Faculties of 1Nursing and 2Medicine and 5Center for Research in Medical Sciences, Autonomous University of Mexico State, Toluca, Mexico; 3Faculty of 
Medicine, Universidad de la Sabana, Chía, Colombia; 4Coordination of High Specialty Hospitals, Ministry of Health, Toluca, Mexico
*Corresponding Author’s e-mail: mezh_74@yahoo.com

العالقة بني َمْنِسب تقييم منوذج االستباب ومستويات األديبونيكنت واللبتني 
واألنسولني ومنسب كتلة اجلسم املرتبط بتعدد األشكال اجلينية عند املراهقني
 ماريا دي لو�ص أنجليس مارتينيز-مارتينيز، هوجو منديتا-زيرون، لوي�ص �شيلي�ص، كري�شتيان ليتون-توفار، روكيو توريه-جار�شيا، لورا قوتريز-بليقو،

انيدا كاملريو-رومريو، هوزي دي جي�شي�شقارديونو-قار�شيا، ماريا كاملريو-رومريو

abstract: Objectives: This study aimed to describe correlations between glucose, insulin and adipokine levels and 
the homeostasis model assessment (HOMA) index with regards to the presence/absence of fat mass and obesity-
associated (FTO) rs9939609 and peroxisome proliferator-activated receptor (PPAR)-y rs1801282 single nucleotide 
polymorphisms (SNPs) as indicators of body mass index in adolescents. Methods: This cross-sectional study was 
conducted between September and December 2016 in Toluca, Mexico. A total of 71 students between 14–18 years 
old were included. Various anthropometric and laboratory measurements were collected, including lipid profile, 
glucose, insulin and adipokine levels and HOMA index. The degree of association between variables was evaluated 
with regards to the presence/absence of the SNPs. Results: Leptin levels were significantly higher among female 
students (P = 0.001), although adiponectin levels did not differ significantly (P = 0.060). There were significant positive 
correlations between insulin levels and HOMA index with FTO (r = 0.391; P = 0.007 and r = 0.413; P = 0.005, respectively) 
and PPARγ (r = 0.529; P = 0.007 and r = 0.537; P = 0.007, respectively) SNPs. Leptin showed a significant positive correl- 
ation in the presence of PPARγ (r = 0.483; P = 0.007) or in the absence of both SNPs (r = 0.627; P = 0.039). However, 
adiponectin was significantly negatively correlated in the presence of FTO, either alone (r = −0.333; P = 0.024) or in 
combination with PPARγ (r = −0.616; P = 0.043). Conclusion: The presence of FTO and/or PPARγ SNPs might be related 
to a genetic predisposition to metabolic syndrome.

Keywords: Obesity; Body Mass Index; Single Nucleotide Polymorphisms; Fat Mass and Obesity Associated Protein, 
Human; Peroxisome Proliferator-Activated Receptor gamma; Adipokines.

ال�شتباب  منوذج  تقييم  وَمْن�ِشب  والديبونكتني،  والأن�شولني  اجللوكوز  م�شتويات  بني  العالقات  الدرا�شة  هذه  ت�شف  الهدف:  امللخ�ص: 
  )FTO rs9939609 و PPARγ rs1801282( ملعرفة وجود اأو عدم وجود تعدد اأ�شكال النوكليوتيدات املفردة ملورثات كتلة الدهن وال�شمنة
بح�شبانها موؤ�رسات لكتلة اجل�شم عند املراهقني. الطريقة: اأجريت هذه الدرا�شة امل�شتعر�شة ما بني �شبتمرب ودي�شمرب عام 2016م يف تولو�شا 
باملك�شيك، و�شارك فيها 71 طالبا يف عمر ما بني 18–14 عاما. ومت اإجراء عدد من القيا�شات الأنرثوبومرتية واملختربية �شملت م�شتويات 
الدهون واجللكوز والأن�شولني والأديبونيكنت وَمْن�ِشب تقييم منوذج ال�شتباب. ومت تقييم درجة العالقة بني تلك املتغريات ن�شبة ً اإىل وجود 
،)P = 0.001( اأو غياب تعدد يف اأ�شكال النوكليوتيدات املفردة. النتائج: كان م�شتوى تركيز اللبتني اأعلى ب�شورة ُمْعَتدٌّة عند الطالبات الإناث 
غري اأن م�شتوى الديبونكتني مل يتغري ب�شورة ُمْعَتدٌّة )P = 0.060(. وكانت هنالك عالقات اإيجابية بني م�شتويات الن�شولني وَمْن�ِشب تقييم 
منوذج ال�شتباب وبني اأ�شكال النوكليوتيدات املفردة ملورثات كتلة الدهن وال�شمنة )r = 0.413; P= 0.005 و r = 0.391; P = 0.007، على 
التوايل(، و PPARγ )r = 0.537; P = 0.007 و r = 0.529; P = 0.007، على التوايل(،  وجد هناك عالقة طردية إيجابية لهرمون الليبتني بوجود 
PPARγ (r = 0.483; P = 0.007) اأو يف غياب �شكلي النوكليوتيدات املفردة (P ;r = 0.627 = 0.039). غري اأن م�شتوى الديبونكتني كان 
.)r = -0.616; P = 0.043( PPARγ اأو يف وجود )r = -0.333; P = 0.024( يتنا�شب �شلبيا مع وجود مورثات كتلة الدهن وال�شمنة، اإما مبفرده 
الوراثي  بال�شتعداد  عالقة   PPARγ ل  املفردة  النوكليوتيدات  اأ�شكال  و/اأو  وال�شمنة  الدهن  كتلة  مورثات  لوجود  يكون  اأن  ميكن  اخلال�صة: 

لالإ�شابة باملتالزمة الأي�شية )ال�شتقالبية(.
الكلمات املفتاحية: ال�شمنة؛ موؤ�رس كتلة اجل�شم؛ تعدد اأ�شكال النوكليوتيدات املفردة؛ الربوتني املرتبط بكتله الدهن وال�شمنة، اإِْن�شان؛ PPARγ؛ الأديبوكاينز.

