exposure to spaceflight produces many physiological changes, including those affecting immune function. For example, total body mass (12, 18,20), thymus and spleen mass (12, 37), circulating corticosterone (9, 24, 27, 47), mitogen-induced proliferation (9, 23, 24, 26, 32, 33, 40), and cytokine production and reactivity (6, 17, 23, 24, 29, 33, 41, 42) are all altered after spaceflight. The mechanisms for these changes are still largely unknown. This is partially due to the limitations and complexities of conducting spaceflight research. Currently, performing animal research aboard the United States Space Shuttle is a relatively rare event. Because of this, there are typically several principal investigators involved with each shuttle mission. This can severely restrict experimental design considerations for any single component. Therefore, it would be advantageous to create a ground-based protocol that can reproduce as many of the spaceflight conditions as possible.

The experiments presented here compare the immunological changes produced by an actual spaceflight with those of several ground-based models. The measure of the immune system taken was flow cytometric analysis of lymphocyte subpopulations. This measure was chosen because the immune response is dependent upon a balance of various immune cells, and changes in population distributions within immunologically competent tissues may impair the ability of the immune system to respond to a pathogenic challenge. We focused specifically on splenic subpopulations because the spleen plays such an important role in lymphocyte recirculation and immune response, and even a small change in splenocyte distribution may reflect a large change in overall immune efficacy. The ground-based models of spaceflight examined were antiorthostatic tail suspension and centrifugation. Antiorthostatic tail suspension was used to simulate several environmental and physiological conditions inherent to all spaceflight experiments involving animals. These include exposure to a novel environment, unloading of the limbs, and cephalic fluid distribution shifts (30, 31, 43). A 1.8-m-diameter centrifuge was used to reproduce launch and landing loads similar to those experienced during a typical spaceflight [2–3 g (gravitational acceleration of the surface of the earth) for 45 min].

Several researchers have investigated the effect of spaceflight on splenocytes and found that total splenocyte counts either decrease or remain the same after spaceflight (3, 41, 42). Some of the more common leukocyte populations investigated have been total T cells, CD4⁺ helper/inducer T cells, CD8⁺ cytotoxic T cells, neutrophils, and macrophages. Depending on the flight, total T cell (W3/13), CD4⁺ (W3/25), and CD8⁺ (OX-8) percentages in the spleen have been reported to increase or remain the same after spaceflight (3, 19, 41, 42). Splenic CD11b⁺ neutrophils and macrophages have been not been examined previously. However, in peripheral blood, neutrophils have been found to increase, whereas monocytes remained constant, after spaceflight (19).

One limitation of these previous spaceflight experiments is that lymphocyte subpopulations were labeled for flow cytometric analysis using a single antibody technique. Because some antibody labels can appear on multiple cell types, analysis resolution is reduced. For example, antibodies against the CD4 surface receptor will appear on both helper/inducer T cells and macrophages in the rat (22). To avoid this limitation, the experiments presented here employed two-color flow cytometry. This approach will help clarify the effect of spaceflight on specific lymphocyte populations.

Ground studies involving lymphocyte populations have also been limited somewhat. Previous investigations involving antiorthostatic tail suspension indicate that the model had no significant effect on T cell, CD4⁺, and CD8⁺ splenocyte distributions (32, 42). However, these studies were limited by the single-label flow cytometric technique already discussed. The effect of centrifugation alone, or centrifugation combined with antiorthostatic tail suspension, on these splenocyte cell populations has not been investigated previously.

MATERIALS AND METHODS

Animals

Two-month-old (starting mass 180–200 g), pathogen-free, male, Harlan Sprague-Dawley rats were used in these studies. Food and water were available ad libitum. Care and use of the animals was in accordance with protocols approved by the University of Colorado Institutional Animal Care and Use Committee.

Procedures

Spaceflight.

Rats (n = 12) flew on the Space Shuttle Endeavor (STS-77) in May of 1996. Six animals were housed in each of two Animal Enclosure Modules (AEMs, supplied by the National Aeronautics and Space Administration) stored in the shuttle middeck for the 10-day mission. Ground controls (n = 16) were kept in the Animal Handling Facilities at Kennedy Space Center. These animals were group housed (n = 4/cage). During the flight, rats had access to water and food ad libitum and were on a 12:12-h light-dark cycle. The AEM temperature averaged 28.6 ± 1.0°C and ranged from 25.6 to 31.3°C. Flight rats were weighed and killed (5 mg/kg xylazine and 75 mg/kg ketamine ip) ∼3 h after the shuttle landed. Estimated launch and landing g loads were 8–10 min at 3 g and 10–15 min at 1–2 g, respectively.

