Caloric restriction attenuates NLRP3 inflammasome activation and oxidative stress in leukocytes of patients with obesity

Summary

Obesity promotes chronic low-grade inflammation through metabolic stress and activation of the NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome. The effects of caloric restriction on this pathway in humans are poorly understood; therefore, we evaluated the impact of a 6-month hypocaloric dietary intervention in 45 adults with obesity by assessing the metabolic, inflammatory, oxidative stress, and inflammasome-related markers in peripheral blood mononuclear cells. Weight loss improved metabolic parameters, reduced oxidative stress, and decreased the expression of nuclear factor κB (NF-κB), NLRP3, apoptosis-associated speck-like protein containing a CARD (ASC), and pro-caspase-1, while there tended to be a reduction of active caspase-1. These changes were accompanied by lower circulating interleukin (IL)-1β (IL-1β) and IL-18 levels, suggesting systemic attenuation of inflammasome activity. Baseline adiposity and insulin resistance influenced the magnitude of the response to dietary intervention. These findings support caloric restriction as a strategy to mitigate obesity-associated immunometabolic dysfunction and identify the redox-inflammasome axis as a potential therapeutic target.

Graphical abstract

Highlights

• •Moderate weight loss improves metabolic health in obesity
• •Caloric restriction reduces oxidative stress in PBMCs
• •Dietary intervention attenuates NLRP3 inflammasome priming
• •Inflammasome-related cytokines decline with weight loss

Obesity medicine; Immune response; Nutrition

Introduction

Obesity is a complex, multifactorial disease characterized by excessive adipose tissue accumulation and a state of chronic low-grade inflammation. This inflammatory milieu is a major contributor to obesity-related comorbidities, including insulin resistance, type 2 diabetes, cardiovascular disease, and certain cancers.¹ Growing evidence highlights the role of the innate immune system in sustaining this inflammation, particularly through the activation of inflammasomes, intracellular complexes that sense metabolic stress and danger signals.²

Among these, the NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome plays a central role in linking metabolic stress to inflammation.³ Its activation involves a priming step, triggered by pattern recognition receptors such as Toll-like receptor (TLR) 4, leading to nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB)-dependent transcription of inflammasome components, followed by an activation step driven by cellular stress signals. This results in caspase-1 activation and the maturation and secretion of interleukin (IL)-1β and IL-18,^3,4 In obesity, chronic exposure to metabolic damage-associated molecular patterns (DAMPs), endogenous molecules or cellular components generated, released, or accumulated in response to metabolic stress or cellular damage, including free fatty acids, ceramides, uric acid, and mitochondrial reactive oxygen species (ROS), sustains both the priming and activation phases of the NLRP3 inflammasome^2,3 in several tissues. Enhanced NLRP3 activity has been reported in circulating leukocytes of individuals with obesity⁵ but also in metabolic tissues including adipose tissue and liver,⁶ supporting its role as a key driver of metabolic inflammation. While this inflammatory response is initially adaptive in the context of nutrient overload, facilitating adipose tissue expansion, remodeling and metabolic compensation, its persistent activation becomes maladaptive and drives insulin resistance, pancreatic β cell dysfunction, and progressive metabolic impairment.^2,7

Although hypocaloric diets are central to obesity management, and moderate weight loss improves metabolic and inflammatory parameters, the extent to which caloric restriction modulates NLRP3 inflammasome activation in human leukocytes remains unclear. Therefore, this study investigates the impact of a hypocaloric diet-induced moderate weight loss on NLRP3 regulation in peripheral blood mononuclear cells (PBMCs) and systemically in individuals with obesity, providing new insights into the immunometabolic effects of dietary intervention.

Results

Moderate weight loss improves metabolic and inflammatory profiles in obesity

We recruited 45 middle-aged patients with a body mass index (BMI) of 45.2 ± 6.1 kg/m², some of whom presented obesity-related comorbidities (Table 1). By the end of the 6-month dietary intervention, the participants achieved a mean weight loss of ≈10%, accompanied by significant reductions in BMI, waist circumference, relative fat mass (RFM), and waist-to-height ratio (WHtR) (p < 0.001), suggesting a decrease in visceral adiposity, which is strongly associated with systemic inflammation and cardiometabolic risk. In fact, blood pressure was also significantly reduced after the intervention (p < 0.05). Biochemical analysis revealed optimized systemic metabolism after weight loss in terms of glycemic control and lipid handling, with decreases in the homeostasis model assessment of insulin resistance (HOMA-IR), insulin, and glycated hemoglobin A1c (HbA1c) levels (p < 0.01), an increase in HDL cholesterol (HDLc) levels, and a drop in triglycerides (TG) (p < 0.05). Furthermore, we detected a significant drop in complement component 3 (C3c) (p < 0.001), and tumor necrosis factor α (TNF-α) (p < 0.05) but no significant alteration of the IL-6 levels.

Table 1.