Correlation of the Homeostasis Model 
Assessment Index and Adiponectin, Leptin and 
Insulin Levels to Body Mass Index-Associated 

Gene Polymorphisms in Adolescents
María D. Martínez-Martínez,1 *Hugo Mendieta-Zerón,2 Luis Celis,3 Cristian F. Layton-Tovar,4 Rocío Torres-García,5 

Laura E. Gutiérrez-Pliego,5 Eneida Camarillo-Romero,5 José D. Garduño-García,5 María D. Camarillo-Romero5

clinical & basic research

Sultan Qaboos University Med J, August 2018, Vol. 18, Iss. 3, pp. e291–298, Epub. 19 Dec 18
Submitted 31 Jan 18
Revision Req. 25 Feb 18; Revision Recd. 13 Mar 18
Accepted 5 Apr 18 doi: 10.18295/squmj.2018.18.03.005

Advances in Knowledge
- Certain single nucleotide polymorphisms (SNPs) have been associated with obesity and insulin resistance. However, current data are 

conflicting regarding the roles of the fat mass and obesity-associated (FTO) rs9939609 and peroxisome proliferator-activated receptor 
(PPAR)-y rs1801282 SNPs in this regard. 



Correlation of the Homeostasis Model Assessment Index and Adiponectin, Leptin and Insulin Levels to 
Body Mass Index-Associated Gene Polymorphisms in Adolescents

e292 | SQU Medical Journal, August 2018, Volume 18, Issue 3

Obesity is a multifactorial disease deter- mined by a complex interaction between genetic and environmental factors. Previous 
research has associated an obesogenic phenotype with 
the presence of specific polymorphisms also linked to 
the development of insulin resistance, diabetes mellitus 
and metabolic syndrome.1 Moreover, pancreatic function 
may be altered by mutations that cause changes in the 
activity or expression of proteins involved in the regul- 
ation of basal energy expenditure; the latter phenomenon 
is explained in part by the various mechanisms of oxid- 
ative phosphorylation. Although the effect of each indiv- 
idual gene or combination of different genetic variants 
on metabolism and thermogenesis is uncertain, this may 
explain why certain individuals have a greater tendency 
to develop metabolic disorders.2

In recent decades, efforts have intensified in the 
search for single nucleotide polymorphisms (SNPs) caus- 
ing observed changes in the signalling pathways that 
regulate metabolic processes, leading to failures in energy 
homeostasis and predisposing individuals to diabetes 
and obesity.3 Peroxisome proliferator-activated receptors 
(PPARs) are a group of nuclear receptor proteins that 
function as transcription factors. These receptors are 
present in different metabolic pathways related to energy 
homeostasis, in addition to the lipid route. So far, three 
PPAR isoforms have been identified; the PPARγ isoform, 
which is highly expressed in white adipose tissue, is 
involved in lipid storage and energy dissipation and is 
recognised to be the master regulator of adipogenesis.4 
The fat mass and obesity-associated (FTO) gene is comp- 
osed of nine exons on chromosome 16 (16q12.2).5 It was 
the first gene to be associated with obesity in genome-
wide association studies.6 Frayling et al. observed that 
adults with the FTO rs9939609 SNP had a 1.67-fold incr- 
eased risk of obesity and weighed an average of 3–4 kg 
more than those without this polymorphism.7

In Mexico, genomic studies have demonstrated 
several genetic ancestries for the local population.8 How- 
ever, this implies that there may be discrepancies in the 
role of some SNPs in the presence of obesity-related 
conditions, such as metabolic syndrome. This study 
aimed to identify correlations between glucose, insulin 
and adipokine levels and the homeostasis model assess-
ment (HOMA) index with regards to the presence or 
absence of FTO rs9939609 and PPARy rs1801282 SNPs 
as indicators of body mass index (BMI) in Mexican 
adolescents.

Methods

This cross-sectional study was conducted from Sept- 
ember to December 2016 at the Autonomous University 
of Mexico State (UAEM) in Toluca, Mexico. All adolesc- 
ents between 14–18 years old attending the Lic. Adolfo 
López Mateos Preparatory School in Toluca were invited 
to participate in the study. The exclusion criteria incl-
uded a history of smoking, being pregnant or having been 
diagnosed with a chronic or acute disease. The sample 
size was calculated as follows:9 

where n is the infinite population of available partic-
ipants (981 adolescents), z is the 95% confidence level 
(1.96), p is the estimated percentage of the studied poly- 
morphisms in the total population (30%), ε is the margin 
of error set at 11% with a replacement rate of 9%, n' is 
the required finite sample size and N is the population 
size. Based on this, the required sample size was set at 
68 adolescents.

A body composition monitor was used to measure 
the subjects’ weight (BC-533 InnerScan Body Comp-
osition Monitor®, Tanita Corp., Tokyo, Japan), while a 
standard stadiometer was used to measure height. BMI 
was calculated as weight in kg divided by height in m2. 
According to gender- and age-specific World Health 
Organization BMI classifications for adolescents, the 
participants were categorised as either normal, over-
weight (≥1 standard deviation [SD] of the z score) or 
obese (≥2 SD of the z score).10 Waist and hip circumfer-
ences were measured using a fibreglass tape to the nearest 
0.1 cm and waist-to-height and waist-to-hip ratios calcul-
ated accordingly. Blood pressure was checked by auscult-
ation using a sphygmomanometer with an appropriately 
sized cuff. Systolic and diastolic hypertension was det-
ermined according to the criteria for high blood pressure 
in children and adolescents from the National High Blood 
Pressure Education Program Working Group.11

After a fasting period of 8 hours, 3 mL of venous 
blood were drawn from participants and collected into 
BD Vacutainer® tubes (Becton, Dickinson and Co., Fran- 
klin Lakes, New Jersey, USA). Using enzymatic methods, 
concentrations of glucose, uric acid, total cholesterol, 

- The current study found that certain adipokines were significantly correlated with FTO and PPARγ expression.

