 |
||||
| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
REGULAR ARTICLES |
*
Department of Endocrinology and Metabolic Disease, University of Pisa,
Italy
Department of Epidemiology and Public Health, University College London,
London, United Kingdom
Department of Internal Medicine, University of Roma
"Tor-Vergata," Rome, Italy
§
La Spezia Hospital, La Spezia, Italy
||
Department of Medicine, University of Milan, San Raffaele Scientific
Institute, Milan, Italy.
Correspondence to Dr. Monica Nannipieri, Dipartimento di Medicina Interna Università degli Studi di Pisa ed Unità Metabolica dell'Istituto di Fisiologia Clinica del C.N.R., Via Savi 8, 56100 Pisa, Italy. Phone: +39 50 583230; Fax: +39 50 553235; E-mail: nannipi{at}nsifc.ifc.pi.cnr.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both of these observations and the elevated concordance in renal destiny among siblings with type 1 diabetes (4,5) are consistent with the hypothesis that genetic factors might influence the development of nephropathy (6).
The pathogenesis of diabetic nephropathy remains largely unknown. Nonetheless, several genes have recently been proposed to explain the susceptibility to this complication (7,8,9). In particular, because a predisposition to hypertension may reflect a risk factor for kidney disease in patients with type 1 diabetes (10,11,12), the genes involved in the control of BP have been more extensively studied (13). However, the findings are controversial and the corresponding relative risk is very low (14,15,16,17,18,19,20,21).
Atrial natriuretic peptide (ANP) plays a central role in the regulation of BP, sodium homeostasis, and vascular permeability (22,23,24), and significant associations between DNA polymorphisms at the pronatriodilatin (PND) locus and hypertension have been reported (25,26,27,28,29,30,31,32,33,34,35). Moreover, ANP could be involved in the regulation of GFR (36,37,38), and hyperfiltration seems to participate in the pathogenesis of glomerular injury. Serum ANP levels have also been reported to be elevated in diabetic patients with micro- and macroalbuminuria (39,40), and salt sensitivity may be a characteristic of these patients (41,42). These observations suggest that the PND gene may be involved in the genetic susceptibility to diabetic nephropathy. Therefore, we examined the relationship between a diallelic polymorphism recently identified in the PND gene with microalbuminuria and overt nephropathy in an ethnically homogeneous cohort of patients with type 1 diabetes. In the same cohort of diabetic patients, additional genotypings have been performed, using known polymorphisms of the PND gene. Finally, in a subset of normotensive patients with type 1 diabetes, we also investigated the relationship of these genetic variants with plasma ANP levels and microvascular permeability to macromolecules.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An age-matched group of 58 nondiabetic healthy individuals (19 men, 39 women) of Caucasian origin was also considered for control purposes. Both cohorts enrolled in the study were carefully recruited to avoid selection bias and geographic differences that may have introduced undue unrecognized stratification in our populations. Using the EURODIAB (43) Family Study questionnaire, we obtained information about geographic origin of the families (at least three generations) of each subject. Only subjects with family origin in Tuscany were recruited for the study. The study protocol was approved by the local ethics committee, and all participants gave informed consent for genetic studies.
Clinical Variables
The clinical variables for the assessment of renal disease were urinary
albumin excretion, plasma creatinine, and treatment for renal failure. Urinary
albumin excretion rate was measured in at least three 24-h urine collections
obtained over a period of 6 mo, after excluding urinary tract infection. Mean
urinary albumin excretion rates <20 µg/min defined normoalbuminuria
(NA). A diagnosis of microalbuminuria (µA) was based on the persistence of
albumin excretion rates between 20 and 200 µg/min. Finally, the clinical
diagnosis of overt nephropathy was based on the coexistence of persistent
macroalbuminuria (MA; AER 
200 µg/min) and either background or
proliferative retinopathy, in the absence of clinical or laboratory evidence
of other causes of increased urinary albumin excretion. Patients with
persistent proteinuria within 5 yr of the onset of type 1 diabetes were not
included in the study. About 10% of normoalbuminuric patients were
hypertensive, and 80% of them were treated with antihypertensive agents
(angiotensin-converting enzyme [ACE] inhibitors or angiotensin II receptor
antagonists in monotherapy). About 80% of microalbuminuric and 10% of
macroalbuminuric patients were on monotherapy (ACE inhibitors, angiotensin II
receptor antagonists, or calcium channel blockers); the large majority of
macroalbuminuric patients required adjunctive treatments (
- or
ß-blockers or clonidine and/or diuretics). Any treatment with ACE
inhibitors, angiotensin II receptor antagonists, or calcium channel blockers
was withdrawn for at least 10 d before collecting urine. In these patients,
three urine collections were obtained over a 2-wk period. However, other
antihypertensive medication (- or ß-blockers or clonidine and/or
diuretics) was allowed, to preserve BP control.
