NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Military Nutrition Research; Marriott BM, editor. Food Components to Enhance Performance: An Evaluation of Potential Performance-Enhancing Food Components for Operational Rations. Washington (DC): National Academies Press (US); 1994.

Cover of Food Components to Enhance Performance

Food Components to Enhance Performance: An Evaluation of Potential Performance-Enhancing Food Components for Operational Rations.

Show details

11The Metabolic Responses to Stress and Physical Activity

Edward S.Horton 1 and William R.Beisel 2


Metabolic responses to stress and physical activity are extremely complex, involving many interacting variables. These multiple factors include endocrinological, physiological (cardiovascular and neuromuscular), biochemical, nutritional, and central nervous system (CNS) components, at a minimum. This chapter will attempt to focus upon several of these interacting factors (i.e., nutrient substrates, endocrine/CNS controls, and physiologic responses during exercise) that seem most important and that also have the support of reliable current data.


Initially, some basic concepts must be made clear. First, there are many different kinds of stress to consider: physical exercise, sleep deprivation, psychological stress, environmental stresses (e.g., heat, cold, high altitude, depth, etc.), trauma, and infection, to name a few. In animal models, there are also standardized stresses, such as immobilization or electrical shock.

Second, the human body responses to the military stresses under discussion are obviously very complex and as noted earlier, involve many systems throughout the body. Despite differences in the kinds of stress, body responses may have some broad or generalized commonalities, or in other situations, body responses to a given form of stress may be very specific. For example, acute-phase responses associated with various infections or trauma are rather stereotyped, whereas metabolic responses to exercise may be far less specific.

Third, it must be remembered that these adaptations by the body to the military stresses under discussion, both acute and chronic, are really designed to help in survival. Yet, some of these adaptations may, at times, actually become deleterious.

Thus, researchers must separate out the positive and the negative responses, and in the words of an old song, “Accentuate the Positive, Eliminate the Negative.” Researchers must identify the adaptations that are beneficial to the organism and attempt to determine how they can be enhanced, but at the same time, no harm must be done, as noted earlier (see Chapter 10). Certainly a recommended nutritional “enhancement” should not interfere with potentially beneficial aspects of the stress response.


Because of the complexities of the metabolic responses to military stresses, researchers must focus on the most important response factors that they are trying to enhance. When evaluating stress situations in a military environment, the goal is to enhance the chances for survival of soldiers. For survival, soldiers must be able to think and to act—sometimes quite strenuously. Errors in thinking have occurred because of sleep deprivation, and maintaining physical performance has been a problem under the circumstances of military stress.

A focus on body fuel metabolism is important, because humans must have fuel available for both the CNS (primarily glucose) and for skeletal muscle (a mixture of glucose and fatty acids) in order to do two things: to think properly and then to act physically on the decisions made.

Metabolic Fuel Homeostasis During Physical Exercise

The metabolic fuel homeostasis that occurs during physical exercise, a militarily important form of stress, involves multiple levels of integration. These multiple levels of metabolic integration will be reviewed in this chapter and compared with experimental models of stress, for (or because) there are many similarities between physical exercise in human subjects and CNS models of stress in laboratory animals. Such animal models may reveal strategies that might help to “accentuate the positive.” John Ivy's discussion of ergogenic aids (see Chapter 12) provides some of the possible strategies.

Fuel Homeostasis Under Baseline Conditions

In the normal, nonstressed individual, after an overnight fast and in the resting state, about two-thirds to three-quarters of glucose uptake from blood occurs in non-insulin-dependent pathways. As much as 50 percent is taken up by the brain and CNS in general, and about 15–25 percent by blood cells and kidneys. Only about one-fourth to one-third of glucose uptake goes to insulin-dependent pathways, with about 15–20 percent going into skeletal muscle.

At rest, skeletal muscle derives about 85 percent of its energy from fatty acid oxidation, only about 10 percent from glucose oxidation, and maybe 1–2 percent from oxidation of branched chain amino acids.

The blood glucose concentration stays very constant in these circumstances, because glucose is being taken up by all of these tissues, and at the same time an equivalent amount is being produced, primarily by the liver. Hepatic glycogenolysis accounts for most (about 75 percent) of liver glucose production, with only about 25 percent coming from gluconeogenesis.

At the same time, lipolysis in fat cells is releasing free fatty acids and glycerol. Lipolysis provides free fatty acids for direct oxidation in muscle, liver, and other tissues. Meanwhile, pyruvate, lactate, and amino acids are being released from the intestine and muscle. All of these nutrients provide substrates that feed gluconeogenic pathways in the liver.

