• BMI;
  • genetic susceptibility


  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgments
  5. References

This study investigated the role of the ENPP1/PC-1 gene K121Q polymorphism in predicting BMI (kg/m2) in non-diabetic individuals. Three independent samples (n = 631, n = 304, and n = 505) of adult whites were analyzed. Selection criteria were fasting plasma glucose level <126 mg/dL, absence of severe obesity (BMI ≥40 kg/m2), and lack of treatment known to modulate BMI. In Sample 1, BMI values were different in individuals carrying the K121/K121 (KK), K121/Q121 (KQ), and Q121/Q121 (QQ) genotypes (25.5 ± 4.3, 25.3 ± 4.1, and 22.8 ± 2.5 kg/m2, respectively (adjusted p = 0.022); BMI values in Samples 2 and 3 also tended to be different, although the differences, after adjustment for age and sex, did not reach statistical significance. When data were pooled, BMI values were 25.8 ± 4.4, 25.6 ± 4.4, and 23.6 ± 3.3 kg/m2 in KK, KQ, and QQ individuals (adjusted p = 0.029). According to a recessive model, QQ individuals had lower BMI values than KK and KQ individuals combined (23.6 ± 3.3 kg/m2 vs. 25.7 ± 4.4 kg/m2; adjusted p = 0.008). These data suggest that the QQ genotype of the ENPP1/PC-1 gene is associated with lower BMI. If similar results are confirmed in prospective studies, the K121Q polymorphism may help identify people at risk for obesity.

Ectoenzyme nucleotide pyrophosphate phosphodiesterase (ENPP1),1 also known as plasma cell membrane glycoprotein 1 (PC-1), is a membrane glycoprotein that reduces insulin action by binding to the insulin receptor and inhibiting subsequent signaling (1) (2). A missense polymorphism (i.e., K121Q) of the ENPP1/PC-1 gene has been described (3). The Q121 variant binds and inhibits insulin receptor more strongly than the K121 variant (4) and is associated with insulin resistance and related metabolic abnormalities in the vast majority of studied populations (3) (5) (6) (7) (8). Recent findings indicate that the Q121 variant is also associated with type 2 diabetes (9) (10) (11) (12) (13), and among diabetic patients with both cardio- and nephrovascular complications (12) (14), all clinical conditions are likely consequences of insulin resistance. More recently, conflicting data have been reported concerning the possibility that this variant also modulates BMI (13) (15) (16). The aim of the present study was to investigate the role of the K121Q polymorphism on BMI, independently of type 2 diabetes.

