Simultaneous exposure to chronic irradiation and simulated microgravity differentially alters immune cell phenotype in mouse thymus and spleen

Abstract

Deep–space missions may alter immune cell phenotype in the primary (e.g., thymus) and secondary (e.g., spleen) lymphoid organs—contributing to the progression of a variety of diseases. In deep space missions, astronauts will be exposed to chronic low doses of HZE radiation while being in microgravity. Ground-based models of long-term uninterrupted exposures to HZE radiation are not yet available. To obtain insight in the effects of concurrent exposure to microgravity and chronic irradiation (CIR), mice received a cumulative dose of chronic 0.5 Gy gamma rays over one month ± simulated microgravity (SMG). To obtain insight in a dose rate effect, additional mice were exposed to single acute irradiation (AIR) at 0.5 Gy gamma rays. We measured changes in proportions of immune cells relative to total number of live cells in the thymus and spleen, alterations in stress level markers in plasma, and change in body weight, food consumption, and water intake. CIR affected thymic CD3+/CD335+ natural killer T (NK-T) cells, CD25+ regulatory T (Treg) cells, CD27+/CD335− natural killer (NK1) cells and CD11c+/CD11b− dendritic cells (DCs) differently in mice subjected to SMG than in mice with normal loading. No such effects of CIR on SMG as compared to normal loading were observed in cell types from the spleen. Differences between CIR and AIR groups (both under normal loading) were found in thymic Treg and DCs. Food consumption, water intake, and body weight were less after coexposure than singular or no exposure. Compared to sham, all treatment groups exhibited elevated plasma levels of the stress marker catecholamine. These data suggest that microgravity and chronic irradiation may interact with each other to alter immune cell phenotypes in an organ–specific manner and appropriate strategies are required to reduce the health risk of crewmembers.

1. INTRODUCTION

Astronauts on deep–space missions will experience continual microgravity (MG) and uninterrupted doses of ionizing radiation (IR) higher than any exposure experienced in the history of human spaceflight. Such missions are outside of low–Earth orbit or LEO (an altitude of above 1,200 miles), including proposed trips to cislunar space (200,000 miles from earth) in the 2020s and to Mars (35,000,000 miles from earth) in the 2030s. During these flights, astronauts will be exposed to uninterrupted low dose and low dose–rate (~1.84 mSv/day) of chronic ionizing radiation (CIR) from galactic cosmic rays (GCRs) with occasional short–term exposure from solar energetic particles (SEPs) during solar flares and coronal mass ejections during their entire time in space (Norbury et al., 2016). It is predicted that during a 3–year mission to Mars, the cumulative effective radiation dose from GCRs and SEPs will be in the range of approximately 50–500 mSv (Cucinotta and Durante, 2006; Cucinotta et al., 2008; Zeitlin et al., 2013; Norbury et al., 2016); equivalent to approximately 0.05–0.5 Gy absorbed radiation dose, which is a dose substantially higher than background radiation on Earth. Importantly, a few biological data sets are available from deep–space missions, specifically from lunar missions, but these are not sufficient to predict the risks to human health; notably, no data are available on alterations in immune cell phenotypes in the lymphoid organs (e.g., thymus and spleen), which can adversely affect the immune response, thus overall human health. Using a ground-based model, Mao et al. reported the effects of coexposure of proton beam (a primary component of SEPs) and simulated microgravity (SMG) on retinal endothelial and hematopoietic cells (Mao et al., 2019). Since exposure to SEPs is typically acute, Mao et al used a single dose of protons after 7 days of SMG and followed the protons with another 7 days of SMG (Mao et al., 2019). To understand the effects of exposure to chronic uninterrupted low doses of ionizing radiation with concurrent microgravity, Mao et al. have used a mouse model of chronic gamma rays and SMG (Mao et al., 2016; Seawright et al., 2017). However, the effects of CIR with SMG on the hematopoietic cells are not yet fully understood. To understand the effects of uninterrupted combined exposure to CIR and microgravity on the immune cells, we used an open gamma ray source with or without SMG. We acknowledge gamma radiation does not exactly mimic the space radiation, which is primarily composed of high linear energy (LET) charged particles with higher relative biological effectiveness (RBE) than low-LET gamma rays, however, while current ground-based facilities are capable of exposing small animals to a mixture of high-LET charged particles as in GCR, fully mimicking the long-term uninterrupted chronic exposures as experienced in deep space is still not possible. We built a facility in which mice were housed around an open gamma ray source to gain a better understanding of the risks associated with concurrent uninterrupted exposure to CIR and SMG on immune cells.

