Exp. Eye Res. (1999) 68, 341–346Article No. exer.1998.0611, available online at http:}}www.idealibrary.com on

Analysis of Sex-mismatched Human Corneal Transplants by

Flourescence in situ Hybridization of the Sex-chromosomes

GREGOR WOLLENSAK* WILLIAM RICHARD GREEN

Eye Pathology Lab, Wilmer Eye Institute, Johns Hopkins Hospital, Baltimore, MD 21287-9248, U.S.A.

(Received Lund 11 June 1998 and accepted in revised form 10 November 1998)

The fate of the cells of corneal transplants has been controversial from the early days of keratoplasty.Various methods such as histological evaluation, radiolabeling of donor cells or Barr-body analysis havebeen applied to clarify the issue. However, the question whether the transplanted cells are replaced orsurvive, remains unsolved.

In this study, we applied fluorescence in situ hybridization (FISH) of the X- and Y-chromosomes inparaffin sections of explanted sex-mismatched corneal transplants to distinguish between host and donorcells. Fourteen sex-mismatched cases with various reasons for explantation and different postoperativetime intervals ranging from 11 months to 30 years were analysed. We found that all cell types, includingepithelium, keratocytes and endothelial donor cells were replaced in most cases as early as 1 year aftertransplantation. In three cases, however, up to 26% of donor keratocytes were still detected up to 4±5years after transplantation, demonstrating a certain individual variability in the process of replacement.Further studies must show if the extent and timing of donor cell replacement in clinically successful,totally clear transplants is different.

Our results are in keeping with the phenomenon of recurrences of corneal dystrophies in the graft, thesignificant postoperative decline of the endothelial cell density, the fact that typical graft rejections usuallytake place within 1–2 years postoperatively and that relatively late rejections can occur in rare casesprobably due to some surviving stromal keratocytes.

Donor cell replacement is a special feature of corneal transplants when compared with other kinds oforgan transplants and might be due to the presence of the same tissue type in the immediateneighbourhood of the graft. # 1999 Academic Press

Key words : FISH; sex chromosomes; sex-mismatch; corneal transplantation; transplant replacement;transplant survival.

1. Introduction

The first successful human corneal transplantation

was performed by Zirm in 1905. From those early

beginnings of keratoplasty until today the fate of the

cells of the corneal transplant has been a controversial

issue. Salzer, Bonnefon, Lacoste, and Loehlein were

strong advocates of the hypothesis that the corneal

transplant is completely replaced by the recipient cells

while Fuchs, Ascher, Elschnig, Castroviejo and Filatov

favored the hypothesis that the corneal transplant

retains its identity and the transplanted cells survive

(Maumenee and Kornblueth, 1948; Salzer, 1990).

These opinions, however, were based mainly on the

histopathological analysis of cases with no reliable

possibility of distinguishing between donor and host

cells. Especially, the obvious survival of Bowman’s and

Descemet’s membrane was a strong argument in favor

of the survival theory while the successful trans-

plantation of non-viable, formalinized corneal trans-

plants backed up the replacement theory (Salzer,

1919; Scherschewskaya, 1940).

Later this question was addressed using the Barr-

body identification in sex-mismatched corneal trans-

plants (Basu and Carre! , 1973; Bourne, 1974; Espiritu,

* Corresponding author: Gregor Wollensak, Wildentensteig 4, D-14195 Berlin, Germany.

Kara and Tabowitz, 1961; Klen and Hradil, 1963;

Kinoshita, Friend and Thoft, 1981). However, the

Barr-body method only allows determination of the

statistical probability of the gender of a piece of tissue

but not of single individual cells. In addition, dis-

appearance of the Barr body has been described with

aging of the cells and the Barr body of the stromal

keratocytes, and mostly also of the endothelium can

only be detected in tangential sections of the cornea or

in cultural outgrowths (Basu and Carre! , 1973; Basu,

Sarkar and Carre! , 1964; Pedler and Ashton, 1955).

