While radiation exposure can be lifesaving in the setting of cancer treatment or bone marrow transplantation for genetic or acquired disease, adverse effects may include cytopenias, abnormal immune function, delayed healing, fibrosis, and secondary malignancies, among others. The benefits of radiation are thought to outweigh the adverse effects for therapeutic radiation exposures. However, biologically significant non-beneficial radiation exposures may also potentially occur due to accidents, intentional non-therapeutic exposures (i. e. terrorist attacks), or prolonged space travel. Thus, it is important to assess the acute and long-term effects of such exposures to facilitate the development of medical countermeasures that can mitigate their harm. Abnormal function of the immune system is one of the adverse effects of radiation exposures that can potentially persist long-term (reviewed in [1]).

Immune responses depend on the presence and proper activity of both innate and adaptive immune cells. Innate immune cells include natural killer (NK) cells, dendritic cells, macrophages, and other myeloid cells that can recognize and immediately respond to a variety of markers (antigens). Adaptive immune cells include T and B cells that develop immune responses when their antigen-specific receptors bind to the corresponding antigens on potential pathogens. While B cells primarily respond to antigens by producing antibodies, T cells can respond either via producing cytokines that regulate immune responses or via direct cytotoxicity. As important regulators of both activation and resolution of immune responses, the numbers and activity of T cells are critical for proper defense against foreign invaders.

T cells develop from hematopoietic progenitor cells that migrate to the thymus, where they generate unique antigen receptors that are selected for appropriate reactivity with foreign antigens presented by other immune cells and the absence of reactivity against self-antigens. The thymus is largest and most active early in life as it establishes the initial peripheral T cell repertoire. Progressive age-related thymic involution occurs subsequently, although the thymus normally continues to produce a low number of T cells throughout the adult life [2–5]. The potentially novel antigen receptor specificities carried by these newly produced naïve T cells (thymic emigrants) have been hypothesized to be critical to maintain broad polyclonal immunity and to achieve full immune reconstitution without the development of “holes” in the immune repertoire following an insult such as chemotherapy or radiation that results in depletion of the previously established peripheral T cell repertoire [6,7]. Thymic function can be assessed non-invasively via direct detection in peripheral blood of phenotypically naïve T cells and/or extrachromosomal circles of DNA called signal joint T cell receptor excision circles (sjTRECs) that are excised during T cell receptor gene rearrangement in the thymus [8]. Since sjTRECs are not replicated during T cell proliferation in the periphery, their presence within cells provides a molecular indicator that those cells were produced de novo in the thymus.

The mechanisms of radiation-induced immune injury and acute recovery have been studied in a variety of small animal and primate models [9,10]. Studies that have focused on late effects of radiation (> 6 months post-radiation) have often shown incomplete hematopoietic and/or thymus recovery, especially in adults. For example, surgically post-menopausal “late-middle aged” female cynomolgus macaques (median age, 20 years; range 15–24 years) exposed to 5 Gray (Gy) radiation demonstrated acute, dose-dependent decreases in hematopoiesis that had not normalized by 6 months post-radiation. Thymus size and function were still decreased at 6 months for animals that had received either 2 or 5 Gy radiation [11]. Thymus tissues from humans exposed to up to 2.9 Gy radiation from the atomic bomb (A-bomb) in Hiroshima showed decreased thymic function at the time of their natural deaths 9–41 years later, compared to age-matched non-irradiated individuals [12]. Studies of how irradiation and thymic recovery affect immune responses to pathogens or vaccines generally have been limited by small cohorts, short follow-up times, and/or potential immune effects of underlying disease. Changes in circulating plasmacytoid dendritic cells were recently documented in 153 Japanese female atomic bomb survivors more than 60 years after radiation exposure [13] suggesting potential effects on immune function. Unfortunately, interpretation of some findings in the human A-bomb survivor studies are often complicated due to age related effects on the immune system. For example, a recent study of the effects of prior A-bomb radiation exposure on immune responses to influenza vaccine was limited by the overall poor antibody responses of both elderly control and irradiated subjects [14]. In addition, the studies in human A-bomb survivors are limited by variable radiation doses, heterogeneous exposures, differing ages at time of exposure, complications by under-nourishment, and retrospective analysis.

