Short day lengths augment stress-induced leukocyte trafficking and stress-induced enhancement of skin immune function

Abstract

Environmental conditions influence the onset and severity of infection and disease. Stressful conditions during winter may weaken immune function and further compromise survival by means of hypothermia, starvation, or shock. To test the hypothesis that animals may use photoperiod to anticipate the onset of seasonal stressors and adjust immune function, we evaluated glucocorticoids and the distribution of blood leukocytes in Siberian hamsters (Phodopus sungorus) exposed to long day lengths (i.e., summer) or short day (SD) lengths (i.e., winter) at baseline and during acute stress. We also investigated the influence of photoperiod and acute stress on a delayed-type hypersensitivity response in the skin. SDs increased glucocorticoid concentrations and the absolute number of circulating blood leukocytes, lymphocytes, T cells, and natural killer cells at baseline in hamsters. During stressful challenges, it appears beneficial for immune cells to exit the blood and move to primary immune defense areas such as the skin, in preparation for potential injury or infection. Acute (2 h) restraint stress induced trafficking of lymphocytes and monocytes out of the blood. This trafficking occurred more rapidly in SDs compared to long days. Baseline delayed-type hypersensitivity responses were enhanced during SDs; this effect was augmented by acute stress and likely reflected more rapid redistribution of leukocytes out of the blood and into the skin. These results suggest that photoperiod may provide a useful cue by which stressors in the environment may be anticipated to adjust the repertoire of available immune cells and increase survival likelihood.

Environmental stressors, such as temperature, and the availability of food, water, and shelter often vary on a seasonal basis. Direct links exist among stressful environmental conditions, disease, and death (1–3). For many nontropical rodents, winter is often more stressful than summer. High thermoregulatory demands for small mammals, in particular, during the relatively cold winter months typically coincide with low food availability. The energetic bottleneck faced by animals during winter has led to the evolution of specific adaptations that allow these animals to conserve energy and cope with winter successfully (4, 5). Siberian hamsters (Phodopus sungorus) stop breeding during the winter by responding to photoperiodic cues that allow individuals to anticipate challenging conditions and prepare accordingly for them. Hamsters also undergo significant reductions of white adipose tissue and body mass (≈25%) during the winter or when housed in short days (SDs) in the laboratory, despite ad libitum access to food and mild temperatures (6, 7). This strategy presumably evolved because maintenance of a smaller body size throughout winter requires less food and increases thermogenic potential, resulting in daily energetic savings (7).

Seasonal adjustments in immune function also may be critical for winter survival. Immune function is compromised in many species during the winter in the wild but is enhanced in the laboratory during SD conditions when all other factors, such as food and temperature, are held constant (4, 8, 9). Stress often suppresses immune function and increases susceptibility to infection and disease (10, 11). However, the notion of suppressed immune function under all stress conditions may be overly simplistic. Stress is an intricate part of life and dealing successfully with stressors prolongs survival (12). In contrast to chronic stress, which is immunosuppressive (13), acute stress in rats and mice appears to enhance skin immune function at a time when injury is likely. Short duration (≈2 h) restraint stress significantly enhances delayed-type hypersensitivity (DTH) responses in the skin by inducing a rapid redeployment of immune cells out of the blood and into the skin (12, 14). Primary immune defense areas, such as the skin, lymph nodes, lungs, and gastrointestinal tract, represent the first line of bodily defense, and infection or injury is most likely to occur first in these peripheral areas (15). Glucocorticoids mediate the trafficking of leukocytes out of the blood and among tissues during stress, whereas adrenalectomy prevents stress-induced trafficking and enhancement of DTH reactions (12–14, 16). In the absence of injury or infection, leukocytes rapidly return to previous baseline levels within the blood. After antigenic challenge, the DTH immune reaction is characterized by inflammation at the site of challenge by an infiltration of monocytes and lymphocytes into the epidermis and dermis, and the physiological role of this system is to provide front-line defense against pathogens (17). A positive correlation exists between the intensity of the swelling and the immune reaction (18), and this model is a standard in vivo measure of cell-mediated immunity (12, 19).