Application to Patient Care
- Determining the presence or absence of FTO and PPARy SNPs among adolescents might help in the design and implementation of more 

intensive obesity prevention strategies in this population.

z2 × p(1-p)

n
z2 × p(1-p)

ε2 N

n = 

n’ = 

ε2

1 +



María D. Martínez-Martínez, Hugo Mendieta-Zerón, Luis Celis, Cristian F. Layton-Tovar, Rocío Torres-García, 
Laura E. Gutiérrez-Pliego, Eneida Camarillo-Romero, José D. Garduño-García and María D. Camarillo-Romero

Clinical and Basic Research | e293

high-density lipoproteins (HDLs), low-density lipoproteins 
(LDLs) and triglycerides were determined according to 
the manufacturer’s instructions for each reagent assay 
kit (Randox Laboratories Ltd., Crumlin, County Antrim, 
UK). The participants’ thyroid profiles were measured 
by radioimmunoassay, whereas adiponectin, insulin and 
leptin levels were determined using the 900 series Invitr-
ogen® enzyme-linked immunosorbent assay (Thermo 
Fisher Scientific Inc., Pittsburgh, Pennsylvania, USA). 
One sample remained frozen at -70 °C until DNA extr-
action. The atherogenic index of plasma (AIP) was calcul- 
ated as total cholesterol divided by HDL. The HOMA 
index for insulin resistance was calculated as follows:12 

where fPI is fasting plasma insulin and fPG is fasting 
plasma glucose. A diagnosis of metabolic syndrome was 
based on the criteria of the International Diabetes Feder- 
ation (IDF) and adjusted for age.13 

DNA extraction was performed using the MagNA 
Pure LC 2.0 Instrument and MagNA Pure LC DNA 
Isolation Kit 1 (Roche Diagnostics GmbH, Manheim, 
Germany). Results were quantified using the N60 Nano- 
Photometer® (Implen GmbH, Munich, Germany), rep- 
orting the concentration (in μg/mL) and purity (the 
ratio of the absorbance of 260/280 nm). Genotyping 
was performed by polymerase chain reaction analysis 
in a Life ECO® thermal cycler (Bioer Technology Co. 
Ltd., Hangzhou, China). The primers and conditions 
for each polymorphism are listed in Table 1.14 Oligo-
nucleotides were designed using the PrimerQuest® 
web tool (Integrated DNA Technologies Inc., Skokie, 
Illinois, USA) and synthesised at the Institute of Bio-
technology, National Autonomous University of Mexico, 
Cuernavaca, Morelos, Mexico. The Basic Local Align- 
ment Search Tool® (National Library of Medicine, 
Bethesda, Maryland, USA) was used to verify the correct 
hybridisation. The final products were visualised in 
2% agarose gel, stained with ethidium bromide and 
digitally documented using an ultraviolet transillum-
inator system (Gel Logic 212 Pro, Carestream Health, 
Rochester, New York, USA) [Figure 1]. Sequencing 
was performed at the National Institute of Genomic 
Medicine in Mexico City to confirm the detection of 
polymorphisms as per previously described methods.14

Statistical analyses were performed using the Stat-
istical Package for the Social Sciences (SPSS), Version 
19 (IBM Inc., Armonk, New York, USA). Descriptive 
continuous data were presented as means and SDs while 
qualitative variables were expressed as percentages. 
Either the Student’s t-test or Mann-Whitney U test were 
used, depending on whether the variables were 
normally distributed. Using Pearson’s correlation coeff- 
icient, the degree of association between glucose, insulin 

and adipokine levels, AIP and HOMA index were eval- 
uated in two different settings, the first being the absence 
of either SNP and the second being the presence of either 
or both SNPs. Multivariate linear modelling was perf- 
ormed to establish the possible effect of the presence 
of polymorphisms and gender on lipid profile. A linear 
regression model weighted by gender was used to det-
ermine whether very-low-density lipoprotein (VLDL), 
LDL, HDL and triglyceride levels were predictors of 
HOMA index. A linear regression analysis was also used 

fPI × fPGHOMA index =
22.5

Figure 1: Agarose gel electrophoresis image of poly-
merase chain reaction amplification products showing 
the presence of a fat mass and obesity-associated (FTO) 
polymorphism. Lane one contains the molecular marker, 
lanes 2–8 contain amplified FTO gene DNA fragments, 
lane nine contains the positive control and lane 10 
contains the negative control.

Table 1: Primers and polymerase chain reaction cond-
itions for the studied polymorphisms14

Polymorphism

FTO PPARγ

Accession 
number*

NG_012969.1 XM_011533843.2

Variant rs9939609 rs1801282

Forward 
primer

tggctcttgaatgaatag- 
gattcagaa

ccaattcaagcccagtcctttc

Reverse 
primer

agcctctctaccatcttat- 
gtccaaaca

cagtgaaggaatcgctttccg

PCR 
conditions

Denaturation - 10 minutes at 95 °C
15 seconds at 95 °C
30 seconds at 60 °C 
30 seconds at 72 °C

Extension end cycle - 5 minutes at 72 °C 

FTO = fat mass and obesity-associated; PPAR = peroxisome 
proliferator-activated receptor; PCR = polymerase chain reaction.
*From the National Center for Biotechnology Information database.

}× 40 cycles



Correlation of the Homeostasis Model Assessment Index and Adiponectin, Leptin and Insulin Levels to 
Body Mass Index-Associated Gene Polymorphisms in Adolescents

e294 | SQU Medical Journal, August 2018, Volume 18, Issue 3

to test BMI as the dependent variable and age as the 
independent variable according to gender and weighted 
by the presence of polymorphisms. A P value of ≤0.050 
was considered statistically significant.

This study was approved by the Institutional Review 
Board of the Medical Sciences Research Center at UAEM 
(#2015/14). All subjects and their parents/guardians 
provided informed consent prior to participation in 
the study. All of the study procedures were in line 
with the ethical standards of the revised Declaration 
of Helsinki as well as the Mexican Standard for the 
Implementation of Projects of Research for Health in 
Humans (#NOM-012-SSA3-2012).