Three independent cohorts of non-diabetic unrelated whites were studied. The first sample (from Gargano, Italy) comprised 631 individuals (men/women, 238/393; mean age, 36.8 ± 11.7 years; mean BMI, 25.4 ± 4.2 kg/m2; BMI range, 18.1 to 39.9 kg/m2). Individuals with BMI values ≥40 kg/m2 were excluded to reduce the risk of including patients affected by severe uncommon forms of obesity. BMI values differed across the three genotype groups: 25.5 ± 4.3, 25.3 ± 4.1, and 22.8 ± 2.5 kg/m2 in individuals carrying the K121/K121 (KK), K121/Q121 (KQ), and Q121/Q121 (QQ) genotypes, respectively (p = 0.022 after adjusting for age and sex) (Table 1). BMI values of QQ individuals (22.8 ± 2.5 kg/m2) were lower than those observed in KK and KQ subjects, considered together according to a recessive model (25.4 ± 4.3 kg/m2; adjusted p = 0.007). Also, waist circumference was lower in QQ vs. KK + KQ individuals (78.6 ± 11.5 cm vs. 82.76 ± 12.3 cm; adjusted p = 0.025). As a replication attempt, a second sample of 304 individuals from Sicily was studied (men/women, 137/167; mean age, 37.6 ± 12.6 years; mean BMI, 25.8 ± 4.6 kg/m2; BMI range, 18.0 to 39.8 kg/m2). In this group, also, BMI values tended to be lower in QQ individuals, although the difference did not reach statistical significance with the present sample size (Table 1). A similar, although not significant, trend (Table 1) was also obtained in a third sample of 505 whites (men/women, 218/287; mean age, 46.0 ± 16.3 years; mean BMI, 26.0 ± 4.5 kg/m2; BMI range, 18.0 to 39.7 kg/m2) from the Dallas, TX, area. Waist circumference was not available for Samples 2 and 3. The association between the K121Q polymorphism and BMI was not different among the three samples (p = 0.50, for interaction between K121Q and place of recruitment in modulating BMI); therefore, the data were pooled and analyzed together. The BMI values were different across the three genotype groups: 25.8 ± 4.4, 25.6 ± 4.4, and 23.6 ± 3.3 kg/m2 in KK, KQ, and QQ individuals, respectively (p = 0.029 after adjusting for age, sex, and place of recruitment) (Table 1). QQ individuals had significantly lower BMI values than KK and KQ individuals considered together, according to a recessive model (23.6 ± 3.3 kg/m2 vs. 25.7 ± 4.4 kg/m2, adjusted p value = 0.008). Levels of fasting glucose, insulin, triglycerides, and high-density lipoprotein-cholesterol were not different across the three genotype groups in any of the samples studied (data not shown). The present study indicates that the Q121 variant of the ENPP1/PC-1 gene is associated, according to a clear recessive model, with lower BMI in non-diabetic whites. We acknowledge that because the proportion of QQ individuals is low among whites, the number of QQ subjects we have been able to study is small and does not allow drawing firm conclusions about the reported association. Nonetheless, the fact that similar data were obtained in three independent samples from both southern Europe and the United States minimizes the risk of spurious results and suggests that, at least among whites, this might be considered a generalized phenomenon. Because diet composition is known to differ greatly between the United States and southern Europe, our findings also suggest that the association between the K121Q polymorphism and BMI is unlikely to be influenced by this potentially important environmental factor. The most likely mechanism accounting for the reduced BMI in Q121 carriers that we and others (16) have observed is the deleterious effect of this variant on insulin resistance (3) (5) (6) (7) (8), which, itself, has been reported to be a predictor of lower BMI in a prospective study (17). As a matter of fact, insulin signaling stimulates lipid storage in adipocytes, and this might explain why adipose tissue-selective, insulin-resistant animals are protected from obesity (18). In addition, treatment with glitazones not only improves insulin sensitivity but also stimulates an increase in body fat (19). Therefore, although our explanation is entirely speculative, it can be hypothesized that at least some genetic determinants improving insulin sensitivity have an opposite deleterious role on body weight, and vice versa. In line with this concept, recent reports have shown that variants of the adiponectin gene predicting higher insulin sensitivity are, in fact, associated with increased BMI (20).

Table 1.  Clinical features of non-diabetic subjects of the three samples studied, according to the ENPP1/PC-1 K121Q genotype
  • Data are means ± standard deviation. Comparisons across genotype groups were adjusted for age and sex and, when the three samples were pooled and analyzed together, also for place of recruitment.

  • *

    p< 0.05 as compared with the other two genotype groups by one-way ANOVA.

  • p< 0.01 as compared with KK + KQ individuals, according to a recessive model.

 Sample 1 (Gargano) (n = 631)  Sample 2 (Sicily) (n = 304)  
ENPP1/PC-1 genotypeKK (n = 465)KQ (n = 152)QQ (n = 14)KK (n = 217)KQ (n = 79)QQ (n = 8)
Sex (M/F)177/28854/987/797/12037/423/5
Age (years)36.8 ± 11.436.7 ± 12.437.9 ± 11.337.5 ± 12.638.4 ± 12.730.0 ± 10.0
BMI (kg/m2)25.5 ± 4.325.3 ± 4.122.8 ± 2.5*25.8 ± 4.725.8 ± 4.724.8 ± 4.2
 Sample 3 (Dallas) (n = 505)  Pooled samples (n = 1440)  
ENPP1/PC-1 genotypeKK (n = 372)KQ (n = 125)QQ (n = 8)KK (n = 1054)KQ (n = 356)QQ (n = 30)
Sex (M/F)159/21355/704/4433/621146/21014/16
Age (years)46.9 ± 16.343.4 ± 16.048.0 ± 19.940.5 ± 14.339.4 ± 14.138.5 ± 14.9
BMI (kg/m2)26.1 ± 4.425.9 ± 4.723.7 ± 3.725.8 ± 4.425.6 ± 4.423.6 ± 3.3*

