- Email: [email protected]
Effect of diabetes andinsulin on rat renal glomerular protocollagen hydroxylase activities
88
Biochimica et Biophysica Acta, 496 (1977) 88--94 ~) Elsevier/North-Holland Biomedical Press
BBA 28137 E F F E C T OF DIABETES AND INSULIN ON RAT R E N A L G L O M E R U L A R P R O T O C O L L A G E N H Y D R O X Y L A S E ACTIVITIES
MARGO P. COHEN and AHMED KHALIFA Departments of Medicine and Physiology, Wayne State University School of Medicine, Detroit, Mich. 48201 (U.S.A.)
(Received July 2nd, 1976)
Summary The effect of diabetes and insulin on the activities of both prolyl hydroxylase (trivial name; proline,2-oxoglutarate dioxygenase, EC 1.14.11.2) and lysyl hydroxylase (trivial name; lysine,2-oxoglutarate dioxygenase, EC 1.14.11.4) in isolated rat renal glomeruli was determined. Three groups of experimental animals were used: age-matched controls, streptozotocin-diabetic, and insulintreated streptozotocin-diabetic. Using 14C-labeled lysine or proline hydroxylase substrate prepared from chick e m b r y o tibiae, glomerular 17 000 × g supernatant enzyme was incubated in a complete hydroxylating syste~n for 60 and 120 min. Lysyl hydroxylase activity was significantly increased in diabetic preparations, but prolyl hydroxylase activity did not differ from control. Administration of insulin to streptozotocin-injected animals completely restored glomerular lysyl hydroxylase to normal levels. The results suggest that the specific elevation of lysyl hydroxylase relates to the biochemical changes contributory to diabetic nephropathy, and that insulin may reverse this process.
Introduction Electron microscopy has established that the characteristic lesion of diabetic nephropathy begins with an increase in the width of the glomerular basement membrane and the basement membrane-like material of the mesangial regions [ 1 ]. While the extent of the glomerular lesion shows a positive correlation with the duration of diabetes [2], the nature of the changes resulting in the deposition of excessive amounts of glomerular basement membrane in diabetes remains undefined. Furthermore, the role of hyperglycemia and/or insulin deficiency in the pathogenesis of this lesion is not well established, and the relation of diabetic control to the development of microangiopathic complications is in dispute.
89 Analysis of glomerular basement membrane from several sources has identified a collagen-like glycoprotein that is rich in carbohydrate and contains greater amounts of hydroxylysine and hydroxyproline than other more common types of collagen [3,4]. It is also characterized by a considerable amount of its hydroxyproline as the 3-hydroxyl isomer, and by a much higher percentage of glycosylated hydroxylysyl residues than interstitial collagen [5]. Attention focussed on the importance of the latter in diabetic nephropathy after Beisswenger and Spiro reported increased hydroxylysine residues and hydroxylysine-linked disaccharide units in diabetic glomerular basement membrane [6, 7]. Other investigators, however, have been unable to confirm these findings. Kefalides found no significant differences in the amino acid and carbohydrate compositions of human glomerular basement membrane from normal and diabetic individuals [8]. Westberg and Michael reported that basement membrane isolated from human diabetic glomeruli contained less cystine and sialic acid than that obtained from non-diabetics, while lysine, hydroxylysine and glucosyl-galactosyl-hydroxylysine contents were not significantly different [9]. Chemical analyses have thus not provided a totally rewarding insight into the problem of diabetic nephropathy, and other approaches seem warranted. Since hydroxyproiine and hydroxylysine are important components of basement membrane collagen, the effect of diabetes on their synthesis in glomerular tissue is of considerable interest. Hydroxyproline and hydroxylysine are both synthesized by separate postribosomal enzymatic hydroxylations that occur in a polypeptide precursor known as protocollagen [10,11]. We recently reported that lysyl hydroxylase activity was elevated in preparations obtained from isolated glomeruli of streptozotocin-diabetic rats, and suggested that chronic hyperglycemia was implicated in these changes, which could reflect increased collagen biosynthesis. The effect of exogenous insulin therapy on glomerular lysyl hydroxylase is now reported, as is the specificity of this change. Prolyl hydroxylase has been correlated with collagen biosynthesis in several tissues, which prompted study of the effect of diabetes and insulin on this parameter in the present report. Materials and Methods
