Insulin: much more than glucose control

Resources, stress and the cost of staying alive

Resources have an external and an internal dimension. Outside the organism are food, nutrient quality, opportunities to obtain food and the predictability of access. Inside it are glycogen and fat stores, functional muscle and the cellular capacity to convert substrates into usable energy. Equal food availability need not mean equal physiological capacity to use it.

Life-history theory provides one way to organize these distinctions. Organisms face trade-offs among growth, maintenance, reproduction and survival because resources and time are finite. Investing in one function can constrain another. I find this perspective useful because it places metabolism within a trajectory: a nutrient has significance for what an organism can do now and what it can build for later. [7]

Stress changes the demands on that budget. Allostasis describes the physiological adjustments that maintain stability as conditions change; allostatic load describes their cumulative cost. McEwen and Wingfield explicitly connected this framework to energy supply and expenditure. They also distinguished overload caused by insufficient supply from overload that can occur despite abundant calories, for example under persistent social demands. Food abundance therefore does not guarantee a low-cost physiological state. [6]

I would frame health here as the capacity to reorganize these investments as circumstances change. Mobilization is appropriate during some challenges; storage and growth are appropriate under other conditions. Vulnerability may emerge when the regulatory response becomes poorly matched to the current demands, or when repeated adjustments carry costs that the organism cannot adequately recover from. This is an organizing interpretation, rather than a complete definition of disease.

Insulin participates in the allocation of resources

At the peripheral level, insulin influences several decisions about nutrient use. In muscle and adipose tissue, it promotes glucose uptake through the recruitment of GLUT4 transporters to the cell surface. It favors glycogen storage, restrains hepatic glucose production and inhibits the release of fatty acids through lipolysis. These actions change the availability of substrates across tissues. They make insulin relevant to the balance between accumulation, use and mobilization of resources. [2]

Inside cells, insulin receptor activation recruits signaling systems such as the IRS–PI3K–Akt pathway. IRS proteins transmit the receptor signal; PI3K and Akt relay it to processes involved in metabolism, cellular survival and growth. Insulin and insulin-like growth factors have related, partly overlapping signaling systems. Their effects depend on the receptor, cell type and physiological context. [1] [2]

Stress hormones modify this coordination. Glucocorticoids and sympathetic activity can affect insulin secretion, tissue responsiveness and substrate mobilization. These relationships vary across tissues and over time. I would avoid reducing them to a contest in which cortisol always defeats insulin: the relevant question is how the combined signals redistribute resources under particular demands. [2]

This makes insulin a candidate mediator of the relationship between resources and biological investment. Its concentration alone is an incomplete description. Access to a tissue, receptor responsiveness and downstream signaling determine what the signal actually does. A developmental change in any of these components could alter resource allocation without producing the same effect in every organ.

Development changes the rules of the response

During development, those signaling systems are being established while they are also regulating ongoing growth. Nutrition and stress can therefore influence both immediate physiology and the later responsiveness of the system. In our 2021 review, we argued for studying insulin alongside glucocorticoids when investigating how early adversity becomes biologically embedded. [2]

My interpretation is that this embedding may include a lasting change in how resources are evaluated and allocated. The later organism could respond differently to the same meal, challenge or opportunity because its regulatory systems developed under different conditions. This possibility connects with the thrifty behavior hypothesis we developed with colleagues: early conditions may affect behavioral routes to obtaining and conserving energy, as well as metabolic routes to using it. [8]

An adaptive explanation is tempting, but it has to earn its place. A lasting effect after fetal adversity does not by itself show that development predicted a future environment or produced a survival advantage. The useful claim is a testable one: early conditions change later physiological and behavioral responsiveness, and the costs or benefits depend on the environment subsequently encountered.

Behavior belongs inside the metabolic model

This is where I would push the argument further. Behavior changes the resource budget. Seeking food brings nutrients into the organism; avoiding a threatening setting may conserve safety while limiting access to those nutrients; exploration expends energy but can reveal new opportunities. Feeding and emotional responses are therefore relevant to the same allocation problem that endocrine regulation addresses.

