La córnea en la enfermedad injerto contra huéspedpatología y tratamiento con células mesenquimales estromales humanas

Resumen

Introduction Graft versus host disease (GVHD) is a common complication of hematopoietic stem cell transplantation (HSCT), an established and potentially curative treatment for malignant and benign hematological disorders. As a part of this treatment, lymphocytes are introduced within an immunocompromised host, where they are able to recognize host antigens as foreign. This recognition has a beneficial effect, termed graft versus leukemia (GVL), which allows destruction of malignant cells and reduces the probability of relapse. However, the morbidity and mortality caused by GVHD limits the use of this treatment in a wide number of disorders. Separation of GVL and GVHD has been tried without success, given that the same immune mechanism is involved in both cases. Ocular involvement is present in 60-90 % of GVHD patients. This ocular GVHD (oGVHD) affects the ocular surface and lacrimal glands, as well as the Meibomian glands, leading to tear film instability. Hence, corneal tissues are exposed to stressful conditions that lead to the loss of the characteristic immune privilege of this part of the eye. The most frequent complication in oGVHD is dry eye, a multifactorial disease characterized by hyperosmolarity of the tear film and inflammation, with damage to the ocular surface. The consequences include the disruption of epithelial barrier, irritation and impaired vision. Although problems derived from oGVHD do not often include loss of vision, they are one of the main causes of deterioration of quality of life. As the treatment of oGVHD is palliative, usually based in tear substitutes, it is necessary to explore new treatments to improve the prognostic of this debilitating disease. Suppression of immune response has shown some good results, but medications like tacrolimus are not recommended for long-term use and some patients just do not respond to treatment. Mesenchymal stromal cells (MSC) are being used in medical trials as reparative treatment, given their ability to limit tissue destruction and enhance repair in various diseases. MSC, which were first described as resident cells in the bone marrow, have a great differentiation potential. In the last years, their immune-modulatory effects have received special attention and they are being used in a number of immune-mediated diseases with promising results. The goal of this thesis is to advance in the knowledge of GVHD events in the eye and explore the therapeutic potential of MSC in corneal damage caused by oGVHD, using a well stablished murine model of the disease. Material and methods Bone marrow and splenic cells from C57BL/6J mice (H2b) were used to induce GVHD in BALB/c mice (H2d) that received total body irradiation (TBI) as conditioning regimen (GVHD mice). No manipulated mice (Control mice) or mice that received only bone marrow cells after TBI (BM mice) served as controls. A group of mice with GVHD were treated with human MSC (hMSC) at day 10 post-transplantation by subconjunctival injection in the right eye (GVHD+MSC eyes). Systemic GVHD and oGVHD were monitored weekly, using previously stablished scoring systems. Tear condition was evaluated through tear osmolarity and tear volume assessment. Finally, we analyzed corneal damage and inflammation using immunofluorescence in sections and qPCR techniques. Transcription factor Pax6 and the corneal envelope protein SPRR1B where analyzed as markers of squamous metaplasia. Inflammation was studied through detection in the cornea of CD3, a marker of T lymphocytes, and expression of TNF-α and IL-1β. Gene expression of MMP9 was also analyzed in mice corneas, as this protease is involved in epithelial barrier disruption. In addition, we performed experiments with a human corneal-limbal epithelial (HCLE) cell line. These cells were exposed to TNF-α and IL-1β for 48 h and the effect in MMP9 levels were analyzed by PCR and gel zymography. PAX6 expression was also evaluated in these cells. Finally, we explored the possibility that MSC were able to engraft in the cornea. For this purpose, hMSC were transfected with a lentiviral vector to express green fluorescent protein (GFP). Hence, the presence of hMSC in corneal tissue was analyzed by immunofluorescence with antibodies against GFP and human mitochondria. In addition, we performed PCR to detect the possible expression of human GAPDH, which would demonstrate the presence of hMSC in the cornea. Results and discussion In the present work we have validated the use of the TearLab osmolarity System in mice. Readings