Results

A total of 71 adolescents were included in the study, 
of which 40 (56.3%) were female and 31 (43.7%) were 
male. The mean age was 15.7 ± 0.7 years and the mean 
BMI was 24.5 ± 3.8 kg/m2. Based on the IDF criteria, 
21.1% of the subjects had metabolic syndrome.13 Never-
theless, 56 adolescents (78.8%) were positive for at 
least one diagnostic criterion of metabolic syndrome, 
with high triglyceride levels being most frequently 

observed (52%). In terms of family history, 14 (19.7%) 
and 16 (22.5%) adolescents had a mother or father, 
respectively, with metabolic disease. Interesting, 39 adol- 
escents (54.9%) had relatives affected by chronic disease; 
however, none of these individuals had been diagnosed 
with hypothyroidism.

According to gender, there was a statistically signif- 
icant difference among the participants in terms of weight, 
height, waist-to-hip ratio and cholesterol, HDL, leptin and 
uric acid levels (P <0.050 each) [Table 2]. Age and leptin 
levels showed a non-parametric distribution. Although 
FTO SNPs were more frequent among females, this 
difference was not statistically significant (χ2 = 2.4671; 
P = 0.116). Similarly, there was a non-significant incr-
ease in PPARγ SNP frequency among males (χ2 = 1.2884; 
P = 0.256). There was no difference according to gender 
in terms of the presence of both SNPs (χ2 = 0.2819; 
P = 0.595) [Table 3]. According to a linear regression 
analysis weighted by gender, VLDL, HDL, LDL and tri- 
glyceride levels significantly influenced HOMA index 
(P ≤0.001). Similar results were observed among adol-
escents with FTO or PPARγ SNPs (P = 0.002) as well 
as those without PPARγ polymorphisms (P = 0.035).

Table 2: Anthropometric and laboratory variables according to gender among adolescents in Toluca, Mexico (N = 71)

Variable Mean ± SD P value

Total Females 
(n = 40)

Males 
(n = 31)

Age in years 15.7 ± 0.7 15.7 ± 0.7 15.8 ± 0.7 0.525

Height in m 1.63 ± 0.09 1.57 ± 0.06 1.70 ± 0.07 ≤0.001

Weight in kg 65.8 ± 13.8 59.4 ± 9.6 74.0 ± 14.2 ≤0.001

BMI in kg/m2 24.5 ± 3.8 23.9 ± 3.8 25.3 ± 3.7 0.111

WHR 0.86 ± 0.06 0.84 ± 0.0 0.88 ± 0.06 0.006

Adiponectin in µg/nL 12.9 ± 4.4 13.8 ± 4.2 11.8 ± 4.4 0.060

Glucose in mg/dL 91.9 ± 6.4 90.7 ± 4.9 93.4 ± 7.8 0.079

Insulin in IU/mL 14.6 ± 8.3 14.5 ± 8.2 14.8 ± 8.5 0.899

HOMA index 3.03 ± 1.83 2.96 ± 1.7 3.11 ± 1.92 0.743

Cholesterol in mg/dL 168.9 ± 37.9 179.3 ± 40.2 156.0 ± 30.7 0.010

HDL in mg/dL 42.5 ± 10.7 45.7 ± 11.1 38.67 ± 8.9 0.006

Leptin in ng/mL 18.0 ± 1.6 23.3 ± 15.7 11.0 ± 14.0 0.001

LDL in mg/dL 107.1 ± 31.8 112.0 ± 34.0 100.9 ± 28.2 0.149

Triglycerides in mg/dL 131.8 ± 72.1 131.2 ± 63.3 132.5 ± 82.9 0.939

Uric acid in mg/dL 4.4 ± 1.4 3.6 ± 1.0 5.3 ± 1.3 ≤0.001

VLDL in mg/dL 19.2 ± 12.2 21.5 ± 12.5 16.3 ± 11.3 0.080

AIP 4.0 ± 0.9 4.0 ± 1.0 4.1 ± 0.8 0.743

SD = standard deviation; BMI = body mass index; WHR = waist-to-hip ratio; HOMA = homeostatic model assessment; HDL = high-density 
lipoprotein; LDL = low-density lipoprotein; VLDL = very-low-density lipoprotein; AIP = atherogenic index of plasma.



María D. Martínez-Martínez, Hugo Mendieta-Zerón, Luis Celis, Cristian F. Layton-Tovar, Rocío Torres-García, 
Laura E. Gutiérrez-Pliego, Eneida Camarillo-Romero, José D. Garduño-García and María D. Camarillo-Romero

Clinical and Basic Research | e295

Overall, the PPARγ SNP was present in 25 adol-
escents (35.2%) and the FTO SNP was present in 46 
(64.8%), including 11 cases (15.5%) in which both PPARγ 
and FTO SNPs were present. Neither SNP was present in 
11 individuals (15.5%). Insulin levels and HOMA index 
were statistically correlated with the presence of FTO 
(P = 0.007 and 0.005, respectively), PPARγ (P = 0.007 
each) and both SNPs combined (P = 0.039 and 0.029, 
respectively). While leptin showed a positive significant 
correlation in the presence of PPARγ (r = 0.483; P = 0.014) 
and in the absence of either SNP (r = 0.627; P = 0.039), 
adiponectin was significantly negatively correlated with 
FTO, either alone (r = −0.333; P = 0.024) or in combin-
ation with PPARγ (r = −0.616; P = 0.043) [Table 4]. A 
multivariate linear model showed that the presence of 
the SNPs were not determinants of BMI and lipid profile. 
However, the effect of gender was significant (P <0.050), 
particularly in the setting of both SNPs together (P = 0.021) 
[Table 5].

Discussion

Worldwide, various studies of different ethnic popul-
ations have shown conflicting results regarding the 
relationship between body weight and FTO and PPARγ 
polymorphisms. For example, the PPARγ Pro12Ala 
rs1801282 polymorphism has been associated with 
obesity and insulin resistance among Asian Indians; 
however, in a Chinese sample, it was reported to be a 
protective factor against type 2 diabetes and obesity.15,16 
Huang et al. found that the presence of the rs2282440-
SDC3T/T genotype alongside the rs1801282 PPARγ2 
carrier genotype was strongly associated with obesity.17 
On the other hand, Queiroz et al. found that PPARγ 
rs1801282 conferred a higher risk of altered insulin 
levels and HOMA index for insulin resistance among 
overweight adolescents.18 To some extent, the latter obs- 
ervation is consistent with the results of the present 
study.