Although a previous study reported on a similar association between the Q121 variant and decreased BMI in both whites and African Americans from the United States (16), two other studies reported the opposite association (i.e., higher BMI in QQ individuals) (13) (15) and suggested that the Q121 allele is a gene variant with pleiotropic deleterious effects on insulin resistance, obesity, and type 2 diabetes. Discordant results in genotype-phenotype association studies are not an uncommon event in the study of complex disorders and can arise both from false-positive or false-negative results and from differences in the genetic and/or environmental background of the populations studied (21). To ascertain whether the Q121 variant of the ENPP1/PC-1 gene behaves not only as a modulator of insulin resistance (3) (5) (6) (7) (8) and type 2 diabetes (9) (10) (11) (12) (13), but also as a negative or positive modulator of BMI, large-scale prospective studies are needed; these should be set up in a collaborative way, thus, providing sufficient power to investigate gene-gene and gene-environment interactions and their differences, if any, among populations.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgments
  5. References

Experimental Subjects

Three independent cohorts of non-diabetic unrelated whites were studied. The first sample comprised 631 healthy unrelated whites residing in the Gargano region (central east coast of Italy) and recruited from the employees of the Scientific Institute Casa Sollievo della Sofferenza in San Giovanni Rotondo, Italy. The second sample comprised 304 unrelated whites from Sicily. Overweight and obese subjects (BMI ≥25 kg/m2) were recruited from the Outpatient Metabolic Clinic of Garibaldi Hospital at the University of Catania, Italy, whereas normal-weight subjects were recruited from the staff of the same hospital. As expected, when compared with normal-weight subjects (n = 149), overweight and obese individuals (n = 155) had a higher proportion of hypertension (34.8% vs. 8.7%, p < 0.0001) and dyslipidemia (26.5% vs. 11.4%, p < 0.001). No other chronic systemic diseases were observed in any subject. The third sample, consisting of 505 non-Hispanic whites, was recruited by public advertisement and the offer of free screening for cardiovascular risk factors at the University of Texas Southwestern Lipid and Heart Disease Risk Management Clinic, Dallas, TX. Height and weight were measured by standard procedures. Selection criteria for all subjects were fasting plasma glucose of <126 mg/dL, absence of severe obesity (BMI ≥40 kg/m2), and lack of treatment known to interfere with body weight or glucose and lipid metabolism. The study was performed according to the Helsinki Declaration and approved by the local ethical committees. All subjects provided written informed consent.

K121Q Genotyping

Genomic DNA was extracted from whole blood by using a DNA isolation kit (Roche Diagnostics, Indianapolis, IN). Genotyping was performed as previously described (12). The Q121 allele frequency was 0.14, 0.15, and 0.14 in Samples 1, 2, and 3, respectively, and was in Hardy-Weinberg equilibrium.

Statistical Analysis

Values are means ± standard deviation. Comparisons between groups were tested by an unpaired Student's t test or Mann-Whitney test, as appropriate, and comparisons among the three groups by one-way ANOVA. Mean values, after adjusting for covariates, were evaluated by analysis of covariance. To model the effect of the polymorphism on bivariate variables, multivariate logistic regression analysis was used, and odds ratios (and 95% confidence intervals) were calculated. Hardy-Weinberg equilibrium was evaluated by χ2 test. Interaction between K121Q and place of recruitment in modulating different variables was tested by general linear model univariate analysis.

A p value of <0.05 was considered significant. All analyses were performed using the SPSS statistical package (version 12; SPSS, Inc., Chicago IL).


  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgments
  5. References

This work was partly supported by an Italian Ministry of Health R.C.2006 grant (to S.P.); NIH Grants K23-RR16075 and RO1 DK072158-01 (to N.A.); and Centers for Disease Control and Prevention Grant H75/CCH523202 and American Heart Association Grant 0465017Y (to M.C.).

  • 1

    Nonstandard abbreviations: ENPP1, ectoenzyme nucleotide pyrophosphate phosphodiesterase; PC-1, plasma cell membrane glycoprotein 1.

