Severing corneal nerves in one eye induces sympathetic loss of immune privilege and promotes rejection of future corneal allografts placed in either eye

Abstract

Less than 10% of corneal allografts undergo rejection even though HLA matching is not performed. However, second corneal transplants experience a threefold increase in rejection, which is not due to prior sensitization to histocompatibility antigens shared by the first and second transplants since corneal grafts are selected at random without histocompatibility matching. Using a mouse model of penetrating keratoplasty, we found that 50% of the initial corneal transplants survived, yet 100% of the subsequent corneal allografts (unrelated to the first graft) placed in the opposite eye underwent rejection. The severing of corneal nerves that occurs during surgery induced substance P (SP) secretion in both eyes, which disabled T regulatory cells that are required for allograft survival. Administration of an SP antagonist restored immune privilege and promoted graft survival. Thus, corneal surgery produces a sympathetic response that permanently abolishes immune privilege of subsequent corneal allografts, even those placed in the opposite eye and expressing a completely different array of foreign histocompatibility antigens from the first corneal graft.

Keywords: Basic (laboratory) research/science; corneal transplantation/ophthalmology; immune regulation; rejection; tolerance: experimental.

Figure 1. Corneal transplantation in one eye abolishes immune privilege in the opposite eye

A) C3H/Hej (H-2^k) or A/J (H-2^a) corneal allografts were transplanted to the right eyes of BALB/c (H-2^d) mice. Sixty days later the rejected grafts were removed and replaced with C57BL/6 (H-2^b) corneal allografts. C57BL/6 corneal allografts in previous recipients of C3H/Hej corneal allografts (median survival time, MST= 7; N= 18) or A/J corneal allografts (MST =16; N= 8). C57BL/6 corneal allografts rejection in naïve BALB/c hosts (MST = 46; P < 0.001; N= 10) B) C57BL/6 corneal allografts were rejected in 100% of the hosts that had been previously grafted with a syngeneic BALB/c cornea graft (MST = 15.5; N= 15). Rejection of C57BL/6 corneal allografts in naïve hosts (MST = 46; P< 0.001; N=10). BALB/c syngeneic corneal grafts were not rejected by hosts that had been previously grafted with syngeneic BALB/c grafts (N= 12). C) C57BL/6 corneal allografts underwent rejection in hosts grafted in the right eyes with either C3H/HeJ (MST = 12.5; N= 12) or A/J (MST = 16; N=7) corneal allografts (P <0.001). D) Syngeneic BALB/c corneal grafts into right eyes first; BALB/c grafts or C57BL/6 corneal allografts grafted to the left eyes 60 days later. 90% of the C57BL/6 corneal allografts were rejected (MST = 20; N=8), which was significantly swifter than the rejection of C57BL/6 corneal allografts in naïve hosts (MST = 46; N= 10; P< 0.032). None of the syngeneic BALB/c corneal grafts underwent rejection (N= 9). All experiments were performed at least twice with similar results.

Figure 2. Circular corneal incision in one eye robs the other eye of its immune privilege

A) Sutures were inserted into the corneas in right eyes of BALB/c mice before applying C57BL/6 corneal allografts into either eye. 100% of the corneal allograft rejection in pre-sutured eyes (MST =18; N=10); 60% rejection in opposite eye of sutures (MST = 34; N=10), P<0.001. Corneal allograft rejection in naïve BALB/c mice= 50%; MST = 33; N= 10; P>0.05. B) “X” shaped incision in left eye prior to applying C57LBL/6 corneal allografts into the right eye. Rejection in “X” shaped corneal incision group: MST = 36; N= 10. Naïve group: MST = 34; N=10; P>0.05). C) 360° corneal incisions 14 days prior to grafting. Corneal allografts opposite eye of the 360° corneal incision: MST = 18; N= 19; corneal allografts placed into the same eye as 360° corneal incisions: MST =25; N=18; P<0.03; D) 360° corneal incisions in left eyes 60 days (MST = 22; N= 9) and 100 days (MST = 22; N = 10) prior to application of C57BL/6 corneal allografts in the right eyes. P>0.05. Naive controls: MST = 34; N = 10 and MST = 34.5; P<0.05 compared to both corneal incisions groups. E) 360° corneal incisions were made in the left eyes of BALB/c mice with clear C57BL/6 corneal allografts that had been in place in the right eyes for 30 days; F) clear C57BL/6 corneal allograft on right eye of a BALB/c mouse 30 days after transplantation and one day before trephining the left eye. G) Same graft shown in Figure F, but 22 days after making circular incisions in the opposite eye. All experiments were performed at least twice with similar results.

Figure 3. Effect of circular corneal incisions on corneal nerves

A) Cornea from a normal BALB/c mouse; B) cornea collected from a BALB/c mouse 96 hr after being scored with shallow360° incisions created by a 2.0 mm trephine. Arrows indicate site of circular corneal incision. C) Cornea collected 96 hr after being scored with an “X” shaped incision. Corneal nerves were stained with rabbit anti-mouse β tubulin III primary antibody (TUJ₁) IgG (Covance, Richmond, CA) and were viewed by confocal microscopy. “C” indicates center of cornea. D) normal cornea. E) Cornea with circular incision (stained with India ink). F) Cornea with “X-shaped” incision (stained with Evan’s blue).

