Plasma level of ATPase inhibitory factor 1 (IF1) is associated with type 2 diabetes risk in humans: a prospective cohort study

doi:10.21203/rs.3.rs-1819108/v1

Download PDF

Research Article

Plasma level of ATPase inhibitory factor 1 (IF1) is associated with type 2 diabetes risk in humans: a prospective cohort study

https://doi.org/10.21203/rs.3.rs-1819108/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Background: Mitochondrial dysfunction is associated with the development of type 2 diabetes (T2D). It is thus of clinical relevance to identify plasma biomarkers of mitochondrial dysfunction associated with the risk of T2D.

ATPase inhibitory factor 1 (IF1) endogenously inhibits mitochondrial ATP synthase activity. Here, we analyzed association of the plasma IF1 level with markers of glucose homeostasis and with the conversion to new-onset diabetes (NOD) in individuals with prediabetes.

Methods: In the IT-DIAB prospective study, the plasma level of IF1 was measured at baseline in 307 participants with prediabetes, defined as a moderately elevated fasting plasma glucose, i.e., 110–125 mg/dL (6.0–7.0 mmol/L). The primary outcome was the incidence of NOD, defined as the first fasting plasma glucose value ≥ 1.26 g/L (7.0 mmol/L) within five years of follow-up. Cross-sectional analysis of the IF1 level was also done in two independent interventional studies. Correlations between plasma IF1 and metabolic parameters at baseline were assessed by Spearman’s correlation coefficients, and the association with the risk of NOD was determined using Cox proportional‐hazards models.

Results: In IT-DIAB, the mean IF1 plasma level was lower in participants who developed NOD than in those who did not (537 ± 248 versus 621 ± 313 ng/mL, p=0.01). The plasma IF1 level negatively correlated with clinical variables associated with obesity and insulin resistance, including the BMI (r = -0.20, p=0.0005) and HOMA-IR (r = -0.37, p<0.0001). Conversely, IF1 was positively associated with plasma markers of cardiometabolic health, such as HDL-C (r = 0.63, p<0.0001) and apoA-I (r = 0.33, p<0.0001). These correlations were confirmed in cross-sectional analyses. In IT-DIAB, the IF1 level was significantly associated with a lower risk of T2D after adjustment for age, sex, and fasting plasma glucose (HR [95% CI] per 1 SD = 0.76 [0.64; 0.94], p=0.012).

Conclusion: We identified for the first time the mitochondrial-related biomarker IF1 as being associated with the risk of T2D. As a biomarker, circulating IF1 can be considered to be a surrogate of mitochondrial energetics. IF1 can potentially be used to select patients for clinical trials of therapeutics that target mitochondrial function.

New-onset diabetes

Type 2 diabetes

ATP synthase

IF1

mitochondria

High-Density Lipoprotein

Type 2 diabetes (T2D) is characterized by chronic hyperglycemia primarily caused by both impaired insulin secretion by pancreatic β-cells (insulinopenia) and defective insulin signaling in metabolically active tissues (insulin resistance). Over time, this chronic condition leads to damage to the heart, vasculature, eyes, kidney, brain, and nerves, resulting in reduced quality of life and life expectancy, with cardiovascular disease (CVD) as the primary cause of morbidity and mortality associated with T2D. Visceral obesity, low physical activity, a high caloric diet, and older age are the main risk factors for T2D. Metabolic disorders, associated with inflammatory and metabolic/oxidative stress, represent a primary defect in the pathogenesis of the disease. Given the essential role of mitochondria in aerobic metabolism and the regulation of oxidative stress, there is increasing evidence associating mitochondrial dysfunction in metabolically active tissues and pancreatic β-cells with the development of T2D [1, 2]. Particularly, the mitochondrial oxidative phosphorylation (OXPHOS) system, comprising the electron transport chain and ATP synthase, was reported to be impaired in muscle samples from obese patients with T2D compared to non-diabetic obese controls [3].

It has been proposed that the OXPHOS dysfunction can lead to decreased β-oxidation and ATP production and increased ROS production, resulting in insulin resistance and defective ATP-dependent insulin secretion [4–8]. It is thus of clinical relevance to identify plasma biomarkers of OXPHOS dysfunction and to explore their association with the risk of T2D.

