(Hypertension. 1996;27:67-71.)
© 1996 American Heart Association, Inc.
Articles |
Polymorphism of the Glycogen Synthase Gene in Hypertensive and Normotensive Subjects
From the First (M.S.), Third (P.K.), and Fourth (C.S.-J., M.T.L., T.T., I.T.) Departments of Medicine and Department of Biochemistry (P.N.-I., X.H., M.L.), Helsinki (Finland) University, and Department of Endocrinology, Malmö General Hospital, University of Lund (L.C.G.), Malmö, Sweden.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Abstract Hypertension and non–insulin-dependent diabetes mellitus (NIDDM) are characterized by a strong genetic component and impaired ability to store glucose as glycogen in skeletal muscle. Impaired insulin activation and altered genetic control of muscle glycogen synthase, the rate-limiting enzyme for glucose storage in skeletal muscle, could provide an explanation for this insulin resistance. We examined whether there is an association between the glycogen synthase gene (Xba I polymorphism) and hypertension in 304 nondiabetic subjects. We examined glucose tolerance with an oral glucose tolerance test and glucose storage in skeletal muscle with the euglycemic insulin clamp technique in combination with indirect calorimetry. The Xba I A2 allele of the glycogen synthase gene was enriched in subjects with hypertension and a family history of NIDDM (48%) compared with normotensive subjects without a family history of NIDDM (6%, P<.0001). The presence of the A2 versus the A1 allele was associated with decreased rates of insulin-stimulated glucose storage in hypertensive subjects (11.2±2.3 versus 16.9±2.6 µmol/kg lean body mass per minute, P=.029) but not in normotensive subjects (28.0±4.6 versus 29.6±3.7 µmol/kg lean body mass per minute). In conclusion, Xba I polymorphism of the glycogen synthase gene identifies a subgroup of hypertensive subjects with a family history of NIDDM. The data suggest that a locus in the glycogen synthase gene region on chromosome 19 may serve as a "thrifty gene," increasing susceptibility for insulin resistance when exposed to other environmental or genetic factors.
Key Words: glycogen synthase • polymorphism, genetics • insulin resistance • blood pressure • non–insulin-dependent diabetes mellitus
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Essential hypertension and NIDDM are both characterized by insulin resistance, manifested as an impaired ability to store glucose as glycogen in skeletal muscle.1 2 3 4 Both conditions have a strong genetic component, and defects in carbohydrate metabolism can be demonstrated already in first-degree relatives of subjects with NIDDM or hypertension.5 6 7 8 9 The key regulatory enzyme for glycogen synthesis in skeletal muscle is GS.10 Decreased rates of insulin-stimulated glucose storage and impaired activation of muscle GS activity by insulin are early events in the pathogenesis of NIDDM.6 7 In nondiabetic subjects increased blood pressure is related to both decreased rates of glucose storage and impaired GS activity.11 The GS gene has been considered a candidate gene in the pathogenesis of NIDDM. Evidence for an association between the GS gene and NIDDM comes from studies in Finnish,12 French,13 and Japanese14 patients as well as Pima Indians.15 Interestingly, both the Finnish and French NIDDM patients identified by Xba I polymorphism of the GS gene had a high prevalence of hypertension.12 13 It is not known whether there is an association between polymorphism of the GS gene and hypertension in nondiabetic subjects. To address this question, we examined 304 nondiabetic subjects, 58 of whom had essential hypertension, for a possible association with the GS gene (Xba I polymorphism). We studied the metabolic consequences of this polymorphism with euglycemic insulin clamp combined with indirect calorimetry.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Subjects
All subjects participating in the study were European and came from the Helsinki area in southern Finland. Liver, kidney, and thyroid function tests were normal in all subjects. Informed consent was obtained from all subjects, and the protocol was approved by the local ethics committee.
