Association between erythrocyte Na⁺K⁺-ATPase activity and some blood lipids in type 1 diabetic patients from Lagos, Nigeria

Diabetes mellitus, a metabolic disease characterized by altered carbohydrate, lipid and protein metabolism and requiring strict glycemic control for delaying or preventing cardiovascular, nephropathic and neurological complications in humans remains a public health problem in developing and developed countries of the world [1]. Several studies in animals and humans have shown that pathogenic course of both Type 1 and Type 2 diabetes mellitus involves alterations in the structures, organization and protein functions of membranes of cells and tissues (e.g. retina, glomerulli, erythrocyte, nerve), culminating in diabetic complications such as retinopathy, nephropathy and peripheral neuropathy [2–4].

Na⁺K⁺-ATPase is a soluble conserved trimeric pump (α, 133 kDa, β, 35 kDa, γ, 10 kDa) involved in transmembrane cation regulation via ATP – dependent efflux and influx of sodium and potassium ions in various cells [5]. The pump is isoenzymic in nature and has its α1 catalytic subunit encoded by the ATP1A1 gene present in both erythrocyte and nerve cells [5, 6]. Na⁺K⁺-ATPase has also been reported as one of the membrane proteins affected structurally and functionally in diabetes mellitus [5].

Numerous experimental studies conducted on streptozotocin-induced diabetic rats show that for instance, the delayed motor nerve conduction velocity, decreased nerve blood flow, degeneration of peripheral nerves, persistent hyperglycemia and other metabolic aberrations involve a common denominator of down-regulated Na⁺K⁺-ATPase [4, 7, 8]. Reduced activity of Na⁺K⁺-ATPase has also been implicated in streptozotocin-induced diabetic rats with nephropathy [9]. De Leo et al [10] and Kowluru [11] also reported impairment in the level of this enzyme in diabetic rats and mice with retinopathy. However, in most of these studies, the observed Na⁺K⁺-ATPase reduction was found with other metabolic derangements which include myoinositol depletion, aldolase reduction, sorbitol accumulation and protein kinase C activity [7, 12]. Subsequent associations of Na⁺K⁺-ATPase with these parameters have been unfolded [8, 12]. Also in humans, several studies have consistently reported a declined Na⁺K⁺-ATPase activity from the membranes of nerve cells and erythrocytes preparations of Type 1 and Type 2 diabetic patients coupled with alterations in membrane protein composition [13–15]. Among diabetic Nigerian patients, defectiveness in Ca²⁺ homeostasis and qualitative and quantitative changes in erythrocyte membrane proteins have also been reported [16, 17]. The work of Okegbile et al [18] also revealed a reduction in erythrocyte Na⁺K⁺-ATPase activity in Nigerian patients with Type 2 diabetes mellitus similar to the finding of DeLuise et al [19] in obese patients. The latter is characterized by dyslipidemia caused by an alteration in the regulations of cholesterol, triglycerides, high density and low-density lipoproteins [20]. Abnormal levels of these lipids in the blood have subsequently been observed in diabetics with or susceptible to cardiovascular complications and microangiopathy [2].

However, research is still needed to enhance our present understanding about factors responsible for the decreased Na⁺K⁺-ATPase observed in diabetes mellitus especially in Type 1 diabetes mellitus, which occurs early in life and requires life-long insulin therapy [1]. Furthermore, this group of diabetics is also susceptible to cardiovascular complications of diabetes mellitus [1–4] and in our recent study we observed microalbuminuria of comparable prevalence rates in Type 1 and Type 2 diabetic cohorts and in whom elevated blood pressure was also found [21].

Understanding the pattern and magnitude of association between Na⁺K⁺-ATPase activity and cardiovascular risk factors of lipid metabolism would provide a leap forward in our understanding of how diabetes mellitus progresses to cardiovascular complications in patients. Clues to this possibility have emanated from few studies. They include the observation by Rabini et al [22] that elevated plasma lysophosphate choline (LPC) due to decreased lecithin-cholesterol acteyltransferase (LCAT) associate strongly with reduced Na⁺K⁺-ATPase activity in diabetic patients. Kiziltunc et al [23] also reported cholesterol as an inhibitor of erythrocyte membrane Na⁺K⁺-ATPase in vitro. Inverse correlation between erythrocyte membrane Na⁺K⁺-ATPase activity and polyunsaturated fatty acid levels has also been reported by Djemli-Shipkolye et al, [8]. Taken together, there is currently a paucity data on the relationship of Na⁺K⁺-ATPase with cardiovascular risk factors such as total cholesterol, triglycerides and low density lipoprotein cholesterol as well as high density lipoprotein cholesterol. These lipid metabolites are more synonymous with cardiovascular complications than LPC in diabetic patients from Nigeria and other parts of the world 21, 24, 25]. Therefore, knowledge of the metabolic effects of these lipid metabolites on Na⁺K⁺-ATPase activity is essential for a better understanding of the pathogenesis of diabetes mellitus and evolution of strategies that would improve the management and prognosis of the disease.

