DOI: 10.31038/EDMJ.2020435

Abstract

Aims: Circulating neurotoxic autoantibodies to the 5-hydroxytryptamine 2A receptor were increased in older adult type 2 diabetes in association with certain neurodegenerative complications. The male Zucker diabetic fatty (ZDF) rat is a model system for studies of obese, type 2 diabetes mellitus. The aim of the current study was to test for (and compare) circulating neurotoxic autoantibodies to the 5-hydroxytryptamine 2A receptor in the Zucker diabetic fatty rat and age-matched lean Zucker rat strains.

Methods: Plasma from lean and Zucker diabetic fatty rat (obtained at different developmental stages) was subjected to protein G affinity chromatography. The resulting immunoglobulin G fraction was tested for neurotoxicity (acute neurite retraction, accelerated neuron loss) in N2A mouse neuroblastoma cells and for binding to a linear synthetic peptide corresponding to the second extracellular loop of the 5-hydroxytryptamine 2A receptor.

Results: The male Zucker diabetic fatty rat (fa/fa) and two Zucker lean strains (+/?) and (fa/+) harbored autoantibodies to the 5-hydroxytryptamine 2A receptor which appeared spontaneously around 7-8.5 weeks of age. The circulating autoantibodies persisted until at least 25 weeks of age in the Zucker diabetic fatty rat and in the Zucker heterozygote (fa/+), but were no longer detectable in 25-week-old lean (+/?) Zucker rats. Autoantibody-induced acute neurite retraction and accelerated loss in mouse neuroblastoma N2A cells was dose-dependently prevented by selective antagonists of the 5-hydroxytryptamine 2A receptor. It was also substantially prevented by co-incubation with antagonists of RhoA/Rho kinase-mediated signaling (Y27632) or Gq11/phospholipase C/inositol triphosphate receptor-coupled signaling.

Conclusions: These data suggest that neurotoxic 5-hydroxytryptamine 2A receptor-targeting autoantibodies increase in the aging male Zucker diabetic fatty rat and in male Zucker lean rats harboring a heterozygous mutation, but not in age-matched, older Zucker lean rats lacking a known leptin receptor mutation. The Zucker genetic strain may be useful in studies of the role of humoral and/or innate immunity in late neurodegeneration.

Introduction

Type 2 diabetes mellitus is associated with an increased risk of late neurodegeneration [1,2] via mechanisms which may involve (in part) increased peripheral and central inflammation. Increased pro-inflammatory cytokines and innate immunity have been associated with early Parkinson’s disease [3] and dementia [4] through complex mechanisms. Leptin is a hormone released by fat cells that is important in energy metabolism. Dysfunction of the leptin system results in classic signs of type 2 diabetes mellitus, such as obesity, insulin resistance and high circulating insulin. Leptin is also a member of the type 1 cytokine family that includes interleukin-6 [5,6]. The leptin receptor is normally expressed on hematopoietic cells where it stimulates helper- T cell, and effector T cell functions and inhibits regulatory T cell function [7].

The male Zucker Diabetic Fatty rat (fa/fa) (ZDF) is a well-recognized animal model of obese type 2 diabetes mellitus and hypertension [8,9] in which hyperphagia causes morbid obesity leading to high level of pro-inflammatory adipocytokines [10]. In the ZDF rat, pro-inflammatory cytokines cause enhanced innate immunity [11], however, owing to its complete leptin receptor deficiency, helper T-cell function is reduced [12]. The lean heterozygous Zucker rat (fa/+) has haplo-sufficiency at the leptin receptor and a two-fold increased plasma leptin concentration [13] compared to the wild-type (+/+) lean Zucker rat. Since hyperleptinemia promoted helper-T cell-mediated autoantibody formation and worsened clinical outcome(s) in a mouse model of systemic lupus erythematosus [14], the lean heterozygote Zucker rat might be expected to exhibit higher autoantibody expression than in the ZDF (fa/fa) rat strain.

In prior studies in older adult (human) obese type 2 diabetes, patients harboring increased circulating autoantibodies to the 5-hydroxytryptamine 2A receptor (5-HT2AR) had an increased prevalence of co-morbid neurodegenerative complications including Parkinson’s disease and dementia [15]. Human plasma autoantibodies targeting the 5-HT2A receptor promoted long-lasting (agonistic) activation of Gq11/phospholipase C/inositol triphosphate receptor/Ca2+ signaling causing accelerated endothelial cell and neuronal cell death [16]. Since the 5-hydroxytryptamine 2A receptor is normally expressed on arterial smooth muscle cells [17] and on neurons in specific brain regions involved in the regulation of mood, thinking, perception, and sleep [18], altered signaling at the 5-HT2A receptor could have diverse peripheral and central effects.

The aim of the present study was to test for spontaneously-occurring autoantibodies to the 5-HT2A receptor in the circulation in male ZDF rat (fa/fa) and age-matched male lean Zucker rats either lacking a known leptin receptor mutation (+/?) or harboring a heterozygous leptin receptor mutation (fa/+). The developmental expression of autoantibody was compared in each genetic strain, and the mechanism of neurotoxicity downstream of 5-HT2AR binding was evaluated in mouse neuroblastoma N2A cells.

Materials and Methods

Animals

All procedures were conducted according to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of the Veterans Affairs Medical Center (East Orange, New Jersey). Male ZDF (N=12) and lean (+/?) Zucker rats (N=12) were obtained from Charles River Laboratories (Kingston, NY) at approximately 6 weeks of age. All rats were single housed upon arrival, with modest enrichment (a PVC tube). Rats were provided ad libitum access to food and water and maintained in a 12 h light/dark cycle with lights on at 0630. All procedures occurred during the light phase of the cycle.

Blood Drawing

EDTA plasma from 10-week ZDF rat and 10-week lean, heterozygous (fa/+) Zucker rats was obtained from Charles River Laboratories. For all other time points, blood was collected from the retro-orbital plexus of ZDF, and lean (+/?) Zucker rats using a retro-orbital blood procedure. The blood was centrifuged at 1400 g x 10 minutes at 4 degrees C. The resulting plasma was stored at -4^oC for up to 2-3 weeks prior to protein G affinity chromatography.

Tail-Nick

Capillary blood glucose was determined from a drop of blood obtained after tail nick at sequential time intervals. A glucose dehydrogenase method (Accucheck Aviva, Roche Diagnostics, Inc) was used for the determination of blood glucose.

Protein G Affinity Chromatography

Protein-G affinity chromatography was carried out as previously reported for protein-A affinity chromatography in human plasma [16]. The pH of the 10 mm Tris binding buffer (6.5) and the 0.1 M citrate elution buffer (2.7) was adjusted to optimize binding and elution of rat immunoglobulin G. Protein-G eluate fractions were stored at 0-4 degrees C.

Mouse N2A Neuroblastoma Cells

Cells were cultured in DMEM with 10% fetal calf serum as previously reported [16].

Neurite Retraction and N2A Cell Survival Assays

Assays were performed as previously described [16].

Enzyme-Linked Immunoassay (ELISA)

The ELISA procedure was carried out as previously reported using an 18-meric linear synthetic peptide Q..N18 corresponding to the second extracellular loop of the human 5-HT2AR as the solid phase antigen [15].

Chemicals

All chemicals were obtained from Sigma Co., Inc. (St. Louis, MO).

Protein determination-Protein levels were determined using a bichinchonic assay method (BioRad, Inc.).

Statistics

Statistical analysis was performed using an unpaired Student’s t-test with an α cutoff level for significance of < 0.05.

Results

Occurrence of Circulating 5-HT2A Receptor Autoantibody in the ZDF Rat: Relation to Obesity and Diabetes

Mean body mass in ZDF rats significantly exceeded that in lean (+/?) Zucker rat at week 7, week 8.5, and week 12 (Figure 1A). There was no longer a significant weight difference between the strains at week 25 (Figure 1A). Diabetes (hyperglycemia) emerged between week 8-10 in the ZDF rat, but at week 7 the ZDF rat was still normoglycemic (Figure 1B). In contrast, lean (+/?) Zucker rats were normoglycemic at all ages tested (Figure 1B), not tested at week 7.

Figure 1: Age-dependent change in body mass(A) and capillary blood glucose concentration (B) in male Zucker fatty (ZDF) vs lean (+/?) (ZDL) rats. A) Body weight increased significantly in the male ZDF rat (starting at week 7) compared to control lean Zucker rat B) capillary blood glucose was determined by tail nick using glucose dehydrogenase method as described in Materials and Methods. Glucose was normal at week 7 in the ZDF rat, but increased significantly compared to ZDL rat starting at week 8.

Detection of plasma immunoreactive 5-HT2AR autoantibody had a similar trajectory to diabetes in ZDF rats. At 7 weeks of age, ZDF rat had no detectable plasma immunoreactive 5-HT2AR autoantibody (Figure 1C). At 10-weeks of age, plasma from ZDF rat contained a two-fold, higher binding to 5-HT2AR (0.14 AU, N=2) compared to background levels (0.07 AU; Figure 2a). Plasma from lean ((+/?) Zucker rats harbored two-fold or slightly higher level of autoantibody binding to 5-HT2AR between 8.5-15 weeks of age which declined to undetectable levels at 25 weeks of age (Figure 2B).

Figure 2: Developmental expression of autoantibody binding to 5-HT2AR synthetic peptide in three Zucker rat strains. A 2 ug/mL concentration of the protein G eluate was tested for binding to the linear synthetic 18-meric 5-HT2A receptor peptide as described in Materials and Methods. Results are “mean +/- SE” in A) ZDF rat: 7-week (N=2); 8.5-week (N=3); 10-week (N=2); 25-week (N=3) B) lean (+/?) Zucker rat: 7-week (N=3); 8.5 week (N=3); 15-week (N=10); 25-week (N=6). C) lean heterozygote (fa/+) Zucker rat: 7-week (N=2); 10-week (N=5); 25-week (N=2).

Spontaneous IgG autoantibody to the 5-HT2A receptor was already present at 7 weeks of age in the heterozygous lean (fa/+) Zucker rat (Figure 2C) and levels at 10-weeks old and 25-weeks old were substantially higher than in age-matched ZDF rat (Figure2C). Autoantibody to the 5-HT2A receptor persisted into older age, i.e. 25 weeks old, in male ZDF and lean heterozygous Zucker rats, but disappeared at 25 weeks of age in the lean (+/?) Zucker rat lacking a known leptin receptor mutation. In a recent report [19], diabetes, obesity and hypertension were associated with substantially increased hazard rates for the occurrence of dementia, Parkinson’s disease and severe depression following traumatic brain injury in older adult veterans.

The present data suggest that male ZDF and heterozygous lean rat strains may be useful in modeling the effects of heightened innate and/or adaptive immunity, respectively, on the occurrence of neurodegeneration.

Zucker Diabetic Fatty Rat Plasma 5-HT2AR Autoantibody Inhibits N2A Cell Survival

We next tested for neurotoxicity in the autoantibody (from 10- week-old ZDF rat plasma) that displayed increased binding to the 5-HT2A receptor peptide. Autoantibody in plasma from the 10-week old ZDF rat caused dose-dependent inhibition of survival in N2A mouse neuroblastoma cells (Figure 3). A two and one-half microgram per milliliter concentration of the IgG caused approximately 40% neuroblastoma cell loss (after 24 hours incubation) compared to control, untreated cells (P< 0.01, Figure 3).

