HFD rats treated with a combination of β-hydroxybutyrate and melatonin for 7 weeks

Effects of β-Hydroxybutyrate and Melatonin on Fasting Blood Glucose and Lipid Profile

Table 3 reports the effects of different treatments on FBG and lipid indices across the experimental groups. Rats fed an HFD (i.e., HFD group) exhibited a remarkable increase in FBG compared to the control group (145 ± 12 mg/dl vs. 82 ± 3 mg/dl, p < 0.001). Treatment with either BHB or MEL significantly reduced FBG levels relative to the HFD group (93 ± 11 mg/dl, p < 0.01 and 102 ± 13 mg/dl, p < 0.05, respectively). Notably, the combined treatment of BHB and MEL produced a further reduction in FBG (97 ± 11 mg/dl, p < 0.05), indicating an additive effect of the two compounds.

Table 3.

Biochemical parameters: fasting blood glucose and lipid profile

Groups → Index ↓	Control	HFD	HFD-BHB	HFD-MEL	HFD-BHB + MEL
FBG (mg/dl)	82 ± 3	145 ± 12***	93 ± 11##	102 ± 13#	97 ± 11§✝
TG (mg/dl)	45 ± 4	79 ± 6**	52 ± 5##	60 ± 4#	38 ± 3§
TC (mg/dl)	50 ± 3.9	92 ± 5.6***	70 ± 4.9##	72 ± 6.4#	54 ± 4.5§✝
HDL (mg/dl)	42 ± 4.1	32 ± 2.3 **	38 ± 3.4#	40 ± 2##	40 ± 2.8
LDL (mg/dl)	36 ± 4	67 ± 6 ***	54 ± 4#	50 ± 4.8##	46 ± 3.1§

• Rat groups Clarification:

HFD: High-fat diet-fed rats

HFD-BHB: HFD rats administered β-hydroxybutyrate (BHB) over 7 weeks

HFD-MEL: HFD rats treated with melatonin (MEL) for 7 weeks

HFD-BHB + MEL: HFD rats receiving combined BHB and MEL therapy for 7 weeks

• Statistical annotations were applied as follows:(*) Significant difference (p < 0.05) between control and HFD groups(#) Significant difference between HFD and intervention groups (HFD-BHB, HFD-MEL, HFD-BHB + MEL)(✝) Significant difference between HFD-BHB versus HFD-MEL or HFD-BHB + MEL groups(§) Significant difference between HFD-MEL and HFD-BHB + MEL groups

• Significance thresholds were stratified:Single symbol (*, #, ✝, §): p < 0.05Double symbols (**, ##, ✝✝, §§): p < 0.01Triple symbols (***, ###, ✝✝✝, §§§): p < 0.001All quantitative data are presented as mean ± standard deviation (SD), with a sample size of n = 6 per group.

Similarly, TG and TC levels were significantly elevated in the HFD group compared to the controls (TG: 79 ± 6 vs. 45 ± 4 mg/dl, p < 0.01; TC: 92 ± 5.6 vs. 50 ± 3.9 mg/dl, p < 0.001). Both BHB and MEL treatments significantly lowered TG and TC levels, with BHB showing a slightly greater effect than MEL (TG: 52 ± 5 mg/dl, p < 0.01 for HFD-BHB vs. 60 ± 4 mg/dl, p < 0.05 for HFD-MEL; TC: 70 ± 4.9 mg/dl, p < 0.01 for HFD-BHB vs. 72 ± 6.4 mg/dl, p < 0.05 for HFD-MEL). The HFD-BHB + MEL exhibited the most substantial improvement, reducing TG and TC levels to near-normal values (TG: 38 ± 3 mg/dl, p < 0.05; TC: 54 ± 4.5 mg/dl, p < 0.05).

Regarding HDL, a significant decrease was observed in the HFD group compared to the control group (32 ± 2.3 vs. 42 ± 4.1 mg/dl, p < 0.01). Treatment with BHB or MEL partially restored HDL levels (38 ± 3.4 mg/dl and 40 ± 2 mg/dl, respectively), with the combined treatment maintaining levels comparable to the HFD-MEL group (40 ± 2.8 mg/dl).

