Differences in gene expression related to the outcomes of obesity treatment , peak oxygen uptake , and fatty acid metabolism measured in a cardiopulmonary exercise test

280 environmental risk.1 Interestingly, the prevalence of obesity has increased in almost all countries.2 Between 1980 and 2016, the prevalence among men has increased from 28.8% to 39% (boys, 8.1% to 19%), and among women, from 29.8% to 40% (girls, 8.4% to 18%). The worldwide prevalence of obesity has tripled since 1975.2 Global data are similar to the epidemiology of the disease report‐ ed in Poland, where 33.4% of the population are obese and 36.7%, overweight.3 The comparison of INTRODUCTION Obesity has become a pandem‐ ic of modern societies. The World Health Orga‐ nization defines obesity as abnormal and exces‐ sive accumulation of fat tissue that presents a risk to health. As a complex systemic disease, it leads to metabolic, endocrine, and cardiopulmonary dysfunction of multiple organs. The main cause of obesity is excessive caloric intake and limit‐ ed physical activity. Recent data on air pollution and metabolic disruptors indicate an additional ORIGINAL ARTICLE

environmental risk. 1 Interestingly, the prevalence of obesity has increased in almost all countries. 2etween 1980 and 2016, the prevalence among men has increased from 28.8% to 39% (boys, 8.1% to 19%), and among women, from 29.8% to 40% (girls, 8.4% to 18%).The worldwide prevalence of obesity has tripled since 1975. 2 Global data are similar to the epidemiology of the disease reported in Poland, where 33.4% of the population are obese and 36.7%,overweight. 3The comparison of Additionally, the results of the obesity treatment using lifestyle modification differ significantly.We hypothesized that changes in gene expression observed in peripheral blood cells related to treatment outcomes may be similar to the changes related to different CPET results, which may help elucidate why some patients do not benefit from the treatment.
The aim of the study was to compare gene expression in lean and obese patients, to examine differences in gene expression related to CPET results, and to evaluate changes in gene expression related to results of obesity treatment in an intensive lifestyle modification program.

PATIENTS AND METHODS
Obese patients treated at the Department of Clinical Nutrition, Medical University of Gdańsk, Poland, who were excluded from bariatric treatment received an option of an intensive lifestyle modification program including diet, physical exercise, and behavioral therapy.The study was performed in the years from 2012 to 2014.The inclusion criteria were body mass index of 30 kg/m 2 or higher, age of 18 years or older, and lack of contraindications to diet treatment, physical rehabilitation, and exercise.The exclusion criteria were neoplastic, autoimmune, and active infectious diseases and chronic heart, liver, and renal failure.The following diseases were considered a contraindication to the treatment: heart, respiratory, renal, and liver insufficiency, other neoplastic or chronic diseases leading to an impaired ability for physical exercise, pregnancy, or lack of compliance.
The size of the study group was assessed using the Altman nomogram 25 with an assumption that gene expression profile characteristics for obesity may be found in 90% of obese and 10% of lean patients, with a test power of 0.8 and a P value of less than 0.05.The estimated number of studied patients was above 12.
The treatment protocol consisted of diet intervention, exercise, and behavioral therapy.Diet intervention was based on the body composition analyzed by bioelectric impedance, resting metabolic rate, and subjective patient preferences.The intensity of the exercise was tailored according to the CPET results, mainly VO 2max , FAT g/h, and FAT HR.Behavioral intervention was focused on the diagnosis of health problems leading to obesity, lifestyle modification, training in self -assessment, and development of healthy habits.The intensive treatment lasted for 12 weeks and consisted of consultations with a physician, dietitian, and medical rehabilitation specialist.The CPET and bioelectric impedance were performed before and after the treatment.The first part of the exercise, which lasted 6 weeks, was performed at the outpatient department of medical rehabilitation, while further rehabilitation treatment was performed by patients at home.
The CPET was performed with a LodeCorival cycloergometer (Lode B.V., Groningen, the results of the Polish studies WOBASZ (conducted in 2003-2005) and WOBASZ II (conducted in 2013-2014) showed a shift of the body mass index category to higher values in both sexes and an increase in obesity rates among men. 4 Obesity is related to an increased risk of several diseases, the most serious being coronary artery disease, stroke, and cancer, which lead to increased mortality in obese patients. 5-7 The magnitude of the obesity pandemic requires development of methods that could be used on a large scale, which is difficult to achieve with bariatric treatment.The CPET is an exercise test combined with an analysis of the respiratory gas exchange concentration.The test measures the response and function of the circulatory, respiratory, and neurologic systems, as well as skeletal muscles and metabolism during physical exercise.The most important parameter analyzing fat metabolism is fatty acid metabolism (FAT) expressed as g/h.It measures the maximal fat mass, which is metabolized over a certain time and which differs according to sex and exercise intensity.Moreover, it is a patient -specific value. 16,17FAT heart rate (HR) is a parameter that enables an assessment of the heart rate at which fat metabolism is maximal.FAT HR is approximately in the mean (SD) range of 58% (3%) of the maximal heart rate.It occurs before the anaerobic threshold point, where anaerobic metabolism dominates aerobic fat metabolism. 16The analysis of gene expression enables an assessment of the functioning of transcripts and their role in biological processes.The whole genome expression enables a study of the function of the most important transcripts. This may be explained by systemic inflammation observed in obesity, which is related to systemic inflammation influencing the function of muscles in fat metabolism. 20,21ombining CPET and gene expression assessment allows a unique insight into obesity pathology and treatment mechanisms, leading to an optimized treatment strategy.The CPET indicates the systemic differences in response to exercise.The observed changes in the CPET may be correlated with differences in gene expression.
ad IJssel, the Netherlands) was used for amplification and labelling of RNA.For each sample, 200 ng of RNA was used.Human HT -12_V3_ expression arrays (Illumina, San Diego, United States) were processed according to the manufacturer's protocol.Slides were scanned immediately using Illumina BeadStation iScan (Illumina, San Diego, California, United States).
Image and data analysis First -line check, background correction, and quantile normalization of the data was performed with Genomestudio Gene Expression Analysis module v 1.0.6 Statistics (Illumina, San Diego, California, United States).Entities containing at least 75% of samples with the signal intensity value above the 20th percentile in 100% of the samples in at least 2 groups were included for further analysis.
Data analysis was performed using Gene-Spring package version 13.0.0(Agilent Technologies Santa Clara, California, United States).Genes with a significantly different expression between compared groups were chosen based on a log 2 -fold change of more than 2 in gene expression and a t-test P value of less than 0.05 corrected for multiple testing by Benjamin-Hochberg false discovery rates of less than 0.01.Functional annotation of genes was described using the Go Process analysis and KEGG pathways with a Genecodis functional annotation web -based tool.
Clinical data for this study were analyzed with Statistica 12.0 (StatSoft, Tulsa, Oklahoma, United States).