Figure 3: Dose-dependent neurotoxicity in the protein G eluate fraction from Zucker diabetes fatty rat plasma. The protein G-eluate fraction from 10-week-old male Zucker diabetic fatty rat plasma was incubated with N2a mouse neuroblastoma cells (at the indicated concentrations) for 24 hours. % N2a cell survival was determined with an MTT assay as described in Materials and Methods. Each point represents the mean + SE of quadruplicate determinations; * P < 0.01 compared to cell survival in untreated cells.

Zucker Diabetic Fatty Rat Plasma 5-HT2AR Autoantibody Causes Acute N2A Neurite Retraction

A 120 nanomolar concentration (1.8 g/mL IgG) of the protein G- eluate fraction of plasma from 10-week-old male ZDF fatty rat caused significant 37% acute neurite length-shortening (after 15 minutes of incubation) of N2A mouse neuroblastoma cells (Figure 4A). Co-incubation with a 200 nanomolar concentration of M100907, a highly selective, potent antagonist of the 5-HT2AR, afforded 86% significant (P < 0.01) protection against autoantibody-induced neurite retraction (Figure 4A).

Figure 4: Neurite outgrowth inhibition (A) and accelerated N2A neuroblastoma cell loss (B) in autoantibody from ZDF rat plasma: neutralization by M100907, a highly selective 5-HT2AR antagonist. A) A 120 nM concentration of the protein-G eluate fraction of 10-week-old male ZDF rat plasma was incubated with N2A mouse neuroblastoma cells in the presence or absence of a 200 nM concentration of the selective 5-HT2AR antagonist M100907. % Basal neurite length was determined after 15 minutes as described in Materials and Methods B) A 160 nM concentration of the protein G-eluate fraction from 10-week-old male Zucker diabetic fatty rat plasma was incubated with N2a mouse neuroblastoma cells in the presence or absence of the indicated concentration of M100907 for 24 hours at 37 deg C. Percent N2A cell survival was determined with an MTT assay as described in Materials and Methods. Each point represents the mean + SE of triplicate determinations.

Pharmacologic Profile of Zucker Diabetic Fatty Rat Plasma 5-HT2AR Autoantibody Toxicity

A ~170 nanomolar concentration (2.5 µg/mL) of the protein G-eluate fraction of plasma from 10-week-old male ZDF rat was incubated with N2A mouse neuroblastoma cells in the presence or absence of a (500 or 1000) nanomolar concentration of M100907 for 24 hours at 37 deg C. M100907 caused significant (dose-dependent) protection against accelerated N2A cell loss, after 24 hours incubation with the ZDF rat IgG (Figure 4B).

Spiperone is a potent, but less selective 5-HT2AR antagonist. Spiperone at 500-1000 nanomolar concentrations afforded significant dose-dependent protection against ZDF rat IgG-induced accelerated N2A cell loss (Figure 5). Ketanserin is a less potent 5-HT2AR antagonist compared to M100907 or spiperone. Substantially higher concentration (2.5 µM) of ketanserin was required to significantly protect against ZDF rat IgG-induced N2A cell loss (Figure 6).

Figure 5: Potent, less selective 5-HT2AR antagonist (spiperone) dose-dependently prevented accelerated neuron loss induced by ZDF rat plasma autoantibody. A 160 nM concentration of the protein G-eluate fraction from 10-week-old male Zucker diabetic fatty rat plasma was incubated with N2a mouse neuroblastoma cells in the presence or absence of the indicated concentration of spiperone for 24 hours at 37 deg C. Percent N2a cell survival was determined with an MTT assay as described in Materials and Methods. Each point represents the mean + SE of triplicate determinations.

Figure 6: Higher concentration of ketanserin, a less potent, but selective 5-HT2AR antagonist dose-dependently prevented accelerated neuron loss induced by ZDF rat plasma autoantibody A-B) 120 nM concentration of the protein G-eluate fraction from 10-week-old male Zucker diabetic fatty rat plasma was incubated with N2a mouse neuroblastoma cells in the presence or absence of the indicated concentration of ketanserin for 24 hours at 37 deg C. Percent N2a cell survival was determined with an MTT assay as described in Materials and Methods. Each point represents the mean + SE of triplicate determinations.

The receptor antagonist profile and concentration associated with significant protection against ZDF IgG autoantibody-induced neuroblastoma cell loss is consistent with relative affinity constants on the 5-HT2A receptor (M100907 < spiperone<< ketanserin).

A one micromolar concentrations of SB-20471, a selective antagonist to the 5-HT2B receptor did not significantly prevent ZDF autoantibody-induced neurite retraction (Table 1). A ten micromolar concentration of either bosentan (an endothelin-1 receptor) or losartan (an angiotensin II type 1 receptor) antagonist did not significantly protect against ZDF IgG-induced acute N2A neurite retraction (Table 1). An 850 nanomolar concentration of prazosin (alpha 1 adrenergic R antagonist) did not afford significant protection against ZDF IgG autoantibody-induced acute neurite retraction (Table 1). Taken together, these data suggest that neurotoxicity in the 10-week old male ZDF rat IgG appears selective for the 5-HT2A receptor.

Table 1: Receptor specificity of prevention of ZDF rat IgG autoantibody-induced acute N2A neurite retraction.

A 160 nanomolar concentration of 10-week-old male Zucker fatty rat IgG autoantibodies was incubated in the presence or absence of the indicated GPCRs. Acute N2A neurite retraction was determined after 15 minutes as described in Materials and Methods. Results are mean +/- SD.

Involvement of Gq11/Phospholipase C/Inositol Triphosphate and Rho Kinase Signaling Pathways

Co-incubation of N2A cells with ZDF plasma IgG autoantibodies together with individual selective antagonists of Gq11 (YM-254890), phospholipase C (U73122), inositol triphosphate receptor (2-APB) signaling or Rho A/Rho kinase (Y27632) mediated signaling completely or nearly completely protected against acute neurite retraction (Table 2) consistent with involvement of Gq11/phospholipase C/inositol triphosphate and Rho kinase signaling pathways in mediating IgG neurotoxicity downstream of 5-HT2A receptor binding.

Table 2: Effect of signaling pathway inhibitors on ZDF rat IgG autoantibody-induced acute N2A neurite retraction.

A 160 nanomolar concentration of 10-week-old male Zucker fatty rat IgG autoantibodies was incubated in the presence or absence of the indicated GPCRs. Acute N2A neurite retraction was determined after 15 minutes as described in Materials and Methods. Results are mean +/- SD as described in Materials and Methods.

Level and Titer of Autoantibody to 5-HT2A Receptor in Male Zucker Heterozygote (fa/+)

An identical (1.5 µg/mL concentration) of the IgG from Zucker heterozygote rat plasma (N=5) displayed two-fold significantly higher binding (to the 5-HT2A receptor peptide) compared to IgG in plasma from five ZDF rat plasmas (N=2, 10-week-old and N=3, 25-week-old rats) (Figure 7A). The titer of autoantibody to the 5-HT2A receptor peptide was more than two-fold higher in the 10-week old lean Zucker heterozygous rat compared to age-matched, 10-week old ZDF rat (Figure 7B).

Figure 7: A) 2 ug/mL concentration of the protein G eluate of ZDF rat and lean Zucker heterozygote plasma was tested for binding to linear, synthetic 18-meric 5-HT2A receptor peptide.
B) Titer of autoantibody binding to 5-HT2A receptor peptide was tested at the indicated concentrations in 10-week lean Zucker heterozygote (N=5) or 10-week ZDF (N=2) protein G eluate. Results are “mean +/- SE”. * P< 0.01

Bioactivity in the Antibody Fraction of Male Zucker Heterozygous (fa/+) Plasma

Dose-dilution curves of the protein-G eluate fraction from three, 10-week-old lean heterozygous Zucker heterozygous plasmas demonstrated potent inhibition of N2A cell survival at all IgG concentrations tested in two of three rats (Figure 8A). In one of the three rats, lean heterozygous Zucker rat #3, a component of significant N2A survival-promoting activity were observed at high IgG dilution (Figure 8A). ‘Functional selectivity’ is a unique property of the 5-hydroxytryptamine 2A receptor [20]-it refers to the ability of different 5-HT2AR ligands to preferentially activate alternative signaling pathways, e.g. PLC/IP3/Ca2+ signaling or a beta-arrestin 2/PI3-kinase survival pathway. Since the selective 5-HT2AR antagonist M100907 significantly protected against lean heterozygous Zucker rat IgG autoantibody-induced accelerated N2A cell loss (Figure 8B) the neurotoxicity is likely due to Gq11/PLC/IP3/Ca2+-mediated signaling occurring downstream of 5-HT2AR binding. More study is needed, however, to determine whether N2A survival-promoting activity in a subset of lean heterozygous Zucker rat IgG autoantibodies may result from selective activation of beta arrestin-2/PI3-kinase-mediated signaling. In a prior report [21], highest level of 5-HT2A receptor-binding occurring in the autoantibodies from a patient with lupus. The lupus IgG autoantibodies were not only neurotoxic, but (at certain concentrations) promoted N2A neuroblastoma cell survival dependent on PI3 kinase-mediated signaling [21].

Figure 8: A) Dose-dilution curves of three Zucker lean heterozygous protein G eluates on survival in N2A mouse neuroblastoma cells; B) Co-incubation of a representative lean heterozygous protein G eluate fraction with 1 µM concentration of the selective 5-HT2AR antagonist M100907. * P< 0.05.

Apparent MW of the Active Component in ZDF Protein G Eluates

The active component in the ZDF protein G eluate (which caused accelerated N2A cell loss) had an apparent MW > 10 kD (Figure 9A) consistent with an antibody or antibody fragment. There was no significant N2A cell survival-inhibitory activity present in the flow-through fraction following membrane dialysis of the ZDF protein G eluate on a 10 kilodalton Molecular Weight (MW) cutoff membrane (Figure 9A). The freshly-isolated, lean heterozygous Zucker rat (fa/+) protein G eluate fraction (SM, starting material) caused significantly increased N2A cell survival. Following dialysis on a 10 kD MW cutoff membrane, more than 90% of the N2A survival stimulatory activity was recovered in the fraction having an apparent MW > 10 kD, i.e. retained on the membrane (Figure 9B). The low MW (< 10 kD) flow-through fractions following membrane dialysis of the ZDF or lean heterozygous Zucker rat protein G eluate did not cause significant inhibition or stimulation of N2A survival (Figure 9A and 9B).

Figure 9: Apparent MW of peak neuroblastoma survival-inhibitory (A) and – stimulatory (B) activity in the plasma protein G eluate from ZDF or heterozygous lean ZDL rat. Apparent MW of peak neuroblastoma survival-inhibitory (A) or -stimulatory (B) activity in the protein G eluate from a representative ZDF (A) and lean heterozygote # 3 Zucker rat (B) plasma. A 160 nM concentration of the protein G-eluate fraction from 10-week-old male Zucker rat plasma was dialyzed on a 10kD MW cutoff membrane. The resulting retained and flow-through fractions or starting material (SM) were incubated with N2a mouse neuroblastoma cells for 24 hours as described in Materials and Methods. Each point represents the mean + SE of triplicate determinations. * P < 0.05 compared to control N2A cell without added test fractions.

