The research team began by examining peripheral blood from healthy human volunteers, stratified into young adults aged 18 to 40 and middle-aged adults aged 41 to 65. When natural killer cells were stimulated in the laboratory with the cytokine interleukin-15, a potent activator of NK cell activity, a striking sex-specific pattern emerged. Cells from middle-aged men expressed significantly lower levels of CD107a, a marker of degranulation that reflects the cell’s ability to release its cytotoxic payload, and produced markedly less interferon-gamma, the signature cytokine that coordinates antiviral and antitumor responses. In contrast, natural killer cells from middle-aged women performed comparably to those from young women across these functional readouts. Enzyme-linked immunosorbent assays of purified CD3-negative CD56-positive cells confirmed that the cytokine secretion deficit in middle-aged men was genuine and not an artifact of intracellular staining.

Surface receptor profiling added further nuance. The investigators measured a panel of receptors known to regulate natural killer cell development and function, including CD150, 2B4, CD84, CD319, NKG2A, NKG2D, CD48, Ly9, Ly108, Tim-3, and CD69. Most of these were unchanged between age groups, but young male donors displayed higher expression of CD48, a ligand involved in activating signals. The authors conclude that reduced cytotoxic function and diminished cytokine production are defining characteristics of natural killer cells in middle-aged men, even in the absence of overt disease.

To dissect the mechanisms behind this human observation, the team turned to a mouse model, comparing 8-week-old young males with 48-week-old middle-aged animals, an age that corresponds roughly to human midlife. Consistent with prior work, the middle-aged mice were substantially heavier, with magnetic resonance imaging revealing elevated fat mass and enlarged epididymal and inguinal white adipose depots. Brown adipose tissue also increased in mass, but its thermogenic browning capacity was significantly diminished. Metabolic cage experiments painted a coherent picture of midlife metabolic decline: oxygen consumption, carbon dioxide production, respiratory exchange ratio, and whole-body energy expenditure all fell significantly, even though spontaneous activity levels were comparable between age groups. Food and water intake were actually reduced, underscoring that the adiposity of middle age reflects a fundamental shift in basal metabolism rather than simple overconsumption.

Flow cytometric analysis across the spleen, bone marrow, liver, peripheral blood, and adipose tissues revealed that natural killer cells were among the most affected immune populations. The relative proportion of NK cells dropped significantly in the spleen and liver, and within adipose tissue the percentage fell in epididymal white adipose tissue, with numbers per gram of tissue reduced across all three fat depots. Maturation, tracked using the classical CD27 and CD11b staging scheme, was also impaired. Middle-aged mice showed an accumulation of immature CD27-positive single-positive cells and a loss of mature CD11b-positive single-positive cells in the spleen and bone marrow, a pattern resembling that previously described in much older animals. In the fat depots, mature subsets were similarly depleted. Notably, the liver appeared relatively spared, suggesting tissue-specific vulnerability. Broader immune profiling using t-distributed stochastic neighbor embedding showed that other lymphocyte populations were largely unchanged, with the notable exception of increased M1 and M2 macrophages in the spleen, reinforcing that natural killer cells represent a particularly sensitive target of the midlife immune environment.

The receptor landscape of natural killer cells shifted in ways that would be expected to blunt surveillance. In the spleen, the inhibitory receptors KLRG1 and TIGIT were downregulated while Ly49A was upregulated, and the activation marker CD69 along with the immature markers CD117 and CD127 were elevated, consistent with a less differentiated, functionally compromised state. Adipose tissue NK cells displayed their own distinctive receptor changes, with broad upregulation of multiple activating and inhibitory receptors in epididymal fat. Survival analysis helped explain the falling cell counts: splenic natural killer cells from middle-aged mice showed increased Annexin V positivity, indicating heightened apoptosis, alongside reduced Ki-67 expression, a marker of proliferation. Proliferation was also reduced in bone marrow, epididymal fat, and brown fat. Single-cell RNA sequencing of splenic and bone marrow NK cells reinforced the functional picture, revealing downregulation of NK cell activation pathways and upregulation of p53-mediated signaling in middle-aged animals.

