A CELL CULTURE INVESTIGATION O FTHEMED IATIONO F COMPENSATORY RENAL GROWTH
€>Pe n e l o pe AnnHansen ,B. S.,M. S.
AThe s i s submitte d in part i alfulfillmen t ofthe requirements for the degr eeof
Doc to rofPh i l osophy
Fa c u l ty of Med i c ine Memoria l Uni versi tyof Ne wf oun d l and
1979
ii ABSTRACT
Although the phenomenon of compensatory renal growth has been under investigation for more than a ce n t u r y , its mediation has not bee n elucidated. The participationof a circulating, kidney-specific growt h factor, either stimulatory or inhibitory,ha s been hypothesized, bu t its existence has not been proved. Theoretical considerations pr e d i c t that
iE..
vitro methods could be profitably used in these cir c ums t a n c e s , since these methods would allow di ssociationof the effe c t s of the postulatedfa c t o r fromth e complexphysiologicalchan g e s whi c h accompany compensatory renal gro wt hiE..
vivo.The work of this thesis consistsof the developmentof a cell cul t u r e system suitablefor testing serum fromprev i o u s l y unlneph- rect omi z e d or sham-operated animals forth e presence of such a growth fac t o r . A method of primaryculture of ad u l tFischerrat kidney epit h e l i um is described. The mor p h o logyof the se cu l t u r e d cellsis compat i b l e with that of proximaltu bul e epithelium
iE..
situ, these being th e cel l s stimulated todivideiE..
viv o by unin e p h re c t omy. The chieffind i ng re ported in th isthesis istha t tre atmentwith serum from rat s un i ne phrec tomized48 ho ursprev i o uslyresu l t s in consls t en t, signif i can t increa s ein upta k eoftri ti a ted thymi din eby cultures , when compared with control sera fromsh am- op e r a tedra t s . Serum fr om rat s uninephrectomized18-36hours previouslyis not consistently stimu l a t o r y. Preliminary investigationwith thisculture systemin d i- catesthat: (a) the differen t i a l effe ctofcontroland uninephre c tomy sera is dueto thepresence of a stimulato ry fa cto r in the lat terratherthan an inhibitor in theforne r ; (b) so l ub le fractions of homog e n a t e s of both kidneyand live r tissue exhibit inhibitory dose- effect s in cultures; (c) xanthopte r in, a compound tha t stimulate s kidneytubule epithelialcell div i s ion
lE.
vivo,has no stimulatory effectlE.
vitro; (d) serum obtaine d from human kidneytransplant donor s 24-72 hours after uninephre ctomystimulates increasing thymidin e uptake in cultu res.Us e of this culture systemha sconfi rmed theexi s tenceofa serum fa c t orin vo lve d in co mpen satiozy renal growth, andshoul dprove to be a valuable tool for its cha racterization.
iv ACKNOWLEDGEMENTS
I have benefited from my association with Dr. Ken nethB. Roberts and Ms . Deborah Hyam during the work for this the s i s. What I have learned fr om my supervisor, Dr. Roberts, about planning, carrying out and interp r e t i n g experimental work will stand me in goodstead always.
My enth u s i a s m for teaching and doing reasearch in physiology has grown from hi s . DeborahIs competency in laboratory work has be e nan ex amp l e for me . Her good humour and pots of tea have been crucialto the research . My supervisorycommittee - Drs. W.H. Marshall , J. D.W.
Tomlin s o n and J. A. Barrowman - have been consistently helpful and encour a g i n g . Judy Power typed this thesis with patience and effici- ency. I was able to depend on Dick Campbell for exp ert care of the rats I us e d . Cliff George was very helpful in producing thefigures for th i s thesis.
Page LIST OF TABLES
LIST OF FIGURES LIST OF ABBREVI ATI ONS
REVIEW AND DISCUSSION OF LITERATURE A. Introdu c t i o n
B. De scri p t i o n of the pheno me n o n I. Organ andcel.l u l a r hypertrophy II. Hy p e rp l a s i a
III . Biochemical changes IV. Functional changes C. Me di a t i o n
vii i
ix xii
2 22 35 48
61
1. Introduction 61
II . Moni toring of nephron loss 62
a. By functional demand 62
b. By mass deficit 64
c. Experimental data 64
d. General consi de ration s 72
III. Translationofnephronlossto a
growthrespon s e 75
a. By a neur a l mechanis m 75
b. By hemodynamicchanges 7~
By a changein workload 80
d. By a chan g e in a circulating
substance 86
MATERI ALS AND METHODS
A. Rat su rgery I. Animals II~ Anaes t he s ia
III. Ne ph r e c tomyand sham operation IV. Bleeding
V. Kidneywe i gh ts B. Cell cu l ture C. Radioisotopetechniques
us
115 115 115 115 118 118 119 122
I. In vitro II.
I!i
vivoD. Cell counting E. Histology
I. Intact kidney II. Cul tured cells III. Microphotography
122 123 125 126 126 126 127
vi
F. Preparationof tissueho moge nate frac ti o ns 128 G. Preparation ofxanthop te r i n sol u t i o n s 129 H. Data presentation and statisti c al ana ly s i s 130
EXPERIMENTAL DESIGN AND RESULTS 131
A. Compensatory renal growthin Fischer rats 131 I. Kidney weights
II. Hyperplasia III. Adre n alglandwe i gh t s
131 132 137 B. Characteri s t icsof primary ratkidney culture s 140
I. Morphology 140
II. Quantitative gro wth chara c t eristics. 144
C. ~vitro radioiso top e assay 148
I. Ef f e ctof dose 148
II. Effec t of cold thymidinech a s e and
proportionuptakeofto t aldose 148 III. Eff e ct of length of expos ure 150 D. Se a r c h for a positive contr o l : xanthopterin 153
I. Op t ima l lengthofexp os u r e II. Dose-response cu r ve
E. Ef f e ctof tissue homo genatefractions on cultur e s
F. Rat serumexperiments
153 155
157 16 0 I. Comp a r i s o n of rat andcalf sera 160 II. Absenceofnon-specific effe c t of surgery 160
III . Effectof uninephrectomy 163
IV. Time co urse of UNxserumef f e c t 163
V. Efforts to optimize UNxeffect VI. Effect of bilateral ne phrectomy
167 171
vii
G. Human serum experiments 174
DI SCUSS IO N OF RESULTS 177
A. Compensatory renal growth in Fischerra t s 177 B. Characteristics of primaryrat kidne ycultures 182
C.
.!!!.
vitro radioisotope assay 186C. Search for a positive control: xanthopterin 189 E. Effect of tissue homogenatefr a c t i o n s on cu .l tures 192 F. Rat serum experiments
G. Human serum experiments
SUMMARY OF EXPERIMENTAL IDRK AND CONCLUSIONS LIST OF REFERENCES
194 206 208 211
LIST OF TABLES
Ta b l e
Increase in contralateral kidney size in human kidney donors.
Increase in nephron sizeafter uninephrectomy in the rat.
Influence of age on contralateral kidney mitotic index after uninephrectomy in the rat.
Early biochemical changes observed in contra- lateral kidney following uninephrectomy in the rat.
Cardiovascular changes following uninephrectomy in the rat.
Change in glomerular filtration rate following uninephrectomy in the rat.
Renal handling of sodium in the rat during the first day after uninephrectomy.
Combined effects on the contralateral kidney of rats of uninephrectomy and intraperitoneal injection of kidney homogenate.
The use of in vitro rat kidney slice preparations in investigaticmsc;f apo s t u l a t e d circulating kidney growth factor.
Page
10
17
30
47
52
57
59
98
111 viii
10
11
Effect ofun i n e p h r e c t omy onad r e na l gland weigh t in Fischer rats.
Differential effect onpr i mar y kidney culturesof treatment with sham-operat edor uninephrectomized rat serum.
139
165
LIST OF FIGURES
Figure
Influence of age on kidney and body weightin Sprague"",Dawley rats.
Compensatory renal growth in rats dur ingthe perinatal period.
Camera lucida tracings of rat nephrons.
Growth rate of Fischer rats.
Scheme of procedureused for primary cul t u r e of rat kidney epithelium.
Effect of uninephrectomy on contralat e ral kidney weight in Fischer rats.
Effect of uninephrectomy on Fischer rat body weight.
Mi totic index of contralateral kidney after uninephrectomy.
Uptake of tritiated thymidine by contralat e ral kidney afte r uninephrectomy.
page
14 16 116
121
133
134
136
138 10
11
12
13
14
Ch a ng e of app earanceof primary kidney cu l t u r e s with time.
