Abstract

Background. Recently, an amino acid variant (K121Q) in the glycoprotein PC‐1 (Q allele) has been associated with faster progression of diabetic nephropathy, as estimated by calculated creatinine clearance. We tested the impact of the PC‐1 (K121Q) variant on loss of glomerular filtration rate (GFR) measured by the [51Cr]EDTA plasma clearance technique.

Methods. We performed an observational follow‐up study of 295 (182 males) type 1 patients with diabetic nephropathy [mean age 37 (SE 0.7) years, mean duration of diabetes 23 (SE 0.5) years]. All patients were followed for at least 3 years, median 8 years (range 3–17), with at least three measurements of GFR using [51Cr]EDTA (median 11 measurements (range 3–32)). Two hundred and seventeen patients had the KK genotype and 78 carried the Q allele (71 KQ and 7 QQ).

Results. Patients carrying the Q allele had a mean rate of decline in GFR during follow‐up of 3.6 ml/min per year (SE 0.4) compared with 4.0 ml/min per year (SE 0.3) in patients with the KK genotype. Other risk factors for progression of diabetic nephropathy were similar in Q carriers and KK carriers. When dividing patients in tertiles based on rate of decline in GFR, we found no difference in distribution of K121Q genotypes. No difference in the number of patients who died or reached end‐stage renal disease during follow‐up according to K121Q genotype were found. A multiple linear regression analysis revealed that albuminuria, mean arterial blood pressure, haemoglobin A1C and serum cholesterol during follow‐up predicted a steeper decline in GFR [R2 (adjusted)=0.27], whereas the PC‐1 genotype did not contribute.

Conclusions. Our study did not reveal an association between the PC‐1 amino acid variant K121Q and progression of diabetic nephropathy.

Introduction

Diabetic nephropathy is a leading cause of end‐stage renal failure, but interindividual loss of kidney function is highly variable. Several factors, including genetic predisposition, metabolic and haemodynamic alterations and various growth factors, have been considered to contribute to the initiation and progression of diabetic nephropathy as reviewed by Parving et al. [1].

Type 1 patients with elevated urinary albumin excretion are insulin resistant [2,3] and this has been suggested to influence the vascular damage occurring in diabetic nephropathy [4,5]. The human plasma–cell membrane differentiation antigen (PC‐1) may inhibit insulin receptor signalling [69] and is involved in insulin resistance [79]. Functional and population‐based studies suggest that an amino acid variant (K121Q) in the glycoprotein PC‐1 is associated with insulin resistance [1012], although this has not been confirmed in a Danish population study [13].

Recently, this genetic variant (Q allele) has been associated with faster progression of diabetic nephropathy, as estimated by calculated creatinine clearance [4]. Since this method is an indirect measure of loss of kidney function, we tested the impact of the PC‐1 (K121Q) variant on loss of glomerular filtration rate (GFR) measured by the [51Cr]EDTA plasma clearance technique.

Subjects and methods

Since 1983, all adult Caucasian type I diabetic patients with diabetic nephropathy attending the Steno Diabetes Center have had their GFR measured routinely approximately once a year starting from onset of diabetic nephropathy (n=584). From 1993, all patients who suffered from diabetic nephropathy were invited to participate in a study of genetic risk factors for the development and progression of diabetic nephropathy. Seventy‐three per cent (423) accepted, of whom 199 had blood drawn for DNA analysis in 1993 and 224 thereafter at their first GFR measurement following diagnosis of diabetic nephropathy. In the present observational follow‐up study, we included all albuminuric patients who had a follow‐up period of at least 3 years, with at least three measurements of GFR (n=297). These patients did not differ from the rest of the patients (n=126) with regard to gender, GFR, mean arterial blood pressure, albuminuria and haemoglobin A1C. The follow‐up period ended on 1 July 2000.

Diabetic nephropathy was diagnosed clinically if the following criteria were fulfilled: persistent albuminuria >200 μg/min in two out of three consecutive determinations; presence of diabetic retinopathy; and no clinical or laboratory evidence of other kidney or renal tract disease, as described previously [14]. End‐stage renal disease (ESRD) was defined as dialysis or renal transplantation.