For LDL, the HFD group showed a significant increase relative to the controls (67 ± 6 vs. 36 ± 4 mg/dl, p < 0.001). Both BHB and MEL reduced LDL levels significantly (54 ± 4 mg/dl, p < 0.05 for HFD-BHB; 50 ± 4.8 mg/dl, p < 0.01 for HFD-MEL). The combined treatment had the greatest effect, lowering LDL to 46 ± 3.1 mg/dl (p < 0.05).

Collectively, these results indicate that BHB and MEL, particularly when administered together, effectively mitigate HFD-induced hyperglycemia and dyslipidemia by reducing FBG, TG, TC, and LDL levels while improving HDL concentrations. This suggests that the combined therapy offers a synergistic benefit in improving systemic metabolic health in obesity-induced rats.

Melatonin and β-Hydroxybutyrate Suppressed the Expression of Pyroptosis-Related Genes

The expression analysis of pyroptosis genes, including CASP1, CASP3, CASP11, GSDMD, NLRP3, NF-κB, and IL-1β, in the hippocampal tissue of HFD-fed rats demonstrated a significant increase compared to the control group (p < 0.0001). Notably, the expression of these genes was significantly reduced in obese rats treated with either BHB or MEL compared to untreated obese rats (p < 0.05). Crucially, the combined treatment (BHB + MEL) produced a more pronounced effect than either treatment alone, restoring the expression of these genes to near-normal levels, with no significant difference compared to the control group, suggesting a potential synergistic interaction between the two compounds.

As shown in Fig. 2, expression of the anti-inflammatory gene IL-10 was reduced in the obese rat group compared to the control group, although this reduction was not statistically significant. However, a significant increase in IL-10 expression was observed in obese rats treated with either BHB or MEL alone (p < 0.05). Remarkably, the combined BHB + MEL treatment elevated IL-10 expression to approximately nine-fold higher than in untreated obese rats (p < 0.001), indicating a synergistic enhancement of the anti-inflammatory response when both compounds were administered together.

Fig. 2.

mRNA expression levels of pyroptotic genes in the hippocampus of the obese rats. The analysed genes are represented as follow: A: CASP1; B: CASP3; C: CASP11; D: NLRP3; E: GSDMD; F: IL-1β; G: IL-10; and H: NFκB. All values are presented as mean ± standard deviation (n = 6). The rats on a high-fat diet (HFD) received various treatments: those treated with β-hydroxybutyrate for 7 weeks are labelled as HFD-BHB; those treated with melatonin for the same duration are labelled as HFD-MEL; and those receiving a combination of β-hydroxybutyrate and melatonin for 7 weeks are referred to as HFD-BHB + MEL

Melatonin and β-Hydroxybutyrate Reduced the Expression of ER Stress-Related Genes

As depicted in Fig. 3, the expression of BIP and CHOP and TXNIP remarkably upregulated in the brain of HFD-fed rats compared to the controls (p < 0.001). MEL or BHB treatment markedly reduced the expression of these markers in HFD-fed rats as compared to HFD rats (p < 0.01). The combination of MEL and BHB could reverse the expression of mentioned genes near to control since no significant difference was found in this group as compared to control. Interestingly, the expression of sXBP1 was noticeably downregulated in HFD group as compared to the control (p < 0.05). Using MEL or BHB reversed its expression to normal. The combined treatment of BHB and MEL indicated even more increase in the sXBP1 expression compared to each compound separately (p < 0.01).