Discussion

The male ZDF rat is a well-studied model of human obese type 2 diabetes mellitus. The animals undergo progressive gain in fat mass, with obesity arising around 4-weeks of age [9]. Hyperglycemia commences around week 8- 9, and chronic marked elevation in blood glucose concentrations (400-500 mg/dL) is associated with early hyperinsulinemia followed by later relative insulin deficiency [10]. Hypertension and severe hypertriglyceridemia manifest around week 9 in the ZDF rat [9,10]. In adult, long-standing human obese type 2 diabetes mellitus from the Veterans Affairs Diabetes Trial, the baseline presence of anti-endothelial cell autoantibodies was significantly associated with lack of baseline insulin therapy (i.e. relative insulin deficiency) and lack of baseline triglyceride-lowering medication [22]. Endothelial cell heparanase expression is upregulated in insulin deficiency [23] and by high level of pro-inflammatory cytokines [24]. Heparanase cleaves heparan sulfate side chains abundant on neurons and vascular cells, and heparan sulfate proteoglycan is a known target of humoral autoimmunity [25]. In older adult diabetes patients evidence suggested that autoantibodies targeting the 5-HT2AR were in many cases cross-reactive with heparan sulfate proteoglycan [26]. Taken together it is possible that certain metabolic derangements in the ZDF rat strain, (insulin deficiency, and high level of pro-inflammatory cytokines) may contribute in part to the long persistence of circulating 5-HT2A receptor autoantibodies at older ages.

Although the heterozygous (fa/+) Zucker rat is lean and non-diabetic, it demonstrated substantially higher titer of 5-HT2A receptor autoantibodies compared to age-matched ZDF rat. One possibility is that enhanced helper T-cell immunity in the lean heterozygote Zucker rat drives autoantibody formation which includes a subset (of autoantibodies) targeting the 5-HT2A receptor. Neurotoxicity in the lean heterozygote Zucker IgG autoantibodies (i.e. acute neurite retraction and accelerated N2A cell loss) was as high or higher than in ZDF IgG autoantibodies consistent with a prior report of a significant correlation between antibody binding level and level of acute neurite retraction in N2A cells [15]. Higher level of 5-HT2AR targeting autoantibodies in the lean heterozygote Zucker rat supports a role for upregulated adaptive humoral immunity (in part) due to a heterozygous leptin receptor mutation causing hyperleptinemia. More study is needed to determine whether high titer of autoantibodies in the lean heterozygous Zucker rat may predispose to immune complex formation that could bias signaling pathway activation downstream of 5-HT2A receptor binding.

In summary, 5-HT2A receptor autoantibodies increased in the circulation in male hypertensive ZDF rat between 8.5-10-weeks of age and persisted up until 25-weeks old or longer. The ZDF rat 5-HT2A receptor-targeting autoantibodies caused acute neurite retraction and accelerated mouse neuroblastoma cell loss by a mechanism involving activation of Gq11/PLC/IP3R pathway and RhoA/Rho kinase coupled signaling. The ZDF and lean heterozygous Zucker rat strains may be suitable animal models for investigations of the roles of obesity-associated inflammation and dysregulated humoral immunity, respectively, in the etiology of certain forms of late-onset neurodegeneration.

Funding

This project was supported by grants from the New Jersey Commission on Brain Injury Research (CBIR19PIL007) to MBZ and the Biomedical Laboratory Research and Development Service of the VA Office of Research and Development (I1BX004561) to KCHP. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.

References

1. Hu G, Pekka J, Siamak B, Riitta A, Jaakko T (2007) Type 2 Diabetes and the Risk of Parkinson’s Disease. Diabetes Care 30(4): 842-847. [crossref]
2. Xu WL, Qiu CX, Wahlin A, Winblad B, Fratiglioni L (2004) Diabetes mellitus and risk of dementia in the Kungsholmen project: a 6-year follow-up study. Neurology 63(7): 1181-1186. [crossref]
3. Prieto GA, Tong L, Smith ED, Cotman CW (2019) TNFα and IL-1β but not IL-18 Suppresses Hippocampal Long-Term Potentiation Directly at the Synapse. Neurochem Res 44(1): 49-60. [crossref]
4. Qin XY, Zhang SP, Cao C, Loh YP, Cheng Y (2016) Aberrations in Peripheral Inflammatory Cytokine Levels in Parkinson Disease: A Systematic Review and Meta-analysis. JAMA Neurol 73(11): 1316-1324. [crossref]
5. Baumann H, Morella KK, White DW, Dembski M, Bailon PS, Kim H, Lai CF, Tartaglia LA (1996) The full-length leptin receptor has signaling capabilities of interleukin 6-type cytokine receptors. Proc Natl Acad Sci U S A 93(16): 8374-8. [crossref]
6. Dozio E, Ruscica M, Galliera E, Corsi MM, Magni P (2009) Leptin, ciliary neurotrophic factor, leukemia inhibitory factor and interleukin-6: class-I cytokines involved in the neuroendocrine regulation of the reproductive function. Curr Protein Pept Sci 10(6): 577- 584. [crossref]
7. Lord, G., Matarese, G., Howard, J. et al. (1998) Leptin modulates the T-cell immune response and reverses starvation-induced immunosuppression. Nature 394, 897-901. [crossref]
8. Clark JB, Palmer CJ, Shaw WN (1983) The diabetic Zucker fatty rat. Proc Soc Exp Biol Med 173(1): 68-75. [crossref]
9. Kurtz TW, Morris RC, Pershadsingh HA (1989) The Zucker fatty rat as a genetic model of obesity and hypertension. Hypertension 13: 896-901. [crossref]
10. Shibata T, Takeuchi S, Yokota S, Kakimoto K, Yonemori F, Wakitani K (2000) Effects of peroxisome proliferator-activated receptor-alpha and -gamma agonist, JTT-501, on diabetic complications in Zucker diabetic fatty rats. Br J Pharmacol 130(3): 495-504. [crossref]
11. Ruth MR, Taylor CG, Zahradka P, Field CJ (2008) Abnormal immune responses in fa/fa Zucker rats and effects of feeding conjugated linoleic acid. Obesity Silver Spring 16(8): 1770-9. [crossref]
12. Tanaka SI, Isoda F., Yamakawa T., Ishihara M, Sekihara H (1998) T lymphopenia in genetically obese rats. Clinical Immunology and Immunopathology 86(2): 219-225. [crossref]
13. Imai T, Cho YM, Takahashi M, Kitahashi T, Takami S, Nishikawa A, Ogawa K (2013) High susceptibility of heterozygous (+/fa) lean Zucker rats to 7,12-dimethylbenz(a)anthracene-induced mammary carcinogenesis. Oncol Rep 29(5): 1914-1922. [crossref]
14. Lourenço EV, Liu A, Matarese G, La Cava A (2016) Leptin promotes systemic lupus erythematosus by increasing autoantibody production and inhibiting immune regulation. Proc Natl Acad Sci U S A 13(38): 10637-10642. [crossref]
15. Zimering MB (2019) Autoantibodies in Type-2 Diabetes having Neurovascular Complications Bind to the Second Extracellular Loop of the 5-Hydroxytryptamine 2A Receptor. Endocrinol Diabetes Metab J 3(4): 1-8. [crossref]
16. Zimering MB (2017) Diabetes Autoantibodies Mediate Neural- and Endothelial Cell- Inhibitory Effects Via 5-Hydroxytryptamine- 2 Receptor Coupled to Phospholipase C/Inositol Triphosphate/Ca2+ Pathway. J Endocrinol Diabetes 4(4): 10.15226/2374-6890/4/4/00184. [crossref]
17. Xu T, Pandey SC (2000) Cellular localization of serotonin (2A) (5HT (2A)) receptors in the rat brain. Brain Res Bull 51: 499-505. [crossref]
18. Alsip NL, Harris PD, Durrani GE (1991) Multiple Serotonin Receptors on Large Arterioles in Striated Muscle. J Vasc Res 28: 537-541. [crossref]
19. Zimering MB, Patel D, Bahn G (2019) Type 2 Diabetes Predicts Increased Risk of Neurodegenerative Complications in Veterans Suffering Traumatic Brain Injury. J Endocrinol Diabetes 6(3). [crossref]
20. Schmid CL, Raehal KM, Bohn LM (2008) Agonist-directed signaling of the serotonin 2A receptor depends on beta-arrestin-2 interactions in vivo. Proc Natl Acad Sci U S A 105: 1079-1084. [crossref]
21. Zimering MB, Nadkarni SG (2019) Schizophrenia Plasma Autoantibodies Promote ‘Biased Agonism’ at the 5-Hydroxytryptamine 2A Receptor: Neurotoxicity is Positively Modulated by Metabotropic Glutamate 2/3 Receptor Agonism. Endocrinol Diabetes Metab J [crossref]
22. Zimering MB, Anderson RJ, Moritz TE, Ge L (2009) Investigators for the VADT. Endothelial cell inhibitory autoantibodies are associated with laser photocoagulation in adults from the Veterans Affairs Diabetes Trial. Metabolism 58(6): 882-7. [crossref]
23. Han J, Woytowich AE, Mandal AK, Hiebert LM (2007) Heparanase upregulation in high glucose-treated endothelial cells is prevented by insulin and heparin. Exp Biol Med (Maywood) 232(7): 927-34. [crossref]
24. Chen G, Wang D, Vikramadithyan R, Yagyu H, Saxena U, Pillarisetti S, Goldberg IJ (2004) Inflammatory cytokines and fatty acids regulate endothelial cell heparanase expression. Biochemistry 43(17): 4971-7. [crossref]
25. Faaber P, Rijke TP, van de Putte LB, Capel PJ, Berden JH (1986) Cross-reactivity of human and murine anti-DNA antibodies with heparan sulfate. The major glycosaminoglycan in glomerular basement membranes. J Clin Invest 77(6): 1824-30. [crossref]
26. M.B. Zimering, A.T. Pulikeyil, CE Myers, KC Pang (2020) Serotonin 2A Receptor Autoantibodies Increase in Adult Traumatic Brain Injury In Association with Neurodegeneration. J Endocrinol Diab 7(1): 1-8. [crossref]

DOI: 10.31038/EDMJ.2020434

Abstract

Aim: To study the clinical and metabolic characteristics of newly diagnosed type 2 diabetes (T2DM) in urban clinics (CDD) and also to compare with the screen detected new diabetes cases (SDD) during an urban population survey.

Methods: Newly diagnosed T2DM (aged 20-60 years, n=741), based on blood glucose and Glycosylated haemoglobin (HbA1c) of ≥6.5% (48 mmol/mol) were selected. Demography, anthropometry, blood pressure, glycaemic and lipids profiles were analysed. Relevant statistical tests were used for group comparisons.

Results: Both groups had young age (45.0 ± 8.6 years) at diagnosis. Fasting blood glucose (p<0.05) and HbA1c (p<0.0001) were higher in CDD. Mean values of HbA1c were 9.1 ± 2.3% (76 ± 20 mmol/mol) in CDD and 8.3 ± 2.4% (67 ± 19 mmol/mol) in SDD (p<0.0001). Values  of HbA1c were higher   than ≥9.0% (75 mmol/mol) in 44.6% of CDD versus 26.4% of SDD (z=4.60, p<0.0001). SDD had higher body mass index (p<0.0001), abdominal obesity (p<0.005), hypertension (p<0.0001), cholesterol (p<0.005) and low density lipoprotein cholesterol (p<0.05) than CDD.

