Embed Size (px)
Citation preview
Steroid Nephropathy in the Rabbit
D. Laurence Wilson, M.D., David A. Rosen, M.D., and Sergio A. Bencosme, M.D.,Kingston, Ontario
Although diabetic glomerulosclerosis occurs frequentlyin diabetic patients, its whole picture has never beenproduced in any form of experimental diabetes. Someinvestigators'"' have been interested in the cortisone-induced glomerular lesions, since these, besides re-sembling those of diabetic glomeruloselerosis, oeeurin animals whose glueose metabolism is disturbed. Dur-ing the past few years several reports''' have stressedthat, in addition to the nodular ehanges of diabetienephropathy, there are other ehanges eoUectively termedthe exudative lesion ( figure i ). While there is nogeneral agreement on the significance of the latterlesion, it has been suggested that it represents an earlystage in the development of nodular diabetic glomeru-losclerosis," although others believe it to be a latemanifestation." In any case it is worthy of note thatthe glomeruli of rabbits treated with cortisone mayshow morphological alterations indistinguishable fromthe exudative lesions of human diabetics. The studies tobe reported in this communication were designed toexplore the histological effects of ACTH and of a num-ber of steroids including cortisone, desoxycorticosterone,prednisone, prednisolone and 9-alpha-fluorohydrocorti-sone, and to investigate conditions under which the ef-fects of these agents might be inhibited or potentiated.Speeifieally, experiments have been designed to exam-ine the influenee on the lesions of dietary faetors,experimental hypertension and alloxan diabetes.
METHODS
Unless otherwise specified, the experiments utilizedmale white New Zealand rabbits weighing 2,400 to2,800 gm., given Sherwood feed unsupplemented byVitamin B,̂ . or aureomycin, and water ad libitum. Bodyweight, blood pressure and blood and urine glucosewere generally determined twice weekly, although notall procedures were carried out in all experiments.
Nonfasting blood sugar was determined by a mod-ification of the Folin-Wu method.'" Urine sugar wasdetermined by Benediets quantitative method ontwenty-four hour urine speeimens preserved with toluol."
From the Departments of Medicine, Ophthalmology andPathology, Queen's University, Kingston, Ontario, Canada.
Blood pressure was measured by auseultation and pal-pation over the abdominal aorta distal to a sphygmo-manometer euff around the abdomen. This relativelyerude method of measurement in our hands yieldedvalues reprodueible within ±15 mm. Hg.
Animals were killed by Nembutal overdosage afteran experimental period of twenty-one days exeeptwhere otherwise speeified.
Tissues were prepared aeeording to methods previ-ously deseribed'" and paraffin seetions eut routinely at2 fj.. Kidneys were stained with hematoxylin, phloxineand saffron (HPS).'" In seleeted eases special stainingprocedures were used such as modified Masson's tri-chrome,'' phosphotungstic acid hematoxylin, periodicacid Schiff's reagent (PAS) counterstained with ironhematoxyiin and orange G, and Bodian's activatedprotargol." Frozen sections were stained with Schar-lach R and Oil Red O for fat.
RESULTS
The experiments were carried out in three categories,A. Control experiments. This group, summarized in
table I, was designed to see whether morphology mightbe altered by certain dietary modifications or by manipu-lations by which we might later try to influence the
) - . . » .
FIG. I. Sfvere nodular lesions from a human case of diabeticglomerulosclerosis. This case also shows some "exudativelesions" characterized by hyaline and fatly deposit(arrow). HPS X 240.
4 0 2 DIABETES, VOL. I I , N O .
D. LAURENCE WILSON, M.D.. DAVID A. ROSEN, M.D., AND SERGIO A. BENCOSME, M.D.
TABLE 1
Control experiments
Average AverageAverage blood maximumweight sugar B.P. 'change mg. per change Steroid
in grams 100 ml. mm. Hg lesions
None
Title ofexperiment
Kidney morphologyunder laboratoryconditions
Effect of diet lacking.supplementary vita-min B-12 and aureo-mycin
Effect of high sodi-um chloride intakeon kidney
Effect of high .-.odi-um intake and uni-lateral nephrectomyon kidney
•
Effect of high sodi-um intake, unilateralnephrectomy andcellophane wrappingof remaining kidneyon glomerular struc-ture
Effect of cortisonevehicle (Upjohn)on kidney
Study of kidneymorphology iniilloxan diabetes
Numberand sex
24 M12F
9 M
30 M20 F
5 M
5 M
12M
! 0 M
Durationof
experimentDays
1-30
21
21
29
M 4
21
42
Experimentalprocedures
Purina rabbitchow
Water ad lib.