All patients had been dependent on insulin treatment from the time of diagnosis and received at least two daily injections of insulin. Patients were on a diabetic diet (45–55% carbohydrates, 30–35% fat, 15–20% protein) without restriction of sodium or protein intake. The local ethical committee approved the experimental design and all patients gave their informed consent.

Methods

GFR and other physiological investigations were carried out a median of 11 (range 3–32) times, during 8 (range 3–17) years of follow‐up. Between the investigations patients had their dose of insulin and antihypertensive treatment adjusted every 2–4 months in the outpatient clinic. The GFR was measured by applying a single intravenous injection of [51Cr]EDTA (3.7 MBq) at 08:30, and thereafter following the plasma clearance of the tracer for 4 h [15]. The results were standardized for 1.73 m2 body surface area, using the patient surface area at the initial GFR measurement. The mean day‐to‐day coefficient of variation in GFR in individuals is 0.04 in our laboratory. Urinary albumin excretion rate was determined by an enzyme immunoassay (ELISA) during the 4 h clearance period [16], before 1991 radio immunoassay (RIA) was used [17]. A very close correlation between RIA and ELISA (r=0.99) was documented before changing the method (linear regression equation: RIA=1.06×ELISA) [16].

Arterial blood pressure was measured with a standard clinical mercury sphygmomanometer, at least three times during each investigation after the patient had been supine for >10 min. Diastolic blood pressure was recorded as the disappearance of the Korotkoff sounds (phase V). Arterial hypertension was diagnosed according to World Health Organization (WHO) criteria until 1995, and antihypertensive medication prescribed if at least three consecutive recordings revealed a systolic blood pressure >160 mmHg and/or a diastolic blood pressure >95 mmHg. After 1995 the American Association of Diabetes criteria of hypertension (blood pressure >140/90 mmHg) was used.

From venous blood samples, haemoglobin A1C (normal range 4.3–6.2%) was determined by high‐performance liquid chromatography (DIAMAT Analyzer; BioRad, Richmond, CA, USA) at each investigation. Serum cholesterol concentration was measured enzymatically using CHOD‐PAP reagents from Boehringer‐Mannheim GmbH (Mannheim, Germany).

Genotyping was carried out on genomic DNA isolated from human leukocytes. The genotype was determined by using AvaII restriction enzyme digestion of polymerase chain reaction (PCR) products obtained with the forward primer 5′‐ctgtgttcactttggacatgttg‐3′ and reverse primer 5′‐gacgttggaagataccaggttg‐3′ (0.7 U of Ampli Taq Gold DNA polymerase, 2.5 mmol/l of MgCl2, 5 pmol of each primer, 1×PCR Buffer II, and an annealing temperature of 65°C). Digested PCR products (KK genotype: 238 bp; heterozygous: 238, 148 and 90 bp; homozygous for the Q allele: 148 and 90 bp) were detected by agarose gel electrophoresis [13].

Statistical analysis

Values are given as means (SE) except for albuminuria, which is expressed as geometric means with 95% CI owing to the skewed distribution. Linear regression analysis (least squares method) was used to determine the rate of decline in GFR for each patient, using all measured values. Clinical and laboratory characteristics at baseline are based on values from the first GFR investigation for each patient. Follow‐up data are based on all values during follow‐up. Normally distributed variables, including logarithmically transformed values of albuminuria, were compared with an unpaired Student's t‐test. Non‐normally distributed variables were compared with a Mann–Whitney test. Chi‐square tests were made to compare proportions of patients.

A multiple regression analysis, with rate of decline in GFR as the dependent variable and mean values of albuminuria, haemoglobin A1C, mean arterial blood pressure and serum cholesterol during the whole observation period in addition to PC‐1 genotype as independent variables, was performed with backward selection. The R2 value is adjusted for the number of variables introduced into the model. A P‐value (two‐sided) of <0.05 was considered significant. All calculations were performed using SPSS (SPSS, Chicago, IL, USA).

Results

PC‐1 genotypes were available in 295 of 297 patients. Two hundred and seventeen (74%) patients had the KK genotype and 78 carried the Q allele (71 (24%) KQ and 7 (2%) QQ). The frequency of the Q allele was 0.14 and the distribution of genotypes were in Hardy–Weinberg equilibrium.