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgments
  5. References
  • 1
    Maddux, B. A., Sbraccia, P., Kumakura, S., et al (1995) Membrane glycoprotein PC-1 and insulin resistance in non-insulin-dependent diabetes mellitus. Nature 373: 448451.
  • 2
    Dong, H., Maddux, B. A., Altomonte, J., et al (2005) Increased hepatic levels of the insulin receptor inhibitor, PC-1/NPP1, induce insulin resistance and glucose intolerance. Diabetes 54: 367372.
  • 3
    Pizzuti, A., Frittitta, L., Argiolas, A., et al (1999) A polymorphism (K121Q) of the human glycoprotein PC-1 gene coding region is strongly associated with insulin resistance. Diabetes 48: 18811884.
  • 4
    Costanzo, B. V., Trischitta, V., Di Paola, R., et al (2001) 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 50: 831836.
  • 5
    Gu, H. F., Almgren, P., Lindholm, E., et al (2000) Association between the human glycoprotein PC-1 gene and elevated glucose and insulin levels in a paired-sibling analysis. Diabetes 49: 16011603.
  • 6
    Rasmussen, S. K., Urhammer, S. A., Pizzuti, A., et al (2000) The K121Q variant of the human PC-1 gene is not associated with insulin resistance or type 2 diabetes among Danish Caucasians. Diabetes 49: 16081611.
  • 7
    Kubaszek, A., Pihlajamaki, J., Karhapaa, P., Vauhkonen, I., Laakso, M. (2003) The K121Q polymorphism of the PC-1 gene is associated with insulin resistance but not with dyslipidemia. Diabetes Care 26: 464467.
  • 8
    Abate, N., Carulli, L., Cabo-Chan, A., Chandalia, M., Snell, P. G., Grundy, SM (2003) Genetic plymorphism PC-1 K121Q and ethnic susceptibility to insulin resistance. J Clin Endocrinol Metab. 88: 59275934.
  • 9
    Kubaszek, A., Markkanen, A., Eriksson, J. G., et al (2004) The association of the K121Q polymorphism of the plasma cell glycoprotein-1 gene with type 2 diabetes and hypertension depends on size at birth. J Clin Endocrinol Metab. 89: 20442047.
  • 10
    Hamaguchi, K., Terao, H., Kusuda, Y., et al (2004) The PC-1 Q121 allele is exceptionally prevalent in the Dominican Republic and is associated with type 2 diabetes. J Clin Endocrinol Metab. 89: 13591364.
  • 11
    Abate, N., Chandalia, M., Satija, P., et al (2005) ENPP1/PC-1 K121Q polymorphism and genetic susceptibility to type 2 diabetes. Diabetes 54: 12071213.
  • 12
    Bacci, S., Ludovico, O., Prudente, S., et al (2005) The K121Q polymorphism of the ENPP1/PC-1 gene is associated with insulin resistance/atherogenic phenotypes, including earlier onset of type 2 diabetes and myocardial infarction. Diabetes 54: 30213025.
  • 13
    Meyre, D., Bouatia-Naji, N., Tounian, A., et al (2005) Variants of ENPP1 are associated with childhood and adult obesity and increase the risk of glucose intolerance and type 2 diabetes. Nat Genet. 37: 863867.
  • 14
    De Cosmo, S., Trevisan, R., Vestra, Dalla M., et al (2003) PC-1 amino acid variant Q121 is associated with a lower glomerular filtration rate in type 2 diabetic patients with abnormal albumin excretion rates. Diabetes Care 26: 28982902.
  • 15
    Barroso, I., Luan, J., Middelberg, R. P., et al (2003) Candidate gene association study in type 2 diabetes indicates a role for genes involved in beta-cell function as well as insulin action. PLoS Biol. 1: E20.
  • 16
    Matsuoka, N., Patki, A., Tiwari, H. K., et al (2005) Association of K121Q polymorphism in ENPP1 (PC-1) with BMI in Caucasian and African-American adults. Int J Obes. 11: 15.
  • 17
    Swinburn, B. A., Nyomba, B. L., Saad, M. F., et al (1991) Insulin resistance associated with lower rates of weight gain in Pima Indians. J Clin Invest. 88: 168173.
  • 18
    Bluher, M., Michael, M. D., Peroni, O. D., et al (2002) Adipose tissue selective insulin receptor knockout protects against obesity and obesity-related glucose intolerance. Dev Cell 3: 2538.
  • 19
    Larsen, T. M., Toubro, S., Astrup, A., et al (2003) PPARgamma agonists in the treatment of type II diabetes: is increased fatness commensurate with long-term efficacy? Int J Obes Relat Metab Disord. 27: 147161.
  • 20
    Bouatia-Naji, N., Meyre, D., Lobbens, S., et al (2006) ACDC/adiponectin polymorphisms are associated with severe childhood and adult obesity. Diabetes 55: 545550.
  • 21
    Ioannidis, J. P., Ntzani, E. E., Trikalinos, T. A., et al (2001) Replication validity of genetic association studies. Nat Genet. 29: 306309.