UnsupplementedSherwood feed
Water ad lib.
UnsupplementedSherwood feed
400 cc.1% NaCl daily
UnsupplementedSherwood feed
1% NaClq.2.d.400 cc./dayUnilateralnephrectomyafter 8 days
UnsupplementedSherwood feed
1% NaClq.2.d.400 cc./dayLeft nephrec-tomy, rightrenal wrapping
UnsupplementedSherwood feed
Water ad lib.3 cc. cortisonevehicle dailyLM.
UnsupplementedSherwood feed
1% NaCl400 cc./dayAlloxan 200mg./kg. I.V.
+419
+5
+ 361
113
153 +24
None
None
None
+5536
None
+526
—480 341
None
None
Conclusions
Unsupplemented dietnot injurious tokidney
Supplementary sodi-um chloride does notaffect kidney mor-phology
No lesions producedby experimentalhypertension
No specific pathol-ogy when hyperten-sion produced in thismanner
Vehicle used in cor-tisone suspension in-nocuous to kidney
Alloxan diabetesproduced tubularchanges and pyelo-nephritis but noglomerular changes
steroid lesions. No differences in renal morphologywere noted between rabbits fed Purina Chow and thosefed an unsupplemented Sherwood feed. High sodiumintake, even when combined with unilateral nephrectomyand its consequent hypertension, failed to induce anychange suggestive of steroid lesions; high sodium plusnephrectomy did lead to some glomerular hypertrophyand hyperplasia with loss of the distinction betweensuperficial and deep glomeruli. When high sodium in-take and unilateral nephrectomy were combined withcellophane wrapping of the remaining kidney, therewere marked cortical fibrosis and inhibition of theusual compensatory hypertrophy, but no specific glomer-ular lesions were produced. The vehicle used for the
cortisone suspension in our experiments produced norenal histological change. Finally, alloxan diabetes failedto cause specific glomerular lesions although it did leadto tubular calcification and pyelonephritis.
B. Production of steroid rtnal lesions. In this groupof experiments, summarized in table 2, we examinedthe effects on renal histology of ACTH and of a num-ber of steroids related to adrenal cortical hormones.The steroids used included a group with predom-inantly "glucocorticoid" effects (cortisone, prednisoneand prednisolone), and some with predominantly"mineralocorticoid" effects ( 9-alpha-fluorohydrocortisoneand desoxycorticosterone). Experiments designed to de-termine the progression and regression of the cortisone
SEPTEMBER-OCTOBER, 1962 4OJ
STEROID NEPHROPATHY IN THE RABBIT
TABLE 2Effects of steroids on rabhit kidney
Title ofexperiment
Effect of cortisoneon kidney
Effect of cortisoneon kidney
Effect of desoxycor-ticosterone acetateon kidney
Effect of 9-alpha-fluorohydrocor-tisonc on kidney
Effect of prednisoneand prednisolone onkidney
Effect of ACTHon glomeruli ofalloxan diabeticrabbits
Effect of ACTHand protamine sul-fate on kidneys ofalloxan diabeticrabbits
Numberand sex
l l M
8 M
6 M
6 M
8 M
4 M
4 M
Durationof
experimentDays
21
21
21
10
26-39
42
42
Experimentalprocedures
UnsupplementedSherwood feed
Water ad lib.Cortisone acetate7.5 mg. l.M.Purina chowWater ad libCortisone acetate7.5 mg. l.M. dailyPurina chow
330-780 cc.1 % NaCl dailyD.C.A. in oil1 mg. l.M. dailyleft nephrectomyat outsetPurina chow400 cc.1 % NaCI daily
9-alpha-fluoro-hydrocortisone5 mg. dailyPurina chowWater ad lib.Prednisolone5 mg. I.P. for21 days thenprednisone15 mg. I.P. for5-18 daysUnsupplemented
UnsupplementedSherwood feed
1 % sodiumchloride400 cc./dayAlloxan diabetesinduced by 200mg./kg. alloxan21 days beforestart of ACTH
Averageweightchange
in grams
+4
+20
+389
—230
+ 160
+4
+ 185
Average Averageblood maximumsugar B.P.
mg. per change100 ml. mm. Hg
126
173
+4835
126
164Dark discol-oration ofurine duringprednisoneadministra-tion
148
265
Steroidlesions
averageseverity*
+ -r +
+ + +
None
fpredni-solone)
( predni-sone)
None
None
Conclusions
Damage to glomeruliof 100% of animalswith this dose ofcortisone
D.C.A. in dosagesufficient to producehypertension, pro-duces none of mor-phological effects ofcortisone
Kidney lesions simi-lar to those of cor-tisone
' 1 w .̂1 l i t hjV^ 1 1 V m.J 1 t i^ 1 U ^ ^ ^
severe renal damage
No glomerularlesions producedby ACTH in alloxandiabetic rabbits
No glomerularlesions producedby ACTH andprotamine inalloxan diabeticrabbits
•Scale of severity of steroid lesions:+ = nodular glomerular lesions rare.