Urinary albumin concentration was measured by a single-antibody RIA (Albumina DA; Techno-Genetics, Milan, Italy) on 2-ml aliquots stored at -20°C. This assay has a sensitivity of 0.5 µg/ml and an interassay coefficient of variation within the working range (2 to 80 µg/ml) of less than 10%.
Sitting arterial BP (Korotkov phase I-V) was taken twice with approximation to 2 mmHg after a 10-min rest, and the arithmetic mean of readings was recorded. In agreement with the Joint National Committee criteria (JNC-VI), hypertension was defined by BP levels higher than 140 (systolic) and/or 90 mmHg (diastolic) on at least three different visits over the 3 mo preceding the study, or by treatment with antihypertensive medication (44).
The presence and severity of retinopathy was evaluated by twofield retinal photography (disc-macula-temporal and disc-nasal) in each eye (45°Canon CRC4-45 nuclear magnetic resonance camera) after pharmacologic mydriasis. Retinopathy was graded as absent, background, or proliferative. The presence of new vessels, fibrous proliferation, preretinal or vitreous hemorrhages, or photocoagulation scars defined the proliferative stage of retinopathy.
Glycated hemoglobin (HbAlc) was measured by HPLC (Diamat; Bio-Rad, Richmond, CA; the normal range in our laboratory is 4.1 to 6.1%), and serum creatinine was determined by a standard laboratory technique. Body mass index (BMI) was calculated as weight/height2 (kg/m2).
Finally, plasma ANP levels and transcapillary escape rate of albumin (TERalb) were measured in a subset of normotensive patients (n = 55) with type 1 diabetes. All patients were on a diet with a controlled daily sodium intake of 120 mmol during the week before the study. Two consecutive 24-h urine collections were performed the day before the study to evaluate sodium excretion. None of these patients was taking any drug other than insulin. On the morning of the study, patients with type 1 diabetes omitted their normal insulin treatment. Peripheral blood was drawn between 8:30 and 9:00 a.m., after at least 30 min of rest in the supine position, with the patients in a fasting state. Plasma ANP concentration was determined by an immunoradiometric assay method as described previously (45). The assay has a sensitivity of 2 pg/ml, and the interassay coefficient of variation within the working range (5 to 2000 pg/ml) was between 10 and 4%.
TERalb, defined as the fraction of the intravascular mass of albumin going through the vascular bed per unit time, was measured as described in detail elsewhere (46). Briefly, a cannula was inserted into the antecubital vein of each arm, and the first blood sample was drawn after 30 min of rest in the supine position.
The TERalb was measured by an intravenous bolus injection of 6 to 8 µCi (222 to 296 kBq) of freshly purified 125I-labeled human serum albumin (SARI-125-A-2; Sorin Biomedica, Vercelli, Italy). From the opposite cannula, 8 ml of blood was drawn after discarding the first 2 ml at 10, 15, 20, 30, 40, 45, 50, 55, and 60 min after the injection. Possible thyroid uptake of radioactive iodide resulting from the catabolism of labeled albumin was blocked by administration of Lugol's solution.
Plasma radioactivity (1 ml in duplicate) was measured in a welltyped gamma counter (Cobra 5000; Packard, Downers Grove, IL). Counting time for each sample was 40 min with less than 0.20% error. Plasma glucose, serum albumin, and hematocrit levels (Coulter Counter S5; Coulter Electronics, Bedfordshire, United Kingdom) were determined in each sample, and the plasma radioactivity was expressed taking into account the small changes in plasma volume occurring during the study.
Transcapillary escape rate (%/h) was compiled from the slope of the linear regression equation of the 125I curve from 10 to 60 min. This measurement was accepted only if the correlation coefficient of the time-plasma radioactivity curve exceeded 0.85.
Genetic Study
Leukocytes were isolated from peripheral blood, and DNA was extracted by
standard techniques (47).
Detection of the C708/T polymorphism
The gene encoding for PND, located on the short arm of chromosome 1
(48), contains three exons
separated by two introns.
Step 1: Analysis to Detect New Point Mutations. As an initial study, a fragment of 640 bp extending from exon I to exon II was amplified by PCR and analyzed to detect unknown point mutations.
PCR Cycling Conditions. The primers used to amplify this fragment, selected by the "OLIGO" Program (Med Probe, Oslo, Norway), were: sense primer 5'-AGACAG-AGCAGCAAGCAGTG-3', complementary to nucleotides 527 to 544; antisense primer 5'-CATTTC-CATCCCCAGTTCC-3', complementary to nucleotides 1148-1166 of the published sequence (48). Primers were synthesized on a Gene Assembler (Pharmacia-Biotech, Uppsala, Sweden). Amplification of template DNA (50 ng) was carried out in a DNA Thermal Cycler 480 (Perkin-Elmer Corp., Norwalk, CT) programmed to allow a two-step cycle: denaturation at 94°C for 5 min, followed by 30 cycles of denaturation at 94°C for 60 s, annealing at 55°C for 60 s, and extension at 72°C for 45 s; the last cycle had extension for 20 min at 72°C. The reaction was performed in a final volume of 50 µl, containing 5 pmol of each primer, 10 mM Tris-HCl, pH 8.4, 50 mM KCl, 1.5 mM MgCl2, 200 nM of dNTP, and 2.5 U of Taq polymerase (Boehringer Mannheim, Mannheim, Germany).