A previous study demonstrated that circulating lipids 
in Mexican children modified the association between 
the PPARγ2 rs1801282 genotype and insulin resistance.19 
In the current study, significant associations were noted 
between HOMA index and the presence of FTO or 
PPARγ polymorphisms, taking into account the role of 
lipids and weighted by gender. In Asiatic populations, 
FTO has been associated with an increased risk of 
obesity and type 2 diabetes.20 Saldaña-Alvarez et al. 
found that FTO SNPs made differential contributions to 
obesity risk, supporting the hypothesis that mechanisms 
involving these variants are gender-dependent and that 
these changes may contribute to disease development.21 
Although the researchers evaluated different FTO SNPs 
(rs1121980, rs17817449, rs3751812, rs9930506 and 
rs17817449) to that of the present study (rs9939609), 
such findings confirm the possibility that gender is a risk 

Table 4: Correlations between glucose, insulin and adipokine levels and homeostasis model assessment index with 
body mass index-related polymorphisms among adolescents in Toluca, Mexico (N = 71)

Variable Polymorphisms 

Neither FTO nor PPARγ 
(n = 11)

FTO 
(n = 46)*

PPARγ 
(n = 25)*

Both FTO and PPARγ 
(n = 11)

r P value r P value r P value r P value

Glucose 0.303 0.365 0.272 0.070 0.090 0.675 0.292 0.413

Insulin 0.600 0.051 0.391 0.007 0.529 0.007 0.627 0.039

HOMA index 0.600 0.051 0.413 0.005 0.537 0.007 0.685 0.029

Leptin 0.627 0.039 0.248 0.096 0.483 0.014 0.221 0.513

Adiponectin −0.323 0.332 −0.333 0.024 −0.110 0.599 −0.616 0.043

AIP 0.721 0.012 0.463 0.001 0.192 0.368 0.487 0.154

FTO = fat mass and obesity-associated; PPAR = peroxisome proliferator-activated receptor; HOMA = homeostasis model assessment; AIP = ather-
ogenic index of plasma.
*Including the 11 adolescents with both polymorphisms.

Table 3: Frequency of body mass index-related polymor- 
phisms according to gender among adolescents in Toluca, 
Mexico (N = 71)

Polymorphism n (%) P 
value

Total Females 
(n = 40)

Males 
(n = 31)

Neither FTO nor 
PPARγ

11 
(15.5)

4 
(10)

7 
(22.6)

0.146

FTO alone 35 
(49.3)

23 
(57.5)

12 
(38.7)

0.116

PPARγ alone 14 
(19.7)

6 
(15)

8 
(25.8)

0.256

Both FTO and 
PPARγ

11 
(15.5)

7 
(17.5)

4 
(12.9)

0.595

FTO = fat mass and obesity-associated; PPAR = peroxisome proliferator 
-activated receptor.



Correlation of the Homeostasis Model Assessment Index and Adiponectin, Leptin and Insulin Levels to 
Body Mass Index-Associated Gene Polymorphisms in Adolescents

e296 | SQU Medical Journal, August 2018, Volume 18, Issue 3

factor for obesity.17 Similarly, these results are supported 
by those of Huđek et al., who found a statistically signif- 
icant positive correlation between frequencies of high-

risk FTO genotypes in obese women (AA rs9939609, 
CC rs1421085 and GG rs17817449).22 

Díaz-Anzaldúa et al. reported that differences in 
mean BMI levels among Mexican patients with bipolar 
disorder were explained by the presence of FTO 
rs8050136 and rs9939609 genotypes.23 Another study 
showed that the locus of the FTO gene was significantly 
associated with increased BMI in indigenous Mexican 
populations.24 Muñoz-Yáñez et al. found that FTO 
rs9939609 polymorphisms were associated with obesity- 
related traits, including elevated BMI, cholesterol and 
LDL levels, tricep skinfolds and an increased waist 
circumference and waist-to-height ratio.25 In the present 
study, an extremely high percentage of adolescents had 
FTO SNPs (64.8%). Given a population of approxim-
ately nine million Mexicans aged 15–20 years old, there 
are an estimated 5,830,200 young people with FTO 
rs9939609 polymorphisms and, therefore, a corresp-
onding risk of obesity.26 

Such findings are alarming for the local healthcare 
system. Fortunately, despite a genetic predisposition 
influenced by polymorphisms, obesity is a modifiable 
condition, even with a positive genetic profile risk score.27

In the PREDIMED-NAVARRA randomised trial, PPARγ 
Pro12Ala rs1801282 carriers exhibited lower telomere 
shortening compared to those with the Pro/Pro geno-
type after five years of adherence to a Mediterranean-
style diet.28 Unfortunately, the molecular effects of a 
traditional Mexican diet have yet to be fully elucidated.

In the current study, mean HDL levels were 
significantly higher among females. However, this may 
be because the mean level for males was below recomm- 
endations reported for the Mexican population.29 In 
addition, female students also had significantly higher 
leptin levels. Although leptin is a significant predictor of 
carotid intima media thickness, previous research has 
confirmed that men suffer from higher cardiovascular-
related mortality compared to women.30,31 Such findings 
undermine the epidemiological evidence for utilising 
leptin as a prognostic tool for cardiovascular disease 
in adolescents. 

As in previous research, females in the current 
study also demonstrated significantly higher cholesterol 
levels.32 A strong correlation has been established bet- 
ween cholesterol and visceral adipose tissue as quantified 
by dual-energy X-ray absorptiometry.33 Unfortunately, 
this technique was not utilised in the present study. 
Finally, the frequency of metabolic disease among the 
parents of the studied adolescents was lower than exp- 
ected; instead, students’ relatives were found to have 
a high rate of chronic disease. However, it is unclear 
whether these individuals had lipid-related abnormalities.

