Retinitis pigmentosa is a group of inherited disorders characterized by progressive peripheral visual field loss, abnormal ERG responses and variable clinical presentation, severity, age of onset, and progression. Early symptoms can occur in childhood or adolescence and usually consist of night blindness due to loss of the rods. It may lead to a gradual reduction of the visual field, and total blindness due to cone involvement in the late stage of the disease (1).
Although several genetic mutations have been identified in patients with RP, the mechanisms through which these mutations lead to photoreceptor apoptosis remain largely unknown.
Rod death is a consequence of genetic mutations and about 80 RP causative genes have been identified. Conversely, cone degeneration is a late event and thought to result from oxidative damage and inflammation caused by rod loss.
Inflammation is now considered a crucial hallmark of chronic disorders, including cardiovascular disease (7), neurodegenerative diseases (8, 9,10), diabetes (11), metabolic syndrome (12), and cancer (13). Furthermore, recent basic and clinical studies have suggested the importance of chronic inflammation in the pathogenesis of neurodegenerative disorders such as Alzheimer’s disease (14), Parkinson’s disease (15), and degenerative retinal diseases, including RP.
Investigations of inflammatory alterations in RP patients have been discussed in recent literature (16) and previous studies have reported the presence of an inflammatory reaction in the eyes of patients with RP, suggesting a potential role of flogosis in RP (17).
Some studies showed the presence of various lymphocyte subsets in vitreous samples obtained from RP patients (18). Yoshida N. et al., by assessing the anterior vitreous of 371 RP patients with slit-lamp biomicroscopy, found a substantial number of cells in the vitreous cavity in 30% of patients with RP. Moreover, they observed that younger patients with an active disease process were more frequently associated with stronger inflammatory reactions and RP patients with more vitreous inflammation had significantly lower visual function. Moreover, the levels of a variety of proinflammatory cytokines and chemokines, including monocyte chemotactic protein-1 (evaluated by a multiplex enzyme-linked immunosorbent assay (ELISA)) increased both in the aqueous humour and vitreous fluid of RP patients compared with controls (19).
These findings suggest that intraocular inflammatory reactions may contribute to the progression of RP.
More studies confirm that the progression of many inherited retinal dystrophies is influenced, among other factors, by the activation of the immune cells and the release of inflammatory molecules such as chemokines and cytokines (20). A recent study revealed that in RP patients intraocular levels of inflammatory cytokines such as IL-2, IL6, MCP1 (monocyte chemoattractant protein1), PIGF (placental growth factor) exceeded the levels of serum, indicating intraocular production (21). A further study confirmed that serum inflammatory cytokines (such as IL8 and RANTES) levels were significantly increased in patients with RP compared with controls and the levels of IL8 were negatively correlated with visual acuity, retinal sensitivity, the central subfield thickness and ellipsoid zone width (22).
In our study, we have investigated the presence and nature of immune cells in the eyes of RP patients using flow cytometry; we also evaluated the functional property of these cells by performing the cytokine profile at single cell level.
We have detected in RP aqueous samples the presence of infiltrating leucocytes with lympho-monocyte characteristics, consisting of T lymphocytes (CD3+, both CD4 + and CD8+, and CD4-CD8-), B lymphocytes (CD19+), natural killer cells (CD16 + CD3-) and monocytes (CD14+), while no cells were present in the aqueous humour of the controls. Of note, red blood cells were virtually absent in all the samples evaluated, thus excluding blood contamination. Therefore, our data support the hypothesis of a role of inflammation in the pathogenesis of RP, in agreement with literature (23)
Interestingly, the cell composition of AH was limited to MNCs; moreover, lymphocyte composition was different from that found in the blood sample of the same patients, with a significant reduction of NK cells, of CD4 + Th cells and with a significant increase of CD3 + CD4-CD8- T cells, mainly made up of TCRαβ lymphocytes.
These data suggest the hypothesis of an impairment of the blood-retinal barrier allowing the migration of circulating inflammatory cells toward the intraocular fluids. We can speculate that these cells migrate from the bloodstream into the eye, recruited by an as yet unknown pathologic process which could be due either to cell damage of degenerative origin, such as rod death, followed by antigen spreading and consequent MNC infiltration, or could be due to a primary autoimmune insult.