+ + = nodular lesions common—but less than 1/3 glomeruli affected.+ + + =: nodular lesions involve over 1/3 glomeruli.
lesions were also included in this group. Where steroidlesions appeared at all, they had the same characteristics,whatever the steroid which induced them. These charac-teristics will be described in detail below. They wereinduced by cortisone (7.5 mg. daily for twenty-onedays), prednisone and prednisolone (5 mg. daily fortwenty-one to twenty-nine days), and 9-alpha-fluoro-
hydrocortisone (5 mg. daily for ten days) but not bydesoxycorticosterone ( i mg. daily for twenty-one days).The lesions seemed independent of the occurrence ofhypertension or of a rise or fall of body weight or ofhyperglycemia. In the latter regard, some of the mostextensive and severe lesions in our experience oc-curred in animals which had received prednisone and
404 DIABETES, VOL. I I, NO. 5
D. LAURENCE WILSON, M.D.. DAVID A. ROSEN, M.D., AND SERGIO A. BENCOSME. M.D.
TABLE 3
Experiments designed to accentuate steroid lesions
Title ofexperiment
Effect of unilateralnephrectomy andprolongation of cor-tisone administrationon "steroid lesion"of surviving kidney
Effect of supplemen-tal .sodium chlorideand unilateralnephrectomy on in-cidence of glomeru-lar lesions
Effect of experimen-tal hypertension onglomerular lesionsof cortisone
Effect of prednisonein alloxan diabeticrabbits
Effect of unilateralnephrectomy ai:dprednisone adminis-tration on glomeru-lar liistology
Durationof
Number experimentand sex Days
60 M30 E
70 M20 E
•42
42
5 M 35
10 5-21
21
Experimentalprocedures
UnsupplementedSherwood feed
Water ad lib.Cortisone acetate7.5mg./dayI.M.Left nephrectomyafter 21 days
UnsupplementedSherwood feed
1% sodium400 cc./daychlorideCortisone acetate7.5 mg./day LM.for 42 daysLeft nephreciomyafter 21 daysUnsupplemented
Sherwood feed1% NaCl q.2.d.350 cc./dayLeft nephrsctomy
at outsetUCA 5 mg./dayCortisone7.5 mg./dayPurina chowWater ¡id lib.Prednisone10 mg. I.P. dailyAlloxan200 mg./kg. I.V.Left nephrectomy
at outset ofprednisone ad-ministration
UnsupplementedSherwood feed
1% NaCI350 cc./dayPrednisone5 mg. I.P. dailyLeft nephrectomy
at outset
Average AverageAverage blood maximum
sugar B.P.mg. per change100 ml.
weightchange
in grams— 245
mm. Hg
Steroidlesionsaverage
severity*
190
182
—223
136 ±1135
-fl90
+ 110 279
136
ConclusionsSlight accentua'iionof glomerular dam-age by prolongationof cortisone
Kidney lesion notaltered by unilateralnephrectomy andsodium supplementa-tion
No effect on glomer-ular lesions
No accentuation ofprednisone-inducedlesions in diabeticrabbits
Glomerular lesionproduced similar tothose of cortisonebut of greater sever-ity
which had no significant elevation of blood sugar.C. Attempts to accentuate steroid lesions. These ex-
periments are summarized in table 3. There was noevidence from these experiments that steroid lesionswere accentuated by unilateral nephrectomy. A highsodium intake appeared to provide some slight protec-tion against steroid lesions in unilaterally nephrecto-mized animals. Alloxan diabetes failed to aggravatethe lesions produced by ptednisone, nor did it enableus to induce any lesions whatever with ACTH, evenin the presence of additional treatment with protaminesulphate, a heparin antagonist which might be expected
to potentiate the effect of steroids on the kidney.'"Morphology of steroid-induced lesions
A. Glomerular Changes. Although rabbits treatedwith cortisone, prednisone, prednisolone and 9-alpha-fluorohydrocortisone showed alterations in the renaltubules, this report is mainly concerned with theglomerular changes.