RESULTS
The study group included 16 participants: 9 obese patients and 7 healthy controls.The demographic data are presented in (TABLE 1).
We analyzed the gene expression profile of obese and lean patients.Additionally, it was analyzed according to the outcome of treatment: participants who gained at least 10% reduction of their body mass were compared with those who did not achieve such a result.Furthermore, the results were analyzed depending on the CPET results (FAT ≥20 g/h and VO 2max ≥1.7 ml/min).The cutoff points were selected according to the results in the study group (TABLE 1) and a previous study. 26In most patients, FAT values were lower than 20 g/h (mean [SD] FAT, 13.65 [7] g/h) and VO 2max was lower than 1.7 ml/min (mean [SD], 1.68 [0.41] g/h).In the control group, the results of FAT typically found in obese individuals were observed in 2 participants, while 1 obese patient had the results typical for healthy controls.VO 2max values were typical for healthy controls in 4 obese patients, while 1 healthy individual had the results typically observed for obese patients.Thus, the comparison was made according to the estimated cutoff points for FAT and VO 2max .
A total of 47 323 transcripts were analyzed, of which 32 379 had a high quality of RNA and were selected for further analysis.the Netherlands).The results were analyzed with Metasoft 3.9 CortexBiophisik software (Cortex-Biophisik GmbH, Leipzig, Germany).Blood pressure was measured with SunTech Tango (Sun Tech Medical, Morrisville, NC, United States), and electrocardiogram with CardioDirect 12.
The study was approved by the Ethics Committee of the Medical University of Gdańsk, (NKEB/41/2010 and NKEBN/151/2010).
Collection of blood samples In all patients, RNA was isolated from the whole blood, using PAXgene Blood RNA Tubes (Qiagen, Venlo, The Netherlands).All tubes were immediately frozen and stored at -20ºC until RNA isolation (maximal period, 2 months).RNA was isolated using PAXgene Blood RNA Kit CE (Qiagen, Venlo, The Netherlands).All RNA samples were stored at -80ºC until labelling and hybridization.
Quality and concentration of RNA were determined using 2100 Bioanalyzer (Agilent, Amstelveen, the Netherlands) with the use of RNA 6000 Nano Kit (Agilent).Samples with the RNA integrity number exceeding 7.5 were used for further analysis on expression arrays.

