Clinical Nutrition ESPEN .sa

Clinical Nutrition ESPEN 20 (2017) 41e46

Contents lists available at ScienceDirect

Clinical Nutrition ESPEN

journal homepage:

Original article

Application of adaptive neuro-fuzzy inference systems (ANFIS) to delineate estradiol, glutathione and homocysteine interactions

Iyyapu Krishna Mohan a, Siraj Ahmed Khan a, Rachel Jacob a, Nooguri Sushma Chander b, Tajamul Hussain c, Salman A. Alrokayan d, Akella Radha Rama Devi b, Shaik Mohammad Naushad b, *

a Department of Biochemistry, Nizam's Institute of Medical Sciences, Panjagutta, Hyderabad 500082, India b Department of Biochemical Genetics, Sandor Lifesciences Pvt Ld, Banjara Hills, Hyderabad 500034, India c Center of Excellence in Biotechnology Research, King Saud University, PO Box 2455, Riyadh, 11451 Saudi Arabia d Department of Biochemistry, College of Science, King Saud University, PO Box 2455, Riyadh, 11451 Saudi Arabia

article info

Article history: Received 4 March 2017 Accepted 8 March 2017

Keywords: Hyperhomocysteinemia B12 Folate Estradiol Glutathione

summary

The rationale of the current study was to elucidate the contributing factors for the gender-based differences in total plasma homocysteine levels. A total of 413 subjects comprising of 293 men and 120 women were enrolled for the study. Chemiluminescence technology for vitamin B12, folate and total plasma homocysteine; ELISA for estradiol and 8-oxo-2-deoxyguanosine; Ellman's method for total glutathione; and PCR-RFLP analysis for the detection of methylene tetrahydrofolate reductase (MTHFR) C677T polymorphism were employed. No statistically significant differences were observed between the men and women in the distribution of age (p ? 0.82), vitamin B12 (p ? 0.23), folate (p ? 0.36) and MTHFR C677T polymorphism (p ? 0.35). However, the total plasma homocysteine levels were higher in men

compared to women (28.4 ? 17.9 vs. 20.6 ? 13.6 mmol/L, p < 0.0001). In order to explain this gender

differences in homocysteine, adaptive neuro-fuzzy inference systems (ANFIS) were developed to understand trivariate interactions among estradiol, glutathione and homocysteine. In the presence of adequate estradiol levels, inverse association was observed between glutathione and homocysteine. This association is lost when estradiol levels were inadequate. Estradiol was found to quench homocysteine mediated oxidative DNA damage. Irrespective of gender, combined deficiency of vitamin B12 and folate showed positive association with hyperhomocysteinemia and vice versa. Homocysteine reduction in response to vitamin status varied according to gender with men responding to folate and women responding to B12. To conclude, gender-differences in homocysteine are attributable estradiol mediated lowering of homocysteine that prevents inactivation of glutathione mediated oxidative defense in women.

? 2017 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Earlier, we have reported 8.6% incidence of hyperhomocysteinemia in Indian newborns [1]. Although, mild eleva-

tion of homocysteine (hcy) (15e30 mmol/L) was more prevalent in Indians, moderate to significant elevation (31e100 mmol/L) was

reported in 10.9% adults [2]. A study on Asian Indians revealed higher incidence of cobalamin deficiency in vegetarians compared

to non-vegetarians (60% vs. 39%) contributing to 2.6 mmol/L

* Corresponding author. Sandor Lifesciences Pvt Ltd, Banjara Hills, Road No 3, Hyderabad 500034, India.

elevation in homocysteine [3]. Dietary folate, B2 and B6 were shown to have inverse association with homocysteine in a South Indian study [4].

Remethylation of homocysteine to methionine requires 5methyl tetrahydrofolate as the methyl donor and methionine synthase-methionine synthase reductase (MTR-MTRR) holo enzyme complex as the catalyst and methylcobalamin as the cofactor. Deregulation in the synthesis of 5-methyl tetrahydrofolate due to C677T and A1298C polymorphisms in the methylene tetrahydrofolate reductase (MTHFR) gene was found be associated with hyperhomocysteinemia [5]. MTR A2756G and MTRR A66G polymorphisms were reported to increase homocysteine synergistically

2405-4577/? 2017 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

42

I.K. Mohan et al. / Clinical Nutrition ESPEN 20 (2017) 41e46

[6]. Hyperhomocysteinemia was reported to be associated with several disorders such as recurrent pregnancy loss [7], birth defects [8], autism [9], coronary artery disease [10], deep vein thrombosis [11], Parkinsons' disease [12].

Lower levels of follicular fluid homocysteine were shown to be associated with better pregnancy outcome [13]. In a rat model, methionine supplementation was shown to induce hyperhomocysteinemia with subsequent increase in systolic, diastolic and mean arterial pressure with an increase in sympathetic simulation and a decrease in parasympathetic modulation [14]. Homocysteine was reported to induce cardiomyocyte dysfunction through the activation of p38 MAPK, downregulation of thioredoxin and increased production of reactive oxygen species [15]. Hyperhomocysteinemia was shown to induce hypermethylation of thrombomodulin thus triggering activation of coagulation cascade [16]. Elevated lipoprotein A level was reported in patients undergoing coronary artery bypass grafting [17]. Lipoprotein A and homocysteine were reported to have significant independent associations with extracoronary atherosclerosis [18]. The coronary artery disease was reported to increase in subjects whose hs-CRP, homocysteine and fibrinogen are in highest quartiles [19].