Kidneys of animals treated with these steroids overa period of ten days or longer, showed in the cortexnumerous small dark brown discolorations about 0.5 mm.in diameter, resembling petechial hemorrhages. Kid-neys of some animals, treated for twenty-one days or
SEPTEMBER-OCTOBER. 1962 405
STEROID NEPHROPATHY IN THE RABBIT
longer, had so many of the.se lesions as to resemblethe so-called "flea-bitten kidney" (figure 2). Thesefoci of discoloration represent engorged glomeruli (fig-ure 11) with or without intracapsular hemorrhage.Their number was closely correlated with the extentot general glomerular damage. There was little increasein severity of the lesions obtained when treatment wasprolonged up to forty-two days, nor were they morenumerous in the animals receiving the larger total dosesof steroids (see table 3).
The evolution of glomerular lesions was followedin a separate group of sixty-four animals sacrificed seri-ally during and after administration of cortisone (7.5mg. i.m. daily up to forty days). Within two days thediameter of most glomerular capillaries had increasedconsiderably, particularly in glomeruli situated in thedeeper portion of the cortex. Dilatation confined toindividual lobules was only occasionally detected. Ascapillary dilatation occurred, the space between adja-
cIgt ^m^ ^^^
-5
FIG. 3. Kidney from a normal rabbit. Masson's Trichrome X900.
FIG. 4. Marked dilatation of all glomspace between adjacent loopsedematous. Masson's Trichrome
appearsX 800.
pillaries.fibrillar
Theand
FIG. 2. Kidneys from cortisone-treated rabbits showing dis-coloration suggestive of "petechial hemorrhages."
cent capillary loops, the so-called mesangial area, becameprogressively more prominent with an increase in thenumber of nuclei and of accompanying fibrillar network(figure 4). Most fibrils in this network stained pink withtrichrome while some stained blue; the latter alsostained with PAS. The perinuclear portion of glomerularepithelial cells appeared larger and more basophilic thanin controls.
When treatment was continued beyond ten days,turther changes occurred such as focal deposition in thebasement membrane of a coarsely nodular, intenselyPAS-positive substance. In adclition, a complex mate-rial witb staining characteristics of protein, fat andmucopolysaccharide was found attached to luminal andepithelial surfaces of glomerular capillaries (figure •)).
The epithelial cells in the vicinity frequently containedthis material in the form of droplets of varying stainingcharacteristics within their cytoplasm (figure 5). Attimes this deposit occurred in striking quantity withinthe Bowman's space. When it was present in smallamounts in Bowman s space, it usually resulted in ad-hesions between parietal and visceral epithelial cells.
Although tbe capillaries of most glomeruli appearedempty, a few were noted to lie stufiFed with erythro-cytes ( figures 6, 7 ). Some of these engorged glomerulishowed large amounts of extracapillary fibrinoid bothin the center and in the periphery of lobules (figure6). Others showed hyaline and fatty deposit with anadmixture of red blood cells (figure 7); such lesionswere, at times, associated with similar deposits inBowman s space. These engorged glomeruli were con-sidered to be responsible for the "flea-bitten" appear-ance of kidneys from steroid-treated rabbits. Although
406 DIABETES. VOL. 11, NO. 5
D. LAURENCE WILSON, M.D.. DAVID A. ROSEN, M.D., AND SERGIO A. BENCOSME. M.D.
FIG. 5. Grossly dilated glomerular capillary showing depositsof a complex material (CM) on the luminal and oxtra-luminal sides of the wall. There are hyaline droplets(H) in the epithelial cytoplasm seen best in a tan-gential section of the epithelial cells. PAS X 800.
this lesion was found in no more than 5 per cent ofglomeruli, its conspicuous occurrence was always asso-ciated with the occurrence of other glomerular lesionsdescribed below.
The most striking alterations in steroid-treated ani-mals usually appeared after the fifteenth day of treat-ment. These consisted of nodules occupying the luminaof dilated peripheral segments of a glomerular tuft.These were composed of material with staining charac-teristics of protein, fat and mucopolysaccharide whiehwas described earlier in other portions of the glornerulus.Their appearance varied from glassy to fibrillar. Inthe fibrillar network there seemed to be some fibrin assuggested by the phosphoningstic acid-hematoxylinstain. Occasional red blood cells were noted withinthese nodules. Many were surrounded by a basement
FIG. 6. Glomerulus showing fibrinoid exúdate. There is alsosome proliferative inside capillaries. These are alsodilated by an increased number of red blood cells.PAS X 500.
