Abstract
The interest in extraterrestrial travel is accelerating, as is evident not only in the expanding governmental programs, but also in private commercial ventures. Preparing the professional astronaut for space travel presents a different challenge than does assessing whether a commercial space traveler is fit enough to sustain the rigors of such an adventure. For both categories of voyagers, however, we must consider the extensive differences between men and women, including both their strengths and vulnerabilities. The population of astronauts that have been trained and assessed is relatively small. Therefore, the personalized assessment and preparation of the individual space traveler is currently a subject of intense interest. This roundtable discussion will consider the many issues surrounding our newest explorations into extraterrestrial flight.
I would like to note that the FAA 20th Annual Commercial Space Transportation Conference took place a few weeks ago in Washington, DC. Some of the human factor considerations included: (1) how heavily we should regulate the commercial space flight industry, (2) how much we should allow individual space flight participants to take the risks of space flight, assuming they are given proper informed consent, and (3) what the medical considerations are for assessment, readiness screening, preparedness, countermeasures, and acceptability to fly. I would also like to explore briefly some of the legal issues raised at the conference and to recognize that the FAA Office of Commercial Space Transportation is leaning toward requiring informed consent, which allows individuals to take these risks while establishing selected restricted conditions. This will be quite different from the way in which government space programs have approached the issue.
As a brief background, my original work in aerospace medicine began at NASA Ames Research Center under the chief medical officer. I was focused on studying the molecular underpinnings of the physiologic and performance phenotypes in extreme environments. Among our study paradigms was the NASA 20G long radius human centrifuge. We've come to realize that many of the health, safety, and performance issues facing professional astronauts will be shared by private citizens who will eventually enter space with commercial space flight providers.
I would like to begin our discussion by providing some background on how my group looks at the convergence of gene-associated or gene-linked molecular networks, and further explore the dynamic convergence between genes, proteins, metabolites, dietary intakes, micronutrient status, drug metabolism, physical activity, and environment in space. In order to establish context, I would like to approach this topic using three representative examples that may provide insight into a novel way in which to view the molecular landscape in humans entering space. From this vantage point, I will then move on to get your insights on how each of you approaches molecular complexity in your space flight research and clinical management paradigms.
The first example I want to explore relates to genes, gender, and the one-carbon molecular network (methylation). One of the genotypes that we are interested in profiling involves the gene for phosphatidylethanolamine methyltransferase, or PEMT. PEMT is an enzyme that transfers a single methyl group to phosphatidylethanolamine in order to form phosphatidylcholine. Expression of this gene is highly estrogen sensitive.
Premenopausal women have sufficient estrogen to activate PEMT and do so even in the presence of a choline-deficient diet. Postmenopausal women often have insufficient estrogen to activate PEMT and have been shown to suffer more symptoms of fatty liver when on a choline-deficient diet. Choline therefore appears to be essential in the diet of a postmenopausal woman who is not on estrogen, in contrast to a premenopausal woman or a postmenopausal woman on estrogen. These dynamics are accentuated when the PEMT gene mutation is present.
We have previously reported (in 2013) how disordered one carbon metabolism—methyl group deficient—may induce DNA instability via reduced conversion of uracil to thymine. This can result in uracil accumulation in the nucleus and endogenous mutational events. Moreover, a subset of astronauts on the International Space Station (ISS) suffers from visual impairment, which has links to one carbon gene and one carbon micronutrient status (methyl donors). The choline-PEMT dynamic is yet one more key element in one carbon metabolism that we believe warrants consideration in space.
Our second example that relates to genes, gender, and potentially radiation is the problem of iron overload in space, which may result from carrying variants of the genes for hemochromatosis (Hfe). In some of the populations that we study, we observe a high prevalence of elevated transferrin saturation, sometimes 70–80% in fairly young adults. This finding is often coupled with heterozygosity or homozygosity of one of the two Hfe genes, as well as with elevated serum ferritin or serum iron.
Why is this important in space? Iron is a known oxidant in humans. It is known that elevated levels of 8-hydroxyguanosine are found in hemochromatosis on Earth—a signature of DNA damage. A question we must answer is whether radiation coupled with a high body burden of redox reactive iron confers additional risk for DNA damage? Moreover, we are also asking what the effect of high iron body burden is on those who enter the variable oxygen conditions (space suit/habitat) that are being proposed by NASA, as detailed in the current Design Reference Missions (DRMs). A DRM is a detailed plan to describe the content, risks, difficulties, technologies, and scale of the mission.
These questions can be further refined to address sex differences, since premenopausal women lose blood (and iron) through menstruation and do not accumulate iron in the same manner as men. The clinical, dietary, and nutritional management of men and women with divergent profiles of Hfe, transferrin saturation, ferritin, serum iron, and age would vary considerably. Optimistically, this affords us the ability to apply personalized countermeasures in space in the same manner that we apply them on Earth.
