Ta

C

FEB(

0d

5

Experimental

emporal Relationship Between Successful Pancreas Transplantationnd Control of Ocular Complications in Alloxan-Induced Diabetic Rats

.T. Spadella, J.L.M. Machado, M.M. Lerco, E.V.P. Ortolan, S.A. Schellini, and E.A. Gregório

ABSTRACT

Purpose. The impact of pancreas transplantation (PT) on the progression of eye diseaseis still controversial. This study evaluated the course of retinopathy in transplanted rats intwo different diabetic stages.Methods. Sixty inbred male Lewis rats were assigned to four experimental groups:NC—15 nondiabetic control rats; DC—15 untreated diabetic control rats; PT1—15diabetic rats that received syngeneic pancreas transplants 2 weeks after alloxan diabetesinduction; PT2—15 diabetic rats that received pancreas transplants 12 weeks after diabetesonset. Clinical and laboratory parameters and lens opacity were examined in all rats priorto treatment and at 1-, 6-, and 12-months follow-up. Nucleated eyes from five rats in eachgroup processed for ultrastructural study of the retinal at 6 and 12 months after PT or atfollow-up.Results. Cataracts were observed in 20%, 60%, and 100% of DC rats at 1-, 6-, and12-months follow-up, respectively. Early PT (2 weeks) significantly reduced the prevalenceof this complication but not late (12 weeks) PT. PT1 rats also showed improvedultrastructure of the superficial and deep capillary plexuses of the retina, and of Müllercells, compared with DC and PT2. In the last group, retinopathy continued to evolvedespite successful PT.Conclusion. Our results suggested that prevention of diabetic ocular lesions by PT was

closely dependent on earlier performance of the procedure.

rom the Departments of Surgery (C.T.S., J.L.M.M., M.M.L.,.V.P.O.), Ophthalmology, (S.A.S) and Morphology (E.A.G.),otucatu School of Medicine, São Paulo State University

Unesp), Botucatu, SP, Brazil.

Research supported by Fapesp.Address reprint requests to C.T. Spadella, MD, Faculdade de

Medicina de Botucatu, Unesp, CP: 18618-970, Botucatu, SãoPaulo, Brazil. E-mail: spadella@fmb.unesp.br

041-1345/08/$–see front matter © 2008 by Elsevier Inc. All rights reserved.oi:10.1016/j.transproceed.2008.01.049 360 Park Avenue South, New York, NY 10010-1710

18 Transplantation Proceedings, 40, 518–523 (2008)

LtcibbcfhpaoI“dott

MA

SrnPwsNg

(mmpa

C

Aaoi

E

E1mctf

apt6

tWw

vwvrflbL

S

Cas

RC

NtIsppdfh

C

CaP(pwa

U

Dgf1

rptlbdlnll

tbM

CONTROL OF OCULAR COMPLICATIONS 519

ONG-TERM STUDIES have strongly suggested thattight control of blood glucose can significantly prevent

he development and retard the progression of chronicomplications in diabetes mellitus.1,2 However, these stud-es are controversial, because retinopathy progression haseen observed in a large number of patients after strictlood glucose control, even among those undergoing suc-essful pancreas transplantation.3,4 This may be due to theact that most recipients received a pancreas graft afteraving had diabetes for over two decades, probably whenancreas transplantation (PT) was no longer effectivegainst ocular lesions. Nevertheless, whole PT remains thenly reliable option to achieve long-term normoglycemia.5

t is therefore important to determine whether there is apoint of no return” for diabetic ocular complicationsespite treatment. This investigation evaluated the coursef retinopathy and cataracts among diabetic rats after PT atwo stages, seeking to contribute to the understanding ofhe pathophysiology and treatment of diabetes.