It is worth noting that various researchers have 
recommended lowering BMI cut-off values for over-

Table 5: Multivariate linear model for body mass index-
related polymorphisms and gender as determinants for 
body mass index and lipid profile

Variable Statistical test Value* F 
value

P 
value

FTO Pillai’s trace 0.066 1.027 0.401

Wilks’ lambda 0.934 1.027 0.401

Lawley-Hotelling 
trace

0.071 1.027 0.401

Roy’s largest root 0.071 1.027 0.401

PPARγ Pillai’s trace 0.107 1.738 0.154

Wilks’ lambda 0.893 1.738 0.154

Lawley-Hotelling 
trace

0.120 1.738 0.154

Roy’s largest root 0.120 1.738 0.154

Gender Pillai’s trace 0.162 2.797 0.034

Wilks’ lambda 0.838 2.797 0.034

Lawley-Hotelling 
trace

0.193 2.797 0.034

Roy’s largest root 0.193 2.797 0.034

FTO × 
PPARγ

Pillai’s trace 0.105 1.697 0.163

Wilks’ lambda 0.895 1.697 0.163

Lawley-Hotelling 
trace

0.117 1.697 0.163

Roy’s largest root 0.117 1.697 0.163

FTO × 
gender

Pillai’s trace 0.040 0.607 0.659

Wilks’ lambda 0.960 0.607 0.659

Lawley-Hotelling 
trace

0.042 0.607 0.659

Roy’s largest root 0.042 0.607 0.659

PPARγ × 
gender

Pillai’s trace 0.032 0.477 0.752

Wilks’ lambda 0.968 0.477 0.752

Lawley-Hotelling 
trace

0.033 0.477 0.752

Roy’s largest root 0.033 0.477 0.752

FTO × 
PPARγ × 
gender

Pillai’s trace 0.178 3.143 0.021

Wilks’ lambda 0.822 3.143 0.021

Lawley-Hotelling 
trace

0.217 3.143 0.021

Roy’s largest root 0.217 3.143 0.021

FTO = fat mass and obesity-associated; PPARγ = peroxisome proliferator 
-activated receptor.
*The degrees of freedom of hypothesis and error were 4.000 and 58.000, 
respectively, for each value.



María D. Martínez-Martínez, Hugo Mendieta-Zerón, Luis Celis, Cristian F. Layton-Tovar, Rocío Torres-García, 
Laura E. Gutiérrez-Pliego, Eneida Camarillo-Romero, José D. Garduño-García and María D. Camarillo-Romero

Clinical and Basic Research | e297

weight and obese categories in Mexican and Asiatic 
populations.34,35 Thus, if a BMI of ≥27 kg/m2 were cons- 
idered to indicate obesity, 31% of the adolescents in 
the current study would have been considered obese. 
Males were significantly heavier than females; this was 
accordingly reflected by slight elevations in glucose 
and insulin levels and HOMA index. However, these 
differences were not significant, possibly due to simil-
arities in BMI in both genders. Therefore, further studies 
are recommended including other variables which could 
influence BMI in adolescents, such as physical activity 
and dietary habits.36 Another limitation of this study 
was the small sample size; nevertheless, investigations 
regarding associations between polymorphisms and 
clinical parameters for metabolic syndrome or obesity 
can be clinically significant in small samples.37 

Conclusion

The results of this study suggest significant positive 
correlations between insulin levels and HOMA index 
with BMI-related FTO rs9939609 and/or PPARγ rs180- 
1282 polymorphisms. Future epidemiological studies 
are recommended to determine the net causative effect 
of such gene variants so as to halt the development of 
a pandemic of obesity-related health problems. Although 
the main causes of obesity, diabetes and metabolic syndr- 
ome are lifestyle factors, the role of such polymorphisms 
could help to explain why some individuals have a greater 
tendency to develop metabolic disorders and are more 
resistant to weight control interventions than others.

c o n f l i c t o f i n t e r e s t
The authors declare no conflicts of interest.

f u n d i n g

This study was funded with the aid of a grant from 
the Mexican Ministry of Education (grant #PROMEP/ 
2013/CA-186/103105/13/9057).

a c k n o w l e d g e m e n t s

The preliminary version of this study was submitted 
as a M.Sc. thesis to the UAEM in 2016. This version is 
available on the university website.

References
1. Kasim NB, Huri HZ, Vethakkan SR, Ibrahim L, Abdullah BM. 

Genetic polymorphisms associated with overweight and obesity 
in uncontrolled type 2 diabetes mellitus. Biomark Med 2016; 
10:403–15. doi: 10.2217/bmm-2015-0037.

2. Kong X, Zhang X, Xing X, Zhang B, Hong J, Yang W. The 
association of type 2 diabetes loci identified in genome-wide 
association studies with metabolic syndrome and its compo-
nents in a Chinese population with type 2 diabetes. PLoS One 
2015; 10:e0143607. doi: 10.1371/journal.pone.0143607.

3. Miranda-Lora AL, Cruz M, Aguirre-Hernández J, Molina-Díaz M, 
Gutiérrez J, Flores-Huerta S, et al. Exploring single nucleotide 
polymorphisms previously related to obesity and metabolic traits 
in pediatric-onset type 2 diabetes. Acta Diabetol 2017; 54:653–62. 
doi: 10.1007/s00592-017-0987-9.

4. Ali AT, Hochfeld WE, Myburgh R, Pepper MS. Adipocyte and 
adipogenesis. Eur J Cell Biol 2013; 92:229–36. doi: 10.1016/j.ejcb. 
2013.06.001.

5. Speakman JR. The ‘fat mass and obesity related’ (FTO) gene: Mech- 
anisms of impact on obesity and energy balance. Curr Obes Rep 
2015; 4:73–91. doi: 10.1007/s13679-015-0143-1.

6. Loos RJ, Bouchard C. FTO: The first gene contributing to common 
forms of human obesity. Obes Rev 2008; 9:246–50. doi: 10.1111/ 
j.1467-789X.2008.00481.x.

7. Frayling TM, Timpson NJ, Weedon MN, Zeggini E, Freathy RM, 
Lindgren CM, et al. A common variant in the FTO gene is assoc- 
iated with body mass index and predisposes to childhood and adult 
obesity. Science 2007; 316:889–94. doi: 10.1126/science.1141634.

