|Home | About | Journals | Submit | Contact Us | Français|
The following study was undertaken to better understand the mechanisms that relate the homeostatic set point of the peripheral T cell population to energy availability in mice. We report that the total number of peripheral naïve and memory CD4+ and CD8+T cells notably declined after one week of malnourishment, a time period too short to be entirely due to malnutrition-induced thymic involution. Peripheral malnourished T cells expressed higher levels of the IL-7 receptor component, CD127, and were less sensitive to death-by-neglect as compared to control T cells. Overall levels of IL-7 were similar in malnourished and control mice. Adoptive transfer studies revealed that CD127 expression did not correlate with increased survival in vivo and that all naïve CD8+T cells upregulated CD127, regardless of initial expression levels. Corticosterone levels were elevated in malnourished mice and this correlated in time with peripheral T cell up-regulation of CD127 and the diminishment of the peripheral T cell pool. Overall, these data suggest a model in which CD127 levels are up-regulated quickly during malnourishment, thereby increasing the scavenge rate of IL-7, and providing a mechanism to quickly adjust the total number of T cells during malnutrition.
Malnutrition is the leading cause of immunodeficiency worldwide . Indeed, malnutrition and infection are intricately linked as malnourished children not only experience a heightened risk of developing infections such as diarrhea and pneumonia, but also experience greater morbidity from infection . Furthermore, the nutrition of infected children is negatively impacted by a multitude of factors, including diarrhea, malabsorption, loss of appetite, diversion of nutrients for the immune response, and urinary nitrogen loss . Consequently, infections further exacerbate malnutrition and immunodeficiency in the malnourished. Quite unfortunately, reports suggest that the interaction between malnutrition and infection contributes to 53% of all child deaths in developing countries . Additionally, malnutrition-related immunodeficiency likely reduces the efficacy of certain vaccines, such as those for rotavirus, polio, and cholera , , .
Malnutrition results in notable changes to the size, composition, and overall function of the T cell compartment . In order for the immune system to respond to any pathogen that may be encountered, the maintenance of a diverse pool of antigen-inexperienced naïve T cells is critical . The size of the naïve T cell pool is determined by the rate of egress from the thymus as well as by peripheral T cell survival and proliferation. During malnutrition, extreme thymocyte depletion occurs due to increases in thymocyte death and decreases in thymocyte proliferation . Interestingly, studies using protein-malnourished mice have demonstrated atrophy of the thymus may be directly caused by increases in glucocorticoid levels and/or decreases in leptin, the satiety hormone , , . The number of peripheral T cells is also reduced in malnourished individuals, and this reduction has been attributed to reduced thymic input . The possibility of additional, peripheral mechanisms that contribute to resetting the homeostatic set point of the naïve CD8+T cell pool during malnourishment has not been addressed.
In addition to diminishment in the size of the T cell population, the T cells that remain in malnourished individuals are functionally impaired . For example, T cell dependent immune reactions such as delayed hypersensitivity responses, are markedly suppressed and are commonly absent in malnourished children . However, existing B cell numbers and function seem to be surprisingly maintained with only new B cell responses to T cell dependent antigens experiencing impairment . This suggests that malnourishment does not affect B cell function directly and only influences antibody responses that require T cell help. Thus, malnourishment impairs the T cell arm of the adaptive immune response by diminishing their number and function.
In the periphery, individual T cells are maintained in response to extrinsic cues rather than an intrinsic survival clock. Specifically, the total number of naïve T cells is continuously adjusted through homeostatic signaling processes that involve self-peptide MHC-complexes and interleukin-7. Indeed, naïve T cells do not survive well in IL-7 or MHC Class I deficient mice , , , . Interestingly, it appears that the dose of IL-7 received by the naïve T cell influences how long that cell can persist in the absence of further survival factors, at least in ex vivo studies .
A key survival factor for T cells, IL-7 is secreted by fibroblastic stromal cells in the bone marrow and lymph nodes as well as by epithelial cells in the thymus and intestine . The available amount of IL-7 in the central lymphoid organs was thought to be primarily regulated by the rate that it is scavenged by T cells . However, recent studies using IL-7 reporter mice suggest that IL-7 expression is increased by lymph node stromal cells during lymphopenia . In certain contexts, some cytokines can also affect production of IL-7 in vivo , , . Thus, although the scavenge rate may influence IL-7 availability, extrinsic stimuli such as cytokines and lymphopenic conditions influence the rate of IL-7 production, at least by stromal cells.