The thymus and spleen harbor multiple types of immune cells with a myeloid (e.g., dendritic cells [DCs], granulocytes, and macrophages) or lymphoid (e.g., B cells, T cells, regulatory T cells (Treg), and natural killer [NK] cells) origin that provide protection against various conditions of oxidative stress, including infections and cancers. The thymus, a primary lymphoid organ, is where bone marrow–derived progenitor cells convert to naïve T cells by sequential differentiation, maturation, and selection before egressing newly developed T cells from the thymus into the peripheral lymph nodes (Stritesky et al., 2012). Notably, thymic myeloid cells play a critical role in this T–cell developmental process by providing necessary signals (Wang et al., 2019). On the other hand, the spleen, a secondary lymphoid organ, is the place for B–cell–mediated capturing of blood–borne pathogens and antigens that subsequently activates the T–cell immune response, macrophage–dependent removal of dying cells, DC–mediated regulation of adaptive immunity, iron metabolism, and erythrocyte homeostasis (Bronte and Pittet, 2013). Importantly, structural, morphological, cellular, and functional alterations occur in these two organs under various conditions of oxidative stress.

Mice subjected to prolonged hind–limb unloading (HLU), a NASA–approved animal model of SMG (Morey-Holton and Globus, 2002; Globus and Morey-Holton, 2016), exhibit substantial atrophy (Wang et al., 2007), alterations in immune cells phenotype (Gaignier et al., 2014), and T–cell activation (Sanzari et al., 2013) in the thymus and/or spleen. Moreover, ground–based studies revealed that exposure to CIR alters immune cell population in the thymus and the spleen (Ina and Sakai, 2005). Notably, long-term coexposure of CIR and SMG in modifying immune cell phenotype in the thymus and the spleen remain unstudied until now. In addition, spaceflight experiments demonstrated that there is a considerable change in the expression of T–cell– and cancer–related genes (Gridley et al., 2013); a decrease in lymphocytes, white blood cells, macrophages, and granulocytes (Baqai et al., 2009); an alteration in cytokine production (Miller et al., 1995); and hypoplasia (Durnova et al., 1976) in the thymus and/or spleen of rodents after a short–term space mission as compared to ground controls. Finally, astronauts were observed to have alterations in blood immune cell phenotype or their functional activities during and after LEO space missions (Taylor and Dardano, 1983; Crucian et al., 2013; Crucian et al., 2015; Bigley et al., 2019). These ground–based or spaceflight studies on rodents and humans demonstrated that space factors alter the phenotype and functional activity of immune cells. However, it is to be noted that these spaceflight experiments on animals and humans were conducted within LEO and for a short period, ranging from a few days to one or two weeks. No data are available on alterations in phenotype and activity of immune cells after deep space missions. Therefore, ground–based animal studies mimicking the deep space environment, as close as realistically achievable, are crucially important to understand the potential effects of various space factors, specifically concurrent exposure to CIR and SMG, on the immune cells present in primary and secondary lymphoid organs.

We have developed a ground–based facility that allows continuous coexposure of CIR and SMG. CIR was performed using a Cs¹³⁷ gamma source and microgravity was simulated by HLU. We subjected C57BL/6 male mice, 8 to 12 weeks old, to CIR and/or SMG for a month, considering a month of mouse age to be equivalent to 3 years of human age (the time required to complete a Mars mission). To obtain insight in a dose rate effect, additional mice were exposed to single acute irradiation (AIR) with gamma rays. We observed substantial changes in certain immune cells in the thymus after coexposure to CIR and SMG. However, spleen immune cells exhibited no such changes after coexposure. The effects of CIR on alterations in the immune cell phenotype and alterations in body weight, food consumption, and water intake are different from AIR exposure to the same total dose. Finally, all treatment groups showed higher levels of catecholamine, a plasma marker of stress compared to the sham group. These findings suggest that appropriate strategies are required to protect immune cells from the adverse effects of long–term deep space missions.

2. Materials and Methods

2.1. Reagents and antibodies

Table 1 lists the fluorophore tagged antibodies that were purchased from BD Biosdences (San Jose, CA, USA) and used for immune cell phenotyping. Brilliant violet staining buffer (Cat. No. 563794) was also purchased from BD Biosciences. Trypsin (Cat. No. 25-200-072), bovine serum albumin (BSA; Cat. No. BP9706100), fetal bovine serum (FBS; Cat. No. 16-140-071), and cell strainers (Cat. No. 22-363-547) were obtained from Fisher Scientific (Pittsburg, PA, USA). Collagenase I (Cat. No. 07416) was obtained from Stem Cell Technologies (Vancouver, BC, V6A 1B6, Canada) and red blood cell (RBC) lysis buffer (Cat. No. IBB198) was obtained from Boston Bioproducts (Ashland, MA, USA).