To detect unknown point mutations within the 640-bp PCR product, single-strand conformation polymorphism (SSCP) analysis was performed in a randomly selected sample of only 100 patients with insulin-dependent diabetes mellitus. Because SSCP has higher sensitivity in the detection of point mutations when DNA fragments are in the range of 100 to 300 bp (49), PCR products (640 bp) were incubated with the AvaII restriction enzyme (Sigma-Aldrich, St. Louis, MO). AvaII digestion produced three fragments of 298, 116, and 226 bp, respectively.
SSCP Conditions. Six microliters of digested samples was denatured by mixing with 2 µl of a stop solution (95% formaldehyde, 20 mM ethylenediaminetetra-acetic acid, 0.05% bromphenol blue, and 0.05% xylene cyanol), heating at 95°C for 5 min, and quenching on ice. Strand separation was carried out using electrophoresis through nondenaturing 6% polyacrylamide slab gels at 20°C (20 W constant for 2 h), and was visualized by silver staining.
Six of 100 samples showed a similar running pattern that differed from the others, and therefore were sequenced by the dideoxy chain termination method.
Sequencing Conditions. A total of 0.1 pmol of amplicons as template in the two-step cycle temperature sequencing reactions, the thermo-Sequenase fluorescein-labeled cycle sequencing kit (Amersham Life Science, Buckinghamshire, United Kingdom) and 1 pmol of fluorescein-labeled sense primer were used, according to the manufacturer's protocol. After heating at 94°C for 2 min, the termination reactions were subject to 28-cycle sequencing reactions under the following conditions: denaturation at 94°C for 20 s, annealing-extension at 55°C for 45 s. Subsequently, half of the final reaction volume of the formamide loading dye was added to each reaction. Finally, reaction products were loaded on a denaturing 6% acrylamide gel in the Automatic Laser Fluorescence DNA Sequencer (Pharmacia Biotech), and the sequence was automatically detected. Direct sequencing showed a new point mutation at position 708 inside the intron 1 of PND gene. The point mutation encodes a thymidine instead of a cytosine (C708/T) (EMBL Nucleotide Sequence Database, accession no. Y11073).
This point mutation introduces a new polymorphic restriction site for BstXI restriction enzyme in the PND gene.
Step 2: Screening for C708/T Polymorphism in a Cohort of Type 1 Diabetic Patients and in Healthy Control Groups. The allele and genotype frequencies of the C708/T point mutation were evaluated in a cohort of 454 unrelated patients with type 1 diabetes and in 58 age-matched nondiabetic healthy control subjects by restriction fragment length polymorphism (RFLP) analysis.
RFLP Conditions. Fifteen microliters of amplicons (640 bp) was incubated at 55°C with 3 µl of dedicated buffer 10x and 5 U BstXI enzyme in a total volume of 30 µl. Digested products were size-separated by electrophoresis through nondenaturing 10% polyacrylamide slab gels and visualized by silver staining.
In samples homozygous for the polymorphism (C708/C), BstXI identified a single restriction site inside the PCR products, and gave origin to two fragments of 442 and 198 bp. If the new mutation was present, two restriction sites were identified: BstXI produced three fragments of 262, 198, and 180 bp in patients homozygous for the polymorphism (T708/T) and four fragments in heterozygote patients (C708/T). In this case, fragments were of 442, 262, 198, and 180 bp, respectively.
Screening for Known Polymorphisms in the Same Cohort of Type 1 Diabetic Patients and in Healthy Control Groups. HpaII Polymorphism, PCR-RLFP Conditions. An amplicon of 1344 bp, derived from nucleotides 946 to 2289 of the published sequence (48), was amplified using the sense primer RSP1: 5'-GGAAGTCAGC-CCAGCCCAGAGAGAT-3' from nucleotides 946 to 970 and the antisense primer RSP2: 5'-GCAGTCTGTCCCTAGGCCCA-3' from nucleotides 2270 to 2289 of the published sequence (48). The PCR method was performed as described elsewhere (28). One-fifth of the PCR product was digested with HpaII, size-separated on a 10% polyacrylamide gel, and silver stained.
A unique two-allele polymorphism in intron 2 was identified by HpaII restriction enzyme producing fragment sizes 666, 193, and 172 bp (constant fragments) and 313 (H1) and/or 282 (H2) bp.