8. Rangel-Villalobos H, Muñoz-Valle JF, González-Martín A, 
Gorostiza A, Magaña MT, Páez-Riberos LA. Genetic admixture, 
relatedness, and structure patterns among Mexican populations 
revealed by the Y-chromosome. Am J Phys Anthropol 2008; 
135:448–61. doi: 10.1002/ajpa.20765.

9. Camacho-Sandoval J. [Sample size in clinical studies]. Acta Med 
Costarric 2008; 50:20–21.

10. de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, 
Siekmann J. Development of a WHO growth reference for school 
-aged children and adolescents. Bull World Health Organ 2007; 
85:660–7. doi: 10.2471/BLT.07.043497.

11. National High Blood Pressure Education Program Working Group. 
The fourth report on the diagnosis, evaluation, and treatment of 
high blood pressure in children and adolescents. Pediatrics 2004; 
114:555–76.

12. Salgado AL, Carvalho LD, Oliveira AC, Santos VN, Vieira JG, 
Parise ER. Insulin resistance index (HOMA-IR) in the differ-
entiation of patients with non-alcoholic fatty liver disease and 
healthy individuals. Arq Gastroenterol 2010; 47:165–9.

13. International Diabetes Federation. IDF consensus definition of 
metabolic syndrome in children and adolescents. From: www.
idf.org/e-library/consensus-statements/61-idf-consensus-defin 
ition-of-metabolic-syndrome-in-children-and-adolescents  
Accessed: Mar 2018.

14. Martínez-Martínez MD. [Association between molecular mar- 
kers and obesity in adolescents of the upper level of the UAEMéx]. 
MSc Thesis, 2016, Autonomous University of Mexico State, Toluca, 
Mexico.

15. Bhatt SP, Misra A, Sharma M, Luthra K, Guleria R, Pandey RM, 
et al. Ala/Ala genotype of Pro12Ala polymorphism in the per-
oxisome proliferator-activated receptor-γ2 gene is associated with 
obesity and insulin resistance in Asian Indians. Diabetes Technol 
Ther 2012; 14:828–34. doi: 10.1089/dia.2011.0277.

16. Wang X, Liu J, Ouyang Y, Fang M, Gao H, Liu L. The association 
between the Pro12Ala variant in the PPARγ2 gene and type 2 
diabetes mellitus and obesity in a Chinese population. PLoS 
One 2013; 8:e71985. doi: 10.1371/journal.pone.0071985.

17. Huang WH, Hwang LC, Chan HL, Lin HY, Lin YH. Study of seven 
single-nucleotide polymorphisms identified in East Asians for 
association with obesity in a Taiwanese population. BMJ Open 
2016; 6:e011713. doi: 10.1136/bmjopen-2016-011713.

18. Queiroz EM, Cândido AP, Castro IM, Bastos AQ, Machado-
Coelho GL, Freitas RN. IGF2, LEPR, POMC, PPARG, and PPA-
RGC1 gene variants are associated with obesity-related risk 
phenotypes in Brazilian children and adolescents. Braz J Med 
Biol Res 2015; 48:595–602. doi: 10.1590/1414-431X20154155. 

19. Stryjecki C, Peralta-Romero J, Alyass A, Karam-Araujo R, Suarez F, 
Gomez-Zamudio J, et al. Association between PPAR-γ2 Pro12Ala 
genotype and insulin resistance is modified by circulating lipids 
in Mexican children. Sci Rep 2016; 6:24472. doi: 10.1038/srep24472. 

https://doi.org/10.2217/bmm-2015-0037
https://doi.org/10.1371/journal.pone.0143607
https://doi.org/10.1007/s00592-017-0987-9
https://doi.org/10.1016/j.ejcb.2013.06.001
https://doi.org/10.1016/j.ejcb.2013.06.001
https://doi.org/10.1007/s13679-015-0143-1
https://doi.org/10.1111/j.1467-789X.2008.00481.x
https://doi.org/10.1111/j.1467-789X.2008.00481.x
https://doi.org/10.1126/science.1141634
https://doi.org/10.1002/ajpa.20765
https://doi.org/10.2471/BLT.07.043497
https://doi.org/10.1089/dia.2011.0277
https://doi.org/10.1371/journal.pone.0071985
https://doi.org/10.1136/bmjopen-2016-011713
https://doi.org/10.1590/1414-431X20154155
https://doi.org/10.1038/srep24472


Correlation of the Homeostasis Model Assessment Index and Adiponectin, Leptin and Insulin Levels to 
Body Mass Index-Associated Gene Polymorphisms in Adolescents

e298 | SQU Medical Journal, August 2018, Volume 18, Issue 3

20. Li H, Kilpeläinen TO, Liu C, Zhu J, Liu Y, Hu C, et al. Association 
of genetic variation in FTO with risk of obesity and type 2 diab-
etes with data from 96,551 East and South Asians. Diabetologia 
2012; 55:981–95. doi: 10.1007/s00125-011-2370-7.

21. Saldaña-Alvarez Y, Salas-Martínez MG, García-Ortiz H, Luckie- 
Duque A, García-Cárdenas G, Vicenteño-Ayala H, et al. Gender- 
dependent association of FTO polymorphisms with body mass 
index in Mexicans. PLoS One 2016; 11:e0145984. doi: 10.1371/
journal.pone.0145984. 

22. Huđek A, Škara L, Smolkovič B, Kazazić S, Ravlić S, Nanić L, 
et al. Higher prevalence of FTO gene risk genotypes AA rs99- 
39609, CC rs1421085, and GG rs17817449 and saliva containing 
Staphylococcus aureus in obese women in Croatia. Nutr Res 
2018; 50:94–103. doi: 10.1016/j.nutres.2017.12.005.

23. Díaz-Anzaldúa A, Ocampo-Mendoza Y, Hernández-Lagunas JO, 
Díaz-Madrid FA, Romo-Nava F, Juárez-García F, et al. Differ- 
ences in body mass index according to fat mass- and obesity-
associated (FTO) genotype in Mexican patients with bipolar 
disorder. Bipolar Disord 2015; 17:662–9. doi: 10.1111/bdi.12328.