Table 1.

List of antibodies from BD Biosciences used for immune cell phenotyping.

Fluorophores	Antibodies	Clone	Cat. No.
BV711	Anti-Mouse CD4	GK1.5	563050
BV786	Anti-Mouse CD3	17A2	564010
APC	Anti-Mouse Ly-6G	1A8	560599
FITC	Anti-Mouse CD11b	M1/70	553310
PE	Anti-Mouse F4/80	T45-2342	565410
BV605	Anti-Mouse CD25	PC61	563061
PerCP-Cy5.5	Anti-Mouse CD27	LG.3A10	563603
APC-Cy7	Anti-Mouse CD11c	HL3	561241
BV650	Anti-Mouse CD19	1D3	563235
AF700	Anti-Mouse CD335 (NKp46)	29A1.4	561169
BV421	Anti-Mouse Foxp3	MF23	562996
-	Anti-Mouse CD16/CD32 (Mouse BD Fc Block™)	2.4G2	553141

2.2. Animal model

All animal studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The Animal Use Protocols (Nos. 3823 and 3886) were approved by the Institutional Animal Care and Use Committee at UAMS. C57BL/6 male mice of 8 to 12 weeks of age were obtained from the Jackson Laboratory (Bar Harbor, ME, USA) and acclimatized at UAMS DLAM facility for two weeks before used for experiments. The mice were housed in conventional cages in a pathogen–free environment with controlled humidity, temperature, and a 12:12 hour light-dark cycle with free access to drinking water and standard chow (Teklad, Madison, WI, USA) during quarantine for 2 weeks.

2.3. Irradiation method and dosimetry

A Mark I Model 25 self-contained Category I irradiator (JL Shepherd and Associates, San Fernando, CA, USA) was commissioned in a basement laboratory. Additional stacks of lead attenuation of one and a half–value layer (7 mm) and one tenth–value layer (23.2 mm) thicknesses were available. Dose rate calculations were performed using broad beam geometry at distances of 1 m from the source, at the treatment console, and at the door and cross-checked based on Papagiannis et al (Papagiannis et al., 2008). Radiation shielding calculations were performed based on the maximum photon energy (i.e., 662 keV) taking into account the room dimensions, concrete wall thicknesses, and occupancy of adjacent rooms in accordance with the National Council on Radiation Protection and Measurements Report No. 49. Additionally, a “sky–shine” attenuator was custom–designed in order to lower radiation exposure to the floor directly above. As a part of acceptance testing of the irradiator, radiation surveys were performed using a calibrated Geiger-Muller survey meter validated the radiation levels. Dose–rate measurement was carried out using an Optically Stimulated Luminescent nanoDot dosimeter (Landauer Inc., Greenwood, IL).

A well–shielded X–ray room, located in the basement of the UAMS Winthrop P. Rockerfeller Cancer Institute, was transformed into a satellite animal facility for the CIR experiments. All the requirements for an animal facility, specifically heating, ventilation, and air conditioning, were fulfilled. A time–controlled lighting system (12:12 hour light–dark cycle) was used to ensure a uniform diurnal lighting cycle in the CIR facility.

The cages were custom–made and trapezoid in shape with a floor area of 175 cm² (which is sufficient for housing a single mouse) as recommended in the “Guide for the Care and Use of Laboratory Animals”; Eighth Edition). Cages were placed in a circle around the CIR source on custom–made two tier plastic tables, so that all animals received a uniform radiation dose (Figure 1). Each table can hold 9 mouse cages (5 cages on the upper shelf and 4 cages on the lower shelf). We placed 6 tables at a time around the source as shown in Figure 1. Two sets of cages were alternately used and changed once a week, so that one set was cleaned while the other set was used. Mice were checked daily, during which the Cs¹³⁷ was lowered for no more than 20 minutes.

Figure 1.

CIR to mice with or without SMG.

We equally randomized 40 mice into five groups: 1) sham, 2) SMG, 3) CIR, 4) SMG + CIR, and 5) AIR. Animals were exposed to the assigned treatments for 1 month. For the CIR and SMG + CIR groups, animals received 0.5 Gy gamma radiation over the period of 1 month, while AIR mice received a single dose of 0.5 Gy gamma radiation on Day 1. Practical constraints limited the number of tissue samples that could be processed in one day. Therefore, on Day 30, 4 mice from each group were sacrificed and tissues harvested. The remainder 4 mice from each group were sacrificed on Day 31. We account for these 2 observation days in the statistical analyses. Mice were housed singly for all treatment groups.