PIG. 7. Glomerulus with several hyaline and fatty material (H)simulating thrombi inside capillaries. These are alsodilated by an increased number of red blood cells.PAS X 500.
membrane and an epithelial cell layer. Some nijdulesseemed to rupture into Bowman's spaee (figure 8).Not infrequently the glomeruli of treated rabbits showedextraluminal nodular lesions with a prominent fibrillarnetwork (figure 9). The fibrillar material was readilydifferentiated from basement membrane and collagenby the use of trichrome (with which it stained red)and PAS. From a comparative study of the special stainsused it was concluded that these lesions arose fromthe coalescence of altered epithelial cells from adjacentloops (figures 9, 10).
In a few instances, trichrome revealed lesions charac-terized by a large accumulation of blue-staining fibrousmaterial, interspersed with nuclei and a few red-stainingfibrils (figures 11, 12). These were not found in eontrolmaterial and therefore are unlikely to represent tan-gential sections of normal intraglomerular arterioles.Judged by their staining properties, these fibrous nodulesseemed to be composed of a mixture of collagen orreticulum, basement membrane, and cell cytoplasm.When surrounded by capillaries, the nodules somewhatresembled in shape the lesions of human diabetieglomeruloselerosis, although they differed from the hu-man lesions in their laek of lamination and laek oflipid deposits.
After withdrawal of eortisone, the size of the glomeru-lar capillaries regressed to normal in three to six days.Within ten to fifteen days after withdrawal most ofthe lesions had disappeared, although an occasionalnodular lesion was still visible after forty-four days.
B. Tubular changes. Although no detailed smdy ofthe changes occurring in tubules was made, there werea few alterations which, because of their consistencyand frequency, are probably worthy of consideration in
SEPTEMBER OCTOBER, 1962 407
STEROID NEPHROPATHY IN THE RABBI!
FIG. 8. There is a round mass composed of a complex materialoccupying a dilated glomerular capillary. This is sur-rounded by basement membrane except where thesubstance of the mass escapes info Bowman s space.Anastomoses between visceral and parietal epithelialcells are seen (arrow). Visceral epithelial cells coveringthe nodular lesion contain hyaline droplets ( H ) . PASX 800.
FIG, 9. In one area there is a conglomeration of severe focallesions apparently involving the epithelial cells ofadjacent loops. Under the microscope the various cap-illary loops containing entrapped red blood cells canbe followed. This nodular lesion is different from thatof figure 8. Masson s Trichrome X 800.
a Study of the pathogenesis ot steroid nephropathy.The large numbers and variety of casts were remarkable.These are summarized in table 4.
About one half of the casts found in steroid-treatedanimals were composed of a mixture of two or morebasic types. In decreasing frequency these basic typesappeared with the Masson's trichrome stain as: ( i )Laminated blue-stained cylinders; ( 2 ) Red-stained ho-mogeneous material with a large number of smallvacuoles witb staining characteristics of fat; (3) Fibril-lar pink material resembling fibrin with the PTA stain;(4) Large masses of agglutinated red blood cells which,instead of staining brigbt red, appeared as a very light
FIG. 10. In the upper right corner there is a nodular andfibrlllar lesion similar to that of figure 9. The silver-stained fibrlllar network appears to be made up ofanastomosing cytoplasmic projections from epithelialcells. Bodlan's acHvated protargol X 8D0.
FIG. I I . Shows an Irregularly shaped avascular area, fibrous incharacter, containing nuclei surrounded by very scanty,irregular cytoplasm. Fibers are primarily bluD, butsome fibrils are pink. Masson's Trichrome X 800.
FIG. 12. Shows a well-circumscribed nodule also similar Innature to that of figure 16, bu^ completely surroundedby a capillary. Masson's Trichrome X 800.
408 DIABETES, VOL. n , NO. 5
D. LAURENCE WILSON, M.D., DAVID A. ROSEN, M.D., AND SERGIO A. BENCOSME, M D.
FIG. 13. Hypercellular glomerui' • ' j ' • ,• "Üic cyto-plasm, pleomorphic nuclei and hyperchromatic nucle-oli. The abnormal cells are probably epithelial althoughparticipation of endotheliat cells cannot be excluded.X 290.
FIG. 14. Glomerulus completely replaced by neoplastic cells.These cells are also invading an adjacent tubule. Sev-eral mitoses are present. X 240.
yellow-green homogeneous mass eontaining a few paleidentifiable erythrocytes; ( 5 ) Stringy masses of ehromo-phobic finely granular material.
As indicated in the table, some of the casts stainedstrongly with PAS.