Phase photomicrographof cultured re nal epi t h e l i a l cel l s
Electron micrographof cul tur e d re na l epi.thelial cel l.
Chang ewith time ofcel l si ze di stribu t i o n in primarykidneycultur es .
Ch a ng e withtim e of cell numbers and upta k eof tritiated thymidine in primary kidneycul t ures .
141
142
143
145
147
LIST
or
FIGURES (corrttdv )rigu r e Page
15
16
17
18
19
20
21
22
23
24
25
26
Tritiated thymidine dose-effect on its uptake in primary kidney cultures.
Effect oflen g t h of exposure to tritiated thymidine on its uptake by primary kidney cultures.
Correlation of uptake of tritiated thymidine with number of cells after treatment of primary kidney cultures with xanthopterin.
Effect of various doses of xanthopterin on uptake of tritiated thymidine in primary kidney cultures.
Dose-effects of soluble fraction of kidney and liver homogenates on uptake of tritiated thymidine by primary kidney cultures.
Scheme of treatment of primary kidney cultures with rat serum.
Comparison of effects of rat and fetal calf sera on primary kidney cultures.
Lack of non-specific effect of surgery on primary kidney cultures treated with rat serum.
Differential effect on primary kidney cultures of treatment with sham-operated or uninephrectomized rat serum at various timesafter surgery. . Comp a r i s o n ofef f e c t s of sham and uninephrectomy sera on primarycu l t u r e s derived from wholekidne y or from cortexalone.
Differential effects of incubating cultur eswith various sera for three to six days.
Effect of bilaterally-nephrectomized rat serum on primary kidney cultures.
149
152
154
156
158
161
162
164
166
168
170
172
Figure Page 27 Effect of control and unilaterally-:-nephrectomized
human sera on J?rimary kidney cultures. 175 28 Differential effect of trea.tment with sham.,..
operated and uninephrectomized rat serum on number
of cells per culture at harvest. 188
29 The structure of xanthopterin. 190
30 Dose effect of rat serum on uptake of tritiated thymidine by primary kidney cultures. 200
ABBREVIATIONS
BiNx bilateral nephrectomy BSS basal salt solution e.p . M. counts per minute
cyclic AMP adenosine 3',5' -monophosphate cyclic GMP guanosine 3',5' -monophosphate D.P.M. disintegrations per minute ECFV extracellular fluid volume EDTA ethylene diamine tetra-acetate G2 second gap phase of mitotic cycle GFR glomerular filtration ra t e HnRNA h.eterogeneous nuclear RNA 3HTdR tri tia ted thymidine JG juxta-glomerular
M mi totic phase of mitotic cycle mRNA messenger RNA
REF renal blood flow REF renal erythropoietic fac tor rRNA ribosomal RNA
RPF re n a l plasma flow
DNA synthetic phase of mitotic cycle
xii
S.E.M.
uue
UUD Xn
standard error of the mean unilateral nephrectomy
unilateral ureterocaval anastomosis unilateral ureteroduodenostomy xanthopterin
A. In t ro d u c t i o n
The ph en ome n o n of compensatoryrenalgrowthha s beenunder inv e s t i gat i o n for more than a century; the liter a tur ecomprises well ov e r a tho u s a n d citations. Most of this efforthas been directed toward s des c r i b i n g the morphological, biochemical and functional eventswhi c h follow unilateral nephrectomy. Relatively few investi- gatorshav e addressed themselves to the question of the control or mediatio nof this growth. Although ev e n t s occurringas soon asfi v e minutesaf t e r unilateral nephrectomyhave been described (Lowenstein and To b a ck,1978),th e nature of the primary stimulus is unknown.
In thi s chapter I shall considerfi r s t the phenomenon of comp ensato r y renal growth, and secondly, the stimulus to growth.
I havebeencritical in both these l e c t i o n and discussion ofpapers, payingpar t i c u l a r attention to such variablesas characteristics of the ani ma l used, samplesize,assaymethod, varian ceof data and us e of approp r ia t e controls. The most recentpublicationswere not necessar i l y sel e c t e din pr e fer e ncetoth e i rpredec essors. Some in f o rma tionwa s deriv ed fr om paper s inwh i c h i t waspresented inciden t a ll y, or inthecou rseof exp erimentspertain i ng to adif- feren t su b jec t . This review was designedto provide a summary of soundly- b ased in f o rm a t i o n con cerning the-p h e nome n onof compensatory renal gro wt h , especiallyas seenin rats an d humans. I have also attemp t ed to present an unb i a s e dbut cr i t i c a l surve yof the more contr o v e r si a l aspe c t s of thesubje ct,whi ch relate to contro lof thegrowt h.
B. Descr i p t i o n of the phenomenon
Org a n and cellular hypertrophy
Wh e n one kidney is removed, the remaining kidney increases in weightduring the succeeding days; this is not due merely to an increase in contained blood volume or to accumulation of fluid, but is caus e d by: (a) an increase in the size of some of the cells (hypert r op h y ) and (b) an increase, by cell divisi.on, in the number of thecells (hyperplasia).
Th i s mass increase has been found in all mammalian species
studied, with the possible exceptions of the baboon (Dicker and Morris,19 7 2 ) and the cat (R. Janicki, in the general discussion include d in Nowinski and Goss, 1969).
Astatistically significant increase in rat kidney mass occurs by 24 hours after unilateral nephrectomy (UNx) (Halliburton and Thomso n , 1966). At the end of a week, the mass increase is in the range of from 30 to 40 per cent whencomp a re d wi.th sham-operated contro l s (KatzandEp s te i n , 1967; Kurnick and Lindsay, 1968a).'fhe time cou r s e of kidneymassincrease in the Fischerrats used in this study is des c r i be d in the Results se ction.
The proportion of dry mass to wet remains constant at about 24 per ce n t during the course of compensatory growth (Ha l l i b u r t o n and Thoms o n , 1965b; Threlfallet al., 1967; Ku r n i c k and Lindsay, 1968a).
Therefo r e , measurement of either wet or dry weight is an equally relia b l e index of growth.
compensat o r y growth is greater: Kaufman et:al. (1974) reportthat remo v a l of about 70 per cent of kidney mass from adult rats results in resto ra t i o n of about 75 per ce ntof the original massby four weeks, orabout 65 pe r centofth ekidneymass in shamanimals at four weeks . Th i s is an increaseof about 2.7 times theweightof the ti s s uewh i c h remained after surgery. UNx rats restored 90 per cent of the ori g i n a l mass and 76 per cent of sham mass, or an increase of about 1.8 times the residualma s s .
The observation thatki d n e ymas s increase aft.erUNx shows an invers e co r r e l a t i o n with age raises several questions : (a) what is the no rma l course of kidney growthin unoperated animals - i.e. , what is th e background of growth upon which compensatorygro wthis superi mpo sed ; and (b) do the relativeproportionsof the hyper- trophicand hyperpl astic compon e n t s of the compensatory growthch a n g e withage? Consideration of thes e que s t i o n s will be limited to investi g a t i o n s on rats and humans.
Ratscontinuetoin c r easein bod y and kidney weight throughou t theirlives , the rate sof ga i n dec reasingwith age. Usingdata on unopera te d Sprague - Da wl e yrats ob t ain ed from Pot.ter et a l. (1 9 6 9 ) , plotsof bodyweight,kidney weig h t, and kidney weight expresse d as a per c e n t a ge of body weightvs. age (Figure1;)show that the rate of body weight gain de creaseswithage,and kidney weight gain does notkee p pace with that of body weight. Kaufman et al. (1974) report that in maleSpragu e-Da wle yrats, kidney weight shows a l inear relat io n s hi pwith bod y wei g ht thro ughoutlife,but that the
400
(gm. ) 200
Body Weight
o
1300 1100 (mg. ) 900
700 500 300
0'6j
0.5
% 0.4
0.3
I4
Right Kidney Weight
Right Kidney Weig ht as % Body Weight
I I I I
6 7 8 9
A ge (weeks)
Fi g u r e 1. Influenceof age on kidney and body weights i.n sprague-Dawley rats. Kidney weight increases at a slower rate than body weight in young malerats. Thes e data are from Potteret al.
1969.
compared with that for smaller animals. Potter et al. (1969), report that Fischer strain rats gain body weight at a slower rate than sprague-Daw1eys, but that their kidneys bear the same proportion to body weight throughout life. Allowance must be made fo r these age correlations when kidney weight data after UNx in rats of different ages and strains is analyzed.
