Circuits and Biomarkers of the Central Nervous System Relating to Astronaut Performance: Summary Report for a NASA-Sponsored Technical Interchange Meeting

Biomarkers, ranging from molecules to behavior, can be used to identify thresholds beyond which performance of mission tasks may be compromised and could potentially trigger the activation of countermeasures. Identification of homologous brain regions and/or neural circuits related to operational performance may allow for translational studies between species. Three discussion groups were directed to use operationally relevant performance tasks as a driver when identifying biomarkers and brain regions or circuits for selected constructs. Here we summarize small-group discussions in tables of circuits and biomarkers categorized by (a) sensorimotor, (b) behavioral medicine and (c) integrated approaches (e.g., physiological responses). In total, hundreds of biomarkers have been identified and are summarized herein by the respective group leads. We hope the meeting proceedings become a rich resource for NASA’s Human Research Program (HRP) and the community of researchers.


Introduction
Astronauts on long-duration space missions (e.g., transits to Mars) will experience the combined, potentially synergistic, impacts of simultaneous exposures to spaceflight hazards that affect the central nervous system (CNS) and operationally relevant behavior and performance [1]. While individual spaceflight hazards are often individually well quantified, in long-duration spaceflight, astronauts will experience multiple hazards simultaneously [2,3].

Sensorimotor Influences on Operational Performance (Leads: S. Rosi, M. Shelhamer)
The goal of Group 1 was to create lists of biomarkers and brain regions and/or neural circuits related to operational performance for constructs that are prioritized in HRP's sensorimotor risk. Group 1 assessed the following 13 key constructs in Table 1: visual function, spatial orientation, vestibular, proprioception, hearing, motion sickness, smell and taste, postural control and balance, locomotion, fine motor control, perception, gaze, and pain. Note that the panel assessed translatability based on the existence of rodent models and did not suggest using non-human primates (NHPs), nor did they identify a construct that should be tested in NHPs.

Summary of Discussions
During discussion of each of the 13 constructs, 10 themes emerged. Although identification of themes was outside the scope of the panel, these themes were applicable to nearly all constructs discussed and, therefore, we define them here: 1. Connections between constructs. Distinctions between the constructs are, in many cases, artificial. Although segregated disciplinary expertise has achieved a great deal in the sensorimotor domain, the different constructs are so closely interconnected that it is hard to discuss them separately in a way that is true to the science and to the operational implications. As an example, vestibular function, gaze control, balance, and locomotion are very closely related, and yet they are often addressed as specific and separable. Another example is perception. Almost all sensorimotor constructs involve perception in some way; vestibular perception-perception of the upright-affects the ability to balance. Perception of upright is influenced by changes that occur in microgravity, which is a vestibular effect. Again, these specific constructs become tightly entangled and it is difficult to separate them in terms of biomarkers and operational relevance.
2. Many spaceflight stressors and sensorimotor effects occur simultaneously with different time courses. Not only do the different constructs interact, they do so with different time courses. The most overt and acute forms of vestibular adaptation (related to space motion sickness) occur over the course of a few days, whereas other vestibular-mediated functions (e.g., the sense of being truly comfortable with the three-dimensional aspects of motion in a weightless environment) develop over several weeks. Some adaptive sensorimotor changes in space occur with similar time courses as those seen in analogous environments on the ground. For example, the changing contributions of vestibular, proprioceptive, and efference copy information during recovery from labyrinthectomy in an animal model [9] have time courses that mimic recovery of motor control during locomotion after spaceflight [10]. Similarly, ground-based studies in animals show that development of efference copy over several weeks mimics the time course of the development of three-dimensional spatial sense in astronauts over the same time period. The similar time courses suggest that these may be aspects of the same underlying process. This might provide translational opportunities from ground-based animal models and may inform a process for preadaptation paradigms for spaceflight.
3. Multi-sensory integration. This is related to the theme of interacting constructs. Most sensorimotor behaviors and perceptions arise from the simultaneous activation of multiple sensory systems. An obvious example is the combination of visual and vestibular information for gaze control (vestibulo-ocular reflex (VOR)). Another is the prevalence of proprioceptive and kinesthetic influences, in addition to vestibular and visual influences, on posture and locomotion. 4. Stress. Spaceflight involves multiple simultaneous stressors-physiological, psychological, and environmental. These have widespread and sometimes unknown influences on sensorimotor function, and likely on the ability to adaptively alter sensorimotor function. The effects of stress on motor learning and on motion sickness are two examples: stress affects motor learning, which alters adaptation, which can change the ability to recover from motion sickness, which can increase stress. 5. Learning. Almost all the individual constructs exhibit adaptive behaviors to spaceflight and these adaptive behaviors may complicate the usefulness of the constructs as biomarkers because the response that is being assessed will change with adaptation to spaceflight. Of course, such adaptation is desirable and should be promoted, but it complicates the use of a biomarker to identify increased risk to astronaut health and performance. This would be especially true in missions of extended duration where the adaptive processes might not be understood. A specific biomarker for learning and adaptation would be desirable. 6. Some constructs might be easily measured but lack relevance. As an example, the angular VOR has been extensively studied and is easy to measure, but little or no evidence exists that it changes significantly due to spaceflight, or that any changes have an operational impact.
7. Neural circuits. Interpretation of neural circuitry is not always straightforward. There is not always a direct analogy between animals (where many circuits have been delineated) and humans; the neural circuitry is different in some cases, and there are also adaptive changes that make the definition of standard circuits difficult. Circuit function is implicitly assessed with behavioral measures, so knowledge of some circuit characteristics such as neurotransmitters and common pathways might aid in the interpretation of behavioral markers.
8. Vestibular Cognition. The relationship between cognition and the vestibular system, and the vestibular effects on cognition, is operationally relevant and directly connects cognition and sensorimotor functions. This connection is seen in many patients with vestibular problems. No specific construct exists for this, and it is difficult to conceive of a specific biomarker.
Overall, the sensorimotor issues of multi-sensory/multi-effector interactions and learning, and their relation to stress, are not yet sufficiently studied, and they likely greatly influence human performance in space. These do not yet lend themselves to direct biomarker identification.