Clinical characteristics of the study population before and after the intervention

	Baseline	Post-intervention
Total N (females %)	45 (73.3)	–
Age (years)	46.2 ± 8.7	–
Weight (kg)	122.6 ± 20.3	110.3 ± 17.7∗∗∗
BMI (kg/m2)	45.2 ± 6.1	40.6 ± 5.1∗∗∗
WL (%)		9.7 ± 4.1
Waist (cm)	123.9 ± 15.1	110.6 ± 21.3∗∗∗
RFM	45.9 (41.6, 50.52)	43.6 (40.4, 47.8)∗∗∗
WHtR	0.75 ± 0.08	0.69 ± 0.01∗∗∗
SBP (mmHg)	135.0 ± 17.8	129.2 ± 15.6∗
DBP (mmHg)	84.6 ± 9.7	78.2 ± 9.0∗∗
HOMA-IR	4.2 ± 2.4	3.4 ± 2.5∗∗
Glucose (mg/dL)	100.6 ± 21.9	96.5 ± 21.9
Insulin (μU/mL)	16.4 ± 8	14.2 ± 8.2∗∗
HbA1c (%)	5.8 ± 0.6	5.6 ± 0.7∗∗
TC (mg/dL)	184.3 ± 35.7	190.0 ± 41.3
HDLc (mg/dL)	43.0 ± 10.4	45.7 ± 11.1∗
LDLc (mg/dL)	114.9 ± 31.6	120.6 ± 36.1
TG (mg/dL)	131 (92, 151)	115 (83, 133)∗
C3c (mg/dL)	142 (119, 159)	124 (109, 138)∗∗∗
IL-6 (pg/mL)	4.1 (3.4, 5.8)	3.6 (2.8, 8.4)
TNF-α (pg/mL)	15.8 ± 6.6	13.7 ± 6.1∗
Treatment
Hypertension % (N)	22.2 (10)	–
Hyperlipidemia % (N)	24.4 (11)	–
T2D % (N)	6.7 (3)	–

Data are expressed as mean ± SD, percentage (n), or median and IQ range (25% and 75% percentiles) for TG. Differences between variables before and after treatment were compared with a paired Student’s t test or Wilcoxon test and were considered significant when ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

BMI, body mass index; C3c, complement component 3; DBP, diastolic blood pressure; HbA1c, glycated hemoglobin; HDLc, HDL cholesterol; HOMA-IR, homeostatic model assessment for insulin resistance; IL-6, interleukin 6; LDLc, LDL cholesterol; T2D, type 2 diabetes; TC, total cholesterol; TG, triglycerides; TNF-α, tumor necrosis factor α; RFM, relative fat mass; WHtR, waist-to-height ratio; SD, standard deviation; SBP, systolic blood pressure; WL, weight loss.

Weight loss reduces metabolic and oxidative stimulus associated with NLRP3 inflammasome activation

In the context of obesity, several metabolic and danger-associated signals act as key inducers of inflammasome activation. Lipopolysaccharide (LPS) is a classical priming signal for NLRP3 inflammasome in obesity, largely due to increased gut permeability. However, no significant changes were observed in the levels of LPS-binding protein (LBP) (Figure 1A), a surrogate marker of metabolic endotoxemia, after the intervention. In parallel, uric acid, especially in its crystalline form, acts as a potent inducer of inflammasome. At baseline, 47.7% of our participants presented hyperuricemia (serum uric acid levels > 6 mg/dL in women and > 7 mg/dL in men) according to current clinical guidelines.⁸ The levels of uric acid in plasma significantly dropped after the intervention (Figure 1B, p < 0.01).

Figure 1.

Metabolic, inflammatory, and oxidative inducers of the inflammasome in individuals with obesity, before and after moderate weight loss

(A–C) Serum levels of LBP (A) (n = 13), uric acid (B) (n = 45), and 8-OHdG (C) (n = 20) were evaluated at baseline and after the 6-month intervention (post).

(D) Superoxide production in PBMCs (D) (n = 33) was quantified using dHE and is expressed as the mean fluorescence intensity per cell.

(E) Representative fluorescence microscopy images acquired at baseline and post-intervention (dHE, red; Hoechst, blue for nuclei visualization; scale bars, 200 μm).

Data are presented as box-and-whisker plots (median and min-max with individual data points). Statistical significance was determined using paired Student’s t test or Wilcoxon signed-rank test, as appropriate. ns, statistically non-significant; ∗, p < 0.05; ∗∗, p < 0.01.

Abbreviations: 8-OHdG, 8-hydroxy-2′-deoxyguanosine; dHE, dihydroethidium; LBP, lipopolysaccharide-binding protein; PBMCs, peripheral blood mononuclear cells; RFU, relative fluorescence units.

Excessive ROS production, characteristic of obesity, has also been identified as a key upstream event in the activation of the NLRP3 inflammasome. Consistent with this, our analysis revealed a significant reduction in cytosolic superoxide production in PBMCs (Figures 1D and 1E, p < 0.01) and in the systemic levels of 8-hydroxy-2′-deoxyguanosine (8-OHdG) (Figure 1C, p < 0.05), which may have mitigated redox-driven inflammasome stimulus.