ScaI Polymorphism, PCR-RLFP Conditions
A fragment of 133 bp was amplified in the region overlapping intron 2 and
the 3' flanking region
(48), using the sense primer
5'-GGCACACTCATACATGAAGCTGACTTTT-3' from nucleotides 2158 to 2185
and the antisense primer 5'-GCAGTCTGTC-CCTAGGCCCA-3' from
nucleotides 2330 to 2349. The PCR method was performed as described elsewhere
(27). A single, two-allele
polymorphism was detected by digestion with ScaI and by size
separation of PCR products on a 15% polyacrylamide gel, and stained with
silver. In the presence of the polymorphic site (allele A2), two
fragments corresponding to sizes 77 and 56 bp were generated; in the absence
of the site, a fragment of 133 bp was observed (allele A1). The
loss of that site caused by mutation leads to an extension of the human ANP by
two additional arginines, so that the original peptide of 28 amino acids is
extended to 30 amino acids.
Statistical Analyses
Data are presented as arithmetic means (± SD). AER and serum
creatinine were not normally distributed and values are given as medians and
95% confidence intervals (95% CI). Normally distributed, continuous variables
were compared among groups using oneway ANOVA and, if differences were found,
using Scheffe F test. Non-normally distributed continuous variables
were compared using the Kruskall-Wallis test and Mann-Whitney U test.
The sign test was used to verify the symmetry of the distributions.
Categorical variables were compared by contingency table analysis. The
relationship between variables was assessed by Spearman rank correlation.
Multivariate stepwise regression analysis was carried out to evaluate the
independence of the relationships (AER was log-transformed before inclusion in
the analysis). ANOVA for repeated measures (one grouping factor and one trial
factor) was applied to verify the constancy of variables such as plasma
glucose, serum albumin, and hematocrit over time during the TERalb
evaluation. Statistical significance was considered at the conventional 5%
level.
Genetic data are presented according to genotype, while allele frequencies
were calculated. The Fisher exact test was used to analyze genotype
distribution in different populations. Allele and genotype frequencies among
patients and control subjects were addressed by the 
2 test
using the values estimated by the Hardy-Weinberg equilibrium association. Odds
ratios were calculated to measure the association of the ANP genotype and
various phenotypes (nephropathy, retinopathy) with the effect of the mutated
allele assumed as dominant (score 0 for wild type and score 1 for heterozygote
and mutated homozygote combined), and a more stringent P value was
used to define significance. Furthermore, the association analysis was
repeated after the whole diabetic cohort was partitioned by duration of
diabetes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In the group of NA patients, 159 subjects had a duration of diabetes less than 15 yr, and 164 subjects showed a duration of disease more than 15 yr. NA patients with long diabetes duration (LDNA) (65 men and 99 women) were older than patients with short diabetes duration (SDNA) and with µA (38.3 ± 10.1 versus 28.5 ± 7.5 and 33.5 ± 8.7 yr, P < 0.001 and P < 0.05, respectively). LDNA had a diabetes duration longer than µA patients (24.6 ± 7.4 versus 21.1 ± 9.0 yr, P < 0.001), and they had an age at onset of diabetes less than SDNA (15.0 ± 9.1 versus 21.1 ± 7.6 yr, P < 0.001). BMI in LDNA was significantly higher than SDNA subjects (23.8 ± 3.0 versus 22.6 ± 2.5 kg/m2, P < 0.001), but lower than healthy control group (23.8 ± 3.0 versus 25.8 ± 3.5 kg/m2, P < 0.001). Systolic BP was higher in LDNA than in SDNA subjects (124.1 ± 17 versus 116.7 ± 12.6 mmHg, P < 0.001). Finally, LDNA subjects had diastolic BP levels (77.9 ± 9.9 mmHg), serum creatinine concentrations (0.80 mg/dl, 0.80 to 0.86 mg/dl), and HbA1c levels (7.9 ± 1.3%) comparable to those of SDNA patients. Prevalence of hypertension in LDNA was 15.9% and prevalence of retinopathy, evaluated as absent, background, and proliferative, was 41.8, 28.7, and 29.5%, respectively. Patients with microalbuminuria had intermediate diabetes duration.
Patients with overt nephropathy had higher systolic and diastolic BP, serum creatinine, and HbA1c levels compared with both NA groups and µA subjects; in these patients, the prevalence of hypertension was also significantly higher than in the other groups. All of the patients with MA had diabetic retinopathy, and the prevalence of proliferative retinopathy was significantly higher when compared with other groups. Patients with retinopathy (background and proliferative, n = 194) were older (37.2 ± 9.8 versus 30.0 ± 8.8 yr), younger at onset of diabetes (14.4 ± 8.8 versus 18.3 ± 8.5 yr), had longer diabetes duration (24.4 ± 8.4 versus 12.9 ± 7.6 yr), and higher systolic and diastolic BP values (128/81 ± 20/12 versus 117/75 ± 13/9 mmHg) (P < 0.0001) than patients with no sign of retinopathy. Prevalence of raised AER levels was 32 and 4%, respectively (P < 0.0001).