For AIR experiments, mice (not anesthetized) were exposed to a single total body irradiation (TBI) dose using a J.L. Shepherd & Associates Mark 1 Model 68A Cs¹³⁷ source irradiator. During irradiation, the mice were placed in a custom–made, well–ventilated aluminum chamber with a Plexiglas lid (J.L. Shepherd & Associates). The chamber was divided into 8 equal “pie slice” compartments by dividers made of Plexiglas. One mouse per compartment provided sufficient space to move during irradiation. The chamber was placed on a turntable rotating at 6 revolutions per minute to ensure uniform radiation dose distribution. The average dose rate was 1.01 Gy per minute. A single dose of 0.5 Gy TBI was delivered and the mice were exposed in the morning to minimize possible diurnal effects for all radiation experiments.

2.4. Method of simulated microgravity

The NASA approved HLU model was used to simulate microgravity as described elsewhere with some modifications (Morey-Holton and Globus, 2002; Chowdhury et al., 2013; Chowdhury et al., 2016; Globus and Morey-Holton, 2016). Mice were briefly anesthetized (for 3–5 minutes) by isoflurane inhalation. Skin–Trac double–sided adhesive orthopedic foam (Cat No. 3874-02, Zimmer Inc., Warsaw, IN, USA) was attached on the dorsal and ventral sides of the tail without covering the tip and the sides of the tail to spare the lateral tail veins. The foam was wrapped with cotton gauge that allowed free air passage and the gauge was draped with another piece of plastic zip–reinforced strapping adhesive tape (Cat No. S-511, Uline, Coppell, TX). One end of the suspension string, which is made out of a bead chain swivel, was attached to the harness with the adhesive plastic tape and the other end was passed through a “split key ring.” The hind–limbs were suspended by passing the key ring onto a metal rod placed horizontally on top of the cage, allowing the animal free movement (Figure 3). The entire procedure took 6–10 minutes to complete. Previous studies using this harness demonstrated no influence on the animal’s ability to maintain a constant core body temperature and animals lost about 10–15% body weight initially, but they started to gain weight after the first 24–48 hours of suspension and were back to normal weight after 3 to 4 days (Chowdhury et al., 2013; Chowdhury et al., 2016).

Figure 3.

Relationship between the cumulative radiation dose (cGy) and the distance (cm) from the radiation source.

2.5. Tissue harvest and euthanasia

For tissue harvest, mice were anesthetized with an intraperitoneal injection of 10–12.5 mg Xylazine and 80–100 mg Ketamine per kg body weight (Henry Schein, Melville, NY, USA). The thymus and spleen were collected and placed in media–filled Eppendorf Safe–Lock microcentrifuge tubes (Thomas Scientific, Swedesboro, NJ) and were kept on ice. After tissue harvest, mice were euthanized by CO₂ asphyxiation followed by cervical dislocation to eliminate any reasonable doubt of survival.

2.6. Flow cytometry protocol

For thymus single cell suspension, tissues were minced with a surgical blade followed by enzymatic digestion in a buffer containing 0.2% collagenase I, 0.25% trypsin, 0.5% BSA in PBS for 5 minutes at room temperature. For spleen single cell suspension, tissues were pressed with the end of a plunger and gently ground against a 40 μm strainer to break apart the tissue fragments. The single cell suspensions were washed and transferred to 1 mL of PBS with 5% FBS. Red blood cells were lysed by adding 9 mL of RBC lysis buffer for 5 min at room temperature. After RBC lysis, the buffers were discarded. Cell numbers were counted using the Countess Automated Cell Counter (Invitrogen, Carlsbad, CA, USA) with trypan blue to discriminate the dead cells. We observed cell death in the range of 2 to 8%. Finally, 3 × 10⁶ live cells were stained with fluorophore tagged antibodies for flow cytometry analysis using the BD LSRFortessa cell analyzer (BD Biosciences).