Hyperplasia of some of the distal tubules and dilata-tion with atrophy of some of the cortical collectingtubules were commonly found in the steroid-treatedanimals. Changes in the proximal convoluted tubuleswere particularly prominent after massive hemorrhagehad occurred in the associated glomeruli. In these in-stances the cells of the proximal convoluted tubulescontained numerous hyaline droplets.
C. Neoplastic changes. Two rabbits treated with cor-
FIG. 15. Kidney section showing several neoplastic nodules, thelargest of which contains an area of necrosis. X 30.
'I
TABLE 4
Summary of staining characteristics of tubular casts
Types of casts1. Laminated
cylinders2. Homogeneous
cylinders3. Homogeneous
cylinders
4. Fibrillarcylinders
5. Agglutinatedred cells
6. Homogen.îouscylinders
7, Stringy ma.-.sesof finely gran-ular material
TrichromeBlue
Red
RedtransparentvacuolesRed
PaleYellow-green
PaleYellow-grjen
Gray
PAS-OrangeG
Red-purple
Orange
OrangetransparentvacuolesOrange
Yellow
Yellow
Pale pink
Mostfrequent
location intubules
Distal &collectingProximal &distalProximal &distal
Proximal &distalProximal,distal &collectingProximal,distal &collectingPapillaryduct
tisone showed distinctive changes. At the end of twenty-one days of cortisone administration (7.5 mg. i.m.daily) the right kidney was removed from each ani-mal. One of these kidneys showed grossly visible tumornodules but the other was not grossly remarkable. Theanimals were then maintained on cortisone for anadditional twenty-one days before they were sacrificed.The left kidney of each animal had gross neoplastienodules.
Mieroseopieally these kidneys showed multieentrieglomerular neoplasia. It was not possible to determinewhether the tumor eells were of epithelial or endo-thelial origin. They were composed of large pale nucleicontaining prominent acidophilic nucleoli and an abun-
SEPTEMBER-OCTOBER. 1962 1 409
STEROID NEPHROPATHY IN THE RABBIT
dant basophilic cytoplasm often crowded with smallclear vacuoles. Mitoses were common. Some glomeruliwere diffusely infiltrated by tutnor cells (figure 13) but
. their general architecture was preserved. In other in-stances, tumor cells had grown to forrn solid nodules.Their cells frequently replaced neighboring proximalconvoluted tubules (figure 14). Elsewhere, neoplasticcells invaded surrounding tissues (figure 15). Largertumor nodules presented areas of necrosis ( figure 15).
DISCUSSIONNot all of the lesions we produced have been iden-
tical with those reported by earlier workers.'""' Withour methods we have detected a variety of changesinvolving all parts of the glomerulus. These includegeneralized glomerular engorgement with hemorrhagesinto Bowman's space, intracellular hyaline droplets inepithelial cells, extracapillary fibrinoid and a varietyof nodular lesions. Numerous tubular casts of varyingcomposition were common. Tubular hyperplastic changeswere not uncommon and in two animals frank neo-plasia occurred in the glomeruli of steroid-treated ani-mals. While the hyperplasia seems clearly to be aneffect of steroid, we cannot, of course, exclude the pos-sibility that the neoplasia seen in these two animalswas conditioned by other factors.
One of the supports of the suggestion of relativeadrenal overactivity in patients with diabetic nephropa-thy and retinopathy, has been the production in rabbitsof glomerular lesions with some of the morphologicalfeatures of hutnan intercapillary glomerulosclerosis." Inconfirming this observation, we have found changeswhich etnphasize the similarity between the experimen-tal and human lesions. In form and in staining reactions(PAS, fat stains) our steroid-induced lesions resembledstrongly the "exudative type" of diabetic nephropathy'"^and, to a lesser degree, the nodular lesions of thisnephropathy.