The third and final example I'd like to review for further context is the dynamic of radiation and DNA repair. Radiation predictably activates a repair protein called poly-ADP ribose polymerase (PARP-1). Once activated by a radiation exposure event, the cellular machinery assembles PARP by scavenging residues of nicotinamide adenine dinucleotide (NAD). Up to 200 molecules of NAD are needed to build one single PARP polymer in order to initiate repair of a single strand of DNA.
As NAD is depleted via PARP activation, NAD-dependent glycolysis becomes impaired—a well-known consequence of PARP activation. Impaired glycolysis can lead to ATP depletion, unless a suitable surrogate energy source is available. NAD depletion can also spread across molecular networks, impacting more than 300 metabolic reactions. This effect may in part account for the unusual sensitivity of tissues such as the brain to radiation exposure, though it is by no means limited to the brain.
Rather than looking at PARP inhibitors in the radiation environment, we are investigating NAD precursors and agonists as countermeasure candidates. Among our NAD-directed countermeasure candidates is nicotinamide riboside, which has been shown to raise NAD in animals and humans.
These three examples represent only a small snapshot of the process we use to query molecular network complexity in an effort to develop countermeasures today. They represent targeted approaches, in which we try to understand known influences that are potentially actionable today or in the near term. Beyond this, we also examine genes and gene-linked molecular networks in an untargeted manner with a goal of better understanding the complex molecular dynamics in space and in space analog environments.
Shifting to research related to commercial space flight, we recognize that individuals will be flying on commercial space flights who have not been preselected. They are not trained, and they will likely have a variety of medical conditions not common among professional astronauts. Most of these individuals will probably fly just once. We understand that these individuals fall into a separate category from the repeat flier—the space scientists, commercial space pilots, space miners, and other types of professionals who will be entering the space environment on a regular basis.
With that introduction, I would like to explore a series of questions. I do not want to restrict our discussion only to these questions, and I am confident that we can follow up on the many interesting comments and thoughts that you will all contribute to the conversation.
My first question for each of you relates to the frequent use of molecular profiling by space medicine scientists and physicians. I'd like to begin by discussing a targeted, hypothesis-driven approach based on preselected analytes. Please also reflect on your use of untargeted, non-hypothesis-driven omics methods.
How have you used these tools to understand group and individual responses to space or space analog conditions? This question obviously applies to humans, but for those of you conducting animal research, how do you deploy these methods, and what are some of the highlights that you can share with us? Dr. Bailey, would you please begin the discussion.
Together, the 10 investigators, collectively referred to as the Twins Study, utilized omics to analyze potential health effects of space flight, specifically: genomics, transcriptomics, epigenomics, proteomics, metabolomics, and microbiomics. It is still too early to say what we are learning from these studies, but certainly changes were observed between the space flight subject and the ground subject. Most of those differences, as you would expect, occurred during the one-year mission while Scott was aboard the ISS.
The other interesting thing we are seeing at this early stage is that the space flight–specific changes observed were by and large temporary. That is, they were only associated with space flight itself, returning to their preflight levels fairly rapidly once Scott returned to Earth.
While it is difficult to say yet what the conclusions of these studies are going to be, it is certainly true that they will facilitate informative correlations between the different endpoints, thereby providing mechanistic insight. For example, if we see changes in methylation patterns that might suggest changes in gene expression, we should see evidence of that in other assays as well. Since our “n” is so small, those kinds of findings give you a lot more confidence in what you are seeing. Overall, I do think that omics is the future of science.
However, such a direction also ushers in a new era of ethical issues and concerns, which we have had to deal with from the beginning with the Twins Study. For instance, NASA now has all this very personal information on easily identifiable individuals. What are they going to do with it? I think that they are still struggling with how best to deal with that question, particularly in light of the fact that astronauts are very public figures. Currently, everything has to be approved by the twins themselves in consultation with a genetic counselor, and they do not have to agree to release any of it. That is one issue.
These types of studies also bring up issues of discrimination on a variety of levels. NASA cannot use personal “omics” information to select crew members, for example. It's not hard to imagine that other kinds of concerns will come up as we move forward in this arena, including gender differences, which also need to be considered. How exactly the data are going to be used is a very interesting question indeed.
As with a lot of the other changes, however, Scott's telomere length returned very quickly to his preflight levels upon return to Earth/gravity. Telomere lengthening seemed to be a transient, temporary effect of space flight.
I emphasize that point because it illustrates how important some of the omics studies are going to be in helping us understand how or why that happened. For example, we are interrogating the epigenomics data for evidence of changes in the methylation pattern of the promoter region of telomerase, which is the enzyme that influences and controls telomere elongation. Differences might indicate that telomerase may well have been activated during space flight. With this example, you can start to appreciate how the omics data are going to help us better understand some of the changes that we observed.