Recommendations
The panel evaluated each specific construct to determine if a good biomarker exists that is operationally relevant for astronauts, and that translates from animal models. The panel also commented on gaps in each construct that would need to be filled to produce an effective biomarker.
1. Visual function is easily measured (acuity, visual fields, etc.), and these measures may help to parse out visual effects from motor effects when there is a functional deficit. Retinal remodeling can be assessed with optical coherence tomography (in flight), and is hence a biomarker. Translatability is clear because many of these aspects can be tested in rodents (e.g., visual acuity in mice and even real-time visual tracking). This is clearly a useful biomarker.
2. Spatial orientation is extremely important. The panel extensively discussed grid cells-the cells in the entorhinal cortex that underlie spatial orientation. The firing of grid cells provides information that can be used to assess spatial orientation as it adapts to alterations in gravity, which is further substantiated as a potential biomarker due to its translational potential as grid cells are present and accessible in rodents. Thus, neural circuits in the hippocampus and medial entorhinal cortex are important. 3. A great deal of information exists on vestibular function in spaceflight. Basic vestibular function is not significantly altered in the microgravity environment of space, although central processing and higher-level derived functions (e.g., spatial orientation, tilttranslation perception) often are. It is, however, important to consider vestibular changes in the context of the integrated spaceflight stressors. So, as noted, the VOR changes little in weightlessness, but it would be useful to assess VOR in the context of other stressors (e.g., radiation, fatigue, etc.); for example, what is the combined impact of multiple stressors?
These aspects need to be elucidated, which can be accomplished through rodent studies (e.g., the narrow balance beam as a viable animal assessment). Taken together, vestibular change (e.g., VOR or balance beam performance) is a suitable biomarker. 4. Proprioception was identified as one of the most strongly interconnected constructs, exhibiting significant overlap with several other constructs. Little is known about the effects of (CBS risks) radiation or other stressors on the peripheral nervous system and, consequently, proprioception (this is a gap in knowledge). A rodent model would provide translational opportunities, as proprioception can be measured in that model (e.g., tape removal test, whisker test). Hence, measures of proprioception are suitable biomarkers. 5. Hearing loss is often a factor associated with spaceflight, perhaps due in part to fluid shifts, and hearing assessment in flight may help to parse out the effect of the fluid shift from noise-induced loss. However, the panel noted that these data are not particularly operationally relevant: hearing loss has not been a functional problem. As such, hearing loss is not a priority biomarker. 6. Motion sickness is a known problem that needs to be further assessed because it can have serious operational impacts [11,12], especially when first experiencing a gravity field after extended weightlessness. Motion sickness susceptibility is still unpredictable. This line of work might be revisited with more recent knowledge on learning and adaptation or might be investigated in relation to the impact on specific operational tasks. We do not know how motion sickness induces stress and how stress feeds back to motion sickness and the overall well-being of astronauts. The interaction of motion sickness, sopite, stress, and crew performance has been studied in other contexts. This work should be reviewed; however, it may still be valuable to investigate these effects in the specific context of spaceflight, with its multiple simultaneous stressors and unique demands. Again, there are several overlapping biomarkers. A drawback in this area is translatability, because it is very difficult to measure motion sickness in rodents. This is a useful biomarker, albeit with some uncertainties as to translational aspects.
7. Smell and taste are particularly important for humans as social creatures and are also clearly important in space. These constructs overlap with the well-being and operational performance of astronauts. Smell and olfaction can be markers for neurodegeneration. Loss of olfaction (anosmia) is an early marker in COVID-19 and Alzheimer's disease, as examples, and is therefore a biomarker for neurodegeneration that can also easily be tested in rodents. This biomarker is rated highly.
8. Posture and balance are important operational issues. They are problematic as biomarkers because, again, their functions cannot be isolated to discrete neural circuits due to the overlap of several circuits for multi-sensory integration and motor control. Rodent models are somewhat problematic because of the difference between neural circuits and functions in organisms with four legs (rodents) relative to two legs (humans). 9. As with posture and balance, locomotion is operationally relevant and important, but good rodent models in spaceflight or microgravity environments are lacking. It might be useful to consider static/dynamic balance control as opposed to posture/locomotion. 10. Fine motor control is difficult to assess because of the large number of confounders. Related factors that can alter fine motor function include changes in proprioception, handeye coordination, and others. Although functionally important, it may not be particularly relevant for operational control tasks, and suitable rodent models are lacking. The many confounders alone make this problematic as a discrete biomarker.
11. Perception is in fact a component of almost all the other constructs because it can include spatial orientation, depth perception, vestibular orientation, time perception, and others. Understanding of this construct is important and would address many of the other constructs, but there are many overlaps. Proprioception may be altered and is a critical issue on its own, but it will be most important to address in the context of other stressors. Specific aspects of perception have been noted in spaceflight and can have operational impacts, and so it would also be beneficial to consider perception in this performance context. Nevertheless, parsing out perceptual effects per se remains difficult. Thus, this was not considered to be a good biomarker.
12. The panel did not rate gaze and pain highly as biomarkers. Gaze largely overlaps vestibular function (and has been studied almost as much), so gaze control can be subsumed under vestibular function. Pain per se is not a good biomarker because of confounders between the perception and the sensation of pain. Nociception can depend on sex and other individual factors. Although biomarkers of inflammation exist, these are associated with pain. Hence, pain itself is not a discrete biomarker.  [24], c-FOS, vestibular hair cells [25], cerebellar nodulus of adult rats [26][27][28], TEM of synaptic ribbons [29][30][31][32][33] Nausea related-cardiac sensitivity to baroreceptor reflex; raised Heart rate; raised cortisol; reduced dominant power on EGG baseline, questionnaire [34,35], Serum: NSE and S100β [36], Otolin-1 [37]. vibration-induced nystagmus [38] (1) Effects of stress on vestibular compensation and adaptation.
(2) Social stress, performance anxiety, other psychological stress-will it impede recovery? (3) Stress impedes motor learning in mice (Fragile X mice).  Isometric pinch grip force between the thumb and index finger [51] (1) Proprioception can be connected to the fine motor control.
(2) Animals have fine motor control, but we need to standardize and develop a model (1) Need to focus on peripheral nervous system and include and utilize blood markers.
(2) Individual pain tolerance is variable The goal of Group 2 was to create lists of biomarkers and brain regions and/or neural circuits that are related to operational performance for constructs that are prioritized in the HRP's Behavioral Medicine (BMed) risk. Group 2 assessed the following key constructs which are summarized below and in Table 2: memory, attention and dual tasking, executive function, working memory, learning and plasticity, social processes, individual behavioral states, arousal and regulatory, emotional regulation, risk taking/tolerance, and stress.

