Exercise and immunity: moderate helps, extreme hurts

The Short Version

Exercise and immunity follow a curve, not a straight line: doing nothing leaves you worse off, regular moderate exercise lowers your risk of getting sick, and very hard or very long efforts push the risk back up for a while. Dr. Roger Seheult calls this the J-curve and backs it with a McMaster University trial in which moderate exercise lowered depression, anxiety and inflammatory markers while high-intensity exercise did less good and raised interferon [1]. Andrew Huberman reaches the same conclusion from immunology: sessions of about an hour or less at a hard-but-not-all-out effort prime the innate immune system for many hours, while marathon-level efforts leave T cell and natural killer cell function well below baseline [2]. Both say the practical rule for cold and flu season is the same: move consistently, keep most sessions moderate and under an hour, don't stack hard efforts, protect sleep, and never train through illness [1][2][3].

Table of contents

What They Said

Roger Seheult: the J-curve and a trial of stressed students

Dr. Roger Seheult, a physician board-certified in internal medicine, pulmonary disease, critical care and sleep medicine and co-founder of MedCram, came to exercise through a question about mental health during the pandemic [1]. He was open about his limits here, saying he is not an expert on mental health and is "just starting to learn about it" [1]. His starting point was that "depression and anxiety are two things that rev each other up," and that a year of pandemic anxiety had taken a toll, including on children in distance learning [1]. Lewis Howes had asked how depression and anxiety feed into a weakened immune system, and Seheult answered with a study rather than a theory [1].

The study was a randomized controlled trial run before COVID by Dr. Jennifer Heisz of the Fit Lab at McMaster University in Ontario [1]. It followed students for six weeks before exams, a period of predictable stress [1]. Students were split into three groups [1]:

• No exercise. They did nothing different.
• Moderate exercise. They raised their heart rate to about 140 beats per minute for about 20 minutes, three times a week, for 18 sessions in all.
• High-intensity exercise. Their heart rates reached around 180.

The results form the backbone of his argument. In the group that did not exercise, depression scores rose sharply over just six weeks [1]. Seheult noted this was not what you would naively expect, since you might think scores would hold steady: "But what that shows is that stress over a period of time increases depression" [1]. In the moderate group, he said, "depression scores went down. Anxiety scores went down. Inflammatory marker scores went down" [1]. His verdict on that arm: "So in other words, there was no bad tradeoff to do a moderate exercise" [1].

The high-intensity group is where the curve bends back. Their depression scores fell, "but not as much as they did in the moderate group" [1]. Self-reported stress did not improve: "they still had actually the same amount of self-reported stress in the high-intensity group" [1]. And "interferon levels went actually up in the high-intensity group" [1]. Seheult presented that interferon rise as a sign of immune stress, placing it alongside the lack of relief from stress as the cost of pushing harder [1].

Lewis summarized the lesson back to him: when your mind or body is under attack, "it's good to have moderate types of consistent, healthy movement," and Seheult answered, "Exactly, exactly" [1]. He then stressed that this one trial is not an outlier: "There is a whole cadre of data that has shown this exact thing. It's called the J-curve" [1]. He described the shape in plain terms: you start at a normal level of risk, and with moderate exercise "the risks go down. But as you do more exercise, the risks go up" [1].

His real-world example was elite athletes. "This is a real problem in people like Olympic athletes," he said; they "have to be very, very cognizant when they are doing high-intensity workouts, that they're not getting sick, because they're actually more susceptible to getting sick" [1]. Lewis added that the stress they are under likely plays a part [1].

Seheult's takeaway was deliberately encouraging: you don't have to be Wim Hof to benefit from hydrotherapy, or Carl Lewis to benefit from exercise [1]. The benefits come at an ordinary, repeatable dose. He added that exercise also improves sleep and eases anxiety [1], both of which feed back into immunity; elsewhere he noted that "cortisol levels have everything to do with immunity and how that affects your immunity" [1].