Conclusion: Both groups had young age at diagnosis. CDD had more severe glycaemia than SDD, probably suggesting that the clinic visits were delayed and therefore had longer period of undiagnosed diabetes. In comparison to CDD, SDD had higher metabolic abnormalities although the HbA1c values were lower.

Keywords

Newly diagnosed type 2 diabetes, South Asians, Clinical characteristics, HbA1c, Metabolic abnormalities

Introduction

Many developing countries show a rising trend in the prevalence and also in the incidence of type 2 diabetes T2DM [1-3]. Recently, a systematic review of studies on trends in the incidence of T2DM among adults in developed countries has shown that between 2006 and 2014, 27% of the reported populations had a stable incidence over time, while 36% reported a declining trend and 36% reported an increasing trend in the incidence of T2DM [4]. The declining trend in the incidence has been shown mostly in the developed countries. A huge clinical burden of newly diagnosed T2DM is present in developing countries. There is only limited real-world data describing the clinical characteristics of such patients [5].

The prevalence of diabetes is increasing rapidly in India, which is estimated to be 77 million adults in 2019 [1]. A recent epidemiological study done in urban India showed that, the prevalence and incidence of diabetes have increased significantly in all areas, including the Peri Urban Villages (PUV) of Tamil Nadu [2]. Using the data from the two studies conducted in 2006 [6] and 2016 [2] to estimate the incidence of diabetes, it was noted that, a sharper increase in the incidence occurred in the urban areas when compared with the villages [2]. A similar observation was made in the INdia DIABetes (INDIAB) study conducted in 15 states of India [7] as well as in a cohort study of 10 years follow-up in the urban area in southern India [8].

Several peculiar features are seen among the newly diagnosed South Asian populations with T2DM such as, younger age at onset, delayed clinical diagnosis due to lack of public awareness regarding the disease, reluctance to undergo periodic medical check-up and a long asymptomatic phase of T2DM [3,9]. The diagnosis of diabetes is often delayed, more so in the developing societies. Previous studies have described the characteristics of patients attending the clinics, who are under treatment [3,6,8]. There is sparse data on the profile of clinically diagnosed new T2DM patients in India.

The aim of this project was to study the clinical and metabolic characteristics of newly diagnosed adult T2DM patients from different clinics (CDD). We compared the profile with that of new T2DM identified during an epidemiological screening of urban population (SDD) [2].

Methods and Materials

Study Subjects

This is a study of newly diagnosed adult T2DM from 12 different centres located in four southern states of India. Patients with a previous history of diabetes or had taken anti-diabetes treatment were excluded from the study by the clinicians. Persons with prediabetes were also excluded. Details of men and women in the selected age group, who were treatment naïve were collected and analysed. For comparison, a group of newly detected T2DM subjects from an epidemiological survey of the urban population was taken.

All participants reported on fasting on the day of testing. In the clinics, the diagnosis of diabetes was based on blood glucose (fasting blood glucose ≥7.0 mmol/l and/2 hour Post Prandial Glucose (PPG)≥11.1 mmol/l) [10]. In the epidemiological study, diagnosis was made based on fasting and 2 h Post Blood Glucose (PBG) values. The 2 hour PBG was measured after giving 75 g of glucose load. We selected only patients who had an HbA1c value of ≥6.5% (48 mmol/mol) [11]. The total number of subjects analysed was 741, comprising of 514 from CDD and 227 from SDD; Identifier: NCT03490136 [2].

All centres followed uniform and standardised procedures for metabolic and clinical assessments. The study details and methodology were discussed in a meeting of the investigators, participating physicians and the researchers prior to the commencement of the study.

The Ethics Committee of the India Diabetes Research Foundation and Dr. A. Ramachandran’s Diabetes Hospitals approved the study. A written informed consent was obtained from all participants prior to the enrollment, at all centres. It was also agreed that the identity of the participants would not be disclosed. All study procedures were carried out in accordance to the ethical guidelines.

Clinical Assessments

Age, sex and presence of family history of diabetes were collected. Height was measured to the nearest centimeter using a stadiometer with the patient standing erect. Weight was measured to the nearest 0.1 kg using a digital weighing scale. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Generalized obesity was defined as BMI ≥23.0 kg/m²., BMI between ≥23.0 and 24.9 kg/m² was considered overweight, and BMI≥25.0 kg/m² was defined as obese. Waist circumference (WC), the smallest girth between the coastal margin and the iliac crest, was measured. Abdominal obesity was indicated by WC of ≥90 cm for men and ≥80 cm for women. Blood pressure was measured in the sitting position using the electronic measuring device (Omron HEM 7132; Omron, Tokyo, Japan). An average of two readings taken at 5-minutes’ interval was recorded. Persons with a history of hypertension and those with newly diagnosed diabetes with blood pressure readings ≥140/90 mmHg were categorized as hypertensive. All the clinical and metabolic details were recorded in a standardised questionnaire.

Biochemical Assessments

Glucose was measured in the respective study centres by the glucose oxidase method. Other biochemical parameters such as HbA1c and lipid profile were estimated at the central laboratory (Dr. A. Ramachandran’s Diabetes Hospitals, Chennai). Glycosylated haemoglobin (HbA1c) was measured by immunoturbidimetry (TINA-QUANT II; Roche Diagnostics Corporation, Germany); a procedure certified by the National Glycohemoglobin Standardization Program. Fasting serum lipid profile was estimated by standard enzymatic procedures using the reagents of Roche Diagnostics, Germany. HDLc was estimated by the direct assay. Samples were shipped to the central laboratory between 2° and 8°C on the same day of collection.

Statistical Analyses

Data are presented as mean ± standard deviation (SD) for continuous variables with a normal distribution, as median with interquartile range for skewed continuous variables and as number and frequency (%) for categorical variables. Independent samples t test, chi-square or Z – test were used to assess group differences for continuous variables and categorical variables respectively. Mann Whitney U test was used for skewed variables. A p value of <0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 21.0.

Results

Table 1 shows the demographic and anthropometric characteristics of the two groups with newly diagnosed T2DM. CDD had higher percentage of men than SDD (p<0.05). Both groups had similar mean age at diagnosis (45.2 ± 8 years). The mean BMI was higher in SDD (p<0.0001). Prevalence of overweight and obesity were similar in both the groups. Women in SDD had higher WC (p<0.005) and the percentage of abdominal obesity (≥80 cm) was also higher in SDD (p<0.05). SDD had higher mean diastolic blood pressure (p<0.0001) and higher percentage of hypertension (p<0.005).

Data are presented as mean ± SD for continuous variables with normal distribution and as frequency (%) for categorical variables. Independent samples t test and chi-square test or Z test were used to test inter group differences for continuous variables and categorical variables.
CDD vs.SDD=^*p< 0.05; ^#p< 0.005; ^$p< 0.0001.
CDD, Clinically Diagnosed Diabetes; SDD, Screen Detected Diabetes; BMI, Body Mass Index; BP, Blood Pressure.

Table 2 shows the metabolic profile of the study groups. SDD had lower FBG (p<0.05) and HbA1c (p<0.0001) when compared with CDD. Among the total cohort of 741, 445 (60.1%) had HbA1c values
≥8.0% (64 mmol/mol), of which 289 (39.0%) had values ≥9.0% (75 mmol/mol).

Data are presented as mean ± SD for continuous variables with a normal distribution, as median with interquartile range for skewedcontinuous variables and as frequency (%) for categorical variables. Independent samples t test and chi-square test or Z testwere usedto test intergroup differences for continuous variables and categorical variables. Mann Whitney U test was used for skewed variables.
CDD vs.SDD=^*p<0.05; ^#p<0.005; ^$p<0.0001
CDD: Clinically Diagnosed Diabetes; SDD: Screen Detected Diabetes; HbA1c: Glycosylated hemoglobin; LDLc: Low Density Lipoprotein Cholesterol; HDLc:HighDensity Lipoprotein Cholesterol.

It was observed that a larger proportion of CDD (67.9%) had values of ≥8.0% (64 mmol/mol) versus 42.3% of SDD (z=6.47, p<0.0001). HbA1c values ≥9.0% (75 mmol/mol) were present in 44.6% in CDD versus 26.4% in SDD (z=4.60, p<0.0001). Among the lipid variables, total cholesterol (p<0.005) and Low Density Lipoprotein cholesterol (LDLc) (p<0.05) were higher in SDD.

Discussion

There have been many reports highlighting the differences in the clinical profile of diabetes between South Asians and the western populations [12,13]. But, there is sparse data on the clinical and metabolic characteristics of newly diagnosed T2DM from India. In this communication we report the clinical and metabolic characteristics of newly diagnosed diabetes patients recruited from different centres in southern India (CDD). The group represents the urban population with diabetes. We compared its clinical profile with a group of undiagnosed diabetes detected during an urban epidemiological survey (SDD).

In our study, the mean age of the newly diagnosed T2DM in both the groups was similar (45.2 ± 8.6 and 45.0 ± 8.6 years in CDD and SDD respectively). Another large, multicentre, cross-sectional study among diabetes patients in India also showed that the mean age at onset of T2DM was 45.4 ± 10.9 years which was similar to the observation made in our study [14]. This was significantly lower than that reported among the newly diagnosed T2DM cases in the United States (55.6 ± 13 years) [5]. Similarly, in two large multi-ethnic studies from UK [15,16], the age at diagnosis of T2DM was found to be lower (51.5 ± 10.42 and 52.6 ± 13.5 years respectively) in the South Asian population when compared to the White population (58.9 ± 10.09 and 63.3 ± 13.8 years respectively). South Asians develop T2DM at least a decade earlier.

Lower BMI with higher HbA1c in the CDD compared to SDD could possibly be due to a reduction in weight usually associated with moderate to high levels of undetected hyperglycaemia. In an earlier study, we had observed that more than 20% of the cases had weight loss prior to the clinical diagnosis of T2DM [17]. Among the Asian populations, Indians have a reluctance to undergo regular health check-up and medical consultation is often sought only when the symptoms or related problems arise due to diabetes [9]. Similar observation had been reported earlier in migrant Asian populations in western countries [15,18]. In the UK, a multiethnic South London study mentioned above showed that HbA1c levels were higher in South Asians than in the Europeans at an early period after the diagnosis. This was partially attributed to a delayed diagnosis in the Asians [15]. In our study, the significantly higher FBG and HbA1c observed in CDD could be related to a delay in the diagnosis of diabetes when compared with those diagnosed during an epidemiological screening. There is a higher likelihood of a few CDD patients having taken treatment for hypertension and hyperlipidaemia prior to visiting a diabetes clinic.

In our study, 44.6% of the patients had significantly higher HbA1c ≥9.0% (75 mmol/mol) when compared with 21.7% of the newly diagnosed patients in the clinics in USA [5]. There may be a number of reasons for high HbA1c in newly detected diabetes. One important reason could be longer periods of undetected T2DM in our population, partially due to the delay on the part of the patients to seek medical help. In India, a larger percentage of men report for medical consultation than women. Therefore, the proportion of men was more among the clinically diagnosed cases.

In an earlier study, we had noted that T2DM remains asymptomatic for a longer period of time and many develop complications such as retinopathy even before diagnosis [19]. We did not record the presence of symptoms or complications in this study.