Other methods, which cannot be used in humans

for ethical reasons, were: radiolabeling of donor cells

with tritiated thymidine (Khodadoust and Silverstein,

1969; Polack, Smelser and Rose, 1964), labeling of

host cells with trypan blue (Maumenee and

Kornblueth, 1948), and karyotyping of host and

donor cells in heterografts with different species (Basu,

Sarkar and Carre! , 1964).

However, most of these studies yielded controversial

results, partly due to the method or animal model

used, but also, we believe because of the relatively

short postoperative time interval after the trans-

plantation.

Recently, we have applied fluorescence in situ

hybridization (FISH) of the X- and Y-chromosome to

ocular tissue. In situ hybridization allows the local-

0014–4835}99}030341­06 $30.00}0 # 1999 Academic Press

342 G. WOLLENSAK AND W. R. GREEN

ization of specific DNA-sequences by using a labeled

complementary DNA-probe. This method is very

specific and has the advantage that it can identify the

gender of a cell to a high percentage in all cell types

present in the eye, without destroying the ocular

texture (Wollensak, Perlman and Green, 1998). In

situ hybridization of the sex-chromosomes has also

already been successfully used in the analysis of sex-

mismatched skin (Brain et al., 1989; Burt et al.,

1989), lung (Yousem and Sonmez-Alpan, 1991) and

cardiac transplantation (Hruban et al., 1993).

In the present study we have applied this approach

to analyse human eyes with sex-mismatched corneal

transplantations. In contrast to earlier studies, we

were able to use specimens with a postoperative time

interval of up to 30 years.

2. Materials and Methods

Corneal Explants

All corneal buttons were retrieved retrospectively

from the archives of the Eye Pathology Laboratory at

the Wilmer Eye Institute in Baltimore, U.S.A.

We selected 14 sex-mismatched corneal buttons

with a wide range of different postoperative time

intervals ranging from 11 months to 30 years (Table

I). Three of the recipients were male (cases 9, 11 and

13) and 12 were female.

The corneal transplants had been explanted for

various reasons as can be seen in Table I. The gender

of the donor of the corneal graft was established either

from the clinical charts if available or by application of

in situ hybridization of the sex-chromosomes on

paraffin sections of the donor corneoscleral rims which

routinely undergo histologic examination at the

Wilmer Eye Institute.

In addition, two control cases with no sex-mismatch

and metaphase spreads of male peripheral lympho-

cytes were tested (Fig. 1). The two control cases were

T I

Results of the analysis of the 14 sex-mismatched cases

Case number Donor keratocytes Time interval Reason for removal

1 none 11 mos pseudopemphigoid2 none 14 mos endothelial failure3 15% (63 of 420 cells) 2 y, 5 mos graft rejection4 20% (90 of 450 cells) 2 y, 6 mos endothelial failure5 none 3 y, 2 mos endothelial failure6 none 4 y endothelial failure7 none 4 y graft rejection8 26% (94 of 360 cells) 4±5 y endothelial failure9 none 9 y recurrent lattice dystrophy

10 none 11 y recurrent keratoconus11 none 12 y recurrent lattice dystrophy12 none 16 y endothelial failure13 none 20 y endothelial failure14 none 30 y endothelial failure

In no case, were there surviving epithelial or endothelial donor cells.

F. 1. Fluorescence photomicrograph of a peripherallymphocyte (right) and a metaphase spread of a maleperipheral lymphocyte. Note the green signal for the Y-(lower arrow) and the red signal for the X-chromosome(upper arrow). (¬416, oil immerson, DAPI counterstain).

cases of female recipients with female donor buttons

with a postoperative time interval of 1±5 years and 4±5years.

Fixation and Hybridization Conditions

All the specimens used in this study had been fixed

in 10% neutral buffered formalin for about 1 day and

embedded in paraffin. Six micrometer thick paraffin

sections were mounted on silanised slides. Three slides

with two sections were tested in each case.

The DNA-probes for the X (α-satellite)- and Y-

chromosome (classical satellite) were from Oncor,

Gaithersburg, MD, U.S.A. The details of the technical

procedure for the fluorescence in situ hybridization

have been described previously (Wollensak, Perlman

and Green, 1999).