At baseline, patients carrying the Q allele were comparable to patients with the KK genotype (Table 1). Eighty‐two per cent of the patients received antihypertensive treatment for the majority of the follow‐up. Sixty per cent received an angiotensin‐converting enzyme inhibitor (ACEi) alone (10%) or in combination with diuretics, beta‐blocker or calcium‐channel blocker (50%), and 22% were treated with non‐ACEi antihypertensive therapy. There were no differences in the agents according to PC‐1 genotype.

The mean (SE) rate of decline in GFR during 8 years of follow‐up in all patients was 3.9 (0.2) ml/min per year. We found no difference in the progression of diabetic nephropathy during follow‐up according to K121Q genotype (Table 2). When dividing patients in tertiles based on rate of decline in GFR, we found 23 patients with the Q allele in the lowest tertile, 26 in the middle and 29 Q‐carriers in the highest tertiles (not significant, NS).

Thirty‐one patients reached ESRD during follow‐up. The frequencies of ESRD were not influenced by K121Q genotype (Table 2).

During follow‐up, 22 patients with the KK genotype died compared with 13 carrying the Q allele (NS). Unfortunately we could not determine the causes of death.

To further avoid influence of selective dropout we tested the impact of the PC‐1 gene in patients who had duration of nephropathy <3 years both at the time of their first GFR measurement and at time of genotyping (n=77). This subgroup analysis also showed a similar rate of decline in GFR in patients carrying the Q allele (n=18) compared with patients with the KK genotype (n=59) (NS).

A multiple linear regression analysis revealed that albuminuria (P<0.001), mean arterial blood pressure (P<0.001), haemoglobin A1C (P<0.01) and serum cholesterol (P=0.02) during follow‐up predicted a steeper decline in GFR [R2 (adjusted)=0.26], whereas PC‐1 genotype did not contribute.

Table 1. 

Baseline clinical and laboratory data in 295 type 1 diabetic patients with diabetic nephropathy according to PC‐1 (K121Q) variant


 
KK genotype
 
Q carriers
 
P‐value
 
Sex (men/women) 131/86 51/27 NS 
Age (years) 38 (1) 36 (1) NS 
Duration of diabetes (years) 23 (1) 24 (1) NS 
Duration of nephropathy (years) 4.0 (0.2) 4.3 (0.4) NS 
BMI (kg/m224.0 (0.2) 23.7 (0.3) NS 
GFR (ml/min/1.73 m291 (2) 89 (3) NS 
Blood pressure (mmHg) 140 (1)/84 (1) 140 (2)/86 (1) NS 
Albuminuria (μg/min)a 559 (489–640) 631 (510–780) NS 
Haemoglobin A1C (%) 9.3 (0.1) 9.1 (0.2) NS 
Serum cholesterol (mmol/l) 5.6 (0.1) 5.7 (0.2) NS 

 
KK genotype
 
Q carriers
 
P‐value
 
Sex (men/women) 131/86 51/27 NS 
Age (years) 38 (1) 36 (1) NS 
Duration of diabetes (years) 23 (1) 24 (1) NS 
Duration of nephropathy (years) 4.0 (0.2) 4.3 (0.4) NS 
BMI (kg/m224.0 (0.2) 23.7 (0.3) NS 
GFR (ml/min/1.73 m291 (2) 89 (3) NS 
Blood pressure (mmHg) 140 (1)/84 (1) 140 (2)/86 (1) NS 
Albuminuria (μg/min)a 559 (489–640) 631 (510–780) NS 
Haemoglobin A1C (%) 9.3 (0.1) 9.1 (0.2) NS 
Serum cholesterol (mmol/l) 5.6 (0.1) 5.7 (0.2) NS 

Values are means (SE).

aGeometric mean with 95% CI.

Table 2. 