He placed exercise inside a wider philosophy. Speaking of the Loma Linda Blue Zone and the Adventist Health Study 2, which he said showed seven to 10 years of extra life, he insisted, "It's not a diet. It's a lifestyle. And part of that lifestyle is exercise" [1]. And he compared lifestyle measures to an operating room, where the surgeon wears a mask, the room is kept under positive pressure and the skin is sterilized: no single measure does the whole job, so healthy habits come in addition to masks and distancing, not instead of them [1]. His broader explanation of why fitness matters for infection is that people with obesity, kidney disease or diabetes are figuratively sitting on the edge of a cliff, compensating and getting by until a virus tips them over, and that prevention comes largely down to diet and activity [1]. His exam analogy fits the exercise point too: we have been cramming for quizzes, and now we are sitting the final exam with nothing [1].

Andrew Huberman: the right dose primes the immune system, too much depletes it

In his episode on colds and flu, Huberman made the dose question central [2]. "I'm a huge believer in exercise," he said, describing his own weekly routine of three cardiovascular sessions (one long, one medium, one short) and three resistance sessions [2]. But he wanted to address what he felt gets too little attention: which kinds of exercise support the innate immune system, and which deplete it [2]. "There are certain intensities and durations of exercise that make us more vulnerable to colds and flus," he said [2].

The helpful dose. Huberman cited a 2019 review in the Journal of Sport and Health Science titled The compelling link between physical activity and the body's defense system [2]. Sessions of 60 minutes or less at a hard but not all-out effort, he explained, increase the exchange between blood and lymph and raise baseline innate immune activity for many hours, perhaps up to 24 [2]. He described exercise as a deliberate, controlled stressor: "the appropriate intensity and duration of exercise can act as a stressor that promotes a bit of inflammation," including more white blood cells and natural killer cells, so that the innate immune system is prompted "to almost think that there's something to battle such that if you ever encounter an infection You can defeat it Right off the bat" [2].

He gave a specific example from the review: people who walk briskly for about 60 minutes a day show increased T cell function, natural killer cell activity and macrophage function [2]. "Cytokines increase But not dramatically, okay, so this is a mild inflammation response," he said, and stress hormones such as cortisol, epinephrine and norepinephrine are deployed as well [2]. He called this "an incentive for getting a regular exercise of 60 minutes or less per day," at an intensity sufficient for the innate immune system to deploy more of those chemicals and for lymph and blood circulation to increase their exchange [2]. The minimum can be small: benefits can come from as little as 20 minutes a day, or even 12 minutes if it is very high intensity, like his weekly sprint session [2].

The harmful dose. Then came the key sentence: "it is absolutely not the case that more is better in fact It's probably the case that less is better" [2]. People who have just run a marathon, or are deep into marathon training, are severely immune compromised, he said, with T cell and natural killer cell function well below baseline and stress hormones and cytokines very high [2]. He was careful to say he is not discouraging marathons, only urging people to understand the trade-off [2].

His own rules. Huberman described a concrete routine: warm up for 5 to 10 minutes, keep workouts to about 50 to 60 minutes, be careful about going beyond 75 minutes, avoid two hard sessions in one day, and take one day a week off [2]. His personal records back this up. He keeps a quick daily log of workouts, their intensity and his sleep, taking about 30 seconds, and looking back at what preceded each serious cold or flu, he found three patterns: two hard workouts in one day, hard workouts followed by cold exposure while traveling (especially overseas), and nights of two hours of sleep or less [2]. His colds have become less frequent over the years, from about once a year a decade ago to roughly once every 18 to 24 months [2].

Sleep comes first. After a single bad night, going back to sleep is better if possible; if not, he advised exercising early at 25 to 50% reduced intensity and duration, and never sacrificing sleep for exercise [2]. He noted that even getting only 50 to 75% of the sleep you need weakens innate immunity [2].