Experimental animals. Three groups of experimental animals (SpragueDawley White Rats} were used: age-matched controls, streptozotocin-di~betic, and insulin-treated streptozotocin-diabetic. Diabetic animals were killed 4½ weeks after intraperitoneal injection of streptozotocin, 85 mg]kg. Lente insulin (2 u/kg per day subcutaneously) was begun the day following streptozotocin injection and was continued until time of sacrifice. Preparation of tissue. Kidneys were obtained immediately after sacrifice by anesthetic overdose, and the renal cortex was separated by gross dissection. Glomeruli were isolated essentially as described by Spiro [3], sequentially using stainless-steel meshes of 100, 60 and 230 gauge (respective pore sizes of 140, 250 and 63/am). Details of the sieving process, and collection and analysis of the glomeruli have been described [13,14]. Glomeruli so isolated are capable of in vitro basement membrane synthesis [14], and basement membrane collagen is the major collagen synthesized by these preparations [15]. The 17 000 × g
90 supernatant, obtained as previously described [12], was used as enzyme source. Kidneys were smaller and glomerular yields were less in the diabetic group than in the normal or insulin-treated groups (0.5--0.8 - 104 versus 1--2 • 104 glomeruli per g of tissue, respectively); homogenization volumes were adjusted so as to give similar supernatant protein concentrations. Protocollagen substrates. Tibiae were obtained from 11 day old chick embryos and incubated for 5 h in Krebs buffer made 10 -3 M in ~,~'-dipyridyl and containing 100 pCi of either [U-14C]lysine or [U-~4C]proline in a total volume of 10 ml. The [~4C]lysine and [14C]proline-rich protocollagen substrates were extensively extracted in 0.5 M acetic acid [16], yielding preparations with the following specific activities: [~C]lysyl-hydroxylase substrate about 1 4 0 0 0 0 cpm/mg, and [~4C]prolyl-hydroxylase substrate about 1 6 5 0 0 0 cpm/mg. Protocollagen hydroxylase assays. Incubations for assay of lysyl and prolylprotocollagen hydroxylases were performed for 60--120 min at 37°C in an atmosphere of O2/CO2 (95 : 5, v/v). The standard incubation mixture contained in a t o t a ! volume of 2 ml: 0.5 mM a-ketoglutarate; 62.5 mM Tris. HC1 buffer (pH 8.0); 2 mM ascorbate; 0.125 mM FeSO4; 0.1 mM dithiothreitol; 1.5 mg/ml bovine serum albumin; 150 pg [~4C]lysyl hydroxylase substrate or 165 pg [14C]prolyl hydroxylase substrate; and 120--250 pg/ml glomerular enzyme. Immediately following incubation, samples were made 6 M in HC1 and hydrolyzed for 18 h at 115°C. The hydrolysates were filtered, evaporated in vacuo, and reconstituted in distilled water. [ ~ C ] H y d r o x y l y s i n e was assayed by periodate oxidation [17] as well by thin layer chromatographic separation [18]; agreement between these two methods was excellent. [~4C]Hydroxyproline was determined by the m e t h o d described by Rojkind and Gonzalez [19] ; reliability of this method had been confirmed by column chromatographic separation of labeled proline and hydroxyproline in preliminary experiments. Results
Animals. Pertinent data concerning the experimental animals is presented in Table I. The streptozotocin-diabetic rats were markedly hyperglycemic and exhibited stunted growth 4½ weeks after injection. The degree of hyperglycemia was significantly diminished, although not to euglycemic levels, by daily insulin administration; growth patterns reverted toward normal in the insulin-treated animals. Lysyl hydroxylase activity. Optimum conditions for protocollagen [14C]lysyl hydroxylation by rat glomerular supernatant fractions have been previously reported [12], and were re-confirmed in the present series of experiments. Essentially we have found that omission of cofactors (a-ketoglutarate, ascotbate, and ferrous sulfate) or of enzyme resulted in markedly reduced levels of hydroxylation. Lysine hydroxylation is proportional to time for up to 120 min o f incubation studied, and to 17 000 × g supernatant enzyme from 50--250 pg/ ml. Protocollagen lysyl hydroxylase activity was significantly increased in preparations from glomeruli of streptozotocin-diabetic rats compared to activity found in glomeruli from age-matched controls. This increase was apparent
91 TABLE BODY
I WEIGHT
Number
AND
BLOOD
GLUCOSE
VALUES
OF EXPERIMENTAL
ANIMAL
GROUPS
of animals given in parenthesis.
Group
B o d y weight (g)
Age-matched controls 3 2 4 -+ 4 (10) Diabetic (4 1/2 weeks post-streptozotocin) 1 4 5 + 12 * (9) Insulin-treated diabetic (4 1/2 weeks post-streptozotocin; Lente insulin (2 u/kg per day) begun 1 day after streptozotocin) 2 7 5 -+ 15 ** (10)
Blood glucose (rag%) 120 +- 10 428 -+ 30 *
2 3 3 +- 25 **
• P < 0 . 0 0 1 , c o m p a r i n g n o r m a l to d i a b e t i c values. • * P < O . 0 0 1 , c o m p a r i n g d i a b e t i c to i n s u l i n - t r e a t e d d i a b e t i c values.