I would like to ask Dr. Thomas Goodwin to comment on our first question about the use of targeted profiling and the various omics tools used to assess changes in space and under space analog conditions. Formerly of NASA, Dr. Goodwin retired in 2016 after a long career there, but he is by no means slowing down.
What we found though were tremendous shifts in gene expression in human cells once they reached a microgravity environment. We tested this in a couple of different ways, looking at cells that were initiated in their growth sequence on the ground, flown up to the ISS via the space shuttle, and then were resident for a period of time in space.
We also wanted to be able to identify the factors related to the launch sequence that might have caused changes in gene expression. We performed another set of studies in which we transported cells in a cryogenic state to the space station, then thawed them, put them into culture, and froze or harvested them in a RNA/DNA preservative before bringing them back to Earth. In this way, we were able to isolate the space environment from any kind of transitional events that might have been going on in the course of the launch sequence.
In both cases, we found some significant changes in gene expression. For example, in kidney cells, we found that the production of erythropoietin was significantly elevated, which was a somewhat unexpected finding.
Another thing that impressed us quite a bit was the rapidity of acclimation to the microgravity environment. Cells began to change their gene expression within 5–10 hours of their arrival or the initiation of experiments in space. I think this is some indication of what is going on physiologically, where the human body immediately begins to change its perception of its surrounding environment and then begins to respond to it in a concordant way.
Why that may be important with regard to commercial space flight is that we are going to have an opportunity to do some things that have not been done previously because the circumstances for collecting data were not available to us. For example, many of the suborbital flight conditions that companies are undertaking are short term in that they last for 3.5–5.5 minutes. This will allow us to study gravitational G-force changes in both men and women that occur in a reasonably short window of time.
G forces are exerted on takeoff and then again on landing, and especially in repeat fliers such as pilots, copilots, and space scientists. So we will be able to obtain repeated samples from the same population over a longitudinal study to assess the effects of these G forces. That is going to give us data that we have never seen before. These data could be very important for understanding how the body acclimates physiologically in both men and women, which we suspect may be different, through these gravitational changes and systematic changes that reflect how the body perceives the outside world.
If someone tells me, “You have lost all of your neurons except one, but you are still functionally fine in terms of cognition,” then for the purposes of our work, you are essentially fine. We rely heavily on behavioral testing to identify functional decrements that are critical, which we think are one of the more concerning things that NASA has to consider.
We have recently uncovered some new data, and I can share some of it with you. It is related to 1-year post exposures to very low doses of helium ions. We have significant evidence using six behavioral paradigms that 5 cGy of helium ions causes persistent deficits in learning and in memory and affects mood disorders, with changes seen in multiple brain regions—the hippocampus, the amygdala, and the prefrontal and perirhinal cortexes.
We perform tests to assess episodic, recognition, and associative memory in rodents in which we analyze their capability to recognize and explore novelty. We have found impairment at 6, 15, and 52 weeks after exposure, and these changes appear to be permanent.
We also use tests to measure anxiety- and depression-like behaviors—tests that have been used by the neuroscience community for decades. Animals tend to show both increased anxiety and depression following exposure.
We have recently developed an important test that is a measure of fear extinction, which is an interesting task. Fear extinction is an active process in which an animal has to unlearn an association between two events. In this case, a conditioning stimulus (tone) is presented with aversive unconditioned stimulus (mild foot shock). The learned “fear” response can be acquired in as little as one experience and can persist for a lifetime. Imagine that a child was standing in an intersection, hears a car horn, and almost gets hit by a car at that moment. If later in life, the adolescent now repeatedly freaks at the sound of a car horn, then that would constitute an inability to extinguish that aversive association between events. Such behavior gone uncorrected could precipitate into more serious mood disorders involving various forms of phobia and obsessive–compulsive behaviors, manifestations of impaired fear extinction.
Interestingly, animals that have been irradiated cannot forget the association between a tone and a mild foot shock 1 year after the exposure. We predicted that this type of impairment in fear extinction would cause anxiety and depression, and in fact it did.
We can also look at the molecular level and interrogate the functional connectivity of neurons in discrete regions of the brain using electrophysiological techniques. For example, my collaborator Ivan Soltesz at Stanford has used a technique in which you stimulate a principal cell (neuron) in the CA1 region of the hippocampus and record from a distal principal cell electrode in the perirhinal cortex. If one tickles the cell in the CA1, >50% of the time the cells in the perirhinal cortex will fire. When we irradiate these animals, that connection probability is virtually abolished 1 year after exposure. Thus, these low doses have completely short-circuited the connectivity between the hippocampus and the perirhinal cortex. This is an astounding finding.
In addition, we find unbelievably high levels of neuroinflammation throughout the brain. If we measure certain cytokine levels, they are increased twofold compared to unirradiated animals. More surprisingly, the levels of activated microglia, which are the innate immune cells of the brain, are elevated three- to fourfold throughout the cortical and hippocampal regions of the brain after one very low dose of helium ions.