Frequencies of Alleles and Genotypes and Association Between
Genotypes and Phenotypes for C708/T Polymorphism
Among the whole cohort with type 1 diabetes, the C708 and
T708 alleles had frequencies of 0.94 (95% CI, 0.92 to 0.955) and
0.06 (95% CI, 0.045 to 0.077), respectively. The genotype frequencies are
consistent with the Hardy-Weinberg equilibrium (2 = 0.26, 1
df, P = 0.88). The frequencies of alleles among the control subjects
did not differ from diabetic patients (C708 = 0.96, 95% CI, 0.90 to
0.98; T708 = 0.04, 95% CI, 0.011 to 0.091), and genotype
distribution was in agreement with the frequencies predicted by the
Hardy-Weinberg equilibrium (2 = 0.07, 1 df, P =
0.96).
Allele and genotype distributions of the PND polymorphism in type 1
diabetic patients were estimated separately in NA, µA, and MA patients
(Table 2). There was no
Hardy-Weinberg disequilibrium either in NA subjects or in patients with µA
and MA, indicating that the distribution over these phenotypes was as could be
expected for a diallelic trait with such an allelic frequency. There were no
significant differences in the allele frequencies between NA and MA patients
(odds ratio for C708/T+T708/T versus
C708/C was 0.78, 95% CI, 0.26 to 2.32, 2 = 0.192, 1
df, P = 0.66). Consistently, no differences were observed between
these groups and healthy control subjects (odds ratio for
C708/T+T708/T versus C708/C were
0.69, 95% CI, 0.24 to 2.06, in NA versus C, 2 = 0.43,
1 df, P = 0.51; and 0.89, 95% CI, 0.21 to 3.75, in MA versus
C, 2 = 0.026, 1 df, P = 0.87).
|
However, the group of patients with µA was in disequilibrium compared
with the allelic frequencies observed within the total group of diabetic
patients (2 = 7.8; 2 df, P = 0.02). Significant
differences were observed in the allele frequencies and genotype distribution
between µA subjects and the other groups
(Table 2). The odds ratio of
the C708/T+T708/T genotype compared with the
C708/C genotype for microalbuminuria versus
normoalbuminuria was 2.78, 95% CI, 1.46 to 5.29, 2 = 10.31, 1
df, P = 0.0013. Allele frequencies and genotype distributions were
reevaluated in the type 1 diabetes cohort with more than 15 yr diabetes
duration (Table 3). No
differences were observed between SDNA (n = 159) and LDNA (n
= 164) patients (odds ratio 1.04, 95% CI, 0.49 to 2.18).
|
The exposure odds ratio of the C708/T+T708/T genotype
compared with the C708/C genotype comparing µA and NA patients
both with long diabetes duration (15 yr) rose to 3.44 (2 =
10.64, 1 df, P = 0.001). No significant associations were seen
between the PND genotypes and diabetic retinopathy (odds ratio of the
C708/T+T708/T genotype versus the
C708/C genotype for absence versus presence of retinopathy
was 1.76, 95% CI, 0.89 to 3.47, NS).
Allele and Genotype Frequencies and Association Between Genotypes and
Phenotypes for HpaII Polymorphism
The H1 and H2 alleles were similarly distributed in
type 1 diabetic patients as well as in healthy control groups (H1
versus H2: 0.02 versus 0.98 and 0.02
versus 0.98, respectively). Both cohorts were in Hardy-Weinberg
equilibrium.
Genotype polymorphism was detected as either H2/H2 and H2/H1, because no H1/H1 was observed. No differences were observed in the allele and genotype distributions over the groups of NA, µA, and MA patients and healthy control subjects (H2: 0.98, 0.97, 0.98, and 0.98; H1: 0.02, 0.03, 0.02, and 0.02, NS). Allele frequencies were superimposable in type 1 diabetic patients with long diabetes duration selected on the basis of AER concentrations.
Frequencies of Alleles and Genotypes and Association Between
Genotypes and Phenotypes for ScaI Polymorphism
Among the whole cohort with type 1 diabetes, the A1 and
A2 alleles had frequencies of 0.18 (95% CI, 0.15 to 0.20) and 0.82
(95% CI, 0.79 to 0.84), respectively. The genotype frequencies are consistent
with the Hardy-Weinberg equilibrium (2 = 4.27, 1 df,
P = 0.12). The frequencies of alleles among the control subjects did not
differ from diabetic patients (A1 0.20, 95% CI, 0.13 to 0.28;
A2 0.80, 95% CI, 0.71 to 0.86), and genotype distribution was in
agreement with the frequencies predicted by the Hardy-Weinberg equilibrium
(2 = 1.12, 1 df, P = 0.57).