The nuclear-encoded protein ATPase inhibitory factor 1 (IF1) is an endogenous inhibitor of the mitochondrial ATP synthase. For a long time, IF1 was thought to act only as an inhibitor of the reverse ATPase activity of the ATP synthase [9]. However, recent data indicate that IF1 also partially inhibits the synthetic activity of the ATP synthase in mitochondria, thus limiting OXPHOS [9]. Meanwhile, IF1 has been found in the systemic circulation, and studies have reported a negative association between circulating IF1 and cardiovascular risk [10–12].

In the present work, we analyzed association of the level of plasma IF1 with lipid and glucose homeostasis parameters in studies of patients of different ages and body weights. We also analyzed association of serum IF1 with the risk of conversion to new-onset diabetes (NOD).

Study participants

The study populations belong to the IT-DIAB (Therapeutic Innovation in Type 2 DIABetes; NCT01218061; and DiabeNord, NCT01432509), MONA (Metabolism Obesity Nutrition Age; NCT02161926), and MAPT (Multidomain Alzheimer Preventive Trial; NCT00672685) studies. These studies were performed in accordance with the ethical principles of the Declaration of Helsinki. All participants received oral information from the investigators and signed an informed consent form before they participated in any of the study procedures.

IT-DIAB

The IT-DIAB study is a 5-year prospective, observational, study designed to identify new biomarkers of T2D risk in a population with prediabetes. Details regarding the study design, recruitment, and procedures have been reported previously [13–15]. Briefly, prediabetes was defined by moderately elevated fasting plasma glucose with a baseline value of ≥ 110 and < 126 mg/dL (6.0–7.0 mmol/L). The follow-up of the patients was originally intended to be for 5 years or until the conversion to NOD (defined by a first fasting plasma glucose value ≥ 126 mg/dL (7.0 mmol/L) and/or a plasma glucose concentration ≥ 200 mg/dL (11.0 mmol/L) after a 2-h 75-g oral glucose tolerance test), introduction of an oral antidiabetic agent or insulin therapy before diabetes occurrence, bariatric surgery, or patient withdrawal, lost to follow-up, or death. The IT-DIAB study protocol was approved by the Ethics Committee of Tours University Hospital (CPP “Ouest I”, approval n ̊2010-S2); the DiabeNord study protocol was approved by the Ethics Committee of Amiens University Hospital (CPP “Nord-Ouest II”, approval n ̊2011/24).

MONA: The MONA interventional study enrolled obese men according to metabolism, obesity, nutrition, and age (MONA) [16]. The individuals were men who had been obese (body mass index between 30 and 40 kg/m²) for more than five years and had achieved a stable weight during the last three months before the study (< 5% change). The exclusion criteria were as follows: dyslipidemia, hypertension, type 1 or type 2 diabetes with insulin therapy, glycated hemoglobin (HbA_1c) > 7%, history of CVD, history of bariatric surgery, dietary restriction, smoking > 10 cigarettes/day, alcohol consumption > 66 g/day, and known malignancy. The study was approved by the French Comité de Protection des Personnes and the Agence Nationale de Sécurité du Médicament et des Produits de Santé.

MAPT: The MAPT interventional study included community-dwelling men and women aged 70 years or older without dementia but who met one of the following conditions: spontaneous memory complaints, gait speed of 0.8 m/s or less, and/or difficulties with performing at least one instrumental daily-living activity [17]. Hypertension and diabetes status, defined as having type 1 or 2 diabetes mellitus or not, were based on the participants’ medical histories and declarative information [18]. The primary objective was to investigate the effect of omega-3 polyunsaturated fatty acids (PUFA) supplementation and a multidomain intervention (including physical activity counseling, cognitive training, and nutritional advice) in preventing cognitive decline [19]. The protocol was approved by the French Health Authority and authorized by the advisory Committee for the Protection of Persons involved in Biomedical Research of the Toulouse University Hospital. Biological samples were stored at the Toulouse Hospital biobank (Toulouse Bio-Ressources).

Data collection

At baseline, all patients from the three studies underwent a medical examination, and a blood sample was taken under fasting conditions to constitute a biocollection. The medical examinations included anthropometric measurements (body weight, height, body mass index (BMI), and waist circumference).