Function followed form. When splenocytes or bone marrow cells were challenged ex vivo with MHC class I-deficient target cells such as YAC-1 and RMA-S, natural killer cells from middle-aged mice produced significantly less interferon-gamma and expressed less surface CD107a than those from young controls. The deficit extended deep into the adipose tissue microenvironment: NK cells isolated from epididymal, inguinal, and even brown fat depots showed markedly impaired degranulation and cytokine production. Imaging flow cytometry using the neutral lipid dye Bodipy 493/503 provided a possible mechanistic clue. Natural killer cells from middle-aged mice accumulated more intracellular lipid than those from young mice, with the most pronounced lipid burden observed in cells residing in epididymal white adipose tissue. This finding echoes earlier reports that lipid droplet accumulation inside NK cells can compromise their cytotoxic machinery, and it suggests that a lipid-enriched adipose microenvironment may directly poison the antitumor capacity of these lymphocytes.

To separate the effects of aging from those of obesity itself, the researchers fed 8-week-old young male mice a high-fat diet deriving 60 percent of calories from fat for 16 weeks. These diet-induced obese animals, though young, mirrored many of the NK cell defects seen in their middle-aged counterparts. Splenic and bone marrow natural killer cells showed reduced interferon-gamma production and degranulation upon target cell stimulation, and cells within all three adipose depots displayed the same functional impairment. The authors note that NK cell dysfunction was, if anything, more pronounced in the high-fat diet group than in middle-aged mice, likely because the dietary model produced even greater adipose expansion. Together with previous reports that dietary restriction can enhance NK cell function, this experiment supports the interpretation that excess adiposity itself, independent of chronological age, is a major driver of the immune decline observed in midlife.

The study has limitations that the authors acknowledge. The comparison between middle-aged obesity and diet-induced obesity is indirect and cannot fully disentangle the two conditions, and the focus on male participants and male mice, justified by the far more pronounced weight gain and adipogenesis seen in males during middle age, leaves sex-specific differences in female biology largely unexplored. Nevertheless, the implications are considerable. Epidemiological data indicate that middle-aged adults with obesity face a higher mortality risk than expected for their conditions, and the loss of natural killer cell quantity and surveillance documented here offers a plausible immunological mechanism linking midlife weight gain to increased vulnerability against cancer and infections. Because white adipose tissue is the first organ to show age-related transcriptomic changes beginning in middle age, and because longevity-promoting pathways such as sirtuins and forkhead box proteins typically suppress adipogenesis, the study positions the expanding fat depot not merely as a passive energy store but as an active remodeler of systemic immunity. If confirmed in larger and more diverse cohorts, these findings suggest that maintaining metabolic health through the middle decades could help preserve the innate immune defenses that guard the body against malignancy and viral disease well before old age arrives.

Subject of Research: Natural killer cell dysfunction during midlife obesity in humans and mice

Article Title: Natural Killer Cell Dysfunction Is Emerging During Midlife Obesity

Article References: Biao, R., Wang, X., Fu, J., Guo, Y., He, J., & Du, J. (2026). Natural Killer Cell Dysfunction Is Emerging During Midlife Obesity. Aging Cell, 25(9), Article e70707. https://doi.org/10.1111/acel.70707

Image Credits: AI Generated

DOI: 10.1111/acel.70707

Keywords: natural killer cells, midlife obesity, immune aging, white adipose tissue, interferon-gamma, lipid accumulation, inflammaging, high-fat diet, basal metabolic rate, NK cell maturation, cytotoxicity, Aging Cell