T cells detect IL-7 via the IL-7 receptor, which is a heterodimer comprised of CD127 (also known as IL-7Rα) and the common gamma chain (γc) . T cell expression of CD127 is thought to be highly regulated; CD127 is up-regulated in the absence of IL-7 and conversely down-regulated after contact with it . In addition, glucocorticoids, type I interferons, and TNF-α can up-regulate CD127 expression; while several cytokines (many of which signal through receptors partially comprised of γc) down-regulate CD127 in T cells .
As aforementioned, maintaining a large diversity of adaptive immune cells is necessary for survival given the unknown nature of pathogens that might be encountered. Considering the energetic challenges that this adaptation poses, lymphocytes are maintained as naïve, quiescent cells in low numbers until particular cells are called upon to defend against a cognate antigen. However, it is less clear how T cell homeostasis is linked to energy availability during drastic energy shortages, as occurs in malnourishment. The following study was undertaken to better understand the impact of short-term malnutrition on T cell homeostasis, with particular attention focused on the CD127/IL-7 signaling dynamic of peripheral T cells.
All mice were housed in the mouse facility at Randolph-Macon College in accordance with approved Institutional Animal Care and Use Committee guidelines and protocols and with the NIH guide for the care and use of laboratory animals. For each individual experiment, the mice were placed on malnourishment as follows. Male and female C57BLJ/6 mice ranging in age from 8 to 15 weeks old were singly housed. The unrestricted intake of chow was monitored for each mouse two weeks prior to the initiation of malnourishment. Mice were randomly assigned to each treatment group. Malnourished (MAL) mice received 35% less Teklad Global 18% Protein Rodent Diet (Harlan Laboratories) by weight while ad libitum (AL) control mice had unrestricted access to the same chow. The dieting period lasted one week for all studies with the exception of the “death-by-neglect” study in which the dieting period lasted 6 weeks. Although some individual malnourished mice lost up to 28% of their body weight, on average, malnourished mice lost 10%, while control mice gained 3% of their original body weight over the course of one week. In all experiments, mice were euthanized via CO2 overdose followed by cervical dislocation.
After 1 week of the malnourishment or control diet, cells were isolated from the spleens of male and female C57BL/6J mice aged 10–13 weeks after a malnourished or control diet for one week. Red blood cells were removed from splenocyte preparations by incubation with ACK lysing buffer (0.15 M NH4Cl; 10 mM KHCO3; 0.1 mM EDTA). The ammonium chloride in the buffer lyses red cells with minimal effect on lymphocytes. After osmotic lysis, one million cells were incubated with Fc-blocking reagent (Purified Anti-Mouse CD16/CD32 (2.4G2)) (Tonbo, California, USA). Surface protein markers used to distinguish particular immune cell subsets are as follows: T cells (CD3+ (145–2C11, eBioscience, California, USA)), B cells (CD19+ (eBio1D3, eBioscience, California, USA)), granulocytes (CD11b+ (M1/70, eBioscience, California, USA) Gr1+ (RB6–8C5, Invitrogen, USA)), and monocytes (CD11b+Gr1-). Naïve (CD44lo) and memory (CD44high) CD4+ (GK1.5, eBioscience, California, USA) and CD8+ (53–6.7, Southern Biotech, Alabama, USA) T cells were distinguished with the surface protein CD44 (1M7, eBioscience, California, USA). After staining, cells were washed and acquired on a BD C6 Accuri cytometer (BD Biosciences, California, USA). Gating strategies used to determine cell percentages are detailed in Fig. S1. The intensity of CD127 surface protein expression was assessed with an antibody to CD127 (A7R34, eBioscience, California, USA). The total number of naïve (CD44lo) and memory (CD44hi) CD4+ and CD8+T cells within the spleen and lymph node was calculated by multiplying the percentage of each subpopulation by the total number of cells in each organ. Replicates from at least two independent experiments were pooled for this analysis. The total number of replicates (combined from each independent experiment) are as follows: n=10, spleen and thymus AL; n=14 spleen and thymus MAL; n=7, lymph node AL; n=5, lymph node MAL.