All laboratory specimens were taken after a 12-hour
fast. Hypertension
was defined as blood pressure above 160/95 mm Hg documented on at
least three occasions during the previous 3 months or known treatment
for hypertension. Blood pressure was measured with a mercury
sphygmomanometer with the subject in the sitting position after a
15-minute rest. Blood pressure values are given as the average of three
measurements. Subjects with secondary forms of hypertension were
excluded from the study. FHx of NIDDM was regarded as
positive if one of the parents or siblings had NIDDM. Table 1
shows the
characteristics of the study subjects.
|
Oral Glucose Tolerance Test
Glucose tolerance was examined in
172 subjects, 136 of whom were
normotensive (115 with A1 allele; 21 with
A2 allele) and 36 hypertensive (25 with A1
allele; 11 with A2 allele) according to World
Health Organization criteria.16 Diagnostic
values for IGT were fasting plasma glucose less than 7.8 mmol/L and
2-hour glucose of 7.8 to 11.1 mmol/L.16 Insulin and
glucose concentrations were measured before and at 30, 60, 90, and 120
minutes after a 75-g oral glucose load. The area under the curve was
calculated for both insulin and glucose.
Euglycemic Insulin Clamp
Insulin sensitivity was measured
with the
euglycemic, hyperinsulinemic clamp
technique in combination with indirect calorimetry and infusion of
[3-3H]glucose for determination of hepatic glucose
production as previously described.6 After three
baseline samples had been taken for measurement of glucose and insulin
concentrations, a primed constant infusion of short-acting human
insulin (Actrapid, Novo-Nordisk) was administered at a rate of 45
mU/m2 per minute (340 pmol/m2 per minute) for 2
hours. Plasma glucose concentration was determined at 5-minute
intervals, and 20% glucose was infused for maintenance of a
constant plasma glucose concentration.
Indirect Calorimetry
Indirect calorimetry was used during 60
minutes in the basal
state and during the last 60 minutes of the insulin clamp for
estimation of glucose oxidation rates.17 A computerized,
open-circuit system was used for measurement of gas exchange
through a transparent plastic canopy (Deltatrac, Datex Inc). Hepatic
glucose production was measured by the isotope-dilution
technique with [3-3H]glucose (Amersham) administered as a
primed (25 µCi) constant (0.25 µCi/min) infusion for 150 minutes
before the insulin clamp was started and was continued throughout the
study. Total body glucose metabolism equals the mean rate
of glucose infusion during the last 60 minutes of the clamp, provided
there is no entry of glucose from the liver. Nonoxidative glucose
metabolism, ie, glucose storage in skeletal muscle, was
calculated as the difference between total body glucose
metabolism and glucose oxidation as determined by indirect
calorimetry. Lean body mass was determined with bioelectrical
impedance.18
Assays
Plasma glucose was measured with the glucose oxidase
method
adapted for the Beckman Glucose Analyzer II. Serum-free
insulin concentrations were measured by double-antibody
radioimmunoassay (Pharmacia). Serum cholesterol,
high-density lipoprotein cholesterol, and
triglycerides were measured by specific enzymatic
assays.
GS Xba I Polymorphism
A nonradioactive polymerase
chain reaction method was developed
for the determination of Xba I polymorphism of the GS
gene. A genomic clone was isolated from a human genomic library
constructed with Lambda dash vector (Stratagene) with the use of
[
-32P]dCTP–labeled human GS cDNA as a
probe.19 The hybridizations were carried out at 42°C
with 50% formamide. The positive clones were confirmed by Southern
blotting and sequenced with Sequenase 2.0 (USB) single-stranded DNA
of the subcloned lambda DNA (Bluescript SK, Stratagene). Genomic lambda
clone GST11 was isolated and subcloned. The polymorphic
Xba I site was in an intron flanked by the exons coding
bases 1806 to 1969 and 1970 to 2050 of the cDNA. Two
oligonucleotides were synthesized:
5'-CTCCTTCCTCTACAGTTTCTG-3', located upstream, and
5'-GTGAGTCTCCTCTTTGGCCA-3', located downstream of the polymorphic
Xba I site. Genomic DNA was extracted from
peripheral blood leukocytes by the standard method. Genomic
DNA (100 ng) was amplified with polymerase chain reaction (Perkin-Elmer
Cetus) in a reaction mixture containing 10 pmol of both primers, 1.25
mmol/L MgCl2, 0.1 U Taq polymerase
(Promega) and 1x Taq polymerase reaction buffer (Promega),
in a total volume of 20 µL. The reaction was carried out at 96°C
for 3 minutes, followed by 35 cycles at 96°C for 1 minute, at 61°C
for 1 minute, and at 72°C for 1 minute and final extension for 10
minutes at 72°C. After amplification, 2 U Xba I
restriction enzyme (Promega), 2 µL of 10x restriction enzyme buffer
(Promega), and distilled H2O were added to a final volume
of 40 µL and incubated for 1 hour at 37°C. The resulting fragments
were separated and analyzed on ethidium bromide–stained
1% agarose gel in 1x TBE buffer (90 mmol/L Tris-borate, 2 mmol/L
EDTA) at a constant voltage of 4 V/cm for 2 hours.