Hypercholestremia seen among the newly diagnosed patients may also be related to the delayed diagnosis of diabetes and higher blood glucose levels. In SDD, screening for diabetes was done among adults with neither a previous history nor the presence of symptoms of diabetes. There is a higher chance of early detection of diabetes among this group than in the clinic diagnosed cases. The likelihood of early detection of hypertension is also higher when screening is done.

Our study describes the clinical and metabolic features of newly diagnosed T2DM in the clinics. The differences with the screen detected diabetes in the population probably indicate late diagnosis in the clinics.

Declarations

Funding

The study was funded by India Diabetes Research Foundation, Chennai.

Conflict of interest

None

Ethics approval

The study was approved by the Ethics Committee of the India Diabetes Research Foundation and Dr. A. Ramachandran’s Diabetes Hospitals.

Consent to participate

A written informed consent was obtained from all participants prior to the enrollment in the respective studies.

Authors’ contributions

Arun R, AR, CS, AN and RV contributed to the study design, developed the protocol, supervised the study, drafted the manuscript and revised it with critical input. AR, CS, Arun R, PS and KS contributed to data preparation and analyses. KS and PS participated and coordinated the field work and data collection. All authors have read and approved the final draft of the manuscript.

Acknowledgements

The authors acknowledge the support rendered by the study physicians of various hospitals for providing the data source of the clinic patients. The help rendered by Mr.M.Karthikeyan and the epidemiology team of the India Diabetes Research Foundation in coordinating the epidemiological screening is greatly acknowledged. We are grateful to all the participants of the study for their co- operation and support.

Abbreviation

BMI: Body Mass Index

CDD: Clinically Diagnosed Diabetes

FBG: Fasting Blood Glucose

HbA1c: Glycosylated Haemoglobin

PBG: Post Blood Glucose

PPG: Post Prandial Blood Glucose

SDD: Screen Detected Diabetes

T2DM: Type 2 Diabetes

WC: Waist Circumference

References

1. International Diabetes Federation.IDF Diabetes Atlas, 9thedn. Brussels, Belgium: 2019. Available at: http: //www.diabetesatlas.org. Last accessed on 11th March 2020.
2. Ramachandran A, Snehalatha C, Shaw JE, Susairaj P, et al. (2019)Response to Comment on Nanditha et al. Secular TRends in DiabEtes in India (STRiDE-I): Change in Prevalence in 10 YearsAmong Urban and Rural Populations in Tamil Nadu. Diabetes Care42: 476-485.[crossref]
3. Misra A, Ramchandran A, Jayawardena R, Shrivastava U, et al. (2014)Diabetes in South Asians. Diabet Med31: 1153-62.[crossref]
4. Magliano DJ, Islam RM, Barr ELM, Gregg EW, et al. (2019)Trends in incidence of total or type 2 diabetes: systematic review. BMJ 366: l5003. [crossref]
5. Pantalone KM, Hobbs TM, Wells BJ, Kong SX, et al. (2016)Changes in characteristics and treatment patterns of patients with newly diagnosed type 2 diabetes in a large united states integrated health system between 2008 and 2013. Clin Med Insights Endocrinol Diabetes9: 23-30. [crossref]
6. Ramachandran A, Mary S, Yamuna A, Murugesan N, et al. (2008)High prevalence of diabetes and cardiovascular risk factors associated with urbanization in India. Diabetes Care 31: 893-898.[crossref]
7. Anjana RM, Deepa M, Pradeepa R, Mahanta J, et al. (2017)ICMR-INDIAB Collaborative Study Group. Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR-INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol 5(8): 585-596.[crossref]
8. Anjana RM, Shanthi Rani CS, Deepa M, Pradeepa R, et al. (2015)Incidence of Diabetes and Prediabetes and Predictors of Progression Among Asian Indians: 10-Year Follow-up of the Chennai Urban Rural Epidemiology Study (CURES). Diabetes Care38(8): 1441-8.[crossref]
9. Mohan V (2019)Expanding the concept of ‘Clinical Inertia’ in diabetes. J Diabetol 10: 1-3.
10. World Health Organization. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. Geneva: World Health Organization 2006; 1-50.[crossref]
11. Use of glycatedhaemoglobin (HbA1c) in the diagnosis of diabetes mellitus. Diabetes Res Clin Pract.93: 299-309.[crossref]
12. Nanditha A, Ma RC, Ramachandran A, Snehalatha C, et al. (2016)Diabetes in Asia and the Pacific: Implications for the Global Epidemic. Diabetes Care39: 472-85.[crossref]
13. Mathur R, Palla L, Farmer RE, Chaturvedi N, et al. (2020)Ethnic differences in the severity and clinical management of type 2 diabetes at time of diagnosis: A cohort study in the UK Clinical Practice Research Datalink. Diabetes Res Clin Pract160: 108006.[crossref]
14. Mohan V, Shah SN, Joshi SR, Seshiah V, et al. (2014)DiabCare India 2011 Study Group. Current status of management, control, complications and psychosocial aspects of patients with diabetes in India: Results from the DiabCare India 2011 Study. Indian J EndocrinolMetab 18: 370-8.[crossref]
15. Winkley K, Thomas SM, Sivaprasad S, Chamley M, et al. (2013)The clinical characteristics at diagnosis of type 2 diabetes in a multi-ethnic population: the South London Diabetes cohort (SOUL-D). Diabetologia56: 1272-81.[crossref]
16. Wright AK, Welsh P, Gill J, Kontopantelis E,et al. (2020)Age-, sex- and ethnicity-related differences in body weight, blood pressure, HbA1c and lipid levels at the diagnosis of type 2 diabetes relative to people without diabetes. Diabetologia 63: 1542-1553.[crossref]
17. Ramachandran A, Snehalatha C, Vijay V, Viswanathan M (1994)Underestimation of obesity in newly diagnosed non – insulin dependent diabetes subjects: A study from South India. Diabetes, Nutrition & Metabolism V7 (4): 251-55.[crossref]
18. Sattar N, Gill JM (2015)Type 2 diabetes in migrant south Asians: mechanisms, mitigation, and management. Lancet Diabetes Endocrinol 3: 1004-16.[crossref]
19. Ramachandran A, Snehalatha C, Vijay V, Viswanathan M (1996)Diabetic retinopathy at the time of diagnosis of NIDDM in south Indian subjects. Diabetes Res Clin Pract 32(1-2): 111-4.[crossref]

DOI: 10.31038/EDMJ.2020433

Abstract

In recent years, obesity has been identified as a risk factor for several cancers including endometrial cancer. This paper identifies molecular mechanisms that are potential causes or effects of the correlation between obesity and endometrial cancer. From public sources, metabolites involved in endometrial cancer were extracted and analyzed with pathway analyses. Several elucidated pathways and mechanisms were involved with both obesity and cancer and were studied to better understand the relationship between obesity and endometrial cancer. These pathways include the Adipocytokine signaling pathway, Insulin resistance pathway, and Glucagon signaling pathway. The Adipocytokine signaling pathway showed a relationship between leptin, a body-weight regulating hormone, and the JAK–STAT3 pathway known to promote cancer cell proliferation. The Insulin resistance pathway shows insulin, an obesity-related hormone, activating the STAT3 protein as well as the PI3k–Akt and the mTOR signaling pathways, all of which can lead to increased cancer risk. The Glucagon signaling pathway gave evidence that glucagon may have anticancer properties as it activates the FOXO signaling pathway, which suppresses cancer cell growth. Metabolic syndrome as well as foods with high glycemic loads may hinder the anticancer potential of glucagon and lead to an increase in endometrial cancer risk. Such results can provide targets for reducing the risk or progression of cancer and provides information on the effects of certain lifestyles on endometrial cancer risk.

Introduction

Endometrial cancer is the cancer of the innermost lining layer of the uterus. Endometrial cancer is the most common cancer of the female reproductive organs. In 2020, the American Cancer Society projections for cancer of the uterus in the United States are 65 620 new cases and 12 590 deathsincluding 59 060 and 11 330 for endometrial cancer accordingly (https://www.cancer.org/cancer/endometrial-cancer/about/key-statistics.html, accessed February 17, 2020). Endometrial cancer has a variety of risk factors, such as advanced age, hormonal imbalance, and diabetes. Additionally, one of the most prevalent risk factors of endometrial cancer is obesity. In the US, an estimated 57% of endometrial cancer cases are directly attributable to being overweight or obese [1]. This correlation has been proven by several meta-analyses, such as the American Institute for Cancer Research’s report that for every increase of five Body Mass Index (BMI) units (kg/m²), there was a 50% increase in the risk of developing endometrial cancer [1]. Among the reasons for this correlation is the fact that fat tissue can alter hormonal balance and increase estrogen levels and is related to insulin resistance, all of which have been directly linked to increased risk of endometrial cancer. Furthermore, several mechanisms involved in obesity have been identified as risk factors for endometrial cancer. The identification of obesity-related mechanisms in endometrial cancer networks has shown the oncogenic effects of various pathways and proteins. Such biological functions within the body can activate cancer-causing pathways and genes or inhibit anticancer processes. The Akt signaling pathway, activated by leptin and other obesity-related hormones, was shown to have a correlation with endometrial cancer when cancer cell proliferation decreased as Akt was inhibited in Ishikawa cells [2]. Obesity-caused activation of the STAT3 protein through leptin and insulin resistance, increases endometrial cancer risk as it promotes induction of oncogenic genes such as c-Myc [3]. Insulin resistance, commonly associated with obesity, is related to activation of TNF and mTOR signaling pathways that promote endometrial cancer [4,5]. Obesity-related proteins and genes, such as c-Myc, TNF-ɑ, S6K, and eIF4E, contribute to cell proliferation, protein synthesis, and oncogenesis [6,7]. Other proteins and genes, such as FOXO1, have tumor-suppression properties [8]. Similar studies demonstrated that cancers of the breast, colon, rectum, esophagus, kidney, and pancreas have also been linked to being overweight or obesity (Figure 1) [9,10].

Figure 1. Relationship of Body Mass Index and risk of endometrial cancer. Graph shows a positive correlation between BMI and endometrial cancer risk from several studies. Adapted from reference [10].

Materials and Methods

Approach Overview

Collected from public sources metabolite data were analyzed by Ingenuity^® Pathway Analysis (IPA^®, Santa Clara, CA; Krämer et al. 2014) [11] program. Created by the IPA program metabolite–gene networks were elucidated by the Database for Annotation, Visualization and Integrated Discovery (DAVID) software and analyzed withGeneFriends package (Figure 2).

Figure 2. Approach overview. Data from public sources are inputted to Ingenuity Pathway Analysis (IPA^®) program and results are analyzed by Database for Annotation, Visualization and Integrated Discovery (DAVID) Functional Analysis Tool yielding significant KEGG pathways. Then results analyzed by GeneFriends program to find co-expressed genes and cluster them.

Metabolite Selection and Analysis

Metabolites specific to endometrial cancer were selected from public sources: articles from PubMed and Google Scholar libraries, Human Metabolite Database (https://hmdb.ca/) and BiGG Models (https://bigg.ucsd.edu). To use significant metabolites only they were cut off by p-value and fold change.