Gene expression
Illumina TotalPrep 96 RNA Amplification Kit (Applied Biosystems, Nieuwerkerk Differences between obese patients and controls: a P = 0.01; b P = 0.005 The differences were not significant between patients who obtained and who did not obtain >10% reduction of the initial weight. Abbreviations: FAT, fatty acid metabolism; VO 2max , maximal oxygen uptake during exercise training of maximal intensity indicated significant results for the following genes: HLA -A29.1,RPS4Y1, XIST, LOC100133 662, HLA -DRB4, EIF1AY, and RPS4Y2 (TABLE 4).
The crucial pathways involving the analyzed transcripts (t test, n = 2455) were indicated with functional analysis (Supplementary material, Table S2).
DISCUSSION Our results showed interesting differences in gene expression related to the CPET results indicating abnormalities in fat oxidation and maximal oxygen uptake.The genes with crucial differences in expression were CLEC12A, genes of the major histocompatibility complex HLA -DRB1, HLA -DRB4, HLA -A29.1, and IFIT1, and a transcript with an unknown function LOC100133662.
A difference in expression of 2 transcripts of the CLEC12A gene was found in patients who achieved a significant reduction of the body mass.The product of the CLEC12A belongs to the family of lectin receptors localized on the surface of myeloid cells, including macrophages.Its role was described in rheumatoid arthritis 27 and Crohn disease. 28Differences in expression of CLEC12A dependent on the patient diet were described. 21he apigenin, which is a natural flavonoid, decreases the expression of CLEC12A in dendritic cells.It reduces systemic inflammation related to obesity. 21Systemic inflammation is one of significant consequences of obesity.The crucial cells involved in systemic inflammation in obesity are the macrophages. 29 -31The correlation of the expression of macrophage activation markers and plasma glucose levels, as well as glycated hemoglobin A 1c , was described. 9Studies on the Nordic diet showed changes in gene expression. 20The diet components may influence the expression of the genes involved in apoptosis, DNA repair, as well as cell proliferation and migration. 19,32The higher expression of CLEC12A in patients who did not achieve a body mass reduction in our study may be related to increased inflammation related to obesity, which impairs the muscle function.
The product of the IFIT1 gene is an inflammatory protein.The expression of the gene was higher in patients with lower VO 2max , which can be related to its role in obesity -related inflammation.The expression of IFIT1 is induced by stimulation with interleukin 6 linked to obesity -related inflammation. 33An increase of the IFIT1 expression was also observed in obese patients with

Results of gene expression in healthy controls and
obese patients A difference in gene expression higher than 2 -fold was observed for the following transcripts: HLA -A29.1,RPS4Y1, XIST, LOC100 133 662, RPS26P11, RPS26L, LOC650646, and RPS4Y2.Significant results corrected for multiple testing were observed for 4 transcripts: RP-S4Y1, XIST, LOC100 133 662, and RPS4Y2.The results are presented in

Results of gene expression according to treatment out-
come A significant difference in gene expression was found for 1751 transcripts, including 8 with fold change >2: LOC388588, CLEC12A (3780385), RPS4Y1, HBG1, CLEC12A (3170 601), HLA -DRB6, HBG2, and LOC100008588.The correction for multiple testing indicated significant results for 2 transcripts of CLEC12A, which showed lower expression for patients who achieved a significant reduction of body mass (>10% of the initial weight) (TABLE 3).
The crucial pathways involving the analyzed transcripts (t test, n = 1751) were indicated with functional analysis (Supplementary material, Table S1).