2.7. Gating strategy for flow cytometry

A staining panel was designed to identify B cells (CD19+), T cells (CD3+), CD4 cells (CD3+/CD4+), NK-T cells (CD3+/CD335+), NK1 cells (CD27+/CD335−), NK2 cells (CD27−/CD335+), DCs (CD11c+/CD11b−), macrophages (Macro; F4/80+CD11b+), granulocytes (Gran; CD11c+Ly6Ghi) and Treg cells (CD25+). The amount of antibodies, the compensated matrix, and the gating strategy were established by running a single stained control, a fluorescence minus one control, and a cocktail control before starting the actual experiments. The instrumental dot plots data were analyzed in the following way: from the forward scatter vs side scatter plot, cell debris and aggregates were excluded, and single cells were identified. Using Texas red live–dead stain, the positively stained dead cells were excluded, and only live cells were used for further analysis. Live cells were then plotted on CD3 vs CD19 axes to identify B (CD3−/CD19+) and T (CD3+/CD19−) cells. The double negative cells (CD3−/CD19−), which do not belong to B or T cell populations, were then plotted on CD335 vs CD27 axes to identify NK1 (CD27+) and NK2 (CD335+) cells. The double negative cells (CD335−/CD27−) were then plotted on CD11b vs CD11c axes. The CD11c+ were identified as DCs, while the CD11b+/CD11c− population was plotted in the context of F4/80 vs Ly6G to identify macrophages (F4/80+CD11b+) and granulocytes, or more specifically, neutrophils (CD11c+Ly6G^hi). For further identification of the T cell subset, the T cells (CD3+/CD19−) population was plotted against CD4, CD335 and CD25 markers. The CD4+ cell population was then isolated and identified as CD4+ T cells, the CD335+ as NKT cells, and CD25+ as Treg cells. Figure 2 shows the gating strategy. Each immune cell type was calculated as a proportion of the total number of viable cells isolated from the spleen or thymus.

Figure 2.

Gating strategy used for the identification of major immune cell populations of lymphoid and myeloid origin in the thymus and spleen of C57BL/6 male mice.

2.8. Catecholamine ELISA

Mouse blood was collected retro-orbitally in BD Microtainer Blood collection tubes (Cat No. 365967) and serum isolated by centrifugation within 30 minutes after collection. QuickDetect™ Catecholamine (Mouse) ELISA Kit (BioVision; Cat No. E4462) was used to estimate the catecholamine level in the serum samples. This kit detects epinephrine, norepinephrine and dopamine. The assay was performed following the manufacturer’s instructions. Briefly, 10 μl of serum were taken and diluting 5 times by adding 40 μl of sample dilution buffer. Standards were prepared by serially diluted the supplied Standard (400 pg/ml) with standard dilution buffer. 50 μl of samples or standards were added in each well and incubated for 30 minutes at 37 °C. After washing with wash buffer, 100 μl HRP-conjugated secondary antibody was added and the plate was incubated for another 60 minutes. The plate was washed again and 100 μl of HRP-substrate was added for 15 minutes. After adding 50 μl of stop solution, the color intensities were measured in a plate reader at 450 nm.

2.9. Statistical analysis

For cell phenotypes, the outcome for a particular cell type was its proportion of the total number of live cells isolated from thymus or spleen (Figure 2B). For thymus and spleen, we analyzed each of the 10 cell types in an analysis of variance (ANOVA) that accounted for the 5 treatments and 2 observation days (a random effect). The main, simple, and interaction effects of SMG and CIR were tested with contrasts within the ANOVA context; these effects did not involve the mean of the AIR group. The interaction of SMG and CIR was a main comparison of interest. The other main comparison of interest was between the CIR and AIR groups; this, too, was tested with a contrast in the ANOVA. We checked normal assumptions with visual inspection of Q-Q plots and a panel of normality tests based on the residuals. In most instances, normal assumptions were reasonable; residuals from some outcomes, however, had skewed distributions. To determine whether the ANOVA inferences were robust for outcomes with assumption violations, we used a bootstrap method. Briefly, we first obtained standardized residual vectors for each of the 40 mice. (Note, each mouse provided 20 observations – 10 cell types from each of the 2 tissues.) Then, with replacement, we randomly drew 40 residual vectors, added back in the original group means, and analyzed each outcome in the aforementioned ANOVA. The point estimates for the comparisons of interest were saved. We repeated this 10,000 times to arrive at an empirical (bootstrapped) distribution of estimates. Finally, we used the 0.005 and 0.995 quantiles to produce bootstrapped 99% confidence intervals (CIs) and 0.025 and 0.975 quantiles for 95% CIs. All results from these analyses are in the Supplementary Material.

When determining whether body weight changes, catecholamine levels, food intake, and water intake differed among the 5 groups, we used a one–factor ANOVA described above. Because there was evidence that variances were not constant across groups, we allowed each group to have its own variance (i.e., Welch’s ANOVA). Similarly, for the catecholamine levels, we used the one-factor ANOVA described for the cell phenotypes. For these outcomes, we report pairwise differences of each experimental group to sham.