Fuel Homeostasis During Exercise

With the onset of exercise, dramatic changes occur in the body because of the rapid increase in energy demands. Initially, rapidly increases, sympathetic CNS activity (Shimazu, 1992), which stimulates glycogenolysis within the muscle itself. As an immediate result, there is a breakdown of muscle glycogen stores, with the production of glucose for anaerobic glycolysis and then, ultimately, for oxidative metabolism in muscle.

At the same time, there is also an activation of lipolysis and a release of free fatty acids and glycerol in and around muscle tissue as well as in adipose tissue. Initially the breakdown of muscle glycogen is more important. In these early moments, before oxidative metabolism begins, lactate may actually be released from muscle, and sometimes a little glucose is also released.

Soon after the onset of exercise, blood flow to the muscle is increased. This allows more substrates to be delivered to muscle. Glucose transport mechanisms in the cell wall are activated also, allowing for an increased uptake of glucose by muscle (Goodyear et al., 1992). However, the blood glucose concentration is maintained at quite a constant level during exercise by a very tightly matched production of glucose by the liver.

Levels of Metabolic Integration During Exercise

The dramatic changes in fuel supply and utilization during exercise are under a very tightly regulated control system that operates at least at five different levels. As previously noted, activation of the sympathetic nervous system is the initial, first-level, response to the onset of exercise (Shimazu, 1992). The very rapid stimulation of sympathetic outflow is measurable by the increase in circulating norepinephrine, and its spillover from the nerve terminals. There is also an activation of a whole cascade of counterregulatory hormone responses over time during exercise, with increases epinephrine, glucagon, cortisol, and growth hormone (Shimazu, 1992). The pattern of hormone responses constitutes the second level of integration. Interactions among insulin, glucagon, and the catecholamines are of major importance in substrate mobilization during exercise.

A third level of integration is the availability of substrates, particularly free fatty acids, lactate, pyruvate, glycerol for gluconeogenesis, and glucose itself. The fourth level of integration is the regulation of blood flow, which controls the delivery rate of substrates to exercising muscle. Fifth, there are local cellular factors that regulate glucose uptake and its intracellular metabolism.


A number of factors influence the pattern of fuel utilization during exercise. These include the intensity of exercise; the duration of exercise; and the effects of prior physical conditioning, of prior dietary intake, and of the existing hormonal milieu.

The first highly important factor is the intensity of exercise. As noted earlier, an oxidation primarily of free fatty acids for energy (with only about 10 percent coming from glucose oxidation) typifies the resting state in skeletal muscle. Exercising at about half the maximum aerobic capacity (i.e., 50 percent of Image pmax.jpg) requires a 50/50 mixture of glucose and free fatty acids, with amino acid oxidation still supplying only 1–2 percent of the energy.

When exercise intensity increases to higher levels, that is 75 percent of Image pmax.jpg or greater, muscles become progressively more dependent on glucose oxidation rather than on fatty acid oxidation.

The duration of exercise also influences the metabolic fuel mixture. The uptake of glucose from blood increases progressively during exercise, peaking after 60–90 min. Then, as exercise persists, free fatty acid concentrations in blood increase, and the muscle gradually shifts over to burning more fatty acids and less glucose.

The effects of physical training tend to cause a more effective adaptation of skeletal muscle, allowing them to oxidize fatty acids more effectively, and thus to spare both muscle and liver glycogen.

The antecedent diet can influence this fuel mixture rather acutely, as can changes in the hormonal milieu, particularly the concentrations of insulin, glucagon, or the catecholamines.


Classical data produced by John Wahren and his group in Stockholm (Wahren et al., 1971) demonstrated the increase in glucose uptake by leg muscles when exercising on a cycle ergometer. Coincident with an increase in oxygen uptake at the onset of exercise was a striking increase in muscle glucose uptake. Glucose uptake was then heightened by increases in the intensity of exercise due to the combination of increased blood flow to the legs and to activation of glucose transport in muscle cells.

This adaptation takes some time to be fully present. The increase in glucose uptake continues for at least 40 min. at moderate intensity exercise, that is about 50 percent of Image pmax.jpg. This observation would suggest an increasing extraction of glucose by the muscles over that period of time. With an increasing duration of exercise, glucose uptake by muscle begins to decline, and there is an increased dependency on free fatty acids for energy metabolism. Free fatty acid uptake becomes progressively greater, so that after 3–4 h of moderate exercise, free fatty acids are the predominant fuel being utilized.

A key to this adaptation is that the liver increases hepatic glucose output to match exactly the peripheral needs, with few exceptions. After 40 min of exercise, the increase in hepatic glucose production is almost entirely due to an increase in glycogenolysis. However, with increasing duration of exercise as hepatic glycogen stores begin to fall, glycogenolysis becomes less significant and gluconeogenesis increases, becoming an important part of hepatic glucose production.