In short, the procedure involved pretreatment with

hot 50%-glycerol for 10 min, and digestion with

proteinase K (0±1 µg ml−") for 45 min at 37°C. Dena-

turation was performed by applying a formamide

SEX-MISMATCHED CORNEAL TRANSPLANTS 343

solution for 10 min at 72°C. Incubation with the

denatured probes was carried out overnight. Dual

detection and dual amplification with a rhodamine-

labeled antibody for the detection of the X- and a

fluorescein-labeled antibody for the detection of the Y-

chromosome was used. The slides were counterstained

with the bluish fluorescent dye 4«-6-diamidino-2-

phenylindole (DAPI).

For the corneoscleral donor rims and the two

control cases the FISH-procedure was identical. The

metaphase spreads of the male lymphocytes were not

digested and were denatured for only 2 minutes.

Fluorescence Microscopy

The specimens were examined under a Zeiss-

axiomat microscope using a triple band pass filter for

fluorescein (460 nm), rhodamine (570 nm) and DAPI

(365 nm). Most of the micrographs were taken with

an attached Zeiss-camera. A few images were captured

by a CCD camera (Photometrics, Tucson, AZ, U.S.A.)

and processed using the Oncor Image analyzing

system.

3. Results

Two red signals for the two X-chromosomes in

female cells, and a green and a red signal in male cells

(Fig. 1) were visualized in about 98% of epithelial (Figs

2 and 3), in about 70% of the stromal keratocytes (Figs

4 and 5) and 85% of endothelial cells (Fig. 5). The

signals had often to be observed at different levels of

focus. Sometimes, distortion of the stromal collagen

fibers occurred but was minimized by placing the slides

on ice at specific steps during the FISH-procedure

(Wollensak, Perlman and Green, 1998).

The FISH-procedure was also successful in the sex

determination of the donor corneoscleral rims (Fig. 6).

In the two control cases with identical gender of the

donor and recipient, only cells of one gender were

found.

In all the 14 sex-mismatched cases, the epithelium

and the endothelium were invariably replaced by cells

of the recipient as early as one year postoperatively

(Table I). The stromal keratocytes were completely

replaced by host cells in 11 cases while a cellular

mosaic (Fig. 4) with 15–26% of donor keratocytes was

present in three cases at 2±5, 2±6 and 4±5 years

postoperatively (cases 3, 4, 8). The surviving stromal

keratocytes were present in small groups of 2–4 cells

which were scattered throughout the corneal stroma

without a preferential localization.

The reason for transplant removal was recurrence

of the underlying dystrophy (3 cases), graft rejection

(2 cases), surface problems (1 case) or endothelial

failure (8 cases) with endothelial counts of up to only

4 nuclei per high power field (¬40). In 2 of the 8 cases

with endothelial failure (cases 5 and 6), pseudophakia

contributed to the endothelial atrophy.

F. 2. Fluorescence photomicrograph of repopulatingfemale epithelium with two red signals per cells (arrows)identifying them as female like the female recipient. (¬416,oil immersion, DAPI counterstain).

F. 3. Fluorescence photomicrograph of repopulatingmale epithelium with a green and red signal per cell (arrows)in the transplanted cornea of a male recipient. (¬488, oilimmersion, DAPI counterstain).

F. 4. Fluorescence photomicrograph of stromal mosaicwith two male keratocytes (green signal for the Y-chromosomes, lower arrows) and one repopulating femalekeratocyte (with two red signals for the X-chromosomes,upper arrow) in a male transplant that was grafted to afemale recipient. The red signals in the male cells are notvisible because they are at another level of focus. (¬496, oilimmersion, DAPI counterstain).

In three cases (cases 1, 2 and 4) a retrocorneal

membrane was present and shown to be of the

recipient gender.

344 G. WOLLENSAK AND W. R. GREEN

F. 5. Fluorescence photomicrograph of repopulatingfemale endothelial and stromal cells with two red signals percell (arrows). (¬496, oil immersion, DAPI counterstain).

F. 6. Fluorescence photomicrograph of trabecularmeshwork of a male donor corneoscleral rim with a greenand a red signal per cell (arrows). (¬488, oil immersion,DAPI counterstain).