Because the onset of winter stressors is predictable, Siberian hamsters may use day length to anticipate their onset and adjust immune function. Evidence that photoperiod can regulate important immune cell functions in Siberian hamsters (20) led us to test the hypothesis that photoperiod may influence immune cell numbers in the blood, as well as the trafficking of immune cells in response to stress. We hypothesized that SD hamsters may have increased numbers of immune cells in preparation for seasonal stressors. Furthermore, if stress-induced leukocyte trafficking enhances immune function in preparation for potential injury, then redeployment of immune cells out of the blood should be more efficient, and skin immune function enhanced, in SDs compared to long days (LDs).

Materials and Methods

Animals.

Adult (4–6 mo of age) male Siberian hamsters (P. sungorus) from our breeding colony were used in this study. Animals were housed in sibling pairs in polypropylene cages (27.8 × 7.5 × 13 cm) in colony rooms with constant temperature and humidity of 21 ± 4°C and 50 ± 10%, respectively, and had constant access to food (Harlan Teklad 8640 rodent diet, Indianapolis) and filtered tap water. Twenty-five hamsters were housed in LD conditions with a reverse 15:9 light/dark cycle (lights on 2400 h Eastern standard time, EST), and 25 hamsters were housed in SD conditions with a reverse 9:15 light/dark cycle (lights on 0600 h EST). All hamsters remained within their respective photoperiod conditions for 10 wk before the start of the experiments.

Leukocyte Analysis.

Five LD and five SD hamsters were lightly anesthetized with isoflurane vapors (Abbott) and bled from the retro-orbital sinus (0.25 ml) into heparinized tubes at 0900 h EST. Handling was kept consistent and time to a minimum (<2 min), and animals were quickly returned to their cages in colony rooms. Total leukocyte numbers and lymphocyte-neutrophil differentials were obtained on a hematology analyzer (F800, Sysmex, McGraw Park, IL). Specific leukocyte subtypes were measured by immunofluorescent Ab staining and analysis by using single color flow cytometry (FACSCalibur, Becton Dickinson). Lymphocyte, neutrophil, and monocyte subpopulations were identified and gated by using forward- vs. side-scatter characteristics. T cells were identified by using CyChr-labeled anti-CD3 (clone 145–2C11), B cells by using phycoerythrin-labeled anti-B220 (RA3–6B2), and natural killer (NK) cells by using phycoerythrin-labeled anti-NK1.1 (PK136). Neutrophils and monocytes were identified by using forward- vs. side-scatter patterns and allophycocyanin-labeled anti-CD11b (M1/70). l-selectin and CD44-positive cells were identified by using phosphatidylethanolamine-labeled anti-CD62L (MEL-14) and FITC-labeled anti-CD44 (IM7), respectively. Each staining panel consisted of a single Ab. All monoclonals were directly conjugated, rat anti-mouse Abs (except for NK1.1, which is a murine Ab) and were obtained from Becton Dickinson-PharMingen. Because Abs directed against hamster immune cells were not available, we adapted crossreactive anti-mouse Abs for these analyses. Ab clones from different suppliers were screened. Staining patterns and relative percentages observed with crossreactive clones were similar to those observed in mice whereas noncross reactive clones showed no positive staining. Relative percentages of independently run T, B, and NK cells panels accounted for >95% of positively stained cells in the lymphocyte gate. In brief, blood samples were incubated with Ab for 20 min at room temperature, washed with PBS, and read on the FACSCalibur; 3,000–5,000 events were acquired from each preparation. Control samples matched for each fluorochrome and each Ab isotype were used to set negative staining criteria. Data were analyzed by using cellquest software (Becton Dickinson).

Restraint Stress and Leukocyte Redeployment.