Five authors who report data on rats undergoing UNx conclude that compensatory renal growth is greaterin young animals than in old, but these data must be looked at in the light of the considera·- tion just discussed.
Dicker and Shirley (1973) used male albino rats of unspecified strain, divided into two groups by age, each including sham-operated and UNx rats: one group consisted of rats five days old at operation, and the other of "adult" rats - their age not stated by the authors.
The data were expressed as perce n t difference between UNx and sham ki d n e y weight expressedas perce n t body weight. There'-laSno difference between age groups untilte n days post-operation, after wh i c h the youngergr o up showedan increasingly greater compensatory growth than the oldergroup untilthelast data point at 70 days.
MacKay et al. (1932) used male albino rats of unspecified strain, of ages 5 to 720 days. They measured t:hedifference between UNx and sham kidney weights expressed as per cent of body surface area 40 days after sham operation or UNx. The authors state, with reference to their own publi shedwork, that theratio of kidney weight to body surface area remains constant at all ages (unlike the ratio of kidney
weigh t to body weight). They founda ra piddecrease in compensatory growth from age five days to sixty days, and then a slow decline until their last data point at 540 days.
The other three papers reportconcl usions,based on absolute weigh t gain of young and adult Sprague-Dawleykidneys, that com- pens a t o r y growth is greater in young animals. However,as we have seen, absolute kidney weight data from different ages of rats may bemi s l e a d i n g , since there is not a direct correlati.on of kidney weigh t with age. From data included in two'p a p e r s (Kaufman et al. , 1975and Galla et al., 1974), I have calculated kidney weight gains as apercentageof body weight. Both authors performed sham and UNx opera t i o n s , and collected datafo u r weeks later. My calculations showthat the UNx kidney is 154 per cent of sham in 50 to 80 gm . .rat s, and 143 per cent in 155 to 210 gm. rats using Galla'sdata;
150pe r cent in 55 gm. rats and 140 per cent in 175 gm. rats using Kauf ma n 'sdata. In the third pape r (Barrows, 1962) no body weight da t aaregiven, sono interpretation may be made. Since the raw da taare not given, it is not possibleto analysetheseresults sta t ist ic a l l y; i t appe ars , howeve r , thatth e y are in li.ne with tho s eof MacKay et al . , and Dicke r and Shirl ey, in showing that after al lowan ce is made for normal growthcharacteristics, there is indeed adecreasedgrowth response toUNx as age advances in the rat. The dec r e a s e in response is probablynot apparentuntil about 10 days after surgery, andthe rate of de crea s e is much greater between bir t h and about 60 day sof age than ther eafter.
The change in response to UNxwhich occurs at puberty may result
size, in both normal and compensatory growth, have be en investigated, primarily in male rats using testosterone. Se v e r a l approaches to the problem have been made : (a) admini strationof testosteroneto normal, unoperated rats; (b) castration; (c) castration with replac ement testosterone treatment.
Most authors have found thatadministration of testosterone to normal male rats results in an increasein kidney weight (Selye, 1940; Lattimer, 1942; Ludden et al., 1949). MacKay (1940), however, found no increase due to testosterone treatment, which may be because he gave less frequentinjections than the other investigators(only five injections in 29 days as op po sedto daily injectionsby others).
This increase in kidney weight has been found to be due to hypertrophy, and not hyperplasia , of the tubular epithelium
(Selye, 1939; Luddenet al.,1949; Kassenaar et al., 1962).
Castration ofma le ratshasbeen re porte d to decreasekidney weight (Korenchevsky an d Ross, 19 40; MacKay, 1940). Korenchevsky an d Ross (1940) al so repo r t thatthedecreaseis reversed by te s t o s t e r on e injection. The atrophyoccurs without any cell loss ; the existing cellsbec o me smaller (Ko c h a kian , 1948; Jelinek et al., 1964).
Testosteronethe r e f o r e app ea rsto have a generalized protein anabolic effect;i tals o cause s, secon d a r i l y , sodium, potassium and
wa t e r retention (Ganong, 1977b). The testosteronestimulus to in c r e a s e in kidney size may be ei ther thedi r e c t protein anabolic eff e c t , or the mechanisms involved with salt and water retention, orboth. Testosterone influences kidney sizeby mean s of tubular cel l hyp e r t r op hy or atrophy, without ef f ect oncell numbe ri in fact, the r e is evidence that testosterone is a mitotic inhibitor in the kid n e y epithelium: as discussed below pubertyand theassociated in c r e a s e in testosterone production inhibits tubule epithelial cell hyperplasia, and yet, the kidneys·do not decrease their growth rate during this·same time. An increase in cellular hypertrophy con- comi t a n t with the decrease in hyperplasia must be occurring.
What then, is the influence of testosteroneon compensatory renal growth? Zumoff and Pachter's (1964) results are discussed be l ow: they found an inhibition of hyperplasia after UNx during pube r t y in rats, with resumption of mitosis in adulthood. Castration pr e v e n t e d the inhibition effectof puberty,an d testosterone admin- istra t i o n reversed theeff e c t of castr a t i o n.
MacKay (1 9 4 0 )ex ami n e d kidney we ightinad u l t male rats, but us e d only two animals per expe r i menta l gro up . He concluded that ca s t r a t i o n had noeffecton, and tha t testosterone administration in c re a s e d , kidney mass afterUNx. Ho wev e r , thereare difficulties wi th this report: in a d e q u a t e samp l e size s , anda lowerbody \...eight gain in testosterone-treat ed rats than in controls.
Morerecently, ther e ha ve bee n repo r t s th a t te s tosteronehas no effect on compens a t o r y renal growth in rat s. Bas ingerand Gittes
int a c t and castrated adult maleSprague-Dawley rats. The authors fo u n d no increase in kidneywei g h t caused by testosterone insh a m- operate d or UNx rats as compar edto controls. Cas t r a t i o n alone decr e a s e d kidney we ightat sev en and thirtydays; testosterone admi n i s t r a t i o n.e l i mi n a t e d theef f e c t. Castrated rats undergoingUNx ha d smaller kidneys than controls at seven days, but not at thirty da y s ; the deficit was reversed by testosterone.
Schlondorff et al. (1977) measured kidney weightin adult male Sp r a g u e - Da wl e y rats one or two weeks after UNx. The authors found no differences in kidney weight gain among control, castrated and testosterone-treatedanimals. Testosterone was administered by pe l l e t implantation; evide nc eis given that this method is more effective than injec t i o n for increasing body weightin female rats.
Thus, the best-controlled exp e r i me n t s with adequate sample sizes indicate that testosteronehas no ef f e ct on compensatorygrowth (ki d n e y mass) afte r UNx. Per haps theopposing eff e c t s of testo- sterone - inhibitionof hy per;::las ia.a n d stimulation of hypertrophy- are balanced inth issi t u a t i on, orperh a p st:estosteronesimply has no effect on the ki dn e yafter sex u a l maturity. The question is far fr om beingresolve d.
Thehuman capability for compensatoryrenalgrowth,and its possible correlation with age, ha sbeen studied by radiographic measurement of kidne y si ze in transplant donors. Measurementof length,or pl a n ime try ofkid neyarea , wa sdone byintravenous
10 uro g r a p h y before and after UNx inth e s e heal thy adults. 'l'his tech n i q u e is an inaccurate index of growth compared with weighing the kidney; the following sources of error are recognized (Dos s etor, 1975 ): (a) a orie-eor two-dimensional measurement only approximat:es thevolume or mass of the organ; (b) the experiment:alerror in mea s u r i n g kidney length is about 0.5 cm. (about four per cent error);
(c) there are physiological changes in kidney size; (d) there is some t i me s an increase in kidney size at the beginning of I. V . gra p h y .
The results of three data collections on heal thy human kidney do n o r s are shown in Table 1. No correlation of kidney size was foun d with age, sexor interval afterUNx, except by Edgren et al.
(1976 ) ,who found an inverse correlation with age, with all patients inthis series showing some compensatory growth.
Tab l e 1. Increase in contralateral kidney size in human kidney donors.
#of Age Interval Parameter Mean Reference
Patients Range After UNx Measured Per Cent
(yrs .) Increase
22 22-63 2 wks-4 yrs . length Boner et al.
(1972)
15 30- 60 6 mos.- length Heideman and
4 yrs. Rosenbaum
(1970)
46 20-74 2 mos.- 23 Edgren et al.
5 yrs. (1976)
There have also been studies ofad u l t patients (Dossetor, 1975) andchildren (Laufer and Griscom,1971) who underwent UNx consequent tounilateral kidney disease (neoplasm, hydronephrosis or pyonephrosis) .