These findings build on our prior work in which, along with these functional decrements in cognition, we have seen a marked deterioration of neuronal structure at the macroscopic and microscopic levels, and these effects do not resolve.1–3 At the macroscopic level, we see remarkable changes in dendritic complexity, that is, the complexity of the neurons is reduced: dendritic length, dendritic area, and the branching patterns are all diminished and degraded. If we zoom in more at the level of the dendritic spine, which is the structural correlate of learning and memory, the density falls through the floor. It is dramatically reduced and shows no signs of recovery. We can look at the different types of dendritic spines, and they are all impacted—mostly the immature spines but also some of the more mature spines such as the mushroom type.
Furthermore, we have recent data from Dr. Dickstein at the electron microscopic level showing reduced myelination along the axons in the hilus region of the hippocampus. She has also found reduced levels of non-perforated synapses after multiple types of charged particle exposures. In all, we see structural changes as revealed by electron microscopy from the nanometer scale all the way up to the micron level with dendritic changes. If you are looking for radiation-induced changes that would be of concern to NASA, those affecting the CNS would be the big ones.
These are the types of results that we are looking at now—trying to piece together the molecular signaling that underlies such alterations. This gets very complicated, but one of the pathways that plays a critical role for NASA-related exposures to heavy ions, as well as for clinical exposures, is the retrograde endocannabinoid signaling pathway in the brain that involves CB1 receptors—the cannabinoid type 1 receptor—which is the most prevalent receptor in the brain.
The natural ligands for these CB1 receptors in the brain are 2-arachidonoylglycerol (2-AG) and anandamide. This pathway is important in setting the balance between the excitatory and inhibitory basal tone of the brain. Radiation disrupts this in a major way, and we are only now beginning to dissect precisely how that happens.
We recently published a paper with Ivan Soltesz of Stanford University looking at a microcircuit, in which we assessed the synaptic properties between specific types of interneurons and pyramidal cells in the CA1 region of the hippocampus. Interestingly, low-dose exposure to protons was found to impact synaptic plasticity of CB1-dependent microcircuits in the hippocampus selectively, acting to decrease the tonic inhibition of gamma-aminobutyric acid release, which is an inhibitory neurotransmitter. 4 We are now finding these longer-range disruptions in the setting of network connectivity.
To be honest, if you had asked me if this is what we would find when we got into this 5 years ago, I would have laughed. I am astounded that we see these types of changes in the brain that persist for so long after such low doses. I never would have been able to predict this.
The role that neurogenesis may play is more questionable at space radiation doses. Amelia Eisch at University of Pennsylvania, Philadelphia, has produced very interesting data showing some age-dependent differences in how neurogenesis impacts pattern separation in animal models. In our prior NASA program project grant, we have had more difficulty establishing a conclusive impact of neurogenesis on functional cognitive decrements in the brain. I am of the opinion that because neurogenesis adds relatively few cells to the brain, the more important impact is on the more mature neurons that show structural impairments after radiation exposure.
I do not mean to imply that radiation-induced changes in neurogenesis are not going to play some role. For example, endocannabinoid blockade using the CB1 antagonist, reverse agonist AM-251, have yielded preliminary data to suggest that this can preserve neurogenesis and cognition. Other treatment approaches that I have seen to protect the stem-cell population from radiation have had mixed results. I believe that preserving a relatively small population in the brain is certainly important, but I do not think it needs to be the overriding concern against a backdrop of 95% post-mitotic cells, at least in terms of neurons.
Having said that, I would interject that interventions targeting the astroglial compartment may be of interest. In fact, we have data showing that activation of microglia over the short term is good, but chronic activation is bad. Lots of data exist to support that in a variety of CNS injury models.
If we eliminate the microglia after exposure to space radiation, then the animals get smarter. The simplest way to achieve this is by a microglial depletion strategy involving the inhibition of colony stimulating factor 1 signaling. We add a certain compound to the animals' diet, it eliminates the microglia, and the animals get smarter. 6 The rationale is that neuroinflammation is being attenuated.
Ultimately, interventions are not likely to be that simple, but you asked what kinds of things we are working on, and those include strategies aimed at attenuating neuroinflammation and those targeted to the endocannabinoid pathway. These at least represent starting points for developing relatively safe therapeutics for astronauts.
It all depends on total dose. If you get a total dose of 40 cGy on a roundtrip to Mars, 7 there is a chance that you will start to develop some of these impairments, perhaps during the mission or soon after you come back. There would be no risk for commercial flights, where you would be going up and back within the low Earth orbit. For someone who is going up in space for fun, and just going in and out for a few minutes, there are a million other things to worry about that pose a greater risk than radiation.