After one week of malnourishment or control diet, the lymph nodes of three AL and 9 MAL mice were isolated. Given the significantly smaller size of the lymph nodes from MAL mice, the lymph nodes from 3 individual MAL mice were pooled. RNA was isolated, DNased, and reverse transcribed with the FastSCRIPT cDNA Synthesis Kit (Tonbo). The cDNA from each sample was subjected to quantitative PCR specific to the IL-7 and β-2 microglobulin genes using the FastPROBE qPCR Hi-ROX Master Mix (Tonbo). IL-7 was targeted with primers and probes included in the PrimeTime Redesigned qPCR assay (Mm. PT.58.10325839, Integrated DNA Technologies) and β2 microglobulin was targeted with the following primers and probe:
5′-/56-FAM/TTC AAG TAT /ZEN/ACT CAC GCC ACC CAC C/3IABkFQ/-3′; primer 1: 5′-GGT CTT TCT GGT GCT TGT CT-3′; primer 2: 5′-ACG TAG CAG TTC AGT ATG TTC G-3′ (Integrated DNA Technologies). Note that “ZEN” refers to an additional quencher placed 9 bases from the fluorophore to reduce background fluorescence.
After 6 weeks of the dieting regimen, CD8+T cells were enriched from the spleens and lymph nodes of male and female C57BL/6J mice aged 8–15 weeks via immunomagnetic separation (EasySep Mouse CD8+T cell isolation kit, Stem Cell Technologies, Vancouver, Canada). Briefly, non-CD8+T cells were labeled with biotinylated antibodies recognizing specific cell surface markers. Cell preparations were subsequently incubated with streptavidin-coated magnetic particles and separated using an EasySep magnet (Stem Cell Technologies). After isolation, CD8+T cells were cultured overnight in complete RPMI 1640 media with 0 or 100 ng/ml of IL-7 (Shenandoah, PA, USA). T cells cultured without survival cytokines, such as IL-7, will undergo cell death, commonly termed “death-by-neglect”. CD8+T cell death was assessed via flow cytometry as above, using Annexin V (BD Biosciences, CA, USA) and 7AAD (7 aminoactinomycin D, BD Biosciences, CA, USA) following overnight culture. Gating strategies used to determine cell percentages are detailed in Fig. S1E. The experiment was repeated at least four times and the data were pooled for analysis. The total number of replicates (combined from each independent experiment) are as follows: AL and MAL, 0 IL-7: n=13/group, AL and MAL, 100 IL-7: n=7/group.
Naïve cytotoxic cells (CD8+CD44low) were sorted into CD127high and CD127low populations via a FACS ARIA (BD Biosciences, California, USA) in the Flow Cytometry Shared Research Core at the Massey Cancer Center (Virginia Commonwealth University). Sorted CD8+CD44loCD127high and CD8+CD44loCD127low cells were labeled with 0.5 and 5 µM CFSE (Carboxyfluorescein succinimidyl ester, Tonbo, CA, USA), respectively. After labeling, CD8+CD44loCD127highCFSElo and CD8+CD44loCD127low CFSEhigh cells were mixed 1:1. Each recipient mouse was injected with the cell population mixture via a single retro-orbital injection (six million cells total/mouse) and then placed on control or malnourishment diet for one week. The recovery rate of CFSEhi and CFSElo cells was determine by flow cytometry as described above. Gating strategies used to determine CD127 expression within each population are detailed in S2 Fig. Male and female C57BL/6J mice aged 14–15 weeks were used in these experiments. This experiment was repeated twice with n=4/treatment level for each experiment.
Corticosterone metabolized in excreta was extracted and quantified using the Corticosterone EIA kit (Cayman Chemical, Michigan, USA) on days 0, 2, an 7 of malnourishment. Samples (0.05 g each) were collected in the morning (all collection times were not more than one hour earlier or later from each other) from each animal and kept at −70 °C until assaying. For extraction, fecal samples were thawed, vortexed for 30 s with 1 ml of 100% methanol, and centrifuged for 10 min at 2500 rpm. The supernatants were collected, diluted in EIA buffer, and quantified via EIA. Male C57B/6 mice (8 weeks of age) were used in these studies. The total number of fecal samples analyzed each day of malnourishment are as follows: day 0, n=9; day 2, n=5; day 7, n=6.