Biochemical analyses

Biological analyses including plasma triglycerides (TG), total cholesterol, HDL-cholesterol (HDL-C), glucose, glycated hemoglobin (HbA1c), aspartate transaminase (AST), alanine transaminase (ALT), gamma-glutamyl transferase (γ-GT), insulin and high-molecular-weight adiponectin (HMW-adiponectin) were performed as previously described [14, 16, 18]. Plasma apoA-I and IF1 were measured by a liquid chromatography–tandem mass spectrometry (LC-MS/MS) method previously developed in our laboratory [20, 21]. Briefly, apolipoproteins and IF1 were quantified in a 40 µL aliquot (EDTA plasma) using trypsin proteolysis followed by analysis of the proteotypic peptides. The intra- and inter-assay variabilities never exceeded 9.4% and 14.2% for apolipoproteins and IF1, respectively. HOMA-IR (homeostasis model assessment of insulin resistance) was defined according to the equation proposed by Matthews et al. [22]. The insulin sensitivity index was calculated from hyperinsulinemic-euglycemic clamp data as: [M value/(Mean blood glucose at clamp steady-state (mM) x 0.18)]/Mean blood insulin at clamp steady-state (µU/mL).

Outcome: Plasma IF1

For the present analysis, the plasma IF1 level was analyzed at baseline in 12 participants from the group aged 30–40 years (mean age 36.0 ± 2.6 years) enrolled in the MONA study and in 307 individuals from the IT-DIAB study, aged 23–79 years (mean age 57.3 ± 9.9 years) who had at least one follow-up visit and for whom the concentrations of apolipoproteins were available (71.4% of the entire cohort) (Fig. 1). Cross-sectional analysis of the plasma IF1 level was also done in 1,097 participants from the MAPT study, aged 70 years or older (mean age 76.3 ± 4.4 years), one year after the beginning of the study (65.3% of the entire cohort) [19].

Statistical analyses

For all of the studies (IT-DIAB, MONA, and MAPT), categorical parameters are expressed as numbers (%) and quantitative parameters are expressed as means ± the standard deviation (SD) or medians (25th − 75th percentile) in case of a skewed distribution. The normality of the distribution was assessed graphically by drawing the histogram. The hypothesis of independence between two categorical parameters was tested using Fisher’s exact test. Between-group differences for quantitative variables were tested using Students’ t-test for unpaired series (Welch in case of heteroscedasticity) or the Wilcoxon rank-sum test, according to distributions. In order to study the relationship between two quantitative parameters, Spearman’s correlation coefficients were determined, and scatter plots were generated, including the linear regression line.

Regarding the follow-up analysis of the IT-DIAB population, Kaplan-Meier curves were generated for the risk of NOD, using the three groups defined by the two IF1 terciles, along with the log-rank test. We also performed a survival analysis using Cox regression models based on the proportional-hazards assumption, with time before NOD as the explanatory variable. All quantitative parameters were systematically standardized. Schoenfeld’s residuals were tested for the main parameter of interest, IF1. The model adjustment was based on background knowledge to observe the influence of the different covariates on the hazard ratio (HR) associated with IF1.

All analyses were performed on available data, using the statistical softwares R, version 4.0.0 [23] and IBM SPSS statistics version 23 (Chicago, IL, USA). No imputation was performed. A p-value < 0.05 was considered to be significant, without correction for multiple testing.

Baseline characteristics of the participants

In the IT-DIAB study, IF1 was measured at baseline in 307 prediabetic individuals who had at least one follow-up visit and for whom the concentration of apolipoproteins was available (71.4% of the entire cohort) (Fig. 1). The median follow-up period was 4.9 years (25th − 75th percentile 2.1–5.1, mean 3.8 years). In Table 1, the baseline characteristics are presented for the entire study population and for the two groups: the “control group” (individuals without NOD during the follow-up, n = 192) and the “NOD group” (individuals with conversion to NOD during the follow-up, n = 115), amounting to a T2D conversion rate of 37.5%, similar to that observed in the entire cohort (35.3%). The entire study population was middle-aged (57.3 ± 9.9 years old) and overweight (BMI: 29.8 ± 6.2 kg/m²), with, by design, impaired fasting glucose (median value of 1.15 g/L, 25th − 75th percentile 1.12–1.19 g/L).