As is common to most spaceflight studies, other experimental questions were being considered in these same animals. Four days before launch, all rats were surgically implanted (subcutaneously in the dorsal surface) with 2-ml Alzet osmotic minipumps (5) containing either insulin-like growth factor I or saline. The osmotic minipump does not pump out fluid, rather, an exchange of fluid occurs with the internal pump solution and extracellular fluid. Therefore, there is no net increase in fluid in pump-implanted animals. Normal immune responses or body weight gains were not effected by minipump implantation (9). One day before launch, all animals were weighed and received intramuscular oxytetracycline injections (20 mg/kg in a saline vehicle). It has been reported previously that these flight animals had elevated serum corticosterone levels and elevated proinflammatory responses [tumor necrosis factor (TNF), interleukin (IL)-6, and nitric oxide secretion; see Ref. 9]. Only data from the saline spaceflight rats (n = 6) are presented in the current paper.

Ground-based simulations.

antiorthostatic tail suspension.

As described in the Morey-Holton et al. (30, 31) suspension protocol, animals were suspended at roughly 30° from horizontal so that their hindlimbs could not be used to support weight. Rat cages (30.4 × 30.4 × 29.2 cm) were designed with a pulley system that allowed the animals to move about the cage on their front paws and have easy access to food and water.

CENTRIFUGATION.

A 1.8-m-diameter centrifuge was used to simulate the launch and landing loads of a shuttle mission. Rats were placed in hardware with an animal-to-surface area ratio similar to that flown on the shuttle. This “Mock AEM” was slightly larger than the flight model so that more animals could be placed in the centrifuge at the same time. Rats were exposed to launch and/or landing loads slightly greater than those of actual shuttle missions: 2-min ramp up to 3 g, 13 min at 3 g, 2-min ramp down to 2 g, 26 min at 2 g, and a 2-min ramp down to nominal g. The centrifuge groups were chosen to allow investigation of the acute effects of landing (2–3 g), the long-term effects of a launch alone, and the possible synergy between launch and landing exposure. To control for day-to-day variability of cell labeling and flow cytometry, control rats were killed, and their cells were assayed on the same day as their respective experimental group.

Rats were exposed to one of the following ground-based simulations.

GROUND-BASED SIMULATION 1: suspension.

The fist suspension protocol consisted of 10 days of antiorthostatic tail suspension (n = 8 and n = 4 time-matched controls; Suspension); the second protocol consisted of launch loads plus 10 days of antiorthostatic tail suspension and landing loads (n = 8 and n = 4 time-matched controls; L + Susp + L).

GROUND-BASED SIMULATION 2: centrifugation.

The centrifugation phase consisted of the following three protocols:1) landing loads only (n = 7 andn = 4 time-matched controls; Landing); 2) launch loads plus 10 days of normal housing (n = 7 andn = 4 time-matched controls; L + 10); and3) launch loads plus 10 days of normal housing and landing loads (n = 7 and n = 5 time-matched controls; L + 10 + L).

Cell Preparation and Flow Cytometry

All animals were killed by brief ether exposure and cervical dislocation. Rats experiencing landing loads were killed 3 h after termination of centrifugation. This delay simulated the time between shuttle landing and animal transport.

After the rats were killed, immune organs (thymus, spleen) and adrenals were removed, weighed, and placed in cold Hanks' balanced salt solution (HBSS; GIBCO). The spleen was then dissociated using a modified tissue homogenizer (Bellco Glass). Cell concentrations and total cell yields were assessed using either a hemocytometer or the coulter counter.

The splenocytes were then divided into two aliquots of cells (2.0 × 10⁶ cells/aliquot). One aliquot was incubated with only FITC-conjugated anti-CD11b for 30 min at 30°C (100 μl at 1:20 in HBSS, WT.5; Pharmingen). The second aliquot of cells was incubated with phycoerythrin-conjugated anti-αβ-T cell receptor (TCR) (100 μl at 1:20 in HBSS, R73, mouse anti-rat IgG₁; Pharmingen) and FITC-conjugated anti-CD4 (100 μl at 1:50 in HBSS, OX38, mouse anti-rat IgG_2a; Pharmingen). Cells were washed one time and resuspended in 500 μl HBSS. To fix the cells, an additional 50 μl 37.6% formaldehyde solution were added to the suspension and incubated at 25°C for 15 min. The cells were then washed and resuspended in 1.0 ml HBSS for fluorescence-activated cell-sorter (FACS) analysis (Becton-Dickinson FACSCAN cell sorter with CellQuest data acquisition software).