Refuel after long fasted sessions. After long fasted morning workouts, especially beyond 60 to 75 minutes, eating complex carbohydrates such as rice, oatmeal or pasta and some fruit within about an hour can reduce inflammatory markers by about 30 to 40% compared with water alone [2]. He also warned that a prolonged caloric deficit, like dieting through winter, can compromise immunity [2].

Don't train through illness. Huberman was firm about the other end of the curve. He admitted the question "If I'm feeling a little bit run down should I exercise or not" has no straightforward answer, because you can't know whether you've been exposed or just feel sluggish [2]. But if you feel whole-body malaise, unexplained fatigue or a throat tickle, he advised going home, taking a hot shower or bath and getting into bed early [2]. If you push into intense activity at that point, "you are going to compromise the function of your innate immune system And it's very likely that you're going to get more sick than you would otherwise" [2]. If you already have symptoms, don't exercise at all, even at home [2]. Only when you are not feeling any of those early signs, he said, is exercise "an excellent way To increase the output of that innate immune system" [2].

Huberman on the nervous system: where exercise fits in the bigger picture

In a separate episode on using the nervous system to strengthen the immune system, Huberman treated exercise as part of the foundation rather than the headline [3]. He listed the generic advice he respects: consistent, high-quality sleep, morning and evening sunlight, about 150 to 180 minutes of zone 2 cardio per week, good nutrition that avoids processed foods, social connection and hydration [3]. Zone 2 cardio is a moderate effort, which puts his baseline recommendation squarely in the low part of Seheult's J-curve.

That episode also adds an explanation that helps make sense of why the dose matters. Huberman described a study in which volunteers who practiced cyclic hyperventilation with breath holds had higher anti-inflammatory IL-10, lower TNF alpha, IL-6 and IL-8, and fewer flu-like symptoms after an endotoxin injection, driven by a large rise in epinephrine (adrenaline) [3]. He used it to explain a familiar experience: pushing through exams or caretaking without getting sick, then falling ill the moment you stop, as adrenaline recedes [3]. In the colds and flu episode he drew the same line between stress types: chronic stress and stress that disrupts sleep impair immunity, but short bouts of stress can enhance it [2]. Exercise is a stressor, so this is the logic of the curve: a short, recoverable dose trains the system, a prolonged or piled-up dose wears it down.

He also explained what happens once you are sick, which clarifies why rest beats training at that point. Sickness behavior, including lethargy and loss of appetite, "is actually a motivated state" designed to accomplish things [3]; resting slows blood circulation, which appears to let the lymphatic system ramp up [3]. Interleukin-1 and interleukin-6 act on serotonin neurons in the dorsal raphe nucleus, which in turn act on sleep-promoting areas of the hypothalamus, producing the sleepiness of early illness [2]. In other words, the tiredness you feel is part of the defense, and training through it works against it [2][3].

Where They Agree, and Where They Don't

Agreement

The core agreement is strong. Both describe a dose-response curve in which some exercise beats none and too much reverses the benefit [1][2]. Seheult's phrasing, that with moderate exercise "the risks go down. But as you do more exercise, the risks go up" [1], and Huberman's, that "it is absolutely not the case that more is better" [2], are the same claim in different words.

They also agree on:

• The benefit is real and immune-related, not just about mood. Seheult reports that inflammatory markers fell in the moderate group [1]; Huberman reports increased T cell, natural killer cell and macrophage activity after an hour of brisk walking [2].
• Elite endurance training carries a cost. Seheult points to Olympic athletes being more susceptible to illness during high-intensity training [1]; Huberman points to marathon runners and those deep in marathon training being severely immune compromised [2].
• Neither says to avoid hard training entirely. Seheult's message is that you don't need to be Carl Lewis to benefit [1], not that intensity is forbidden. Huberman says he isn't discouraging marathons and does a weekly sprint session himself [2].
• Stress is the thread. Seheult notes the stressed students who didn't exercise got more depressed, and that Olympic athletes face stress on top of training load [1]. Huberman distinguishes short, helpful stress from chronic, sleep-disrupting stress [2], and ties the falling ill when you stop pattern to adrenaline [3].
• Exercise is one layer among several. Seheult's operating-room analogy [1] and Huberman's foundations list of sleep, light, nutrition, connection and hydration [3] make the same point: exercise works alongside other habits, not in place of them.
• Sleep matters alongside exercise. Seheult says exercise improves sleep [1]; Huberman says never sacrifice sleep for exercise and that short sleep preceded his own illnesses [2].

