Kinetics of fast haemoglobin in diabetic rats

Diabetologia, Apr 1982

This study was designed to examine the appearance and disappearance kinetics of glycosylated haemoglobin during abrupt changes of blood glucose in the rat. The concentration of the fast haemoglobin component, which has similar chromatographic and electrophoretic profiles to human haemoglobin A1, was measured after the induction of diabetes by streptozotocin and its cure by syngeneic intraportal islet transplantation. Fast haemoglobin was increased in 12 diabetic rats compared with 22 controls(15.8±0.8 versus 8.2±0.3%, mean ± SEM). In a group with mild diabetes (n= 8, blood glucose < 22 mmol/l), fast haemoglobin rose to 13.7±1.0% by week 8. In a group with severe diabetes (n=4, blood glucose > 22 mmol/l), fast haemoglobin rose more quickly (in 3 weeks) to a higher level (18.2±3.3%) and changed little thereafter. This suggests a saturable system in which the rate of increase and final value depend upon the degree of hyperglycaemia. After islet transplantation, fast haemoglobin returned to normal in 4 weeks (n=5, 17.6±1.4 to 9.4±0.9%). This delay is shorter than expected from the red cell lifespan (around 60 days), suggesting that haemoglobin glycosylation may be partly reversible. These results suggest that in unstable diabetes the interpretation of haemoglobin A1 levels is not as simple as was supposed previously.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1007%2FBF00281303.pdf

Kinetics of fast haemoglobin in diabetic rats

Diabetologia Kinetics of Fast Haemoglobin in Diabetic Rats B. Vialettes l 0 L. Vovan 0 M. C. Simon ~ 0 V. Lassmann I 0 E. Altomare I 0 Ph. Vague 0 0 1Department of Internal Medicine, Hrpital Universitaire Michel Levy and 2Haemoglobinopathy and Paediatric Haematology Laboratory, Hopital Universitaire La Timone , Marseille , France Summary. This study was designed to examine the appearance and disappearance kinetics of glycosylated haemoglobin during abrupt changes of blood glucose in the rat. The concentration of the fast haemoglobin component, which has similar chromatographic and electrophoretic profiles to human haemoglobin Ab was measured after the induction o f diabetes by streptozotocin and its cure by syngeneic intraportal islet transplantation. Fast haemoglobin was increased in 12 diabetic rats compared with 22 controls (15.8 + 0.8 versus 8.2 + 0.3%, mean + SEM). In a group with mild diabetes (n = 8, blood glucose < 22 mmol/1), fast haemoglobin rose to 13.7 ___ 1.0% by week 8. In a group with severe diabetes (n = 4, blood glucose > 22 mmol/1), fast haemoglobin rose more quickly (in 3 weeks) to a higher level (18.2 + 3.3%) and changed little thereafter. This suggests a saturable system in which the rate of increase and final value depend upon the degree of hyperglycaemia. After islet transplantation, fast haemoglobin returned to normal in 4 weeks (n = 5, 17.6 ___ 1.4 to 9.4 + 0.9%). This delay is shorter than expected from the red cell lifespan (around 60 days), suggesting that haemoglobin glycosylation may be partly reversible. These results suggest that in unstable diabetes the interpretation of haemoglobin A1 levels is not as simple as was supposed previously. Glycosylated haemoglobin; streptozotocin diabetes; islet transplantation; rat 9 Springer-Verlag 1982 Increase of haemoglobin Alc (HbAlc) in diabetic patients was first described by Rahbar et al. [ 21 ]and later linked to hyperglycaemia by Trivelli et al. [ 27 ]. There is convincing evidence that glycosylation of HbA to form HbAlc is secondary to hyperglycaemia and occurs during the entire erythrocyte lifespan [ 2, 5, 6, 24 ]. Accordingly, HbA1 (HbAl(a+b+c)) determinations have been proposed as a means of evaluating blood glucose control over a prolonged period of time [ 7, 9, 16, 19 ]. Unfortunately, despite clinical studies [ 3, 4, 28 ], the kinetics of HbA1 level during abrupt changes of diabetic control remain unknown. In an attempt