with this system were similar to those obtained with other methods and we obtained good values for intraclass correlation coefficient (ICC), suggesting a good reliability of the procedure. Two forms of GVHD were noticed, because of the use of two different doses of splenic cells to induce the disease. Mice transplanted with 5 x 106 splenocytes (5M) developed a more severe GVHD, where the treatment with MSC had no visible effect. Among the corneal alterations caused by the GVHD, thinning of the epithelium and loss of Pax6 was common. In the most serious cases, total loss of Pax6 in epithelial cells and epithelial thickening was observed, a characteristic of the pathological process of squamous metaplasia. Furthermore, we saw evidence of severe inflammation, with T cell invasion in almost all corneas analyzed. Transplantation with 4 x 106 splenocytes (4M) resulted in a milder form of GVHD. These mice showed an improvement when they were treated with MSC as reflected by a better evolution of the ocular disease score. Tear osmolarity was increased and tear volume was decreased in GVHD mice compared with BM mice. These are evidences that a process of ocular dryness takes place in this model of GVHD. In contrast, BM and GVHD+MSC eyes showed similar values in both parameters. A thinner epithelium was also observed in 4M mice, which was not affected by the treatment with MSC. However, detection of Pax6 in corneal epithelium showed a similar pattern in GVHD and Control mice, and its gene expression was enhanced in GVHD corneas. In contrast, Pax6 was reduced in GVHD+MSC corneal sections and a non-statistically significant decrease was observed in gene expression. As Pax6 loss has been proposed as a marker for squamous metaplasia, we analyzed SPRR1B to know if keratinization was occurring. Our results showed an increase in SPRR1B staining in GVHD eyes that was lower in GVHD+MSC mice, which demonstrate that corneal keratinization is less frequent after treatment with hMSC. These results indicate, also, that variations in the expression of Pax6 do not always correlate with the appearance of keratinizing proteins. Immune cell invasion also occurred in 4M mice, as shown by the presence of CD3+ cells in the cornea of GVHD mice. Interestingly, no GVHD+MSC cornea presented CD3+ cells. Gene expression of IL-1β was augmented in some non-treated eyes but not in any treated one. TNF-α expression was highly increased in all corneas but in Control and GVHD+MSC. These results demonstrate that hMSC exerted an anti-inflammatory effect in the cornea of mice with GVHD. As TNF-α was increased also in MO mice, we used animals that only received the TBI to know if irradiation was underlying this effect. Our results confirmed this, as well as the effect of irradiation on Pax6 expression, which was also increased in these TBI mice. The presence of hMSC in the cornea of treated mice was determined by the immunodetection with antibodies against GFP and human mitochondria. Only two 5M animals, strongly affected by the disease exhibited positive staining for both markers. No staining was found in 4M mice neither by immunofluorescence nor by PCR detection of the transcript for human GAPDH. Therefore, MSC do not migrate to the cornea and the observed effects might be brought about by an action at distance. Given the important role of MMP9 in epithelial barrier disruption, we analyzed its expression in the cornea of our animals. We found a decrease in MMP9 expression in MO and GVHD corneas, compared to Control group, but not in GVHD+MSC corneas. Since our results were contrary to what was observed in previous studies, we also performed experiments with HCLE cells, to know if the exposure to inflammatory cytokines could explain the decrease in MMP9 expression. These cells were exposed to TNF-α and IL-1β for 48 h and the effect in MMP9 levels were analyzed by PCR and gel zymography. The results point to a complex regulation of the expression of MMP9, given that the presence of IL-1β results in a decrease in gene expression but an increase in extracellular active MMP9. In summary, we found that the use of MSC could be an effective treatment in oGVHD, as some of its more relevant consequences were reverted or avoided. MSC-treated mice showed better results in tear film measurements, and they had fewer keratinization features. More importantly, inflammation was reduced, decreasing the expression of TNF-α and avoiding T-cell infiltration. However, the treatment had no effect in a more severe form of GVHD, generated with a higher dose of splenic cells. The observed effects were not attributable to migration of MSC to corneal tissue.