Flow Cytometric Analyses

For the samples labeled with αβ-TCR and CD4, fore scatter vs. αβ-TCR dot plots were used to choose splenocyte populations. Two gates were used. In the first, all cells were bitmapped, excluding background (debris, etc., within the first decade of both αβ-TCR and Fore Scatter histograms). This gate was used to include the total splenocyte population. In the second, the largest concentration of αβ-TCR⁺ populations within the first gate was gated and bitmapped. This gate was used to differentiate the splenic T cell populations. Single-parameter (CD4) histograms were used to determine CD4⁺ and CD4⁻ populations within the T cell population gate. Both cell populations were normalized to the total splenocyte populations of the first gate.

For the samples labeled with CD11b, fore scatter vs. side scatter dot plots were used to gate the total counted splenocyte population. Single-parameter (CD11b) histograms were used to determine CD11b⁺ and CD11b⁻ populations within the total splenocyte population.

Statistical Analyses

Thymus mass was normalized to body weight using the following equation: wet thymus mass/total body mass. Results from the flight experiment were analyzed (StatView) using a two-group ANOVA (flight vs. vivarium control) and Fisher's protected least significant difference (PLSD) post hoc two-group comparison. Ground-based simulation experiment 1 was analyzed using a two (Suspension vs. L + Susp + L) times two (simulation vs. control) ANOVA and Fisher's PLSD post hoc two-group comparisons. Ground-based simulation experiment 2 was analyzed using a three (Landing vs. L + 10 vs. L + 10 + L) times two (simulation vs. control) ANOVA and Fisher's PLSD post hoc two-group comparisons. A reliable statistical difference was defined at the level of P < 0.05. Figures 1-3 show means ± SE.

RESULTS

Spaceflight

As depicted in Table 1, there was a significant spaceflight-induced increase in the total body mass (P < 0.001) and thymus wet mass (P < 0.05) compared with their respective vivarium ground control values. The increase in thymus wet mass was no longer statistically significant when normalized to total body mass at death (P = 0.07).

Spaceflight produced several significant changes in the percentage of splenocyte subpopulations. As depicted in Fig.1, spaceflight reduced the total percentage of T cells (TCR⁺) found in the spleen (P < 0.01). Similarly, the percentage of splenic helper/inducer T cells (TCR⁺/CD4⁺; see Ref.22) was reduced significantly after spaceflight (P < 0.001). Although there was a trend for a decrease in splenic TCR⁺/CD4⁻ cytotoxic T cell percentages after spaceflight, this difference was not significant (P = 0.08). The CD11b⁺ neutrophil and macrophage population (44) was significantly reduced (P < 0.05) after exposure to spaceflight (Fig. 1).

Total splenocyte counts significantly increased (P < 0.05) after spaceflight (spaceflight, mean = 27.8 + 4.1 × 10⁶ cells/ml of media vs. vivarium ground control, mean = 17.1 + 2.9 × 10⁶ cells/ml of media). There were, however, no significant differences in any of the individual phenotype cell counts examined (data not shown).

Ground-Based Simulation 1: Suspension

As shown in Table 1, both the Suspension and L + Susp + L groups had a significant reduction in total body mass compared with home cage controls (P < 0.0001). Although thymus wet mass did not change, normalized thymus mass reliably increased after both suspension conditions (P < 0.05).

Suspension and L + Susp + L produced very few changes in splenocyte subpopulations. There were no significant differences in total splenocyte counts (data not shown) or in any of the subpopulation counts and percentages after either suspension condition (Fig.2, A and B).

The percentage of CD11b⁺ cells, however, did change slightly. There was a reliable interaction of Suspension vs. L + Susp + L simulation in the percent of CD11b⁺ cells (P < 0.05).

Ground-Based Simulation 2: Centrifugation

As shown in Table 1, there were no significant differences in total body mass or thymus mass after any of the centrifugation conditions. There was a significant overall reduction in normalized thymus mass (P < 0.05) after exposure to the centrifugation models. Fisher's PLSD post hoc comparisons revealed significant decreases in normalized thymus mass after the Landing and L + 10 + L conditions.

There were no significant changes in total splenocyte count (data not shown). Similarly, there were no differences seen in the percentages of total T cell (TCR⁺), helper/inducer T cell (TCR⁺/CD4⁺), cytotoxic T cell (TCR⁺/CD8⁻), or neutrophil/macrophage (CD11b⁺) subpopulations after any centrifugation manipulation (Fig. 3, A-C). In contrast, Landing alone resulted in a significant increase in splenic T cell (TCR⁺) and helper/inducer T cell (TCR⁺/CD4⁺) counts when compared with time-matched controls (Fig. 3D; P < 0.05).