In our analysis the frequency of NK cells was significantly inferior in the AH compared to the PB, rendering improbable a possible role of these cells in the local inflammatory process. The eye is characterised by immune-privilege, and the absence of a lymphatic system is important to isolate it from the immune system and to avoid a possible damage due to immune reaction. Alteration of this physiologic condition leads to immune-mediated disease of the eye involving both innate and adaptive immune response. Moreover, the corneal endothelium and retina do not express the Major histocompatibility complex (MHC) class I molecules and are a potential target of cytotoxic activity by NK cells; this could represent the starting point or the progression of an inflammatory condition. It has been shown that the aqueous humour contains a wide range of soluble factors inhibiting immune response, in particular NK-mediated cytotoxicity, the main being represented by TGF-b (23, 24). We could speculate that the ocular micro-enviroment contributes to reducing the frequency of NK cells in order to maintain tissue protection and support the immunological privilege.
Similarly, the frequency of the CD3 + CD4 + T cell (T helper Lymphocytes) subset was significantly reduced in the AH compared to the PB with a very strong increase of the CD3 + CD4-CD8- T cell subpopulation.
Double negative (DN) CD4-CD8- T cells, which compose approximately 1–3% of total T cells in healthy human Peripheral Blood Mononuclear Cells (26), have been recently described as a cell population of circulating mature T cells of unclear origin. According to emerging data these cells may be involved in the pathogenesis of different kinds of autoimmune/inflammatory systemic diseases, including systemic lupus erythematosus (SLE), Sjögren's syndrome, and psoriasis, through the induction of systemic inflammation and tissue damage (27).
In particular, DN T cells neither express γδTCR, nor NK cell markers, they do not have regulatory T cell markers such as FOXP3, CD25, or CTLA-4 (27, 28). Moreover, these cells show a reduced/absent expression of the co-stimulatory molecule CD28 (28). Of note, CD28-deficient T cells are generally considered antigen-experienced and differentiated elements; accordingly, DN T cells are CCR7 + CD45RA−, thus they display phenotypic characteristics of terminally differentiated and “exhausted” T cells (27). To our knowledge this is the first description of DN CD4-CD8- T cells in RP patients.
With regard to CK production by in vitro stimulated T cells, we obtained only preliminary data showing some interesting trends, with no statistical significance due to the low number of experiments (n 5) in which the cell number was sufficient to perform the cytokine stimulation assay. For the same reason the CK measurement was performed on CD3 + CD8- and CD3 + CD8 + T cells, but not on the less frequent CD3 + CD4-CD8- T cell subpopulation. In particular, we found a slight increase in the frequency of CD161 + CD4 + T cells producing IFN-ɣ in the AH, as compared with circulating lymphocytes, and a trend toward a decrease in CD3 + CD8 + T cells producing IL-4 and double producing IFN-ɣ plus IL-4. These data together are in favour of a cytotoxic cell profile in the AH, that is consistent with a pathogenetic role for these infiltrating lymphocytes. Of note the CD161 + CD4 + IFN-ɣ+ T cell population has already been described, by our group and others, as a Th17 derived phenotype, involved in the tissue damage during various autoimmune/inflammatory diseases (29, 30). The Th1-oriented response leads to monocyte/macrophage activation, with subsequent oxidative stress of the tissue.
In general, all our data agree with the hypothesis of a pathogenetic role for immune cells, and therefore inflammation, in RP. The initial mechanism which may lead to ocular inflammation in RP patients is not clear at the moment, but some hypotheses may be made.
First, most dystrophic and degenerative diseases are accompanied by low-grade inflammation. It is well known that in RP increased retinal lipofuscin fluorophores may determine damage, disturbed polarity, death of the RPE, and apoptosis of photoreceptors (31). In response to this stimulation, the RPE synthesises and releases a wide variety of inflammatory molecules such as cytokines and chemokines (32), with microglial activation contributing to a proinflammatory phenotype. These events may promote the recruitment of inflammatory cells that leak into the vitreous and may reach the aqueous, as there is no barrier separating the posterior from the anterior segment (33). Secondly, as blood retinal barrier breakdown occurs both in retinal vessels and in the RPE (34,35), even the blood aqueous barrier (BAB) may be affected, leading to an increased number of inflammatory components in the aqueous.