Insulin has access to the neural processes through which these priorities are expressed. Chen and colleagues’ independent review describes its effects in hypothalamic circuits involved in energy regulation and in mesolimbic circuits connecting the ventral tegmental area with the nucleus accumbens. In the latter, insulin can influence dopamine neuron activity, dopamine release and transporter-mediated clearance. Its effect depends on the circuit and the route through which it acts. [1]

The mechanisms extend beyond dopamine neurons themselves. In mouse experiments discussed in the review, deleting insulin receptors from astrocytes reduced evoked dopamine release and altered behavior in anxiety- and depression-related tests. Insulin also affects synaptic transmission and plasticity: studies in the hippocampus describe changes in the trafficking of NMDA and AMPA glutamate receptors, which influence how neurons communicate and how connections change with experience. [1]

My hypothesis is that such effects help translate metabolic state into the relative priority of reward seeking, learning and responding to threat. This is more specific than proposing that insulin simply makes an animal eat less or feel safer. I suspect the important outcome is a change in the weighting of competing behavioral options, with the direction depending on the circuit, the animal’s developmental history and the demands of the situation.

Our early-overfeeding study provides an example of why this distinction matters. Male rats raised in small litters showed altered insulin signaling in the ventral tegmental area and dopamine-related protein changes. Their ordinary chow intake was unchanged, yet they ate more sweet food during a brief challenge and differed in learning about a location paired with sweet food. The pattern concerns behavior in particular contexts, rather than a uniform increase in appetite. [3]

A developmental link between metabolism, inflammation and emotion

Inflammatory responses alter metabolic demands and can influence neural function. Microglia, the brain’s resident immune cells, participate in these responses and also support the development and remodeling of neural circuits. Their functions extend beyond inflammation. I think the relationship between insulin signaling, neuroinflammation and circuit development deserves a more central place in developmental models of behavior. [1]

Our fetal-growth-restriction study gives this proposal a concrete starting point. We examined insulin responses in the medial prefrontal cortex at birth, around weaning and in adulthood. Gene responses differed with age, sex and prenatal growth, and some insulin responses were reduced after growth restriction. In control males, the insulin-regulated genes were enriched for genes associated with microglia; this pattern was absent in growth-restricted males. Immune, metabolic and developmental pathways, including Wnt/β-catenin signaling, appeared among the affected processes. [4]

I interpret this as a reason to investigate whether early adversity changes the coordination between metabolic signals and the cellular work of brain development. One possibility is that altered insulin responsiveness changes how neuroinflammatory signaling and synaptic remodeling respond to later challenges. The microglial gene enrichment makes this a candidate mechanism to test; it does not demonstrate ongoing inflammation, establish a particular microglial state or prove that microglia mediate the behavioral effects. [4]

The behavioral experiment is especially relevant to this argument. Direct insulin infusion into the adult medial prefrontal cortex changed emotional responses in a sex- and context-dependent manner. Growth-restricted animals given insulin spent more time in a brightly lit compartment, consistent with reduced avoidance. A developmental effect remained responsive to a regional intervention in adulthood. [4]

My reading is that insulin-sensitive prefrontal processes may help regulate the balance between avoidance and engagement. An effect on prefrontal–amygdala communication is one plausible route to investigate. The study did not establish that circuit mechanism, and it would be premature to describe the infusion as restoring a failed prefrontal brake. What it established is enough to motivate the next experiment: changing insulin signaling in a defined brain region can change behavior after fetal adversity. [4]

Vulnerability may involve a loss of coordinated flexibility

In our UK Biobank work, early adversity was associated with the coexistence of psychiatric and cardiometabolic conditions. That study did not measure insulin signaling. It describes a human pattern that a developmental account of shared vulnerability needs to explain. [5]

I suspect that part of this vulnerability may lie in a loss of coordinated flexibility across metabolic and behavioral systems. A response appropriate to scarcity, repeated threat or an early developmental constraint may carry different costs when food availability, social demands or opportunities change. Insulin could participate in that persistence through altered tissue sensitivity, neural signaling and interactions with stress regulation and inflammatory processes. This is a proposed route to comorbidity, rather than a claim that one hormone explains the entire pattern.

The proposal makes predictions. After early adversity, insulin responses in the brain and peripheral tissues may diverge; changes in behavior may become most apparent when the animal must choose between competing demands; and the effects of altering insulin signaling should vary with developmental stage, sex and context. Some of our findings are consistent with these predictions. Testing the integrated model requires measuring these processes together and manipulating the relevant pathways. [2] [3] [4]

I would prioritize experiments that connect cell-specific insulin responsiveness to inflammatory signaling, broader microglial gene activity, circuit function and behavior within the same developmental model. They could distinguish whether insulin is helping organize the response, compensating for another change, or simply marking it. That distinction is where a broader account of insulin can become a stronger explanation of development and lifelong health.