The statistical significance of any differences among treatment groups for all aspects of this study was determined in GraphPad Prism by a one-way ANOVA and/or unpaired t-tests depending upon the number of treatment groups and as indicated in the Assumptions regarding equal variance between two groups were checked with an F test in GraphPad Prism. Variations within treatment levels were equal in all cases, except for the expression of CD127 on naïve CD8+T cells between MAL and AL groups in Fig. 2A, the death-by-neglect rates in Fig. 2D, and the corticosterone concentrations in Fig. 5A.
After a week of malnourishment, the total number of cells in the spleens and lymph nodes of malnourished mice (MAL) was only one third and one fifth that observed in control mice (ad libitum, AL) (Fig. 1A and B). Mesenteric lymph nodes showed a similar trend, with malnourished mice demonstrating one fifth the number of cells as compared to control mice (MAL mean±SEM – 2.3×106±0.6; AL mean±SEM – 10.3×106±0.6; p=0.0007 by unpaired t test, n=3/group). The cellularity of the thymus was similarly reduced, with malnourished mice exhibiting less than one fourth the total number of cells as control mice (Fig. 1C). An assessment of major immune cell subpopulations was carried out to determine if one subpopulation was predominately affected by malnourishment.
Overall, no single subpopulation was responsible for the dramatic reduced cellularity of the spleen and lymph nodes. In the spleen, the percentages of B cells (CD19+) and granulocytes (CD11b+Gr1+) remained the same; while, the percentage of monocytes (CD11b+Gr1-) was halved and the percentage of CD3+T cells was modestly increased in malnourished mice (Fig. 1D). In the lymph node, the percentages of B cells were unaffected by malnutrition, but T cells were diminished by more than 10% (Fig. 1E).
The reduced cellularity of the thymus appeared to be attributed to a selective diminishment of the CD4+CD8+(double positive, DP) population. In particular, the percentage of double positive thymocytes decreased by almost half; while, the percentages of CD4SP and CD8SP thymocytes increased three- and five- fold, respectively (Fig. 1F). However, the total number of CD4SP, CD8SP, and DP cells were significantly reduced in malnourished mice as compared to controls (CD4SP mean±SEM – AL: 8.5±1.6, MAL: 2.9±0.5, p=0.01, by unpaired t test; CD8SP mean±SEM – AL: 2.1±0.5, MAL: 0.8±0.1, p=0.014, by unpaired t test; DP mean±SEM – AL: 77±14, MAL: 12±3, p=0.00003, by unpaired t test; AL, n=10, MAL, n=14).
The total numbers of memory and naïve CD4+ and CD8+T cells were also assessed in the spleens and lymph nodes of mice from each group. Malnourished mice exhibited no more than half the number of naïve and memory CD4+ and CD8+T cells in the spleen (Fig. 1G). The diminishment of absolute numbers of peripheral T cells in the lymph nodes was considerably more dramatic than in the spleen, with naïve T cells being more affected. Naïve CD4+ and CD8+T cells were reduced by almost 15-fold in the malnourished mice as compared to controls, while memory CD4+ and CD8+T cells were diminished by five-fold (Fig. 1H).
Given the dramatic reduction in T cell numbers during malnutrition, the expression levels of CD127, the alpha (α) subunit of the IL-7 receptor, and IL-7 were evaluated. All T cell subsets demonstrated higher levels of CD127 in the spleens of malnourished as compared to ad libitum-fed mice (Fig. 2A). The expression of CD127 was also significantly up-regulated in naïve and memory CD4+ and CD8+T cells in the lymph nodes of malnourished mice by 30–80%, with the largest up-regulation observed in naïve CD4+ and CD8+T cells (Fig. 2B). Even though the number of T cells was quite different in lymph nodes isolated from malnourished mice, the total levels of IL-7 mRNA was not significantly different than those observed in control, ad libitum mice (Fig. 2C).
These findings raised the question of whether the increased expression of CD127, yet relatively constant levels of IL-7, may relate to individual T cells consuming more IL-7 per cell during malnourishment. Others have shown that the IL-7 dose received by T cells in vivo correlates with increased survival time in vitro when the cells are cultured without survival factors . Therefore, we compared the propensity of malnourished and control T cells to “die-by-neglect” in vitro as a way of assessing the dose of IL-7 likely received in vivo. When IL-7 was not present in overnight culture, splenic CD8+T cells from MAL mice exhibited significantly less death-by-neglect than cells from AL mice. Indeed, 15% of CD8+T cells from AL mice underwent death-by-neglect while less than 5% of CD8+T cells from MAL mice did so (Fig. 2D). As expected, IL-7 supplementation rescued the majority of AL CD8+T cells from death while it had little effect on MAL CD8+T cell death (Fig. 2D). These data support the hypothesis that malnourished T cells receive a higher “dose” of IL-7 than control T cells.