4. Discussion

Using in situ hybridization of the sex-chromosomes,

we were able to identify, unequivocally, most of the

cells of the explanted cornea, including epithelium,

stromal keratocytes and endothelium without destroy-

ing histological integrity.

We found a complete and relatively early cell

replacement of the donor cells in the majority of the

cases. In three cases (cases 3, 4 and 8) 15–26% of the

donor keratocytes were still detectable 2±5–4±5 years

after corneal transplantation. These data suggest that

a complete replacement of the donor cells occurs over

time, although there seems to be some individual

variability in terms of the extent and timing.

Factors that might contribute to a shortened cell

cycle or early necrosis of donor cells are : preoperative

cell death in cadaver donor tissue, mechanical trauma

during the operation, attrition from aging, the absence

of epithelial stem cells in the transplant, neuroparalytic

mechanisms, disturbed metabolism because of the

circular scar tissue and the sutures, chronic corneal

edema, postsurgical inflammation, elevated intra-

ocular pressure or pseudophakia.

Similarly, even after minor traumas like corneal

deepithelialization (Campos et al., 1994), photo-

refractive keratectomy (PRK) or laser-assisted in situ

keratomileusis (LASIK) (Helena et al., 1998) apoptosis

of keratocytes and subsequent repopulation has been

described.

All our cases were failed grafts from a clinical point

of view. It is probable that the extent and the timing of

the donor cell replacement in our cases was influenced

by individual features of the cases like corneal edema,

graft rejection etc. Ideally, totally clear grafts from

post-mortem eyes should have been included in the

study. This was not possible as autopsy eyes are

usually fixed longer than two days making the

application of the FISH-technique impossible because

of the increased number of crosslinks caused by

formaldehyde (Wollensak, Perlman and Green, 1999).

In the cases with recurrent corneal dystrophy (cases 9,

10 and 11) there was no corneal edema and the

cornea was clear except for the deposits respectively

apical scarring. Therefore, we think that replacement

of donor cells by recipient cells probably may be a

normal and gradual process which can also be found

in clear, clinically successful grafts and does not

necessarily lead to graft failure.

It is not surprising that the corneal epithelium was

replaced in all cases by recipient cells because the

donor epithelium is often removed prior to trans-

plantation so that repopulation originating from the

limbal stem cells of the host is almost certain.

Replacement of the donor epithelium has also been

observed in animal experiments 12 weeks to 6 months

after corneal transplantation (Khodadoust and

Silverstein, 1969; Kinoshita, Friend and Thoft, 1981).

Similar results in terms of the keratocyte replace-

ment have been reported by others who observed a

stromal replacement as early as 6 months post-

operatively and a maximum survival time of 6±5 years

for the donor cells (Basu and Carre! , 1973; Klen and

Hradil, 1963). Also Maumenee’s classical experiments

in rabbits on the immune reaction against corneal

transplants have demonstrated that an immune

reaction could only be induced up to 6 weeks after

transplantation. Maumenee concluded that the donor

stromal cells were no longer present after this period

(Maumenee, 1951). In addition, Maumenee was able

to destroy transplanted corneas by freezing them and

found repopulation without clouding of the graft from

as early as 11 days after the operation whereas the

grafts became cloudy if they were frozen earlier.

Maumenee concluded that in clear, successful grafts

viable donor keratocytes must be present until the

tissue is firmly attached to the recipient cornea.

Thereafter, repopulation and replacement by host

keratocytes can occur without opacification of the

graft. Formalinized, non-viable donor corneas usually

result in cloudy grafts. Therefore, viable donor

material should be used for corneal transplantation

even though eventually the donor cells are replaced.

SEX-MISMATCHED CORNEAL TRANSPLANTS 345

During the time of the gradual replacement, viable

donor cells seem to be necessary to avoid irreversible

scarring of the cornea (Maumenee and Kornblueth,

1948).