The epidemiological picture is unambiguous. Global analyses pooling 3,663 population-representative studies covering 222 million children and adults show that obesity rates have more than doubled since 1990, and forecasts extending to 2050 project continued escalation, particularly in low- and middle-income countries that now bear the greatest economic and clinical burden. Against this backdrop, cohort studies tracking hundreds of thousands of participants for decades reveal a consistent pattern: obesity in midlife, typically defined as the forties and fifties, carries a particularly high risk of later cognitive decline and dementia. Landmark work following Kaiser Permanente members showed that central obesity in midlife nearly tripled the risk of dementia more than three decades later, and more recent mega-analyses encompassing nearly six million participants confirm that abdominal adiposity predicts cognitive impairment even after accounting for body mass index. Notably, the relationship is time-dependent, and some studies suggest that weight loss in later life does not fully erase the risk accrued during middle age, implying a window of vulnerability that may never completely reopen.

The review is careful to acknowledge complexity. In some elderly cohorts, higher body mass index in the final years of life appears protective against dementia, a finding many researchers attribute to reverse causation, as incipient neurodegenerative disease suppresses appetite and causes weight loss long before diagnosis. Sex also matters: observational data from Japanese, Chinese and Norwegian cohorts reveal divergent dementia risk patterns between men and women, likely reflecting differences in fat distribution, hormonal milieu and vascular disease. These nuances complicate simple messaging but do not undermine the central conclusion that midlife obesity is a modifiable risk factor for cognitive decline, one that the 2024 Lancet Commission on dementia prevention explicitly recognizes alongside hearing loss, depression and physical inactivity.

What makes obesity neurologically destructive? The authors trace a cascade that begins in the hypothalamus, the brain’s metabolic command center. In healthy physiology, leptin released from adipose tissue and insulin released from the pancreas signal satiety to arcuate nucleus neurons, curbing food intake. Chronic overnutrition disrupts this loop in a feed-forward spiral: elevated leptin and free fatty acids trigger inflammatory activation of hypothalamic microglia and astrocytes, which impairs leptin and insulin signaling, dulls satiety perception and permits further weight gain. Animal studies reveal that hypothalamic neurons accumulate iron, extracellular matrix remodeling becomes pathologically fibrotic, and the gliovascular interface that regulates blood flow to the hypothalamus is altered in ways that promote hypertension. Even more striking, recent work in humans and macaques demonstrates that these hypothalamic changes are not limited to rodents fed laboratory high-fat diets; they mirror the transcriptomic and structural alterations seen in obese patients undergoing neuroimaging and, in some cases, in postmortem tissue.

Beyond the hypothalamus, obesity inflicts damage on the hippocampus and prefrontal cortex, the twin pillars of memory and executive control. Neuroimaging in the UK Biobank and other large cohorts shows that higher body mass index and central adiposity are associated with reduced gray matter volume, degraded white matter integrity and lower cerebral myelin content, changes that partially mediate the link between adiposity and poorer cognitive test performance. Mechanistically, obese adipose tissue becomes inflamed and insulin resistant, releasing a chronic trickle of cytokines, free fatty acids and adipokines into the circulation. Some of these signals cross or disrupt the blood-brain barrier, activating resident microglia and provoking neuroinflammation in memory circuits. Experimental work in mice demonstrates that lipid-droplet-accumulating microglia, a dysfunctional proinflammatory state previously linked to aging, appear in the obese hippocampus, while astrocytes lose their process arborization and glutamate buffering capacity. Mitochondrial dysfunction, oxidative stress, altered microRNA expression and epigenetic changes further erode synaptic plasticity, and activation of the cGAS-STING innate immune pathway, typically associated with cellular senescence, has emerged as a shared mechanism linking metabolic stress to both obesity-related and age-related neurodegeneration.