Differences

The sources differ mostly in what tips exercise from helpful to harmful.

Intensity versus volume. Seheult's evidence is about intensity: the harmful arm in the McMaster trial was defined by heart rates around 180 [1]. Huberman's is mostly about duration and accumulated load: marathons, sessions beyond 75 minutes, two hard sessions in one day [2]. Huberman even endorses a very high-intensity session of just 12 minutes as beneficial [2], and describes his helpful range as hard but not all-out, for 60 minutes or less [2]. These are not really in conflict, since both describe a large total stress as the problem, but they put the emphasis in different places. A reader following Seheult alone might avoid all high heart-rate work; a reader following Huberman alone might think a short all-out sprint is always fine. Taken together, the fair reading is that short, occasional high-intensity efforts can fit, while frequent, prolonged or stacked hard efforts are what push people up the far side of the curve.

What the evidence measures. Seheult's trial measured depression, anxiety, self-reported stress, inflammatory markers and interferon over six weeks in students [1]; it tells us about mood and inflammation rather than directly counting colds. Huberman's cited review looks at immune cell function and cytokines after exercise sessions [2], and his personal log tracks actual illnesses, though only in one person [2]. Neither source presents a trial that counted infections across exercise levels, so the J-curve is supported by mechanism, markers and observation rather than by a single definitive infection study in these conversations.

Interpreting inflammation. There is a subtle difference in how they treat inflammatory signals. Seheult presents the rise in interferon in the high-intensity group as a cost [1]. Huberman stresses that some inflammation after exercise is the point: a "mild inflammation response" is how the innate immune system gets primed [2], and he calls inflammation a tool you want available, just not so much that it wrecks sleep and gut health [2]. These views fit together once you see that both are talking about degree: a small, brief spike trains the system, while a large or lasting one reflects strain.

Setting and motivation. Seheult framed exercise primarily around the mental toll of the pandemic and its link to immunity [1]; Huberman framed it around colds and flu season directly [2]. Seheult was candid that he was new to the mental health side [1], while Huberman's advice mixes cited research with his own tracking and admits where it is personal [2].

Why It Works

The sources give several mechanisms, from immune cells to hormones to sleep.

Exercise as a training signal for the innate immune system. Huberman describes the innate immune system as the fast, nonspecific layer of defense, with neutrophils, natural killer cells and macrophages rushing to a site of invasion and cytokines like interleukin-1, interleukin-6 and TNF-alpha calling in help [2][3]. Often, he says, the innate system beats a virus before you ever get full-blown symptoms [2]. A moderate session produces a small version of that response: more white blood cells and natural killer cells, mild cytokine release, and stress hormones like cortisol and adrenaline [2]. The system behaves as if there were something to fight, so it is readier when a real virus arrives [2].

Lymph and blood exchange. Huberman singles out the lymphatic system, which works alongside the blood by filtering material and helping produce useful chemicals, as especially relevant to exercise [2]. Moderate sessions increase the exchange between blood and lymph, which he links to raised baseline innate immune activity lasting many hours, perhaps up to 24 [2].

Too much stress suppresses the same cells. At the extreme, the direction reverses: after a marathon or during heavy marathon training, T cell and natural killer cell function fall well below baseline while stress hormones and cytokines climb very high [2]. The same cells that a moderate session boosts are the ones that an extreme effort leaves depleted, which is the biology behind the rising arm of Seheult's J-curve [1][2].