Results of gene expression according to FAT value
A significant difference in gene expression between patients with FAT lower than 20 g/h compared with those with FAT of 20 g/h or higher was found for 2455 transcripts, including 8 with fold change >2: HLA -A29.1,RPS4Y1, XIST, LOC100133 662, HLA -DRB6, HLA -DRB4, EIF1AY, and RPS4Y2.The correction for multiple testing the platelet -derived growth factor signaling pathway was identified in the group of genes with differences in expression in the present study.
The proteins which are involved in this pathway are an important component of metabolic syndrome. 38The integrin as well as Wnt pathways play a role in the adiponectin signal transduction. 39,40The pathological imbalance of αv/β1 integrin subunits is implicated in obesity. 41The comparison of gene expression in patients according to the peak oxygen uptake showed significant differences in the endothelin pathway-a strong vasoconstrictive and atherogenic factor related to obesity and its complications such as hypertension and obstructive sleep apnea. 42A crucial step in gene expression is related to RNA isolation.The PAXGene Tubes system used in this study is a standardized method that provides results highly similar to in vivo conditions and reduces the possible errors related to cell sorting and sample storage.It was used, for example, in the Framingham Heart Study, proving the high diagnostic value of the method. 22 study analyzing differences in gene expression found in peripheral blood in patients with obesity and metabolic syndrome had been published previously. 23This study confirmed differences related to starvation and lipid -lowering diet. 23 However, the current study did not identify the genes asthma. 33IFIT1 is also involved in the immune response to a viral infection and is induced by T helper -2 derived cytokines. 34atients with a lower VO 2max also presented higher expression of HLA -DRB1.The genes of the major histocompatibility complex were studied in relation to diabetes, hypertension, and coronary artery disease. 32The polymorphism of HLA -DRB1 increases the risk of type 2 diabetes. 35he gene is localized in the region 6p22 of chromosome 6.This area was related with the risk of type 2 diabetes in a recent multicenter genome--wide association study. 32Additionally, the genes HLA -A 29.1 and HLA -DRB4, which also showed differences in the analyzed group, are localized in the neighboring 6p21 region.Genes localized in this part of the genome show high expression in the adipose tissue. 36he transcripts RPS4Y1, RPS4Y2, XIST, and EI-F1AY, which showed a significant difference in expression, are associated with the sex differences in the study group.
The pathway analysis may provide insights into pathophysiological mechanisms related to the differences in gene expression observed in our study.A group of 16 transcripts differentiating patients who achieved a body mass reduction of at least 10% of the initial body mass, plays a role in the proteins Gi alpha and Gs alpha.This pathway is crucial for the adipogenesis and functioning of fat tissue. 37It also plays a role in thermogenesis and cold tolerance. 37The mutation of the studied genes leads to the development of Albright syndrome, a rare form of genetic obesity. 37 group of 17 of the 127 proteins cooperating in  The analysis of both cohorts was combined and done in a few steps.Based on previous publications of cooperating centers, several genes of interest were identified, the polymorphisms of which might have been related to the effects of treatment and further weight regain.The results on a group of 3597 participants showed that allele Ala 12 of the PPARG gene is related to short-and long--term effects of treatment.The short -term effect was also related to the polymorphism of LY-PLAL1, GNPDA2, and MTCH2 genes. 45The long--term effect was related to the polymorphism of NEGR1 and BDNF genes.A further study was performed in a group of 5730 patients including 1724 participants of the Diabetes Prevention Program and 3906 patients from the Look AHEAD study. 8A significant difference was shown only in a single gene, MTIF3, whose polymorphism was related to a significant body mass reduction. 8A recent Iranian study reported that the polymorphism of the DYRK1B gene may be related to metabolic syndrome. 46t is not possible to indicate a set of genes that could be used as pharmacogenetic biomarkers based on the current knowledge in the field.
The current study has some limitations, such as the use of whole blood in the analysis.Nonetheless, our study gives a unique insight into in vivo gene expression.Based on the data by Liu et al, 44 we can estimate which cell type is responsible for the observed changes in gene expression.Similarly, there are tools for a post hoc extrapolation of mixed -cell population lines in epigenetic studies. 47However, in our future studies we are planning to focus on selected cell types and selected genes studied using the quantitative polymerase chain reaction method.
In conclusion, the results of the current study indicate that obesity is linked to a difference in expression of the genes CLEC12A, IFIT1, HLA -DRB1, HLA -DRB4, and HLA -A 29.1, as well as genes involved in platelet -derived growth factor, endothelin, and G -protein signaling pathways.CPET results provide a unique insight into the pathology of obesity.The test can also be used as a predictor of treatment outcomes.The differences in expression may be responsible for the effects of the treatment related to inflammation caused by obesity, which influences the muscles, fat tissue, and fatty acid metabolism.
SUPPLEMENTARY MATERIAL Supplementary material is available with the online version of the article at www.pamw.pl.

TABLE 1
Clinical data of obese patients and controls participating in the gene expression study

TABLE 2
Genes with different expression in obese patients and controls

TABLE 4
Genes with different expression according to FAT (g/h) results 20,23The current study proved the role of the CLEC12A gene.Further functional studies of this gene are needed.Gene analyses were part of large cohort studies on lifestyle modification program in obesity performed in the United States: Diabetes Prevention Program 45 and Look AHEAD. 8