Effect of PC‐1 (K121Q) variant on selected clinical and biochemical variables in 295 type 1 diabetic patients; based on all values during 8 years of follow‐up


 
KK genotype
 
Q carriers
 
Mean difference (95% CI)
 
P‐value
 
Follow‐up (years)a 7 (3–17) 8 (3–17)  NS 
Decline in GFR (ml/min per year) 4.0 (0.3) 3.6 (0.4) 0.4 (−0.6 to 1.4) NS 
End‐stage renal disease (%) 11 2 (−6 to 11) NS 
Death (%) 10 17 −7 (−17 to 3) NS 
Blood pressure (mmHg) 139 (1)/80 (1) 139 (1)/81(1) 0 (−4 to 4)/−1 (−3 to 1) NS 
Albuminuria (μg/min)b 437 (368–518) 541 (437–671) −24% (−70 to 9) NS 
Haemoglobin A1C (%) 9.2 (0.1) 9.2 (0.1) 0.0 (−0.2 to 0.3) NS 
Serum cholesterol (mmol/l) 5.5 (0.1) 5.5 (0.1) 0.0 (−0.3 to 0.3) NS 
Antihypertensive therapy (%) 82 82 0 (−10 to 10) NS 

 
KK genotype
 
Q carriers
 
Mean difference (95% CI)
 
P‐value
 
Follow‐up (years)a 7 (3–17) 8 (3–17)  NS 
Decline in GFR (ml/min per year) 4.0 (0.3) 3.6 (0.4) 0.4 (−0.6 to 1.4) NS 
End‐stage renal disease (%) 11 2 (−6 to 11) NS 
Death (%) 10 17 −7 (−17 to 3) NS 
Blood pressure (mmHg) 139 (1)/80 (1) 139 (1)/81(1) 0 (−4 to 4)/−1 (−3 to 1) NS 
Albuminuria (μg/min)b 437 (368–518) 541 (437–671) −24% (−70 to 9) NS 
Haemoglobin A1C (%) 9.2 (0.1) 9.2 (0.1) 0.0 (−0.2 to 0.3) NS 
Serum cholesterol (mmol/l) 5.5 (0.1) 5.5 (0.1) 0.0 (−0.3 to 0.3) NS 
Antihypertensive therapy (%) 82 82 0 (−10 to 10) NS 

Values are means (SE).

aMedian (range). bGeometric mean with 95% CI.

Discussion

Our 8‐year observational follow‐up study did not reveal an association between the PC‐1 amino acid variant (K121Q) and progression of diabetic nephropathy. The allele frequencies in our study were similar to previous reports [4,11,13].

An impact of the Q allele in PC‐1 amino acid variant K121Q on progression of diabetic nephropathy has been reported by other authors. In a retrospective study of 77 type 1 patients from Italy and the UK (6.5 years of follow‐up), the rate of decline in kidney function was estimated by creatinine clearance derived from the Cockcroft–Gault formula [4]. Q carriers had a rate of decline in GFR of 7.2 ml/min per year compared with 3.7 ml/min per year in patients with KK genotype, and significantly more patients carrying the Q allele belonged to the highest tertile of GFR decline [4]. However, direct measures of GFR are the gold standard when assessing kidney function. To obtain a valid determination of the rate of decline in GFR in patients with chronic, progressive kidney disease, the method should have good accuracy and precision, repeated GFR determination should be performed, and the observation period should be extended to at least 2 years as discussed by Levey et al. [18]. These requirements were fulfilled and extended in our observational follow‐up study of a large, well defined cohort of type 1 diabetic patients with diabetic nephropathy. Our observed 95% confidence interval of the difference in the rate of decline in GFR between patients with different K121Q genotypes indicates that the gene has no clinical importance in progression of diabetic nephropathy.

A difference in the genetic background of patients from the present study compared with patients from the study from Italy and the UK [4] may contribute to the conflicting results. In support of this, the reported impact of the Q allele on insulin resistance in Caucasians from Italy has not been confirmed in the Danish population [13], although a Swedish family‐based study [12] and a functional study [10] found an association. On the other hand, as mentioned above, the Q allele frequency is similar in our population to that in the population from Italy and the UK. A different genetic composition in populations could cause diverse interaction between other genes and the amino acid variant (K121Q), but potential mechanisms need further investigation.