Statistical Analysis
Data are expressed as mean±SEM.
Statistical analyses
were performed with a BMDP computer program. The significance of
differences between group means was tested with the Mann-Whitney rank
sum test (nonparametric). In addition, ANCOVA was performed
with BMI and age (OGTT data) and BMI, age, and fasting
plasma-glucose (clamp data) as covariates to adjust for differences
in these parameters between the groups studied. The
significance of the frequency difference of GS Xba I
alleles was tested by 
2 analysis with
Yates' correction.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Xba I Polymorphism of the GS Gene
With the use of the restriction enzyme Xba I, two polymorphic patterns can be demonstrated in the GS gene. The A1 allele lacks the Xba I site and gives a fragment of approximately 600 bp in size, whereas the A2 allele is cleaved into two fragments of 500 and 100 bp in size.
Of 304 unrelated, nondiabetic subjects, 253 (83%) had the genotype A1A1 and 51 (17%) the A2 allele. Most subjects with the A2 allele (n=48) were heterozygous, whereas homozygosity for the A2 allele (A2A2) could be demonstrated in only 3 subjects.
The characteristics of study subjects
are given in Table 1
.
Hypertensive subjects were characterized by significantly higher
fasting insulin values, higher total cholesterol and
triglyceride concentrations, lower high-density
lipoprotein cholesterol concentrations, and increased BMI
and age compared with normotensive subjects (Table 1).
The A2 allele was twice as frequent among the
hypertensive as among the normotensive subjects (28% versus 14%,
P=.024) (Fig 1). The highest frequency of the
A2 allele (48%) was observed among hypertensive
subjects with FHx of NIDDM and the lowest in normotensive
subjects without FHx of NIDDM (6%); the frequency in
subjects with isolated FHx of NIDDM (26%) or hypertension
(16%) was intermediary (2=33.7, 3 df,
P<.0001) (Fig 2).
|
|
Characteristics of Subjects With the A1 or
A2 Allele
Characteristics of normotensive and hypertensive
subjects with the
A1 or A2 allele are given in Table 2. A
reliable FHx of hypertension was
obtained for 103 subjects. Of these subjects, 57% of those with the
A2 allele and 46% with the A1 allele
had FHx of hypertension (P=NS). Sixteen percent
of subjects with the A2 allele and 13% of those with
the A1 allele were taking antihypertensive drugs.
|
Glucose Tolerance
Normotensive subjects with the
A2 allele were
younger than normotensive subjects with the A1 allele
(39.0±2 versus 45±2 years, P=.047). Glucose
tolerance was
similar in normotensive subjects with the A1 or
A2 allele (P=NS). The areas under the curve
for glucose (27.3±0.5 versus 25.3±1.6 mmol/Lx120 minutes,
P=.40) and insulin (1441±87 versus 989±118
pmol/Lx120
minutes, P=.10) did not differ significantly between
subjects with the A1 or A2 allele,
respectively (adjusted for age and BMI).
In hypertensive subjects, the prevalence of IGT was significantly higher in subjects with the A2 compared with those with the A1 allele (36% versus 4%, P=.04). The areas under the curve for glucose (32.4±2.1 versus 29.8±0.9 mmol/Lx120 minutes, respectively; P=.57) and insulin (1446±240 pmol/L versus 1598±190x120 minutes, respectively; P=.68) did not differ significantly between the two groups (adjusted for age and BMI).