The rhesus macaque non-human primate (NHP) model has been shown to be an excellent model for study of immune system homeostasis and function [15] including studies that assess radiation countermeasures [11,16–18]. We used this model to test the hypothesis that previous irradiation would impair hematopoiesis and/or humoral and cell-mediated immune responses years after irradiation, using both phenotypic and functional analyses. Hematopoietic recovery was assessed in a large cohort of rhesus macaques at a median of ~5 years post-acute radiation exposure. The capacity of a subset of these animals to mount antibody responses was determined by challenge with rabies, Streptococcus pneumoniae, and tetanus toxoid antigens. Individual variation in immune recovery was assessed through measurement of complete blood counts, lymphocyte subpopulation analysis, and thymus function (via signal joint T cell receptor excision circles, sjTRECs) in the larger cohort and by post-vaccine response antibody titers in the antigen-challenged subset.

This study of hematopoietic recovery and immune function several years after radiation exposure shows that the irradiated macaques experienced dose-related thrombocytopenia, increased total white blood cell counts, and decreased percentages of T cells in peripheral blood, as a prolonged or delayed effect of radiation exposure. Effects of prior radiation on production and export of new T cells by the thymus gland was associated with age, with evidence of interaction of age and radiation exposure dose for CD8+ T cells. Irradiated and control animals generally developed similar humoral responses to protein antigens and most of the polysaccharide antigens tested, although irradiated animals uniformly failed to make antibodies against polysaccharides from serotype 5 pneumococci. Trends toward decreased serum IgM responses to tetanus and slower peak antibody responses to rabies were also observed, which need to be confirmed by larger studies with increased statistical power. Taken together, this study suggests that while irradiated animals respond to most antigens, minor impairment of immune responses persisted in these relatively young subjects despite ~ 2 years of recovery, particularly relative to polysaccharide antigens.

Irradiated macaques in the larger, broader age range cohort demonstrated long-standing decreases in platelet counts as well as dose-related increases in total peripheral blood white blood cell counts relative to control animals. Prior radiation exposure has been documented to increase peripheral white blood cell counts and chronic inflammation, as demonstrated by increased erythrocyte sedimentation rates in human survivors of atomic bomb blasts [41]. These increases in inflammation have been documented to occur via a variety of mechanisms, including direct damage to cells, cellular stress responses, endovascular injury that results in vascular leak, excessive activation of M1 pro-inflammatory macrophages, and the senescence-associated secretory phenotype (reviewed in [42]). The relative contributions of any of these specific mechanisms to the differences seen in this study are unknown. In addition, since these animals were studied cross-sectionally and were heterogeneous with respect to age, age at irradiation, time since irradiation, and radiation dose, additional follow-up studies will be required to confirm whether the observed differences are representative across populations.

Irradiated macaques demonstrated dose-related decreases in the percentages of T cells present in peripheral blood. Since T cells are generated in the thymus, we investigated several markers that can provide insight into thymus gland function (i. e., thymopoiesis). The data presented here show that levels of sjTRECs were highest in younger animals and were not associated with radiation dose. The generation of new naïve T cells in the thymus depends both on availability of precursor cells that migrate from the bone marrow and thymus-intrinsic factors that affect the output of T cells. Effects of either radiation or aging on thymopoiesis are multi-factorial and depend on survival and repopulation of bone marrow hematopoietic stem cells (HSC), potential radiation-induced skewing of functional responses with respect to cell lineage repopulation properties, migration to the thymus, appropriate seeding of thymic niches, and completion of the thymocyte/T cell maturation processes. The relative contributions of each of these factors is still poorly understood. However, the data suggest that animals irradiated early in life are more resilient for recovery of thymopoiesis, compared to those irradiated later in life.