To invoke a role of the adrenal gland in tlie patho-genesis of the human lesions, one need not assume anabsolute elevation of the steroid output in the diabetic—for which, in any case, there is little evidence.Rather, we may think of a qualitative abnormality ofadrenal secretion, or even a "permissive" role of nortnalsteroid output. The work of Sommers and Haley'"lends .sotne support to the concept that the steroid anddiabetic lesions have a similar pathogenesis. They havedetected a substance in the walls of capillaries in theintercapillary spaces of glomeruli of patients treatedwith cortisone, which is identical in ultraviolet lightabsorption properties to the one present in diabetesmellitus. Tbey suggest tbe possibility that cortisone tnay
be one factor essential for the development of diabeticglomerulosclerosis. Since this substance which stainswith the PAS reagent is removed by hyaluronidase, itis felt to be a mucopolysaccharide. Serum polysaccharidemeasured as galactose mannose was found significantlyelevated in diabetics with well established, clinicallydetectable degenerative vascular disease.'" In this labora-tory the galactose and mannose of kidney extracts fromsteroid-treated animals were found to be markedlyincreased. Tbe presence of iron-porphyrin compoundsin the nodular and exudative lesions of diabetic glomeru-losclerosis" is of interest in the light of our findings ofcasts composed of lysed red blood cells and of increasedamounts of uroporphyrin"'" in cortisone-treated rabbits.The significance of hyperlipemia and local concentra-tions of lipids in both cortisone-treated animals" and inhumans with diabetes and glomerulosclerosis' or with-out diabetes' has been emphasized. Wilens and others"have pointed out that glomerular lipid accumulationsare also a feature common to human diabetic nephropa-thy and cortisone nephropathy in the rabbit. Similaralterations in serum lipids have been detected in humandiabetics with nephropathy,"" and in both humans andanimals given cortisone.
There are, however, a number of stumbling blocksin the path of one who wishes to propose a close rela-tion between the human and experimental kidney le-sions. First, the experimental lesions which we arediscussing have been reported only in rabbits. We havefailed to produce them with cortisone in dogs, cats,mice, rats and guinea pigs and, of course, they haveyet to be found in human Cushing's syndrome. Secondly,the experimental lesions reported here occurred quiteindependently of gross diabetes. Many of our animalshave not been grossly diabetic and the production ofexperimental diabetes has failed to augment the fre-quency or the severity of the lesion. Thirdly, the lesionshave been induced by 9-alpha-fluorohydrocortisone, asteroid known to suppress glucocorticoid activity. Inthe light of these difficulties, the resemblance betweenthe experimental nephropathy and the human diabeticlesions is not sufficiently close to support a prominentrole for the adrenal cortex in the pathogenesis of thehuman lesion.
SUMMARYThe histological characteristics of steroid nephropathy
were studied in rabbits treated with ACTH, cortisone,prednisone, prednisolone, desoxycorticosterone, and 9-alpha-fluorohydrocortisone. A separate group of animalswas sacrificed serially during administration of cortisonefor periods up to forty days, and thereafter following
4 1 0 D I A B E T E S , V O L . I I , N O . 5
D. LAURENCE WILSON. M.D., DAVID A. ROSEN, M.D., AND SERGIO A. BENCOSME. M.D.
its withdrawal in order to record the evolution andregression of the lesions. All the steroids used, exceptdesoxycorticosterone, induced glomerular lesions dom-inated initially by dilatation of capillary loops, andlater, after about fifteen days of treatment, by the devel-opment of nodules occupying the lumina of dilatedperipheral segments of a glomerular tuft; within fifteendays after withdrawal of steroid most of the lesions haddisappeared. The lesions were not increased in fre-quency or severity by unilateral nephrectomy with orwithout supplemental salt administration, by unilateralnephrectomy plus salt plus desoxycorticosterone, or byalloxan diabetes. The glomerular changes included atendency to hyperplasia, and in two cases, neoplasia. Incomparing steroid nephropathy with human diabeticglomerulosclerosis, it was concluded that the evidencenow available does not support an important role ofadrenal hyperfunction in the pathogenesis of the humanlesions.
SUMMARIO IN INTERLINGUANephropathia Steroidic in le Conilio
Le characteristicas histologie de nepbropathia steroidicesseva studiate in conilios tractate con ACTH, cortisona,prednisona, prednisolona, disoxycorticosteroide, e 9-alpha-fluorohydrocortisona. Un separate gruppo de ani-males esseva sacrificate serialmente durante le admin-istration de cortisona (in cursos de usque a quarantadies) e post su suspension con le objectivo de registrarle evolution e le regression del lesiones. Omne le ste-roides usate (con le exception de disoxycorticosterona)induceva lesiones glomerular dominate initialmente perle dilatation de ansas capillari e plus tarde, post circadece-cinque dies de tractamento, per le disveloppamentode nodulos occupante le lumines de dilátate segmentospériphérie de un glomerulo malpighian. Intra dece-cinque dies post le suspension del steroide le majoritatedel lesiones habeva disparité. Le lesiones non essevaaugmentate in frequentia o in severitate per nephrec-tomia unilateral con o sin le administration de salsupplementari o per nephrectomia unilateral in com-bination con sal e disoxycorticosterona o per diabetea alloxano. Le alterationes glomerular includeva untendentia hyperplastic e, in duo casos, neoplastic. Incomparar nephropathia steroidic con human glomerulo-sclerosis diabetic, il esseva concludite que le datos cur-rentemente disponsibile non supporta le these de unimportante rolo de hyperfunction adrenal in le pathoge-nese del lesiones human.