ATERIALS AND METHODSnimals and Groups

ixty inbred male Lewis rats, approximately 3 months old, wereandomly assigned to four experimental groups (n � 15): NC—ondiabetic control group; DC—untreated diabetic control group;T1—diabetic group receiving syngeneic pancreas transplants at 2eeks after diabetes induction; and PT2—diabetic group receiving

yngeneic pancreas transplants at 12 weeks after diabetes onset.ormal inbred Lewis rats were used as donors for both transplanted

roups.Diabetes was induced by intravenous administration of alloxan

Sigma Chemical Co, St. Louis, Mo, USA) in a single dose of 42g/kg body weight. Only diabetic rats showing severe clinical andetabolic alterations were included in the experiment. PT was

erformed according to an original procedure described by Lee etl6 as modified in our laboratory.7

linical and Laboratory Analysis

t the beginning of follow-up or prior to transplant and at 1-, 6-,nd 12-month follow-up, body weight, water and food intake, urineutput, blood glucose, glycated hemoglobin (Hb1Ac), and plasma

nsulin were documented for all groups.

ye Ultrastructural and Morphometric Analyses

yes from five rats in each subgroup were enucleated after 6- or2-month follow-up. The ocular globe dissected under a specularicroscope was sagitally sectioned at the equatorial level. The

rystalline was harvested; retina and choroid were separated fromhe sclera. The retina cut at the posterior pole was sliced into smallragments for fixation in 2.5% glutaraldeyde.

Eye tissue fragments were postfixed in osmic acid, dehydrated,nd embedded in epoxy resin (Araldite). Five tissue blocks wererepared from each eye. Ultrathin sections were examined byransmission electron microscopy at final magnifications between500 and 21,000�.About 16 electron micrographs were performed on each eye

issue block, resulting in a total of 80 micrographs at each time.herever possible, electron micrographs were taken of structures

ith linear images and without angles or artifacts. For retinal c

ascular analysis, five superficial plexus and five deep layer vesselsere selected from each examined eye. Tangentially sectionedessels were excluded from analysis. In this investigation weecorded cataract prevalence index and ultrastructural findingsrom all retinal layer cellular components, mainly Müller, endothe-ial, and pericytic cells, as well as inner and outer capillaryasement membranes (BM). Digital image morphometry witheica Quin Lite 2.5 was used to measure the BM thickening.

tatistical Analysis

linical, laboratory, and morphometric data were evaluated bynalysis of variance, complemented by the Tukey-Kramer test, withtatistical significance set at P � .05.

ESULTSlinical Findings

ondiabetic control rats (NC) showed clinical and labora-ory parameters within normal ranges throughout the study.n contrast, untreated diabetic control rats (DC) revealedignificantly increased water and food intakes, urine out-uts, blood glucose, and glycated hemoglobin levels com-ared to NC rats. Also, plasma insulin was significantlyecreased in DC rats. All metabolic parameters in success-ul pancreas-transplanted rats (PT1 and PT2), includingemoglobin A1c (Figure 1) were restored to normal levels.

ataract Index

ataracts were observed in 20%, 60%, and 100% of DC ratst 1-, 6-, and 12-month follow-up, respectively. SuccessfulT significantly reduced the prevalence of this complicationP � .01) when performed earlier (PT1) but not whenerformed later (PT2). Cataract indices in these groupsere: 0%, 26.6%, 33.3% and 20%, 53.3%, 93.3% at 1, 6,nd 12 months after transplantation, respectively.

ltrastructural and Morphometric Analyses

iabetes affected all retinal cellular components in the DCroup. Ocular changes, which were already visible in ratsrom this group at 6-month follow-up, had become severe at2 months after diabetes onset.The ultrastructural alterations seen in Müller (M) cells

epresented what occurred in most other rat retinal com-onents. In these animals, M cells take up almost the entirehickness of the retina from the external to the internalimiting membrane. These cells are easily distinguishedecause of their nuclei arranged in two layers, their higherensity, and their typical hexagonal shape (Fig 2A). The

ong cytoplasmic prolongations of M cells permeate theuclei of the photoreceptor cells in the external nuclear

ayer and the synaptic endings in the external plexiformayer.