when activity was expressed either as cpm [14C]hydroxylysine formed per unit time or cpm [~4C]hydroxylysine synthesized per mg of supernatant protein (Fig. 1). The administration of insulin on a daily basis to streptozotocin-injected rats resulted in a complete restoration of glomerular lysyl hydroxylase activity to normal levels. This correction, which, was also significant when expressed either as cpm/time or as cpm/mg protein (Fig. 1), occurred despite the absence of complete normalization of the blood glucose to euglycemic values. Prolyl-hydroxylase activity. Prolyl-protocollagen hydroxylase activity was readily detectable in homogenates of rat renal glomeruli. Conditions were com.E
~o20~
.~_ °
-I-__
i
~m16 -~IE
g •
•~~ 4-8-
~0 ~)< 16-
~
~
N SD ISD 60 minutes
~~E 4-8-
N SD ~SD 120 minutes
N SD ISD 60 minutes
N SD ISD 120 minutes
Fig. 1. G l o m e r u h r l y s y l h y ~ o x y h ~ . R e s u l t s e x p r e ~ d as c p m [ 1 4 C ] h y d r o x y l y s ~ e s y n t h e s i z e d p e r m g o f e n z y m e p r o t e ~ . V a l u e s r e p r e s e n t m e ~ ~ S.E. o f 8 o b ~ r v a t i o n s . I n c u b a t i o n s p e r f o r m e d f o r 6 0 a n d 1 2 0 rain. N, n o r m M ( a g e - m a t c h e d controls)~ SD, s t r e p t o z o t o c i n ~ i a b e t i c ~ I S D , i n s u l i n - t r e a t e d s t r e p t o z o t o c i n - d i a b e t i c ; *, S i g n i f i c a n t l y d i f f e r e n t f r o m N, P ~ 0.01~ # S i g n i f i c a n t l y d i f f e r e n t f r o m SD, P ~ 0 . 0 1 . Fig. 2. G l o m e ~ prolyl hy&oxyh~. R e s u l t s e x p r e s ~ d as c p m [ 1 4 C ] h y d z o x y p r o l ~ e s y n t h e s i z e d p e r m g of e n z y m e p r o t e ~ . V ~ u e s r e p r e s e n t m e ~ ~ S.E. of 4 o b ~ r v a t i o n s in e a c h g r o u p d e p i c t e d . I n c u b a t i o n s p e r f o ~ e d f o r 60 ~ d 1 2 0 m ~ . N, n o r m a l ~ SD, s t r e p t o z o t o c i n - d i a b e t i c ~ I S D , ~ s ~ i n - t r e a t e d strept o z o t o c ~ diabetic.
92 parable to those observed with lysyl-protocollagen substrate, and similar cofactor requirements and time curves were demonstrated. After 2 h of incubation with 165 pg of [14C]prolyl-protocollagen substrate and 250 t~g/ml of 17 000 × g supernatant protein, 24.3% of proline residues were hydroxylated. Hydroxylation was only 3.9% when substrate and enzyme were incubated without cofactors, and 3.4% when superaatant protein was omitted. In contrast to the findings with [14C]lysyl-protocollagen substrate, glomerular prolyl-hydroxylase activity in preparations from streptozotocin-diabetic rats did not differ significantly from that of control animals (Fig. 2). Daily insulin therapy to streptozotocin-injected rats did not significantly alter glomerular prolyl hydroxylase activity when compared to activity found in homogenates from either normal or untreated streptozotocin-diabetic animals (Fig. 2). The percent hydroxylation of the [~4C]proline substrate did n o t differ among the three experimental groups when similar time periods and enzyme protein concentrations were compared; mean percent [ ~ C ] h y d r o x y p r o l i n e formed after 2 h of incubation with 250 pg/ml of supernatant protein was 25, 22, and 23 respectively for normal, streptozotocin-diabetic, and insulin-treated diabetic preparations. Differences in total glomerular prolyl hydroxylase (calculated activity for each glomerular pool) were also not significant among the experimental groups. Discussion Several lines of evidence have suggested that alterations in glomerular basement membrane biosynthesis and secretion, which may be of importance in the structural changes which occur in diabetic nephropathy, are ameliorated with treatment of the diabetes. Renal cortical glucosyltransferase, which effects the transfer of glucose from UDPglucose to galactosyl-hydroxylysine is elevated in kidney homogenates from alloxan-diabetic rats [20]; with exogenous insulin therapy, glucosyltransferase activity reverts toward normal. Basement membrane synthesis, determined by the appearance of labeled lysine and hydroxylysine in acetic acid insoluble basement membrane obtained from sonication of glomeruli incubated in the presence of [14C]lysine, is increased in preparations from streptozotocin-diabetic animals [21]. Progressive diminution of basement membrane synthesis occurs in incubations containing glomeruli from insulintreated diabetic rats (Cohen and Khalifa, in preparation, and ref. 15). The data presented here support the contention that metabolic control is an important factor in allaying the development of diabetic nephropathy. Glycosylation of hydroxylysine occurs subsequent to lysine hydroxylation; both hydroxylation and glycosylation are accomplished after the incorporation of lysine into peptide linkage but before the molecule leaves the cell [10]. Lysyl hydroxylase and glucosyltransferase are therefore sequentially acting enzymes which are critical to the synthesis of basement membrane collagen. In many biosynthetic pathways with multiple enzymatic processes that occur in sequence, operational control is frequently exerted at the level of one of the earliest enzymatic steps in that pathway. Thus elevated glucosyltransferase probably occurs consequent to increases in lysine hydroxylation. However, the increased lysyl hydroxylase observed in diabetic glomeruli may not represent