to elucidate this point, an experimental rat model was designed to achieve successively the rapid induction of hyperglycaemia, stable sustained diabetes, and rapid normalization of plasma glucose. We have studied a fast minor haemoglobin component in rats which has similar electrophoretic and chromatographic profiles to human HbAv The animals were rendered diabetic by streptozotocin and subsequently cured by intraportal islet transplantation. Material and Methods The animals used were female inbred Lewis rats (CNRS Centre de Srlection et d'Elevage d'Animaux de Laboratoire, Orlrans) weighing 170-210 g at the beginning of experiment. Diabetes was induced by IV injection of 65 mg/kg streptozotocin (Sigma) in 0.005 mol/l citrate buffer (pH 4.5). Eighteen rats were injected. No death occurred. Four animals did not become diabetic. Fourteen became diabetic and of them 12 were followed in this study. Severe diabetes was defined as a mean plasma glucose above 22 mmol/1 (n = 4) during follow-up in any animal. Mild diabetes was taken as a mean plasma glucose between 11 and 22 mmol/1 (n = 8). Islets of Langerhans were isolated from rats as described by Lacy and Kostianowsky [ 17 ].The donor pancreases were distended in vivo by Hanks' balanced salt solution (pH 7.4), removed, minced and digested with 15 mg of collagenase (type IV, Worthington Biochemicals) for every two pancreases for 5-9 min with hand shaking in a waterbath at 37 ~C. After two washes in Hank's solution, the islets were counted and hand-picked under a dissecting microscope. For transplantation, 500-1000 islets were slowly embolized into the liver by infusion into a mesenteric vein tributary of the portal vein under ether anaesthesia. Blood (500 lxl) was collected into heparinized capillary tubes from a tail vein of non-fasting normal or diabetic rats and was microfuged (Eppendorf microfuge) for 2 rain. A 20 ~tl aliquot of plasma was kept frozen at - 2 0 ~ for subsequent glucose assay (glucose oxidase method, peroxidase-glucose oxidase diagnostica, Roche). The remaining plasma was discarded and the ceils were washed three times in 0.154 tool/1 saline. Erythrocytes were haemolysed by the addition of two volumes of distilled water and two H1 R Hs + II II II1 I i~I 'i;I'I; I I I I" l I,I! !II! ,J A 2 c A I i| i! I I Ii II i I "" III1! I!I R3 drops of toluene. After vortex mixing the blood was kept for one night at 4 ~ The solution was then spun at 10000 rev./min for 1 min. Electrophoretic isolation of haemoglobin fractions was done on cellulose acetate (cellogel 2.5 x 17 cm band) in Tris-EDTA-borate buffer (pH 9.0) under 200 volts for 2 h. Bands were stained after migration by amido-Schwartz colouration. Fractions were then located, cut out and dissolved in acetic acid 80% (v/v). Optical density (OD) was read at 625 nm. Four main fractions could be isolated from haemolysate by this procedure (Fig. 1). The major component, R3, migrated as h u m a n HbA and represented approximately 50% of total Hb. R2 migrated coincidently with h u m a n HbS of sicklecell disease and R1 as carbonic anhydrase. R2 and R1 represented about 45% of whole Hb. Fast Hb (R4) migrated as human glycosylated Hb and its level was determinated by the following calculation: I ,,, A Sixty b l o o d samples were collected from 12 diabetic animals regardless o f the date o f streptozotocin injection (Fig.3). A correlation was f o u n d between the plasma glucose and fast H b level (r = 0.30,p < 0.05). When samples collected before, 20 days and 50-60 days after streptozotocin injection were considered separately, a statistically significant correlation was found only for the last group (r = 0.58, p < 0.05; Fig. 4). . .. "% ,| .::..:  9 "e"l . ~" | 15. ,~ 105153272115921 15 9 3 E 8 3J~ 21- day 0 Kineticsof Fast Haemoglobin ChangesAfter Induction of Diabetesby Streptozotocin Two groups of diabetic rats were isolated according to severity of hyperglycaemia (Fig. 5): group 1 with severe diabetes and group 2 with mild diabetes. In the first group stable hyperglycaemia was established by day 5 following the streptozotocin injection (mean E E 8 4:= 1510oJ 28. 