Figure 2.

Protein expression of NLRP3 inflammasome components in PBMCs of individuals with obesity, before and after moderate weight loss

(A–E) Protein levels of NF-κB p65 (A) (n = 18), NLRP3 (B) (n = 25), ASC (C) (n = 23), pro-caspase-1 (D) (n = 22), and the active caspase-1 p10 (E) (n = 22) were quantified in PBMCs by western blot at baseline and after the 6-month intervention. Protein levels were normalized to actin.

(F) Representative western blot images. Approximate molecular weights (kDa) are shown beside the immunoblots. Representative uncropped western blot membranes are provided in Figure S1.

Data are presented as box-and-whisker plots (median and min-max with individual data points). Statistical significance was determined using paired Student’s t test or Wilcoxon signed-rank test, as appropriate. ∗, p < 0.05; ∗∗, p < 0.01.

Abbreviations: ASC, apoptosis-associated speck-like protein containing a CARD; kDa, kilodalton; NLRP3, nucleotide-binding oligomerization domain-like receptor family pyrin domain-containing 3; NF-κB, nuclear factor κ-light-chain-enhancer of activated B cells; PBMCs, peripheral blood mononuclear cells.

Weight loss attenuates NLRP3 inflammasome priming in circulating PBMCs

In parallel, we explored whether the amelioration of metabolic inducers of the NLRP3 inflammasome influenced the overall activation of this pathway in PBMCs (Figure 2). Immunoblotting assays (Figure 2F) revealed a decrease in NF-ĸB p65 (Figure 2A, p < 0.05), the transcriptional factor involved in inflammasome priming. This mediator induces the transcription of key inflammasome components such as NLRP3, pro-caspase-1, and pro-interleukins. In this sense, there was a significant reduction in NLRP3 protein expression after the intervention (Figure 2B, p < 0.05). Furthermore, we detected a significant decrease in apoptosis-associated speck-like protein containing a CARD (ASC) (Figure 2C, p < 0.05), a key adaptor protein in the assembly and activation of the inflammasome complex. In parallel, we observed a significant drop in the protein expression of pro-caspase-1 (Figure 2D, p < 0.01), while the expression of its active form caspase-1 p10 also tended to decline (Figure 2E, p = 0.07), suggesting the attenuation of inflammasome priming together with a possible reduction in inflammasome activation following the intervention. Representative uncropped western blot membranes are provided in Figure S1.

Systemic inflammasome-derived cytokines decline after dietary intervention

Upon activation, the enzymatic activity of caspase-1 is known to mediate the cleavage and maturation of the pro-inflammatory interleukins IL-1β and IL-18. Accordingly, we observed a significant decrease in serum levels of both IL-1β (Figure 3A, p < 0.05) and IL-18 (Figure 3B, p < 0.01) following the intervention. Although these systemic cytokines are derived from multiple cell types and tissues, their reduction is consistent with an overall attenuation of inflammasome activity after weight loss.

Figure 3.

Circulating cytokines derived from inflammasome activation, before and after caloric restriction

(A and B) Serum levels of IL-1β (A) (n = 13) and IL-18 (B) (n = 21) were measured using high-sensitivity immunoassays at baseline and after the 6-month intervention.

Data are presented as box-and-whisker plots (median and min-max with individual data points). Statistical significance was determined using paired Student’s t test or Wilcoxon signed-rank test, as appropriate. ∗, p < 0.05; ∗∗, p < 0.01.

Abbreviations: IL-1β, interleukin 1β; IL-18, interleukin 18.

Baseline metabolic status influences the inflammasome response to caloric restriction

To explore interindividual variability in the response to dietary intervention, correlation analyses were performed between baseline clinical parameters and relative changes of inflammasome and oxidative stress markers (Figure 4). Baseline measures of adiposity, including BMI, waist circumference, and WHtR, were inversely associated with changes in key inflammasome-related proteins—ASC, pro-caspase-1, caspase-1, and IL-1β—indicating a blunted downregulation of inflammasome markers in individuals with greater central adiposity. Similarly, baseline markers of insulin resistance—insulin, HOMA-IR, and HbA1c—showed negative associations with the changes in pro-caspase-1, ASC, and superoxide levels. In contrast, changes in NLRP3 expression correlated positively with NF-κB, ASC, and IL-1β, supporting a coordinated modulation of the inflammasome pathway. Overall, these findings suggest that higher baseline adiposity and metabolic dysfunction limit the anti-inflammatory and antioxidant benefits of caloric restriction.

Figure 4.