The questions being discussed include, for example, what are the probable combinations of different types of exposures in space? What types of disclosures to space flight participants will be needed? When the time comes to make a disclosure to a commercial space flight participant, each individual participant will have to determine what risk he or she is willing to take. Those disclosures by flight providers will need to be appropriately made in terms of historical adverse events and also in terms of what we know about exposure intensity, duration, and other parameters in the different space conditions. The conditions will be inherently variable, whether it be a suborbital flight, a Bigelow orbiting habitat, an ISS visit, a lunar orbit, or some other mission dynamic.
In addition, from the very beginning of ISS, we started collecting measures of oxidative stress, as part of the crew's overall medical assessment of health. These included measures of lipid peroxidation products such as the aldehydes 4-hydroxy-trans-2-nonenal and malondialdehyde, levels of glutathione and superoxide dismutase (SOD), and measures of total antioxidant capacity. In addition to the biodosimetry were measures of the DNA adducts. These tests assessed both the byproducts of oxidative stress and the potential oxidative stress defense mechanism changes in the crew.
As has been discussed here, with the relatively low doses of radiation exposure in low Earth orbit, these indicators have not been particularly sensitive in trying to characterize the level of oxidative stress. Yes, we can see increases in lipid peroxidation products, and we can see decreases in SOD, total antioxidant capacity, and levels of glutathione. These are not, however, changes that we would describe as being clinically significant.
In terms of measuring biodosimetry—looking at chromosomal aberrations and radiation—from a research perspective, a lot of the literature has focused on single-strand breaks, double-strand breaks, translocations, aneuploidy, and so on.
DNA adducts have also been looked at, as mentioned before, and in fact, we have been measuring DNA adducts in the crew for quite a while now, looking at (8OHdG) 8-hydroxydeoxyguanosine. We have some limited data to suggest that the radiation exposure in low Earth orbit is being manifest in terms of elevated oxidative stress and the formation of DNA adducts, and we can measure a biological response in terms of a biodosimetry profile based on increases in chromosomal aberrations. All of these changes are relatively small in magnitude in low Earth orbit compared to what we expect to see outside the geomagnetosphere, as Dr. Limoli has pointed out.
What can we do in terms of the studies we perform in low Earth orbit to improve their sensitivity and to see if potential countermeasures could be effective? Obviously, having countermeasures for the radiation insult that we expect to affect both the nervous system and all the other cells of the body—especially stem cells, for example in the bone marrow and the gastrointestinal tract—we mostly want to protect the stem cells. What can we do to achieve that?
We want to have a countermeasure that has been validated in space flight, and the only place right now to do that is in low Earth orbit. We need to identify more sensitive measures to allow us to understand the effect of countermeasures if we want to test them in low Earth orbit.
We can look at things such as histone gamma-H2AX and ataxia teleangiectasia mutated, as there are a number of areas in which the expression of these types of molecules may be more sensitive and specific indicators of radiation injury. We could also assess various “-omics” profiles, such as proteomic changes, because we know that lipid peroxidation may affect the post-translational modification of proteins, for example. Certainly, if lipid peroxidation occurs, it changes membrane stiffness, and it can also produce some DNA damage—the adducts we talked about.
Genomic and transcriptomic profiles could reveal patterns that are specific for radiation injury and might be more sensitive indicators. We could also look at metabolomics, which may be something that we could assess using a urine sample, for example, to make it less invasive and to improve sample availability for crew members. We could easily take measurements in flight and post flight.
We might be able to develop a radiation “-omics” profile that would serve as a more sensitive indicator for radiation exposure and related damage due to space flight in the low Earth orbiting environment. We could later translate that for use in measuring and monitoring omics profiles during exploration-type missions outside the geomagnetosphere. We could also use it to determine the efficacy of countermeasures in low Earth orbit and to select which countermeasure might be employed on exploration-class missions.
We have had a proposal in place for several years to collect these kinds of data as part of a research study in conjunction with colleagues at the Department of Defense who have similar interests when it comes to troops and sailors who are on nuclear submarines and nuclear aircraft carriers and troops who might be exposed to weapons in the dirty bomb arena. We are also working with the Department of Homeland Security and the Department of Energy, which deal with radiation workers and the potential for a radiation accident such as what happened in Japan and at Chernobyl, for example.
There is a lot of interest in developing better monitoring tools for these types of military personnel and being able to develop countermeasures that might be effective in these arenas. Using space as a platform for doing this has proved to be challenging to some degree, so I am not sure that NASA is convinced that this type of strategy should be applied across the board, especially with the ethical considerations that Susan raised earlier.
This is also where monitoring telomere length longitudinally also fits into the picture because telomeres also respond to the stress induced by radiation exposure. These examples are very much along the same lines as you are talking about, and hopefully we will be able to expand on some of this. As you point out though, I do not know how interested NASA really is in doing these kinds of things. They have pretty much shut down their biodosimetry chromosome aberration studies.
I think there is a lack of awareness at least within the HRP community that this type of scientific assessment would be of value in monitoring crew members who are exposed to extended periods of microgravity. The question remains whether the main concern is microgravity, radiation, or a combination of the two. Are there synergistic effects?