In DC rats, M-cell nuclear chromatin was less condensedhan in controls of 2-week pancreas-transplanted rats (PT1),ut not 12-week transplanted rats (PT2). The DC group-cell nuclei also showed eccentric circular electron-dense

ondensations near the karyoteca. In the cytoplasm, edem-

adolldaalu1

tesCa

ncl“

grwm

odp

atstp

(1ijhovotEwpwr(i

Fmef

520 SPADELLA, MACHADO, LERCO ET AL

tous mitochondria, increased glycogen, dense bodies, cellebris, and electron-dense granulations were frequentlybserved. Cytoplasmic prolongations, especially perivascu-

ar ones, often appeared to be completely filled withysosomes and residual ovoid bodies with variable electronensity patterns (Fig 2B, 2C). All these M-cell changes werelso observed in NC and pancreas-transplanted rats, mainlyt 12-month follow-up. However, lesions were significantlyess intense in NC and PT1 than DC and PT2 groups. M-cellltrastructure in these groups did not differ at 6- and2-month follow-up.The rat retinal capillary net had ultrastructural charac-

eristics shared by both vascular plexuses in the fourxperimental groups. Capillaries were of a continuous type,ince endothelial (E) cells are joined by tight junctions.ytoplasm projections toward the vascular lumen couldlso be observed close to the E-cell junctions.

DC rat E cells showed elongated nuclei with disperseduclear chromatin and electron-dense granulations. In theytoplasm there were pinocyotic vesicles, dense bodies, andysosomes with increased glycogen and vacuoles containingmyelin figures.”

Pericytic (P) cells were also severely affected in the DCroup. These cells had smaller nuclei than those from NCats, with condensed chromatin. In the cytoplasm thereere also an increased number of giant cells and edematous

ig 1. Mean � SD glycated he-oglobin (Hb1AC) in the four

xperimental groups throughoutollow-up.

itochondria, dense bodies, glycogen deposits, and vacu- fi

les containing cell debris. Some P cells showed intenseisorganization, with complete hyalinization of the cyto-lasm.E- and P-cell changes in DC rats were also significantly

meliorated in PT1 rats at 6 and 12 months after transplan-ation. In this group age-dependent ocular lesions wereimilar to NC rats. Otherwise, there were no benefits fromhe procedure on E- and P-cell lesions when PT waserformed at 12 weeks after diabetes onset (PT2).BM, between (inner BM) and external to E and P cells

outer BM), were thicker in DC than control rats, mainly at2 months after diabetes onset. This thickening was morentense in the outer BM, especially in regions close tounctions between the outer and inner BM. BM also showedeterogeneous osmiophilic accumulations located inside orutside this structure. Some long-term diabetic rats re-ealed clear and electron-transparent spaces inside theuter BM. There were occasional outer BM projectionsoward adjacent glial cells, which were not accompanied by

and P cells. Red blood cells, leukocytes, and plateletsere seen inside the vascular lumen, but never in theerivascular space. Retinal bleeding and microaneurysmsere not also seen in diabetic rats. Similar to other analyzed

at retina capillary net structures, early (PT1) but not laterPT2) pancreas transplant improved all cellular alterationsn BM membranes. Figure 3 illustrates the ultrastructural

ndings in E and P cells from retinal superficial vascular

pHo1Dsmg

D

Tohqmro

alpcooagco

c

tsir

oitbartsglieeoSdcab

seis

F(e 0,00

CONTROL OF OCULAR COMPLICATIONS 521

lexuses in diabetic (A, B, C), PT1 (D, E, F), and PT2 (G,, I) groups. Table 1 shows morphometric measurements

f the outer BM areas in all four experimental groups at2-month follow-up. There was marked BM thickening inC rats compared to NC rats. PT1 group rats, however,

howed significantly reduced changes (P � .001) at 12onths after transplantation compared to DC and PT2

roups.