Summary of Discussions
Many of the themes that arose during this panel's discussion were also discussed by the sensorimotor group (Group 1), including learning and plasticity for assessing an astronaut's general level of adaptability. The panel also discussed the importance of studying individual differences in these different behaviors, in addition to various modifying factors, such as sex, age, the impact of stress, and immune status. The panel also highlighted the importance of general biomarkers that are not specific to any construct, behavior, or tissue, but could provide a more accurate reflection of overall behavioral health.
Behavior is a biomarker. One major theme that emerged from the discussion was the fact that behavior is an important biomarker. Although biomarkers and brain regions and neural circuits are important for understanding the biological basis of changes in operational performance, the behavior itself needs to be studied as an indicator of changes in operational performance. Variations in behavior, such as increases in variability of response and instability in performance, are often the most sensitive indicators of degradation of operational performance [69,70]. Furthermore, marked inter-individual differences exist in these domains, some of which appear to be phenotypic [70,71]. However, limited knowledge exists regarding the biological basis of these individual differences and how they are modulated by spaceflight stressors. For several constructs, the panel noted specific behavioral changes that should be considered as biomarkers and gave examples of potential neuroimaging modalities that could be used to investigate underlying brain regions and neural circuits. More studies of human behavior in spaceflight are needed. Behavioral tests with greater ethological relevance to animal models would most likely yield better translation of findings to human operational performance. The panel discussed similarities between attention tasks and dual tasking; performance instability, increases in the variability of responding, and increased impulsivity are all behavior markers indicating a problem [70,72,73]. These changes can be subtle, which highlights the importance of knowing the organism's baseline performance for a task, so that changes to that baseline will then indicate a problem. Finally, behavioral biomarkers can be used to determine when an organism-from rodents to humans-is unable to use new information in the environment to adapt their behavior; these results have been obtained primarily from reversal learning and extinction tasks that are highlighted under General Brain Plasticity below.
Common measurements for studying brain biomarkers. Various neuroimaging modalities were discussed for most of the constructs, and because the panel focused on measures that could be assessed during spaceflight and across species, electroencephalogram (EEG) and event-related potentials were regarded as valuable for identifying markers associated with several constructs, including memory, working memory, attention, dual tasking, and learning and plasticity. The use of whole-brain and region-specific EEGs were both considered useful, with whole-brain EEG being particularly important for learning and plasticity [74,75]. Region-specific EEGs were regarded as most useful when coupled with a behavioral task dependent on that region, such as frontal cortex activity and attention or performance on an adaptive N-back test to assess working memory. Near-infrared spectroscopy (NIRS) and functional NIRS were also regarded as useful for assessing underlying neural targets during task performance during spaceflight.
Magnetic resonance electroencephalography and other frameworks for integrating multiple imaging modalities should also be investigated, such as joint imaging markers from simultaneous magnetic resonance imaging (MRI) and EEG (e.g., temporal volume, cor-tical thickness) that are associated with cognitive status in healthy individuals, pathophysiological changes in neurodegenerative diseases, and after traumatic brain injury [76][77][78][79][80][81]. The panel contended that these simultaneous recordings could provide a more accurate diagnosis of pathology than either modality alone.
Overlapping markers among constructs. The panel agreed that many biomarkers overlap among the constructs, such as the gastrointestinal (GI) microbiome, immune markers, and the influence of steroid hormones. As such, these markers could be general markers of behavioral health. For translational studies, most of these markers can be measured in animal models and have supporting preclinical evidence to demonstrate their relevance to human CNS function and disease.