Correlations between baseline metabolic parameters and diet-induced changes in inflammasome and oxidative stress markers

Heatmap displaying Spearman’s correlation coefficients between the baseline (B) anthropometric and metabolic variables (BMI, waist circumference, WHtR, HOMA-IR, insulin, HbA1c) and the relative changes (Δ) in inflammasome-related proteins (NF-κB, NLRP3, ASC, pro-caspase-1, caspase-1, and IL-1β) and oxidative stress markers (superoxide). Relative changes were calculated as Δ = [baseline – post-intervention] to facilitate interpretation, such that positive values reflect reductions after the diet, and then transformed to Z scores. Squares are color coded according to the correlation coefficient (blue for negative correlations, and red for positive correlations), with color intensity proportional to the magnitude of r. Correlations were analyzed using Spearman’s correlation coefficients. Statistical significance is indicated by asterisks within the corresponding squares (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001).

Abbreviations: ASC, apoptosis-associated speck-like protein containing a CARD; BMI, body mass index; HbA1c, glycated hemoglobin A1c; HOMA-IR, homeostasis model assessment of insulin resistance; IL-1β, interleukin 1β; NF-κB, nuclear factor κ-light-chain-enhancer of activated B cells; NLRP3, NOD-like receptor family pyrin domain-containing 3; WHtR, waist-to-height ratio.

Discussion

Obesity is a major global health challenge, with a prevalence that has doubled since 1990 according to the WHO. Caloric restriction remains one of the most effective and accessible strategies to promote weight loss and counteract obesity-associated pathophysiological mechanisms, which include chronic low-grade inflammation. The findings of the present study provide novel evidence that moderate weight loss achieved through a hypocaloric diet attenuates NLRP3 inflammasome signaling in PBMCs and at systemic level in individuals with obesity. This downregulation occurred in parallel with systemic metabolic improvements and reduced oxidative stress, highlighting the interplay between metabolism, redox homeostasis, and innate immune activation.

In our study, the 6-month hypocaloric intervention resulted in a ∼10% reduction in body weight, accompanied by significant decreases in the markers of total and central adiposity. Blood pressure, insulin sensitivity, and lipid metabolism were also improved, which is in line with previous findings.^9,10 Although IL-6 levels remained unchanged in our cohort, declines in C3c and TNF-α supported the anti-inflammatory effects of caloric restriction.^11,12

Visceral adiposity is closely linked to the accumulation of metabolic danger signals that prime and/or activate the NLRP3 inflammasome in several tissues, thus linking nutrient excess to immune and metabolic dysregulation.^2,3 In obesity, lipid-derived molecules act as endogenous DAMPs that enhance inflammasome activity.¹³ In parallel, dysfunctional adipose tissue releases TNF-α, priming NLRP3 inflammasome through TLR-4-NF-ĸB and stimulating caspase-1 activation in adipocytes and immune cells.¹⁴ In our cohort, improvements in lipid profile and reductions in circulating TNF-α and C3c may, therefore, have contributed to attenuated NLRP3 inflammasome signaling. Despite significant reductions in several inflammatory markers, LBP, an acute-phase protein that binds to bacterial LPS, was unaltered by the diet, though the relatively small sample size available for this particular analysis warrants a cautious interpretation of this finding. Nevertheless, it is consistent with reports showing reductions in LBP after longer interventions or when combined with probiotics.^15,16

Uric acid, in both its crystalline and soluble forms, is also a potent DAMP that triggers NLRP3 inflammasome activation. Monosodium urate crystals prime NLRP3 via TLR-4 activation, and, once internalized by macrophages, they can activate inflammasome through lysosome destabilization, cathepsin B release, and ROS generation,¹⁷ whereas soluble uric acid disrupts oxidative phosphorylation, increasing mitochondrial ROS and promoting IL-1β maturation.¹⁸ Nearly half of our participants exhibited elevated uric acid levels at baseline, consistent with an obesity-associated hyperuricemic state. In line with previous findings, uric acid levels declined markedly after weight loss,¹⁹ which might have reduced urate-driven inflammasome stimulus.

Oxidative stress is another key upstream regulator of NLRP3 inflammasome activation. In obesity, chronic nutrient overload and hyperglycemia act as potent metabolic stressors that amplify inflammasome priming and activation specially via mitochondrial dysfunction and excessive ROS.^3,20 Weight loss interventions can reduce oxidative stress in obesity,²¹ though a few studies have examined this effect in human leukocytes. In our cohort, superoxide production in PBMCs was significantly reduced after moderate weight loss, consistent with reports following short-term dietary restriction²² or bariatric surgery.²³ ROS-induced oxidative damage promotes mitochondrial depolarization and release of oxidized mitochondrial DNA (mtDNA), which can activate the NLRP3 inflammasome through direct binding to the NLRP3 pyrin domain²⁴ or by its extracellular recognition through TLR-9.²⁰ In our study, we observed decreased plasma 8-OHdG levels after the intervention. As mtDNA constitutes a major fraction of total circulating DNA²⁵ and is particularly vulnerable to oxidative stress, this decrease likely reflects attenuated mitochondrial injury and reduced availability of oxidized mtDNA. These findings align with previous reports showing that caloric restriction alleviates oxidative damage markers and enhances mitochondrial function in obesity,^12,26,27 thereby modulating innate immune activation.