Fig. 3.

mRNA expression levels of endoplasmic reticulum stress-related genes in the hippocampus of the obese rats. Panels A, B, C and D correspond to the genes TXNIP, sXBP1, BIP and CHOP, respectively. The rats on a high-fat diet (HFD) received various treatments: those treated with β-hydroxybutyrate for 7 weeks are labelled as HFD-BHB; those treated with melatonin for the same duration are labelled as HFD-MEL; and those receiving a combination of β-hydroxybutyrate and melatonin for 7 weeks are referred to as HFD-BHB + MEL. All values are presented as mean ± standard deviation (n = 6)

Histopathological Findings in the Hippocampal Formation

Histological analysis of the hippocampal formation, specifically the DG and CA1 subfields, was performed across the experimental groups. Representative photomicrographs stained with H&E at magnifications of X100 and X400 are presented in Fig. 4. The BHB-treated group (Fig. 4A and B) demonstrated moderate preservation of hippocampal architecture. While the overall laminar organization of the CA1 and DG was largely maintained, a reduction in neuronal density was evident compared to controls. Occasional pyknotic nuclei were observed, suggesting some degree of neuronal injury; however, the general structural integrity remained intact. Rats subjected to an HFD (Fig. 4C and D) exhibited pronounced histopathological alterations. There was severe neuronal degeneration characterized by numerous pyknotic nuclei and neuronal ballooning. The CA1 pyramidal layer showed marked cell loss, disorganization, and vacuolization, confirming successful induction of hippocampal injury. Histological evaluation of the HFD + Treatment group (Fig. 4E and F) revealed a clear neuroprotective effect. Neuronal density was notably improved compared to the HFD-only group. Although some degenerating neurons persisted, the laminar architecture was better preserved, and the CA1 subfield showed partially restored structural integrity. Sections from the control group (Fig. 4G and H) exhibited intact hippocampal cytoarchitecture. The CA1 pyramidal layer and DG displayed tightly packed, healthy neurons with clear nuclei and prominent nucleoli. No evidence of neuronal degeneration, pyknotic nuclei, or cell loss was observed, indicating normal histological features. Sections from the melatonin-treated group (Figures I, J) displayed relatively well-preserved DG and CA1 regions. While subtle neurotoxicity was suggested by minor histopathological changes, there was a slight increase in nuclear density, which may indicate a potential neuro-stimulatory effect of MEL. Overall, the hippocampal architecture remained largely intact.

Fig. 4.

Histopathological evaluation of the hippocampal formation (DG and CA1 subfields). A, B) Histopathological evaluation of the BHB Group shows moderate preservation of hippocampal architecture. Neuronal density is reduced compared to control (G, H), with occasional pyknotic nuclei, but laminar organization is mostly maintained. (H&E, X100, X400); C, D) Histopathological evaluation of the HFD Group shows severe neuronal degeneration. Numerous pyknotic nuclei (black arrowheads) and neuronal ballooning (white arrowheads) are evident. The CA1 pyramidal layer shows marked cell loss, disorganization and vacuolization (black arrows), indicating successful injury induction. (Hematoxylin and Eosin, X100, X400); E, F) Histopathological evaluation of the HFD Group shows evident neuroprotective effect. Neuronal density is improved compared to injury-only group. Some degenerating neurons remain (arrowhead in E), but overall laminar architecture is better preserved. CA1 shows partially restored integrity. (Hematoxylin and Eosin, X100, X400); G, H) Histopathological evaluation of the Control Group shows Intact hippocampal cytoarchitecture with tightly packed, healthy pyramidal neurons. No signs of degeneration or cell loss are observed. (Hematoxylin and Eosin, X100, X400); I, J). Histopathological evaluation of the Melatonin Group shows relatively well-preserved dentate gyrus and CA1 region, indicating subtle neurotoxicity. Slight increase in nuclear density suggests potential neuro-stimulatory effect. (Hematoxylin and Eosin, X100, X400) (Scale bars in X100 images represents 100 micrometer and in X400 images represents 20 micrometer)

Together, the control group maintained normal hippocampal structure, while the HFD group exhibited severe neurodegeneration. Both BHB and MEL groups showed moderate preservation of neuronal architecture, with MEL possibly exerting a mild neuro-stimulatory effect. The HFD + Treatment group demonstrated significant neuroprotection, as evidenced by improved neuronal density and preservation of laminar organization compared to the HFD group.