In order to determine whether CD127 up-regulation occurred as a result of selective survival, in which CD127high cells were selected for in a malnourished environment, or if CD127 was up-regulated in malnourished T cells, CD127high and CD127low naïve CD8+T cells stained with different concentrations of CFSE dye (CFSElo and CFSEhigh) were adoptively transferred into MAL or AL recipient mice (Fig. 3A). After one week, the expression levels of CD127 were evaluated on endogenous and adoptively transferred CD8+T cells. As expected, endogenous CD8+T cells isolated from both the spleens and lymph nodes of MAL mice expressed significantly more CD127 compared to AL mice (Fig. 3B and C). Interestingly, CD127 expression was up-regulated on CD8+T cells adoptively transferred into MAL mice regardless of whether the cells initially expressed high or low levels of the receptor. CD127lowCFSEhigh and CD127highCFSElow adoptively transferred naïve CD8+T cells isolated from both the spleens and lymph nodes of MAL recipients expressed 1.5 times the level of CD127 as cells isolated from AL recipient mice (Fig. 3D and E).
The recovery rates of adoptively transferred CD127lowCFSEhigh and CD127highCFSElow CD8+T cells from AL and MAL recipient mice were quite varied. However, it was apparent that initial expression levels of CD127 on adoptively transferred naïve CD8+T cells had little influence on their later recovery from either the spleen or lymph node during malnourishment, as there was no difference in the recovery rates of CD127lowCFSEhigh and CD127highCFSElow CD8+T cells from malnourished recipients (Fig. 4A and B). Likewise, the intensity of CD127 expression on naïve CD8+T cells adoptively transferred into MAL recipients did not correlate with differences in cell death rates, as assessed by 7AAD staining, on cells recovered from the lymph nodes or spleen (Fig. 4C and D).
We reasoned that increased expression of CD127 on T cells in malnourished mice could be caused by increased levels of glucocorticoid. Therefore, corticosterone concentrations and CD127 expression in T cell populations were evaluated at days 0, 2, and 7 of malnutrition. Prior to malnutrition, corticosterone metabolites in fecal samples were quite low. Corticosterone levels were increased by day two and by day seven, levels were ten-fold higher than at baseline (Fig. 5A). Increasing corticosterone levels correlated with increasing CD127 expression in CD4+ and CD8+T cells isolated from the spleen and lymph nodes (Fig. 5B–E). Finally, both corticosterone levels and CD127 expression were inversely correlated with the total number of CD4+ and CD8+T cells in the spleens and lymph nodes (Fig. 5).
In well-nourished organisms, homeostatic mechanisms maintain a diverse population of naïve and quiescent T cells so that any foreign antigen encountered may be appropriately recognized. Presumably, the particular number of T cells maintained is established by mechanisms that balance pathogen readiness with energy conservation. During low energy contexts, such as malnutrition, the balance is shifted, and the total number of T cells declines. Previous studies have shown that reductions in thymic size and output contribute to the reduction of peripheral T cell numbers of malnourished individuals , . Here we report that the total numbers of CD8+T cells notably declined after just one week of malnourishment. Yet, mature T cells persist in the periphery weeks to months after leaving the thymus . If thymic mechanisms were uniquely responsible for reducing the homeostatic set point during malnourishment, one would expect T cell numbers to diminish in a much more gradual manner over the course of one week. Therefore, the possibility of additional mechanisms at work in the periphery should be considered.
Malnourished individuals experience dramatic atrophy of the thymus as well as reductions in peripheral lymphocyte numbers, and this unfortunate consequence has also been observed in various animal models of malnutrition and fasting , , , . Several studies indicate that changes in glucocorticoid and/or leptin levels during malnourishment are likely responsible for massive thymocyte apoptosis during malnutrition . For example, fasted mice that have been adrenalectomized do not experience a decline in lymphocyte numbers although fasted intact mice do, suggesting that hormones derived from the adrenal glands, such as glucocorticoids, play a role in reducing T cell numbers during times of energy shortages . Additionally, leptin deficiency has been associated with reduced counts of lymphocytes, and leptin restoration reverses malnutrition-associated effects on immune cell homeostasis , . As has been reported by others, we observed a decrease in the proportion of double positive cells (CD4+CD8+thymocytes) and an increase in more mature, single positive cells. It is possible that this could be due to a reduced number of T cell precursors arriving to the thymus from the bone marrow coupled with increased differentiation. On the other hand, others have proposed that the selective depletion of CD4+ CD8+(double-positive) thymocytes during malnutrition may be the result of abnormal, early export of these cells from the thymus, as occurs in chronic infections .