Neuroinflammation is currently considered an early event in the pathophysiology of many neurodegenerative disorders: despite its essential role in protecting tissue at the beginning of disease, the continuous presence of proinflammatory stimuli eventually induces cellular damage (36, 37). It is generally accepted that astrocytes and microglia are the cells that in the central nervous system (CNS) play a critical role in the neuroinflammation preceding neurodegenerative diseases. In this scenario, activated microglia contribute to the release of different inflammatory mediators, including cytokines, chemokines, reactive oxygen species (ROS), and nitric oxide (NO), all of them participating in maintaining an oxidative stress and chronic neuroinflammatory milieu that ultimately could be responsible for oxidative stress and neurotoxic damage in the CNS (38).
In this context, hyperactivation of microglial cells has been shown to play an important role in photoreceptor neurodegeneration in animal models of RP (39). A recent study using live-cell imaging in the rd10 mouse model of RP has identified that the initiation of rod degeneration is accompanied by early infiltration of microglia, upregulation of phagocytic molecules in microglia, and presentation of “eat-me” signals on mutated rods. During the early stages of the disease microglia is able to migrate, interact with, and phagocyte nonapoptotic photoreceptors, after which it becomes hyperactivated and promotes the loss of non- and apoptotic photoreceptors (40). The Authors of this study propose that primary microglial phagocytosis may be a contributing mechanism underlying cell death in retinitis pigmentosa and microglia as a potential cellular target for therapy. Since activated microglia releases neurotoxic and/or inflammatory mediators including TNF-α, interleukin-1β (IL-1β), IL-6, and glutamate and increases the expression of inducible nitric oxide synthase (iNOS), all of them exacerbating the death of retinal neurons, this could contribute to the breakdown of the BRB. The outer BRB expresses immunoregulatory molecules that inhibit lymphocyte activation, while the RPE of the BRB secretes immunomodulatory mediators that control the immune and inflammatory responses within the eye/into the aqueous humour (41). Therefore, in physiological conditions, PB immune cells are not able to enter the retina to mediate endogenous insults; instead, after BRB breakdown a recruitment of inflammatory cells from the bloodstream occurs, attracted by chemokines probably produced by activated microglial cells, leading to an increased number of inflammatory components in the aqueous. A simplified hypothesis is that the underlying genetic defect is the trigger of rod degeneration, which drives a concomitant microglial activation and local increase in production of inflammatory species. In the long run, inflammation may become detrimental for all the cells, and particularly for the cones, known to be especially vulnerable because of their elevated metabolism and high degree of specialisation. This would concur to their degeneration, thus conferring to a secondary process (inflammation) a major role in cone vision loss, which is the most severe consequence of RP for humans. The importance of research aimed at understanding the function of the different inflammatory processes in the retina, and especially the contribution of the microglial-mediated neuroinflammation that precedes neurodegeneration, could provide useful knowledge to implement new therapies. Potentially, primary microglial phagocytosis could be a potential cellular target for therapy. However, more research is needed to detect further molecular mechanisms involved in microglial activation in degenerative retinal diseases.
Unfortunately, in our study we could not find any correlation in RP between the humoral and cellular immunological alterations and the genotype and clinical phenotype. We can hypothesise that these flogistic abnormalities could be more significant in patients with macular oedema (commonly considered of flogistic origin) but at present we have no conclusive data on this issue.
We are aware of some limitations of our study. First, the sample size is relatively small, but we must consider that RP is a rare disease and that aqueous samples can be obtained only during surgery. Moreover, our groups are not age-matched and the controls are significantly older than the RP patients. However, this difference should not interfere with the report on inflammatory cells in the aqueous humour of RP patients, as it has been ascertained in literature that increasing age is associated with increasing levels of intraocular cytokines and probably of inflammatory cells (42).
Sustained chronic inflammatory reaction may therefore underlie the pathogenesis of RP, suggesting interventions modulating ocular inflammatory reactions as potential treatment for patients.
Lymphocytes are critically involved in the pathogenesis of several neurodegenerative diseases, such as multiple sclerosis, and clarifying the role of lymphocytes in RP pathogenesis may lead to the identification of a common signature of lymphocytes in neurodegeneration and thus show the way towards new treatment options.