Discussion

BHB and MEL emerge as promising neuroprotectants in this study, with evidence that they blunt obesity-driven neurodegeneration and point toward tangible therapeutic avenues. These findings corroborate and extend existing evidence regarding the anti-inflammatory, antioxidant, and neurotrophic properties of both compounds. Substantially, this investigation provides novel insights into the molecular mechanisms underlying their protective actions, thereby offering a more comprehensive understanding of their multifaceted benefits.

Obesity is well-recognized for its deleterious impact on the hippocampus, as the brain region essential for learning and memory, through multiple pathophysiological mechanisms [21]. It is commonly associated with hippocampal atrophy and chronic neuroinflammation, driven largely by adipose tissue–derived pro-inflammatory cytokines that can disrupt neurovascular integrity and contribute to neuronal injury [22]. Consistent with this framework, our histological analyses revealed reduced hippocampal volume and pronounced inflammatory responses in HFD-fed rats. Notably, administration of BHB and MEL, either alone or combined, significantly ameliorated inflammation and restored hippocampal volume. These histopathological improvements were robustly supported by molecular data demonstrating marked downregulation of pyroptotic markers elevated in the hippocampi of obese animals.

The potential reversibility of hippocampal damage with metabolic intervention underscores the critical role of weight management in preserving cognitive health throughout life [23]. Weekly body weight monitoring indicated significant reductions following BHB and MEL treatments, with combination therapy exerting particularly robust effects in attenuating HFD-induced weight gain. In accordance with previous studies, BHB may facilitate weight reduction primarily by enhancing satiety signaling through the modulation of hypothalamic neuropeptides, including proopiomelanocortin (POMC) and neuropeptide Y (NPY). Elevated circulating ketone levels have also been suggested to exert a direct appetite-suppressing effect, as evidenced by findings that ketone ester (KE) beverages reduce plasma ghrelin concentrations, subjective hunger sensations, and the motivation to eat [24, 25], while also improving lipid metabolism via upregulation of fatty acid oxidation pathways and modulation of adipose tissue function. Notably, BHB promotes the browning of white adipose tissue (WAT) by activating thermogenic markers including uncoupling protein 1 (UCP1), peroxisome proliferator-activated receptor gamma coactivator-1 alpha (PGC-1α), cell death-inducing DFFA-like effector A (CIDEA), PR domain containing 16 (PRDM16), and fibroblast growth factor 21 (FGF21) [25, 26]. Moreover, we previously uncovered that the combined oral administration of BHB and MEL successfully prevented the whitening of BAT in obese rats fed an HFD [27].

MEL contributes by regulating circadian rhythms, which optimize metabolic processes, and by increasing energy expenditure through the activation of sympathetic outflow to BAT. Moreover, MEL might reduce ER stress by downregulating the PERK, IRE1α, and CHOP signaling pathways, thereby alleviating obesity-related metabolic dysfunction. Crucially, MEL could prevent the whitening of BAT, a process characterized by decreased UCP1 and mitochondrial activity that impairs thermogenesis and contributes to weight gain [28].

When administered in combination, BHB and MEL might exert complementary and synergistic effects on both metabolic and neuroendocrine pathways. BHB predominantly influences peripheral lipid metabolism, fatty acid oxidation, and fat browning, while MEL enhances mitochondrial dynamics, circadian regulation, and anti-stress responses. Together, these compounds could converge on central energy-sensing pathways such as AMP-activated protein kinase (AMPK), mammalian target of rapamycin (mTOR), and sirtuin 1 (SIRT1), resulting in an accelerated and more pronounced reduction in body weight compared to either treatment alone [29–31].