After a 10-day recovery period (after baseline samples), animals were placed individually into well ventilated, Plexiglas, restraint tubes at 0900 h EST. Animals were prevented from significant movement but not compressed or squeezed. This procedure has been used to approximate a stressful experience that is largely psychological in nature because of the response to confinement on the part of the animal (21). At time points 0 min (baseline), 10 min, and 120 min, animals were lightly anesthetized and bled from the retro-orbital sinus (0.25 ml) into heparinized tubes. Handling was kept consistent and time to a minimum (<2 min) for each sampling. Animals were quickly returned to restraint chambers immediately after 0- and 10-min blood sampling and returned to their cages after sampling at 120 min. Total leukocyte numbers and lymphocyte-neutrophil differentials were obtained as before from all blood samples on a hematology analyzer, and specific leukocyte subtypes were measured and identified by using the same markers and Abs as described above.

RIA Procedures.

After FACS analysis for baseline and stress procedures, remaining blood aliquots (≈200 μl) were centrifuged at 4°C for 30 min at 2,500 rpm, and the supernatant was collected and frozen at −70°C until assayed for cortisol and testosterone by RIA. Plasma cortisol concentrations were determined in a single assay by using a Diagnostic Products (Los Angeles) ¹²⁵I double Ab kit (Inter-Medico, Markham, ON). Cortisol represents the primary glucocorticoid in this species (22). Plasma testosterone concentrations were determined in a single assay by using a Diagnostic Systems Laboratories ¹²⁵I double Ab kit (Webster, TX). These kits have previously been validated for use in Siberian hamsters (22, 23). Average assay sensitivities for cortisol and testosterone were 10.17 ng/ml and 0.31 ng/ml, respectively. The intra-assay coefficients of variation were <10% in both cases.

Induction of DTH.

DTH was induced by application of the antigen, 2,4-dinitro-1-flourobenzene (DNFB; Sigma), to the pinnae of a separate group of previously unmanipulated hamsters after initial immunization to the dorsum. On day 1, 20 LD and 20 SD naïve hamsters were anesthetized with isoflurane vapor, and an area of ≈2 × 3 cm was shaved on the dorsum. Twenty-five microliters of DNFB [0.5% (wt/vol) in 4:1, acetone/olive oil vehicle] was applied to the shaved skin on days 1 and 2. The thickness of both pinnae was measured on day 1 before sensitization by using a constant-loading dial micrometer (Mitutoyo, Tokyo). On day 5, baseline pinnae thickness was again measured. On day 6, 10 LD and 10 SD hamsters were restrained as described above for 2 h. Ten LD and 10 SD hamsters remained undisturbed in home cages for 2 h. Immediately after restraint stress or control procedures, all animals were lightly anesthetized with isoflurane, and 20 μl of DNFB [0.2% (wt/vol) in 4:1, acetone/olive oil] was applied to the skin of the dorsal surface of the right pinna. Left pinnae were treated with vehicle. Pinna thickness was measured every 24 h for the next 6 days at 1000 h EST, and all measurements were made on the same relative region of the pinna.

Data Analysis and Statistics.

At the conclusion of all procedures, all hamsters were killed by rapid cervical dislocation. Paired testes were removed, cleaned of connective tissue and fat, and weighed. Body and testis masses, and baseline leukocyte and hormone data were analyzed between groups by using two-tailed t tests. To control for the possibility that cell numbers changed as a function of reduced body mass, and thus blood volume, in SD hamsters, we analyzed each cell type as a percentage of body mass and compared between groups by using two-tailed t tests. Hormone concentrations and leukocytes were analyzed between groups as a function of time during stress by using repeated measures ANOVA. DTH reactions were analyzed as percentage increases in pinna thickness over baseline for each animal and compared between groups by using repeated measures. Posthoc (Tukey's honestly significant difference) tests were performed to further distinguish among groups, and all differences were considered statistically significant if P < 0.05.

Results

Baseline Leukocyte Distribution.

