Methods for assessing change in brain plasticity at night and psychological resilience during daytime between repeated long-duration space missions

This study was designed to examine the feasibility of analyzing heart rate variability (HRV) data from repeat-flier astronauts at matching days on two separate missions to assess any effect of repeated missions on brain plasticity and psychological resilience, as conjectured by Demertzi. As an example, on the second mission of a healthy astronaut studied about 20 days after launch, sleep duration lengthened, sleep quality improved, and spectral power (ms2) co-varying with activity of the salience network (SN) increased at night. HF-component (0.15–0.50 Hz) increased by 61.55%, and HF-band (0.30–0.40 Hz) by 92.60%. Spectral power of HRV indices during daytime, which correlate negatively with psychological resilience, decreased, HF-component by 22.18% and HF-band by 37.26%. LF-component and LF-band, reflecting activity of the default mode network, did not change significantly. During the second mission, 24-h acrophases of HRV endpoints did not change but the 12-h acrophase of TF-HRV did (P < 0.0001), perhaps consolidating the circadian system to help adapt to space by taking advantage of brain plasticity at night and psychological resilience during daytime. While this N-of-1 study prevents drawing definitive conclusions, the methodology used herein to monitor markers of brain plasticity could pave the way for further studies that could add to the present results.

. Frequency-domain measures of heart rate variability (HRV).

Frequency-domain measures (units, ms 2 ) Frequency range (Hz) Description Related references Brief physiologic correlation
Conventional frequencydomain measures of HRV TF-HRV 0.0001-0. 50 Variance of all N-N intervals over 3-h interval 29 Index suggestive of anti-aging or longevity VLF is the most predictive of adverse outcomes, including all-cause motality. Historically, VLF may reflect both vagal control of heart rate and also the effect of the renin-angiotensin system. Recently, VLF rhythm appears to be produced by the heart itself and may be an intrinsic rhythm that is fundamental to health and well-being  19,37 Index primarily related to brain's DMN activity, in medial prefrontal and precuneus/posterior cingulate cortex parts MF1-band 0.05-0. 10 Power in MF1-band range, ISO Index primarily related to brain's DMN activity, mainly in the thalamus and basal ganglia MF2-band 0.10-0. 15 Power in MF2-band range Index primarily related to brain's DMN activity, mainly in the orbitofrontal, insular and temporal cortex parts HF01-band 0. 15-0.20 Power in HF1-band range Index primarily related to medial orbitofrontal cortex (mOFC)/medial prefrontal cortex (mPFC)-guided core integration system HF02-band 0.20-0. 30 Power in HF2-band range 17,19,[31][32][33][34]38 Index reflecting relative vagal modulation of heart rate in response to respiration HF03-band 0. 30 www.nature.com/scientificreports/ Statistical analyses. Data shown in Table 2 are expressed as mean ± standard error (SE). The ECG recording was started at 13:25 during the 1st space-flight mission, and at 17:50 during the 2nd space-flight mission. For comparison of HRV indices, statistical analyses were applied on hourly averages of the 5-min estimates in order to minimize serial correlation. Paired hourly HRV indices were compared between the two spaceflights, focusing on ISS01 (days 18 and 21 after launch, respectively), using the paired t-test. Cohen's distance was determined to assess effect size. The Stat Flex (Ver. 6) software (Artec Co., Ltd., Osaka, Japan) was used. P-values less than 0.05 were considered to indicate statistical significance.

Results
Sleep performance. While sleep duration around day 20 after launch during ISS01 cannot be statistically compared between the two missions without knowing its day-to-day variation for this astronaut, it was more than one hour longer on the second than on the first mission (374 vs. 300 min during ISS01 and 365 vs. 295 min during ISS02). Such large differences between the two missions are not seen before launch (297 vs. 289 min) and after return to Earth (330 vs. 360 min). Sleep quality may also have been improved on the second compared to the first spaceflight, as suggested by a clear increase in spectral power of the HF-HRV, Fig. 1.