In alignment with prior reports emphasizing the anti-neuroinflammatory properties of ketone bodies and MEL, our results reveal suppression of the TXNIP–NLRP3–CASP1 signaling pathway, a key mediator of pyroptosis in obesity-associated brain injury. In addition, lower GSDMD mRNA expression was detected in the present study, which curtailed the subsequent cascade of pyroptotic cell death characteristic of obesity-related neurodegeneration. While previous studies implicated TXNIP and NLRP3 as critical mediators of obesity-induced neurotoxicity [32], our investigation uniquely elucidates how pharmacological suppression by BHB and MEL translates into measurable neuroprotection.

Moreover, although the neuroprotective effects of BHB and MEL have been previously linked to the attenuation of ER stress across various pathological contexts, this study specifically demonstrates their efficacy in lowering ER stress markers, BiP, CHOP, and TXNIP, in the hippocampus under obese conditions. The concomitant upregulation of sXBP1, particularly following BHB and MEL treatment, is noteworthy given its role in the adaptive UPR [33, 34]. Enhanced sXBP1 expression promotes protein folding homeostasis and mitigates accumulation of misfolded proteins, thereby protecting neurons from ER stress-induced apoptosis [34]. These findings implicate ER stress modulation and inflammasome inhibition as interconnected pathways underpinning the neuroprotective effects observed.

Supporting these findings, Montiel et al. demonstrated that ketone bodies modulate autophagy-lysosomal pathways and reduce ER stress in stroke and glucose deprivation models [35], while Camberos-Luna et al. reported that BHB enhances autophagic flux and neuronal survival under glucose deprivation in cortical neurons [36]. Sethuraman and colleagues showed that chronic ketosis modulates hypoxia-inducible factor 1-alpha (HIF-1α)–mediated inflammatory signaling in the brain [37], with Parveen et al. further confirming that ketosis confers neuroprotection through HIF-1α regulation of the TXNIP/NLRP3 axis, characterized by reduced pro-inflammatory cytokines and increased IL-10 following stroke [38].

Guo et al. provided additional mechanistic insight, revealing that a ketogenic diet suppresses NLRP3 inflammasome activation via inhibition of Drp1-mediated mitochondrial fission and ER stress, paralleling the BHB-mediated TXNIP/NLRP3 inhibition demonstrated here [39]. Jin and colleagues further reported that BHB modulates microglial metabolism and attenuates amyloid-β oligomer–induced inflammation, reinforcing BHB’s relevance in neurodegenerative disease contexts [40].

Similar neuroprotective effects have been observed with MEL; for instance, MEL mitigates acute neuronal injury post-ischemic stroke by suppressing ER stress-dependent autophagy via the PERK and IRE1 pathways [41]. Its capacity to inhibit inflammasome activation positions MEL as a promising candidate for AD therapeutics. Wongprayoon et al. demonstrated MEL’s protective effects against methamphetamine-induced ER stress and apoptosis in SH-SY5Y cells [42], while Song et al. observed MEL-mediated suppression of ER stress-induced neuronal death under insulin-resistant conditions [43].

In models of STZ-induced diabetes, MEL inhibited pyroptosis and autophagy markers, including NLRP3, cleaved caspase-1, GSDMD-N, and IL-1β, reducing neuronal death both in vivo and in high-glucose-treated neurons [44]. Furthermore, in amyloid-β–treated SH-SY5Y cells, MEL exerted anti-inflammatory effects mediated by inflammasome-related mechanisms central to cytokine activation and pyroptotic cell death [45].

Consistently, our study demonstrated elevated hippocampal mRNA expression of ER stress markers BiP, CHOP, and TXNIP in obese rats, accompanied by downregulation of sXBP1. Treatment with MEL and BHB reversed these alterations, reducing BiP, CHOP, and TXNIP levels while restoring sXBP1 expression. Additionally, obesity-induced elevations of pyroptosis-associated transcripts, CASP1, CASP3, CASP11, GSDMD, NLRP3, NFκB, and IL-1β, were significantly mitigated by BHB and MEL supplementation, confirming their capacity to attenuate molecular drivers of neuroinflammation and neuronal injury.