Only recently, a family‐based associations study (abstract ADA 2001) reported an increased prevalence of Q carriers in patients with ESRD after short duration of diabetes compared with patients with normoalbuminuria and patients with ESRD after longer duration of diabetes [19]. Unfortunately, details are not yet published. We found no impact of the PC‐1 genotype on development of ESRD, although our study was not designed to evaluate this end point.

The impact of the PC‐1 gene on development of diabetic nephropathy has been addressed in one study. We recently found no differences in the distributions of K121Q genotypes in patients with normoaluminuria compared with patients with macroalbuminuria [20].

A limitation of our study is that we identified and genotyped patients after the onset of diabetic nephropathy. The median survival times in these patients are now exceeding 15–17 years after onset on diabetic nephropathy [14]; however, selective drop out is possible. A preliminary report from Kusuda et al. [21] (abstract ADA 2001) suggests an increased frequency of the Q allele in Japanese non‐diabetic and type 2 diabetic patients with myocardial infarction. This could lead to higher mortality in Q carriers. We did not find any difference in death rate during follow‐up according to K121Q genotype, although numbers were small. Furthermore, since the allele frequencies in our study are similar to reports from normal subject and type 2 patients [11,13], and since all known deleterious risk factors were similar among patients with different K121Q genotypes, the effect of selection may be of less importance. In agreement with this, the subgroup analysis of patients identified and genotyped <3 years after the onset of diabetic nephropathy confirmed the findings in all participating patients.

Our study revealed that elevated blood pressure, albuminuria, haemoglobin A1C and serum cholesterol are independent risk factors for excessive loss of GFR. We thereby confirm what has previously been shown in a subset of patients from the present study (n=231) [22] and in other reports, as reviewed by Rossing [14]. This supports the evidence of a beneficial effect of antihypertensive treatment [14], since our patients had a rate of decline in GFR of 4 ml/min per year in a group where 82% received antihypertensive treatment and the remaining patients were spontaneously normotensive. In comparison, the original studies of diabetic nephropathy carried out before antihypertensive treatment was established revealed a loss of GFR of 12 ml/min per year, as reviewed by Rossing [14].

In conclusion, when applying direct measures of GFR, we could not find an association between the PC‐1 amino acid variant K121Q and progression of diabetic nephropathy in a large, well defined cohort of Danish type 1 diabetic patients.

Correspondence and offprint requests to: Peter Jacobsen, Steno Diabetes Center, Niels Steensens Vej 2, DK‐2820 Gentofte, Denmark. Email: pkjacobsen@dadlnet.dk

The authors wish to thank Birgitte V. Hansen, Tina R. Juhl, Berit R. Jensen, Ulla M. Smidt and Inge‐Lise Rossing for technical assistance. This study was supported by the P Carl Petersens Foundation (Copenhagen), The Danish Medical Association Research Fund, The Danish Medical Research Council, The Velux Foundation and the Danish Diabetes Association.

References

1
Parving H‐H, Østerby R, Ritz E. Diabetic nephropathy. In: Brenner BM, Levine S, eds.
The Kidney
 . WB Saunders, PA
2000
:
1731
–1773
2
De Cosmo S, Bacci S, Piras GP et al. High prevalence of risk factors for cardiovascular disease in parents of IDDM patients with albuminuria.
Diabetologia
 
1997
;
40
:
1191
–1196
3
Yip J, Mattock MB, Morocutti A et al. Insulin resistance in insulin‐dependent diabetic patients with microalbuminuria.
Lancet
 
1993
;
342
:
883
–887
4
De Cosmo S, Argiolas A, Miscio G et al. A PC‐1 amino acid variant (K121Q) is associated with faster progression of renal disease in patients with type 1 diabetes and albuminuria.
Diabetes
 
2000
;
49
:
521
–524
5
Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease.
Diabetes
 
1988
;
37
:
1595
–1607
6
Maddux BA, Sbraccia P, Kumakura S et al. Membrane glycoprotein PC‐1 and insulin resistance in non‐insulin‐dependent diabetes mellitus.
Nature
 