Of19 adult patients 60 per cent showed no ev i de n ce ofcomp e n s a t o r y growth after intervals of one month to eight years,and 40 per cen t showe d a 3 per cent meanincrease in kidney length. It is quite pr obab l e that some of these pati e ntshad undergonecompensatory
growth befo reUNx due to destruc t i on of kidney tissue in the dise as e proc e s s.
In a series of 24 children born with one mul ticystic dysplastic
kidne y and studied by Laufer and Griscom (1971), the contralateral kidn ey was of normal size at birth. After UNx was performed during thefirstmonth of life, compensatorygrowth occurred over the next 18 months, resulting in a totalren a l volume (a p p ro x imat e d by the ki dne y length) of 100 per centofnormal. This length increase, from whic h volume was calculated, isab o u t 120 per centof the expected norma l. The observation that thenormal kidneyin these children wasnot hypertrophiedat birthwill be di scussednext.
Theseadmittedlyimperfec t measureme n ts leadone to the con- clusi on tha t the capabi l i t y for compe n s a t o r y kidney growth in humans exi s t.sth rough outlife,wi t h a possi blein v e r s e correlation withage.
Certai nly, in the ne onatalper iod, thegrowth re sponse afte r UNx is la r ge enough to compensate fully fo r thelossin renalmass.
We have seen that neonatal rats and humans respond to UNx with
vigo r o u s compensatory growth. Can we extrapolate to fetal animals?
The presence of kidney tiss ue inthemammalianfetus is not es s en t i al for its surviva l: renalhomeo s t a t i c functi o n s ar ecarr i edout via theplacenta by the mother ' s kidn e y s. The fetal ki.dneysdo excrete
12
a dil u t e urine, contributing to amniotic fluid; oligohyd ramnio s is associ a t e d with renalagenesis (Laufer and Gr i s c om, 1971; Goss and Walker , 1971).
Twoexper i mentshave beenreport e d on 1:h eeffect ofUNxonfe tal ratkidneys and there is a paper reportingobservations at birth of human s with unilateral non-functioningkidney. Rollason (1969) uninephre c t o mi z e d inbred albino ra t s on day18~of gestation. All incis i o n s were sutured, although the fetuseswere not replacedin the ut e r u s . Umbilical cords remained patent. Sham operation s wer e doneon some fetuses, and others were untouched and used as controls.
Feta l bodyweightsand contralateral ki d n e y weightswere determined 24 and 48 hoursaf te r surgery. Theauthorsplotted kidney vs.body weight for each animal, resultingin a linear distributionwith all poin ts fallingcloselyaboutthe mea n. Nei t h e r kidney no~body weight is decreased due to deleteriou s effec t s of surgery in UNx orsham an imal s compared with unop e r atedcontrols. Al though the number offetuses pergroup is smal l - five to ei g h t - there isa clear indic a tion that no comp e n s ato r y growthhasoccurred due toUNx.
Gos s and Walk e r (1971 )_re po rt a si mi l a r expe r i me n t usingSp rague- Dawley rats. Th e ydidsh a m or UNx operationson t:h e fet u ses on the 19th day of gestation . The kidney and bodywei g h t s of these and of unoper a t edcon t rolanima l s were determi ne d 24 , 48 or 72 hours late r.
(The last time was the final day ofges t a t i o n - day 22.) Unlike the prev i o u s investig a tors, thesedidnotsuture thefetusesor ute ru s.
The ra te of surv i v a l waslow(from25 - 50 per cent with an inver s e
rela t io n s h i p to timeaftersurg e ry);but suf f i c i e n t operations were donetoyi e l d at least50 surviv or s ineac h group. Operations wer e also do ne on rats onth e i r day of birth. Theserats were killed one, two orthreedays later . Theauthorscalcu l a t e d (for each group) theme a n values of kidney weightexp re s s e d relative to body weight.
Theydid a statistical comparisonof groupsand plotted the relative
wei ghts vs. time. They found asi g n i f i c a n t increase in relative kid n e yweight after UNx only on day 21 of gestation, and on days one,
two and three postnatally. The y conclud ethat fetalrats can com- pen s ateafter UNx, although themagnitude of the response is less tha n in the neonatal period. However, if these data are plotted
(Fi g u re 2) in a fashionsimilar to thatemp l o y e d by Rollason(1969), the evi de n c e for fetalkidney hypertrophyafterUNx seems to disappear.
Thera t e of gain of bothbody andkidney weigh t after sham or UNx sur g ery is lessth a n inunope ra tedcontrols,and there are two points ofinte res t in this diff e re nc e : (a) therelative decreasein weight gai n mus t be anon - s pec i ficre s u l t of surge r y sincesham and UNxrats sho w simi l a r we igh t s ; (b) ki d neyweig htcor r e l a t e d with body weight inspite of rate chang es, as sho wn bythepoi n ts falling in a linear pl o t .
The·l o s s of bodywei g h t and high rateof mortality, increasing
withtime, indi cate s th a t thefet u s e s were in distressas a result ofthe surgery. Apar a l le l si t u a t io n in older rats may bethat of sta r v a t i o n : body weig ht and kidn e y weigh t decrease concomitantly both in 10- 36day oldrats (Fra se r andA~l eyn e, 1974) and in adult rats
(Ku r n i c k , 1955) . Starvat i o n ha s in hi bi t ed compen satorygrowth after
14
70
60
_ 50
§ en
° ~4 cu
$:
.§
~30 Q :E
~
O'l20
10
o
o
• 0
• o u noperated control
o sham-operated
• uninephrectomized
4 6
Body Weight (gm.)
8 10
Fig u r e 20 Compensatoryrenal growth in rats during the perinatal pe.r i.od, Sprague-Dawleyfetal ratsdo not appearto undergocompensatory renalgrowthwhenuni n e p h r e c tomi z e d threeda y s be fo r e te rm.
When uninephrectomized oneday after birth, their contralateral kidneys increase in mass mo r e than those of sham-operat.ed animals. These data arefrom Goss and Walker,1971.
UNx either partially (Wachtel an d Co le, 19 6 5) or to t a ll y (Williams, 1962),or the we i ghtof the contra lat eral kidney has even de cre ased
(Goldma n , 1971).
Neit he r experimenton fetalratshassh own goodev i de n ce of comp e ns a t o r y growth: theanimals se emed to be in good condition in one (Rollason, 1969), and therewaslowvar i a n ce in thedata, even thoughthe number of animals us ed wa s small; in the other (Goss and Wa l k er , 1971) sufficie ntanimalswereused,.but their poor condition
mi g h t have preclud edthe capabil i ty to re s pondto UNx. \fuat isneeded for co n c l u s i v e evidenc e is a larg e samp le of animals, surgical techniq u e s which result in low mortality, and an experimental design in whic h fet u s e s rece ive UNx and ar ekilled at 24 hour intervals over the per i n a talperi od up to threeor fo u r daysafter birth. Such an experime n t would bedifficult, butfea s i b le , to carry out.
Mi c r o d i s sectionpr epara t i on s giveaqual i tat i ve but clear illustr a t ion of the en l a rgeme ntof ra t kidn eytubulesthat occurs afteruninephr e ctomy . Figu r e3.is a reproduc t i o n of cameralucida tra cing s ofnephronsfrom normaland hypertrophi e d rat kidneys prepa r e d by Jean Oliver (19 44).
Quantitativ e mea s u reme n ts ha v e been reported by Hayslettet al.
(1968 ). Table2. givesthepercenta g eincre a se sin size of adult ma l e Sprague~Dawleyrat nephrons two to fou rweeksafter uninephrec- tomy . The lengths we r edetermi n e d fr om formalin-fixed,micro- diss ec tedspecimens, diamete rs were measu re d fromse c t i o n e d , rapidly- froz e n specimens, vol umes were calculated fro m length and diameter
Normal rat
1 '/
( j
Figure 3. Carrera lucida tracings of rat nephrons. These preparations, made by Jean Oliver (1944), show the enlargement of rat kLdney tubules 110 days after removal of three-quarters of the renal mass ~ (on the right),compared with tubules froman unoperated rat of the same age (on the left).