ISCUSSION

he metabolic changes caused by diabetic status affect allrgans and systems. The retina, one of the organism’sigh-metabolism tissues, is also sensitive to the conse-uences of these alterations. Diabetes in rats inducesorphological and ultrastructural changes in crystalline and

etinal vessels similar to those observed in humans andther mammals.8

In our study M, E, and P retinal cells were equallyffected in all untreated diabetic control rats. The diabeticesions were characterized by an accelerated involutionrocess, involving abnormal accumulation of substances inell cytoplasm, such as glycogen, dense bodies, cell debris,smiophilic corpuscles, and an elevated number of pinocy-tic vesicles and vacuoles. Alloxan-diabetes also induced E-nd P-cell losses with significant changes in nuclei, or-anelles, and the inner and outer BM between E and Pells. These diabetic alterations have also been observed inther experimental studies.9,10

The increased number of intracytoplasmic vesicles in

ig 2. (A) Internal nuclear layer of the retina in a normal control6500�) (B, C) Müller-cell cytoplasmic prolongations (M) in diabelectron-transparent spaces (*), and numerous lysosomes (Ly) (1

ontinuity with E or P plasma-cell membranes suggest a l

ransport function. Studies by Raviola and Butler11 haveupported the idea that these structures may be implicatedn the removal of potentially toxic molecules from theetinal interstitium to the intravascular lumen.

In addition, glycogen deposits and increased necroticrganelle remnants, cell debris, and dense bodies have been

nterpreted as signs of cellular degeneration.12 Similarly,he increased number of vacuoles and lysosomes seems toe necessary not only for removal of necrotic substances butlso for the breakdown of excess glycogen accumulated inetinal cells under diabetic conditions.13 Since M cells havehe ability to digest and eliminate residual bodies and toxicubstances and are responsible for the storage and supply oflycogen to the retina, it is easy to understand why there arearge numbers of lysosomes, residual bodies, and glycogenn these cells. Several mechanisms have been postulated toxplain the retinal cell alterations in diabetes; namely,xcessive fatty acid oxidation and exacerbated productionf advanced glycation end products (AGEs). According totitt et al,14 AGEs can initiate blood-retinal barrier break-own even under normoglycemic conditions. Otherwise, aoncomitant increase in the expression of the vasoperme-bilizing agent, vascular endothelial growth factor, in dia-etes can lead to vascular lesions in the retina.Nuclear alterations observed mainly in M-cell nuclei,

uch as descondensed chromatin and the presence oflectrodense condensation close to the karyoteca, may benterpreted as elevated protein synthesis or a sign of celluffering and death.15

Several types of cell accumulations involving collagen,

t 12-month follow-up showing Müller (M) and amacrins (A) cellsts at 12-month follow-up showing ovoid dense bodies (arrows),0 to 11,400�).

rat atic ra

ipids, ferritin, and laminin have also been identified in rat

rt

rcaamstoa

bpgsr

l2cp

Fpnptw

522 SPADELLA, MACHADO, LERCO ET AL

etinal BM.16 These accumulations may result in BMhickening as observed in our animals.

Nevertheless, the ocular changes observed in our diabeticats were also seen in older animals from the normalontrol group. They have been considered to be a sign ofge-related involution.12 Diabetes, however, seemed toccelerate this process, as all retinal lesions were markedlyore intense in diabetic than control animals. Despite the

everity of diabetic retinopathy (DR), the clinical control ofhis pathology is one of the most challenging problems inphthalmology; the pathogenesis and optimal treatments

ig 3. Capillaries of the retinal superficial plexuses (21,000�)erycitic (p) cells of a DC rat with vascular lumen (L) containingumerous pinocyotic vesicles (v), vacuoles containing cell debriscells of a PT1 rat showing complete relief of all the ultrastructur

hickening. (G, H, I) e and p cells of a PT2 rat showing that pancrhen pancreas transplant was performed later (12 weeks).

re not fully understood. e

Evidence suggests that normalization of glycemia haseneficial effects on the course of DR.1,2,17 However,rogressive DR has been reported despite normal bloodlucose values after diabetes treatment.18 These controver-ial results only suggest that there is probably a “point of noeturn” in ocular complications despite treatment.