•
Oxidative stress markers. The panel considered transthyretin (TTR) as a biomarker of neuronal stress that could be useful for assessing general CNS health, irrespective of a specific BMed construct. Although TTR is possibly inaccessible for spaceflight (e.g., choroid plexus TTR, lumbar puncture for cerebrospinal fluid), recent work suggests serum levels could be indicative of CNS pathology [82]. • Microbiome. The GI microbiome is connected to the brain through the gut-brain axis and the panel regarded this as an important system to assess potential biomarkers indicative of CNS pathology. Recent research demonstrates a vital role of the GI microbiome in CNS pathology and psychiatric disorders [83][84][85] and the microbiome has important implications for health during long-duration spaceflight [86,87].
Incorporate modifying factors into biomarker studies. The panel discussed additional factors important for spaceflight, and differences in many of the BMed constructs that were not included on the worksheet, such as sex, age, stress, immune status, steroid hormone levels, and prior experiences. The panel noted that any findings regarding the usefulness of the various biomarkers should also include tests of these biomarkers under these additional conditions to determine if the markers were relevant when these other factors are included. For example, a biomarker might be useful for males, but not females, or the menstrual cycle phase could impact the usefulness of the biomarker in females. Studying biomarkers under combined spaceflight factors in analog environments [88] was also viewed as being important to determine the usefulness of these biomarkers, given that individuals might respond differently to various spaceflight factors.
Default mode network (DMN). The panel discussed the importance of the DMN in both normal and pathophysiological processes as it relates to several of the BMed constructs, and they considered DMN to be a marker that might overlap among constructs (e.g., changes in DMN could indicate memory and attention problems, in addition to sensorimotor changes). The DMN is a brain system that is preferentially activated when the brain is at wakeful rest [89,90]. Core regions of the DMN include the medial prefrontal cortex, posterior cingulate cortex, and parts of the precuneus, as well as the hippocampus, retrosplenial cortex, and angular gyrus [91]. Changes in activation of the DMN have been associated with several psychiatric conditions, including post-traumatic stress disorder, Alzheimer's disease, autism, depression, and chronic pain [92][93][94][95][96]. DMN activation can be modulated by different interventions and physiological processes, including physical activity and exercise, sleeping, resting wakefulness, sleep deprivation [97][98][99], and age [100]. The panel regarded the DMN as an important biomarker of brain function, and given its relationship to other cognitive functions (e.g., attention), they thought it could be useful for understanding changes in operational performance. Because the DMN could be an important marker associated with multiple constructs (e.g., memory, working memory), the panel suggested it could also be an important marker for integration of these constructs and/or how modifying factors influence these constructs (e.g., sleep/wake and sleep deprivation). The DMN seems to be essential to the social understanding of others and could provide a biomarker for spaceflight-associated changes in social cognition and behavior.