The relation between caloric restriction, redox state, and inflammasome activation is partially mediated by AMP-activated protein kinase (AMPK), a central nutrient-sensing kinase that orchestrates mitochondrial quality control.^11,12,28 Previous studies have shown enhanced AMPK signaling, increased autophagic flux, and mitochondrial remodeling following caloric restriction in immune cells,²⁹ whereas impaired autophagy has been linked to NLRP3 activation in adipocytes.¹⁴ Together, this suggests that AMPK-driven auto/mitophagy mediates the protective effects of caloric restriction on mitochondrial integrity and limits inflammasome priming.

Collectively, these redox-immunometabolic adaptations to caloric restriction converged to limit inflammasome priming and activation in our cohort. Specifically, we observed a marked downregulation of NLRP3 inflammasome pathway components in PBMCs. The expression of NF-κB p65, NLRP3, ASC, and pro-caspase-1 decreased significantly, indicating suppression of the priming step, while the cleavage of pro-caspase-1 into its active form, caspase-1 p10, tended to decline. In parallel, plasma concentrations of IL-1β and IL-18 were reduced. Although these circulating cytokines are produced by multiple cell types and tissues, their reduction is consistent with an overall attenuation of systemic inflammasome activity following weight loss. Consistent with our results, previous studies support that caloric restriction directly modulates inflammasome activity and contributes to metabolic improvements. In obese mice, pharmacological inhibition or genetic deletion of inflammasome components reproduced the benefits of dietary restriction, including reduced adipose inflammation, improved insulin sensitivity, and restored mitochondrial function.^4,30 In humans, short-term fasting reduced NLRP3 inflammasome activation in immune cells, indicating that nutrient availability directly influences its activity.²⁸ Moreover, Herrero-Aguayo et al. described a decline in mRNA expression of components of the inflammasome in patients with obesity who underwent bariatric surgery.³¹ Similarly, the CALERIE-2 trial reported that moderate long-term caloric restriction improves metabolic health and broadly suppresses inflammasome pathways in adipose tissues and macrophages,³² without impairing cell-mediated immunity.³³

In this sense, adipose tissue is a major site of NLRP3 activation, where both adipocytes and infiltrating macrophages sustain a chronic inflammatory loop that contributes to systemic metabolic dysfunction.^4,34,35 Although less tissue specific, PBMCs represent a practical and informative model for assessing inflammasome signaling in humans, allowing the evaluation of systemic immunometabolic adaptations to dietary intervention^36,37 through a less invasive approach. In the present study, the coordinated reduction in circulating cytokines, including TNF-α, IL-1β, and IL-18, together with the suppression of inflammasome priming in PBMCs, suggests that the molecular changes observed in the circulating immune cells mirror broader immunometabolic adaptations occurring in adipose tissue and other metabolically active organs.

The metabolic effects of inflammasome are largely mediated by IL-1β and IL-18 cytokines. The post-weight loss reduction of these cytokines is, therefore, likely to have systemic metabolic benefits. IL-1β impairs insulin receptor signaling and β cell function, whereas IL-18 contributes to endothelial dysfunction, reduced lipid oxidation, and peripheral insulin resistance.^3,4,7,38 Sustained high levels promote a pro-inflammatory, metabolically inflexible state favoring hyperglycemia and dyslipidemia. Elevated glucose levels can exacerbate NLRP3 inflammasome activation by overloading the mitochondrial electron transport chain²⁸ and increasing ROS production, establishing a self-perpetuating inflammatory-metabolic loop. In this context, the post-intervention reduction of circulating IL-1β and IL-18, together with improved glucose and lipid profiles, suggests partial disruption of this vicious cycle and supports the link between metabolic restoration and inflammasome suppression.

Pre-existing metabolic status may influence the response to caloric restriction. Indeed, previous studies have reported that individuals with higher baseline insulin resistance and inflammatory cytokine levels present a limited metabolic and inflammatory improvement after dietary weight loss.^39,40 In our cohort, higher baseline adiposity and insulin resistance markers were associated with smaller reductions in inflammasome-related proteins and superoxide, suggesting that a more adverse metabolic profile limits the downregulation of this pathway. This could reflect a state of entrenched metabolic-inflammatory coupling sustaining NLRP3 activation, indicating that individuals with severe metabolic impairment may require longer or combined interventions. Beyond metabolic status, experimental evidence indicates that estrogens exert predominantly anti-inflammatory effects by attenuating NLRP3 inflammasome priming and activation through multiple mechanisms.^41,42 Although reproductive hormonal status was not recorded in our predominantly female cohort, the relatively short duration of the intervention makes it unlikely that hormonal status was the primary driver of the coordinated metabolic and inflammatory improvements observed, which are more plausibly explained by the weight loss induced by caloric restriction. Given the association between estrogen decline and enhanced NLRP3 inflammasome activation, our findings suggest that dietary weight loss is particularly beneficial to post-menopausal women. This hypothesis warrants investigation in future studies.