Table 1

IT-DIAB study: Baseline characteristics of the population, according to status for NOD at the end of the follow-up.
	Whole population n = 307	No NOD, n₁ = 192	Conversion to NOD, n₂ = 115	p-value	Available data (n₁/n₂)
Clinical
Age (years)	57.3 ± 9.9	57.2 ± 10.4	57.5 ± 9.0	0.77	(192/115)
Women	92 (30.0%)	54 (28.1%)	38 (33%)	0.37	(192/115)
Body mass index (kg/m²)	29.8 ± 6.2	28.8 ± 5.9	31.4 ± 6.3	0.001	(192/115)
Obesity¹	129 (42.0%)	69 (35.9%)	60 (52.2%)	0.006	(192/115)
Waist circumference (cm)	99.0 ± 14.6	96.8 ± 14.0	102.7 ± 14.9	0.001	(191/114)
Lipid-lowering therapy²	80 (26.1%)	47 (24.5%)	33 (28.7%)	0.42	(192/115)
Hypertension	117 (38.2%)	69 (36.1%)	48 (41.7%)	0.33	(191/115)
Biological (fasting)
Glucose homeostasis
Glucose (mg/dL)	115 (112–119)	114 (112–118)	116 (113–121)	0.002	(192/115)
HbA_1c (%)	5.9 ± 0.4	5.8 ± 0.4	6.0 ± 0.4	< 0.0001	(191/115)
Insulin (mIU/L)	11.2 (8.1–17.9)	10.4 (7.3–14.3)	14.9 (9.4–22.1)	< 0.0001	(182/103)
HOMA-IR	3.14 (2.28–5.24)	2.86 (1.99–3.94)	4.39 (2.88–6.72)	< 0.0001	(182/103)
Adiponectin (µg/mL)	3.37 (2.18–4.58)	3.62 (2.36–4.94)	3.03 (2.04–4.28)	0.009	(183/103)
Lipid profile
Triglycerides (g/L)	1.19 (0.86–1.67)	1.08 (0.83–1.54)	1.37 (0.94–1.73)	0.006	(191/113)
Total cholesterol (g/L)	2.17 ± 0.40	2.18 ± 0.40	2.15 ± 0.40	0.55	(191/114)
LDL-C (g/L)	1.37 ± 0.36	1.38 ± 0.37	1.36 ± 0.35	0.63	(188/112)
HDL-C (g/L)	0.53 ± 0.15	0.55 ± 0.16	0.50 ± 0.13	0.015	(191/113)
ApoA-I (g/L)	1.23 (1.06–1.43)	1.24 (1.06–1.44)	1.22 (1.05–1.42)	0.61	(192/115)
ApoB100 (g/L)	0.56 (0.44–0.71)	0.56 (0.45–0.72)	0 .54 (0 .42–0 .67)	0.43	(192/115)
Liver parameters
AST (IU/L)	24.8 ± 10.2	23.4 ± 8.5	27.0 ± 12.0	0.040	(95/65)
ALT (IU/L)	33.2 ± 19.3	32.0 ± 18.5	35.3 ± 20.7	0.17	(191/114)
γ-GT (IU/L)	49.4 ± 42.0	49.4 ± 45.4	49.6 ± 36.0	0.96	(187/114)
IF1 (ng/mL)	589 ± 293	621 ± 313	537 ± 248	0.010	(192/115)
Categorical parameters are expressed as the population size (%). Quantitative parameters are expressed as means ± standard deviation (SD) in the case of Gaussian distribution, otherwise as medians (25th − 75th percentile). The hypothesis of independence between two categorical parameters was tested using Fisher’s exact test. Between-group differences for quantitative variables were tested using Students’ t-test for unpaired series (Welch in case of heteroscedasticity) or the Wilcoxon rank-sum test, according to distributions.
ApoA-I: apolipoprotein A-I; ApoB100: apolipoprotein B100; ALT: alanine aminotransferase; AST: aspartate aminotransferase; BMI: body mass index; HDL-C: high-density lipoprotein cholesterol; γ-GT: gamma-glutamyl transferase; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; ISI: Insulin Sensitivity Index; LDL-C: low-density lipoprotein cholesterol; NOD: New-Onset Diabetes.
¹ Obesity was defined as BMI ≥ 30 kg / m²
² Statins and/or fibrates.

Regarding other cardiovascular risk factors, 117 participants (38.2%) were hypertensive and 80 (26.1%) were treated for dyslipidemia (mostly with statins: 87.5%). There was a higher percentage of obese subjects (BMI ≥ 30 kg/m²) in the NOD group (52.2% versus 35.9% in the control group, p = 0.006). There were noticeable differences in glucose homeostasis parameters between the two groups, with, as expected, significantly higher levels of fasting plasma glucose, insulin, HbA_1C, and HOMA-IR in the NOD group compared with the control group. Conversely, adiponectin was significantly lower in the NOD group. Significant higher AST levels were observed in the NOD group. In terms of plasma lipids, the TG level was significantly higher (p = 0.006) and the HDL-C level significantly lower (p = 0.015) in the NOD group. No difference between groups was observed for apoA-I.