Table 2. Increa s e innephro n size aft e r unineph rec t omy in the rat*
luminal ou t s i de len g t h volume diameter diameter
%incre a s e in 17% 14% 35% 96 %
proxima l convolute d tubule
% inc re a s e in 12% 10% 17% 25%
distal convoluted tubule
* Dataar e from Haysl ett et al. (1968), co mp a r i ng meantubule dim ensionstwo to four weeks afterUNx with those of control an i ma l s .
meas ureme n t s . Thenear - d o ub l i ng of proximal tubule volume is partic u l a r l y inter e sti ng: this impliesth e ability of the nephron to ha n d l e twice the no rm a l volume of glomerular filtrate. If this is so, then themode s t mass increa s e of 30 to 40 per cent seen after un i neph re ct omyrepresent sa doubli ng of func t i o n a l ca p a c i t y .
The large si ze increa s e in pr o ximalrelativeto distal t:ubule cor r e l at e s with thehighermitotic in d i c e s observedin proximal cells.
The volume of the glomeruli also increasesafter UNx. Shea (197 8 ) , using two-dimen sionalmeasurementsof ultrathin sectionsof glome r u l i , reports thatmea n glome rular vo l ume has increa s ed 2.75 timesat 21 weeks aft e r UNx in the ad u l trat . During the same in t e rv al , sham-op erated ra tsI gl omeru l arvol ume ha s do u b led , due
18
to no nna l kidney growth, so thattheeff e ctof UNxis to cause a 38 percentgreaterincrease in volume.
In the 50 yea r s around the tur n of the cen turytherewere at leasttwenty publishe d reportson the quest i o n of whetheror not new nep h r o n s are formed during compensatoryrenalgrowth. Various ages, so me t i me s unspecified , of fou rmamma l ian sp e c i e s (rats,rabbi t', guinea pig and dog) wereusedforexper iment s . The authors arriv ed at dif fere n t conclusions, probably pa r t l y becau set.heyusedverysmall sample sizes. Thefour authorswhich have been most commonly quot ed in dis c u s s i o n s of this question(Arata k i , 1926; Shiels, 1926-27;
Jackso n and Shiels, 1927; Saphi r,19 27) all concluded that no new nephrons areformedaft erUNx. Noneof these investigations included suffic i e n t animals (therewere one to four per group), and the technique of serial sectionswhich wasus e d is not suitable becaus e the in v e s t i g a t o r s did not take intoacc o u n t changes in morphology in the gr o wi n g cortex. Jacksonand Sh i el s noted a "nephrogenic"zone at thepe r i mete r of the kid neycor t ex. Th is undi fferentiatedtissue, thought tobe the sourceof new nephrons , wa s of greates tmass in newbor n rats , and had fu l lydifferenti a ted by 20 days. Jackson and Shiels' conc l u s i o n was bas edon the data from oneseven-day-old rat using,asa control, the datafr om on e othe r ratwhoseglomerulus countwas done by anothe r person. Ar atak i (1926),using the same str a i n of rat but starting his exp erimentsat 20 days,also found no newnephrons. Sa p h i r usedeigh t li t te nna te rabbits of unspecifi ed age (but yo un g on e s , as evi de ncedby theirbody wei.qht.s ) ; lie killed themfrom three da y s to'o neye a r afterUNx . There were thre e control
rabbit s. Glomeru lus co u n ts werema d einevery tenthser i a l sect i on ; theautho r st a t e s that sometimes gl o me r u l i we r e counte d twice using this met h o d. Although the incre a s e in numbe r of glomerulicou nted is pro p o rt i o n a l to thekidney mas sincrea s e, theauthordiscounts the res ul t , due to theprobable error s in counting.
Eve n though the conclusion that no newnephrons wereformed was basedon su c h inadequateexperimentaldata, further work was not repo r t e duntil the1960' s. Hiramotoet al. (1962) treatednine rats ofbody weig h t range 300 - 400 qms ,with rabbit, anti-rat kidne y
(gl o me ru l a r membrane), antiserum. After 24 hours they performed unin e p hrec t omi e s on si x of the animals, keeping three as controls.
Seve nmonthslater, theyremovedth e remainingkidneys,sectioned them and staine d withfl u o res ce i n- l a b e led anti-rabbit antiserum, and do uble-st a i n e d withrh o d ami n e - l a be led anti -rat kidney antiserum.
Any gl o me r u l i formedafterthe UNx su r ge r y would have st a i n e d rho d ami nered, andal l pre-existing ones would have double-stained yel l o w (rhod amin e red plus fluore sce ingreen). They found no new nep h ro ns . The i rmetho d wassucce ssful insh o wi n g the developmen t ofnew ne phr on s in theoute r zo ne of the cort e x in veryyoungrats ; Hi r a mo to andhiscol leagu e s ex ami ned t:i s suefrom six-weekold rats only ; and thus did not make any obs ervation s about the timecours e ofnew nephron formationin the neonatalperiod.
Bonvaletet al. (1972) examined two groups of rats, one of ages four to seven weeksandthe other greaterthan sevenweek s at the ti meofUNx, andfound thatthe youngergroup but:not theolde r fo rm e d newnephrons. He ob t ai ne dconsisten t data from two diff eren t
20
me t h o d s : (a) a count of glomeruli; and (b) calculation based on who leki d n e y and single nephron glomerular filtration rate data.
Since the report is an abstract of a paper, there is no information availabl e about sample size ormet h o d s. Bonvalet (1978) has reviewed this su b jec t and stated that young mice, but not guinea pigs, areable to pro d uce new nephrons, but no data is given in support of these finding s .
Thereis an indication from these reports that the formation of
newnephrons,or the capability to do so, might disappear at some pointin the late neonatal period - an hypothesis which is compatible withthe known ontogeny of the mammalian kidney. Canter and Goss
(1975) tested this hypothesis. Th e y used a method wherein a sus- pensionof the glomeruli from kidney tissue was used to obtain an accu r a t e count of nephrons. They found that in unoperated Sprague- Dawleyrats the numbero~nephronsstops increasing by about 40 days of age. They showedan increasein nephron formation due to UNx betwee n birth and50 days, butnot at 70 days. 'l'hiseffect of UNx had an inv ersecorr el a tionwithage.
Kunes eta1. (1976) used 18 and 80 day-old rats for UNx, and killedthem27 dayslater. Usin gtechniques si mi l a r to those of Canter and Goss the y found sevenper cent more nephrons than controls in theyounger group, and no dif f erencefrom controls in the older rats. Administration of a highsal t dietproducedno change in the younger controls or operatedrats, but resulted in an i.ncrease in nephro n number in all older rats. This work confirmed that of Canterand Goss withres p e c t tothesi tuati.on after UNx but presents
some pr o v o c a t i v e ne wqu estion s abou t the formationof nephrons in the adu l t rat.
Itseems reasonableto make the fo l l owi n g s tat.ement;s regarding post-natal nephron formation inrats: (a) theproduction of nephrons continuesfor about six weeks afte r birth in the outer zone of the cortex; (b) UNx during this pe riod , and probably for a short time after i t, stimulates production of newnephrons.
22 II Hyper p l a s i a
There nalgrowth following UNxis ofte nrefe r re d to as compensa- tory ren a l hypertrophy .".a cor r ectte r m iforgan hypertro phy isbeing referred to , but a misleadingonewithrespe c t to cellulargrowth, sincebot h cellular hypertrophy and hy perplasiaoccur.
unti l 1949, when Rollason reported otherwise,i t was thepre- vailingopinion that renal ep i t h e l i um , like neurons,did not have the ca p a c i t y for celldivision. In 19 6 6,Joh n son an d VeraRoman published experime nt a l re sultsand a discussionof therelative contributions of hyper pl a s i a and hyp ertroph y af te r UNx in mice. They derived a calcul u s eq u a t i o n whos esolution gives the percentageincrease in numberof kidn eyepi the l i a l cells. Theda tarequiredare fraction of epi t h elia l cells label edat various times after UNx or sham operat i on, and the estimated mean duration of the synthetic phase of thecellcycle . Theyfo u n d an increas e in cellsof 4.4 percen t at three days, and 7. 0per cen t at fiv eday s. At five days, the kidneydryweightha d inc reasedby about30 per cen t ; therefore about on e- fo urt h of thema ssincre a s e was dueto hyperplasia.
Th eseauth orsused thedataofGoss and Ra nk i n (1 9 6 0 ) to calcul a t e the samepar amete rs for the rat (young adult,male, Sprague -Da wley ) , and found a 5.6percentincrea seafter three da y s, and 6.8percent after four days. Kidne yweightswere not reported, so thecontr i b u t i on of hyp erplasiacannot be ca lcu l a te d ; however, i t wou l d beof the sameorderas th atfo rmic e .