To our knowledge, this is the first study to show that moderate weight loss through a hypocaloric diet attenuates the NLRP3 inflammasome in circulating PBMCs of individuals with obesity and induces a systemic reduction in inflammasome-derived cytokines. A major strength is the integrative assessment of systemic, metabolic, and cellular parameters involved in inflammasome regulation, together with the evaluation of inflammasome-related protein expression and downstream cytokine release—rather than solely mRNA levels—which more accurately reflects final pathway activation. The translational approach of the study provides novel mechanistic insights into how caloric restriction modulates innate immune signaling in humans.

These results underscore the interplay between metabolic dysregulation, redox imbalance, and innate immunity in obesity and suggest that the baseline metabolic status influences the magnitude of the response to weight loss. Understanding the mechanisms underlying the beneficial effects of caloric restriction beyond the obvious health outcomes may guide the development of nutritional or pharmacological strategies that reproduce its immunometabolic benefits.

Limitations of the study

Some limitations of this study should be acknowledged. First, our analysis focused solely on PBMCs; so, the role of adipose tissue-resident macrophages or other immune cell populations in the observed systemic effects is yet to be elucidated. Second, although the study included both female and male participants, it was not designed or statistically powered to evaluate the potential influence of sex or menopausal status on the measured outcomes, and reproductive hormonal status was not recorded. Finally, the number of samples available differed across some analyses because of technical constraints related to sample availability and assay-specific requirements; thus, certain assays may have had limited statistical power to detect modest biological effects. Importantly, all available biological materials were analyzed without investigator-driven sample selection, thereby minimizing the risk of selection bias. In addition, the observational design precludes causal inference. Future randomized studies are needed to confirm our findings and further outline the immunometabolic benefits of dietary intervention.

Resource availability

Lead contact

Requests for further information and resources should be directed to the lead contact, Dr. Milagros Rocha (milagros.rocha@fisabio.es).

Materials availability

This study did not generate new unique reagents.

Data and code availability

The data reported in this paper will be shared by the lead contact upon reasonable request. Individual participant data cannot be publicly deposited because of ethical and privacy restrictions associated with clinical human data. This paper does not report original code. Any additional information required to re-analyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

The authors thank Rosa Falcon (Fisabio) and Begoña Zaragoza (University Hospital Doctor Peset) for their technical support. The authors also thank Brian Normanly (University of Valencia, Spain/CIBEREHD) for his editing assistance. This work was supported by the Carlos III Health Institute, Spain and co-funded by the European Union through the European Regional Development Fund (ERDF) “A way to build Europe” (grant numbers CD23/0006 [S.L.-D.], FI23/00141 [L.P.-G.], FI23/00070 [C.L.-M.], PI25/00219 [V.M.V.], PI25/01841 [M.R.], and CIBEREHD (CB06/04/0071)]; Ministry of Education, University and Occupation, Generalitat Valenciana, Spain (grant numbers CIGE/2024/82 [S.L.-D.], CIGRIS/2021/112 [J.H.], CIGRIS/2022/172 [J.C.], and CIPROM/2022/32 [V.M.V. and M.R.]); and HORIZON EUROPE EU Programme (grant number 101136516 [M.R. and M.P.-S.]). The funding institutions have not been involved in any of the following: study design; collection, analysis, and interpretation of data; writing of the report; or in the decision to submit the article for publication.

Author contributions

Conceptualization, S.L.-D., V.M.V., and M.R.; funding acquisition, S.L.-D., V.M.V. and M.R.; project administration, M.R.; resources, S.V., C.M., V.M.V., and M.R.; methodology, S.L.-D., S.V., C.M., and M.R.; investigation, S.L.-D., L.P.-G., Z.A.-J., J.H., M.F.-R., C.L.-M., J.C., S.V., and C.M.; data curation, S.L.-D., L.P.-G., M.P.-S., Z.A.-J., and M.R.; formal analysis, S.L.-D., L.P.-G., Z.A.-J., J.H., and M.P.-S.; validation, S.L.-D. and M.P.-S.; visualization, S.L.-D., M.P.-S., Z.A.-J., J.H., L.P.-G., M.F.-R., C.L.-M., and J.C.; supervision, S.L.-D., V.M.V., and M.R.; writing – original draft, S.L.-D. and L.P.-G.; writing –review & editing, S.L.-D., M.P.-S., V.M.V., and M.R.