Assessment of circadian and circasemidian components of HRV endpoints.
Circadian and circasemidian amplitudes of HRV endpoints are shown in Table 3. On the first spaceflight, the circadian amplitude of HR increased more than two-fold both during ISS01 (255%) and ISS02 (271%) compared to pre-flight, as observed previously 28 . This was not the case on the second spaceflight. On the first mission, the circadian amplitude of the intrinsic cardiovascular regulatory system function (β) also increased during both ISS01 (303%) and ISS02 (233%), compared to pre-flight. The adaptation behavior of the 12-h component of HRV endpoints was remarkably larger than that of the 24-h rhythm, particularly for TF-HRV, seen in both amplitude and phase. The 12-h amplitude of TF-HRV increased up to 574% and 473% during ISS01 and ISS02, respectively, on the first spaceflight, although similar changes were not clear on the second spaceflight.
An apparent phase shift of the 24-h and 12-h components of HRV endpoints in response to spaceflight was observed after fitting a single 24-h or 12-h cosine curve separately to the 20 HRV measures by cosinor. Results are summarized in Table 4, where misaligned circadian phases occurring at unusual times (such as day-night reversals), are shown in bold. On the first but not on the second spaceflight, quite a few HRV endpoints show circadian misalignment pre-flight (Table 4, left), suggesting that circadian desynchrony due to social jetlag was larger on the first than on the second mission.
The recovery process of such internal desynchrony of the 20 HRV indices during spaceflight is illustrated in Fig. 3. It depicts the time course of the circadian (Fig. 3, top) and circasemidian (Fig. 3, bottom) acrophases during the first (Fig. 3, left) and second (Fig. 3, right)