Overall, SD hamsters (n = 25) weighed significantly less and had significantly regressed testes compared to LD hamsters (n = 25) after 10 wk of photoperiod (P < 0.001, for both; Fig. 1). Total white blood cells and lymphocyte counts were higher in SD than LD hamsters (n = 5, both photoperiod groups) (P < 0.01, for both; Fig. 2). Monocyte and neutrophil cell numbers did not differ between groups. Specific T cell and NK cell lymphocyte subsets also were significantly higher in SD compared to LD (P < 0.01 and P < 0.001, respectively) (Fig. 3), whereas B cells did not differ between groups (P > 0.05). Total hematocrit did not differ between groups (LD = 43.38 ± 2.01; SD = 43.14 ± 0.52; P > 0.05). Reanalysis of data when corrected as percentage of body mass did not change the statistical significance of any measure (data not shown).

Figure 1.

Mean (± SEM) body mass (g) (A) and paired testes mass (mg) (B) of LD (n = 25) vs. SD (n = 25) hamsters; *, P < 0.05.

Figure 2.

Mean (± SEM) baseline blood leukocyte numbers in LD (n = 5) vs. SD (n = 5) hamsters. White blood cells (WBC), lymphocytes (LYM), monocytes (MONO), and neutrophils (NEU); *, P < 0.05.

Figure 3.

Mean (± SEM) baseline blood lymphocyte subtype numbers in LD (n = 5) vs. SD (n = 5) hamsters; *, P < 0.05.

Hormones.

Cortisol concentrations were significantly higher in SD than LD hamsters (n = 5, both groups) at baseline and at each time point throughout the restraint stress period (P < 0.05 for all; Fig. 4A). Cortisol increased significantly during stress (10 and 120 min) compared to baseline in both LD and SD hamsters (P < 0.01). Testosterone concentrations increased numerically during stress in LD hamsters, but this difference is not statistically significant (P = 0.08; Fig. 4B). Testosterone concentrations were undetectable in plasma of SD hamsters.

Figure 4.

Stress-induced changes in mean (± SEM) plasma cortisol (A) and testosterone (B) concentrations in LD (n = 5) vs. SD (n = 5) hamsters. Separated circle symbols on each graph represent baseline [B] concentrations from blood plasma samples 1 wk before restraint stress. *, Significantly different from LD group; #, significantly different from 0 min baseline; P < 0.05.

Stress-Induced Changes in Leukocyte Distribution.

Hamsters displayed a robust stress-induced decrease in the number of blood lymphocytes and monocytes after 120 min of restraint (P < 0.05, for all; Fig. 5). l-selectin- (CD62L) and CD44-positive lymphocyte counts were significantly higher in SD than LD animals (n = 5, both groups) at 0 min (P < 0.01, for both; Fig. 5 A and D; respectively). All lymphocytes decreased significantly from blood after 10 and 120 min of stress in SD but decreased in LD animals only after 120 min (P < 0.01, for both). SD animals also showed a stress-induced increase in blood CD62L-positive neutrophils after 120 min (P < 0.01; Fig. 5C), whereas LD cell numbers did not change.

Figure 5.

Stress-induced changes in mean (± SEM) blood leukocyte numbers in LD (n = 5) vs. SD (n = 5) hamsters. Baseline [B]. SD hamsters have higher baseline numbers of lymphocytes (LYM) and exhibit a greater stress-induced decrease in cell numbers, than LD hamsters (A and D). Monocyte (MONO) numbers significantly decrease during stress in LD and SD animals (B and E), whereas CD62L+ neutrophils (NEU) increase after 120 min in SD animals only (C). *, Significantly different from LD group; #, significantly different from 0 min baseline; P < 0.05.

DTH.

Nonstressed SD hamsters (n = 10) exhibited an enhanced DTH response compared to nonstressed LD hamsters (n = 10; P < 0.05; Fig. 6). Similarly, stressed SD hamsters (n = 10) exhibited an enhanced DTH response compared to stressed LD hamsters (n = 10; P < 0.05). Acute stress significantly increased the inflammatory response during days 2–3 post-DNFB challenge compared to nonstressed animals in SD hamsters only (P < 0.01).