Allele and genotype distributions of the ScaI polymorphism were determined separately in type 1 diabetic patients with NA, µA, and MA and in healthy control subjects (Table 2).
A significant difference in the distribution of the A1 and
A2 alleles over the groups was observed (P < 0.0001).
Genotype distribution was significantly different between NA and MA patients
(odds ratio 0.194, 95% CI, 0.08 to 0.46, 2 = 15.99, 1 df,
P = 0.0001). Consistently, differences were observed between NA and µA
groups (odds ratio 0.48, 95% CI, 0.24 to 0.77, 2 = 8.32, 1
df, P = 0.0039); no difference was found between µA and MA
subjects (odds ratio 0.44, 95% CI, 0.16 to 1.2, 2 = 2.65, 1
df, P = 0.1).
Allele frequencies and genotype distributions were reevaluated in the type
1 diabetic cohort with more than 15 yr diabetes duration
(Table 3), and a difference in
the distribution of A1 and A2 allele was observed in the
groups (P = 0.001). The exposure odds ratio of the
A1/A1+A1/A2 genotype
versus the A2/A2 genotype comparing NA and MA
patients both with long diabetes duration (15 yr) was 0.23 (95% CI, 0.09
to 0.58, 2 = 10.91, 1 df, P = 0.001); odds ratio
between NA and µA subjects with long duration of disease was 0.455 (95% CI,
0.22 to 0.90, 2 = 5.13, 1 df, P = 0.023). Odds ratio
between µA and MA patients with a diabetes duration of more than 15 yr was
0.51 (95% CI, 0.18 to 1.48, 2 = 1.54, 1 df, P =
0.21).
A stepwise regression analysis including several clinical and biochemical features (gender, age, diabetes duration, BMI, presence of hypertension [as dummy variable], HbA1c, the C708/T and A1/A2 genotypes) compared to the presence of diabetic nephropathy, considered as the dependent variable, was carried out. Hypertension, the A1/A2 genotype, diabetes duration, and age were significantly and independently associated with kidney disease in the diabetic cohort as a whole (step 1: r = 0.32, F test 24.2; step 2: multiple r = -0.38, F test 10.46; step 3: r = 0.40, F test 11.21; step 4: r = -0.43, F test 8.7).
The extent of disequilibrium in pairwise combinations of alleles at the ANP
locus was estimated by means of the maximum likelihood from the frequency of
diploid genotypes. The ScaI and the BstXI sites are in
positive linkage disequilibrium in the coupling phase with a P value
for the 2 test of zero disequilibrium < 0.001. No
significant associations between the A1/A2 genotype and
diabetic retinopathy were observed.
Association Between PND Gene Polymorphisms and Plasma ANP
Concentrations and TERalb
Plasma ANP concentrations, urinary sodium excretion, and TERalb
were measured in a subset of 55 normotensive type 1 diabetic patients selected
on the basis of C708/T genotype (C708/C, n =
31; C708/T, n = 24)
(Table 4). Thirty-three type 1
diabetic patients (C708/C, n = 19; C708/T,
n = 14) had AER in the range of normoalbuminuria and 22 patients
(C708/C, n = 12; C708/T, n = 10) had
AER in the range of microalbuminuria. As expected
(50), TERalb was
higher in µA patients than in NA subjects (7.7 ± 1.9 versus
6.2 ± 2.2%/h, P = 0.0092).
|
No difference in plasma ANP levels was observed between NA and µA patients (12.1 ± 7.5 versus 12.6 ± 6.9 pg/ml, NS). In the whole cohort, plasma ANP concentrations were significantly higher in C708/C compared with C708/T (15.3 ± 7.7 versus 8.3 ± 3.9 pg/ml, P = 0.0003) (Figure 1a). The difference in plasma ANP concentrations between C708/C and C708/T genotypes also persisted when NA and µA patients were analyzed separately (NA: 15.1 ± 7.6 versus 7.8 ± 4.9 pg/ml, P = 0.005; µA: 15.6 ± 8.2 versus 9.0 ± 2.1 pg/ml, P = 0.03; (Figure 1b).
|
TERalb was also higher in C708/C than in C708/T patients (7.8 ± 2.0 versus 5.5 ± 1.7%/h, P = 0.0001) (Figure 2a). A significant difference in TERalb between C708/C and C708/T genotypes also persisted when NA and µA patients were analyzed separately (NA: 7.3 ± 1.9 versus 4.7 ± 1.5%/h, P = 0.0003; µA: 8.6 ± 1.9 versus 6.7 ± 1.3%/h, P = 0.015) (Figure 2b).
|
In the same subset of type 1 diabetic patients, ScaI polymorphism (A2/A1: 22 subjects, A2/A2: 33 subjects) was also evaluated in relation to plasma ANP levels and TERalb values. No differences were observed between patients with A2/A1 and A2/A2 genotype for gender, age, diabetes duration, BP levels, fasting plasma glucose, HbA1c, previous 2-yr mean HbA1c, and 24-h urine sodium excretion in patients with A2/A2 and A2/A1 genotypes (Table 4).