Declaration of interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

This article is original and was not generated or written with artificial intelligence assistance.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Mouse monoclonal anti-ASC	Santa Cruz	Cat# sc-514414; RRID: AB_2737351
Mouse monoclonal anti-Caspase 1	Santa Cruz	Cat# sc-56036; RRID: AB_781816
Mouse monoclonal anti-NF-κB	Thermo Fisher Scientific	Cat# 33–9900; RRID: AB_2533153
Rabbit monoclonal anti-NLRP3	Cell Signaling	Cat# 13158; RRID: AB_2798134
Rabbit polyclonal anti-α-actin-1	Sigma-Aldrich	Cat# A2066; RRID: AB_476693
Goat Anti-Rabbit IgG Secondary Antibody	Vector Laboratories	Cat# PI-1000; RRID: AB_2916034
Goat anti-Mouse IgG Secondary Antibody	Thermo Fisher Scientific	Cat# 31430; RRID: AB_228307
Chemicals, peptides, and recombinant proteins
Ficoll-Paque ® PLUS	Cytiva	Cat# 17–1440–02
Dihydroethidium (DHE)	Thermo Fisher Scientific	Cat# D11347
Hoechst 33342	Thermo Fisher Scientific	Cat# 62249
SuperSignal™ West Pico PLUS Chemiluminescent Substrate	Thermo Fisher Scientific	Cat# 34580
RIPA Lysis and Extraction Buffer	Thermo Fisher Scientific	Cat# 89901
Halt™ Protease and Phosphatase Inhibitor Cocktail	Thermo Fisher Scientific	Cat# 78430
Critical commercial assays
MILLIPLEX® Human High Sensitivity T cell Magnetic Bead Panel	Merck	Cat# HTCMAG-28SK
MILLIPLEX® Human Cytokine/Chemokine Magnetic Bead Panel A	Merck	Cat# HCYTA-60K
Human LBP ELISA Kit	Hycult Biotech	Cat# HK315
DNA/RNA Oxidative Damage (High Sensitivity) ELISA Kit	Cayman Chemical	Cat# 589320
Pierce™ BCA Protein Assay Kit	Thermo Fisher Scientific	Cat# A55861
Software and algorithms
Belysa™ Immunoassay Curve Fitting Software	Merck	Version 1.0
ScanR	Olympus	Version 2.03.2
Bio1D	Vilber	Version 15.03a
IBM SPSS Statistics	IBM	Version 20.0
GraphPad Prism	GraphPad Software	Version 9.0.0
Other
LUNA-FL™ Dual Fluorescence Cell Counter	Logo Biosystems	Cat# L20001
Luminex® 200 Instrument	Luminex Corporation	N/A
Fusion FX5 Imaging System	Vilber	N/A

Experimental model and study participant details

Human participants

In this longitudinal observational study, participants were consecutively recruited from the Outpatient’s Department of Endocrinology and Nutrition at University Hospital Dr. Peset (Valencia, Spain) between January 2017 and December 2019. Eligible participants were adults (≥18 years) with obesity (body mass index [BMI] ≥30 kg/m²) seeking nutritional treatment for obesity and with stable body weight (±2 kg) during the three months before enrollment. Exclusion criteria included pregnancy or lactation, severe or inflammatory disease (including cancer, hepatic or renal impairment).

All participants underwent a structured dietary intervention under nutritional supervision, as described in the following section. The study protocol was approved by the hospital Ethics Committee (CEIm ID: 96/16), registered at ClinicalTrials.gov: NCT0507139, and conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent.

A total of 45 adults with obesity were included in the study. Participants were of Caucasian ancestry and had a mean age of 46.2 ± 8.7 years. The cohort comprised 33 females (73.3%) and 12 males (26.7%). Sex and gender were not predefined analytical variables, and the study was not designed or statistically powered to assess sex-specific differences in outcomes.

Method details

Dietary intervention

Following the initial assessment, participants underwent a 24-week dietary intervention consisting of a 6-week very-low-calorie diet (VLCD) followed by an 18-week low-calorie diet (LCD). During the VLCD phase, participants replaced their three daily meals with a liquid formula (Optisource Plus®, Nestlé S.A., Switzerland) providing 654 kcal/day (2738 kJ/day), including 52.8 g of protein, 75.0 g of carbohydrates, 13.5 g of fat, and 11.4 g of fiber, together with essential vitamins and minerals. After 6 weeks, participants received individualized dietary counseling and followed a 18-week LCD based on the Mediterranean diet, with an individualized energy intake of 1200–1800 kcal/day (5023–7535 kJ/day), calculated according to sex, height, weight, and physical activity level. Participants received written and verbal guidance on food quantities and preparation, were advised to consume at least 2 L/day of calorie-free fluids, preferably water, and were encouraged to maintain their usual physical activity. Medication prescriptions remained unchanged throughout the study.

Dietary adherence was assessed by interview with the dietician after the VLCD and at the end of the intervention. Participants were re-evaluated after completion of the 24-week intervention.

Blood sampling, anthropometric and biochemical determinations

Anthropometric measurements were recorded at baseline and 6 months after treatment, including body weight, height, waist circumference, and blood pressure. BMI was calculated as weight (kg)/height (m2). Relative fat mass (RFM) was calculated according to the equation proposed by Woolcott & Bergman. The waist-to-height ratio (WHtR) was obtained as waist circumference (cm)/height (cm).

Blood samples were collected after 12h overnight fast and biochemical determinations were performed in the Clinical Analysis Department of the hospital using standardized laboratory procedures. Remaining samples were centrifuged for 10 min at 1,500g and 4°C for serum isolation, and then preserved at −80 °C for future determinations.

