Journal Logo



Dixon, Janet A.2; Reynolds, Irena2,3; Zeggini, Eleftheria2; Tullo, Andrew B.3; Chantal Hillarby, M.2,3,4

Author Information
  • Free



Corneal transplant survival rate is high with only a 10% failure rate in the first year after transplantation, raising to 30% in the first 7 years (1). The main cause of transplant failure is endothelial rejection, accounting for 34% of corneal transplant failures. The high acceptance of corneal grafts is thought to be due in part to the eye as a site of immune privilege. This was originally believed to be due solely to the limited vascularization of the eye, sequestering it from the systemic circulation and thus protecting it from the immune system. It has since become clear that the eye does have some immune response albeit different from the classic immune response, suggesting that other mechanisms are involved. On such mechanism is Fas-induced apoptosis (2). FasL is expressed on the surface of corneal epithelium and endothelium cells. The FasL on these cells binds the Fas expressed on the surface of infiltrating cells that induces apoptosis of the infiltrating cell, therefore preventing infiltration and ultimately inflammation.

It is likely that FasL expression in the cornea, protects donor cornea from graft rejection. This has been demonstrated in experiments using FasL/Fas-deficient mice. Corneas that do not express FasL are rejected when transplanted into normal mice and Fas-deficient recipients reject normal cornea (3). However, immunological studies have shown that mutations of either the FasL or Fas genes in humans, similar to those of FasL/Fas-deficient mice, are rare and lead to lymphoproliferative syndromes. Therefore any defect in either FasL or Fas in cases of corneal transplant rejection must be tissue and time specific rather than a due to a complete loss as seen in the FasL/Fas-deficient mice. Our purpose was to determine if Fas mRNA expression in the recipient’s blood could be correlated with graft rejection and outcome.



A group of 88 patients undergoing corneal transplantation at the Manchester Royal Eye Hospital was recruited prospectively over 18 months, including 11 patients who subsequently suffered one rejection. A blood sample was taken from the recipients at the time of transplant, at subsequent clinic visits, and during and after episodes of graft rejection. Only classical episodes of rejection were included, i.e., epithelial rejection lines, subepithelial lesions, and frank keratic precipitates and/or Khodadoust lines associated with corneal oedema. Patients were categorized according to potential risk of rejection, i.e., low or high risk. Blood samples were also collected from 36 normal controls. The demographics of the subjects studied is summarized in Table 1.

Table 1
Table 1:
Demographics of normal controls and corneal graft patients

RNA extraction.

RNAzol B (Biogenesis, Poole, UK) (800 μl) was mixed with 200 μl of fresh blood and 100 μl chloroform. Total RNA was extracted from the RNAzol mix according to the manufacturer’s instructions. The resulting RNA pellet was washed three times with ice-cold 70% ethanol, dried, and finally resuspended in 20 μl 0.5% NP-40 containing 1:100 dilution of RNase inhibitor (Boehringer Mannheim, Lewes, UK).

Reverse transcription-PCR.

Fas and β-actin mRNA expression in blood was detected by RT-PCR from corneal recipients at the time of transplant, at routine clinic visits and during episodes of rejection. RT-PCR was performed using ready to go RT-PCR beads (Amersham Pharmacia Biotech, Bucks, UK) according to the manufacturers instructions. Fas was amplified with primers; Fas-F 5′tgg aaa taa act gca ccc gga cc3′ and Fas-R 5′ggc ttc att gac acc att ctt tcg3′ by 35 cycles of 45 sec at 94°C, 60 sec at 62°C, and 45 sec at 72°C, producing a product of 443 bp. The expression of the housekeeping gene β-Actin was used as a control of RT efficiency. β-Actin was amplified with primers; actin-F 5′cgt tgc tat cca ggc tgt gc3′ and actin-R 5′gta gtt tcg tgg atg cca ca3′ by 25 cycles of 45 sec at 94°C, 60 sec at 50°C, and 45 sec at 72°C, producing a 434-bp product. Of mRNA (0.5 μl) was used for the β-actin PCR and 0.5, 1, 2, and 3 μl of mRNA for the Fas PCR. The difference in volume of mRNA used in the Fas PCR reactions was taken into consideration when calculating results. The level of Fas gene expression was determined by comparison to β-actin expression by gel densitometery using the GeneTools gel analysis package (Syngene, Cambridge, UK). This semiquantitative measurement was represented as the ratio of Fas/β-actin. One of the control samples was used as a standard and added to each PCR/gel run to allow comparisons between samples run at different times.

Statistical methods.