There are two circumstances where blood glucose concentrations are not maintained within a very narrow range. These situations are during brief, high-intensity exercise and during very prolonged, exhaustive, marathon-type endurance exercise.

During an 8–12 min maximum stress test, blood glucose values of a cyclist will overshoot baseline values for a brief time. This is the result of intense sympathetic nervous system stimulation, with an activation of hepatic glucose production that transiently exceeds peripheral glucose utilization.

During prolonged, exhaustive exercise, the liver is simply not able to keep up with the glucose demands. When glycogen stores become depleted in both muscle and liver, the marathon runner must slow down, become hypoglycemic, or ingest glucose.


Sympathetic nervous system activation, which stimulates glycogenolysis within muscle and lipolysis within fatty tissues, is the initial endocrine response to exercise stress. This response is accompanied by an activation of both the adrenocortical axis and the adrenomedullary axis, with rises in both plasma cortisol and epinephrine values. Epinephrine, in turn, stimulates an increase in glucagon secretion and a suppression of insulin secretion.

With exercise above 50 percent of Image pmax.jpg, both norepinephrine and epinephrine concentrations in plasma rise quite linearly in proportion to the intensity of exercise. Epinephrine, as a counterregulatory hormone, also responds to low blood glucose concentrations. Both of these catecholamines stimulate lipolysis, but epinephrine is also a major factor in mobilizing hepatic glucose production.

Plasma insulin concentrations actually fall during exercise, due to stimulation of alpha adrenergic receptors that suppress insulin secretion. When exercise stops, there is a rebound release of insulin and then a gradual return to baseline values.

Glucagon responds more slowly as a function of both the intensity and duration of exercise. Rising catecholamine values are believed to stimulate glucagon release. Glucagon is the primary hormone responsible for stimulating hepatic glucose output, although catecholamines from plasma and/or from direct sympathetic innervation of the liver may also contribute.

As noted, glucagon stimulates hepatic glucose production, while insulin suppresses it. The declining insulin glucagon ratio during exercise is a contributing factor in stimulating hepatic glucose output. On the other hand, the insulin norepinephrine ratio is the key regulator of the lipolytic response, with a decline in the ratio-stimulating lipolysis, as seen during exercise.


To sort out effects of individual counterregulatory hormones on hepatic glucose production, Alan Cherrington and his colleagues at Vanderbilt University (Cherrington et al., 1993; McGuinness et al., 1993) have used the pancreatic clamp technique in the conscious dog model, with implanted hepatic vein and peripheral vascular catheters. By measuring artery-vein concentration differences and by using tracer infusions, rates of glycogenolysis and gluconeogenesis can be estimated. Then, in addition, a pancreatic clamp can be established by infusing sufficient somatostatin to maintain basal insulin and glucagon values. With this technique, effects of each individual counterregulatory hormone can be measured.

When an excess of glucagon is infused (Cherrington et al., 1993; McGuinness et al., 1993), a prompt and sustained increase of arterial glucose concentration is seen, followed by a more gradual increase in peripheral vein glucose. About 70–75 percent of this glucogenic response is due to glycogenolysis; a very small increase in hepatic gluconeogenesis is more delayed.

When epinephrine is infused and all other hormones held constant, an increase in arterial blood glucose values is also seen. This increase is smaller than the one caused by glucagon, but it is also due to an increase in hepatic glucose production. To accomplish this, epinephrine also increases glycogen breakdown and initiates a slow increase in gluconeogenesis (Cherrington et al., 1993; McGuinness et al., 1993). However, epinephrine has a far greater effect on gluconeogenesis than does glucagon. This can be explained by the lack of peripheral effects of glucagon, whereas epinephrine stimulates peripheral lipolysis and increases the delivery of gluconeogenic substrates to the liver, for example glycerol, lactate, and alanine (Cherrington et al., 1993; McGuinness et al., 1993).

Norepinephrine also causes a rise in plasma glucose concentrations, with a gradual slight increase in hepatic gluconeogenesis and a gradual slight decline in glycogenolysis. The effects of norepinephrine are primarily on non-hepatic tissues, where, like epinephrine, norepinephrine increases the production of gluconeogenic substrates for delivery to the liver (Cherrington et al., 1993; McGuinness et al., 1993).

Cortisol alone has little short-term effect on hepatic glucose production or on peripheral utilization (Cherrington et al., 1993; McGuinness et al., 1993).