Additional cross-sectional analyses were also performed in two independent interventional studies, MONA and MAPT, for which the baseline characteristics are reported in Supplementary Tables 1 and 2, respectively. The study population from the MONA study comprised obese young men without diabetes (n = 12, age 35.9 ± 2.7, BMI: 34.1 ± 2.8 kg/m²), while the individuals from the MAPT study (n = 1097) comprised both sexes (63.8% women), elderly (age: 76.3 ± 4.4 years), and slightly overweight (BMI: 26.2 ± 4.1 kg/m²). Comparison of the individual data for the MAPT study according to the diabetes status at baseline is presented in Supplementary Table S3. Statistical differences were found in terms of sex distribution, obesity, hypertension, and treatment for dyslipidemia. There was a higher percentage of obese subjects among the participants with diabetes (36.4% versus 14.5% in individuals without diabetes group, p < 0.0001). The proportion of individuals diagnosed with hypertension and treated for dyslipidemia was higher in diabetics than in non-diabetics (62.6% versus 48.6%, p = 0.006 and 48.6% versus 30.0%, p < 0.0001, respectively). Regarding plasma parameters related to lipids, the level of TG was significantly higher in participants with diabetes (p = 0.025), while total cholesterol (p < 0.0001), LDL-C (p < 0.0001), HDL-C (p < 0.0001), and apoA-I (p = 0.006) were lower.

Plasma IF1 level according to diabetes status, NOD conversion, and baseline characteristics

A lower level of plasma IF1 was observed in diabetic participants from the MAPT study (Supplementary Table S3, 521 ± 262 versus 580 ± 277 ng/mL in non-diabetic participants, p = 0.036) and in participants from the IT-DIAB study who developed NOD (Table 1, 537 ± 248 versus 621 ± 313 ng/mL in the control group, p = 0.010). Table 2 lists the differences between the IF1 level and several of the categorical variables at baseline in the IT-DIAB and MAPT studies. A lower median concentration of IF1 was observed in obese participants (p = 0.035 and 0.005, respectively). The median IF1 level was higher in women compared with men, albeit without reaching statistical significance in IT-DIAB (p = 0.11 versus < 0.0001 in MAPT). In both IT-DIAB and MAPT, no difference in the IF1 level was observed according to lipid-lowering treatments or hypertension.

Table 2

IT-DIAB and MAPT study: IF1 level according to the baseline characteristics
Variable	n (%)	IF1 level (ng/mL) in the subgroups (median (25th − 75th percentile))
Variable	n (%)	No	Yes	p-value
IT-DIAB (n = 307)
sex (women)	92 (30.0%)	524 (399–692)	631 (372–827)	0.11
Obesity¹	69 (35.9%)	574 (422–757)	496 (365–673)	0.035
Lipid-lowering therapies²	80 (26.1%)	545 (399–718)	512 (368–694)	0.36
Hypertension	117 (38.2%)	545 (400–711)	531 (379–749)	0.74
MAPT (n = 1097)
sex (women)	700 (63.8%)	518 (350–649)	607 (410–748)	< 0.0001
Obesity¹	182 (16.7%)	585 (394–718)	524 (359–679)	0.005
Lipid-lowering therapies²	349 (31.8%)	527 (393–710)	521 (380–731)	0.955
Hypertension	548 (49.9%)	578 (390–710)	571 (384–716)	0.598
Categorical parameters are expressed as the population size (%). Quantitative parameters are expressed as medians (25th − 75th percentile). Between-group differences for quantitative variables are tested using the Wilcoxon rank-sum test.
¹ Obesity was defined as BMI > 30 kg / m²
² Statins and/or fibrates.

Association between the plasma IF1 level and metabolic and anthropometric parameters

Figure 2 presents scatter plots of IF1 levels with different biological parameters and Spearman’s correlation coefficients (r) in the IT-DIAB, MAPT, and MONA cohort studies. In all studies, the IF1 level exhibited a negative association with BMI and plasma TG, while it was positively correlated with plasma HDL-C and apoA-I (Fig. 2). Figure 3 and Table 3 report IF1 association with other parameters that were only available in one of the studies. IF1 negatively correlated with HOMA-IR (IT-DIAB and MONA), HbA_1c (IT-DIAB), and fasting plasma insulin (IT-DIAB), while it was positively associated with adiponectin (IT-DIAB) and the insulin sensitivity index, as assessed during a hyperinsulinemic-euglycemic clamp (MONA). In addition, IF1 was found negatively correlated with ALT but not with AST (IT-DIAB, Table 3). To further explore the relationship between individuals’ anthropometric or metabolic parameters and IF1, the plasma IF1 level in MAPT and IT-DIAB was considered as the dependent variable in simple linear regression analyses (Supplementary Table S4). In both studies, obesity, BMI, and TG were negative determinants of IF1, whereas HDL-C and apoA-I were positively associated. To a lesser extent, total cholesterol was positively associated with IF1 in MAPT but not in IT-DIAB. No association was found with hypertension, both in MATP and IT-DIAB. In MAPT, diabetes status contributed positively and significantly to IF1 variability, while IF1 levels were not influenced by the intervention group. Similar observations were found in multiple linear regression analyses, adjusted for age, sex (both studies), and intervention group (MAPT).