93 the earliest change but may simply reflect an effect of diabetes on some more fundamental step in basement membrane collagen synthesis, such as stimulation of transcription or translation mechanisms governing basement membrane production. If glomerular collagen synthesis is generally stimulated as a result of the diabetic state, an increase in prolyl hydroxylase levels would be anticipated since protocollagen proline hydroxylase has been shown to correlate with collagen biosynthesis in several tissues. For example, elevated rates of collagen biosynthesis in fibrotic liver tissue and granulating wounds are accompanied by correspondent increases in the level of prolyl hydroxylase activity [22,23], and proportionate decreases in dermal collagen synthesis and prolyl hydroxylase activity are seen after administration of glucocorticoids [24]. The lack of effect of streptozotocin-diabetes on glomerular proline hydroxylation is puzzling and may be due to the use of interstitial rather than basement membrane protocollagen substrate. Although attempts to prepare the latter from glomeruli incubated with ~,~'-dipyridyl have yielded a collagen precursor of low specific activity and questionable purity, increased prolyl hydroxylation is being sought with other substrates. In the present studies, changes in prolyl hydroxylase would not be detected if the maximal possible hydroxylation of interstitial substrate were a limiting factor. It is possible that prolyl hydroxylase activity does not correlate with collagen biosynthesis in glomerular tissue, or that changes in hydroxyproline-containing units of the glomerular basement membrane are not of particular importance in the diabetic state. The recent report by Grant et al. indicating that [~4C]hydroxyproline synthesis is markedly increased in glomeruli from streptozotocin-diabetic rats makes this unlikely [25]. No protein fraction with an unusually high hydroxylysine/hydroxyproline ratio has been isolated to date from glomerular basement membrane [26], and we have been unable to demonstrate lysine overhydroxylation in basement membrane isolated from incubated glomeruli. However, the exact chemical model of basement membrane is controversial, and changes in a few critically occurring residues may not be detectable until further characterization of normal basement membrane is available. Acknowledgments Presented in part at the Midwest Meeting of the American Federation for Clinical Research, Chicago, 1975. Supported by grants-in-aid from Wayne State University, the American Diabetes Association, the Weight Watchers Foundation, the Detroit General Hospital Research Corporation, and by a Research Grant from the Michigan Department of Public Health. This work was performed in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Physiology, Wayne State University School of Medicine (A.K.). References I K i m m e l s t e i l , P., K i m , O . J . a n d Beres, J. ( 1 9 6 2 ) A m . J. Clin. P a t h . 3 8 , 2 7 0 - - 2 7 9 2 O s t e r b y - H a a s e n ~ R. ( 1 9 6 5 ) D i a b e t o l o g i a 1, 9 7 - - 1 0 0
94
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Spi~o, R.G., (1967) J. Biol. Chem. 242, 1915--1922 Kefalides, N.A. and Denduchis, B. (1969) Biochemistry 8, 4613--4621 Kefalides, N.A. (1973) Int. Rev. Conn. Tiss. Res. 6, 63--104 Beisswenger, P. and Splro, R.G. (1970) Science 168, 596--598 Beisswenger, P. and Spiro, R.G. (1973) Diabetes 22, 180--93 Kefalides, N.A. (1973) Adv. Metab. Disorders Suppl. 2, 167--177 Westberg, N.A. and Michael, A.F. (1973) Acta Med. Scand. 194, 39--47 Grant, M.E. and Prockop, D.J. (1972) New Eng. J. Med. 286, 194--199 Grant, M.E. and Prockop, D.J. (1972) New Eng. J. Med. 286, 242--249 Khalifa, A. and Cohen, M.P. (1975) Biochim. Biophys. Acta 386, 332--339 Westberg, N.A. an d Michael, A.F. (1970) Biochemistry 9, 3837--3846 Cohen, M.P. and Vogt, C.A. (1975) Biochim. Biophys. Acta 393, 78--87 Cohen, M.P. and Khalifa, A. (1976) Diabetes 26, Abstr. Rhoads, R.E., Roberts, N.E. and Udenfriend, S. (1970) Arch. Biochem. Biophys. 139, 329--339 Blumenkrantz, N. and Prockop D.J. (1969) Analyt. Biochem. 39, 377--385 Cohen, M.P. and Vogt, C.A. (1972) Biochem. Biophys. Res. Comm. 49, 1542--1546 Rojkind, M. and Gonzalez, E. (1974) Analyt. Biochem. 57, 1--7 Spiro, R.G. a n d Spi~o, M.J. (1971) Diabetes 20, 641--648 Cohen, M.P. and Khalifa, A. (1974) Diabetes 23, 357 (abstr.) Takeuchi, T., Kivirikko, K.I. and Prockop, D.J. (1967) Biochem. Biophys. Res. Commun. 28, 9 4 0 - 944 Stein, H.D. and Keiser, H.R. (1971) J. Surg. Res. 11, 277--283 Cut~oneo, K.R., Stanford, R., Hwang, P., Hodnette, B. and Dobson, J. (1973) Fed. Proc. 32, 759 (abstr.) Grant, M.E., Ha~wood, R. and Williams, I.F. (1976) J. Physiol. 257, 56--57. Hudson, B.G. and Spiro, R.G. (1972) J. Biol. Chem. 247, 4 2 2 9 - - 4 2 3 8
Biochimica et Biophysica Acta, 496 (1977) 88--94 ~) Elsevier/North-Holland Biomedical Press
BBA 28137 E F F E C T OF DIABETES AND INSULIN ON RAT R E N A L G L O M E R U L A R P R O T O C O L L A G E N H Y D R O X Y L A S E ACTIVITIES
MARGO P. COHEN and AHMED KHALIFA Departments of Medicine and Physiology, Wayne State University School of Medicine, Detroit, Mich. 48201 (U.S.A.)