Ingenuity^® Pathway Analysis

Ingenuity^® Pathway Analysis (IPA^®; QIAGEN Inc., Redwood City, Calif., USA; Krämer et al. 2014; https://www.qiagenbioinformatics.com/products/ingenuity-pathwayanalysis) [11] is a web-based analysis software that allow to understand complex ‘omics data. Using a list of genes or metabolites, it allows identifying signaling and metabolic pathways, interaction molecular networks, and their functions.

David

The Database for Annotation, Visualization and Integrated Discovery (DAVID) is an online server that consists of the DAVID Knowledgebase and five integrated functional annotation tools for bioinformatics data. It allows, among other functions, functionally classify and annotate genes, convert gene IDs, search for interesting and related genes, highlight them on signaling and metabolic pathways, and view these genes dynamically.

GeneFriends

GeneFriends is a tool that creates a large co-expression map from a wide variety number of datasets. There are two variants to this tool: a dataset consisting of microarray datasets [12] and a dataset consisting of RNA-seq samples [13]. The RNA-seq map currently consists of 46 475 human samples and 34 332 mouse samples while the microarray map currently consists of 4164 human microarray datasets and 3571 mouse datasets. This tool is used to elucidate clustered co-expressed genes since genes that are activated simultaneously tend to share similar transcriptional regulation. Novel genes probably related to the disease of study can be elucidated based on a guilt-by-association approach when observed to be co-expressed with known candidate genes. The microarray version of the program GeneFriends was used to elucidate the genes coexpressed with our genes of interest using over 1000 microarray datasets [12].

Results

Metabolites Selection

Selected metabolites underwent two cutoffs:

• byp-value (p). Metabolites with p<= 0.05 considered as significant.

• by fold change (FC). To reduce noise, the metabolites with 1.33 < FC < 0.67 were used.

The list of metabolites selected is presented in Table 1.

Gene-metabolite Network Analysis

The metabolites were analyzed by the IPA^® program, elucidating gene-metabolitenetworks including the selected metabolites, genes, and other entities (Figure 3).

Figure 3. Ingenuity Pathway Analysis networks 1 (a) and 2 (b) including endometrial cancer metabolites. Metabolites from the public sources [14] inputted into IPA, colored in blue. Proteins, colored yellow, were extracted and inputted into DAVID Functional Annotation Tool.

Many genes in these networks have a dual function involved in obesity/insulin resistance/diabetes mechanisms along with clear involvement in cancers including endometrial cancer (Table 2).

Network 1

In this network, a number of genes have dual function and are involved in cancer and obesity mechanisms. We will describe some examples of such dual genes (Figure 3a).

PDPK1 (PDK1)

PDPK1 (PDK1) (3-Phosphoinositide Dependent Protein Kinase 1) as a gene phosphorylating AKT1 and involved in MYC1 pathway is very important in mechanisms of various cancer including endometrial cancer [14,15].PDPK1 gene is also a key player in the obesity. It phosphorylates AKT1, which is important in insulin signaling. Inhibiting of PDPK1 affects AKT1 function and consequently insulin signaling and weight gain [16].Intervention of a lifestyle decreasing insulin resistance led to significant degrease of PDPK1 (PDK1) gene expression confirming its dependence on obesity-related parameters [17].

IGFR1

IGFR1 (Insulin-like growth factor 1 receptor) is an important gene activating PI3K/ACT signaling pathway. Interesting to note that its expression growth proportionally to BMI and is leads tocancer prognosis [18]. Levels of activation of IGFR1 were found to be significantly higher in patients with endometrial cancer and insulin-resistant Diabetes II [19].

PPT1

Palmitoyl-protein thioesterase 1 (PPT1) is a lipoprotein;it has been shown that PPT1 gene is overexpressed in insulin-resistant patients [20]. PPT1 also significantly promotes tumor growth; PPT1 plays an important role in autophagy in cancer cells along with activation of mTORC1 [21].

SOCS3

As pointed by Sutherland and colleagues, silencing of the suppressor of cytokine signaling‑3 (SOCS3) by hypermethylationmay promote the oncogenic transformation of epithelial tissues and support growth of tumors [22]. It brings this gene in the set of tumor-suppressing genes. In obesity, SOCS3 plays a role of negative regulator of leptin-activation signaling actually increasing the leptin-resistance [23]. So, we can speculate that in both cases—cancer development and obesity methylation—silencing of SOCS3 would lead to disease enhancement.

BTK

BTK (Bruton’s tyrosine kinase) functions as a dual regulator of apoptosis and plays numerous other roles in cancer development [24]. BTK also plays a significant role in obesityleading to insulin resistance [25].

TTK

TTK protein kinase is a dual specificity protein kinase associatedwith cell proliferation. TTK has been found to be expressed in cancers such as pancreatic ductal adenocarcinoma, where cell proliferation was weakened when TTK was knocked down [26]. High expression levels of TTK was also found in clear cell renal cell carcinoma cells, and cell proliferation and invasion were suppressed with the knockdown of TTK [27].

SGK1

SGK1 (Serum/glucocorticoid regulated kinase 1) is involved in cellular stress response. SGK1 is known to be stimulated by insulin-related diseases as well as hyperglycemia [28]. SGK1 has also been found to have positive correlations with cancer progression and metastasis [29].

Network 2

Network 2 also contains dual-function genes. Several examples are presents below (Figure 3b).

TP53

Function of TP53 as a tumor suppressor is well known and does not need more explanation. In the same time not very obvious its role in obesity. Interesting findings demonstrated Morikawa and colleagues [30]. They found that TP53 expression changes to increase (positive) in patient with close to normal BMI. In the patients in obese category TP53 mostly has decreased expression (negative).Among the patients of invasive bladder cancer normal weight, overweight, and patients with obesity had a TP53 mutation frequency of 68.4%, 44.8%, and 25%, respectively (P < 0.05)” [31]. It means that mutational frequency of this gene is lower in obese patients. These fact that cancer rate is higher and TP53 expression is lowerin patients with lower BMI leads to a suggestion that obesity plays a significant role in cancer independently of the TP53 mutations. It can be related to higher probability of cancer proliferation in the adipose cells, more intensive development of blood vesicles in the tumors of obese patients, and other factors.Interesting to note that we found the same relations in some viral proteins impact to cancer development. We elucidated that presence of HPV viral proteins in patients with ovarian cancer leads to cancer development requiring less cancerogenic mutations than ovarian cancer without HPV (Shang, 2020 in press).

FOXO

FOXO (Forkhead box O3) phosphorylation leads to insulin resistance initiated by ER stress [32]. FOXO genes family has various roles in cancer development. Inactivation of FOXO function leads to uncontrolled cell proliferation and accumulation of DNA damage associated with cancer development [33].

PPARA

PPARA (Peroxisome proliferator-activated receptor-alpha) polymorphisms lead to cancer development [34]. It is also known that liver oncogenesis is significantly dependent on miRNAimpact on PPARA gene [35].

HIF1

HIF-1 (Hypoxia-inducible factor-1) regulates cellular response to hypoxia.As a transcription factors it increases concentration of genes involved in angiogenesis, cell survivaland other cancer-related activities [36]. In the same time HIF1 activation leads to white adipose tissue expansion in obese patients [37].

PHB

PHB (Prohibitin) is a “pleiotropic protein that has roles in both adipocytes and immune cells” [38]. PHB has been found to be upregulated in tumor cells from leukemia and lymphoma and it is associated with cancer progression [39]. PHB expression has been linked to both diabetes and cancer [38].

BDNF

BDNF (Brain Derived Neurotrophic Factor) has been found to play a role in cancer cell growth and progression through the stimulation of oncogenic pathways [40]. BDNF expression and mutation have also been associated with obesity [41].

Signaling Pathways Analysis

Genes elucidated the IPA networks as interacting with the endometrial cancer-related metabolites were analyzed by the DAVID program (Table 2). As expected, several KEGG pathways involving cancer were identified, including proteoglycans in cancer, prostate cancer, and pathways in cancer (Table 3). Moreover, several pathways that were identified were associated with obesity, which could reveal underlying mechanisms involving obesity and endometrial cancer.In order to understand the relationship between obesity and endometrial cancer, the adipocytokine signaling pathway, insulin resistance pathway, and glucagon signaling pathway from the list of these pathways were closely analyzed.

Adipocytokine Signaling Pathway

There is a high correlation between obesity and leptin concentration levels. Leptin is a hormone that regulates appetite, body mass composition, and energy expenditure. This hormone, while promoting lower food intake and fat loss, is often found in high concentrations alongside obesity when the hormone fails to induce weight loss. This effect is known as leptin resistance. Hyperleptinemia, where high levels of leptin are present in the bloodstream, is almost exclusively observed in obese animals. Leptin is primarily secreted in adipocytes and its expression is increased by obesity-related factors, such as overfeeding and insulin resistance [42,43]. The expression of leptin can be a risk factor for cancer. Leptin activates JAK, which then phosphorylates STAT3 (Figure 4). The activation of STAT3 is shown to significantly increase cell proliferation in endometrial cancer cells. When the phosphorylation of STAT3 was blocked with an inhibitor, the cell growth of the cancer cells decreased [2]. The activation of STAT3 leads to the induction of several oncogenic proteins, including c-Myc. In endometrial cancer cells, expression of the c-Myc gene increases cancer cell proliferation, and when the gene is knocked down, proliferation of the tumor cells dramatically decreases [6].

Figure 4. KEGG pathway of adipocytokine signaling pathway selected by DAVID program. Leptin is a hormone that regulates obesity-related functions, such as body mass, appetite, and food intake. Leptin is shown to activate JAK–STAT3 pathway known to drive the proliferation of cancer tumor cells.

Insulin Resistance Pathway

Insulin resistance characterized by resistance to the effects of insulin on glucose uptake is a commonly known result of obesity and is a fundamental aspect of type 2 diabetes. In obesity, weight gain causes an increase in islet β cells, which in its turn increases insulin production. It has been shown that with increased insulin production, hyperinsulinemia occurs and induces insulin resistance [44]. Insulin resistance is related toactivation both the STAT3 and PI3k–Akt pathways (Figure 5). Through IL-6, obesity activates the STAT3 protein, which is known to increase cell proliferation in endometrial cancer through induction of the c-Mycprotein [6]. Additionally, insulin resistance is related to activation the PI3k–Akt and the mTOR signaling pathways (Figure 5). mTOR plays an important role in glucose metabolism and is activated with increased levels of insulin and glucose. Hyperactive mTOR occurs with obesity and nutrient overload, likely due to hyperglycemia or hyperinsulinemia—bothinvolved in the development of insulin resistance and obesity [45]. The activation of mTOR can lead to an increased risk of cancer. mTOR activates S6K and eIF4E (Figure 5), which are responsible for protein synthesis and have been shown to promote cancer cell survival [45]. Obesity leads to the expression of TNF-ɑ (Figure 5). TNF-ɑ can stimulate estrogen synthesis and has a direct impact on cell proliferation in the endometrium; overexpression of TNF-ɑ has been linked to endometrial cancer risk [7].

Figure 5. KEGG pathway of insulin resistance from DAVID. Insulin resistance, which is commonly associated with obesity, activates both STAT3 and PI3k–Akt pathways. mTOR, activated by hyperinsulinemia and hyperactive in the obese population, can also lead to increased cancer risk.