One of the patterns we find in elite performers is that of micronutrient deficits, coupled with variant single nucleotide polymorphism (SNP) profiles. As I noted in my introduction, these elements of gene variants, micronutrient deficits, and environmental conditions may converge to influence the phenotype in ways more complex and more widely distributed than if a single variable is altered.
One of the common patterns we see involves altered magnesium status. I have chosen this example because we have been talking about DNA damage and radiation, and because magnesium is also relevant to other areas such as musculoskeletal function. Magnesium has an important role in stabilizing the structure of DNA and is essential for the activity of DNA repair enzymes. First, the three negatively charged phosphate moieties on ATP are stabilized by the +2 charge of magnesium and one water molecule. Second, magnesium is a central atom in ATP synthase. Third, magnesium is involved in virtually all DNA repair processes via the requirement of DNA repair enzymes for ATP, but also because most DNA repair enzymes require magnesium as a structural or functional element.
Deficits in cellular magnesium are common findings in our elite athlete population and in some populations entering extreme environments. Recently, we have been reviewing the molecular and physiologic phenotype findings from our Mayo Clinic Study on Mt. Kilimanjaro. Magnesium deficit is probably one of the more repeatable findings in our elite cohorts who are typically training in extreme environments.
One area of focus for us is attending to the status of DNA stability and DNA repair before individuals even enter the space environment. We have been asking questions about what influence SNPs, insertions and deletions, and other DNA alterations have on DNA stability and repair. Further, how does micronutrient status come to bear on functional changes and the altered functional capability of DNA repair enzymes before entering space, during a space mission, and upon return from space?
Astronauts spend 2.5 hours of their 8-hour duty day in activities assigned to exercise. This includes time for hygiene, setup, and data collection, so they are not exercising all that time. I mention this though to point out that it is a very mature countermeasure and is something that is actionable today. We can do studies of 6- and 12-month duration, and we are at a stage where we are optimizing these countermeasures.
Over the past 10 years, we have made dramatic advances not only in the exercise hardware, making it more reliable, allowing for a greater variety of exercises, and allowing for an intensity of exercise that researchers expect to be effective for all of the body systems, but also in seeing some research come to fruition and some operational changes in how exercise is prescribed. We are now starting to realize better outcomes with respect to things such as bone mineral density, muscular strength and endurance, and aerobic capacity. The behavioral health counterparts consider exercise as their most potent psychological and behavioral countermeasure, and so it really cuts over a lot of these disciplines.
We have spent several years looking through all of the exercise data in terms of matching what the astronauts do on a day-to-day basis with outcomes, and initially focusing simply on effectiveness. Not surprisingly, as we have improved the exercise hardware and refined the prescriptions, on average we have seen stepwise improvements in the effectiveness of this countermeasure. We are starting to hit the point where for 6-month missions, we are having a hard time detecting decreases in fitness. In other words, the exercise programs are becoming increasingly more effective as you look at the group averages.
What we do see—and it relates to what the group has been talking about—is a lot of individual variability. After we had stopped patting ourselves on the back about the group means looking really great, we started looking at individual variability among the astronauts, and we observed that some actually had improvements in their bone mineral density, their muscle strength, and their cardiovascular endurance with space flight. Others held their own, and others experienced fairly significant losses.
The astronauts were all following approximately the same program. Yet, you can imagine that there were individual differences because they have quite a bit of free choice in what they do. They have some free choice in how much they exercise, what equipment they use, how they experience the stress, how much and what they eat, how well they are sleeping, how they respond to the radiation, how they respond psychologically, and so on. It was no doubt a very poorly controlled study, but the best we can in the operational space flight environment.
We have also had the opportunity to conduct some 70-day bed-rest studies, which we consider to be long duration, in order to evaluate the same exercise program. In the bed-rest studies, we took care to control what I would call the usual suspects believed to contribute to the variability in the group. For example, the diet was very precisely controlled. In fact, the subjects all ate the same foods on the same day with the calories calculated for their actual metabolic cost, not estimated. If subjects did not eat all of the food, it was returned to them in hourly increments until they did eat it all. The light/dark cycles were controlled. The level of stress was roughly controlled inasmuch as they were exposed to the same staff and conditions and they did not have operational work.
I could go on and on describing all of the controls, but suffice it to say that it was as well-controlled a human study as I could imagine anyone being able to do, and we still saw tremendous variability in all of our measures of cardiovascular, muscle, and bone health, as well as performance. The control subjects, who had no countermeasure whatsoever, had a range in which the best responding control subjects fared better than the worst exercise countermeasure subjects, so there was even group overlap. The factors that commonly account for group variability, such as differences in the diet or in sleep versus awake time, were very well controlled. Yet, the individual variability remained, and it was quite dramatic.