Plasma ANP levels were significantly higher in patients with A2/A2 genotype than in patients with A2/A1 genotype (14.5 ± 7.4 versus 8.5 ± 4.7 pg/ml, P = 0.0024). Plasma ANP levels were also higher in NA patients with A2/A2 genotype (n = 19) compared with A2/A1 genotype (n = 14) (14.6 ± 7.6 versus 8.4 ± 5.7 pg/ml, P = 0.018), but only slightly higher in µA subjects with A2/A2 genotype (n = 13) than in A2/A1 genotype (n = 9) (14.3 ± 7.4 versus 8.7 ± 2.7, P = 0.06).
TERalb was increased in patients with A2/A2 genotype (n = 33) compared to patients with A2/A1 genotype (n = 22) (7.3 ± 2.1 versus 6.05 ± 2.2%/h, P = 0.044). The difference on the basis of genotype persisted in NA patients (A2/A2 versus A2/A1, 6.83 ± 1.9 versus 5.3 ± 2.2%/h, P = 0.04), but disappeared in the group of µA patients (8.1 ± 2.2 versus 7.5 ± 1.2, P = NS).
Furthermore, a positive linear correlation was found between plasma ANP concentrations and TERalb in the whole cohort (r = 0.37, P = 0.0068). Finally, a stepwise regression analysis including several clinical and biochemical features, such as age, diabetes duration, BMI, systolic BP, diastolic BP, HbA1c, previous 2-yr mean HbA1c, log-transformed AER (or presence of microalbuminuria), and the C708/T and ScaI polymorphisms, versus TERalb, considered as the dependent variable, was performed. The C708/T polymorphism and AER levels (or presence of microalbuminuria, as dummy variable), but no other variables, were significantly and independently associated with albumin leakage in the whole diabetic cohort (step 1: r = -0.49, F test 20.05; step 2: multiple r = 0.63, F test 11.79).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The frequency distribution of this polymorphism was evaluated by restriction enzyme length polymorphism in a cohort of 454 type 1 diabetic patients and a group of nondiabetic healthy control subjects of Caucasian origin.
In the total group of diabetic patients studied, the frequency of the C708/T genotype was about 12%. In the group with type 1 diabetes, the T708 allele was more common in patients with microalbuminuria, although not in patients with overt nephropathy, thereby failing to show an association of the corresponding chromosomal region with this long-term complication of diabetes.
It is thought that microalbuminuria may predict established nephropathy with a positive predictive value greater than 80% (52,53), but this has recently been questioned in patients with duration of diabetes longer than 15 yr (54). This heterogeneity in the chance for progression is supported by recent epidemiologic findings. After 30 yr of diabetes, the prevalence of nephropathy levels off at 30%, whereas the prevalence of microalbuminuria and diabetic nephropathy is 58%. This 28% excess of patients with presistent microalbuminuria after 30 yr of diabetes suggests that many patients with microalbuminuria may never progress to overt proteinuria and only about 50% do so (43,55). In our study, the T708 allele was apparently more common among patients with microalbuminuria and long-term diabetes. Only a longitudinal observation will establish whether this subset of patients will eventually show any progression to overt nephropathy. In the present cross-sectional study, there were stringent criteria to define renal disease and any chance that this may be due to diabetes; it is therefore unlikely that the detection of an association among patients with microalbuminuria rather than patients with proteinuria may reflect an inaccurate definition of reference phenotypes. Furthermore, genotype and allele frequencies did not differ significantly in patients with nephropathy compared with control subjects, short-duration (<15 yr) normoalbuminuric patients who still had a full risk of developing kidney disease, and long-standing normoalbuminuric subjects, whose risk of developing nephropathy is expected to be relatively low (52,53). Patients with microalbuminuria are in disequilibrium with respect to all diabetic cohorts, and the exposure odds ratio of the C708/T versus C708/C genotype is higher than in patients with normoalbuminuria. It is unlikely that this could be due to an unrecognized stratification of our population because both cohorts of type 1 diabetic patients and healthy control subjects came from the same geographic location and were probably homogeneous. So, we suggest that it could be likely related to the presence of microalbuminuria, which has also been considered to be an independent risk factor of microvascular damage in type 1 diabetes (54). It is tempting to speculate that the mutated allele might theoretically be involved in conferring protection from the progression of microvascular damage and also glomerular impairment.
ANP seems to be involved in the regulation of GFR and in the genesis of hyperfiltration (36,37,38), and some studies have demonstrated that ANP infusion increases renal and vascular permeability to proteins in diabetes (24,56,57).