Figure 4. Circular corneal incisions induce SP expression in the opposite eye, which abolishes immune privilege of corneal allografts in both eyes

360° incisions were made in the right eyes of BALB/c mice and EIAs were used to quantify: A) SP, B) α-MSH, C) VIP, and D) CGRP expression in the anterior tissue of the left eye. Horizontal line represents baseline expression of each neuropeptide in eyes of normal mice. E) SP (1.0 pg) was injected subconjunctivally into right eyes of BALB/c mice one day before placing a C57BL/6 corneal allograft onto either the right or left eye (N= 15; MST = 19.5; P = 0.02 and N = 8 ; MST = 17; P = 0.02, respectively) in each group); F) SP (1.0 pg) was injected i.v. into BALB/c mice either 1 or 100 days prior to their receiving C57BL/6 corneal allografts (N= 15; MST = 18.5 and N = 9; MST = 28, respectively); SP groups vs untreated (N = 10; MST = 34) P = 0.04. G) SP (0.1 or 1.0 pg) was injected i.v. into BALB/c mice one day prior to the application of C57BL/6 corneal allografts. N = 10 for 0.1 pg group; N= 9 for 0.01 pg group. P= 0.001. H) 360° corneal incisions (trephine) were made in the left eyes of BALB/c mice. Sixty days later mice Spantide II (72 μg/day; i.p. N = 8 MST = 23) treatment was initiated and mice were grafted with C57BL/6 corneal allografts 24 hr after Spantide treatment was initiated. Mice received Spantide II for 60 consecutive days. Spantide II group vs. other two groups. (60 day trephine N = 10; MST = 19; P = 0.017 .Naïve recipients N = 10; MST = 34 P > 0.05). All experiments were performed at least twice with similar results.

Figure 5

Effect of exogenous α-MSH on corneal allograft survival. A) 360° incisions (trephine) were made in the right eyes of BALB/c mice on day 0 and α-MSH (2 μg; i.p.) was injected on days -1 and every 3 days thereafter for 15 days. C57BL/6 corneal allografts were transplanted to the left eyes on day 0 (N =12) The right eyes of other BALB/c mice received 360° incisions and C57BL/6 corneal allografts were transplanted to the left eyes one day later, but the mice were not treated with α-MSH (N= 19). P>0.05. B) BALB/c mice were either untreated (N= 10) or treated with α-MSH (2μg; i.p.)(N = 12) beginning on day -1 and continuing every 3 days thereafter for 15 days. This experiment was performed twice with similar results.

Figure 6. Blockade of SP does not enhance immune privilege of first-time corneal allografts

A) C57BL/6 corneal allografts were transplanted to BALB/c mice that were either treated with Spantide II or were injected with vehicle (untreated). Spantide II was administered i.p. at a dose of 36 μg/day beginning one day prior to the creation of circular incision and continued until day 60 post transplantation. N= 10; MST = 34 for untreated group; N= 8; MST = 30 for Spantide II treated group. P >0.05. B) BALB/c corneal allografts were transplanted to either C57BL/6 wild-type (WT) mice; N=20 or SP^−/− (SP KO) mice; N=9. P>0.05. These experiments were performed twice with similar results.

Figure 7. SP induced by the severing of corneal nerves prevents the generation of T regs in corneal grafted hosts

A) CD4+ CD25+ T cells from BALB/c mice that had accepted C57BL/6 corneal allografts, rejector mice, or BALB/c mice that had been treated with circular incisions and that rejected C57BL/6 corneal allografts on the contralateral eye were tested for suppressive activity in a conventional LAT assay. B) CD4+ T cells were isolated from untreated BALB/c mice or mice subjected to 360° corneal incisions (trephine) fourteen days prior to receiving C57BL/6 corneal allografts on the opposite eye and were tested in a LAT assay using CD4+CD25+ T regulatory cells from graft acceptor mice. C) CD4+ CD25+ T cells were isolated from untreated BALB/c mice or mice injected subconjunctivally with SP (1.0 pg) one day prior to corneal transplantation in the other eye. CD4+CD25+ T cells from untreated graft acceptor mice, untreated graft rejector mice, and SP-treated mice (100% graft rejection) and were tested in a LAT assay. D) CD4+ CD25+ T cells were isolated from naïve BALB/c mice that had surviving C57BL/6 corneal allografts on day 30 and used in a LAT assay. SP (1.0 pg) was co-injected into the ears along with the other cells to determine if T reg suppression would be abolished. All experiments were performed twice with similar results. N=5 mice/group.

Figure 8. Effect of circular corneal incisions on the expression of CTLA-4, GITR, TGF-β1, and Foxp3 in CD4+CD25 T cells

A) FACS profile for T reg markers. B) % of cells expressing T reg markers. 360° corneal incisions (trephine) were made in the left eyes of BALB/c mice on day 0 and C57BL/6 corneal allografts were transplanted to the right eye 14 days later. Untreated mice that did not reject their corneal allografts by day 21 were deemed to be “acceptors”. CD4+CD25+ T cells were isolated from the spleens 21 days after transplantation and examined by flow cytometry for cell surface expression of CTLA-4, GITR, or TGF-β and cytoplasmic expression of Foxp3. These experiments were performed twice with similar results.