In the present study, erythrocyte ghost membrane preparations from Nigerian patients with Type 1 diabetes mellitus were evaluated for protein content and Na⁺K⁺-ATPase activity in association with duration of diabetes and their plasma levels of total cholesterol, triglycerides, LDL-cholesterol and HDL-C. The kinetics of the enzyme based on reaction with ATP in vitro was also investigated.

The kinetic profile of erythrocyte Na⁺K⁺-ATPase in the ghost membrane preparations based on reciprocal transformation of data in the Lineweaver-Burk plot is presented in Figure 1. The Na⁺K⁺-ATPase activity with ATP as a substrate exhibited V_max of 39.2, 26.3, 16.7 and 16.1 micromole Pi/mg/h in control, good glycemic control, poor glycemic and all the diabetic patients respectively. Their respective K_m values were 0.454, 0.256, 0.167 and 0.154 mM.

Correlation analyses revealed significant (P < 0.05) inverse relationship between FBS (r = -0.784), TC (r = -0.703) and Na⁺K⁺-ATPase activity in all the diabetic patients. In type 1 diabetic patients with poor glycemic control, an additional inverse significant relationship between TAG (r = -0.797), LDL-C (r = -0.707) and membrane enzyme activity was found. These associations were absent in type 1 diabetic patients with good glycemic control.