Rates of Insulin-Stimulated Glucose Storage
To examine
whether the A2 allele was associated
with decreased rates of glucose storage, we performed
euglycemic insulin clamp studies in combination with
indirect calorimetry (Table 3) in matched subgroups of
normotensive subjects (A1 allele: 4 women/4 men; mean
age, 39±2 years; BMI, 25.0±1.4 kg/m2; A2
allele: 4 women/4 men; mean age, 38±2 years; BMI, 24.6±0.8
kg/m2) and hypertensive subjects (A1
allele: 1 woman/6 men; mean age, 49±5 years; BMI, 27.8±1.6
kg/m2; A2 allele: 7 men; mean age, 53±4
years; BMI, 28.3±1.4 kg/m2). Insulin sensitivity and
glucose storage rates did not differ between normotensive subjects with
the A1 or A2 allele (Table 3). Among
hypertensive subjects, subjects with the A2 allele had
lower rates of insulin-stimulated glucose storage
(P=.029) compared with subjects with the A1
allele (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Hypertension is already observed in about half of NIDDM patients at diagnosis. Both hypertension and NIDDM are characterized by insulin resistance manifested as a decreased ability to store glucose as glycogen in skeletal muscle.2 4 We12 and others13 have previously demonstrated an association between polymorphism of the GS gene and NIDDM. It is not known whether hypertension in these subjects represents a secondary phenomenon or predisposes to the development of NIDDM. The present study examined GS Xba I polymorphism in a nondiabetic population, and the A2 allele of this polymorphism was highly enriched in subjects having overt hypertension and FHx of NIDDM. On the basis of results from twin and adoption studies and statistical analyses of blood pressure in various pedigrees, approximately 20% to 60% of the population variability in blood pressure is estimated to be genetically determined.20 21 GS Xba I polymorphism seems to identify a subgroup of subjects with a genetic background involving both hypertension and NIDDM. High blood pressure frequently associates with IGT, indicating, at least in part, a common genetic background for these disorders.22 23 24 25 The significant association with the GS gene in subjects having both hypertension and FHx of NIDDM implies that this may represent a distinct disease entity, the susceptibility of which is increased by some genetic defect or defects within or in the vicinity of the GS gene.
IGT was more frequent in hypertensive subjects with the A2 than with the A1 allele. No such difference was observed in normotensive subjects. However, caution is warranted in the interpretation of these data because normotensive subjects with the A2 allele were younger than those with the A1 allele. Therefore, this could have masked an expected age-dependent increase in the frequency of IGT in these subjects.
In subjects with NIDDM the presence of the A2 allele was associated with impaired insulin-stimulated glucose storage.12 In keeping with this finding we observed a significant decrease in the rate of insulin-stimulated glucose storage also in hypertensive subjects with the A2 allele. However, this difference was not seen in the normotensive subjects. In support of this, there was no significant difference in insulin responses during the OGTT in normotensive subjects. Insulin responses during the OGTT were similar also in hypertensive subjects. If anything, insulin responses tended to be diminished in the subjects with the A2 allele. How could these data be reconciled? First, changes in insulin sensitivity explain only 30% of the variance in insulin concentrations during the OGTT.26 Second, the defect was confined to the glucose storage pathway of glucose metabolism, whereas glucose oxidation was normal. These changes in the intracellular partitioning of glucose may be too subtle to be reflected by changes in insulin concentration during the OGTT. Third, impaired insulin secretion was recently reported in hypertensive subjects with IGT independent of insulin resistance.27 This could explain why the hypertensive subjects with the A2 allele did not show an increase in the insulin response during the OGTT.