Table 3

IT-DIAB study: Spearman’s rho correlation coefficients between IF1 and the parameters of interest at baseline
Variables	Spearman’s rho correlation coefficients	p-value
Clinical
Age	0.17	0.003
BMI	− 0.20	< 0.001
Waist circumference	− 0.24	< 0.0001
Biological (fasting)
Lipid profile
Triglycerides	− 0.41	< 0.0001
Total cholesterol	0.06	0.27
LDL-C	− 0.03	0.64
HDL-C	0.63	< 0.0001
ApoA-I	0.33	< 0.0001
ApoB100	− 0.14	0.015
Glucose homeostasis
Glucose	0.08	0.17
Insulin	− 0.38	< 0.0001
HbA_1c	− 0.13	0.026
HOMA-IR	− 0.37	< 0.0001
Adiponectin	0.39	< 0.0001
Liver function
AST	0.04	0.63
ALT	-0.12	0.029
γ-GT	-0.03	0.66
ApoA-I: apolipoprotein A-I; ApoB100: apolipoprotein B100; ALT: alanine aminotransferase; AST: aspartate aminotransferase; BMI: body mass index; γ-GT: gamma-glutamyl transferase; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; LDL-C: low-density lipoprotein cholesterol

Plasma IF1 level and risk of T2D

The results of the simple and multiple Cox model analyses in IT-DIAB are presented in Table 4. The baseline plasma level of IF1 was negatively associated with NOD. The unadjusted HR per 1 SD increase [95% CI] was 0.82 [0.67; 1.00] (p = 0.055). The association of IF1 with NOD was statistically significant after adjustment for age and sex (Model 2, HR per 1 SD = 0.80 [0.65; 0.98], p = 0.035) and after further adjustment on fasting plasma glucose (Model 3, HR per 1 SD = 0.76 [0.64; 0.94], p = 0.012). This association did not remain significant when TG (p = 0.081) or HbA_1c (p = 0.24) were introduced in the model (Models 4 and 5).

Table 4. IT-DIAB study: Association between baseline IF1 and the incidence of NOD during

follow-up

Models - Events/Population size (112/302 (37.1%))	HR for IF1 (+1 SD) [95% CI]	p-value
Model 1: IF1 alone	0.82 [0.67; 1.00]	0.055
Model 2: Model 1 + age + sex	0.80 [0.65; 0.98]	0.035
Model 3: Model 2 + glucose	0.77 [0.63; 0.95]	0.017
Model 4: Model 2 + triglycerides	0.83 [0.67; 1.02]	0.081
Model 5: Model 2 + HbA_1c	0.88 [0.71; 1.09]	0.24

Survival analysis using Cox models based on the proportional-hazards assumption, with time before NOD as the explanatory variable. All quantitative parameters were systematically standardized. Schoenfeld’s residuals were tested for the main parameter of interest, IF1. P-values are calculated for IF1 using the Wald test. HR: hazard ratios. NOD: New-Onset Diabetes; HR (95% CI): Hazard Ratio (95% confidence interval).

The present study was aimed at exploring association of the circulating IF1 level with lipid profiles, glucose homeostasis markers, and the risk of new-onset T2D in individuals with prediabetes.

In three distinct population studies covering people of different ages and body weight categories, the IF1 level was associated with a cardioprotective plasma lipid profile, being inversely correlated with TG and positively associated with the HDL markers HDL-C and apoA-I. Interestingly, these associations between IF1 and plasma lipids have been previously reported in the general population and in patients with coronary artery disease (CAD), in whom IF1 has been inversely and independently correlated with cardiovascular mortality [11, 12]. This reciprocal relationship between IF1 and HDL markers could be attributed to the inhibitory action of IF1 on the ATP synthase expressed at the cell surface as a receptor for apoA-I (herein conventionally referred to as ecto-F₁-ATPase) and whose extracellular ATPase activity contributes to HDL metabolism and HDL-mediated vasculoprotection [24–29], indicating a possible extended function of IF1 at the cell surface. Then, the inverse correlation between plasma IF1 and TG could be related to the well-documented inverse relationship between TG and HDL-C, mainly due to the activities of lipases and lipid transfer proteins, such as cholesteryl ester transfer protein (CETP), that contribute to HDL remodeling [30, 31].