The observed complete cell replacement of the donor

endothelium in the early postoperative years is in good

accordance with clinical studies that have shown that

the endothelial cell count in corneal transplants

decreases up to 65% during the first two postoperative

years which is much more than expected from normal

aging or due to pseudophakia (Waring et al., 1982).

Probably, the repopulating host endothelial cells

migrate towards the center of the tranplanted cornea

with flattening and enlargement of the cells and only

little actual mitotic proliferation (Waring et al., 1982).

It has also been found that the recipient age has an

unexpected significance as a predictor of endothelial

cell survival in the respect that endothelial cell counts

decrease with increasing recipient age regardless of

the donor age (Linn et al., 1981; Williams et al.,

1997). In animal experiments, survival of endothelial

cells has been observed from 5 weeks to 12 months

postoperatively. (Bourne, 1974; Polack et al., 1964)

and replacement by host endothelial cells between 4 to

7 months postoperatively (Espiritu, Kara and Tabo-

witz, 1961).

The time scheme which emerges from our results, is

in good accordance with the clinical observation of

graft rejection being mostly in the first two years but

also happening later in a few cases. In addition, these

results are in keeping with numerous reports about

recurrences of various corneal dystrophies like

Meesmann’s epithelial (Chiou et al., 1998), Reis-

Bu$ ckler’s (Olson and Kaufman, 1978), keratoconus

(Bechrakis et al., 1994), granular (Lyons et al., 1994),

macular (Klintworth et al., 1983), lattice (Klintworth

et al., 1982) or posterior polymorphus dystrophy

(Boruchoff et al., 1990) as early as 2–3 years after

corneal transplantation or later in some cases.

Moreover, recurrences of corneal dystrophies are a

clear clinical proof of the phenomenon of repopulation

of the graft by host cells.

The fact that the donor cornea seems to serve

primarily as a matrix or framework with a complete

replacement of the donor cells by recipient cells in the

long term makes corneal transplantation unique in

comparison with other organ transplants like lung

(Yousem and Sonmez-Alpan, 1991), kidney (Sedmak

et al., 1988; Williams et al., 1969), liver (Gouw et al.,

1987), small bowel (Iwaki et al., 1991) or cardiac

transplantats (O’Connell et al., 1991) where usually

only the tissue macrophages, lymphocytes and the

endothelium of the blood vessels are replaced by the

recipient cells. In addition, in other organ transplants,

no cadaver donor organs are used and vascular

anastomoses are necessary.

The nature of corneal transplantation can perhaps

be compared to blood transfusion (Salzer, 1900) in

contrast to a bone marrow transplantation. Interest-

ingly, also skin grafts (Burt et al., 1989; Plenat et al.,

1992), epidermal (Brain et al., 1989; Gielen et al.,

1987) and fascial allografts (West, Crawford and

Basu, 1969) are completely replaced by the recipient

cells.

In summary, our study has shown that all cell types

of corneal transplants tend to be replaced by recipient

cells in the long term. Epithelium and endothelium are

replaced relatively early within the first postoperative

years. Small proportions of stromal keratocytes can

survive for several years but are probably replaced

eventually. Individual variability in the process of

replacement exists. Further studies must determine if

there is a higher rate of surviving donor cells in totally

clear, clinically successful grafts. Our findings can help

to explain and understand better many features of

corneal transplantation like recurrences of corneal

dystrophies, postoperative decrease of endothelial cell

density and timing of graft rejections.

Acknowledgements

This work was supported by the Deutsche Forschungs-gemeinschaft and The International Order of Odd Fellows,Winston-Salem, North Carolina, U.S.A.

References

Basu, P. K. and Carre! , F. (1973). A study of cells in humancorneal grafts. Can. J. Ophthalmol. 8, 1–7.

Basu, P. K., Sarkar, P. and Carre! , F. (1964). Use of thekaryotype as biologic cell marker. Am. J. Ophthalmol.58, 569–72.

Bechrakis, N., Blom, M. L., Stark, W. J. and Green, W. R.(1994). Recurrent keratoconus. Cornea 13, 73–7.

Boruchoff, S. A., Weiner, M. J. and Albert, D. M. (1990).Recurrence of posterior polymorphous corneal dys-trophy after penetrating keratoplasty. Am. J. Ophthalmol.109, 323–8.