22,4. 162'5 E o = 11,1 NO_ 5,6 1 {',, Days 25.9 _+ 0.8 mmol/1). In the second group the rise in plasma glucose concentration was less regular and led to a lower level (20 + 2.2 mmol/1). The mean plasma glucose level was significantly different in the two groups (p < 0.005). Fast Hb changes followed those of plasma glucose with different kinetics in each group. In group 1, fast Hb level increased rapidly towards a maximum (18.2 + 3.3%) at 20 days and was no higher (16.8 4- 0.7%) at 50 days. In contrast, in group 2 the increase was more progressive and led to a lower value (13.7 _+ 1%) 60 days after streptozotocin injection. There was a significant difference between the two groups (p < 0.05) in fast Hb levels at 2 months. Changes of Fast HaemoglobinAfter Islet Transplantation Five diabetic rats with high plasma glucose and fast H b levels sustained for a least 2 weeks (24.7 _+ 1.4 mmol/l and 17.6 + 1.4%, respectively) were transplanted intraportally with islets of Langerhans isolated from adult syngeneic donors. Patterns of plasma glucose and fast Hb changes are represented in Figure 6. Plasma glucose dropped rapidly by the day after transplantation and became normal between weeks 2 and 3. Fast Hb levels diminished more slow2015E lOg .s LL 028_ 22,~ E E 16,7. 8 ~ 11,1 E g 5,6 i i 0 20 Days 4O 60 ly, became comparable with non-diabetic values about 5 weeks after transplantation and thereafter remained stable throughout the observation period of 60 days. Discussion Although extrapolation of results from animal experiments to man requires caution, some similarities exist between our animal model and human diabetes. Rat fast Hb seems to be analogous to human HbA1 on both chromatographic and electrophoretic study. The increase in fast Hb induced by hyperglycaemia is an additional argument in favour of similarities in structure and glycosylation. The lifespan of rat erythrocytes is not established but results of isotopic studies range between 45 and 70 days with an average of 60 days [ 8, 22 ]. Diabetes induced by beta-cytotoxic drugs in rats displays the same metabolic and structural features observed in human diabetes, including insulin deficiency [ 11, 22 ], hyperglucagonaemia [ 12 ], hypertriglyceridaemia [ 23 ] and glomerulopathy [ 18 ]. Koenig et al. [ 14, 15 ] have studied other animal models from which an analogue of human HbAlc has been isolated, These include C57B1/Ksj mice, among which bearers of the db/db gene are spontaneously diabetic, and mice rendered diabetic by the administration of the B cell-cytotoxic drugs, alloxan and streptozotocin. Studies by these authors in chemically induced diabetes suggested that the HbAlc levels, although increased in the diabetic population, did not correlate with either the degree of hyperglycaemia or the duration of diabetes. In contrast, our results show that fast Hb changes in the diabetic rat depend on both the plasma glucose level and the duration ofhyperglycaemia, at least in the early stages. Hyperglycaemia induced by streptozotocin injection caused a progressive rise in fast Hb levels over the succeeding weeks. The degree of hyperglycaemia seemed to act more upon the speed of formation of fast Hb than upon its final level. A maximum value of fast Hb was obtained by week 3 in the group with severe diabetes, whereas in the group with mild diabetes the fast Hb level was still rising up to 8 weeks after streptozotocin. In db/db mice studied by Koenig et al. [15], HbA] rose in parallel with the plasma glucose from 11 to 40 mmol/1 with a delay of 4 weeks. The rapidity of fast Hb formation and the sustained high values after the first month in our group 1 rats and in mice [ 15 ] do not agree with the kinetics deduced from the persistence of a hypothetical glycosylation throughout the