Recommendations
The panel evaluated each specific construct to determine if a good biomarker exists that is operationally relevant for astronauts and that translates from animal models. The panel also commented on gaps in each construct that would need to be filled to produce an effective biomarker.
1. Attention. The panel identified several important behavioral markers from attention tests, primarily the psychomotor vigilance test, including increased variability in responses, decreased psychomotor speed, impulsivity, instability in performance, and lapses of attention. Several of these performance measures have been studied on the International Space Station (ISS) and in various analogs of the spaceflight environment [101,102].
2. Dual tasking. This construct overlaps BMed and sensorimotor effects and demonstrates the interconnectedness of numerous constructs relevant to operational performance. Furthermore, dual tasking is argued to be a useful behavioral method for assessing changes in cognitive reserve [103][104][105] during spaceflight and after g-transitions after landing [72,73]. Dual tasking measurements during long-duration spaceflight have identified long-term deficits in visuomotor performance and that cognitive reserve is reduced, possibly due to continued sensorimotor adaptation and stress [72]. Dual tasking measures could be useful behavioral biomarkers of how individuals adapt to the spaceflight environment.
3. Procedural memory. This form of memory [106] was not specifically identified in the two different memory constructs, but the panel felt that it is essential for operational performance and should be mentioned as a subheading under the memory construct.
4. General brain plasticity as an important biomarker of adaptability or lack of adaptability. Operational performance requires a brain that can adapt to stressors under various spaceflight conditions. As such, alterations in brain "adaptability" could be a useful biomarker indicating degradation in operational performance [107]. For example, simple adaptation to repetitive stimuli or general adaptation across multiple tasks (not only task-specific changes) might indicate how the nervous system is faring in a space-like environment (i.e., whether the brain is able to adapt to this new environment, and whether this adaptability is changing over time). This construct is important because it integrates across all measures, can be translated between rodents and humans, and clinical markers of brain damage exist that could be useful biomarkers (e.g., blood brain-derived neurotrophic factor [88]). In addition, learning and plasticity are constructs that have been tested in animal models relevant to astronaut performance (e.g., reversal learning, extinction learning), including after space radiation exposure [108,109].
5. Reversal learning is used extensively in animal models to assess cognitive flexibility and translates well between rodents and humans [110,111]. The panel suggested that reversal learning under stress or under multiple spaceflight stressors could be paired with neuroimaging (e.g., EEG) to identify factors that impair brain adaptability, and to allow translation from rodents to humans.
6. Although social processes were listed as a standalone construct, the panel noted that social interactions are important for the other constructs, and can be affected by the way individuals interact, the way the crew interacts, and how they perceive the interactions of others or the emotional states of others. This is not trivial and is not necessarily easy to assess, but it is integrated into all other constructs. These interactions highlight the need to consider how these individual states impact the group, and the need to determine if there are biomarkers of these interactions, and/or if those interactions then change the individual biomarkers.
7. Inclusion of additional constructs. When the panel took a broad view of the worksheet, they concluded that additional constructs should be added. Although many of these additional constructs were embodied within some of the other constructs, the panel thought they should be discussed as discrete constructs and how they affect operational performance.
Emotion regulation. This includes dysregulation that is subclinical, but not psychiatric disorders such as depression or anxiety, because those are included in the individual behavioral states construct.
Executive function. Assays to measure executive function were included in the attention construct, but executive function, irrespective of attention, is important to operational performance.
Risk taking/tolerance. The Balloon Analog Risk task is included within the astronauts' Cognition Test Battery test, and the panel thought that risk taking/tolerance should be a discrete construct and not embedded within another construct. Risk taking/tolerance is also important for social interactions and group dynamics [112] and should be examined in animal models under different spaceflight stressors.
Stress. For example, astronauts' self-reported stress ratings increased during 6-month ISS missions [102,113] and these changes could have important implications for the usefulness of biomarkers throughout the mission.
The panel identified the following gaps in knowledge: Lack of integrated approach. The panel noted several gaps that could be addressed by first taking an integrated approach to these different constructs. For example, sleep loss or stress will most likely affect all constructs on the list. The constructs are intertwined, and many things can affect them, and for this reason, our group suggested the use of more general biomarkers, instead of construct-specific biomarkers; for example, a "general health" biomarker or a "vulnerability" biomarker that would indicate an individual's status on some continuum of functioning within the spaceflight environment. What remains unknown is whether the biomarkers that have been identified are informative under all conditions, or if these markers will change as external stressors and internal conditions change.
Importance of stress. The panel noted several modifying factors, but stress emerged as a critical factor that probably deserves its own category on the worksheet.
Lack of sex differences or inclusion of sex. Sex needs to be considered throughout all the constructs. It was not included in any construct and could have important implications for determining what biomarkers are relevant and useful.
Inclusion of microbiome. This appears to be important to brain function, and as such, could affect the majority of the BMed constructs. A better understanding of the specific bacteria, dysbiosis, etc., and how they relate to cognition and the different performance constructs, would be useful for biomarker development.
Lack of measurements for individual differences. The panel noted the importance of inter-individual differences for these constructs and their likelihood of affecting operational performance. All individuals can be trained with the same techniques, but it is not known, nor can we currently predict, how each individual will continue to perform in the spaceflight environment. This is especially true when hazards such as radiation exposure and isolation are combined. Methods are required to measure these differences and to understand how they might impact operational performance.
Additional gaps. These include the need for better technology to quantify biomarkers during spaceflight, and greater understanding of the differences between diurnal humans and nocturnal animal models (e.g., rodents) and how this influences the biomarkers we identify and study.    [122]. Salivary: immune markers.
(2) Integrative approach       The goal Group 3 was to use a systems-biology approach to generate lists of biomarkers and signaling pathways related to CNS circuitry and operational performance that will be important to monitor in astronauts during spaceflight and after return to Earth. To achieve this goal, the integrated approaches team (a) reviewed and identified a broad array of biomarkers of important mechanisms known from space research (i.e., what is known); this panel discussed research on biomarkers and signaling pathways in animals and humans that could be used to assess the effects of acute or long-duration exposure to spaceflight stressors on operationally relevant performance; (b) considered knowledge from other CNS-health studies that could be repurposed for assessing astronauts (e.g., aging, disorder, disease); and (c) documented open questions and research gaps in the knowledge base that connect genes and biological pathways to brain regions and neural circuits that link to operational performance (i.e., what is not known, needed experiments). Discussions are summarized below and in Tables 3 and 4. The goal of this integrated approaches team was to provide recommendations regarding the availability, validity, and limitations of biomarkers and signaling pathways to be examined in future research. Table 3. The major observations and points discussed by the panel.