Histopathological examination also provided strong morphological validation of the molecular findings in this study; the hippocampal fields CA1 and DG within the HFD group exhibited widespread neuronal degeneration in the form of pyknotic nuclei diffusely, ballooning of the neurons, cytoplasmic vacuolization, and laminar disorganization, and thus indicating widespread neuroinflammatory injury. These structural alterations were consistent with the increased expression of pyroptosis-associated genes CASP1, GSDMD, and NLRP3 and ER stress markers BiP, CHOP, and TXNIP, which suggested that chronic ER stress and inflammasome activation collectively led to neuronal cell death. On the other hand, BHB and MEL treatment, particularly in combination, greatly preserved neuronal morphology, with increased neuronal density and recovery of normal hippocampal cytoarchitecture. This histological protection is in accordance with downregulation of ER stress and pyroptotic pathways that were observed, implying molecular modulation of these processes leads to tangible structural neuroprotection. These findings emphasize that BHB and MEL not only steer clear of molecular determinants of neurodegeneration but also maintain the physical integrity of hippocampal neurons responsible for sustaining cognition.

Together, the combined administration of BHB and MEL might produce a synergistic effect, offering greater benefits than either treatment alone. This potential synergy was reflected in both molecular and physiological findings. For instance, IL-10, a potent anti-inflammatory cytokine involved in suppressing excessive immune responses and supporting neuronal survival, was upregulated nearly nine-fold in the BHB + MEL group compared to the HFD group alone. This marked increase might indicate that the combination therapy not only counteracts pro-pyroptotic pathways but also enhances protective anti-inflammatory signaling. Additionally, the combination treatment appeared to result in earlier and more pronounced reductions in body weight, which could have contributed to improved metabolic balance and reduced neural stress. These observations suggest that BHB and MEL might act through complementary mechanisms with BHB primarily mitigating inflammasome activation and pyroptosis, while MEL potentially provides antioxidant and mitochondrial stabilizing effects. The convergence of these actions could lead to a broader neuroprotective outcome, highlighting the possible therapeutic value of their co-administration in managing obesity-associated neurodegeneration.

Although the present study primarily focused on ER stress pathways, it is important to acknowledge that HFD-induced obesity may simultaneously activate multiple molecular processes beyond the UPR. These include lipotoxicity, autophagy/lysosomal dysfunction, and mitochondrial-related oxidative stress, all of which may contribute to neuronal damage and cognitive decline. For instance, saturated fatty acids such as palmitic acid, which are elevated in HFD-fed animals, have been shown to disrupt the autophagic flux and induce autolysosomal dysfunction, thereby increasing cellular stress and predisposing neurons to injury through excessive lipid accumulation and metabolic imbalance [46].

Given these findings, it would be valuable for future studies to include direct assessments of mitochondrial function and oxidative stress markers, such as mitochondrial reactive oxygen species (mito-ROS), proteins involved in mitochondrial dynamics (Drp1, Mfn2), and biogenesis regulators like PGC-1α. Similarly, evaluating autophagy-related markers (LC3-II/I, p62, LAMP1) could clarify whether the neuroprotective benefits of BHB and MEL are mediated through the restoration of autophagic flux in addition to ER stress alleviation. Moreover, additional metabolic evaluations such as glucose tolerance tests (GTT) and insulin tolerance tests (ITT), along with profiling circulating cytokines, might help to distinguish between central and systemic effects of these treatments and establish clearer links between metabolic health, lipid dysregulation, and brain pathology.

Given the multifactorial nature of obesity-induced brain injury, BHB and MEL might be most effective when used as part of a multimodal therapeutic strategy rather than as monotherapies. For example, glucagon-like peptide-1 receptor agonists (GLP-1RAs), such as liraglutide or semaglutide have demonstrated metabolic benefits (e.g., weight loss, improved insulin sensitivity) and emerging neuroprotective effects that may reduce neuroinflammation and cognitive decline [47]. Co-administration of BHB or ketogenic interventions with GLP-1RAs might produce complementary benefits [48]; GLP-1RAs could improve systemic metabolic drivers of brain injury, while BHB and MEL could more directly modulate central stress responses. Such a combined approach might therefore address both peripheral metabolic dysfunction and central cellular stress more effectively than any single agent.