We conducted all analyses in SAS/STAT software, version 9.4, SAS System for Windows® (SAS Institute, Cary, NC, USA). The data and code are available in the Supplementary Material.

3. Results

3.1. Radiation Dosimetry Reveals a Correlation Exists Between the Radiation Source and Dose Uniformity

We first measured the radiation dose at different distances around the source (Cs¹³⁷ gamma source) to determine the dose uniformity. The distance of 1 meter from the CIR source was optimum with low variation of dose distribution over the size of the cage (Figure 3), therefore, we placed the animals 1 meter away from the source.

3.2. Effects of CIR on Certain Immune Cell Populations Depend on SMG in the Thymus, but not the Spleen

We equally randomized 40 mice into five groups: four combinations of CIR (sham-irradiated control and 0.5 Gy) and SMG (yes or no); and 0.5 Gy of AIR without SMG. After 1 month, we used multicolor flow cytometry to determine the proportion of total cells for these immune cell populations in the thymus and spleen. Analyzed cells included: B cells (CD19+), T cells (CD3+), CD4 cells (CD3+/CD4+), NK-T cells (CD3+/CD335+), NK1 cells (CD27+/CD335−), NK2 cells (CD27−/CD335+), DCs (CD11c+/CD11b−), macrophages (Macro; F4/80+CD11b+), granulocytes (Gran; CD11c+Ly6G^hi) and Treg cells (CD25+). The detailed gating strategy for identifyning the different types of immune cells is described in the Materials and Methods section.

The most significant changes in proportions of immune cell phenotypes were seen for Treg, NKT, NK1 and DCs in the thymus (Figure 4). A primary objective is to determine whether chronic irradiation effects are additive across normal loading and SMG; this is tested with CIR×SMG interaction. The interaction compares the change due to chronic irradiation in mice undergoing normal loading (i.e., CIR group vs Sham group) to chronic irradiation-induced changes in mice undergoing simulated microgravity (i.e., CIR+SMG group vs SMG group). Because the CIR vs CIR+SMG comparison is also of importance, in Figure 4 we reframe the interaction into the equivalent comparison of Sham vs SMG to CIR vs CIR+SMG.

Figure 4.

Mean proportions (± pooled SD) for thymic (A) Treg cells, (B) NK-T cells, (C), NK-1 cells, and (D) dendritic cells. All groups had 8 mice each (N = 40 total). Indicated comparisons were made within ANOVA context. Actual estimates and confidence intervals are in the Supplementary Material.

In the thymus, adding SMG to CIR decreased Treg, NK-T and NK1 cell proportions (Figure 4; Supplemental table 1 & 2). The magnitudes of these decreases were greater than SMG-induced changes under sham irradiation (i.e., Sham vs SMG); that is to say, the CIR×SMG interactions in these cell types were significant. Though not significant at the traditional 0.05 level in the ANOVA results, we also note that SMG had a small effect on the proportion of DCs in mice exposed to CIR (p=0.065). For those thymic cell types that did not have evidence of a CIR×SMG interaction (macrophage, T, B, NK2 cells), we note that CIR decreased B cell levels when averaged over SMG and no SMG mice (i.e., main effect of CIR). Simulated microgravity did not have any main effect on these 5 cell types. See the supplementary material (Supplemental table 1 & 2) for results not presented in Figure 4.

In the spleen, we did not find evidence of a CIR×SMG interaction. We did find that CIR increased B and CD4 cells and decreased NK1 and NK2 cells over sham irradiation; i.e., a main effect of chronic irradiation (Supplemental table 1 & 2). We also found that SMG increased granulocytes and decreased T cells from levels observed in mice in the normal loading conditions; i.e., a main effect of simulated microgravity (Supplemental table 1 & 2).

3.3. CIR Exerts Differential Effects on Immune Cell Populations in the Thymus and in the Spleen Compared to AIR of the Same Radiation Dose

The other primary objective is to compare whether CIR (under normal loading) affected immune cell populations differently than an acute exposure to radiation; i.e., the CIR vs AIR comparison. During deep space missions, astronauts will be primarily exposed to CIR and microgravity concurrently; with rare or no probability of AIR. However, to gain better insight in radiation dose-rate effects in altering immune cell phenotype we included AIR data in our current study. In the thymus, mice in the CIR group had lower proportions of DCs and higher proportions of Treg cells than those in the AIR group (Figures 4a & 4d; Supplemental table 1 & 2). In all other cell types in both tissues, there was little to no evidence that the AIR and CIR groups differed (Supplemental table 1 & 2).