A two-tailed t test for two samples assuming unequal variances was used to determine the significance of the frequency difference between the test groups and normal controls (significance level 5%).


Fas mRNA was detected in the blood samples of controls and patients at their first routine clinic visit and at times of rejection episodes. Figure 1 shows the varying levels of Fas mRNA in these patients. Fas mRNA levels expressed as a Fas/β-actin mRNA ratio for the three groups studied is illustrated in Figure 1. Samples taken at routine clinic visits from patients not suffering corneal graft rejection showed there was no statistical difference in Fas mRNA expression levels (Fas/β-actin mRNA ratio: mean=0.52; SEM=0.032) from the normal control group (Fas/β-actin mRNA ratio: mean=0.49 SEM=0.063). However, samples taken at the time of a rejection episode showed Fas mRNA levels (Fas/β-actin mRNA ratio: mean=0.3; SEM=0.042) were significantly lower in these patients than either normal controls (P =0.017) or corneal transplant recipients not undergoing graft rejection (P =0.00052).

Figure 1
Figure 1:
Fas mRNA expression. a, Fas (a) and β-actin (b) RT-PCR products separated by agarose gel electrophoresis. The level of Fas gene expression was measured by densitometry and represented as the ratio of Fas/β-actin mRNA. M: DNA markers, 1–9: patient samples. b, Levels of Fas mRNA levels in recipient blood samples, expressed as a ratio of Fas:β-actin. All samples in the nonrejecting group were taken at the first routine clinic visit, 3 to 6 months after grafting. The lines indicate the mean and SEM. [circo], No previous graft; [trio], previous graft in fellow eye; [squlo], previous graft in same eye; [circf], normal control. c, Longitudinal studies of Fas mRNA expression for five corneal graft recipients who suffered an episode of graft rejection (CG-2, GC-7, GC-65, CG-110, and GC-155). The * symbol marks the time of the episode of rejection. After treatment, the rejection was resolved in all five patients.

Table 2 and Figure 1c summarizes the data on patients under going graft rejection. It is of interest that one patient (CG-2) in particular has an extremely low level of Fas mRNA. This 22-year-old patient suffering from keratoconus was considered at low risk of graft rejection. However, he has had two episodes of rejection. The first rejection episode was 17 days after grafting and the second after 95 days. During the second period of rejection levels of Fas mRNA were found to be 10% of the mean for the patients suffering no rejection, unfortunately no sample was available for the first episode of rejection. A sample taken at a visit 1 month after the second rejection episode was successfully treated showed an increase in the Fas/β-actin mRNA ratio to 0.53.

Table 2
Table 2:
Corneal graft patients undergoing episodes of rejection.

Serial samples were available for five patients who suffered an episode of graft rejection, including samples before and after the rejection episode. Fas mRNA levels for these samples are represented in Figure 1c. The level of Fas mRNA is low during the rejection episode and increases after treatment. There is a significant increase in Fas mRNA levels in the rejection group between time of rejection (mean=0.3) and after the rejection is resolved (mean=0.64, P =0.022). There is also temporal variance in Fas mRNA levels in the controls and nonrejecters group. However, in nonrejecting group where there is no significant difference between samples taken at the first and second clinic visit post graft (approximately 3–6 months apart) or with controls were there is no significant difference in Fas mRNA levels between samples taken 5 months apart.


The level of Fas mRNA is significantly lower in corneal graft patients during rejection compared to either normal controls or non-rejecting graft patients. It is unclear whether this decrease in Fas mRNA contributes to graft rejection or is a consequence of the rejection. However, decreased Fas mRNA appears to be common to all patients suffering immunological rejection, at both high and low risk. Changing levels of Fas mRNA occur naturally in normal individuals, for example the expression of Fas on T cells increases with age (4). All patients experiencing corneal graft rejection had Fas mRNA levels lower than the normal average, including patients between the ages 16 and 80 yr, which suggests that low Fas mRNA levels found in graft rejection is not age related.

Fas mRNA expression in normal controls and nonrejecting corneal graft patients showed a similar wide range of Fas mRNA levels, whereas patients undergoing corneal graft rejection demonstrated a lower range of Fas mRNA expression. However, the lowest Fas mRNA levels found in the nonrejecting graft patients over lapped the range of the rejecting patients, demonstrating that a low Fas mRNA level may contribute to graft rejection, but is not the only factor involved. Acceptance of a corneal graft relies on the immune privilege of the site it is grafted into, as well as FasL-induced apoptosis. Therefore, other factors involved in rejection are likely to include defects in the immune privilege of the recipient’s eye.