Bourne, W. M. (1974). In vivo survival of cryopreservedendothelial cells in primates. Arch. Ophthalmol. 92,146–8.

Brain, A., Purkis, P., Coats, P., Hackett, M., Navsaria, H. andLeigh, I. (1989). Survival of culturedallogeneic keratino-cytes transplanted to deep dermal bed assessed withprobe specific for Y-chromosome. Brit. Med. J. 298,917–19.

Burt, A. M., Pallett, C. D., Sloane, J. P., O’Hare, M. J.,Schafler, K. F., Yardeni, P., Eldad, A., Clarke, J. A. andGusterson, B. A. (1989). Survival of cultured allograftsin patients with burns assessed with probe specific for Y-chromosome. Brit. Med. J. 298, 915–17.

Campos, M., Szerenyi, K., McDonnell, J. M., Lopez, P. F. andMcDonnell, P. J. (1994). Keratocyte loss after cornealdeepithelialization in primates and rabbits. Arch.Ophthalmol. 112, 254–60.

Chiou, A. G.-Y., Florakis, G. J., Copeland, R. L., Williams,V. A., McCormick, S. A. and Chiesa, R. (1998). Re-current Meesmann’s corneal epithelial dystrophy afterpenetrating keratoplasty. Cornea 17, 566–70.

Espiritu, R. B., Kara, G. B. and Tabowitz, D. (1961). Studieson the healing of corneal grafts. II. The fate of theendothelial cells of the graft as determined by sexchromatin studies. Am. J. Ophthalmol. 52, 91–5.

346 G. WOLLENSAK AND W. R. GREEN

Gielen, V., Faure, M., Mauduit, G. and Thivolet, J. (1987).Progressive replacement of human cultured epithelialallografts by recipient cells as evidenced by HLA class Iantigen expression. Dermatologica 175, 166–70.

Gouw, A. S. H., Houthoff, H. J., Huitema, S., Beelen, J. M.,Gips, C. H. and Poppema, S. (1987). Expression of majorhistocompatibility complex antigens and replacement ofdonor cells by recipient ones in human liver grafts.Transplantation 43, 291–6.

Helena, M. C., Baerveldt, F., Kim, W.-J. and Wilson, S. E.(1998). Keratocyte apoptosis after corneal surgery.Invest. Ophthalmol. Vis. Sci. 39, 276–83.

Hruban, R. H., Long, P. P., Perlman, E. J., Hutchins, G. M.,Baumgartner, W. A., Baughman, K. L. and Griffin,C. A. (1993). Fluorescence in situ hybridization for theY-chromosome can be used to detect cells of recipientorigin in allografted hearts following cardiac trans-plantation. Am. J. Pathology 142, 975–80.

Iwaki, Y., Starl, T. E., Yagihashi, A., Taniwaki, S., Abu-Elmagd, K., Tzakis, A., Fung, J. and Todo, S. (1991).Replacement of donor lymphoid tissue in small-boweltransplants. Lancet 337, 818–19.

Khodadoust, A. A. and Silverstein, A. M. (1969). Thesurvival and rejection of epithelium in experimentalcorneal transplants. Invest. Ophthalmol. Vis. Sci. 8,169–79.

Kinoshita, S., Friend, J. and Thoft, R. A. (1981). Sexchromatin of donor corneal epithelium in rabbits.Invest. Ophthalmol. Vis. Sci. 21, 434–41.

Klen, R. and Hradil, I. (1963). Beitrag zur Biologie desHornhautpfropfes. Sex-chromatin-Charakter der im-plantierten Scheibe. Graefes Arch. Clin. Exp. Ophthalmol.166, 180–7.

Klintworth, G. K., Ferry, A. P., Sugar, A. and Reed, J.(1982). Recurrence of lattice corneal dystrophy type 1in the corneal grafts of two siblings. Am. J. Ophthalmol.94, 540–6.

Klintworth, G. K., Reed, J., Stainer, G. A. and Binder, P. S.(1983). Recurrence of macular corneal dystrophywithin grafts. Am. J. Ophthalmol. 95, 60–72.