We wanted to understand this because we have optimized these exercise countermeasures on average to protect the group. In reality, however, it appears that we will need to recommend exercise programs to protect individuals. We now realize the potential for variability, even when you very precisely control the exercise, frequency, and dose, as in the bed-rest study, where every session was supervised by the same set of trainers every time. The force of every footfall was recorded on the treadmill. I mean to emphasize the study was exquisitely well controlled, and we can say definitively that the participants were doing the same intensity of exercise. Yet, the response to exercise training was very different, with some people responding better and others worse.
Further complicating the situation, we did not identify overall responders or non-responders. If an individual scored very well on measures of bone mineral density, it did not necessarily mean that he or she also fared very well on muscle measures. If they did well in one muscle group, that result did not necessarily carry over to another muscle group. This was really quite confusing to us.
This is definitely an area that requires further research, and I would even recommend that it be a future area of emphasis. We may need more of the precision that an individualized exercise prescription can provide, and going forward, for very long-duration missions, we will need to understand how to monitor and customize the countermeasure on an individualized basis. Another important consideration will be what to expect and do in the event that a countermeasure cannot be completed because the person sustains an injury, the equipment fails, there is not enough power for some period of time, or whatever reason may arise. Customization will be our next big leap in the development of exercise countermeasures.
In general, the exercise world seems relatively behind in omics research. A few groups are doing some clever work with elite athletes in predicting injury susceptibility. In terms of taking this to the next practical level, however, we have some technological hurdles to accomplish, and that is where the future lies for this operational countermeasure work.
If we go back to neuroplasticity in our earlier discussion on radiation susceptibility, one of the areas that we have been working on extensively is the BDNF molecular network. We presented this work in the NASA HRP meeting a few years ago. One of the interesting things about BDNF concerns its role as the main activity-dependent neurotrophin in the human brain, at least as far as we understand it today. Activity dependence can relate to physical activity of course, but it can also refer to cognitive activity.
Physical activity as a driver of BDNF is really quite crucial, and it may be one of the strongest means by which physical activity influences mood and cognition.
Preclinical studies have examined BDNF expression, as well as looking at things such as cytokine expression with exercise as a prophylactic in these different radiation exposure paradigms. Some of these studies used radiation doses that were much higher than what Charlie Limoli usually uses. I think the lowest in a specific BDNF radiation study was 2 Gy, increasing to 30 Gy. The control group, with no exercise and with each dose of radiation exposure, had a substantial reduction in BDNF and TrkB expression—the primary BDNF receptor—as well as in the expression of the transcription factor cAMP-response element binding protein. They also showed reductions in dendritic branching, synaptic spine density, and the kinds of things that Charlie measures, though using less sophisticated methods than those Charlie is using today.
Interestingly, with the exercise as a prophylactic, the treatment group showed reduced adverse effects on brain morphological changes and increases in tissue BDNF. While, the concept of exercise as a countermeasure for muscle and bone effects is crucial, it appears to also be important for brain effects, especially in terms of mood and cognition. Mechanistically, exercise may be one of the crucial drivers for maintaining neuroprotection, at least in part through its impact on BDNF gene expression.
If we then couple this with the radiation-associated PARP activation phenomenon that depletes NAD, and possibly the need for NAD precursors—not PARP inhibitors but NAD precursors or modulators—it further illuminates the idea of convergent variables. Our approach to these molecular networks is to attempt to understand better the multiple converging variables and how they may lend themselves to countermeasures.
As we have touched on in our omics discussion, a reliable way to describe these dynamics will increasingly involve some combination of targeted and untargeted analyses applied to various space and space analogue paradigms.
Lori, using what you have described as these different types of responders, have you examined how we might develop machine learning–based predictive models, which we may then be able to apply to new individuals? We could conceivably identify their molecular, physiological, and performance parameters, plug those into the predictive model, and be able to say, “You may be more like one group versus another group,” or “You may need one type of countermeasure versus another type.” That is the dream that I think we are all hoping to realize, but we will need the molecular, physiological, and phenotypic data to get to that point.
Other work I would like to mention is that of Ruth Globus at NASA Ames, who has some data to suggest that mechanical loading in the form of exercise is radio protective for bone. Again, this lends support to a multi-organ system effect of exercise. We have so much maturity in terms of studying exercise countermeasures that trying to leverage as much of that knowledge as possible and to understand what it can tell us about these other systems seems efficient.
In 1998, male rats were sent on NASA's STS-90 NeuroLab mission. Myostatin gene expression increased significantly after 17 days based on myostatin mRNA levels in muscle. This was associated with muscle loss. Myostatin is known to inhibit muscle protein synthesis. Similar findings were reported in a more recent limb unloading experiment. In the latter study, comparing loaded to unloaded limbs, a 35% increase in myostatin gene expression in the unloaded limb was observed.
This would support extensive whole-body exercise over any kind of an isolation exercise. Lori, I would like to get your thoughts on these findings for myostatin and what they might mean for physiology in general.