Why would the presence of the A2 allele be associated with impaired skeletal muscle glycogen synthesis in patients with NIDDM and hypertension but not in lean, normoglycemic individuals? A recent animal study may shed some light on this discrepancy. The trait of developing diabetes during fat feeding was recently linked to the GS gene locus on chromosome 7 in the C57BL/6J mouse.28 The diabetes-prone mouse was further characterized by impaired GS activity in skeletal muscle and elevated fasting insulin concentrations (as a measure of insulin resistance). When kept on a normal diet, the mice maintained normal weight and normal glucose and insulin concentrations.28 These data suggest that no metabolic abnormalities could be discerned until the "thrifty gene" was exposed to an environment of high energy intake. If the same applies for Xba I polymorphism of the GS gene in humans, the presence of the genetic marker in an otherwise lean and healthy person should not be associated with the metabolic abnormalities of insulin resistance. However, we do not know what the pathogenetic link is between this polymorphism and insulin resistance. Since insulin resistance was observed mainly in obese (NIDDM12 or hypertensive) subjects with the A2 allele, obesity may be the common denominator predisposing to the other conditions.
In conclusion, Xba I polymorphism of the GS gene identifies a subgroup of subjects with both hypertension and FHx of NIDDM. Since this polymorphism is associated with insulin resistance only in obese subjects with hypertension and/or NIDDM, it may represent a "thrifty gene," which has to be exposed to an affluent environment before the associated metabolic abnormalities are unmasked.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by grants from the Sigrid Jusélius Foundation, the Finnish Diabetes Research Foundation, Finska Läkaresällskapet, the Perklén Foundation, and Svenska Kulturfonden. We are grateful to Maija Parkkonen, RT, and Esa Laurila, RT, for technical assistance.
|
Footnotes |
|---|
Reprint requests to C. Schalin-Jäntti, MD, IV Department of Medicine, Helsinki University Hospital, Unioninkatu 38, FIN-00170, Helsinki, Finland.
Received April 17, 1995; first decision July 25, 1995; accepted September 6, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Reaven GM. Role of insulin resistance in human disease. Diabetes. 1988;37:1595-1607. [Abstract]
2. DeFronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia and atherosclerotic cardiovascular disease. Diabetes Care. 1991;14:173-194. [Abstract]
3. Haffner SM, Valdez RA, Hazuda HP, Mitchell BD, Morales PA, Stern MP. Prospective analysis of the insulin-resistance syndrome (syndrome X). Diabetes. 1992;41:715-722. [Abstract]
4. Ferrannini E, Buzzigoli G, Bonadonna R, Giorico MA, Oleggini M, Graziadei L, Pedrinelli R, Brandi L, Bevilacqua S. Insulin resistance in essential hypertension. N Engl J Med. 1987;317:350-357. [Abstract]
5. Eriksson J, Franssila-Kallunki A, Ekstrand A, Saloranta C, Widén E, Schalin C, Groop L. Early metabolic defects in persons at increased risk for non-insulin dependent diabetes mellitus. N Engl J Med. 1989;321:337-343. [Abstract]
6. Schalin-Jäntti C, Härkönen M, Groop L. Impaired activation of glycogen synthase in people at increased risk for developing NIDDM. Diabetes. 1992;41:598-604. [Abstract]
7. Vaag A, Henriksen JE, Beck-Nielsen H. Decreased insulin activation of glycogen synthase in skeletal muscles in young nonobese Caucasian first-degree relatives of patients with non-insulin-dependent diabetes mellitus. J Clin Invest. 1992;89:782-788.
8. Alleman Y, Horber FF, Colombo M, Ferrari P, Shaw S, Jaeger P, Weidmann P. Insulin sensitivity and body fat distribution in normotensive offspring of hypertensive parents. Lancet. 1993;341:327-331. [Medline] [Order article via Infotrieve]
9. Ohno Y, Suzuki H, Yamakawa H, Nakamura M, Otsuka K, Saruta T. Impaired insulin sensitivity in young, lean normotensive offspring of essential hypertensives: possible role of disturbed calcium metabolism. J Hypertens. 1993;11:421-426. [Medline] [Order article via Infotrieve]
10. Leloir LF, Olavaria JM, Goldenburg SH, Carminatti H. Biosynthesis of glycogen from uridine diphosphate glucose. Arch Biochem Biophys. 1959;81:508-520. [Medline] [Order article via Infotrieve]
11. Schalin-Jäntti C, Laurila E, Groop L. Blood pressure is related to skeletal muscle glycogen synthesis and muscle glycogen synthase activity. Diabetologia. 1995;38(suppl 1):A164,636. Abstract.
12.