Discussion
Repeated HRV monitoring over 24 h of a healthy astronaut during two space missions, 4 years apart, served to illustrate methods to assess changes in brain plasticity and psychological resilience. Whereas conclusions cannot be derived from data from a single astronaut monitored for only 24 h at discrete times before, during and after a mission in space, the methodology used herein could be used in future studies to determine whether neural adaptation improves on repeated missions, as observed herein around day 20 after launch. During nighttime, sleep improved, and HRV activity co-varying with brain neural activity in the SN accelerated, while decelerating during daytime. HRV endpoints reflecting DMN activity showed no differences between the two space missions.
Stimulating environment and brain plasticity. Brain plasticity refers to the capacity of neurons and of neural circuits in the brain to change, structurally and functionally, in response to experience. This property is fundamental for the adaptability of behavior, for learning and memory processes, brain development, and brain repair. Exposure to stimulating environments has repeatedly been shown to strongly influence brain plasticity. Thus, it is a crucial underlying component of the enormous challenge of space adaptation for astronauts. Neural plasticity can take place at several levels, from synaptic plasticity at the (sub)cellular level to plasticity at the system and network levels [44][45][46] . Brain plasticity can be studied with a number of methods, such as electroencephalography (EEG)/evoked potentials (ERPs), structural and functional MRI and transcranial magnetic stimulation (TMS). In addition to functional brain response 47-50 illustrated herein, recent work showed structural changes in the brain after long-duration space flights 4-7 resulting from alterations in sleep performance or functional brain networks. Both aspects were estimated by HRV in a healthy astronaut who also took part in repeated space missions 20,21,28 .
Our observation of improved sleep agrees with results from a previous study 28 where we assessed sleep quality based on sleep-related changes in RR-intervals and HRV-HF, which we found to be improved in space, and to be associated with increased parasympathetic activity, contrary to previous investigations [51][52][53][54] . Sleep quality was assessed as changes in sleep performance and HRV behavior in specific frequency regions for interpretation in terms of functional brain networks, as done in previous studies 20, 25,26,28,42 . Previous investigations reported shorter sleep duration and inadequate sleep quality of astronauts during spaceflight aboard the ISS. These results were Figure 1. Estimation of sleep span and assessment of sleep quality. RR-intervals (first two rows) and HF-HRV (last two rows) of the first (left) and second (right) spaceflight assess sleep duration and sleep quality during ISS01 and ISS02, respectively. HF-HRV spectral power, reflecting sleep quality, is clearly larger on the second than on the first spaceflight during both ISS01 and ISS02. Sleep-related increase in RR-intervals also appears to be larger on the second than on the first spaceflight. Effect of nighttime HRV changes on brain plasticity in space. Despite increased interest in the effect of spaceflight on the human central nervous system (CNS) 15,55 , not much is known thus far about the functional and morphological effects of microgravity on the human CNS. Previous studies have shown that CNS changes occur during and after spaceflight in the form of neuro-vestibular problems, alterations in cognitive function and sensory perception, problems with motor function, cephalic fluid shift, and psychological disturbances 56,57 . In the past few years, advances in structural and functional neuroimaging techniques have shown spaceflight-induced neuroplasticity in humans in several brain regions, including the insular cortex, the temporo-parietal junction, and the thalamus, in relation to short-and long-duration spaceflight 1,2,14 .
HRV indices that co-vary with SN activity are of particular interest since the SN is linked to the autonomic nervous system function and is sensitive to environmental challenges. The SN is mainly centered on the dorsal anterior cingulate, extending into the perigenual anterior cingulate cortex, and orbital fronto-insular cortices, but it also encompasses the limbic and brainstem areas. Relevance to HF-HRV is suggested by the inclusion of known autonomic nervous system control areas in the SN, and by this vagal marker's putative role in switching between rest and activity and between internal and external focus of attention.
Further investigations are needed to examine whether acceleration of SN activity starts with nighttime sleep, as observed herein. It would suggest that brain plasticity may be initiated at night. The sensitivity of vagallyinduced heart rate reactions to event salience might further suggest relationships between the SN and HF-HRV, as might the apparent overlap between nodes of the SN and areas related to autonomic control.
Identified as related to HF-HRV, the mPFC is important both as a node in the DMN and in the SN 58 . Anatomically, the mPFC is known to connect to pre-autonomic cell groups in the hypothalamus, periaqueductal gray, and brainstem 59,60 . If diffuse attention is a major aspect of the functionality of the DMN, then the overlapping membership of the mPFC in the two networks would provide an anatomical site for shifting from DMN activation to SN activation. Some evidence supports the view that DMN activation is switched to SN activation when an interoceptive or environmental stimulus is encoded as significant 61 . Table 2. Comparison of heart rate variability between the two spaceflights suggests role of brain functional network for faster adaptation to microgravity on second spaceflight. Tests applied on hourly averages of 5-min intervals in order to eliminate or at least reduce serial correlation. n number of hourly averages (of 5-min intervals), SE standard error. P-values not adjusted for multiple testing; P < 0.05 after adjusting for mutiple testing highlighted in bold. www.nature.com/scientificreports/ Daytime HRV fluctuations associated with brain resilience in space. Because HRV may be associated with neural structures that are involved in the appraisal of threat and safety, HRV can be considered a potential marker of stress. HRV reflects the status of one's ongoing adjustment to constantly changing environmental demands. Previously, under stressful environments, such as performing tasks during a spaceflight mission, HRV was found to be decreased 17 . Increased HF-HRV is considered to be associated with a positive mood, absence of negative affect, and an alert readiness to engage with the physical and social environment 62,63 .
Much recent research has found that psychological resilience is mediated by spontaneous brain activity measured with resting-state functional MRI. Although Waugh et al. 64 found that when faced with a threat, participants had prolonged changed activity in the insula in response to aversive stimuli, psychological resilience is a complex construct that likely involves different brain functions. Other studies provided evidence that brain resilience is related not only to the insula, but also to the mPFC, OFC, PCC, ACC, and thalamus [65][66][67][68][69][70][71] . In the extant literature, the most consistent brain area related to psychological resilience is the ACC, perhaps because the ACC is associated with many important emotional functions, including motivation, emotion regulation, and attention or adaptation to a novel environment, such as space [72][73][74][75] . Previous investigations on resilience speculated that local activity in the ACC (such as fractional amplitude of low-frequency fluctuations measured by fMRI) would be negatively associated with psychological resilience 23,75,76 .
The bi-directional connections between heart and brain enunciated by Claude Bernard can be studied by analyzing HRV 17,77 . Over the past several years, many neuroimaging studies examined the association of HRV endpoints with fluctuations in brain functional connectivity 18,19,[32][33][34][35]59,60,78 . They confirmed the existence of intimate connections between the different brain regions and HRV endpoints. They also posited that any changes in brain functional networks, which dynamically adjust the structure of their global and local network connectivity, should affect and change HRV activities in their respective frequency bands. "HRV is like a mirror reflecting the strength of activities of humans' brain and mind" 17,19,77 .
Astronauts' motivation aboard the ISS is also expected to reflect changed activities in the respective HRV frequency bands. Several investigations reported a relation between levels of psychological well-being and HRV 35,79 , which confirmed a statistically significant negative correlation between life satisfaction and HF-HRV activities 35 . Should our observation of decreased spectral power of HF-HRV, HF-component, and the series of the HF-band  Role of biological rhythms in the adaptation to the space environment. Whereas the circadian system plays a key role in the adaptation to a novel environment, such as microgravity in space 19,20,[25][26][27][28] , ultradian components provided an evolutionary advantage for almost all life forms, from bacteria to humans [80][81][82][83][84] . These ultradian rhythms can be expected to be important for the rapid adaptation to microgravity in space. The 12-h (circasemidian) component in particular may be involved [85][86][87][88][89][90] . It may reflect the function of two stress response pathways reacting to unfolded protein in the endogenous endoplasmic reticulum (ER) and mitochondria. A 12-h (circasemidian) component characterizes the ER-and mitochondria-associated "unfolded protein response (UPR) cycle" [88][89][90][91][92][93] . Several potential roles of the circasemidian clock in coordinating human health have been proposed, such as maintaining metabolic homeostasis 87 , coordinating sleep quality of slow wave sleep 94,95 , and mediating aging, especially in the prevention of aging-related metabolic decline 87,88,96,97 .
Based on our observations herein, the following hypothesis comes to mind. First, when faced with a new environment in space, the 12-h response appears faster and is larger than the circadian response (Table 3). Second, strong 12-h clock regulation might help repair circadian desynchrony (Table 4 and Fig. 3). The more severe internal desynchrony is (Table 4, Flight 1), the larger is the activation of the 12-h component (Table 3 and Fig. 3, Flight 1). Third, a milder circasemidian response during the second than during the first mission suggests that spaceflight-induced neuroplasticity may be present in the astronaut's brain during the second mission.
Harmonic oscillations of 24 and 12 h likely provide evolutionarily adaptive advantages. The 12-h (circasemidian) component may contribute to consolidating a strong circadian system in space, and may contribute to a better adaptation in space by taking advantage of brain plasticity at night and psychological resilience during daytime.
Limitations. This investigation has several limitations. First, the study is limited to a single astronaut, and results were only compared between missions on a single day (ISS01). Factors other than adaptation to space environment (such as exercise, nutrition, mission tasks, and interpersonal stress) likely contributed in part to the results. As such, results herein do not provide inferential information about the effect of repeated missions of many days flown by a "population" of astronauts. Future studies should be designed to also estimate the uncertainty due to variation between astronauts and between mission days for each astronaut.
Despite the medium to large effect size of changes observed in this illustrative case, serial correlation, reduced by considering hourly averages instead of the original 5-min HRV endpoints, remains an issue preventing the derivation of generalizable inferences. As similar data from other astronauts become available, individual estimates can be used as imputations that no longer depend on the sampling interval.
In view of the importance of the circadian rhythm, there is merit in recording ECG around the clock. Demanding schedules and inconvenience of implementing the monitoring have been limiting factors to obtaining more data or data covering spans longer than 24 or 48 h. As technology advances, ECG monitors may become easier to use for longer spans, and as space exploration expands, more space travelers may participate in similar studies in the future.
Space adaptation of human neural cardiovascular coordination remains a challenge, as mechanisms are diverse and complex. Second, brain oscillatory activity data are lacking. Several studies, however, showed that HRV is associated with structures and functions of the neural network, and HRV is a biomarker reflecting  www.nature.com/scientificreports/ activities of the brain integration system. These associations are extremely complex, however, and have not yet been fully confirmed. Future investigations are needed to directly assess the brain's oscillatory activity in space. The methodology used in this investigation may help address these complex issues in future studies.