The immune privilege of the anterior chamber of the eye relies on many contributory factors including the antiinflammatory environment of the aqueous humor. Tumor growth factor-β, tumor necrosis factorα, interleukin 4, interleukin 10, and α melanocyte stimulating hormone (5,6) are involved in maintaining this anti-inflammatory environment. The loss or down-regulation of any of these antiinflammatory factors may contribute to the rejection of a corneal graft. In converse the presence of the proinflammatory cytokines, interleukin- (IL) 1β, IL-2, IL-12, and interferon-γ, are known to inhibit ocular immune privilege and their up-regulation may contribute to graft rejection (5). Peripheral lymphocytes with reduced levels of Fas on their surface can evade FasL- induced apoptosis and therefore be able to survive in the anterior chamber. The antiinflammatory environment of the aqueous humor would normally render these cells harmless, however, when combined with a reduction of the antiinflammatory environment of the aqueous humor their presence may lead to corneal graft rejection.

Defects in the donor cornea may also contribute to graft rejection. The presence of donor dendritic cells (7) and the lack of FasL expression (3) have both been implicated in corneal graft rejection. The presence of either of these factors combined with low levels of Fas on lymphocytes may lead the graft rejection. Unfortunately the unavailability of human corneas at the time of rejection prevents further investigation of this hypothesis.

The low levels of Fas mRNA may be due to a decrease in expression prior to the rejection episode. The longitudinal study of patient CG-7 showed a normal level of Fas mRNA at the time of grafting which was reduced during the rejection episode. To confirm that decreased Fas mRNA is rejection specific, more graft patients, including those who do not reject, need to be studied longitudinally. Samples taken after rejection episodes had been successfully treated showed a significant increase in Fas mRNA and two patients from the longitudinal study showed an increase of Fas mRNA to prerejection levels (samples CG-7 and CG-155). It is unclear if the increase in Fas mRNA is a direct consequence of the treatment or if the increased Fas mRNA is facilitating resolution of the rejection. Glucocorticoids are known to induce apoptosis, though it is unclear whether the apoptosis is induced via a FasL/Fas pathway (8,9).

These preliminary findings indicate defects in Fas-mediated immune privilege may play a role in corneal transplant rejection. Future studies should include more longitudinal studies and further analysis of the cells responsible for this decrease in Fas mRNA. Although whole blood samples were investigated in this study, it is likely that only a subset of cells is involved that may be the same cells, CD8 and CD4 positive T cells, found in rejecting cornea (10). It is possible that Fas expressing cells leave the circulation to infiltrate the eye during rejection episodes and it is the loss of these cells, which leads to the reduction Fas mRNA in the blood.


The authors thank Dr Phil Dyer for help and advise during the preparation of this manuscript.


1. Williams KA, Muehlberg SM, Lewis RF, Coster DJ. How successful is corneal transplantation—a report from the australian corneal graft register. Eye 1995; 9: 219.
2. Griffith TS, Brunner T, Fletcher SM, Green DR, Ferguson TA. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 1995; 270: 1189.
3. Stuart PM, Griffith TS, Usui N, Pepose J, Yu XH, Ferguson TA. CD95 ligand (FasL)-induced apoptosis is necessary for corneal allograft survival. J Clin Invest 1997; 99: 396.
4. Aggarwal S, Gupta S. Increased apoptosis of T cell subsets in aging humans: altered expression of Fas (CD95), Fas ligand, Bcl-2, and Bax. J Immunol 1998; 160: 1627.
5. Rocha G, Deschenes J, Rowsey JJ. The immunology of corneal graft rejection. Crit Rev Immunol 1998; 18: 305.
6. Taylor AW, Streilein JW, Cousins SW. Identification of alpha-melanocyte stimulating hormone as a potential immunosuppressive factor in aqueous-humor. Curr Eye Res 1992; 11: 1199.
7. He YG, Niederkorn JY. Depletion of donor-derived Langerhans cells promotes corneal allograft survival. Cornea 1996; 15: 82.
8. Seki M, Ushiyama C, Seta N, et al. Apoptosis of lymphocytes induced by glucocorticoids and relationship to therapeutic efficacy in patients with systemic lupus. Arthritis Rheum 1998; 41: 823.
9. Horigome A, Hirano T, Oka K, et al. Glucocorticoids and cyclosporine induce apoptosis in mitogen-activated human peripheral mononuclear cells. Immunopharmacology 1997; 37: 87.
10. Larkin DFP, Alexander RA, Cree IA. Infiltrating inflammatory cell phenotypes and apoptosis in rejected human corneal allografts. Eye 1997; 11: 68.
© 2001 Lippincott Williams & Wilkins, Inc.