entire erythrocyte lifespan. These results suggest a saturation of glycosylation beyond a certain value. This saturation level could be reached sooner or later depending on the degree of hyperglycaemia. This phenomenon could explain the lack of difference in H b & level between severely and mild diabetic mice found by Koenig et al. [ 15 ]. Saturable glycosylation of HbA has also been suspected in man by Graf et al. [ 10 ]. Clinical studies by Dunn et al. [ 4 ] and Vague et al. [ 28 ] have shown that in insulin treated diabetic patients, HbA1 correlated well with the last month's glycosuria but less well with those of the two preceding months. The second part of this work confirmed the ability of syngeneic islet transplantation to reverse the metabolic consequences of chemically induced diabetes [ 13, 20, 25, 29, 30 ]. Islet transplantation also prevents the development of specific glomerular lesions of diabetes in the rat [ 18 ]. This study shows that rapid normalization of the HbAa analogues is another feature of this treatment. The rate of change of fast Hb was again more rapid than expected. The short delay of 5 weeks necessary to achieve normalization of fast Hb does not agree with the hypothetical very slow reversibility of haemoglobin glycosylation described by Fluckiger and Winterhalter in human eythrocytes in vitro [ 6 ].A simple progressive disappearance of erythrocytes previously exposed to hyperglycaemia would take longer to normalize HbA1. Clinical studies are in agreement with the results presented here. In cases of remission of juvenile diabetes, Ditzel and Kjaergaard [ 3 ] and Vague et al. [ 28 ]reported that normalization of HbA1 occurred by the first month following the return of plasma glucose to normal values. There are two possible explanations for these findings. First the decay curve ofglycosylatedhaemoglobin following normalization of blood glucose would be expected to be sigmoidal since the earliest cells to disappear are the oldest (and hence more glycosylated) ones and the last to go the least glycosylated.The other explanation is that glycosylation of haemoglobin may at least in part be reversible. Svendsen et al. [ 26 ] have demonstrated with human HbAlc isolated by chromatography that rapid changes in vivo or in vitro are not related to the stable HbAlo (ketoamine linked glucose), but to reversible adduct of glucose to haemoglobin. Variations of this adduct may, at least in part, account for the rapid decrease in glycosylated haemoglobin found after glucose normalization in both animals and man. Acknowledgements.The authors are grateful to S.Halimi and J. Celotto for preparation of the manuscript. 1. Beccaria L , Chiumello G , Gianazza E , Lippis B , Righetti PG ( 1978 ) Haemoglobin Alc repartition by isoelectric focussing . AmJ Hematol 4 : 367 - 374 2. Bunn HF , Haney DN , Kamin S , Gabbay KH , Gallop PH ( 1976 ) The biosynthesis of human haemoglobin Alo:slow glycosylation of haemoglobinin vitro . J Clin Invest 57 : 1652 - 1659 3. Ditzel J , Kjaergaard JJ ( 1978 ) Haemoglobin Aicconcentrations after initialinsulintreatment for newly discovered diabetes . Br Med J 1: 741 742 4. Dunn PJ , Firoozabadi H , Cole RA , Younger D , Graham CA , Gleason RE , Soeldner JS ( 1978 ) Relative influence of remote and proximal glycaemic events on glycosylated haemoglobin % . Diabetes 27 : 434 (Abstract) 5. Fizgibbons JF , Koler RD , Jones RT ( 1976 ) Red cell age-related changes of haemoglobin Ala+band Aloin normal and diabetic subjects . J Clin Invest 58 : 820 - 824 6. Fluckiger R , Winterhalter KH ( 1976 ) In vitro synthesis of haemoglobinAle . FEBS Lett 71 : 356 - 360 7. Gabbay KH , Hasty K , Breslow JL , Ellison RC , Bunn HF , Gallop PM ( 1977 ) Glycosylated haemoglobins and longterm blood glucose control in diabetes mellitus . J Clin Endocrinol Metab 44 : 859 - 864 8. Ganzoni AM ( 1970 ) Kinetikund Regulation der ErythrocytenProduktion . Springer, Berlin Heidelberg New-York 