3.4. Coexposure Results in Relatively Higher Decline in Body Weight, Food Consumption, and Water Intake than Singular exposure

We measured body weight, food consumption, and water intake twice a week until tissue harvest. Compared to sham (without SMG), mice experiencing combined injury (CIR with SMG) did not gain as much weight, eat as much food, or drink as much water (Figure 5A-5C; Supplemental table 3 & 4). Mice in the CIR alone group (CIR without SMG) and SMG alone group (sham with SMG) also did not gain as much weight as sham mice; but did not statistically differ from sham mice in their food and water intake (Table 2). Notably, we did not observe any differences in these parameters between the sham and AIR groups (Figure 5, Table 2).

Table 2.

Means of weight change, food intake, and water intake from the 5 treatment groups over the 1 month period. The differences (and 95% CIs) of each of the 4 active treatment groups from sham are also displayed. If 0 is not contained in the 95% CI, the group significantly differs from Sham at p < 0.05

Measure	Group	Mean	Difference from sham (95% CI)
Body weight gain (g)	Sham	4.3	Reference
	AIR	4.1	0.1	(−1.3, 1.6)
	CIR	2.0	2.3	(0.4, 4.1)
	SMG	1.3	2.9	(1.9, 3.9)
	CIR + SMG	−0.8	5.1	(4.2, 6.0)
Food intake (g)	Sham	117.9	Reference
	AIR	122.3	−4.4	(−21.2, 12.5)
	CIR	120.1	−2.3	(−21.4, 16.9)
	SMG	107.0	10.8	(−1.5, 23.2)
	CIR + SMG	86.8	31.1	(0.8, 61.3)
Water intake (mL)	Sham	119.4	Reference
	AIR	117.4	2.0	(−14.7, 18.7)
	CIR	111.1	8.3	(−10.2, 26.7)
	SMG	94.4	25.0	(−7.3, 57.4)
	CIR + SMG	93.2	26.2	(9.9, 42.5)

Figure 5.

Effects of CIR with and without SMG or AIR on body weight, food consumption, and water intake.

3.5. No Difference in Stress Level following Singular exposure or Coexposure

Catecholamines— a group of monoaminergic neurotransmitters, which include dopamine, noradrenaline, and adrenaline, are released by the sympathetic nervous system to mediate adaptive responses to acute stressors. We measured plasma catecholamine concentration to determine the stress level in mice one month after various challenges. Compared to the sham group, mice exposed to AIR and to CIR, regardless of SMG, experienced increased levels of plasma catecholamine. When comparing catecholamine levels among singular and co-exposures of CIR and SMG, we found no statistical or substantive differences, suggesting these various challenges elevated stress level to a similar extent (Figure 6; Supplemental table 5 & 6).

Figure 6.

Mean (± pooled SD) levels of plasma catecholamines. All groups had 8 mice each (N = 40 total). Indicated comparisons were made within ANOVA context. Actual estimates and confidence intervals are in the Supplementary Material.

4. Discussion

At the 2005 inception of the Human Research Program, NASA identified altered immune response as one of the most crucial risk factors during a space mission. Indeed, blood samples collected from astronauts after LEO missions demonstrated dysregulation of the immune system, which included suppression of functional activities of T lymphocytes and NK cells (Taylor and Dardano, 1983; Crucian et al., 2015; Bigley et al., 2019), a decline in the number of NK cells before and immediately after return to Earth (Bigley et al., 2019), and an elevation of serum pro–inflammatory cytokine levels (Crucian et al., 2013). Alterations in immune responses might modify host–microorganism interactions, change wound healing, and enhance allergic reactions. Notably, data on immune dysregulation after long-term space missions beyond LEO are not available.