Stress hormones cut both ways. Huberman explains that cortisol is not simply bad: too little is bad for immune function because glucocorticoids help activate natural killer cells and deploy interleukins [2]. Short bouts of stress can enhance immunity, while chronic stress and stress that disrupts sleep impair it [2]. Seheult makes a parallel point in a different context, saying "cortisol levels have everything to do with immunity" [1]. A moderate workout gives a short hormonal pulse; very hard, frequent training keeps the system under sustained stress. Huberman's adrenaline explanation adds a twist: adrenaline during a demanding stretch can hold illness at bay, and people often get sick the moment the demand stops and adrenaline recedes [3], which may help explain why athletes and stressed people fall ill after big efforts.

The mood and inflammation loop. Seheult's reasoning links mental state to immunity: depression and anxiety rev each other up [1], stress over time increases depression [1], and in the McMaster trial moderate exercise lowered depression, anxiety and inflammatory markers together [1]. Huberman's nervous system episode supports that link from the other side: sickness behavior closely mirrors major depression, with shared features including lethargy, loss of appetite and libido and elevated interleukin 6 and TNF alpha [3]. If mood and inflammation move together, an exercise dose that lowers both does double duty, while a dose that leaves stress unchanged and raises interferon, as in the high-intensity arm, gives up some of that benefit [1].

Sleep as the multiplier. Seheult says exercise improves sleep [1]. Huberman says even partial sleep loss weakens innate immunity [2] and that deep sleep is when the brain's glymphatic clearance system is most active, which matters for recovery from infection [3]. That makes sleep the hinge between exercise helping and exercise hurting: training that costs sleep undermines itself, which is why he says never to trade sleep for a workout [2].

Why rest wins once you're sick. When the innate immune system is already fighting, interleukin-1 and interleukin-6 act on serotonin neurons in the dorsal raphe nucleus to drive sleepiness [2], and resting slows circulation in a way that appears to let the lymphatic system ramp up [3]. Adding a hard workout at that point stacks another stressor on a system that is already deployed, which is why Huberman says it will likely make you sicker [2].

Where It Applies, and Where It Doesn't

It applies most to people who are well and active through the winter. Huberman's exercise-for-immunity advice is explicitly for when you are not yet feeling malaise, a throat tickle, a burning in the nasal passages or watering eyes [2]. He singles out people who are out and about a lot during the winter, interacting with many people through work, public transport, the gym, teaching, or kids bringing colds home [2].

It does not apply once you feel sick. If you feel whole-body malaise, unexplained fatigue or a throat tickle, rest instead [2]; with symptoms, don't exercise at all [2]. The J-curve's benefits assume a healthy starting point.

It applies strongly to people under stress. Seheult's evidence comes from students facing exams during a stressful stretch, and he applies it to pandemic anxiety [1]. The group that didn't exercise got worse over six weeks, which suggests the stressed and sedentary have the most to gain from starting moderate exercise [1].

It applies with special force to endurance athletes and people in heavy training. Olympic athletes [1] and marathon runners or those deep in marathon training [2] sit on the risky side of the curve. The message for them is not to stop but to understand the trade-off [2] and be "very, very cognizant" about getting sick during high-intensity blocks [1]. Travel adds risk: Huberman found hard workouts followed by cold exposure while traveling, especially overseas, preceded his own serious colds [2].

It applies to people with chronic conditions, as prevention. Seheult's concern about kidney disease, diabetes and obesity leaving people on the edge of a cliff [1] frames regular activity as part of a longer-term resilience strategy, alongside diet. That is a lifestyle argument, built over months and years, not a quick fix for a cold next week [1].