(Received July 2nd, 1976)
Summary The effect of diabetes and insulin on the activities of both prolyl hydroxylase (trivial name; proline,2-oxoglutarate dioxygenase, EC 1.14.11.2) and lysyl hydroxylase (trivial name; lysine,2-oxoglutarate dioxygenase, EC 1.14.11.4) in isolated rat renal glomeruli was determined. Three groups of experimental animals were used: age-matched controls, streptozotocin-diabetic, and insulintreated streptozotocin-diabetic. Using 14C-labeled lysine or proline hydroxylase substrate prepared from chick e m b r y o tibiae, glomerular 17 000 × g supernatant enzyme was incubated in a complete hydroxylating syste~n for 60 and 120 min. Lysyl hydroxylase activity was significantly increased in diabetic preparations, but prolyl hydroxylase activity did not differ from control. Administration of insulin to streptozotocin-injected animals completely restored glomerular lysyl hydroxylase to normal levels. The results suggest that the specific elevation of lysyl hydroxylase relates to the biochemical changes contributory to diabetic nephropathy, and that insulin may reverse this process.
Introduction Electron microscopy has established that the characteristic lesion of diabetic nephropathy begins with an increase in the width of the glomerular basement membrane and the basement membrane-like material of the mesangial regions [ 1 ]. While the extent of the glomerular lesion shows a positive correlation with the duration of diabetes [2], the nature of the changes resulting in the deposition of excessive amounts of glomerular basement membrane in diabetes remains undefined. Furthermore, the role of hyperglycemia and/or insulin deficiency in the pathogenesis of this lesion is not well established, and the relation of diabetic control to the development of microangiopathic complications is in dispute.
89 Analysis of glomerular basement membrane from several sources has identified a collagen-like glycoprotein that is rich in carbohydrate and contains greater amounts of hydroxylysine and hydroxyproline than other more common types of collagen [3,4]. It is also characterized by a considerable amount of its hydroxyproline as the 3-hydroxyl isomer, and by a much higher percentage of glycosylated hydroxylysyl residues than interstitial collagen [5]. Attention focussed on the importance of the latter in diabetic nephropathy after Beisswenger and Spiro reported increased hydroxylysine residues and hydroxylysine-linked disaccharide units in diabetic glomerular basement membrane [6, 7]. Other investigators, however, have been unable to confirm these findings. Kefalides found no significant differences in the amino acid and carbohydrate compositions of human glomerular basement membrane from normal and diabetic individuals [8]. Westberg and Michael reported that basement membrane isolated from human diabetic glomeruli contained less cystine and sialic acid than that obtained from non-diabetics, while lysine, hydroxylysine and glucosyl-galactosyl-hydroxylysine contents were not significantly different [9]. Chemical analyses have thus not provided a totally rewarding insight into the problem of diabetic nephropathy, and other approaches seem warranted. Since hydroxyproiine and hydroxylysine are important components of basement membrane collagen, the effect of diabetes on their synthesis in glomerular tissue is of considerable interest. Hydroxyproline and hydroxylysine are both synthesized by separate postribosomal enzymatic hydroxylations that occur in a polypeptide precursor known as protocollagen [10,11]. We recently reported that lysyl hydroxylase activity was elevated in preparations obtained from isolated glomeruli of streptozotocin-diabetic rats, and suggested that chronic hyperglycemia was implicated in these changes, which could reflect increased collagen biosynthesis. The effect of exogenous insulin therapy on glomerular lysyl hydroxylase is now reported, as is the specificity of this change. Prolyl hydroxylase has been correlated with collagen biosynthesis in several tissues, which prompted study of the effect of diabetes and insulin on this parameter in the present report. Materials and Methods