In view of the documented evidence of lower homocysteine levels in women compared to men, we have hypothesized that estradiol might be associated with this lowering. The prima facie evidence in favor of this hypothesis is the demonstration of formation of estrogen-homocysteine conjugates, which might reduce oxidative stress by preventing homocysteine mediated inactivation of glutathione antioxidant defense [20]. To substantiate this further, we have constructed neuro fuzzy models to study the interactions among estradiol, homocysteine and glutathione in women; and also to elucidate changes in oxidative DNA damage with respect to different combinations of estradiol and homocysteine. Furthermore, the contribution of B12, folate and MTHFR C677T polymorphism towards the gender differences in homocysteine was also evaluated.

2. Materials and methods

To execute retrospective study focused on the contribution of various factors towards gender differences in homocysteine, the baseline data on the distribution of MTHFR C677T polymorphism in both genders (5.5% vs. 11.0%) was used to calculate the sample size [5]. As per the calculation, the required sample size was 271 with an alpha error of 1% and beta error of 50%. In total, we recruited 413 subjects comprising of 293 men and 120 women. The mean age of the studied population was 56.47 ? 16.0 y. The recruitment was done at the outpatient department of Nizam's Institute of Medical Sciences, Hyderabad. All the subjects are of Dravidian ethnicity. The inclusion criteria were: i) subjects in the age group of 20e80 yrs; ii) with no history of anti-folate therapy. Patients with history of any malignancy were excluded from the study. The study protocol was approved by the Institutional ethical committee of Nizam's Institute of Medical Sciences, Hyderabad. Informed consent was obtained from all the study subjects. Whole blood samples were collected in EDTA following overnight fasting. Plasma was separated immediately after centrifugation at 3000 RPM for 10 min.

2.1. Biochemical analyses

Vitamin B12, Folate, and homocysteine are performed on ADVIA Centaur? XPImmunoassay system using kits manufactured by Simens Healthcare Diagnostics Inc., U.S.A. The assays are competitive immunoassays using direct chemiluminescent technology. Glutathione was determined by Ellman's method. Plasma estradiol (Thermo Fisher Scientific, USA) and 8-oxo-2-deoxyguanosine

(Northwest Life Sciences Specialties, USA) were estimated by ELISA techniques.

2.2. Statistical analyses

Student t was performed to assess the differences between continuous variables in both genders. The data was represented in the form of mean ? standard deviation. The genetic data was segregated as categorical variable and chi-square test was used to obtain statistical significance of the distribution of this polymorphism in men and women. Folate and B12 data in men and women was segregated into tertiles and changes in the homocysteine across different tertiles was computed as mean ? standard deviation. MATLAB 7.10.0 (R2010a) version was used to generate adaptive neuro-fuzzy inference systems (ANFIS). ANFIS 1 and 2 were generated using tertiles of folate and B12 as inputs and homocysteine as output corresponding to men and women, respectively. ANFIS 3 was generated using estradiol and glutathione as inputs and homocysteine as output. ANFIS 4 was generated using estradiol and homocysteine as inputs and 8-oxo-2-deoxyguanosine as output. These models were trained based on hybrid method with grid partitioning approach to generate FIS with 0.0001 as error tolerance and 3000 epochs. The surface plots were generated based on fuzzy IF-THEN rules. All the other statistical analyses were carried out using computational webpage . A `P' value of 14.7 ng/ml and vitamin B12 > 823 pmol/ L), significant reduction in homocysteine levels were observed compared to those with vitamin status in the lowest tertile (men:

22.1 ? 15.3 mmol/L, p < 0.0001 and women: 16.5 ? 7.2 mmol/L,

p ? 0.07) (Table 2). As shown in Figs. 1 and 2, homocysteine reduction in response to vitamin status varied according to gender with men responding to folate and women responding to B12.

As shown in Fig. 3, inverse association was observed between homocysteine and glutathione when estradiol levels were adequate. As shown in Fig. 4, positive association was observed between homocysteine and 8-oxo2deoxyguanosine when estradiol levels were low. Presence of adequate estradiol was shown to reduce 8-oxodG irrespective of homocysteine elevation.

4. Discussion

In the current study, despite significant elevation of homocysteine in men compared to women, no significant differences were observed in plasma folate, vitamin B12 and MTHFR C677T. However, in both genders, combined deficiency of folate and B12 lead to hyperhomocysteinemia. Homocysteine levels were inversely associated with folate and B12 in men and women, respectively. Such gender difference in response to vitamin status was reported by

I.K. Mohan et al. / Clinical Nutrition ESPEN 20 (2017) 41e46

43

Table 1 Gender based differences in the distribution of variables.

Variable

Men (n ? 293)

Women (n ? 120)

P value

Age (yr) Vitamin B12 (pg/ml) Folate (ng/ml) Total plasma

homocysteine (mmol/L) MTHFR C677T CC-genotype CT-genotype TT-genotype

56.3 ? 16.6 799 ? 742 11.7 ? 7.5 28.4 ? 17.9

243 50 0

56.8 ? 14.5 893 ? 637 12.5 ? 7.6 20.6 ? 13.6

90 23 1

0.82 0.23 0.36 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download