It is currently unknown whether the influence of glucocorticoids on T cell homeostasis involves peripheral and/or central mechanisms. Certainly glucocorticoids concentrations affect thymus size. Peripheral mechanisms in which glucocorticoids act directly on mature T cells are possible, as mature T cells have been shown to be sensitive to the effects of glucocorticoids . In addition, multiple studies have shown that glucocorticoids can modify expression of CD127 , . Therefore, we suspect that the significant up-regulation of CD127 in malnourished T cells observed in our study, regardless of initial receptor expression, is a direct result of increased glucocorticoid levels.
IL-7 plays an important role in the homeostatic fitness of T cells in the thymus and in the periphery . The dose of IL-7 received by a T cell in vivo has been shown to correlate with the length of time the cell can survive when deprived of survival factors in vitro . For example, T cells from lymphopenic mice are thought to experience less competition for IL-7 and therefore obtain a higher dose of the cytokine in vivo. Conversely, T cells conditionally deficient for CD127 do not receive any IL-7 signals in vivo whatsoever. When placed in an ex vivo death-by-neglect assay, CD8+T cells that received a high dose of IL-7 in vivo were less sensitive to death-by-neglect than cells that received a lower dose . In our study, malnourished T cells, though fewer in number, were also less sensitive to death-by-neglect. In addition, the level of IL-7 was similar in malnourished and control lymph nodes. When considered together, these results indicate that peripheral malnourished CD8+T cells experienced a higher dose of IL-7 than control T cells in vivo. We can infer from these two findings that each T cell receives a higher dose of IL-7 than would typically be available in a well-nourished animal, explaining the malnourished T cell's reduced reliance on IL-7 exposure ex vivo.
Based upon these results, we propose that reductions in the number of peripheral T cells during malnourishment may, at least partially, result from alterations in the peripheral IL-7/CD127 signaling axis. Specifically, glucocorticoids released from the adrenal gland in response to the stress of malnutrition may act directly on peripheral T cells, resulting in the up-regulation of CD127 expression. As a consequence, individual T cells with closest proximity to IL-7 will consume more of it, leaving less IL-7 for other T cells. Those T cells not fortunate enough to obtain a dose of IL-7 will perish, consequently resetting the homeostatic set point for the T cell population. Future studies manipulating the CD127/IL-7 signaling axis in malnourished mice will best test this hypothesis.
Although it has not been heavily studied, the available data indicate that many of the immunologic consequences of malnourishment can be reversed upon restoration of a normal diet , , . However, until chronic malnourishment can be eradicated through social, economic, and political change, there remains a pressing need for optimizing vaccination and treatment strategies for the malnourished. An improved understanding of the molecular pathways involved in malnourishment-induced immunodeficiency is an important first step towards this goal.
We thank Julie Farnsworth at the VCU Massey Center Flow Cytometry Core for her expert assistance with the FACS Aria (partially supported by the Massey Cancer Center Core NIH grant: P30CA16059). We also thank Seth Litvin, Victoria Robinson, Josh Anoff, Brittany Mihalcoe, Alex Koppelman, Samantha Scott, and Talbot Weston for their technical help on various aspects of this project. This work was funded by the Randolph-Macon College Chenery, Cragie, and Rashkind grants, the Jeffress Memorial Trust (J-1049), and the Mednick Fellowship. SM, KP, and TT received support from the R-MC Schapiro Undergraduate Research Fellowship and KP and TT also received support from a Virginia Federation of Independent Colleges/Carilion Fellowship.
1MAL – malnourished; AL – ad libitum; LN – lymph node; MHC I – major histocompatibility complex I; IFN-γ – interferon gamma
Appendix ATransparency document associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. bbrep.2016.06.006.
Appendix BSupplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. bbrep.2016.06.006.