Experimental animals. Three groups of experimental animals (SpragueDawley White Rats} were used: age-matched controls, streptozotocin-di~betic, and insulin-treated streptozotocin-diabetic. Diabetic animals were killed 4½ weeks after intraperitoneal injection of streptozotocin, 85 mg]kg. Lente insulin (2 u/kg per day subcutaneously) was begun the day following streptozotocin injection and was continued until time of sacrifice. Preparation of tissue. Kidneys were obtained immediately after sacrifice by anesthetic overdose, and the renal cortex was separated by gross dissection. Glomeruli were isolated essentially as described by Spiro [3], sequentially using stainless-steel meshes of 100, 60 and 230 gauge (respective pore sizes of 140, 250 and 63/am). Details of the sieving process, and collection and analysis of the glomeruli have been described [13,14]. Glomeruli so isolated are capable of in vitro basement membrane synthesis [14], and basement membrane collagen is the major collagen synthesized by these preparations [15]. The 17 000 × g
90 supernatant, obtained as previously described [12], was used as enzyme source. Kidneys were smaller and glomerular yields were less in the diabetic group than in the normal or insulin-treated groups (0.5--0.8 - 104 versus 1--2 • 104 glomeruli per g of tissue, respectively); homogenization volumes were adjusted so as to give similar supernatant protein concentrations. Protocollagen substrates. Tibiae were obtained from 11 day old chick embryos and incubated for 5 h in Krebs buffer made 10 -3 M in ~,~'-dipyridyl and containing 100 pCi of either [U-14C]lysine or [U-~4C]proline in a total volume of 10 ml. The [~4C]lysine and [14C]proline-rich protocollagen substrates were extensively extracted in 0.5 M acetic acid [16], yielding preparations with the following specific activities: [~C]lysyl-hydroxylase substrate about 1 4 0 0 0 0 cpm/mg, and [~4C]prolyl-hydroxylase substrate about 1 6 5 0 0 0 cpm/mg. Protocollagen hydroxylase assays. Incubations for assay of lysyl and prolylprotocollagen hydroxylases were performed for 60--120 min at 37°C in an atmosphere of O2/CO2 (95 : 5, v/v). The standard incubation mixture contained in a t o t a ! volume of 2 ml: 0.5 mM a-ketoglutarate; 62.5 mM Tris. HC1 buffer (pH 8.0); 2 mM ascorbate; 0.125 mM FeSO4; 0.1 mM dithiothreitol; 1.5 mg/ml bovine serum albumin; 150 pg [~4C]lysyl hydroxylase substrate or 165 pg [14C]prolyl hydroxylase substrate; and 120--250 pg/ml glomerular enzyme. Immediately following incubation, samples were made 6 M in HC1 and hydrolyzed for 18 h at 115°C. The hydrolysates were filtered, evaporated in vacuo, and reconstituted in distilled water. [ ~ C ] H y d r o x y l y s i n e was assayed by periodate oxidation [17] as well by thin layer chromatographic separation [18]; agreement between these two methods was excellent. [~4C]Hydroxyproline was determined by the m e t h o d described by Rojkind and Gonzalez [19] ; reliability of this method had been confirmed by column chromatographic separation of labeled proline and hydroxyproline in preliminary experiments. Results
Animals. Pertinent data concerning the experimental animals is presented in Table I. The streptozotocin-diabetic rats were markedly hyperglycemic and exhibited stunted growth 4½ weeks after injection. The degree of hyperglycemia was significantly diminished, although not to euglycemic levels, by daily insulin administration; growth patterns reverted toward normal in the insulin-treated animals. Lysyl hydroxylase activity. Optimum conditions for protocollagen [14C]lysyl hydroxylation by rat glomerular supernatant fractions have been previously reported [12], and were re-confirmed in the present series of experiments. Essentially we have found that omission of cofactors (a-ketoglutarate, ascotbate, and ferrous sulfate) or of enzyme resulted in markedly reduced levels of hydroxylation. Lysine hydroxylation is proportional to time for up to 120 min o f incubation studied, and to 17 000 × g supernatant enzyme from 50--250 pg/ ml. Protocollagen lysyl hydroxylase activity was significantly increased in preparations from glomeruli of streptozotocin-diabetic rats compared to activity found in glomeruli from age-matched controls. This increase was apparent
91 TABLE BODY
I WEIGHT
Number
AND
BLOOD
GLUCOSE
VALUES
OF EXPERIMENTAL
ANIMAL
GROUPS
of animals given in parenthesis.
Group
B o d y weight (g)
Age-matched controls 3 2 4 -+ 4 (10) Diabetic (4 1/2 weeks post-streptozotocin) 1 4 5 + 12 * (9) Insulin-treated diabetic (4 1/2 weeks post-streptozotocin; Lente insulin (2 u/kg per day) begun 1 day after streptozotocin) 2 7 5 -+ 15 ** (10)
Blood glucose (rag%) 120 +- 10 428 -+ 30 *
2 3 3 +- 25 **
• P < 0 . 0 0 1 , c o m p a r i n g n o r m a l to d i a b e t i c values. • * P < O . 0 0 1 , c o m p a r i n g d i a b e t i c to i n s u l i n - t r e a t e d d i a b e t i c values.