The diploid DNA content of cells is const:ant(Vendrely, 1955), sothere is a direct correlation between number of cells in the kidney and its total DNA content. Enesco and Leblond (1962) have quantitated this relationship for rat cells: number of nuclei in kidney:= total kidneyDNA (mg.) /6.2 x 10-9,
since each diploid nucleus contains .6.2x 10-9
mg. DNA. A comparison of the per cent increase in total DNA wi t h mass increase at various times after uninephrectomy wouLd
yiel d an estimate of the proportion of growth due to hyperplasia.
The assumption would have to be made that all cells have the same mass, and that no multinucleated cells are present; in fact,recent1y- divid e d cells are smaller than others, so the amount of hyperplasia
would be underestimated by this method. Multinucleated cells are rarein kidney tissue (unpublished personal observations) .
Threlfall et al. (1967), using inbred male rats aged six to eight weeks and of body weight 90 to 120, gm., found an increase of 25 per cent in total DNA content and of 35 per cent in kidney wet or dry we i g h t 14 daysafter UNx. The contributionof hyperplasia in these young adult animals is then 25 (100)/35:= 71 per cent.
Uninephrectomy in the rat results in an elevation of the mi.t:otic indexof thecontralateralkidney with a peak at 40 to 48 hours, and a sec o nd , smaller peak at three to four days (Williams, 1961; Goss, 1964; Saetren, 1970). There is no significant increase in mitotic in d e x in male Sprague-Dawley rats 24 hours after UNx (Go s s and Rankin, 1960 ) . Dividing cells are most frequently seen in the proximal conv o l u te d tubule epithelium, much les s frequently in the distal co n v o l u t e d tubule epithelium, and very rarely in the renal corpuscle,
24
bloodvessels, connective tissue, loops of Henle and collecting ducts; therefore, the hyperplasia occurs almost entirely inthe cortex (Wi l l i a ms , 1961).
DNAsynthesis , as measuredby tritiated thymidine uptake , shows a sligh t increaseone and fo urdaysaf t e r UNx, anda fi ve- f o l d increas e at two and three days,according to Mayfield et al. (l 9 67).
sinceonly two rats per day were used, these resultscan be ta k e n only as a general indication of DNA syntheticactiv i t y . Threlf all et al. (1967) re p o r t e d a similar experiment,again us i n g only two rats pergroup: they found a five-fold increaseon daytwo , smal l e r increase s on day s three and four, and no differen cefrom cont r ol s on days one, sevenand fourteen.
On e may conclude only thatth e r e is some corresponden cebetween mito t i c index and DNA synthesis after UNx'inthe ra t . Althoughthe t imeseq u e n c eof change in DNA synthesisis predict a b le from exis ting data onmitotic rate s , correlat i o nof the two pheno mena ha snot been wellinvestigated.
The normal ratki d ne y has a ver y low rate of cel l divis i on ; Messier and Leblond (19 6 0 ) classifiedthe kidney asha v ing an
"expan d i n g " cell population: Le.,one that,althoughnormally nearly quiescent, has the capacityfor proliferatio nince r ta i n ci r c ums t a n c e s (a n d thereforemightbe betternam ed "e x pand a ble").
There are no stem cells, but each tubule cell (p r e s uma b l y) ha s the capaci ty for division.
A daily mitotic activity in the range of 0.8 to 1.5 per cent in adult male rats is just sufficient for cell number to keep pace with body weight increase (Leblondand Walker, 1956). The parenchymatous cells of the kidney fall into this range. McCreight and Sulkin (1962) report a mitotic index (M. I.) of 0.10 per cent among proximal convoluted tubule cells. Goss and Rankin (1960) report 0.024 per cent in cortical cells. The latter data are based on observation of about 20,000 cells per kidney, while the former on only about 2,000, so the figure of 0.024per cent is more reliable. Since the duration of mitosis in the rat kidney tubule is about 45 minutes (saet ren , 1970), thepr~portionof kidney cells dividing in 24 hours can be approximated from the index at any point in time: there are 16 intervals of 90 minutes each per day and an average of 0.024 per cent M.l. at anyone time, so the daily index will be about 0.77 per cent, since any discrete observation will reveal half the number of cells dividing over a total range of 90 minutes.
The normal low level of cell division in the rat kidney exhibits a diurnal rhythm. This variation has not been studied in a very rigorous manner: only three intervals of two hours each (out of 24) have been compared. Williams (1961) found the mitotic index at 2 to 4 p.m. twice as great as at 6 to 8 a.m. Blumenfeld (1938) reported a higher index at 2 to 4 p.m.than at 10 - 12 p.m. This indication of a maximum in the afternoon may also help to explain the discrepancy between the mitotic indices given above: McCreight and Sulkin' s rats were killed at 3 p.m., Goss and RankinIs at 10 a.m.
26
Theuse of live animals inex p er i me n tson comp en s atory renal growth intr o d u ce sa larg enumber of variab l e s. Failure to appreciate or co n tr o l theseha s contributedto variability in results and confusion in inte r p r et a t i on. The ex i s te nc eof diurnal variation in the mitotic rateof ki d n e ycel l s isonly one examp le of su c h a variable. The daily maxi mummitotic indexis at leasttwice as large as the minimum, andthe wa v e of hyperplasiafoll owingUNx has beenvariously reported asonly about a six-fold increase (Go ss andRankin, 1960), a seven- fold increase (McCr eightand Su l k i n , 1959) or an eight-fold in c rea se
(Arg y r is , 1961).
saet.ren (1972) looked specifically at the influence of the
endo g e n o us diurnal rh ythmon thewa v e of hyperplasia fo l l o wi n g UNx.
Hefo u n d and analy z e d a largeeffec t : the highest peaks were three tim e sthe lowest, and reached the irmaxima about eight hours earlier.
Thehighest peak soccu rr edarou nd noo n and wezedue to nephrectomies per fo rmed two da y s earl ier at three to ninep.m. This finding is i ll u str a ti v e of th ela rg e influe n c eth a t biological variables can have onex pe r imen t al data.
It isimpo rt a nt toexa mi ne the in f l ue nceof age on the relative pr o p o r t i o n s ofhyperpl a sia and hy p ertrophyin both normal and comp en s- ato r y kidneygrowth of the kidney . Recognitionof these age-related ch a n g e s is necessary for critical"interpretation of experiments which us emitotic indexor DNAsynthesi s as amea sureof comp ensatorygrowth.
Fiveinvestigations of normal kidney growth in rats agree that the rel a t i v e contribution of cell division, compared with increase in cellsize, decreases at aboutsi x or seven weeks of age. Shirley
(1971) fo u nd that the concentrationof DNA in kidney t:issuedecreases from bir t h to six weeks, when i t has reached adult values. 'foback and Lowen s t e i n (1974b) report that DNA concentration in rat kidney cortexdecreases from birth to eight weeks, decreasing at a faster rate betwe e n about six and eight weeks,when i t has reached its adult va l u e . The total amount of DNA should parallel that of cell number. To interpret this information,one wouLdhave to calculate the to t a l number of cells per kidney from DNA concentration and kidneyweight; then the change in rate of cell division with age would be obvi o u s . Therefore, the only conclusion that may be drawn from ShirleyIs experiment isthat a decreasein cell division occurs at about six weeks, since therate of kidney weight gain remains constant.
Potteret al. (1969) determined kidney weight and total DNA in Sprague - Da wl eyma le s , and from the s edata calculated cell number and mean cel l size. Theyco n c l u d e d th athyperplasia contributed to kidneygrowth up to about sevenwe eks; thereaft:er (unti l. three months, whenth ei r experimentended) growt hwasdue to hypertrophy.
Winick and Noble (l965) reportedtheir very carefully-obtained and compl.e t.e data on the increases in weight and protein, DNA and RNA cont e n t of various organs, including the kidney, during growth of the Sprague - Da wl e y rat. Kid ney DNAcontent increased linearly from birth tosix weeks, when there was a one-week duration of rate decrease,
28
after whi c h DNA increased onlyve r y slightly until the experiment ended at 14 weeks. Weight per nucleus,an indication of cell size, decrease d slightly until about two weeks,and then increased steadily.
These ra t e s of change in DNA and weightmay be interpreted to mean that onl y hyperplasia is occurring during the first two weeksafter birth; then both hyperplasia and hypertrophy occur until about seven weeks, after which growth is almost pure hypertrophy.
Zumo f f and Pachter (1964) measured total cell number directly in kid ne y s of growing Sherman rats. They prepared suspensions of nucleiisolated from whole kidney homogenates. When they plot ted numberof kidney cells vs. rat body weight, they found a plateau in the rate of increase - and there for e a cessation of hyperplasia - fromabout 60 to 120 gm. in males, and 75 to 140 gm. in females.