Glucagon Signaling Pathway

Several meta-analyses have shown that a high glycemic load diet can increase the risk for endometrial cancer [3,46,47]. Glycemic load (GL) measures the effect a carbohydrate has on blood sugar levels as well as the amount consumed. Diets containing high GL with foods having high glycemic index are commonly associated with weight gain and obesity [48]. High GLs increases blood glucose and insulin levels, which leads to a decline in blood sugar levels within hours, ultimately creating a state of hunger that causes higher food intake and weight gain [49]. During the postprandial stage, a meal with the same nutrition but a higher GL can increase blood sugar levels two-fold, causing hyperglycemia. Hyperglycemia stimulates insulin production to lower blood sugar levels, while inhibiting glucagon release, which in turn would increase blood sugar levels [50]. Increased insulin production could potentially lead to hyperinsulinemia and insulin resistance, increasing the risk of cancer. Furthermore, glucagon has been shown to have anticancer properties on tumor growth as well as breast cancer [51,52]. Glucagon activates the FOXO signaling pathway and phosphorylates FOXO1 (Figure 6). FOXO1 is considered a tumor suppressor gene that is deactivated by insulin and the PI3k–Akt pathway. FOXO1 expression suppressed endometrial cancer cell proliferation in xenograft models and inhibited cell migration in Ishikawa endometrial cancer cells [8]. Diets with a high GL increase insulin levels while inhibiting glucagon release, reducing the activation of FOXO1 and its tumor suppressing properties, causing an increased risk of cancer.

Figure 6. KEGG’s glucagon signaling pathway selected by DAVID program. Foods with high glycemic loads cause hyperinsulinemia and inhibit glucagon release, potentially leading to insulin resistance and obesity. Glucagon, having anticancer properties, can lead to an increased risk of cancer when inhibited.

Similarly Expressed Genes Analysis

To understand more comprehensively the common molecular mechanisms standing behind similarities in endometrial (among the others) cancer with obesity and related events, we performed analysis with the program GeneFriends [12,13]. We submitted to the program our list of genes elucidated by IPA program on the basis of endometrial cancer metabolites. The resulting co-expression networkconsists of four clusters having increased numbers of interconnections (Figure 7). Gene Ontology (GO)—analysis of existing clusters of similarly expressed genes brought the following results.

Figure 7. Similarly expressed genes for the genes elucidated based on endometrial cancer metabolites. Green colored nodes represent genes from the original list.

Cluster 1

This contained mostly the genes related to immune response and related pathways activation (Table 4). These results support already pointed fact that immune response related processes are common for cancer and obesity. As noted by Pérez de Heredia and colleagues [53],“Obesity … is known to impair the immune function, altering leucocyte counts as well as cell-mediated immune responses. In addition, evidence has arisen that an altered immune function contributes to the pathogenesis of obesity”. Here, we see the obvious positive feedback when immune response alterations and obesity enhance each other leading to increase of both manifestations. In all cancers, inflammation is a part of the cancer development mechanism and immune response is common to tumor cells. Obesity is also related to inflammation. It is considered a condition causing chronic low-grade inflammation in patients [54]. Adipose tissue produced pro-inflammatory molecules including interleukins 1 and 6, TNFα, IFNγ, etc. and anti-inflammatory interleukins 3, 4, 10, and Ra [54,55].

Cluster 2

This contains a number of biochemical reactions-related processes including interactions of various molecules within the cell environment, metabolic processes (Table 5). We will show in some examples that many ofthese processes are common for obesity and cancer.

Lipid Metabolic Process(GO:0006629)

Long and colleagues [56] pointed that “…energy metabolism, especially lipid metabolism, is significantly elevated during carcinogenesis.Furthermore, abnormality of lipid metabolism promotes cancer development, invasion and metastasis…”. In breast cancer the concentration of total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) are significantly higher than in normal tissue. Moreover, concentration of TC and TG significantly higher in patients with metastases [56-58].

Oxoacid (Ketoacid) Metabolic Process

It was shown that key ketolytic enzymes BDH1 and OXCT1 can be biomarkers for possible ketogenic diet for cancer treatment [59]. Low expression level of these genes leads to sensitivity of cancer cells for low carbohydrates treatment.

Cluster 3

Thiscontains mostly processes-related to chromatin assembly, DNA processing, and epigenetic regulation (Supplemental Table S1). Experimental clinical studies of Ebot and colleagues [60] confirmed a link between obesity and prostate cancer through chromatin regulation. Their study demonstrated that in tumor tissues of very overweight/obese vs. healthy weight men, fifteen gene sets were enriched, while they did not change in normal tissues. They noted that patients with high tumor expression of chromatin-related genes had worse clinical characteristics and increased risk of lethal disease independent of grade and stage [60]. Wang with colleagues found that disruption of the bromodomain containing protein 2 (Bdr2), which plays a role in chromatin remodeling, leads to metabolically healthy obesity, with hyperinsulinemia, but enhanced glucose tolerance and low blood glucose, without T2D [61,62].Obesity causes changes in the acetylation distribution in specific cis-regulatory regions. This was supported by the fact of increased gene expression in the carcinogenic genes [63]. Authors also noted that premalignant lesions in obese patients’ colons have more probability to evolve to malignancy [63].

Cluster 4

.This contains mostly processes related to cell cycle (Supplemental Table S2). For obese patients, a number of cell-cycle-related genes were overexpressed including genes mostly involved in adipogenesis that most probably led to expansion of the adipose tissue during obesity [64]. Overexpressed genes promoting cell cycle were also found by Baranova and colleagues [65]. In cancers,defects of cell cycle is one of the most important molecular mechanisms of the disease: “Deregulation of the cell cycle engine underlies the uncontrolled cell proliferation that characterizes the malignant phenotype” [66]. Otto &Sicinski [67] pointed that cell cycle proteins are a promising target for anticancer drugs.

Discussion

Endometrial cancer is the cancer of the endometrium, contributing to an estimated to more than 10000 deaths in 2020. Obesity is recognized as one of the largest factors of increased risk and severity among different cancers, including endometrial cancer. There have been numerous population studies and meta-analyses supporting that an increase in BMI can significantly increase endometrial cancer risk. However, analysis of the molecular mechanisms involving both obesity and endometrial cancer can help elucidate their relationship and provide possibilities for new endometrial cancer therapies. Metabolites related to endometrial cancer obtained from the public sources were analyzed by the IPA program that created the gene–metabolites networks including the genes that interact with inputted metabolites. From the given list of endometrial cancer related genes and metabolites, DAVID Functional Annotation Tool identified several KEGG pathways involved in obesity, including insulin resistance, glucagon and adipocytokine signaling pathways.These pathways revealed potential targets for cancer treatment. For example, the STAT3 pathway, which was involved in both insulin resistance and adipocytokine signaling, could be targeted or inhibited which may reduce the effects of obesity on cancer risk. Lifestyle alterations, such as avoiding excessive food intake and high GL diets, can reduce metabolic syndrome, hyperinsulinemia, and weight gain, which can reduce endometrial cancer risk.In addition, further study on shared obesity and endometrial cancer biological processes via co-expression GO analysis may elucidate how their disease risks are positively correlated.

Data Availability

All data generated are presented in the article and supplemental materials.

Conflicts of Interest

The authors declare that there are no conflicts of interest.

Authors’ Contributions

VLK and IFT designed the study concept; KQW extracted the data; KQW created protein networks and extracted signaling pathways; BKP elucidated the genes with the similar expressions and did GO-analysis of the results; KQW, IFT, and VLK analyzed results; KQW, VLK, and IFT were involved in drafting the manuscript; IFT critically reviewed the manuscript for the intellectual quality of the study. The authors had full access to all data in the study. All authors have given their final approval to the final version of the manuscript.