when activity was expressed either as cpm [14C]hydroxylysine formed per unit time or cpm [~4C]hydroxylysine synthesized per mg of supernatant protein (Fig. 1). The administration of insulin on a daily basis to streptozotocin-injected rats resulted in a complete restoration of glomerular lysyl hydroxylase activity to normal levels. This correction, which, was also significant when expressed either as cpm/time or as cpm/mg protein (Fig. 1), occurred despite the absence of complete normalization of the blood glucose to euglycemic values. Prolyl-hydroxylase activity. Prolyl-protocollagen hydroxylase activity was readily detectable in homogenates of rat renal glomeruli. Conditions were com.E
~o20~
.~_ °
-I-__
i
~m16 -~IE
g •
•~~ 4-8-
~0 ~)< 16-
~
~
N SD ISD 60 minutes
~~E 4-8-
N SD ~SD 120 minutes
N SD ISD 60 minutes
N SD ISD 120 minutes
Fig. 1. G l o m e r u h r l y s y l h y ~ o x y h ~ . R e s u l t s e x p r e ~ d as c p m [ 1 4 C ] h y d r o x y l y s ~ e s y n t h e s i z e d p e r m g o f e n z y m e p r o t e ~ . V a l u e s r e p r e s e n t m e ~ ~ S.E. o f 8 o b ~ r v a t i o n s . I n c u b a t i o n s p e r f o r m e d f o r 6 0 a n d 1 2 0 rain. N, n o r m M ( a g e - m a t c h e d controls)~ SD, s t r e p t o z o t o c i n ~ i a b e t i c ~ I S D , i n s u l i n - t r e a t e d s t r e p t o z o t o c i n - d i a b e t i c ; *, S i g n i f i c a n t l y d i f f e r e n t f r o m N, P ~ 0.01~ # S i g n i f i c a n t l y d i f f e r e n t f r o m SD, P ~ 0 . 0 1 . Fig. 2. G l o m e ~ prolyl hy&oxyh~. R e s u l t s e x p r e s ~ d as c p m [ 1 4 C ] h y d z o x y p r o l ~ e s y n t h e s i z e d p e r m g of e n z y m e p r o t e ~ . V ~ u e s r e p r e s e n t m e ~ ~ S.E. of 4 o b ~ r v a t i o n s in e a c h g r o u p d e p i c t e d . I n c u b a t i o n s p e r f o ~ e d f o r 60 ~ d 1 2 0 m ~ . N, n o r m a l ~ SD, s t r e p t o z o t o c i n - d i a b e t i c ~ I S D , ~ s ~ i n - t r e a t e d strept o z o t o c ~ diabetic.
92 parable to those observed with lysyl-protocollagen substrate, and similar cofactor requirements and time curves were demonstrated. After 2 h of incubation with 165 pg of [14C]prolyl-protocollagen substrate and 250 t~g/ml of 17 000 × g supernatant protein, 24.3% of proline residues were hydroxylated. Hydroxylation was only 3.9% when substrate and enzyme were incubated without cofactors, and 3.4% when superaatant protein was omitted. In contrast to the findings with [14C]lysyl-protocollagen substrate, glomerular prolyl-hydroxylase activity in preparations from streptozotocin-diabetic rats did not differ significantly from that of control animals (Fig. 2). Daily insulin therapy to streptozotocin-injected rats did not significantly alter glomerular prolyl hydroxylase activity when compared to activity found in homogenates from either normal or untreated streptozotocin-diabetic animals (Fig. 2). The percent hydroxylation of the [~4C]proline substrate did n o t differ among the three experimental groups when similar time periods and enzyme protein concentrations were compared; mean percent [ ~ C ] h y d r o x y p r o l i n e formed after 2 h of incubation with 250 pg/ml of supernatant protein was 25, 22, and 23 respectively for normal, streptozotocin-diabetic, and insulin-treated diabetic preparations. Differences in total glomerular prolyl hydroxylase (calculated activity for each glomerular pool) were also not significant among the experimental groups. Discussion Several lines of evidence have suggested that alterations in glomerular basement membrane biosynthesis and secretion, which may be of importance in the structural changes which occur in diabetic nephropathy, are ameliorated with treatment of the diabetes. Renal cortical glucosyltransferase, which effects the transfer of glucose from UDPglucose to galactosyl-hydroxylysine is elevated in kidney homogenates from alloxan-diabetic rats [20]; with exogenous insulin therapy, glucosyltransferase activity reverts toward normal. Basement membrane synthesis, determined by the appearance of labeled lysine and hydroxylysine in acetic acid insoluble basement membrane obtained from sonication of glomeruli incubated in the presence of [14C]lysine, is increased in preparations from streptozotocin-diabetic animals [21]. Progressive diminution of basement membrane synthesis occurs in incubations containing glomeruli from insulintreated diabetic rats (Cohen and Khalifa, in preparation, and ref. 15). The data presented here support the contention that metabolic control is an important factor in allaying the development of diabetic nephropathy. Glycosylation of hydroxylysine occurs subsequent to lysine hydroxylation; both hydroxylation and glycosylation are accomplished after the incorporation of lysine into peptide linkage but before the molecule leaves the cell [10]. Lysyl hydroxylase and glucosyltransferase are therefore sequentially acting enzymes which are critical to the synthesis of basement membrane collagen. In many biosynthetic pathways with multiple enzymatic processes that occur in sequence, operational control is frequently exerted at the level of one of the earliest enzymatic steps in that pathway. Thus elevated glucosyltransferase probably occurs consequent to increases in lysine hydroxylation. However, the increased lysyl hydroxylase observed in diabetic glomeruli may not represent