Consistent with this, increased percentages of both CD4+ and CD8+ naïve cells detected by flow cytometric immunophenotyping were also significantly associated with younger animals. However, linear regression analysis also suggested that a significant interaction between age and prior radiation dose was limited to the CD8+ naïve T cell population. It is important to note that our analysis was based on age at the time of blood draw, and that animals were irradiated at ages of 5.9 to 15 years and their thymus function was assessed 44 to 70 months later. Also, while T cell percentages varied by dose, it is very difficult to separate variability associated with the dose range used (6.6 to 8.4 Gy) given the relatively small number of animals assessed and assay variability. The animals were exposed to radiation at different research sites that also confounds dose delivery to target sites, relative dosimetry etc. Additional studies may be necessary to fully address these issues in more controlled settings.

Previous work in ovariectomized female cynomolgus macaques showed that total body irradiation sufficient to cause acute hematopoietic syndrome (e.g. 2–5 Gy) resulted in decreased thymus weight with marked atrophy of thymic cortex and medulla and decreased sjTRECs that remained evident at least 24 weeks post-irradiation [11]. The data presented here using a broader age range of male rhesus macaques showed that age had a stronger influence on the production and export of new T cells by the thymus than total body irradiation doses when assessed ~5 years post-irradiation, although there was evidence of significant interaction with radiation dose for CD8+ T cells (p = 0.05). At first glance, these results may seem discordant with prior work that demonstrated significantly decreased long-term thymic function in humans exposed to atomic bomb radiation [12] and in experimentally irradiated mice [43]. However, the studies presented here differed in several ways from those previously published, including age at irradiation, radiation doses, the time elapsed between radiation and assessment relative to the lifespan of the organism, and the methods used to assess thymus function. The thymus of young mice has been shown to rapidly regenerate in weight and production of T cells following acute radiation exposure (reviewed in [43]). However, thymic regeneration is slower in older mice and humans and may not reach pre-irradiation baseline levels (reviewed in [44]). The available data in humans suggests that radiation exposure may not necessarily prevent full thymus regeneration acutely, but may accelerate age-related thymus involution [12]. Thus, the full effects of radiation on the thymus may not yet be evident in the cohort of macaques studied, where most animals were irradiated in “late adolescence” and studied as young adults, when thymic reserve is still substantial.

Although antibody responses to most of the administered vaccine antigens were statistically similar in irradiated and control animals, one exception emerged–the response to the polysaccharide antigens from pneumococcus serotype 5 –where 4 of 4 control animals produced antibodies versus 0 of 9 irradiated animals. Immune responses to polysaccharides require a more developed immune system, such that strong responses to these antigens are not typically seen in humans until about two years of age. The “hole in the immune repertoire” toward the serotype 5 polysaccharides demonstrated by irradiated animals (p = 0.001) may thus reflect subtle radiation-associated decreases in immune function. This interpretation is supported by the non-significant trend toward decreased IgM levels produced against tetanus toxoid, the delayed peak antibody responses to rabies that were also observed in irradiated animals, and the observation that one irradiated macaque who responded to the vaccine protein antigens had only transient responses to 5 of the polysaccharide antigens and this response had disappeared by the 16 week timepoint. Interestingly, the polysaccharide expressed by serotype 5 pneumococcus is unique among the strains contained in the 23-valent pneumococcal vaccine, by the replacement of glucose in the repeating unit of its capsular polysaccharide with galactose [45,46]. How this structural variation might influence immunogenicity in irradiated animals is unclear, and whether skewing of the recovery of thymopoiesis relative to recovery of other hematopoietic cell types contributes to this observation is unknown. It is also unknown whether the antibodies made by irradiated animals differ in their functional (opsonic) activity and thus their ability to protect against disease compared to control animals. Importantly, radiation-associated immune deficits may become more pronounced as the animals age, since aging has independently been associated with decreased thymic T cell production, frequent clonal expansion of memory T cell populations, increases in autoimmune diseases/syndromes, and diminished response to vaccinations [47–50].

In summary, the data presented show that dose-related changes in peripheral blood cells and in immune responses to antigens were detected at 2 to at least 5 years after exposure to radiation. Additional studies and longer term follow-up research is needed to determine whether these changes persist and the impact of the alterations on other systems of the body. These data are relevant to the question of whether pathologies become more evident with increasing time since radiation, particularly as subjects begin to develop aging-related deficits in immune function.