Regarding the other part of your question, it probably will not surprise you at all that I am a strong advocate for whole-body exercise as a countermeasure for all of the systems, and more highly targeted exercises for some than for others. The side effects of exercise, apart from a rare orthopedic injury, are all very positive.
We saw a lot of forearm fatigue in crew members who we studied in the extravehicular activity (EVA) setting Those crew members were especially subject to fatigue when using a handgrip for a potential EVA maneuver, and we saw some improvement when using a countermeasure against oxidative stress. This led to interest in the potential to use metabolomics to identify individuals who might be at risk and to determine specific molecular targets and what countermeasures might be most effective. We would look at which individuals had particular SNPs that would make them more vulnerable to fatigue and possibly more responsive to fatigue reduction protocols, and be able to personalize their exercise prescription in that way.
Finally, regarding the whole issue about PARP activation, we in the oncologic community are now using PARP inhibitors to reduce the growth of neoplastic cells. Could it be possible to allow for the survival of transformed cells if we are able to target countermeasures in that way?
Regarding the question of what to do if there were a failure of systems that did not allow people to exercise with the same frequency or intensity, I would suggest relying on some of the pharmacological countermeasures in those circumstances.
In addition to addressing PARP in circumstances that may involve transformed cells, we are also working with it in a couple of other areas. We have a study underway with the Mayo Clinic involving collegiate football players, looking at PARP activation in traumatic brain injury (TBI). There is an observed depletion of NAD and a disruption of glycolysis in TBI, and we are looking at what effect an NAD modulator in the form of nicotinamide riboside has on recovery from TBI. We are also conducting a study with the US military. All participants are engaged in a program that has a historically high and predictable rate of concussion. Our question is whether an NAD modulator favorably influences the trajectory of brain acceleration injury.
Addressing the transformed cell question, it is important to consider the nature of the injury and the known or anticipated rate or presence of transformed cells. Then, one needs to decide whether simply to inhibit PARP or to allow PARP to proceed with DNA repair while modulating the NAD depletion concomitant with the repair process. The answer to that question will have to be worked out over time.
However, I do believe this approach may be important to consider in protection against space flight radiation. However, I would also like to take a look at these circumstances in terms of muscle, because like reduced NAD in the brain, one must suspect that reduced NAD due to radiation may simultaneously be occurring in muscle. Unique to muscle is the fact that physical activity drives NAD synthesis via upregulation of nicotinamide phosphoribosyltransferase (NAMPT). In this regard, sustained whole-body physical activity in space may be one means to counteract the effect of radiation-induced PARP activation (and coincident NAD depletion).
Thus, one might use exercise as one method to build NAD levels via NAMPT activation and couple that with a second method to build NAD levels, using the NAD precursor nicotinamide riboside. This converging variable approach may add further confidence that NAD levels are being managed in the radiation environment that drives PARP activation. These hypotheses of course must be tested in space. But they exemplify the value in developing a refined understanding of molecular networks in the formulation of space flight countermeasures.
Someone mentioned the work of Ruth Globus, and I would like to add a bit to that because I have published extensively with her. Some of the countermeasures regarding weight, exercise, and so on certainly offset some of the effects of radiation, but to what extent remains to be seen.
One of factors to consider is that radiation impairs trabecular bone density and micro-CT structure by stimulating osteoclastogenesis and inhibiting osteoblastogenesis.8–10 This is not my work, but I am sure that exercise will have a beneficial impact on that.
Another side topic that I thought we would talk about in more detail is the differences between the sexes. We have some new data, and I also know that Cindy Lemere at Brigham and Women's Hospital has data showing that females are more resistant to the effects of cognitive dysfunction caused by space radiation exposure. This is a fascinating topic that we have not hit on regarding radiation, and our data suggest that if we are going to go to Mars, we need to send women! However, I must admit that from the carcinogenic standpoint, females have also been shown to be more prone to radiogenic cancers, as Susan points out—so again there are simply no easy answers in which we can overly generalize.
What is the value of altering DNA repair? We know that radiation impacts DNA repair, and we know that the repair process is differentially impacted by heavy ions. Suppose you have perfect repair in space. How much is that going to mitigate a risk for cancer? I do not know. People may disagree, but I do not think that worrying about transformation events in space is absolutely critical at this point in time.
Basically, it talks about the fact that there is some legislation on the state level regarding liability for private manned space flight, but none on the federal level. I think Virginia was enacting legislation about this as early as 2007, California did so in 2012, and a few other states did so in between.
They are basically doing it to attract business, but it is questionable as to how this would unfold in a real-life scenario. It is very questionable as into what direction we are headed, but looking at the big picture, I do not believe we have much time left. It is hard to predict exactly how this will unfold on a federal level, if at all, moving forward. Legal scholars are becoming more aware of this area and talking about it.