9. Gonen B , RubensteinAH ( 1978 ) Haemoglobin A1and diabetes mellitus . Diabetologia 15 : 1 - 8 10. Graf ILl , Halter JB , Porte D ( 1978 ) Glycosylated haemoglobin in norm/tl subjects and subjects with maturity onset-diabetes. Evidence for saturable system in man . Diabetes 27 : 834 - 839 11. Junod A , Lambert AE , Stauffacher W , Renold AE ( 1969 ) Diabetogenic action of streptozotocin: relationship of dose to metabolic response . J Clin Invest 48 : 2129 - 2139 12. Katsilambros N , Rahman YA , Hinz M ( 1970 ) Action of streptozotocin on insulin and glucagon responses of rat islet . Horm Metab Res 2 : 268 - 270 13. Kemp CB , Knight MJ , Scharp DW , Lacy PE , Ballinger WF ( 1973 ) Transplantationof isolated pancreatic islets into the portal vein of diabetic rats . Nature 244 : 447M48 14. Koenig RJ , Cerami A ( 1975 ) Synthesis of minor haemoglobin in normal and diabetic mice: A potential model of basement membranethickening . Proc Natl Acad Sci USA 72 : 3687 - 3 691 15. Koenig RJ , Araujo DC , Cerami A ( 1976 ) Increased haemoglobin Ale in diabetic mice Diabetes 25 : 1 5 16. Koenig RJ , Peterson CM , Jones RL , Saudek C , Lehrman M , Cerami A ( 1976 ) Correlation of glucose regulation and haemoglobinAlcin diabetes mellitus . N Engl J Med 295 : 417 - 420 17. Lacy PE , Kostianowsky M ( 1967 ) Method for the isolation of intact islets of Langerhans from rat pancreas . Diabetes 16 : 35 - 39 18. Mauer SM , Suthefland DER , Steffes MW , Leonard ILl , Najarian JS , Michael AF , Brown DM ( 1974 ) Pancreatic transplantation . Effects on the glomerular lesions of experimental diabetes in the rat . Diabetes 23 : 748 ~ 53 19. Paulsen EP , Koury M ( 1977 ) HaemoglobinAI~level in insulindependent and independent diabetes mellitus . Diabetes 26 : 890 896 20. Pipeleers-Marichal MA , Pipeleers DC , Cutler J , Lacy PE , Kiphis DM ( 1976 ) Metabolic and morphologic studies in intraportal-is!et-transplanted rats . Diabetes 25 : 1041 - 1051 21. Rahbar S , Blumenfeld O , Ranney HM ( 1969 ) Studies of an unusual haemoglobinin patients with diabetes mellitus . Biochem Biophys Res Commun 36 : 838 - 843 22. Schalm OW ( 1965 ) Veterinary hematology . Lea Lebiger , Philadelphia 23. Schnatz JD , Formaniak JM , Chlouverakis C ( 1972 ) The origin of hypertriglyceridaemia in streptozotocin diabetics rats . Diabetologia 8 : 125 130 24. Spicer KM , Allen RC , Hallet D , Buse MG ( 1979 ) Synthesis of haemoglobin Ale and related minor haemoglobin by erythrocytes: in vitro study of regulation . J Clin Invest 64 : 40 - 48 25. Steffes MW , Sutherland DER , Mauer SM , Leonard RJ , Najarian JS , Brown DM ( 1975 ) Plasma insulinand glucose levels in diabetic rats prior to and following islet transplantation . J Lab Clin Med 85 : 75 - 81 26. Svendsen PAA , Christiansen JS , Soegaard U , Welinder BS , Nerup J ( 1980 ) Rapid changes in chromatographically determined haemoglobin Aleinduced by short-term changes in glucose concentration . Diabetologia 19 : 130 - 136 27. Trivelli LA , Ranney HM , Lai HT ( 1971 ) Haemoglobin compogents in patients with diabetes mellitus . N Engl J Med 284 : 353 - 357 28. Vague PH , Vovan L , Vialettes B , Boustiere C , Serin N ( 1979 ) Concentration sanguine en h6moglobine glycosyl6e (Ale), indice du contr61e du diab6te dans le mois pr6c6dent . Nouv Presse Med 8 : 491 - 494 29. Vialettes B , Sutherland DER , Matas AJ , Payne WD , Najarian JS ( 1979 ) Amelioration of streptozotocin induced diabetes in rats. Effect of islet isografts on plasma lipids and other metabolic abnormalities . Metabolism 28 : 489 - 494 30. Ziegler MN , Reck~trdCR, Barker CF ( 1974 ) Long term metabolic and immunological consideration in transplantation of pancreatic islets . J SurgRes 16 : 575 - 581


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2FBF00281303.pdf

B. Vialettes, L. Vovan, M. C. Simon, V. Lassmann, E. Altomare, Ph. Vague. Kinetics of fast haemoglobin in diabetic rats, Diabetologia, 1982, 264-268, DOI: 10.1007/BF00281303