Other candidate pairings include the established metabolic drugs and agents targeting oxidative or autophagic pathways. For instance, metformin and SGLT2 inhibitors have been reported to exert anti-inflammatory or neuroprotective effects in preclinical models and could synergize with BHB/MEL by further improving insulin signaling and reducing systemic drivers of lipotoxicity. Likewise, therapies that restore autophagic flux or improve mitochondrial dynamics (e.g., modulators of Drp1/Mfn2, or interventions that boost PGC-1α) might complement BHB’s and MEL’s effects on cellular bioenergetics and proteostasis [49, 50]. Future preclinical studies should therefore test rational combinations (e.g., BHB + MEL + GLP-1RA or + metformin) and include readouts of systemic metabolism (GTT/ITT), plasma cytokines, and central markers of ER stress, mito-ROS, and autophagic flux to delineate additive versus truly synergistic mechanisms [51]. Their multi-targeted effects may complement current anti-diabetic and anti-obesity therapies, presenting a comprehensive strategy to ameliorate obesity-related metabolic and neurodegenerative disorders. This multi-faceted approach is especially pertinent in the context of HFD-induced lipid dysregulation, autophagic impairment, and neuroinflammation, where BHB and MEL could synergize with pharmacological interventions to restore metabolic and inflammatory equilibrium.

Conclusion

In the current investigation, BHB and MEL have been recognized as promising neuroprotective agents capable of counteracting the complex pathophysiological processes underlying obesity-induced brain injury and neurodegeneration. Through the convergence of anti-inflammatory, anti-pyroptotic, and ER stress relieving mechanisms, coupled with neurotrophic modulation, these compounds provide a comprehensive therapeutic approach for preserving neuronal structure and function. The findings not only reinforce existing evidence but also extend the therapeutic potential of BHB and MEL within the context of obesity-related neurodegeneration. Further research is warranted to facilitate clinical translation, particularly focusing on their synergistic capacity to modulate intersecting metabolic and neuroimmune pathways. Such multi-targeted interventions may offer substantial benefits for patient-centered neurotherapeutic strategies aimed at mitigating obesity-associated neurodegenerative disorders.

Abbreviations

AD

Alzheimer’s disease

ATF6

Activating transcription factor 6

BHB

β-hydroxybutyrate

BiP

Binding immunoglobulin protein

DG

Dentate gyrus

ER

Endoplasmic reticulum

GSDMD

Gasdermin D

HFD

High-fat diet

IL-1β

Interleukin-1β

MEL

Melatonin

NLRP3

Nucleotide-binding domain and Leucine-rich repeat containing (NLR) family pyrin domain containing 3

PD

Parkinson’s disease

PERK

Protein kinase RNA-like ER kinase

CHOP

C/EBP homologous protein

IRE1α

Inositol-requiring enzyme 1 alpha

sXBP1

Spliced X-box binding protein

TXNIP

Thioredoxin-interacting protein

UPR

Unfolded protein response

Author Contributions

All experiments, statistical analyses, and figure preparation were conducted by ZZ, MHM, AD, MJRK, FO, ARD, and OV. The initial draft of manuscript was written by ZZ and MHM. All tests were set up by ZZ and MHM, and second draft of manuscript was written by MHM and OV. A final manuscript proof was edited and completed by OV and SMS. Project administration and supervision were conducted by SMS. All authors have read and agreed to the final version of the manuscript.

Funding

This work was supported by the Vice-chancellery for Research Affairs of Shiraz University of Medical Sciences [Grant number 30077].

Data Availability

All datasets analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical Considerations

The Research Ethics Committee of Shiraz University of Medical Sciences has ethically confirmed the study under the Ethics code of IR.SUMS.ACE.1403.056).

Consent to Publish

Not applicable, as this research was an animal study.

Competing interests

The authors declare no competing interests.