93 the earliest change but may simply reflect an effect of diabetes on some more fundamental step in basement membrane collagen synthesis, such as stimulation of transcription or translation mechanisms governing basement membrane production. If glomerular collagen synthesis is generally stimulated as a result of the diabetic state, an increase in prolyl hydroxylase levels would be anticipated since protocollagen proline hydroxylase has been shown to correlate with collagen biosynthesis in several tissues. For example, elevated rates of collagen biosynthesis in fibrotic liver tissue and granulating wounds are accompanied by correspondent increases in the level of prolyl hydroxylase activity [22,23], and proportionate decreases in dermal collagen synthesis and prolyl hydroxylase activity are seen after administration of glucocorticoids [24]. The lack of effect of streptozotocin-diabetes on glomerular proline hydroxylation is puzzling and may be due to the use of interstitial rather than basement membrane protocollagen substrate. Although attempts to prepare the latter from glomeruli incubated with ~,~'-dipyridyl have yielded a collagen precursor of low specific activity and questionable purity, increased prolyl hydroxylation is being sought with other substrates. In the present studies, changes in prolyl hydroxylase would not be detected if the maximal possible hydroxylation of interstitial substrate were a limiting factor. It is possible that prolyl hydroxylase activity does not correlate with collagen biosynthesis in glomerular tissue, or that changes in hydroxyproline-containing units of the glomerular basement membrane are not of particular importance in the diabetic state. The recent report by Grant et al. indicating that [~4C]hydroxyproline synthesis is markedly increased in glomeruli from streptozotocin-diabetic rats makes this unlikely [25]. No protein fraction with an unusually high hydroxylysine/hydroxyproline ratio has been isolated to date from glomerular basement membrane [26], and we have been unable to demonstrate lysine overhydroxylation in basement membrane isolated from incubated glomeruli. However, the exact chemical model of basement membrane is controversial, and changes in a few critically occurring residues may not be detectable until further characterization of normal basement membrane is available. Acknowledgments Presented in part at the Midwest Meeting of the American Federation for Clinical Research, Chicago, 1975. Supported by grants-in-aid from Wayne State University, the American Diabetes Association, the Weight Watchers Foundation, the Detroit General Hospital Research Corporation, and by a Research Grant from the Michigan Department of Public Health. This work was performed in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Physiology, Wayne State University School of Medicine (A.K.). References I K i m m e l s t e i l , P., K i m , O . J . a n d Beres, J. ( 1 9 6 2 ) A m . J. Clin. P a t h . 3 8 , 2 7 0 - - 2 7 9 2 O s t e r b y - H a a s e n ~ R. ( 1 9 6 5 ) D i a b e t o l o g i a 1, 9 7 - - 1 0 0
94
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Spi~o, R.G., (1967) J. Biol. Chem. 242, 1915--1922 Kefalides, N.A. and Denduchis, B. (1969) Biochemistry 8, 4613--4621 Kefalides, N.A. (1973) Int. Rev. Conn. Tiss. Res. 6, 63--104 Beisswenger, P. and Splro, R.G. (1970) Science 168, 596--598 Beisswenger, P. and Spiro, R.G. (1973) Diabetes 22, 180--93 Kefalides, N.A. (1973) Adv. Metab. Disorders Suppl. 2, 167--177 Westberg, N.A. and Michael, A.F. (1973) Acta Med. Scand. 194, 39--47 Grant, M.E. and Prockop, D.J. (1972) New Eng. J. Med. 286, 194--199 Grant, M.E. and Prockop, D.J. (1972) New Eng. J. Med. 286, 242--249 Khalifa, A. and Cohen, M.P. (1975) Biochim. Biophys. Acta 386, 332--339 Westberg, N.A. an d Michael, A.F. (1970) Biochemistry 9, 3837--3846 Cohen, M.P. and Vogt, C.A. (1975) Biochim. Biophys. Acta 393, 78--87 Cohen, M.P. and Khalifa, A. (1976) Diabetes 26, Abstr. Rhoads, R.E., Roberts, N.E. and Udenfriend, S. (1970) Arch. Biochem. Biophys. 139, 329--339 Blumenkrantz, N. and Prockop D.J. (1969) Analyt. Biochem. 39, 377--385 Cohen, M.P. and Vogt, C.A. (1972) Biochem. Biophys. Res. Comm. 49, 1542--1546 Rojkind, M. and Gonzalez, E. (1974) Analyt. Biochem. 57, 1--7 Spiro, R.G. a n d Spi~o, M.J. (1971) Diabetes 20, 641--648 Cohen, M.P. and Khalifa, A. (1974) Diabetes 23, 357 (abstr.) Takeuchi, T., Kivirikko, K.I. and Prockop, D.J. (1967) Biochem. Biophys. Res. Commun. 28, 9 4 0 - 944 Stein, H.D. and Keiser, H.R. (1971) J. Surg. Res. 11, 277--283 Cut~oneo, K.R., Stanford, R., Hwang, P., Hodnette, B. and Dobson, J. (1973) Fed. Proc. 32, 759 (abstr.) Grant, M.E., Ha~wood, R. and Williams, I.F. (1976) J. Physiol. 257, 56--57. Hudson, B.G. and Spiro, R.G. (1972) J. Biol. Chem. 247, 4 2 2 9 - - 4 2 3 8