Conclusion
We examined the hypothesis proposed by Demertzi et al. 14 that second-time flyers adapt more quickly and are less prone to microgravity-induced problems 14,16,21 . This demonstration is a simple illustration of methodology aimed to assess changes in brain plasticity and psychological resilience in a single astronaut, limited to comparing Table 4. Circadian and circasemidian phase changes in heart rate variability indices in response to space flight differ between the two long-term missions. Bold cells mean circadian phase misalignment induced by astronaut social jetlag on Earth before spaceflight mission. Circasemidian acrophases expressed in (negative) degrees, with 360° ≡ 12 h, 0° = 00:00. www.nature.com/scientificreports/ HRV endpoints between missions on a single day. Results nevertheless confirm earlier findings that sleep duration lengthened and sleep quality improved in space. The methodology used herein outlines how HRV behavior, which estimates the process of neural adaptation 17,20,21,28 , can serve to interpret changes in terms of brain functional networks. In the case examined herein, we find that brain plasticity during nighttime and psychological resilience during daytime may help with the adaptation to space's environment. The 12-h component may have played a role in the adaptation process since it underwent larger changes than the 24-h component in response to the space environment, as assessed around day 20 on the ISS. HRV in the HF spectral region may be critical to assess microgravity-induced brain plasticity and psychological resilience, because HF-HRV reflects the adaptation process. Further studies are needed to examine how adaptation to the microgravity environment in space occurs. The role of functionally integrating the SN, consisting of neural centers (ACC, OFC, Amygdala and Insula), which involves and responds in a task-dependent manner to interceptive-autonomic and reward processes in a task-independent manner to emotional and homeostatic stimuli of personal salience 23,35,71,72,75,[98][99][100] may be particularly important, as our data suggest.

Data availability
Restrictions from Japan's Aerospace Exploration Agency (JAXA) apply to the availability of the data supporting the findings of this study. The data were used under license for the current study. Although data are not publicly available, they are available to collaborating parties under ethical approval from JAXA.