Figure 6.

SD hamsters exhibited an enhanced DTH response [mean (± SEM) percentage increase in pinna thickness] during both stress (n = 10) and nonstress (n = 10) conditions compared to LD hamsters (n = 10, each group). Restraint stress significantly increased the inflammatory response during days 2–3 post-DNFB challenge in SD hamsters only. *, Significantly different from LD (within group); #, significantly different from nonstressed SD group; P < 0.05.

Discussion

Seasonal changes in the onset and severity of stressors may influence survival (24, 25). Energy availability throughout the year is not constant, and numerous adaptations have evolved in nontropical species such as Siberian hamsters to help survive the energetic bottleneck of winter. We hypothesized that hamsters use SD lengths to signal the onset of winter to enhance immune function when environmental stressors are typically high. SD hamsters have higher numbers of circulating blood leukocytes, total lymphocytes, T cells, and NK cells compared to LD animals. During stressful challenges, however, it may be adaptive for immune cells to exit the blood and traffic to immune defense areas such as the skin, in preparation for potential challenges or injury (13). Acute stress in hamsters induces a rapid deployment of lymphocytes and monocytes out of the blood, with a greater and more rapid decrease of lymphocytes from the blood in SD compared to LD animals. SD hamsters also show enhanced skin DTH responses compared to LD hamsters, and this effect is augmented by acute stress in SD hamsters only.

Neural and endocrine factors may influence immune function by inducing changes in the functional activity of cells, or by inducing changes in immune cell numbers and distribution between various immune compartments, or both (26). Previous research indicated that Siberian hamsters enhance spontaneous blastogenesis in whole blood and isolated lymphocytes, and enhance NK cell cytolytic capacity in SDs compared to LDs (20). Peripheral blood immune cells represent a small proportion of total body leukocytes; however, the blood is a necessary conduit by which leukocytes travel among tissues. Thus, an analysis of cell phenotypes and populations within the blood provides an important representation of the distribution of leukocytes throughout the body (27). Increased blood leukocyte numbers in SD hamsters are coincident with significant decreases in testes mass, testosterone concentrations, and body mass. Importantly, the SD increase in cell numbers cannot be attributed to lower body mass, or differential blood volume-to-body ratio. All leukocyte number differences between LD and SD animals are statistically significant regardless of correction for body mass, and blood hematocrit levels in all animals are equivalent. Rather, our results, consistent with previous reports (20, 28), suggest that physiological and metabolic adjustments during winter in this species include specific, organized changes in immune surveillance.

We suggest that augmentation of basal leukocyte numbers in SD hamsters may occur in preparation for seasonal stressors such as low temperatures and reduced food availability, which would otherwise increase susceptibility to infection. Photoperiod may provide a useful cue by which stressors in the environment are anticipated. Importantly, augmentation of leukocyte numbers appears to occur by means of alterations in glucocorticoid production and stress reactivity during SD exposure. Restraint stress represents a general stressor, and reliably activates the hypothalamic-pituitary-adrenal (HPA) axis (16, 29) and sympathetic nervous system (30), and increases glucocorticoids in rats and mice (26, 27). As has been demonstrated previously in rats and mice (16, 21, 26, 27), restraint stress significantly and rapidly influences the distribution and proportions of leukocytes within the blood of hamsters. Contrary to the idea that glucocorticoids necessarily suppress immune function, acute stress in mice and rats appears to enhance skin immune function by leukocyte trafficking to the skin (12, 16). During acute stress, immune cells are redeployed throughout the body from storage areas such as the spleen, into areas such as the skin, and this redistribution occurs by means of the bloodstream. After antigenic challenge, stressed animals exhibit a more robust DTH response in the skin compared to unstressed controls. Importantly, in the absence of challenge or injury, leukocyte trafficking out of the blood during stress is rapidly (minutes to hours) reversible after the cessation of stress in rats (27). Thus, blood leukocyte numbers quickly return to previous baseline levels upon the termination of restraint stress, strongly indicating that reduced blood leukocyte numbers during stress reflect a dynamic redistribution of cells, rather than a loss of cells.