Our study demonstrated a positive effect of PT on eyeesions in diabetic rats, when surgery was performed within

weeks after diabetes onset. In contrast, cataract and DRontinued to evolve in PT rats when the procedure waserformed later (12 weeks), despite normoglycemia at all

s at 12-month follow-up showing: (A, B, C) endothelial (e) andcells (h), edematous mitochondria (m), dense bodies (arrows),

nd marked basement membrane thickening (bm). (D, E, F) e andions observed in the DC rat, including significantly improved bmransplant did not change any ultrastructural lesions in the retina

in ratblood(*), aal leseas t

xperimental times. These findings have been reported by

MurovnD5th

nrc4ksmTc

dptpeh

R

I

gP

ip1

dd

oC

t

ctG

rR

MG

a

tt

to

lI

tn

ql1

ImE

L

rpC

ps

dv

oS

pI

pa

tb

tl

ac

NDPP

CONTROL OF OCULAR COMPLICATIONS 523

iao et al19 in studies of spontaneous diabetic Torii ratsndergoing syngeneic PT. By means of fluorescein angiog-aphy and retinal immunohistochemistry, these authorsbserved that diabetic cataracts and retinopathy were pre-ented or improved where PT had been performed at 5 butot at 10 weeks after diabetes onset. The efficacy of PT forR has also been demonstrated in a canine model with

-year follow-up of segmental duct-occluded pancreas au-ografts and in alloxan-induced diabetic rats undergoingeterotopic PT with 24-month follow-up.20,21

In humans, it has been difficult to determine a “point ofo return” for diabetic ocular complications. Therefore, theesults of most studies on the efficacy of PT for DR areonflicting. Giannarelli et al,22 studying the course of DR in8 diabetic patients undergoing simultaneous pancreas-idney transplantation, concluded that despite a relativelyhort follow-up, successful PT was associated with improve-ent and/or stabilization of DR in the majority of patients.his finding has been confirmed by other authors,23,24 butontradictory results have also been observed.3,4,25,26

In conclusion our results suggested that prevention ofiabetic ocular lesions by PT was closely dependent on earlyerformance of the procedure. Thus, glycemia normaliza-ion probably does not have a positive influence on DRrogression, if it is already too advanced. The risk-benefitvaluation of surgery and chronic immunosuppression,owever, must always be considered in this setting.

EFERENCES

1. Bretzel RG: Intensive insulin regimens: evidence for benefit.nt J Obes Relat Metab Disord, 28(suppl 2):S8, 2004

2. Cerveny JD, Leder RD, Weart CW: Issues surrounding tightlycemic control in people with type 2 diabetes mellitus. Annharmacother 32:896, 19983. Wang Q, Klein R, Moss SE, Klein BE, Hoyer C, et al: The

nfluence of combined kidney-pancreas transplantation on therogression of diabetic retinopathy. A case series. Ophthalmology0:1071, 19944. Ramsay RC, Goetz FC, Sutherland DE, et al: Progression of

iabetic retinopathy after pancreas transplantation for insulin-ependent diabetes mellitus. N Engl J Med 318:208, 19885. Lee TC, Barshes NR, Agee EE, et al: The effect of whole

rgan pancreas transplantation and PIT on diabetic complications.

Table 1. Area (mean � SD) of the Outer Basement Membrane(�m2) of Vessels From the Retinal Superficial Plexus in the

four Experimental Groups Throughout Follow-up

Groups

Follow-up (mo)