Plasma ANP concentrations and microvascular permeability to albumin have been evaluated in a selected subset of strictly normotensive type 1 diabetic patients, to avoid confounding factors such as any, even slight, increase in BP. As described in other studies (24,47), our findings showed a positive correlation between plasma ANP levels and TERalb, supporting a role for ANP in the regulation of microvascular permeability. Furthermore, we observed that normotensive type 1 diabetic patients with the T708 allele had significantly lower plasma ANP concentrations and TERalb values than patients with the C708 allele, independently of the presence of normo- or microalbuminuria. Because functional endothelial injury is likely to have a role in the pathogenesis of microvascular complications and because TERalb is a reasonable, although not perfect, method for assessing endothelial function in vivo in humans (58), these findings seem to support the hypothesis of a protective role for the recently identified PND gene variant in the progression of microvascular damage.
Moreover, data about additional polymorphisms in the PND gene have been obtained. Interestingly, although the HpaII polymorphism does not seem to play a role in the development of microvascular complications, the ScaI polymorphism shows a strong association with diabetic nephropathy.
Indeed, in type 1 diabetes, the A1 allele was relatively more frequent in normoalbuminuric patients with short- and long-term diabetes duration and progressively decreased in micro- and macroalbuminuric patients, showing an association of the corresponding chromosomal region with this long-term complication of diabetes. Furthermore, stepwise regression showed that the ScaI polymorphism, hypertension, diabetes duration, and age are independently related to diabetic nephropathy. These findings, taken as a whole, suggest an important protective role for the A1 allele in the development and progression of kidney damage.
The A1 allele has been reported to cause the loss of the regular stop codon, leading to an extension of the human natriuretic peptide by two additional arginines, i.e., the peptide of 28 amino acids is extended to 30 amino acids (27). Lower plasma ANP levels have been found in normotensive subjects with type 1 diabetes with the A1 allele, independently of normo- and microalbuminuria, but a week association has been found between the ScaI polymorphism and microvascular permeability, measured by TERalb.
Because raised ANP levels have been described in patients with micro- and macroalbuminuria (39,40), our data could suggest that the ScaI polymorphism may play a major role particularly in the regulation of GFR and the genesis of hyperfiltration (36,37,38).
In conclusion, in type 1 diabetes, the C708/T PND gene variant is associated on the one hand with microalbuminuria and on the other with decreased plasma ANP levels and microvascular permeability. The A1 allele of ScaI polymorphism, related to lower plasma ANP concentrations, seems to play a major role in the susceptibility to diabetic nephropathy, conferring protection from the development of this complication. Whether these phenomena are associated with reduced risk of progression to kidney disease and/or microvascular complications remains to be seen in conjunction with renal function evaluation as well as prospective follow-up studies of microalbuminuria and extrarenal diseases.
|
Acknowledgments |
|---|
This study was supported by a grant from "III Programma di Ricerca Sanitaria Finalizzata" (342/94) of the Regione Toscana. We thank all of the insulin-dependent diabetic subjects who participated in this study.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C of nucleotide 2238 in the gene of atrial
natriuretic peptide (ANP) precursor and heterogeneity of sodium-sensitive
hypertension: Preliminary report. Pol Arch Med Wewn98
: 501-509,1997[Medline]
This article has been cited by other articles:
 |
|
 |
|
  C. Vassalle and M. G. Andreassi Genetic Polymorphisms of the Natriuretic Peptide System in the Pathogenesis of Cardiovascular Disease: What Lies on the Horizon? Clin. Chem., May 1, 2009; 55(5): 878 - 887. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Vassalle, M. G. Andreassi, C. Prontera, M. Fontana, L. Zyw, C. Passino, and M. Emdin Influence of ScaI and Natriuretic Peptide (NP) Clearance Receptor Polymorphisms of the NP System on NP Concentration in Chronic Heart Failure Clin. Chem., November 1, 2007; 53(11): 1886 - 1890. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. G. Ewens, R. A. George, K. Sharma, F. N. Ziyadeh, and R. S. Spielman Assessment of 115 Candidate Genes for Diabetic Nephropathy by Transmission/Disequilibrium Test Diabetes, November 1, 2005; 54(11): 3305 - 3318. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Roussel, D.-A. Tregouet, S. Hadjadj, X. Jeunemaitre, and M. Marre Investigation of the Human ANP Gene in Type 1 Diabetic Nephropathy: Case-Control and Follow-up Studies Diabetes, May 1, 2004; 53(5): 1394 - 1398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nannipieri, M. Manganiello, A. Pezzatini, A. De Bellis, G. Seghieri, and E. Ferrannini Polymorphisms in the hANP (Human Atrial Natriuretic Peptide) Gene, Albuminuria, and Hypertension Hypertension, June 1, 2001; 37(6): 1416 - 1422. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