With an insulin-glucose infusion, there is an increase in peripheral glucose utilization and a suppression of hepatic glucose production (Cherrington et al., 1993; McGuinness et al., 1993).

Combinations of counterregulatory hormones can also be studied, and over longer intervals of time. In an attempt to study the more chronic effects of these hormones, Cherrington's group (Cherrington et al., 1993; McGuinness et al., 1993) infused a combination of epinephrine, norepinephrine, glucagon, and cortisol over a 3-d period. Arterial glucose and insulin concentrations increased, as did the concentrations of all infused hormones. There was also an increase in glucose turnover during this 3-d period, with increased rates in both the production and utilization of glucose.

Such an increase in glucose turnover rates is part of the normal stress response, which appears to be driven by increased hepatic glucose production. There is also an increase in glucose utilization in the periphery, despite some possible degree of insulin resistance.

In other studies of chronic responses, Horton's group infused norepinephrine into rats for 10-d (Lupien et al., 1990). They also found an increased glucose turnover under basal conditions. This increase in glucose turnover persisted even under conditions of insulin stimulation, with no evidence of insulin resistance. The chronic norepinephrine infusion had apparently changed the glucose transport system by increasing blood flow to skeletal muscle as well as to brown fat (Lupien et al., 1990).

Some of the mechanisms by which exercise stress increases glucose turnover thus include changes in hepatic metabolism, changes in blood flow to skeletal muscle, and changes in the glucose transport system itself.


Another model of stress (stress caused by intracerebro-ventricular carbachol infusions) resembles the stress caused by exercise. In experiments done by Mladen Vranic and his group in Toronto (Miles et al., 1991), carbohydrate metabolism was studied after an injection (via a small catheter) of carbachol (an analogue of acetylcholine) into the third cerebral ventricle of dogs, thus stimulating paraventricular nucleus neurons and other nearby brain areas.

This simple injection produced a mild stress response, with activation of the adrenocortical axis, the adrenomedullary axis, and the sympathetic nervous system (Miles et al., 1991; Yamatani et al., 1992). Increases in cortisol, epinephrine, norepinephrine, and glucagon occurred, but insulin values remained unchanged (possibly because of a balanced alpha-, and beta-adrenergic stimulation of pancreatic beta cells). These hormonal changes were accompanied by a very large increase in hepatic glucose production rates, a surprising, matching increase in peripheral glucose uptake and only a small increase in blood glucose (Miles et al., 1991).

The surprising increase in peripheral glucose uptake, in the absence of an increase in plasma insulin, may have resulted from an increased blood flow to muscles and/or possibly to an activation of the glucose transport system in muscle cells.

To determine if the increase in glucose clearance was insulin-dependent or not, Vranic's group did other studies before and after the production of partial alloxan diabetes in the dogs (Miles et al., 1991; Yamatani et al., 1992). A submaximal insulin infusion was used to stabilize both insulin and glucose values in the normal baseline range. In these stabilized diabetic dogs, the carbachol infusion produced marked hyperglycemia, far in excess of the slight rise in blood glucose seen in nondiabetic dogs. The diabetic dogs showed a similar increase in hepatic glucose production, but they lacked the increase in peripheral glucose clearance seen in normal dogs. These findings suggest that at least a permissive amount of insulin (i.e., an amount sufficient to trigger some cellular uptake of glucose) must be present in order to see an increase in muscle glucose uptake in response to this mild model of stress (Miles et al., 1991; Yamatani et al., 1992).


Data on glucose homeostasis during exercise stress reveal a combination of stimulated hepatic glucose production and activated peripheral glucose uptake. This combination is obviously useful, because it provides the glucose energy needed for muscular contractions.

The animal stress model with CNS simulation-activation leads to an increase in glucose turnover, which could provide additional glucose for CNS function and also initiate a stimulated muscle cell uptake of glucose in case muscular exercise would also ensue. However, in the absence of physical activity, little benefit would result from creating severe hyperglycemia. There must be, then, a mechanism for increasing peripheral glucose clearance in stress situations that is independent of the increase in peripheral glucose uptake stimulated by physical exercise.

The skeletal muscle's uptake of glucose is an important area for study, because glucose transport in skeletal muscle is the major rate-limiting step in glucose utilization. Since skeletal muscle is the largest organ system in the body responsible for using glucose, it becomes important to understand how glucose is transported across the plasma membrane and is utilized for energy production in muscle tissue.

Glucose transport occurs primarily by a carrier-mediated pathway involving a whole family of glucose transporter protein isoforms, named Glut 1 through Glut 5. The key transporter isoform in skeletal muscle appears to be Glut 4. The glucose transport system responds to mediators that include insulin, muscle fiber contraction (in skeletal muscle), growth hormone, and glucose itself.