Vascular damage compounds the neuronal injury. Obesity promotes endothelial dysfunction, cerebral small-vessel disease and blood-brain barrier leakage, all of which deprive brain tissue of adequate perfusion and permit peripheral immune cells and inflammatory mediators to infiltrate the parenchyma. Cerebrovascular dysfunction is particularly relevant because the same pathology underlies vascular dementia and worsens Alzheimer’s disease progression. Studies of middle-aged mice show that visceral adiposity alone, independent of diet-induced weight gain, is sufficient to impair cerebral blood flow regulation and cognitive performance, and human neuroimaging confirms reduced cerebral blood flow in obese adults, an effect partially reversed by physical activity. Brain insulin resistance adds another layer: neurons in the hippocampus and prefrontal cortex require insulin for synaptic plasticity, and selective insulin resistance in these regions, documented in obese humans using functional imaging, correlates with impaired memory encoding and altered food decision-making, creating a vicious cycle in which metabolic dysfunction degrades exactly the circuits needed to regulate eating behavior.

Perhaps the most consequential insight from the review is that much of this damage may be preventable or partially reversible. Weight loss interventions consistently improve cognition, with meta-analyses showing that obese and overweight individuals who lose weight exhibit gains in memory, attention and executive function. Dietary strategies ranging from the Mediterranean and MIND diets to intermittent fasting and ketogenic protocols have shown cognitive benefits in randomized trials, though effects vary in magnitude and durability. Exercise deserves particular emphasis: aerobic training restores brain insulin sensitivity in sedentary obese adults, improves executive function and working memory across age groups, enhances cerebrovascular health and, in animal models, prevents obesity-induced white matter damage even independently of weight loss. Bariatric surgery, the most effective weight-loss intervention available, produces sustained cognitive improvements measurable years after the procedure, and observational cohorts suggest reduced incidence of mild cognitive impairment and Alzheimer’s disease and related dementias among surgical patients compared with matched controls.

Pharmacotherapy is now entering the picture with unprecedented momentum. Glucagon-like peptide-1 receptor agonists, originally developed for diabetes and now widely prescribed for obesity, appear to exert neuroprotective effects beyond weight reduction. Preclinical studies show that semaglutide and exenatide dampen hypothalamic and hippocampal neuroinflammation, restore blood-brain barrier integrity and improve cognition in diabetic and obese rodent models, while clinical data from the REWIND trial and a phase 2b liraglutide study in Alzheimer’s disease hint at cognitive benefits in humans. Most notably, the large phase 3 EVOKE and EVOKE+ trials testing oral semaglutide in early-stage symptomatic Alzheimer’s disease have recently reported results, marking the first time an obesity-related metabolic therapy has been rigorously evaluated as a dementia treatment. The review also highlights emerging strategies targeting cellular senescence, hypothalamic inflammation and the gut-brain axis, including fecal microbiota changes induced by fasting that boost microglial function.

The authors close by identifying critical gaps. Most human studies are observational, leaving causality incompletely established; the optimal timing, intensity and duration of interventions remain unknown; and virtually no trials have tested whether preventing midlife obesity lowers dementia incidence decades later. Sex-specific mechanisms, the contribution of childhood adiposity to adult brain structure, and the long-term cognitive effects of GLP-1 therapies all require dedicated investigation. Yet the message for clinicians and the public is already actionable: the brain is a metabolic organ, and midlife is the decisive window. As global obesity rates continue their relentless climb, the review argues that protecting cognition in an aging world may depend less on new drugs for the dementia ward than on the unglamorous work of preventing and treating obesity decades before memory begins to fail.

Subject of Research: The effects of obesity on brain health, cognitive function and dementia risk

Article Title: Effects of obesity on brain health and cognition

Article References: McEntee, C. M., Savelieff, M. G., Noureldein, M. H., Eid, S. A., Grisold, W., Hassenstab, J. J., & Feldman, E. L. (2026). Effects of obesity on brain health and cognition. Nature Reviews Neurology. https://doi.org/10.1038/s41582-026-01251-6

Image Credits: AI Generated

DOI: 10.1038/s41582-026-01251-6

Keywords: obesity, brain health, cognition, dementia, cognitive decline, hippocampus, hypothalamus, neuroinflammation, blood-brain barrier, GLP-1 receptor agonists, bariatric surgery, midlife obesity