•
Oxidative stress-associated mitochondrial dysfunction has been shown in many cells, tissue and organ system, their impacts have to be further investigated.

•
The role of diet in mitigating oxidative stress associated with spaceflight. • Epigenetic clock measurements in astronauts and related to time in space or deep space and their association with oxidative stress-induced aging. • miRNA signatures and exosomes in identifying oxidative stress biomarkers and as novel biomarkers in brain pathogenesis.

•
Animal to human correlation (circadian and sleep system differences, rhythm differences, immune differences, white-matter differences, vasculature differences).

•
Applying cell-free DNA and subsequent methylation analysis can give high sensitivity measurement of BBB integrity, cell breakdown and inflammation in the brain.    (1) Neurotransmitters provide a direct readout of CNS functionality at multiple levels.

One-carbon metabolism
(2) Cross-species correlated (chemical changes) rodents-NHP-Humans. Should be translated to lipidomic and metabolomic findings.          It should be emphasized at the outset that the results of this integrated approaches exercise did not reveal any biomarker (or combination thereof) that was uniformly responsive across different regions of the brain to a single or given combination of spaceflight stressors. The panel raised the following distinct, yet overlapping questions: 1.
Does the literature provide any useful insight regarding if or how combined exposure to spaceflight stressors might interact to alter (additive, synergize, diminish) biomarkers and signaling pathways involved in CNS function? 2.
What experiments need to be performed to inform how these combined stressors interact and affect biomarkers and signaling pathways associated with CNS function? 3.
What are the challenges that need to be addressed for data collection and storage? 4.
What information do we need for successful biomarker repurposing? 5.
What new experiments, analysis, and techniques are needed? 6.
What information about biomarkers and signaling pathways is needed to identify and implement effective spaceflight countermeasures that will minimize CNS decrements associated with the long-duration spaceflight beyond Earth's protective magnetosphere?
Below is a summary of the key issues that were raised by the integrated approaches panel.

1.
First and foremost, all group members recognized the need for standardizing certain aspects of the experimental protocol across laboratories; in particular, standardizing (a) factors related to the degree of exposure to a spaceflight stressor (e.g., space radiation (Galactic Cosmic Radiation simulation), dose, dose rate, and energy; isolation/confinement; altered gravitational levels (Mars, lunar or Earth)); (b) the type of animal models used (e.g., age, sex, strain, species; see below) and the time of tissue collection. This approach will permit meaningful comparisons and interpretations of data from different endpoints collected among investigators.

2.
The panel overwhelmingly agreed that a paucity of information exists on how CNSrelated neurocognitive performance is affected in laboratory animals that have been exposed to space-relevant radiation (e.g., a low-dose (<0.5Gy)/low-dose-rate of simulated galactic cosmic rays) and that such effects have not yet been systematically studied.