But the main point, in terms of exercise stress, is that insulin and exercise independently stimulate glucose transport in skeletal muscle. Contractions of skeletal muscle, in in vitro systems, can activate the glucose transport system in a total absence of insulin.

Thus there are two major (and apparently independent) stimuli. One is insulin, which normally activates the glucose transport system in a rested, fed state, and during recovery from exercise. And the second, the contraction stimulus, which is responsible for increasing muscle uptake of glucose during exercise and which can act in the absence of insulin.

With this dual system, plasma insulin values fall during exercise. This decrease allows for an increased hepatic glucose production, and at the same time, the contraction stimulus is enough to activate glucose uptake in skeletal muscle. This beautifully coordinated system thereby allows for both the increase in hepatic gluconeogenesis and the increase in skeletal muscle uptake of glucose.


  • Glucose homeostasis is an important factor in attempting to preserve or enhance function during the stress response.
  • It is very important to maintain blood glucose concentrations and the hepatic output of glucose during prolonged stress. However, any attempt to change either hepatic glucose output or peripheral glucose uptake by manipulating the CNS with current levels of knowledge may induce deleterious effects.
  • Recommendations about ergogenic aids and the enhancement of performance during endurance exercise are contained in Chapter 12.
  • Additional research is needed to understand fully the role of the sympathetic nervous system, centrally mediated hormone responses, and direct neuronal effects on both hepatic glucose production and peripheral glucose utilization. The central role of the CNS, and how it might be enhanced, remains unclear.
  • Additional research is also needed on the effects of various neurotransmitters, beta-blockers, and beta-receptors on metabolic homeostasis during stress.


  • Cherrington, A.D., R.W.Stevenson, K.E.Steiner, C.C.Connolly, M.Wada, and R.E. Goldstein 1993. Acute hormonal regulation of gluconeogenesis in the conscious dog. Pp. 199–208 in New Concepts in the Pathogenesis of NIDDM, S.Efendie, editor; , C.J.Ostensen, editor; , and M. Vranic, editor. , eds. New York: Plenum Publishers. [PubMed: 8249683]
  • Goodyear, L.J., M.F.Hirshman, and E.S.Horton 1992. The glucose transport system in skeletal muscle: Effects of exercise and insulin. Pp. 201–215 in Integration of Medical and Sports Sciences. Y.Sato, editor; , J.Poormans, editor; , I. Hashimoto, editor; , and Y.Oshida, editor. , eds. Basel, Switzerland: Karger.
  • Lupien, J.R., M.F.Hirshman, and E.S.Horton 1990. Effects of norepinephrine infusion on in vivo insulin sensitivity and responsiveness. Am. J. Physiol. 259(2 Pt1):E210-E215. [PubMed: 2116726]
  • McGuinness, O.P., T.Fugiwara, S.Murrell, D.Bracy, D.Neal, D.O'Connor, and A.D. Cherrington 1993. Impact of chronic stress hormone infusion on hepatic carbohydrate metabolism in the conscious dog. Am. J. Physiol. 265(2 Pt1):E314–E322. [PubMed: 8368302]
  • Miles, P.D.G., K.Yamatani, H.L.A.Lickley, and M.Vranic 1991. Mechanism of glucoregulatory responses to stress and their deficiency in diabetes. Proc. Natl. Acad. Sci. 88:1296–1300. [PMC free article: PMC51004] [PubMed: 1996330]
  • Shimazu, T. 1992. The hypothalamus and neural feed-forward regulation of exercise metabolism Pp. 20–32 in Integration of Medical and Sports Sciences, Y.Sato, editor; , J.Poormans, editor; , I. Hashimoto, editor; , and Y.Oshida, editor. , eds. Basel, Switzerland: Karger.
  • Wahren, J., P.Felig, G.Ahlborg, and L.Jorfeldt 1971. Glucose metabolism during leg exercise in man. J. Clin. Invest. 50:2715–2725. [PMC free article: PMC292221] [PubMed: 5129319]
  • Yamatani, K., Z.Q.Shi, A.Giacca, R.Gupta, S.Fisher, H.L.A.Lickley, and M.Vranic 1992. Role of FFA-glucose cycle in glucoregulation in total absence of insulin. Am. J. Physiol. 263(4Pt 1):E646–53. [PubMed: 1415684]



Edward S.Horton, Joslin Diabetes Center, Boston, MA 02215


William R.Beisel, Johns Hopkins School of Hygiene and Public Health, Baltimore, MD 21205

Copyright 1994 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK209038


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (6.6M)

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...