24. León-Mimila P, Villamil-Ramírez H, Villalobos-Comparán M, 
Villarreal-Molina T, Romero-Hidalgo S, López-Contreras B, et al. 
Contribution of common genetic variants to obesity and obesity- 
related traits in Mexican children and adults. PLoS One 2013; 
8:e70640. doi: 10.1371/journal.pone.0070640. 

25. Muñoz-Yáñez C, Pérez-Morales R, Moreno-Macías H, Calleros-
Rincón E, Ballesteros G, González RA, et al. Polymorphisms 
FTO rs9939609, PPARG rs1801282 and ADIPOQ rs4632532 and 
rs182052 but not lifestyle are associated with obesity related-
traits in Mexican children. Genet Mol Biol 2016; 39:547–53. 
doi: 10.1590/1678-4685-GMB-2015-0267. 

26. National Institute of Statistics and Geography. Population. 
From: www.beta.inegi.org.mx/temas/estructura/  Accessed: Mar 
2018.

27. Celis-Morales CA, Lyall DM, Gray SR, Steell L, Anderson J, 
Iliodromiti S, et al. Dietary fat and total energy intake modifies 
the association of genetic profile risk score on obesity: Evidence 
from 48 170 UK Biobank participants. Int J Obes (Lond) 2017; 
41:1761–8. doi: 10.1038/ijo.2017.169.

28. García-Calzón S, Martínez-González MA, Razquin C, Corella D, 
Salas-Salvadó J, Martínez JA, et al. Pro12Ala polymorphism of 
the PPARγ2 gene interacts with a Mediterranean diet to prevent 
telomere shortening in the PREDIMED-NAVARRA random-
ized trial. Circ Cardiovasc Genet 2015; 8:91–9. doi: 10.1161/CIRC 
GENETICS.114.000635.

29. Bibiloni MD, Salas R, De la Garza YE, Villarreal JZ, Sureda A, 
Tur JA. Serum lipid profile, prevalence of dyslipidaemia, and 
associated risk factors among Northern Mexican adolescents. 
J Pediatr Gastroenterol Nutr 2016; 63:544–9. doi: 10.1097/
MPG.0000000000001325.

30. Velarde GP, Sherazi S, Kraemer DF, Bravo-Jaimes K, Butterfield R, 
Amico T, et al. Clinical and biochemical markers of cardio-
vascular structure and function in women with the metabolic 
syndrome. Am J Cardiol 2015; 116:1705–10. doi: 10.1016/j.amj 
card.2015.09.010.

31. Bots SH, Peters SA, Woodward M. Sex differences in coronary 
heart disease and stroke mortality: A global assessment of the 
effect of ageing between 1980 and 2010. BMJ Glob Health 2017; 
2:e000298. doi: 10.1136/bmjgh-2017-000298.

32. Lotufo PA, Santos RD, Sposito AC, Bertolami M, Rocha-Faria J Neto, 
Izar MC, et al. Self-reported high-cholesterol prevalence in the 
Brazilian population: Analysis of the 2013 National Health Survey. 
Arq Bras Cardiol 2017; 108:411–16. doi: 10.5935/abc.20170055.

33. Miazgowski T, Kucharski R, Sołtysiak M, Taszarek A, Miazgowski B, 
Widecka K. Visceral fat reference values derived from healthy 
European men and women aged 20-30 years using GE Healthcare 
dual-energy X-ray absorptiometry. PLoS One 2017; 12:e0180614. 
doi: 10.1371/journal.pone.0180614.

34. Mexico Ministry of Health. [NORMA Official Mexican NOM-
174-SSA1-1998: For the integral management of obesity]. From: 
www.salud.gob.mx/unidades/cdi/nom/174ssa18.html  Accessed: 
Mar 2018.

35. Ko GT, Chan JC, Cockram CS, Woo J. Prediction of hyper-
tension, diabetes, dyslipidaemia or albuminuria using simple 
anthropometric indexes in Hong Kong Chinese. Int J Obes Relat 
Metab Disord 1999; 23:1136–42. doi: 10.1038/sj.ijo.0801043.

36. Liu W, He MZ, Wang Y, Wang Y, Zhou Y, Wu M, et al. 
Differences in health-related behaviors between middle school, 
high school, and college students in Jiangsu province, China. 
Asia Pac J Clin Nutr 2017; 26:731–7. doi: 10.6133/apjcn.072016.06.

37. Saldaña-Rivera E, Careaga-Castilla MJ, Olvera-Cárdenas GD, 
Pérez-Soto E, Sánchez-Monroy V. Mitochondrial T16189C poly- 
morphism is associated with metabolic syndrome in the Mexican 
population. Dis Markers 2018; 2018:3981315. doi: 10.1155/2018/ 
3981315. 

https://doi.org/10.1007/s00125-011-2370-7
https://doi.org/10.1371/journal.pone.0145984
https://doi.org/10.1371/journal.pone.0145984
https://doi.org/10.1016/j.nutres.2017.12.005
https://doi.org/10.1111/bdi.12328
https://doi.org/10.1371/journal.pone.0070640
https://doi.org/10.1590/1678-4685-GMB-2015-0267
https://doi.org/10.1038/ijo.2017.169
https://doi.org/10.1161/CIRCGENETICS.114.000635
https://doi.org/10.1161/CIRCGENETICS.114.000635
https://doi.org/10.1097/MPG.0000000000001325
https://doi.org/10.1097/MPG.0000000000001325
https://doi.org/10.1016/j.amjcard.2015.09.010
https://doi.org/10.1016/j.amjcard.2015.09.010
https://doi.org/10.1136/bmjgh-2017-000298
https://doi.org/10.5935/abc.20170055
https://doi.org/10.1371/journal.pone.0180614
https://doi.org/10.1038/sj.ijo.0801043
https://doi.org/10.6133/apjcn.072016.06
https://doi.org/10.1155/2018/3981315
https://doi.org/10.1155/2018/3981315