Linn, J. G., Stuart, J. C., Warnicki, J. W., Sinclair, R. A. andMarsh, G. M. (1981). Endothelial morphology in long-term keratoconus corneal transplants. Ophthalmology88, 761–70.

Lyons, C. J., McCartney, A. C., Kirkness, C. M., Ficker, L. A.,Steele, A. D. M. and Rice, N. S. C. (1994). Granularcorneal dystrophy. Visual results and pattern of re-currence after lamellar or penetrating keratoplasty.Ophthalmology 101, 1812–17.

Maumenee, A. E. and Kornblueth, W. (1948). Symposium:Corneal transplantation. VI. Physiopathology. Am. J.Ophthalmol. 31, 1384–93.

Maumenee, A. E. (1951). The influence of donor recipientsensitization on corneal grafts. Am. J. Ophthalmol. 34,142–52.

O’Connell, J. B., Renlund, D. G., Bristow, M. R. andHammond, E. H. (1991). Detection of allograft en-dothelial cells of recipient origin following AB0-com-patible, non-identical cardiac transplantation. Trans-plantation 51, 438–42.

Olson, R. J. and Kaufman, H. E. (1978). Recurrence of Reis-Bu$ ckler’s corneal dystrophy in a graft. Am. J. Oph-thalmol. 85, 349–51.

Pedler, C. and Ashton, N. (1955). Sex of nuclei in oculartissues. Brit. J. Ophthalmol. 39, 362–7.

Plenat, F., Vignaud, J.-M., Guerret-Stocker, S., Hartmann, D.and Duprez, A. (1992). Host-donor interactions inhealing of human split-thickness skin grafts onto nudemice: In situ hybridization, immunohistochemical, andhistochemical studies. Transplantation 53, 1002–10.

Polack, F. M., Smelser, G. K. and Rose, J. (1964). Long-termsurvival of isotopically labeled stromal and endothelialcells in corneal homografts. Am. J. Ophthalmol. 57,67–78.

Salzer, F. (1910). Beitra$ ge zur Keratoplastik II. UeberImplantationvonisolirtenSchichtenconservirterPferde-hornhaut in die Cornea des Kaninchens. Arch. f.Augenheilk. 65, 214–28.

Salzer, F. (1900). Kritische und litterarische Studien u$ berTransplantation in Hinblick auf die Frage der Kerato-plastik. Zeitschr. f. Augenheilk. 4, 124–45.

Scherschewskaya, O. J. (1940). Keratoplastische Versuchemit Anwendung von Formollappen. Ophthalmologica99, 4–15.

Sedmark, D. D., Sharma, H. M., Czajka, C. M. and Ferguson,R. M. (1988). Recipient endothelialization of renalallografts. Transplantation 46, 907–10.

Waring, G. O., Bourne, W. M., Edelhauser, H. F. and Kenyon,K. R. (1982). The corneal endothelium. Normal andpathologic structure and function. Ophthalmology 89,531–90.

West, R. D., Crawford, J. S. and Basu, P. K. (1969). The fateof the cellular elements in fascial grafts. Can. J.Ophthalmol. 4, 64–71.

Williams, G. M., Krajewski, C. A., Dagher, F. J., TerHaar,A. M., Roth, J. A. and Santos, G. W. (1969). Hostrepopulation of the endothelium in allografts of kidneyand aorta. Surg. forum. 20, 293–4.

Williams, K. A., Muehlberg, S. M., Lewis, R. F. and Coster,D. J. (1997). Influence of advanced recipient and donorage on the outcome of corneal transplantation. Brit. J.Ophthalmol. 81, 835–9.

Wollensak, G., Perlman, E. and Green, W. R. (1999).Interphase fluorescence in situ hybridization analysis ofthe X- and Y-chromosome in human eyes. Cytogenet.Cell Res. forthcoming.

Yousem, S. A. and Sonmez-Alpan, E. (1991). Use of abiotinylated DNA probe specific for the human Ychromosome in the evaluation of the allograft lung.Chest 99, 275–9.