1 6 12

C 31.2 � 2.3a 29.1 � 1.9a 33.8 � 2.7aC 40.3 � 3.2b 39.7 � 3.4b 47.7 � 4.2bT1 30.0 � 2.2a 29.9 � 2.0a 34.0 � 2.4aT2 38.3 � 3.4b 38.6 � 3.6b 46.4 � 4.3b

urr Diab Rep 6:323, 2006 1

6. Lee S, Tung KS, Koopmans H, et al: Pancreaticoduodenalransplantation in the rat. Transplantation 13:421, 1972

7. Spadella CT, Mercadante MCS, Machado JLM, et al: Pan-reaticoduodenal transplantation in rats with microsurgery. Theechnique and results following 12 years of investigation. Arqastroenterol 33:158, 19968. Sima AA, Chakrabarti S, Garcia-Salinas R, et al: The BB-

at—an authentic model of human diabetic retinopathy. Curr Eyees 4:1087, 19859. Hori S, Mukai N: Ultrastructural lesions of retinal capillaryüller cells in streptozotocin-induced diabetic rats. Albrecht Vonraefes Arch Klin Exp Ophthalmol 213:1, 198010. Schellini SA, Gregório EA, Spadella CT, et al: Müller cells

nd diabetic retinopathy. Braz J Med Biol Res 28:977, 199511. Raviola G, Butler JM: Morphological evidence for the

ransfer of anionic macromolecules from the interior of the eye tohe bloodstream. Curr Eye Res 4:503, 1985

12. Schellini SA, Gregório EA, Padovani CR, et al: Ultrastruc-ural and morphometric aspects of ageing in the retinal capillariesf rats. J Submicrosc Cytol Pathol 29:275, 199713. Hori S, Nishida T, Mukai N: Ultrastructural studies on

ysosomes in retinal Müller cells of streptozotocin-diabetic rats.nvest Ophthalmol Vis Sci 19:1295, 1980

14. Stitt AW, Bhaduri T, McMullen CB, et al: Advanced glyca-ion end products induce blood-retinal barrier dysfunction inormoglycemic rats. Mol Cell Biol Res Commun 3:380, 200015. Barron KD, Dentinger MP, Krohel G, et al: Qualitative and

uantitative ultrastructural observations on retinal ganglion cellayer of rat after intraorbital optic nerve crush. J Neurocytol5:345, 198616. Das A, Frank RN, Zhang NL, et al: Increases in colagen type

V and laminin in galactose-induced retinal capillary basementembrane thickening prevention by aldose reductase inhibitor.xp Eye Res 50:269, 199017. Wang PH: Tight glucose control and diabetic complications.

ancet 342:129, 199318. Henricsson M, Berntorp K, Berntorp E, et al: Progression of

etinopathy after improved metabolic control in type 2 diabeticatients. Relation to IGF-1 and hemostatic variables. Diabetesare 22:1944, 199919. Miao G, Ito T, Uchikoshi F, et al: Stage-dependent effect of

ancreatic transplantation on diabetic ocular complications in thepontaneously diabetic Torii rat. Transplantation 77:658, 2004

20. Hawthorne WJ, Wilson TG, Williamson P, et al: Long-termuct-occluded segmental pancreatic autografts: absence of micro-ascular diabetic complications. Transplantation 64:953, 1997

21. Orloff MJ, Lee S, Charters AC 3rd, et al: Long-term studiesf pancreas transplantation in experimental diabetes mellitus. Annurg 182:198, 197522. Giannarelli R, Coppelli A, Sartini M, et al: Effects of

ancreas-kidney transplantation on diabetic retinopathy. Transplnt 18:619, 2005

23. Koznarova R, Saudek F, Sosna T, et al: Beneficial effect ofancreas and kidney transplantation on advanced diabetic retinop-thy. Cell Transplant 9:903, 2000

24. Konigsrainer A, Miller K, Steurer W, et al: Does pancreasransplantation influence the course of diabetic retinopathy? Dia-etologia 34(suppl 1):S86, 199125. Bandello F, Vigano C, Secchi A, et al: Effect of pancreas

ransplantation on diabetic retinopathy: a 20-case report. Diabeto-ogia 34(suppl 1):S92, 1991

26. Petersen MR, Vine AK: Progression of diabetic retinopathyfter pancreas transplantation. The University of Michigan Pan-reas Transplant Evaluation Committee. Ophthalmology 97:496,

990