DISCUSSION

Rats exposed to spaceflight aboard the Space Shuttle Endeavor (STS-77) had increases in total body and thymus masses compared with vivarium ground controls; however, there was no effect of spaceflight if thymus mass was normalized to body mass (thymus mass/%body mass). Reductions in total body mass and thymus mass (due to chronically elevated glucocorticoids) are classic indicators of chronic stress (38, 39). Because spaceflight rats failed to show a decrease in body and/or thymus mass, it is likely that rats exposed to 10 days of spaceflight were not chronically stressed.

At first glance, the significant increase in total body mass in rats flown on STS-77 seems to be inconsistent with results from past flights. In fact, the impact of spaceflight on body mass is variable. Depending on the flight, body mass has either increased (9), significantly decreased (12, 18, 20), or remained the same (2, 9, 10, 46, 47). Thus, given the variability of this effect, a significant increase in total body mass over vivarium controls seen after STS-77 is not extraordinary compared with previous flights with similar control and landing conditions.

The flight-induced increase in thymus wet mass found after STS-77 also appears to be inconsistent with results from previous spaceflight experiments. Thymus wet masses or nomalized thymus wet mass from animals of past flights have either decreased (12, 20) or remained constant (9, 10, 17). However, given that STS-77 did not alter normalized thymus wet mass and that there are inconsistencies in the literature due to differences in launch vehicle and/or controls, these results are not unexpected.

The experience of spaceflight resulted in large and significant changes in splenocyte subpopulations (Fig. 1). There was a decrease in the percentage of total splenic T cells (TCR⁺) in the flight animals compared with vivarium ground controls. This was due primarily to a reduction in the helper/inducer T cell (TCR⁺/CD4⁺) subset. Because helper/inducer T cells are important for many aspects of adaptive immunity, a large chronic decrease in this population would allow a pathogen to spread more aggressively than it otherwise would on the ground. A similar reduction in TCR⁺ T cells and CD4⁺ T cells has been previously reported after exposure to acute stress (4,14), suggesting this decrease could be due to an acute response to launch and/or landing loads rather than a chronic response to microgravity. Spaceflight also resulted in a decrease in the CD11b⁺ neutrophil/macrophage percentage (Fig. 1). In contrast, the percentage of cytotoxic T cells (TCR⁺/CD4⁻) did not reliably change after spaceflight. Interestingly, the large changes in the percentages of splenic lymphocyte populations were not reflected in changes in splenic cell numbers. This could suggest that the decreases in percentages of T cells, helper/inducer T cells, and CD11b⁺ cells were due to an increase in some other unlabeled cell population (e.g., B cells). However, it is more likely that the variability in cell counts made it difficult to detect reliable changes.

The effect of spaceflight on splenic subpopulation distributions is equally unclear. Depending on the flight, splenic total T cell (W3/13), CD4⁺ (W3/25) cell, and CD8⁺ (OX-8) cell percentages have all been reported to increase or remain the same after spaceflight (3, 19, 41, 42). As with body and thymus masses, these inconsistencies are probably due to the launch vehicle, flight hardware, and/or controls. The flights that resulted in cell percentage increases were both COSMOS missions. The stress inherent to the harsh landing conditions of these flights may have contributed to the reported increases in cell populations (41, 42). Similarly, no change in cell percentages was detected in shuttle flights that used Rodent Animal Handling Facilities for both spaceflight and ground control animals (3, 19). These differences in flight conditions, combined with the single-label flow cytometric analysis used on previous flights (as discussed previously) may account for differences in the results across studies.

There are at least three possible explanations for the spaceflight-induced shifts in phenotype percentages reported here. They are the following: a shift in white blood cell hematopoiesis; an increase in systemic tissue damage; and a change in splenocyte homing patterns. Spaceflight results directly related to hematopoiesis have been limited and contradictory. For example, the ability of femoral bone marrow cultures to respond to macrophage colony-stimulating factor (M-CSF) or granulocyte/monocyte-colony stimulating factor has either decreased (41, 42) or remained the same (9,40) after previous flights. After the IMMUNE.03 flight, there was actually an increase in the response to M-CSF (S. Chapes, personal communication). To date, there has been no conclusive evidence for a consistent change in white blood cell hematopoietic mechanisms.

Similarly, there is little evidence to suggest animals flown aboard the space shuttle would experience an increase in systemic tissue damage (i.e., due to musculoskeletal atrophy or radiation exposure), causing a shift in lymphocyte population requirements. However, due to the hazardous nature of the spaceflight environment, this possibility cannot be ignored. Radiation exposure aboard most Shuttle flights is minimal and therefore is not a probable etiological factor for the immune changes reported (16, 45).