In the population studied, we also confirmed a negative association between circulating IF1 and BMI and waist circumference [11], which are two primary anthropometric variables used to assess the risk of diabetes in routine clinical practice [32]. Accordingly, the plasma IF1 level was lower in obese participants, while no difference was observed according to hypertension or lipid-lowering treatments, as previously reported [11]. Additionally, we here report a strong negative association between IF1 and markers that reflect glucose homeostasis, including insulin, HbA_1c, and HOMA-IR. Consistent with this, IF1 positively correlated with adiponectin, a protein secreted from adipocytes whose plasma concentration is lower in obesity and T2D [33] as well as ALT, an indicator of liver disease. The relationship between IF1 and glucose homeostasis is also supported by genetics in humans. In a Korean male population, the IF1 genetic variant rs3767303 was independently associated with a lower prevalence of obesity and lower BMI, waist circumference, and HbA_1c [34]. More importantly, our study found that the plasma IF1 level was lower in diabetic participants from the MAPT study and in pre-diabetic individuals from the IT-DIAB study who later developed T2D. In this latest population at high risk, a significant negative association was observed between baseline IF1 and the incidence of NOD, independently of age and sex. It is noteworthy that this association was strengthened following further adjustment on fasting plasma glucose, while it was attenuated, becoming borderline significant, when additionally adjusted on TG.

Our study is the first one to report an inverse relationship between the plasma IF1 level and the risk of T2D. The plasma IF1 level may reflect biological processes that are not captured by conventional risk factors for diabetes or bio-clinical variables and current biomarkers of glycemic control. First, mitochondrial IF1 has been reported to partially inhibit the synthetic activity of ATP synthase and to repress OXPHOS, thereby resulting in an attenuated cellular ATP content in pancreatic β-cells, which in turn leads to impaired glucose-stimulated insulin secretion [35–37]. Second, IF1 protein has been reported to be overexpressed in skeletal muscles of obese and T2D patients, and its function in limiting OXPHOS in myotubes derived from these patients was associated with anaplerotic perturbations of the cellular metabolome, leading to limited β-oxidation of fatty acids and enhanced secretion of TNF-α, which could contribute to insulin resistance [38]. Altogether, these recent studies indicate that mitochondrial IF1 activity may participate in the development of T2D [9].

At present, it can be hypothesized that increased circulating IF1 can indicate optimal OXPHOS function and mitochondrial energetics. As a biomarker, circulating IF1 could be considered to be a surrogate of OXPHOS activity and mitochondrial energetics, with a high level in plasma indicating low mobilization of IF1 in mitochondria and optimal ATP synthase activity [9]. Several mitochondrial stress-related mitokines such as fibroblast growth factor 21 (FGF21), growth differentiation factor 15 (GDF15), and humanin (HN) have emerged as disease-specific plasma biomarkers [39–42]. Given that several studies have found evidence of mitochondrial dysfunction in insulin-resistant (pre)-diabetic humans [4–6], plasma IF1 quantification, either alone or as part of a panel with other mitokines, could serve to assess individual mitochondrial energy metabolism associated with protection against insulin resistance. Future studies should assess the potential of plasma IF1 quantification as a companion diagnostic test in clinical trials of antidiabetic drugs that have been reported to improve mitochondrial functions [4, 43, 44]. For instance, one could explore whether the plasma IF1 level helps identify patients who will best respond to SGLT2 inhibitors, for which the proposed mechanisms of action include improved ATP generation, energy metabolism, and mitochondrial functions [43, 45].