3.
Although studies using several species (e.g., rats, mice) have provided important information about how spaceflight stressors may affect behavior and cognitive function, extrapolating data from rodents to humans is an imperfect science. Notably, the translational value of larger size animals (e.g., NHPs) used in various research domains, including neurobiological, neurobehavioral, and complex cognitive processes, has been validated and established over many decades. These successes are based on numerous factors including (1) the considerable overlap in the genetic, physiological, pharmacokinetic, neurobiology, and neurobehavioral effects in NHPs and humans; (2) the proven reliability of NHPs as subjects in long-duration (i.e., longitudinal) neurobehavioral and cognitive studies; and (3) the ability to use powerful within-subject designs that are similar to those used in human studies, which permit meaningful conclusions or inferences by evaluating all treatment effects in individuals as well as in groups. Considerations such as these suggest that NHPs are especially well-suited for ground-based study of the acute and long-term neurobehavioral effects induced by spaceflight stressors, either alone or in combination, and for translating effects to astronauts. Thus, there was considerable appreciation in the group that the use of appropriate animal models, especially targeted studies in NHPs to confirm or advance observations in rodents, should be carefully considered by NASA in future work.

4.
The panel recognized that an integrated "omics" profiling strategy using technologies such as genomics, proteomics, and metabolomics is desperately needed to further expand understanding of the underlying brain systems/mechanisms that may be affected by exposure to spaceflight stressors. This multimodal approach will be highly beneficial to determine biomarker datasets of differentially expressed genes, proteins, or metabolomic/lipidomic signatures and the pathways that lead to pathological and possible degenerative changes in the brain. An omics-based molecular phenotyping approach for characterizing biosignatures associated with low-dose space radiation, simulated microgravity, and other space environmental stressors will provide a deeper understanding of the underlying mechanisms responsible for brain structure and pathophysiological changes. This approach will also provide critical information about how individual sensitivity (e.g., genetic, epigenetic, previous injury, age, and sex/gender) will influence how spaceflight stressors affect operational performance. However, as stated above, it will be critical for protocols and metadata from experiments in different laboratories to be standardized and processed on a uniform pipeline.

5.
A need was identified for longitudinal studies that provide information about changes within the brain (i.e., acute to chronic). This is especially germane for determining if exposure to spaceflight stressors produces short-or long-term neurobiological (or degenerative) adaptations that affect operationally relevant behavioral and neurocognitive performance. A major complication associated with determining how the brain responds to stress insults is the latency between exposure and the expression of injury (e.g., cell loss or dysfunction). Thus, it is essential that longitudinal studies are conducted to meaningfully quantify the development and progression of the CNS injury response. 6.
At present, few studies have examined the combined impact of spaceflight stressors on operational performance and/or associated neurobiological changes in the brain. Thus, it is critical that future studies use ground-based animal models that incorporate stressors that are inherent to the spaceflight environment, i.e., space-like radiation exposure and other spaceflight environment stressors including high pCO2, fluid shifts, microgravity, environmental constraints, emotional stress, and circadian misalignment/sleep deprivation. This will permit data to be extrapolated more accurately to estimate potential risks encountered by astronauts during deep space missions.
Ground-based studies to examine the impact of combined spaceflight conditions and the underlying mechanism(s) of potential interaction on structural and functional deficits in the brain are very limited. 7.
The panel overwhelmingly agreed that significant effort and resources are needed to develop new cutting-edge techniques to identify brain biomarkers that may indicate operationally relevant neurocognitive performance. Novel imaging techniques that provide an early detection of the subtle changes in the brain and identify the target population and biomarkers for intervention are essential. Thus, to improve knowledge about anatomical, physiological, and functional changes to the brain, especially for longitudinal evaluation, an effort is needed to develop advanced computerized tomography scan, functional magnetic resonance imaging (fMRI), positron emission tomography scan, EEG, magnetoencephalography, and transcranial magnetic stimulation scan imaging technologies.
The panel members agreed that a critical need exists to use data better and carefully from flown astronauts to evaluate the actual acute and long-term health risk of the spaceflight environment. Importantly, there was appreciation that human data could be better related to outcomes from animal studies, which may help characterize alterations in circadian rhythm and sleep, immune system, neurotransmitters, neurobiology (i.e., brain structure and function), and vasculature. If used carefully, follow-up analysis of omics, biochemistry, imaging, and a battery of behavior and neurocognitive testing will provide critical human data that may be used to evaluate the actual acute and long-term health risk of the space environment. Table 3 highlights the major observations and points of discussion that were addressed by the integrated approaches panel. Although it is likely that exposure to combined space-flight stressors will alter a wide range of biomarkers in different endpoints in animals and humans, ultimately, it is critical that these biomarkers are consistently and reliably linked with changes in operationally relevant behavior and neurocognitive performance. Evidence so far suggests that specific neurocognitive impairments may manifest under evolving mission scenarios (i.e., increased cognitive load) and, therefore, assessing the impact of spaceflight hazards on a wide range of operationally relevant behavioral and neurocognitive tasks is critical. Moreover, the panel suggested that NASA should explore both novel and trained paradigms with increased difficulty of determining the level of impairment. Finally, to promote translation between animal models and humans, parallel behavioral and neurocognitive testing paradigms exist between rodents ↔ NHPs ↔ humans that should be further exploited.