Because spaceflight did produce changes in cell population counts, a change in splenic homing patterns is perhaps the most likely possibility of the three hypotheses suggested. There are several potential mechanisms that could be responsible for spaceflight-induced changes in splenic homing patterns. For example, the production of cytokines (IL-1, -2, -3, and -6, TNF-α and -β, and interferons-α and -γ) is significantly altered in mitogen-stimulated splenocyte cultures from flight animals (17, 23, 24, 29, 33). Several of these cytokines are also known to regulate lymphocyte homing patterns (1, 21, 34, 36). In addition, hormonal changes produced by spaceflight could also alter lymphocyte homing patterns. For example, plasma corticosterone levels have been shown to increase after spaceflight (9, 24, 27, 43, 47). This stress hormone is known to moderate lymphocyte homing by acting either directly on endothelial homing receptors (7, 11, 35) or indirectly via changes in cytokine production (8, 11, 15, 25). Last, a change in lymphocyte homing receptors could be responsible. Animals flown in microgravity have an increased expression of lymphocyte function-associated antigen (LFA)-A and LFA-B in peripheral blood lymphocytes. There is also a decrease in reactivity of antibodies against two L-selectin clones. In contrast, there were no significant differences in antibody reactivity to intracellular adhesion molecule-1 (19). Although these epithelial adhesion receptors are probably not directly involved with homing to the spleen, long-term changes in homing to the lymph nodes and other lymphoid organs may lead to systemic changes in splenic lymphocyte subpopulations. The current experiments cannot address which of these mechanisms are responsible for the changes in splenic subpopulations found after spaceflight.

The ground-based models tested here failed to reproduce the changes in body mass or thymus mass generated by spaceflight. Of all of the models tested, the L + Susp + L manipulation most closely simulated the spaceflight conditions. Animals in this group experienced the initial g forces of launch, unloading of the limbs, and cephalic fluid shifts of simulated microgravity and the final g forces of landing. In addition, animals in this condition did not experience Coriolis forces that may result in symptoms of motion sickness (i.e., vertigo, disorientation, or nausea; see Ref. 13). However, contrary to what occurred in the spaceflight animals, body mass actually decreased, and thymus mass remained unchanged. Because exposure to any of the centrifugation conditions alone (without suspension) had minimal effect on body mass and thymus mass, the decreases seen in L + Susp + L body mass were most likely due to suspension alone. This is verified by the Suspension results presented here and elsewhere (28, 30,32). The slight overall decrease in normalized thymus mass seen after centrifugation suggests a possible stress response to the launch and landing loads. However, this change does not reflect what is seen in animals flown aboard the space shuttle.

Exposure to the ground-based models also failed to reproduce the changes seen in splenocyte subpopulations after spaceflight. The Landing alone condition produced a reliable change in T cell numbers (Fig. 3B); however, this change was in the opposite direction to that of spaceflight. The spaceflight-induced changes in T cell, helper/inducer T cell, and neutrophil/macrophage percentages were not reproduced by any of these models. Interestingly, although not significant, there were trends for decreases in T cell and helper/inducer T cell percentages in the L + Susp + L group (Fig.2B). This is similar to what was seen after spaceflight. Perhaps with additional modifications, this combination of centrifugation and suspension could be successfully used as a ground-based model.

In conclusion, the ground-based models presented here failed to induce the same immunological changes produced by spaceflight. It is important to note, however, that changes in splenic subpopulations are just one of many immunological consequences of spaceflight. Other effects, such as changes in mitogen-induced proliferation and cytokine production, may be more easily reproduced using tail suspension and centrifugation. In addition, there are a number of possibly critical differences between these models and spaceflight conditions. The partial unloading modeled in tail suspension is not the same as the free-floating, full-body unloading of microgravity. Hence, the animals may adapt less readily to tail suspension than to the completely novel environment of microgravity. There are also unique factors inherent to the launch vehicle and space environment not simulated by these models. These include the noise and vibrations due to surrounding hardware, astronaut movement, and shuttle positioning thrusters. In any case, the large spaceflight-induced shifts in splenic lymphocyte population distributions will be difficult to effectively simulate on the ground using the current methods.

FOOTNOTES

• Financial support for these studies was provided by the Chiron Corporation of Emeryville, CA, and by National Aeronautics and Space Administration Grant NAGW-1197.

• Address for reprint requests and other correspondence: M. Fleshner, Dept. of Kinesiology and Applied Physiology-CB#354, Univ. of Colorado-Boulder, Boulder, CO 80309-0354 (E-mail:[email protected]colorado.edu).

• The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.