To add to our current knowledge of the risk associated with space missions, we have recently developed a ground–based system that can be used to study long-term effects of CIR and SMG simultaneously. Lymphoid organs, such as the thymus and the spleen, exert numerous immune functions. A number of studies have demonstrated that the alteration in immune cell phenotypes occurs both in the thymus and the spleen after spaceflight or when subjected to ground–based models of the space environment (Miller et al., 1995; Ina and Sakai, 2005; Wang et al., 2007; Baqai et al., 2009; Gridley et al., 2013; Novoselova et al., 2015). For example, a significant decrease in thymic and/or splenic lymphocyte counts was observed in C57/BL6 mice flown on a 30–day space high–orbit satellite mission (i.e., BION-M1) (Novoselova et al., 2015). Moreover, Baqai et al. reported a decline in mouse splenic lymphocyte counts after 13 days of a space shuttle mission (STS-118) as compared to ground controls (Baqai et al., 2009). On the other hand, Gaignier et al. observed no change in mouse splenic lymphocytes after 21 days of HLU (Gaignier et al., 2014). Lastly, Wang et al. demonstrated 3 days of HLU differentially reduced lymphocyte populations in the thymus and the spleen (Wang et al., 2007). The authors observed significant lymphocyte reduction in the spleen, not in the thymus (Wang et al., 2007). Unfortunately, since these prior studies report their results as total cell counts in thymus and spleen, and we determined individual cell populations as proportions of the total number of viable cells in the thymus and spleen, we cannot make a direct comparison between the results of the prior studies and our current outcomes.

SMG is known to cause a decrease in food consumption and water intake as reported by various groups. This decrease in food and water intake could be due to the partial restraint from free movement and consequential anxiety because of HLU. However, the effects of coexposure of CIR with SMG on food and water intake are seldom reported, and data for long-term coexposure are not available. We observed SMG alone significantly decreased food and water intake, which is in line with previous publications (Chowdhury et al., 2013; Chowdhury et al., 2016). However, coexposure did not cause further decrease in food and water intake compared to SMG alone, suggesting these changes are a primary consequence of SMG. Decrease in food and water consumption is expected to negatively affect the body weight. Indeed, we observed a decrease in body weight in the SMG and the CIR with SMG groups, when compared to sham group. Importantly, space missions are associated with cumulative loss of bodyweight over time and a strong correlation exists between mission duration and weight loss (Matsumoto et al., 2011). Motion sickness has been identified as one of the major causes of weight loss during spaceflight. Similar to space studies, we also observed coexposure significantly reduced body weight when compared to singular treatment groups.

Importantly, CIR and AIR exert differential effects on alterations in immune cell phenotypes, body weight, food consumption, and water intake. Compared to AIR, CIR causes greater reduction in the proportion of thymic DCs, while AIR results in higher depletion of thymic Treg than CIR. Compared to AIR, CIR also causes greater reduction in body weight, food consumption, and water intake. These data suggest that the radiation dose rate has a significant effect on various biological parameters. These differences could be due to the variation in the number of “hits” (total radiation dose deposited) per cell, which again further depends on the turnover time of that particular cell. In the case of CIR, the deposition of events within the cell is much lower than AIR (Brooks et al., 2016). However, whether fewer deposition events reduce the risk of pathogenesis of various pathological conditions has yet to be established.

Finally, compared to sham, radiation (CIR or AIR) with or without SMG significantly elevated the stress level in mice as estimated by measuring plasma catecholamine concentration, however, no difference in the stress level was observed among exposure groups, suggesting coexposure did not cause any further increase in stress level than singular exposure. Catecholamine levels mainly indicate sympathetic activity during stress. For a more comprehensive analysis of stress due to HLU, future studies should add parameters such as circulating corticosterone levels or spleen and thymus mass.

The mechanistic basis of alterations of immune cell phenotypes in the thymus has not received much attention in this work, so it will be our subject for a future study. Additionally, we need to determine the influence of dose rate on the key events in crucial pathways to exert differential biological effects. The question of whether CIR results in a lower biological stochastic response (e.g., cancer development) and consequently, a lower risk than the same dose delivered at a high dose rate has remained a puzzle in the scientific community for the last two decades and needs to be answered. Finally, our CIR study using a gamma–radiation source does not exactly mimic the radiation in deep space, primarily composed of high–energy charged particles that deposit more energy per unit distance than gamma rays and display higher relative biological effectiveness (RBE) than gamma rays. The 0.5 Gy dose of gamma rays used in the current study will not be equivalent to 0.5 Gy of GCR. Therefore, this discrepancy in radiation quality should be taken into consideration before reaching to any conclusion based on the ground–based studies where gamma– or X–rays have been used.

In conclusion, using our recently developed CIR facility at University of Arkansas for Medical Sciences (UAMS), we demonstrated that the effect of CIR on immune cell phenotypes depend on whether load bearing is reduced (i.e., SMG) or not. Further, these CIR effects, in the presence or absence of SMG, depend on the organ; namely, the thymus is affected whereas no such effects were found in the spleen. Additionally, under normal loading, the effects of CIR are different from AIR to the same total dose with the biological parameters studied, signifying the contribution of radiation dose rate in modifying various biological endpoints. Finally, CIR plus SMG negatively impact body weight, food consumption, and water intake compared to sham control condition.