Recommendations
The panel identified the following gaps in knowledge: • How can data be integrated across many biology scales for CNS endpoints? • How can system biology approaches with new technologies-organ cultures, organson-a-chip made from normal human cells, integrated "omics" (genomics, proteomics, metabolomics) and cutting-edge brain imaging techniques-be used to estimate acute CNS risks to astronauts from space environment? • How can knowledge of space environment-induced biomarkers/pathways in neuroinflammation, blood-brain barrier function, vasculature, glia activation be integrated towards better understanding of their impact on acute pathophysiological changes in the brain and late neurodegeneration? • What is the likelihood of increases in the brain susceptibility to later development of neurological disorders as results of observed changes? • What is the relationship between neurochemical biomarkers and operationally relevant performance? • What are the temporal and regional differences in neurochemical biomarkers and their influence on operationally relevant performance? What is the right neurochemical balance? • What CNS neurotransmitter metabolites can be measured peripherally? Can wearable devices/sensors be used instead of blood? • Is personalized nutrition (i.e., B-vitamin supplementation) a viable SANS countermeasure? • Do recurring cycles of sleep deprivation affect performance/vestibular/sensorimotor changes, recovery, and biomarkers? • What is the role of individual susceptibility-genetic, epigenetic, previous injury, age, and sex/gender-in addressing CNS risk?
Information that is lacking includes astronaut data to monitor the level of DNA damage over time; miRNA signatures as neurodegeneration markers for acute/chronic injury; data from integrated phenotypic studies in models; and omics to identify molecular changes at the synaptic level.

Overall Summary and Recommendations
In total, hundreds of biomarkers have been identified and synthesized through this effort. Synthesizing across all three topical groups, the following common responses emerged as general themes:

•
Biomarkers span all levels of data from molecules to behavior.

•
Integrated stressors and integrated effects should be studied, including studies using multi-sensory approaches, for example, combined sleep and radiation exposure.
o Note combined effects of HZE radiation exposure and sleep fragmentation in rodent models show dramatic effects specific to brain regions [109]. o Integrated sensorimotor and cognition effects should be considered for study, e.g., olfaction and vestibular.

•
The responses themselves will have multiple downstream impacts. Treatment may not be successful following a reductionist manner.
• Modifying factors should be identified and tracked throughout assessment, e.g., cognitive load, stress, circadian aspects, and sex, and their impacts on executive function and attention.

•
Learning and plasticity were highlighted as critical areas to assess during spaceflight to determine the astronaut's general level of cognitive and sensorimotor adaptability. • Biomarkers were recommended not just for immediate predictiveness, but also for long-term predictiveness of damage (late effects that can follow the initial injury by months or longer). As an example, some omics biomarkers may precede pathologies by months.

•
Studying appropriate animal models in parallel with astronauts is extremely valuable for determining applicable constructs/responses, and to better understand the astronaut's condition.
We hope this effort yields usable knowledge and an effective tool for HRP and the CBS Project to improve monitoring and management of astronaut cognitive and behavioral health.
Author Contributions: All co-authors played a critical role in collecting and disseminating the knowledge in this article. Meeting organizers, discussion leads, participants, and facilitators are itemized in Appendix B. All authors have read and agreed to the published version of the manuscript.

Appendix A. Agenda of Meeting
A NASA translational working group TIM titled Circuits and Biomarkers of the Central Nervous System Relating to Astronaut Performance (Biomarker TIM) was held virtually between 21-25 September 2020, and was supported by the NASA HRP's Human Factors and Behavioral Performance Element in conjunction with Space Radiation Element and the Human Health Countermeasures Element. The goals of this Biomarker TIM were to (1) identify relevant brain regions, neural circuits, functions, and associated biomarkers that relate to operationally relevant performance and (2) identify any critical needs for new biomarker knowledge ("gaps") that can be filled by additional focused and translational animal experiments that include a plausible pathway toward eventual biomarker validation in humans.
Deliverables addressing these goals may ultimately inform countermeasure strategies to maintain performance standards and identify performance limits for astronauts. To address the goals, 22 extramural experts from 19 academic institutions and 26 intramural experts from various NASA centers contributed to 15 talks reviewing findings from biomarker research on animals and humans in response to terrestrial and spaceflight stressors, and then participated in virtual thematic breakout sessions to systematically and qualitatively review biomarkers and associated brain circuits for 30 cognitive or behavioral constructs or physiological responses. The topics of the breakout sessions were sensorimotor influences (Group 1), behavioral medicine influences (Group 2), and integrated approaches to understanding operationally relevant performance (Group 3), and respective behavioral constructs listed in Table A1. Before the TIM, a portfolio of documents and scientific literature was shared with participants to frame the workshop and help the participants prepare.