Thesebody weights probably correspondto age ranges of three to seven weeks in males and four to eight weeks in femal es of the Sherman strain, although the authors do not give age/weight correlations.
This is theonly re p ortin whichthere is an indication that mitosis resumesafter a periodof cessat ion . Since the hiatus in cell division coinci d e s with puberty, the authors su s p e c t e dsexhormones might be invo lved . They castrateda serie sof animals of both sexes, and found thatthe inhibition of hyperplasi awas abolished; hormone replacement:
res t o r e d the plateau, and extended i t as long as the hormone was given.
There is a consensus that kidney cell division occurs until six orseven weeks after birth in the normal rat. There are indicat:ions th at following puberty the rateof cell division remains low, the
decre as e being causallyrelated to theincreasein blood sex.h o r mo n e level s .
Kn o wi n g the background of normal kidney growth, i t is possible to exami n e compensatory growth which is s uoer i.mpo se d on it. As we have seen, there is agreement among published results that kidney cell hyp e r p l a s i a after UNx decreases in rats after sexual maturity.
Two groups of researchers assayed protein,RNA and DNA in rat ki d n e y s remaining after UNx at various ages. Paulini et al.
(1970) used two to thirteen month-old males four weeks after UNx.
TotalDNAper kidney increased steadily with age in control rats;
there wa s no change in the rate of increase at puberty. These rats were of a strain not used in other work (SIV-50, obtained in Germa n y ) , which mayaccount for the differences between these results and th o s e discussedabove. The DNA content of contralateral kidney four we e k s after UNx (when compensatorygrowth would have been accompl i s hed ) , increasedat thesamera t e as controlsat all ages except in animals of ages four to sevenmonths,when the DNA content decrea s e d with age. Theserats, thre e to six months old when surgery was per f o rm e d , must have been post-pubescent.
Karp et al. (1971) measured totalkidney protein, RNA and DNA inmale Sprague-Dawley rats two weeksafter UNx. Three groups of rats were used: four days, six weeks, or three to four months of ageat surgery. Hyperplasia, as determined by rela ti ve amounts of DNA, RNA and protein, occurred to an equal extent in the two younger gro up s of rats, but to a far lesserex t e n t in the adult group. The
30
author s state that Sprague-Dawleyrats attain sexual maturity at 40 days of age. These results were from groups consisting of three, three and ten rats, respectively.
Two groups of investigatorsfound that the mitotic index of kidneyepithelium in control, unoperated rats decreased with age.
Uninephr e c t omy caused a similar "degree" of increase in mitotic indexin all ages of rat studied, but since the increase was super- imposed on a decreasing baseline, the absolute index after UNx also decrea s e d with age. McCreight and Sulkin (1959), using a small number of rats (a total of ten) of the C.F. Nelson strain, found a seve n - f o l d increase in mitotic index 72 hours after UNx in four month - o l d rats, and a five-fold increase in senile rats aged 3.2 years (see Table 3,). The kidneys removed at uninephrectomy served as the controls.
Table3. Influenceof ageon contralateral mitotic index after uninephrectomy in therat:
Referen ce #of Age Operation Mi totic Index
Animals (percentage)
McCreig h t 4 mos. pre-UNx 0.10
and Su l k i n UNx 0.69
1959 3.2 yrs. pre-UNx 0.03
UNx 0.14
Phil l i p s 4 wks. 0.056
and Leong 4 wks. UNx 0.30
196 7 4 mos. 0.0062
4 mos. UNx 0.088
Phillips and Leong (1967) usedfour week and fourmonth old spra g u e - Da wl e y males, and found a five-fold increase in mitotic
index36 hours after UNx in the yo u ng e r rats, and a nine-fold incre a se in theolder ones (seeTa b l e 3). Rats receivingno oper a t i o n were con tr ols. There were fi verats per group .
Both of these experimentsca n be criticized for the small sampl e sizes; adequate numbers of rats are especially necessary
when measur ing such small values - as low as 62mi t o se sin 10, 000 cell s .
Zumoff and Pachter (1964) determined the totalnumb e r of nu clei per ki d n ey in Sherman rats. Asdesc r i b e d above, they fo u n dan inc r e a s e with age except during the timeof puber t y (a t. thr e e to sevenweeks), during which there was a plateau. Castrationabo lished this cessation of hyperplasia. These authors al s orepo r ted the effect s of UNx on kidneynuclei counts. TheSherman ra ts use d for thisser i esofex p e r i men t s showed a normal kidney-cell-numberplateau betweenabout 80and 200 gramsbod y weight. Ratswe r eunine p h r e c t o m- izedat 160 qms, (t ento eleven weeks). Sixteenda y s late r the remain i n g kidne y s were assayed, and no increase in cell numbe r wa s found . Othergroups were castrated at the time of UNx: these ani ma l s showed a 27 per cent incr e a s e in cell munber. .'freatment wit h isosexua l hormo n e (t es t o s t e ron e prop i o nate in mal es)gave resu l ts similar to uncastrated animals.
32
Itis not possible to draw any conclusions from thesefive reportsbecause of variation among techniques and strains of rat, and inad e q u a t e sample sizes. However, there is indication that the cap a c i t y for a hyperplasticres p o n s e t:o UNx is compromised at pub e r t y due to the increase in sex hormones, an effect which probab l y continues throughout the remainder of life. Because the co r r e l a t i o n among age, body weight,time of puberty and capaci ty for hyperplasia vary with strain of rat, in f o r ma t i o n regard i n g strain is necessary for interpretation of mitotic index andDNA synthesis data.
Both hyperplasia and hypertrophy of the remaining kidney after UNxare modest compared with liverregeneration: more than 90 percentof hepatocytes undergo mitosis,and the liver returns to itspre-partial hepatectomy mass (Bucher and Malt,1971 ). There areseveralpossible explanationsfor the difference in growth resp o n se seen in these two organs.
Compensatory liver gr o wt h aft erpartial ablation is not true re g en e r atIo n- the ex c ised lobesdo not regrow,but the residual ti s sueenlarges. The rate ofcel l divisionparallels that of mass increa s e , and all the cell typesex h i b i t hyperplasia. New lobules are formed, so that normal micro- anatomical architecture is preserved.
In the rat kidney, no new nephrons are formed after about si x weeks of age. The contribution of hyperplasia to the mass inc re a s e falls at puberty, so that compensatory renal growth in adult
rats is pri ma r i ly a matter ofcel l u l a r hype rt rophy. Both hyper- plasiaand hypertrophyoccur to th egreate stextent in the proximal tubular ep i t h e l i um . Thus the onlypo s s i b i l i t y for kidneygro\"th in adu l t animalsl iesinen l argemen t of existing nephrons, andev e n this gro wt h is limitedby thenarrowrangeof surface-volume rel a t io n s h i p s permissable for thefunction of the'nephron (Hayslett)
et al.,1968). A teleologicalview is that the growth of the nep h r o n is limited in orderto pre s ervei tsfunctional efficiency. Fr o manotherviewpo int,mech ani sms fo r thelimitation of growthhave evolved si n c e the r e may be aselec t i v e advantagefor a li mi ted nep h r o n si z e . Themechanismsmay involve a low mitotic capacity co uple d with amaximumpo s s iblecell size,asdiscus s edby Mitchi son
(1971).
Anotherpossibil ity is th at thetough, fibrous tissue capsule covering the kidn e y l imi ts itsgrowth . Anyone who performs uninephr e c tomies in ratshas ob s e r v ed that the kidneycapsul eis not ableto be stret ched. It is comp o s e d of dense, collagenous co n n e c tiv e.ti s s ue , thefibre s of wh ich are inelasticand randomly arranged. Inthe mongre l do g , theel as tic modul us (f o r ce/u n i t area re q u i r e d todouble the lengthof thesp e c i me n ) of the kidney caps u l e was found toexc e ed that of the aorta (He b e r t et al.,1976). Thes e authors proposethat thecapsuleisa major determinant of who l e - k i d ne y volume/pressurerel ation ships.
In general , the amo u n t of connect i ve tissuecollagen increase s with age. Although thereis no info rmationavailable specifically
34
abou t theinfluence of ageon the ki d ne ycaps u lel i tmay be hypo- thes i zedthat one factor respo n s ibl e for thedecr eas ein compensatory rena l gr owth(mas sin creas e) withageis an increase indensityand dec r ea sein elasti ci tyof the ca p su l e .