In the current report, cortisol concentrations, the primary glucocorticoid in Siberian hamsters, rise significantly in all animals after 10 min of restraint and remain elevated for 2 h. However, SD hamsters, in addition to higher baseline concentrations, exhibit a greater cortisol response to 2 h of acute stress than do LD hamsters. This difference is accompanied by an accelerated decrease in total blood leukocyte numbers during stress, and enhanced DTH reactions after stress, in SD compared to LD hamsters. DTH reactions are antigen-specific, cell-mediated immune responses, involving T cell function and immunological memory, and play a significant role in the development of resistance to bacterial and viral infections (18, 31, 32). After antigenic challenge, locally released inflammatory cytokines work to increase blood vessel permeability and recruit accessory cells to the site. Changes in cell-surface adhesion molecules (i.e., CD62L and CD44) are important markers of immune status (17). For instance, leukocytes bearing specific l-selectin (CD62L) molecules are involved in cell adhesion and mediate rolling interactions on the endothelium during immune activation (33). Similarly, CD44 levels increase on memory T cells on initial exposure to antigen, which then aid in lymphocyte homing to tissues after a second exposure (34).

During stress, glucocorticoids may help prepare an organism for a possible challenge by increasing the expression and/or affinity of adhesion molecules on leukocytes and/or endothelial cells. These processes may enhance leukocyte recruitment to the site of challenge (14, 35). The numbers of CD62L and CD44 lymphocytes are initially higher in SD compared to LD hamsters, but these cells exit the blood more rapidly in SD in response to stress. Blood neutrophil numbers tended to increase or remain constant during stress in hamsters, consistent with a previous report in mice (35). Neutrophils are the initial responders during wounding and inflammation and appear to remain in the blood to aid in the transport of immune cells out of the blood vessel and into surrounding tissue (35). Interestingly, these cell numbers increased by the end of the stress session in SD hamsters only.

Taken together, our results suggest that Siberian hamsters are prepared to mount a greater and more rapid immune response during SD compared to LDs. Importantly, stress responses and the production of glucocorticoids should be considered as part of the mechanisms underlying seasonal changes in immune function. Because several weeks of exposure to SD photoperiod are necessary to observe significant changes in immune function, it is possible that the rise in cortisol and stress itself provide the signals for immunoenhancement in Siberian hamsters. Glucocorticoids act on multiple targets to enhance or inhibit various cellular activities, actions adapted to provide the altered metabolic, endocrine, nervous and immunological needs necessary for survival (36). Many species exhibit seasonal alterations in glucocorticoid secretion (37–39). Studies in other rodent species, however, have indicated that immune changes in response to photoperiod occur in the absence of elevated glucocorticoids (9). Stress responses and glucocorticoid secretion have generally been described to work in concert with reproductive state and the sex steroid hormone environment (i.e., high glucocorticoid concentrations coincident with breeding and reproduction). However, an opposite pattern has been described for species inhabiting extreme environments (e.g., arctic and xeric) (40). Stress hormones generally suppress reproductive axis hormones, and those animals with limited breeding opportunities may have evolved mechanisms to suppress their stress responses during breeding to maximize reproductive success (40). These alterations may also reflect changes in metabolism, in which elevated glucocorticoid concentrations work to mobilize the usage of energy stores throughout the body. Future studies should address the extent to which day length and glucocorticoids may influence the onset and severity of disease in multiple species, to help inform better treatment and prevention.

Abbreviations

LD

long day

SD

short day

DTH

delayed-type hypersensitivity

DNFB

2,4-dinitro-1-flourobenzene

EST

Eastern standard time

NK

natural killer