1995
;
373
:
448
–451
7
Frittitta L, Youngren J, Vigneri R et al. PC‐1 content in skeletal muscle of non‐obese, non‐diabetic subjects: relationship to insulin receptor tyrosine kinase and whole body insulin sensitivity.
Diabetologia
 
1996
;
39
:
1190
–1195
8
Frittitta L, Youngren JF, Sbraccia P et al. Increased adipose tissue PC‐1 protein content, but not tumour necrosis factor‐α gene expression, is associated with a reduction of both whole body insulin sensitivity and insulin receptor tyrosine‐kinase activity.
Diabetologia
 
1997
;
40
:
282
–289
9
Frittitta L, Spampinato D, Solini A et al. Elevated PC‐1 content in cultured skin fibroblasts correlates with decreased in vivo and in vitro insulin action in nondiabetic subjects: evidence that PC‐1 may be an intrinsic factor in impaired insulin receptor signaling.
Diabetes
 
1998
;
47
:
1095
–1100
10
Costanzo BV, Trischitta V, Di Paola R et al. The Q allele variant (GLN121) of membrane glycoprotein PC‐1 interacts with the insulin receptor and inhibits insulin signaling more effectively than the common K allele variant (LYS121).
Diabetes
 
2001
;
50
:
831
–836
11
Pizzuti A, Frittitta L, Argiolas A et al. A polymorphism (K121Q) of the human glycoprotein PC‐1 gene coding region is strongly associated with insulin resistance.
Diabetes
 
1999
;
48
:
1881
–1884
12
Gu HF, Almgren P, Lindholm E et al. Association between the human glycoprotein PC‐1 gene and elevated glucose and insulin levels in a paired‐sibling analysis.
Diabetes
 
2000
;
49
:
1601
–1603
13
Rasmussen SK, Urhammer SA, Pizzuti A et al. The K121Q variant of the human PC‐1 gene is not associated with insulin resistance or type 2 diabetes among Danish Caucasians.
Diabetes
 
2000
;
49
:
1608
–1611
14
Rossing P. Promotion, prediction and prevention of progression of nephropathy in type 1 diabetes mellitus.
Diabet Med
 
1998
;
15
:
900
–919
15
Bröchner‐Mortensen J, Rödbro P. Selection of routine method for determination of glomerular filtration rate in adult patients.
Scand J Clin Lab Invest
 
1976
;
36
:
35
–45
16
Feldt‐Rasmussen B, Dinesen B, Deckert M. Enzyme immunoassay: an improved determination of urinary albumin in diabetics with incipient nephropathy.
Scand J Clin Lab Invest
 
1985
;
45
:
539
–544
17
Miles DW, Mogensen CE, Gundersen HJG. Radioimmunoassay for urinary albumin using a single antibody.
Scand J Clin Lab Invest
 
1970
;
26
:
5
–11
18
Levey AS, Gassman J, Hall PM, Walker WG. Assessing the progression of renal disease in clinical studies: effects of duration of follow‐up and regression to the mean.
J Am Soc Nephrol
 
1991
;
1
:
1087
–1094
19
Canani LH, Warram JH, Krolewski AA. Plasma cell differentiation antigen (PC‐1) K121Q polymophism and the development of end‐stage renal disease (EASD) in type 1 diabetes.
Diabetes
 
2001
;
50 [Suppl 2]
:
A173
(Abstract)
20
Tarnow L, Grarup N, Hansen T, Parving HH, Pedersen O. Diabetic microvascular complications are not associated with two polymorphisms in the GLUT‐1 and PC‐1 genes regulating glucose metabolism in Caucasian type 1 diabetic patients.
Nephrol Dial Transplant
 
2001
;
16
:
1653
–1656
21
Kusuda Y, Hamaguchi K, Yamashita T, Hara M, Sakata T. Implication of the membrane glycoprotein (PC‐1) gene polymorphism (K121Q) in type 2 diabetes and coronary artery disease in a Japanese population.
Diabetes
 
2001
;
50 [Suppl 2]
:
A492
(Abstract)
22
Hovind P, Rossing P, Tarnow L, Smidt UM, Parving HH. Progression of diabetic nephropathy.
Kidney Int
 
2001
;
59
:
702
–709

Comments

0 Comments