Choie and Richter (1973) have reportedthat renal decapsulation stimu la tes a six-fold increasein tr i t ia t e dthymidineincorporation int o pr o x i ma l and distaltubule cel l s of adultfema l e Sprague- Dawle yrats. The maxi mum rateofDNAsyn th esi s occurr edat about 42hou r s , and label e dcellswe r ese enmost frequ entlynearthe outer cortex . Therewas noef f e c t on thecontralateralkidney when compa r e d with sham-operated controls. No trauma due t.othe decap- sulati on procedur e was detecte d.
Uninephrectomywith simulta n eou s decapsulation of the contra- lat e r a l kidney is anexpe r i men t th a t obviously should bedone.
III Bi o c h e mi c a l changes
Oneof the components of the phenomenon of compensatorygrowth following UNx is the biochemistryof the synthesisof new RNA, DNA and prote i n by thecells of the co n t r a l a te r a l kidney. Thesechanges, the necess a r y basis force l l u l a r hypertrophyand hyperplasia,have been desc r i be d in ever-increasingdetail by biochemists and molecular biologists . It is frustrating that in spite of our knowledgeof these events, which occur beginning justa few minutesaf t e r removal of one kidney, thestimulus which sets themin motionis not yet under- stood.
It ha sbe e nes t a b l i s h e d that adenosine 3',5' -mo n o p h o s p h a t e (cyclic AMP ) , functions in cellsas a second messenger mediating a varietyof non-steroidhormonal effects (Su t h e r l a n d et al. , 196 8).
With resp e c t to thekidney, this compound has be enimplicated in the actionof anti-diur etichormone (vasopressin),parathyroid hormone, calci ton i n , glucagon and catecholamine s (Dousaand Barnes,1977).
Intracellu lar le v e l sof cyclicAMP are ra i s e d by thesehormones. Intracel lu l a r phosphodie ste r a se, whichde gra des cyclic AMP,is inhibited by methyl xant hi ne s. Ad ministrat i on of thes e compounds, caffeineor theop hyl l i n e for exampl e, re s u l t s in augmentation of hormo n a l effects mediated via cyclic AMP.
Cyclic AMP and cyclic GMP (guanosine 3' ,5' -monophosphate) also playa role in regulationof cellulargrowth. In general,an increas e in cyclic AMP content of cel lsisas soc iatedwith
36
inhibi tion of growth, and an increase in cyclic GMP content with cell growth and proliferation. The two compounds show a reciprocal relationship. A good review of the role of cyclic nucleotides in growth stimulation of cultured cells by various serum factors has been written by Schonhofer and Peters (1977).
Three groups have reported the effects of UNx on the cyclic nucleotide content of contralateral kidney tissue. Schlondorff and Weber (1976 and 1978) used adult male Sprague-Dawley rats.
They found a decrease in cyclic GMP to 20 per cent of normal levels at 15 minutes, followed by an increase to 200 to 300 per cent of normal from 1 to 72 hours. Cyclic AMP decreased to about 75 per cent of normal during 1 to 8 hours after UNx, and returned to normal by 24 hours. The authors ascribe the increase in cyclic GMP to an increase in its synthesis rate.
Dicker (1977 and 1978) reported similar results using adult male Wistar rats, except he found a different temporal pattern to the changes in cyclic nucleotides: cyclic GMP increased by ten minutes, fell to less than normal levels by 20 minutes, and returned to baseline by 90 minutes. Cyclic AMP had decreased by ten minutes, and did not rise again to baseline values until three hours.
Solomon et al. (1978), also using adult male Wistar rats, found no significant change in either cyclic nucleotide between two hours and seven days after UNx.
These dissimilar results, using apparentlysimilar experimental condi t i o n s , probably indicate that a variabie is not being recognized and co n t r o l l e d .
Another phenomenon, possibly related to cyclic nucleotide induc t i o n of growth, has been reported by Lowenstein and Toback
(1978): rate of synthesis of acid-insoluble phospholipids in kidney cortic a l cells is increased by five minutes after UNx in rats. This may repre s e n t increased synthesis of membranes, associated with amembrane-mediated second-messenger stimulation mechanism.
Altern a t i v e l y , i t could, as the authors suggest, be a prelude to prolife r a t i o n of membranes of mitochondria anden d op l a s mi c reticulum.
A group of aliphatic amines - putrescine, spermidine and spermine (collect i v e l y named polyamines) - are normalcell constituents of kidneyand various other tissues. Byvirt~eof their basic nature, polyami n e s interact with the acidic phosphategroups of nucleic acids.
Theseco mp o u n d s exert ast a b i l i z i n gan d stimulatoryeffect on nucleic acid sy n t h e s i s : (a) they effect changes in the secondary structure of RNA wh i c h facilitate it s removal fromDNA after transcription;
(b) theyst i mu l a t e theDNA-replicatingenzyme, DNA-primed DNA polyme r a s e .
Putrescine is derived by decarboxylation from ornithine, and sperm i d i n e and spermine are metabolic products of putrescine. The acti v i t y of the enzyme ornithine decarboxylase and the amount of po l y a mi ne s are markedlyincreased in tissu eswhich are undergoing comp e n s a t o r y growth. These changes are amongthe earliest detectable
38
af t e r grow t h stimulation, and probablyform a link in the chainof ev e n t s wh e r e i n the growth stimulantelicits the second messengers (cy c l i cAMP and GMP), whicheffectan increas e inor n i t h in e de c a r boxylaseactivity,which stimulates po l ya mi n e synthesis and th e r e b ynucleic acidsynthesis.
The r e have be e n tworeportsof investigationsof ornithi n e de c a r bo xylaseact i vityandpolyaminesy n t hesisdur i ng comp en s a t o r y ki d n e ygrowth; the resultsdo not supportthe hypothes is tha t these eve n t s are among the earl ies t dete c tabl e afteruninephr e ctomy, alt h o ugh the experimentaldesign may be at fault.
Bra nd t et al. (1972),using young adult Holtz ma nrat s ,measured orn i thine decarboxylase activi ty at six-hourintervals aft erUNx.
Th e y fou nd a negligibleincrease at six hours,a maximu m,fi ve-fold in c r e a s eat 24 hours, a drop to normal at 30 hours,and a sec ond th r e e - foldmaximum at 36hours. Sham-operated anima ls showed"s ma l l"
in c r e a s esin activity. Spermidineconcentration inkidne y t is s u e sh o we d asharp ma x i mum at 36 hours, twiceas high as normalvalue s.
Me l v i n and Thomson (19.72) found a six-foldin c reasein enz yme ac tivity in ratki d n e y fourho urs af t e r UNx, retu r n i n g to normal by twe l v e hours. Sh a moperation produ c ed a simi larly - t i med but small er re s po nse, andth e authors ar eof theopinion thatsomeof the enzyme act i vityis duetothe non-speci ficstressof surgery .
It is possib le th atthefirs t gr o up of inves t ig a t o rs missed asharp risein enzymeactivitybefore six hours. Onewould like to see en z ymeactivi ty assays ma deat freq uen t interval sover the fi rst fewhours afterUNx, sinc e i ti.sduring this time thatan in c r e a s esho u l d be seen if thepo l y a minesare playing a role in th e growth st i mu l a t i o n .
It isan interestingasidetha t Brandt et al. (1972) found a consistent l y higher , by 30 per cent , ornithine decarboxylase
activity in leftkidne yscompared toright in sham or unoperated rats. Theright kidne y is about 2.5percent heavier than the left inSprague-Dawley rats (Kaufman et al., 1974). It is difficult to specu l a te on there a son"f o r eit her differ enc e. The possibility mustbe ruled out th at theinvestigators introduc eda bias by alwaysremovi n gthesa me -s i de d kidneyfirst, thus changing the contrala te ra l ren alblood flow. If t.heamount of blood contained intheremai ningkidneywere in c re a sed immediatelyafter UNx, then thema ss of tha t kidney wou l dbe inc r ease d .
The r e hasbeen no inv e stiga ti onof t:he ef fec t of blood flow
onorn i t hi ne decarbo x ylaseac t ivit y. Si t u a t i o ns of rapid growth wh i c hareac compan.ied by inc r e a s eint:his enz yme's act.ivity also incl u d e an incre a seinbl o o d flow, e.g. , in liver remnant after parti al hepa te c tomy . It is possible that therelationship may be causal. Also cyclicAMP mediat e s many cardiovascular control syst em s (Sut he rlandet al., 196 8). The combination of shaky hypothesisand