Groop LC, Kankuri M, Schalin-Jäntti C, Ekstrand
A, Nikula-Ijäs P, Widén E, Kuismanen E, Eriksson J,
Franssila-Kallunki A, Saloranta C, Koskimies S. Association
between polymorphism of the glycogen synthase gene and
non-insulin dependent diabetes mellitus. N
Engl J Med. 1993;328:10-14.
13.
Zouali H, Velho G, Froguel P. Polymorphism
of the glycogen synthase gene and non-insulin-dependent
diabetes mellitus. N Engl J Med. 1993;328:1568. Letter.
14. Kuroyama K, Sanke T, Ohagi S, Furuta M, Furuta H, Nanjo K. Simple tandem repeat polymorphism in the human glycogen synthase gene is associated with NIDDM in Japanese subjects. Diabetologia. 1994;37:536-539. [Medline] [Order article via Infotrieve]
15. Majer M, Mott DM, Pedersen O, Knowler WC, Bennett PH, Bogardus C. Association of NIDDM with the glycogen synthase locus on 19q13 in the Pima Indians. Diabetes. 1995;44(suppl 1):18A,60. Abstract.
16. World Health Organization. Diabetes Mellitus: Report of a WHO Study Group. Geneva, Switzerland: World Health Organization; 1985. Technical Report Series No. 727.
17. Ferrannini E. The theoretical basis of indirect calorimetry: a review. Metabolism. 1988;37:287-301. [Medline] [Order article via Infotrieve]
18.
Deurenberg P, Weststrate JA, van der Kooy K.
Body composition changes assessed by bioelectrical impedance
measurements. Am J Clin Nutr. 1989;49:401-403.
19.
Browner MF, Nakano K, Bang AG, Fletterick RJ.
Human muscle glycogen synthase cDNA sequence: a negatively charged
protein with an asymmetric charge distribution. Proc Natl
Acad Sci U S A. 1989;86:1443-1447.
20. Spence MA. Segregation analysis. In: Rimoin DL, Emery AEH, eds. Principles and Practice of Medical Genetics. 2nd ed. Edinburgh, UK: Churchill-Livingstone; 1990:115-120.
21. Kurtz TW. Genetics of essential hypertension. Am J Med. 1993;94:77-84. [Medline] [Order article via Infotrieve]
22.
Jarrett RJ, Keen H, McCartney M, Fuller JH, Hamilton
PJS, Reid DD, Rose G. Glucose tolerance and blood pressure in
two population samples: their relation to diabetes mellitus and
hypertension. Int J Epidemiol. 1978;7:15-24.
23. Modan M, Halkin H, Almog S, Lusky A, Eshkol A, Shefi M, Shitrit A, Fuchs Z. Hyperinsulinemia: a link between hypertension, obesity and glucose intolerance. J Clin Invest. 1985;75:809-817.
24. Reaven G. Relationship between insulin resistance and hypertension. Diabetes Care. 1991;14:33-38.
25. Lind L, Jakobsson S, Lithell H, Wengle B, Ljunghall S. Relation of serum calcium concentration to metabolic risk factors for cardiovascular disease. Br Med J. 1988;297:960-963.
26.
Laakso M. How good a marker is insulin level for
insulin resistance? Am J Epidemiol. 1993;137:959-965.
27. Abel ED, Ledingham JG. Impaired glucose tolerance in hypertension is associated with impaired insulin release independently of changes in insulin sensitivity. J Hypertens. 1994;12:1265-1273. [Medline] [Order article via Infotrieve]
28. Seldin MF, Mott D, Bhat D, Petro A, Kuhn CM, Kingsmore SF, Bogardus C, Opara E, Feinglos MN, Surwit RS. Glycogen synthase: a putative locus for diet-induced hyperglycemia. J Clin Invest. 1994;94:269-276.
This article has been cited by other articles:
 |
|
 |
|
  J. St-Onge, D. R. Joanisse, and J.-A. Simoneau The Stimulation-Induced Increase in Skeletal Muscle Glycogen Synthase Content Is Impaired in Carriers of the Glycogen Synthase XbaI Gene Polymorphism Diabetes, January 1, 2001; 50(1): 195 - 198. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||