Studies conducted to date in humans and animals have consistently shown that diabetes mellitus induces a reduction in erythrocyte membrane Na⁺K⁺-ATPase activity, which results in heamodynamic dysfunction due to altered microvascular blood flow, rheological abnormalities precipitated by decrease erythrocyte defomability and raised fluidity and complications such as nephropathy, neuropathy, cardiovascular disorders and microangiopathy [36, 37, 1]. The present study has also demonstrated decreased erythrocyte membrane Na⁺K⁺-ATPase activity in Nigerian patients with type 1 diabetes mellitus. Similar findings have been reported in previous studies from other parts of the country involving both type 1 and type 2 diabetic patients [16–18]. However, in these studies the role of glycemic control and metabolic influence of cardiovascular risk factors such as blood pressure on the membrane enzyme activity were not investigated. Meanwhile, studies conducted in other countries have observed decreased enzyme activity in diabetic patients in association with microalbuminuria, hyperkalaemia and elevated serum sialic acid [38, 39, 14]. In Nigeria, some of these complications have also been found but independently among diabetic patients [21, 40]. In the present study, we observed the decreased erythrocyte membrane Na⁺K⁺-ATPase activity to have a significant negative correlation with fasting blood glucose (FBS) and plasma total cholesterol irrespective of the level of glycemic control in all our diabetic patients. Our observation with respect to FBS is supportive of previous findings in which glucose in diabetic state exhibited direct toxicity on erythrocyte membrane protein including Na⁺K⁺-ATPase through non-enzymatic glycosylation [41, 42]. Although, the level of glycosylated hemoglobin (HbA1c) was not determined in our diabetic cohort, the greater FBS-Na⁺K⁺-ATPase association observed in our patients with poor glycemic control is suggestive of advanced glycation of the enzyme since this non-enzymatic process has been associated with greater reduction in Na⁺K⁺-ATPase activity and other cellular signaling proteins such as protein kinase C owing to their proneness to glycability [43]. The greater reduction in Na+K+ATPase observed among our type 1 diabetic subgroup with poor glycemic control may also be due to greater susceptibility of this protein to oxidative stress [44]. The fact that glycation can be minimized through glycemic control as demonstrated in previous studies [9], there is huge possibility of reducing the magnitude of Na⁺K⁺-ATPase inevitable reduction and subsequent complications in diabetic patients in order to reduce morbidity and mortality rates, currently on the increase in Nigeria [45]. The poor glycemic control observed in 20 of the 34 patients investigated may not be unconnected with poor compliance to insulin therapy, the use of which is determined in part by socioeconomic factors which include poverty, access to treatment and quality of care in most developing countries [21]. Our observation of the plasma total cholesterol as the only cardiovascular risk factor to correlate negatively with Na⁺K⁺-ATPase activity in our diabetic cohort irrespective of glycemic control is indicative of an exceeding clinical importance of this lipid metabolite in type 1 diabetic patients. Earlier studies in this regard are mostly in vitro in which decreased erythrocyte membrane cholesterol content and cholesterol:phospholipid ratio were found in erythrocyte ghost preparations from diabetic patients [8, 46]. Since cholesterol modulates erythrocyte membrane fluidity, its loss into systemic circulation may therefore contribute to the increased fluidity observed in diabetic patients [37]. However, there is a strong indication from this study and previous ones that poor glycemic control may exacerbate the reported cholesterol loss, which when added to the contributions from diet and endogenous synthesis may increase the risk of hypercholesterolemia in diabetes and this we have found only in our IDDM patients with poor glycemic control. The same explanation can be adduced for the incidence of hypertriglyceridemia and elevated LDL-cholesterol in this group of type 1 diabetic patients since they also exhibited significant Na⁺K⁺-ATPase – TAG/LDL-C association not found in their good glycemic control counterparts. The observed significant positive relationship between enzyme activity and plasma HDL-cholesterol in normal and type 1 diabetic patients with good glycemic control validates not only the anti-atherogenic function of this lipid metabolites but also implicates its presence as a protective factor against cardiovascular complications in this subgroup. It raises the possibility of using glycemic control as a prerequisite for achieving lipid control in type 1 diabetic patients as recently reported by Erciyas et al [47]. Our observed lack of association between duration of diabetes, membrane protein and Na⁺K⁺-ATPase activity is in agreement with previous findings [3]. However, Adewoye et al [17] had previously reported changes in erythrocyte membrane protein composition in diabetic Nigerians to involve loss of spectrin and preponderance of ankyrin. The erythrocyte membrane protein content of 4.3 – 5.1 μgmL in diabetics and 5.5 μg/mL in healthy humans reported by these workers are comparable to our findings and thus validate their results. In this study, our non-diabetic controls were matched for age with the diabetic groups to minimize the impact of age on Na⁺K⁺-ATPase activity and validate of the results obtained. Also, the effect of hematological abnormalities on Na⁺K⁺-ATPase activity [48] was annulled by inclusion of subjects with hematocrit level ≥ 33% and hemoglobin level of 11 g/dL and above.

We also observed reduced Vmax for Na⁺K⁺-ATPase from diabetic erythrocyte ghost membrane preparations compared to the control using ATP as the substrate. Also, greater reduction in the kinetic constant was observed in poor glycemic control patients. Our data are in concordance with the finding of Okegbile et al [18] in type 2 diabetic patients. Reduced Vmax connotes a reduction in the enzyme ability to hydrolyze ATP in vitro, which undoubtedly has physiological implications in the regulation of transmembrane cation transport in the erythrocytes. While diabetes may be a factor for this event, numerous secondary factors have been reported. They include decreased number of pump units on the erythrocyte membrane, altered lipid – protein interaction, depleted membrane anionic charge and enzyme glycation and peroxidation [9, 10, 20, 42]. We also observed lower Km values for the membrane enzyme in our diabetic patients compared to the control and this suggest increased affinity for ATP in vitro. This observation is contrary to what Okegbile et al [18] observed in type 2 diabetic patients. However, Okegbile et al [18] conducted their kinetic assay in the presence of digoxin and this may explain the difference in our observations. In this study, assays conducted in the presence of digoxin were carried out to compare the sensitivity of Na⁺K⁺-ATPase from both diabetic and non-diabetic subjects to this digitalis. We found the diabetic enzyme to exhibit an increased sensitivity ranging from 74.3 – 75.6% compared to 69.5% in the non-diabetic control. This indicates a greater inhibition of diabetic Na⁺K⁺-ATPase enzyme by digoxin similar to observations with Ouabain [49].

Based on the results of this study, we conclude that the reduced erythrocyte Na⁺K⁺-ATPase activity which occurs in type 1 diabetes mellitus is exacerbated in Nigerian patients with poor glycemic control in association with atherogenic blood lipids.