ACKNOWLEDGMENTThese studies were supported by grants from the
Canadian Life Insurance Officers Association. The Na-
tional Cancer Institute of Canada and by ProvincialPublic Health Grant No. 605-7-100.
REFERENCES'Rich, A. R., Berthrong, M., and Bennett, I. L., Jr.: The
effect of cortisone upon rhe experimental cardiovascular-renallesions produced by anaphylacric hyperactiviry. Bull. JohnsHopkins Hosp. 87:549, 1950.
"Becker, B.; Diabetic rerinopathy. Ann. Int. Med. 37:273, 1952.
"Wilens, S. L., and Stumpf, H. H.: Nodular and fattyglomerular lesions in rabbits on cortisone. Am. J. Path. 31:275, 1955.
' Bloodworth, J. M. B., and Hamwi, G. J.: Experimentaldiabetic glomerulosclerosis. Diabetes 5:37, 1956.
° Lambie, A. T., and Macfarlane, A. : A clinico-pathologicalstudy of diabetic glomerulosclerosis. Quarr. J. Med. 24:125.
"Smirh, J. F., Bolton, J. R., and Turnbull, A. L.: Renalcomplications of diabetes mellitus. J. Path. & Bact. 70:475.
' Shea, S. M., Robbins, S. L., and Mallory, K. : Diabeticnephroparhy. Arch. Path. 68:447, 1959.
* Lynch, M. J. G., and Raphael, S. S. : The nature of diabetic(Kimmelsriel-Wilson) glomerulosclerosis. Diabetes 6:488.
° LeCompte, P. M. : The "Kimmelstiel-Wilson lesion." Dia-betes 7:495, 1958.
'"Hawk, P. B., Oser, B. L., and Summerson, W. H.: Prac-tical physiological chemistry. Ed. 2, New York, Blakiston, 1954.
" Todd, J. C, and Sanford, A. H. : Clinical diagnosis bylaboratory methods. Philadelphia, W. B. Saunders Co., 1943.
'" Bencosme, S. A.: Srudies on the methods of staining rheislet cells of the pancreas. Arch. Path. 53:87, 1952.
" Bencosme, S. A.: Trichrome staining methods for routineuse. Am. J. Clin. Path. 24:1324, 1954.
" Bencosme, S. A.: Studies on the terminal autonomie ner-vous system with special reference to the pancreatic islets. Lab.Invest. 8:629, 1959'
'" Rosen, D. A., Becker, B., Maengwyn-Davies, G. D., andFriedenwald, J. S. : Influence of heparin on cortisone nephropa-thy of the rabbir. Bull. Johns Hopkins Hosp. 95:144, 1954.
"Bencosme, S. A., Wilson, D. L., and Rosen, D. A.:Glomerular lesions produced in the rabbit by prednisone andprednisolone. Arch. Path. 65:340, 1958.
"̂ Sommers, S. C, and Haley, K. H. : Similarity of glomeru-lar ultraviolet absorptions in diabetes mellitus and after corti-sone therapy. Proc. Soc. Exper. Biol. & Med. 9r:263, 1956.
""Berkman, J., Rifkin, H., and Ross, G.: Serum poly-saccharide in diabetes patients with and without degenerativevascular disease. J. Clin. Investigation 32:415, 1953.
" Bencosme, S. A., Rosen, D. A., Wilson, D. L., Runeckles,V. C, and Frei, J. : Influence of steroids on carbohydrate con-rent of rabbit kidney: A Chromatographie study. Proc. Soc.Exper. Biol. & Med. 93:587, 1956.
^ Wyllie, J. : Personal communication. ! I°̂ Engelberg, H., Gofman, J., and Jones, H. : Serum lipids and
lipoproteins in diabetic glomerulosclerosis. Metabolism r:3oo.-Adlersberg, D., Schaefer, L. E., and Drachman, S. R.:
Studies on hormonal control of serum lipid partition in man.J. Clin. EndocrinoL 11:67, 1951.
'"Wong. C, Schaefer, L. E., and Adlersberg, D.: Experi-mental srudies of the relations between adrenal cortex, plasmalipids and arteriosclerosis. Endocrinology 56:628, 1955.
SEPTEMBER-OCTOBER, 1962
Copyright of Diabetes is the property of American Diabetes Association and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.