References

1. Onstad MA, Schmandt RE, Lu KH (2016) Addressing the role of obesity in endometrial cancer risk, prevention, and treatment. Journal of Clinical Oncology 34: 4225-4230. [crossref]
2. Sharma D, Saxena NK, Vertino PM, Anania FA (2006) Leptin promotes the proliferative response and invasiveness in human endometrial cancer cells by activating multiple signal-transduction pathways. Endocrine-Related Cancer 13: 629-640.[crossref]
3. Xu WH, Xiang YB, Zhang X, Ruan Z, HuiCai, et al. (2014) Association of dietary glycemic index and glycemic load with endometrial cancer risk among Chinese women. Nutrition and Cancer67: 89-97.[crossref]
4. Thangaraju S, Subramani E, Chakravarty B, Chaudhury K (2012) Therapeutic targeting of the TNF superfamily: A promising treatment for advanced endometrial adenocarcinoma. Gynecologic Oncology127: 426-432.
5. Bajwa P, Nielsen S, Lombard JM, Rassam L, PravinNahar, et al. (2017) Overactive mTOR signaling leads to endometrial hyperplasia in aged women and mice. Oncotarget8: 7265-7275. [crossref]
6. Subramaniam KS, Tham ST, Mohamed Z, Woo YL, Adenan NAM, et al. (2013) Cancer-associated fibroblasts promote proliferation of endometrial cancer cells. PloS One8: e68923. [crossref]
7. Dossus L, Becker S, Rinaldi S, Lukanova A, Tjønneland A, et al. (2011) Tumor necrosis factor (TNF)‐α, soluble TNF receptors and endometrial cancer risk: The EPIC study.International Journal of Cancer129: 2032-2037. [crossref]
8. Zhang Y, Zhang L, Sun H, Lv Q, Qui C, et al. (2017) Forkhead transcription factor 1 inhibits endometrial cancer cell proliferation via sterol regulatory element-binding protein 1. Oncology Letters13: 731-737. [crossref]
9. Azvolinsky A (2016) The obesity–cancer link: A growing connection. Journal of the National Cancer Institute 108: djw243.
10. Renehan AG, Egger M, Zwahlen M (2010) Body mass index and cancer risk: The evidence for causal association. The Open Obesity Journal2: 12-22.
11. Krämer A, Green J, Pollard J, Tugendreich S (2014) Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics30: 523-530.
12. van Dam S, Cordeiro R, Craig T, van Dam J, Wood SH, et al. (2012) GeneFriends: An online co-expression analysis tool to identify novel gene targets for aging and complex diseases. BMC Genomics13: 535. [crossref]
13. van Dam S, Craig T, de Magalhães JP (2015)GeneFriends: A human RNA-seq-based gene and transcript co-expression database. Nucleic Acids Research43: D1124-1132. [crossref]
14. Bahado-Singh R (2017) U.S. Patent No. 0003291. Washington, DC: U.S. Patent and Trademark Office.
15. Gagliardi PA, Puliafito A, Primo L (2018) PDK1: At the crossroad of cancer signaling pathways. Seminars in Cancer Biology48: 27-35. [crossref]
16. Chakraborty A, Koldobskiy MA, Bello NT, Maxwell M, Potter JJ, et al. (2010) Inositol pyrophosphates inhibit Akt signaling, thereby regulating insulin sensitivity and weight gain. Cell 143: 897-910.[crossref]
17. Miranda DN, Coletta DK, Mandarino LJ, Shaibi GQ (2014) Increases in insulin sensitivity among obese youth are associated with gene expression changes in whole blood. Obesity22: 1337-1342. [crossref]
18. Joehlin-Price AS,Stephens JA, Zhang J, Backes FJ, Cohn DE, et al. (2016) Endometrial cancer insulin-like growth factor 1 receptor (IGFLR) expression increases with body mass index and is associated with pathologic extent and prognosis. Cancer Epidemiology, Biomarkers & Prevention25: 438-445.[crossref]
19. Dai C, Li N, Song G, Yang Y, Ning X (2016) Insulin-like growth factor 1 regulates growth of endometrial carcinoma through PI3k signaling pathway in insulin-resistant type 2 diabetes. American Journal of Translational Research8: 3329-3336.[crossref]
20. Yang X, Pratley RE, Tokraks S, Bogardus C, Permana P (2002) Microarray profiling of skeletal muscle tissues from equally obese, non-diabetic insulin-sensitive and insulin-resistant Pima Indians. Diabetologia45: 1584-1593. [crossref]
21. Nicastri MC, RebeccaVW, AmaravadiRK, Winkler JD (2017) Dimericquinacrines as chemical tools to identify PPT1, a new regulator of autophagy in cancer cells. Molecular & Cellular Oncology5:e1395504. [crossref]
22. Sutherland KD, Lindeman GJ, Choong DY, Wittlin S, Brentzell L, et al. (2004) Differential hypermethylation of SOCS genes in ovarian and breast carcinomas. Oncogene23: 7726‑7733.[crossref]
23. Wunderlich CM, Hövelmeyer N, Wunderlich FT (2013) Mechanisms of chronic JAK-STAT3-SOCS3 signaling in obesity. JAK-STAT2: e23878. [crossref]
24. Vassilev AO, Uckun FM (2004) Therapeutic potential of inhibiting Bruton’s tyrosine kinase, (BTK). Current Pharmaceutical Design10: 1757-1766. [crossref]
25. Winer DA, Winer S, Shen L, Wadia PP, Yantha J, et al. (2011) B cells promote insulin resistance through modulation of T cells and production of pathogenic IgG antibodies. Nature Medicine17:610-617. [crossref]
26. Kaistha BP, Honstein T, Müller V, Bielak S, Sauer M, et al. (2014) Key role of dual specificity kinase TTK in proliferation and survival of pancreatic cancer cells. British Journal of Cancer111: 1780-1787.[crossref]
27. Liu XD, Yao DW, Xin F (2019) TTK contributes to tumor growth and metastasis of clear cell renal cell carcinoma by inducing cell proliferation and invasion. Neoplasma66: 946-953. [crossref]
28. Lang F, Görlach A, Vallon V (2009) Targeting SGK1 in diabetes. Expert Opinion on Therapeutic Targets13: 1303-1311. [crossref]
29. Liu W, Wang X, Wang Y, Dai Y, Xie Y, et al. (2018) SGK1 inhibition-induced autophagy impairs prostate cancer metastasis by reversing EMT. Journal of Experimental & Clinical Cancer Research37: 73. [crossref]
30. Morikawa T, Kuchiba A, Liao X, Imamura Y, Yamauchi M, et al. (2012) Tumor TP53 expression status, body mass index and prognosis in colorectal cancer. International Journal of Cancer13: 1169-1178. [crossref]
31. Ecke TH, Schlechte HH, Gunia S, Lenk SV, Loening SA (2008) Body mass index (BMI) and mutations of tumor suppressor gene p53 (TP53) in patients with urinary bladder cancer.Urologic Oncology26: 470-473. [crossref]
32. Zhang W, Hietakangas V, Wee S, Lim SC, Gunarathe J, et al. (2013) ER stress potentiates insulin resistance through PERK-mediated FOXO phosphorylation. Genes & Development27: 441-449. [crossref]
33. Arden KC (2006) Multiple roles of FOXO transcription factors in mammalian cells point to multiple roles in cancer. Experimental Gerontology 41: 709­-717. [crossref]
34. Golembesky AK, Gammon MD, North KE, Bensen JT, Schroeder JC, et al. (2008) Peroxisome proliferator-activated receptor-alpha (PPARA) genetic polymorphisms and breast cancer risk: A Long Island ancillary study. Carcinogenesis29: 1944-1949.[crossref]
35. Drakaki A, Hatziapostolou M, Polytarchou C, Vorvis C, Poultsides GA, et al. (2015) Functional microRNA high throughput screening reveals miR-9 as a central regulator of liver oncogenesis by affecting the PPARA-CDH1 pathway. BMC Cancer15: 542. [crossref]
36. Kafshdooz L, Tabrizi AD, Mohaddes SM, Kafshdooz T, Akbarzadeh A, et al. (2014) The polymorphism of hypoxia-inducible factor-1a gene in endometrial cancer. Asian Pacific Journal of Cancer Prevention15: 10393-10396. [crossref]
37. Drareni K, Ballaire R, Barilla S, Mathew MJ, Toubal A, et al. (2018) GPS2 deficiency triggers maladaptive white adipose tissue expansion in obesity via HIF1A activation. Cell Reports24: 2957-2971. [crossref]
38. Mishra S, Nyomba BG (2017) Prohibitin – At the crossroads of obesity-linked diabetes and cancer. Experimental Biology and Medicine242: 1170-1177. [crossref]
39. Ross JA, Robles-Escajeda E, Oaxaca DM, Padilla DL, Kirken RA (2017) Theprohibitin protein complex promotes mitochondrial stabilization and cell survival in hematologic malignancies. Oncotarget8: 65445-65456. [crossref]
40. Meng L, Liu B, Ji R, Jiang X, Yan X, et al. (2019) Targeting the BDNF/TrkB pathway for the treatment of tumors. Oncology Letters17: 2031-2039. [crossref]
41. Sandrini L, Minno AD, Amadio P, Ieraci A, Tremoli E, et al. (2018) Association between obesity and circulating brain-derived neurotrophic factor (BDNF) levels: Systematic review of literature and meta-analysis. International Journal of Molecular Sciences19: 2281. [crossref]
42. Khan BB, Flier JS (2000) Obesity and insulin resistance. Journal of Clinical Investigation106: 473-481. [crossref]
43. Yang R, Barouch LA (2007) Leptin signaling and obesity:Cardiovascular consequences. Circulation Research101: 545-559. [crossref]
44. Ye J (2013) Mechanisms of insulin resistance in obesity. Frontiers in Medicine7: 14-24. [crossref]
45. Yoon M (2017) The role of mammalian target of rapamycin (mTOR) in insulin signaling. Nutrients9: 1176. [crossref]
46. Mulholland HG, Murray LJ, Cardwell CR, Cantwell MM (2008) Dietary glycaemic index, glycaemic load and endometrial and ovarian cancer risk: A systematic review and meta-analysis. British Journal of Cancer 99: 434-441. [crossref]
47. Nagle CM, Ibiebele TI, Webb PM (2013) Glycemic index, glycemic load and endometrial cancer risk: Results from the Australian National Endometrial Cancer study and an updated systematic review and meta-analysis. European Journal of Nutrition52: 705-715. [crossref]
48. Brand-Miller JC, Holt SHA, Pawlak DB, McMillan J (2002) Glycemic index and obesity.The American Journal of Clinical Nutrition76: 281S-285S. [crossref]
49. Bell SJ, Sears B (2003) Low-glycemic-load diets: Impact on obesity and chronic diseases. Critical Reviews in Food Science Nutrition 43: 357-377. [crossref]
50. Chiu C, Taylor A (2011) Dietary hyperglycemia, glycemic index and metabolic retinal diseases. Progress in Retinal and Eye Research30: 18-53. [crossref]
51. Sawin CT, Silbert CK, Mitchell ML (1975) The effect of glucose on the growth hormone response to glucagon and propranolol-glucagon in normal subjects. Metabolism24: 1009-1014. [crossref]
52. McCarty M (1999) Vegan proteins may reduce risk of cancer, obesity, and cardiovascular disease by promoting increased glucagon activity. Medical Hypotheses53: 459-485.
53. Pérez de Heredia F, Gómez-Martínez S, Marcos A (2012) Obesity, inflammation and the immune system. Proceedings of the Nutrition Society71: 332-338. [crossref]
54. Lee H, Lee IS, Choue R (2013) Obesity, inflammation and diet. Pediatric Gastroenterology,Hepatology& Nutrition16: 143-152. [crossref]
55. Trayhurn P, Wood IS (2004) Adipokines: inflammation and the pleiotropic role of white adipose tissue. British Journal of Nutrition92: 347-355. [crossref]
56. Long J, Zhang CJ, Zhu N, Du K, Yin YF, et al. (2018) Lipid metabolism and carcinogenesis, cancer development. American Journal of Cancer Research8: 778-791.[crossref]
57. Wei LJ, Zhang C, Zhang H, Wei X, Li SX, et al, (2016) A case-control study on the association between serum lipid level and the risk of breast cancer. Chinese Journal of Preventive Medicine50: 1091-1095.[crossref]
58. Zhu YR, Aupperlee MD, Zhao Y, Tan YS, Kirk EL, et al. (2016) Pubertal and adult windows of susceptibility to a high animal fat diet in Trp53-null mammary tumorigenesis. Oncotarget7: 83409-83423. [crossref]
59. Zhang J, Jia PP, Liu QL, Cong MH, Gao Y, et al. (2018) Low ketolytic enzyme levels in tumors predict ketogenic diet responses in cancer cell lines in vitro and in vivo. The Journal of Lipid Research59: 625-634. [crossref]
60. Ebot EM, Gerke T, Labbé DP, Sinnot JA, Zadra G, et al. (2017) Gene expression profiling of prostate tissue identifies chromatin regulation as a potential link between obesity and lethal prostate cancer. Cancer123: 4130-4138. [crossref]
61. Wang F, Liu H, Blanton WP, Belkina A, LeBrasseur NK, et al. (2009) Brd2 disruption in mice causes severe obesity without type 2 diabetes. Biochemical Journal425: 71-83. [crossref]
62. Wang F, Deeney JT, Denis GV (2013) Brd2 gene disruption causes “metabolically healthy” obesity: epigenetic and chromatin-based mechanisms that uncouple obesity from type 2 diabetes. Vitamins and Hormones91: 49-75. [crossref]
63. Li R, Grimm SA, Chrysovergis K, Kosak J,Wang X, et al. (2014) Obesity, rather than diet, drives epigenomics alterations in colonic epithelium resembling cancer progression. Cell Metabolism19: 702-711. [crossref]
64. Rodríguez-Acebes S, Palacios N, Botella-Carretero JI, Olea N, Crespo L, et al. (2010) Gene expression profiling of subcutaneous adipose tissue in morbid obesity using a focused microarray: Distinct expression of cell-cycle- and differentiation-related genes. BMC Medical Genomics3: 61. [crossref]
65. Baranova A,Collantes R, Gowder SJ, Elariny H, Schlauch K, et al. (2005) Obesity-related differential gene expression in the visceral adipose tissue. Obesity Surgery15: 758-765. [crossref]
66. Williams GH, Stoeber K (2012) The cell cycle and cancer. Pathology226: 352-364. [crossref]
67. Otto T, Sicinski P (2017) Cell cycle proteins as promising targets in cancer therapy. Nature Reviews. Cancer17: 93-115. [crossref]
68. Shang Z, Kouznetsova VL, Tsigelny IF (2020) Human papillomavirus (HPV) viral proteins substitute for the impact of somatic mutations by affecting cancer-related genes: meta-analysis and perspectives. Journal of Infectiology 3(1): 29-47. [crossref]