Limits of the evidence. The McMaster trial Seheult describes lasted six weeks and measured mood, stress and inflammatory markers in students [1], so it doesn't directly show fewer infections, and it may not generalize to every age group. Huberman's personal log is a sample of one [2]. The sources also do not give precise thresholds that apply to everyone: the heart rates in Seheult's trial (about 140 versus about 180) [1] and Huberman's 60- and 75-minute markers [2] are guides from specific studies and personal practice, not universal cut-offs. Heart rate targets in particular depend on the person, and the sources give no formula for adjusting them.

Exercise is not a substitute for other protections. Seheult is explicit that lifestyle measures come in addition to masks and distancing, not instead of them [1]. Huberman's own playbook puts exercise alongside keeping hands away from the face, nasal breathing, fermented foods, sleep and sauna [2]. Anyone who is immune-compromised, pregnant, has a heart condition or is returning to exercise after illness should get a doctor's advice before changing their training, since none of the sources give guidance tailored to those situations.

What To Do With It

1. Exercise most days at a moderate level. Aim for something like the McMaster moderate arm, a raised heart rate for about 20 minutes three times a week [1], or Huberman's baseline of about 150 to 180 minutes of zone 2 cardio per week [3]. Even 60 minutes of brisk walking raises T cell, natural killer cell and macrophage activity [2], and benefits can start at about 20 minutes a day [2].

2. Keep most sessions to an hour or less. Follow Huberman's structure: warm up for 5 to 10 minutes, train for about 50 to 60 minutes, and be careful about going past 75 minutes [2]. His reason: "it is absolutely not the case that more is better" [2].

3. Use high intensity sparingly and briefly. A short, very hard session like Huberman's weekly 12-minute sprint can help [2], but frequent sessions at near-maximal heart rates did less for mood and stress and raised interferon in Seheult's trial [1]. Make hard efforts the exception, not the default, especially in cold and flu season.

4. Never do two hard sessions in one day, and take a day off each week. Two hard workouts in a day was one of the three patterns that preceded Huberman's serious colds, and he takes one day a week off [2].

5. Guard sleep above training. After a bad night, go back to sleep if you can; if not, train early at 25 to 50% less intensity and duration, and never sacrifice sleep for exercise [2]. Short sleep of two hours or less was another pattern before his illnesses [2].

6. Eat after long fasted workouts. If a fasted morning session runs beyond 60 to 75 minutes, eat complex carbohydrates like rice, oatmeal or pasta plus some fruit within about an hour; Huberman says this can cut inflammatory markers by about 30 to 40% compared with water alone [2]. Avoid dieting hard through winter, since a prolonged caloric deficit can compromise immunity [2].

7. Stop at the first signs of illness. Whole-body malaise, unexplained tiredness or a throat tickle mean go home, take a hot shower or bath, and get into bed early [2]. If you have symptoms, skip exercise entirely, even at home [2].

8. Be extra careful around big efforts, races and travel. If you're training for a marathon or in a heavy block, accept that you are more vulnerable [1][2], and take extra care with sleep, hand-to-face contact and exposure in the days after big efforts. Watch the combination of hard training plus cold exposure while traveling [2].

9. Log your training, sleep and illnesses. Copy Huberman's 30-second daily note of workout, intensity and sleep [2]. Over months it shows your own pattern: which loads or combinations tend to come right before you get sick.

10. Treat exercise as one layer of a lifestyle, not a shield. As Seheult put it, "It's not a diet. It's a lifestyle. And part of that lifestyle is exercise" [1]. Pair moderate exercise with good sleep, low-sugar fermented foods, nasal breathing and keeping hands away from your eyes [2][3], and keep using other protections when illness is circulating [1]. If you have a chronic condition or are unsure what intensity is safe for you, talk it through with your doctor first.

Sources

1. The Facts About Vaccines, Improving Immunity, and The Medical Side of Mental Health with Dr. Roger Seheult · The School of Greatness
2. How to Prevent & Treat Colds & Flu · Huberman Lab
3. Using Your Nervous System to Enhance Your Immune System · Huberman Lab