The number of samples analyzed varied across experiments according to sample availability and assay-specific technical requirements. Differences in sample size resulted exclusively from the availability and quality of biological material and technical limitations of individual assays, and not from investigator-driven sample selection.

Levels of systemic cytokines, endotoxemia and oxidized DNA

Levels of tumor necrosis factor α (TNFα), IL-6, IL-1β and IL-18 were evaluated in plasma with multiplex magnetic immunoassays in customized kits (Merk Millipore, Darmstadt, Germany). Samples were assessed in duplicate with a Luminex® 200 instrument (Luminex Corp., Austin, TX, USA), according to the manufacturer instructions. Results were analyzed with Belysa Immunoassay Curve Fitting Software (version 1.0). To evaluate the levels of lipopolysaccharide (LPS) binding protein (LBP) and 8-hydroxy-2′-deoxyguanosine (8-OHdG), serum aliquots were analyzed using the Human LBP ELISA Kit (HycultBiotech, Uden, The Netherlands) and the DNA/RNA Oxidative Damage HS ELISA Kit (Cayman Chemical, Ann Arbor, MI, USA), following the respective manufacturer’s protocol.

Isolation of PBMCs

PBMCs were isolated by standard Ficoll density gradient centrifugation (Ficoll® Paque Plus/Cytiva, Merck Millipore, Darmstadt, Germany). In brief, citrated blood samples were diluted in PBS, layered over Ficoll solution and centrifuged for 25 min at 650g and room temperature. PBMCs on the interface between plasma and Ficoll were collected, washed twice with PBS and resuspended in PBS. Cells were then counted with the LUNA-FL Dual Fluorence Cell Counter (Logos Biosystems, Anyang, Republic of Korea). Some cells were prepared for in vitro experiments, while other aliquots were pelleted and cryopreserved at −80°C for subsequent western blot analysis.

Measurement of superoxide in PBMCs

To detect cytosolic superoxide production, fresh PBMCs were seeded in a 48-well plaque and incubated with 5 μM Dihydroethidium (dHE) fluorescence probe and Hoechst 33342 (for nucleus visualization) for 30 min at 37 °C (Thermo Fisher Scientific, Waltham, MA, USA). Twenty non-overlapping fields per well were automatically acquired using an IX81 Olympus fluorescence microscope equipped with the ScanR system (version 2.03.2; Olympus, Hamburg, Germany) and a 10× objective. Image acquisition was performed using a predefined acquisition template with identical imaging settings for all samples within each experiment. Autofocus was automatically achieved using the Hoechst signal. Images were analyzed using a predefined analysis protocol in ScanR software, in which nuclei were automatically identified based on Hoechst staining and the mean dHE fluorescence intensity per cell was calculated by normalizing the total fluorescence intensity of each field to the number of detected nuclei.

Immunoblotting

PBMCs were lysed for 15 min on ice with RIPA buffer combined with a phosphatase/protease inhibitor cocktail (Thermo Fisher Scientific; Waltham, MA, USA). Solution was sonicated for 30s, vortexed and centrifuged for 15 min at 14,000g and 4°C. Protein concentration was determined using a BCA protein assay (Thermo Fisher Scientific, Waltham, MA, USA).

Equal amounts of protein (25 μg) were resolved by 4–20% SDS-PAGE and transferred onto nitrocellulose membranes. Following a 1h blocking step with 5% skimmed milk in TBS-T, membranes were incubated overnight at 4°C with the specific antibodies against apoptosis-associated speck-like protein containing a CARD (ASC), Caspase 1, NF-κB, NLRP3 and actin for loading control. Blots were then washed and further incubated with the appropriate secondary antibodies for 1h at room temperature. Complete relation of antibodies used can be found in key resources table. Specific antibody-protein signals were visualized and analyzed using an enhanced chemiluminescence reagent (Super Signal West Pico Plus; Thermo Fisher Scientific, Waltham, MA, USA) in the signals were acquired using a Fusion FX5 system and quantified using Bio1D software v15.03a (VilbertLourmat, Marne La Vallée, France).

Quantification and statistical analysis

Sample size (n) for each experiment is indicated in the corresponding figure legend and was calculated to achieve a power of 80% and to detect a 20% difference in the primary outcome – protein detection by immunoblotting – assuming a common standard deviation of 30 units. A minimum of 18 subjects were required for the study. Data were analyzed with IBM SPSS Statistics v20.0 (IBM Corp., Chicago, IL, USA) and Graphpad Prism v9.0.0 (GraphPad Software, Boston, MA, USA) (see key resources table).

Normality of variables was assessed using the Shapiro–Wilk test. Pre- and post-intervention comparisons were performed using paired Student’s t test or Wilcoxon signed-rank test, as appropriate. To quantify the magnitude of change of molecular biomarkers, Δ values were calculated [Baseline - post-intervention] and standardized using z-scores. Correlations were analyzed using Spearman’s coefficients. Statistical significance was established at p < 0.05. Statistical significance in the figures is indicated as follows: ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

Additional resources

The study was registered at ClinicalTrials.gov: NCT0507139.