At this stage, the organ or tissue sources of plasma IF1, as well as the cellular mechanisms that govern IF1 release in the bloodstream, remain unknown. The functional importance of mitochondrial IF1 is indicated by its presence in all tissues and organs [9]. In humans, IF1 has been reported to be highly expressed in tissues with high energy demands, such as brain, heart, kidney, liver, and skeletal muscle [9, 38, 46], and it can be hypothesized that these organs may be the greatest contributors to the circulating level of IF1. Moreover, it is not known whether circulating IF1 can exert autocrine or endocrine actions. The absence of the mitochondrial targeting sequence (MTS) in the circulating form of IF1 [9, 10] suggests that IF1 first reaches the mitochondria before being secreted and that plasma IF1 is then no longer able to again reach the mitochondria (in the absence of an MTS). However, it is possible that plasma IF1 inhibits the activity of ecto-F₁-ATPase at the cell surface [25]. In this respect, recombinant IF1 (rIF1), produced without an MTS has been reported to compete the with apoA-I binding site on ecto-F₁-ATPase [28] and to control different physiological processes. For instance, treatment of mice with rIF1, which solely targets ecto-F₁-ATPase, has been shown to inhibit hepatic HDL uptake and apoA-I-mediated nitric oxide production while also improving glucose tolerance [24, 25, 34]. As a marker, plasma IF1 may reflect both cell surface and mitochondrial ATP synthase activity. Irrespective of the mechanisms and pathways beyond plasma IF1 secretion and activity, IF1 appears to have ample potential for T2D risk prediction.

The study had several limitations. The relatively small size of the IT-DIAB prospective cohort limited the statistical power. Consequently, the plasma IF1 level should be analyzed in larger prospective cohorts to further explore the inverse association between IF1 and T2D risk beyond conventional risk markers. Nevertheless, association between IF1, lipids, and glycemic markers were confirmed in cross-sectional analyses consisting of individuals of different ages and weight ranges. Additionally, our study does not allow definition of an optimal level of IF1 that would confer T2D risk protection. However, a major strength of our study is that IF1 was quantified using an LC-MS/MS analytical method, thus enabling accurate quantification of IF1 in human plasma and that can hence be proposed as a reference method suitable for clinical practice [20].

Conclusions

Our study is the first one to identify plasma IF1 as a determinant of T2D onset in high-risk populations, independently of age, sex, and fasting plasma glucose levels. IF1 measurements are foreseen within the framework of other prospective cohorts of individuals at different risks of T2D to more firmly establish the predictive value of IF1 measurements in the assessment of T2D risk along with established risk factors. Further studies are needed to validate the circulating IF1 level as a proxy of mitochondrial energy metabolism associated with protection against insulin resistance. In the future, determination of plasma IF1 levels may help to better stratify T2D risk in populations at high risk and/or in the general population, in addition to being used to evaluate future mitochondrial-targeted therapies.

Ethics approval and consent to participate

The IT‐DIAB (Innovation Thérapeutique‐Diabète) study is registered as NCT01218061 and NCT01432509. The IT-DIAB study protocol was approved by the Ethics Committee of Tours University Hospital (CPP “Ouest I”, approval n ̊2010-S2); the DiabeNord study protocol was approved by the Ethics Committee of Amiens University Hospital (CPP “Nord-Ouest II”, approval n ̊2011/24).

The MONA (Metabolism Obesity Nutrition Age) study is registered as NCT02161926. The MONA study protocol was approved by the French Comité de Protection des Personnes and the Agence Nationale de Sécurité du Médicament et des Produits de Santé.

The MAPT (Multidomain Alzheimer Preventive Trial) study is registered as NCT00672685. The MAPT study protocol was approved by the French Health Authority and authorized by the advisory Committee for the Protection of Persons involved in Biomedical Research of the Toulouse University Hospital.

All of the reported investigations have been carried out in accordance with the principles of the Declaration of Helsinki as revised in 2008. All participants provided their written informed consent for research use of the anonymized data collected according to local regulation.

Consent for publication

All authors consent to this publication.

Availability of data and materials

The datasets associated with the current study are not publicly available but may be provided by the corresponding author on reasonable request.

Competing interests

There are no competing interests related to this work to disclose.

Funding

The present work was performed in the context of the Inspire Program, a research platform supported by grants from the Région Occitanie Pyrénées-Méditerranée (Reference number: 1901175) and the European Regional Development Fund (ERDF) (Project number: MP0022856) and the HEPATOCARE Program, funded by the Région Occitanie Pyrénées-Méditerranée (Reference number: R20018BB no. 19014226). This study received funds from Alzheimer Prevention in Occitania and Catalonia (APOC Chair of Excellence - Inspire Program).

The IT-DIAB study was supported by a grant from Société Francophone du Diabète (SFD) and by an internal Grant of CHU Nantes (RC17_0470). The MAPT study was supported by grants from the Gérontopôle of Toulouse, the French Ministry of Health (PHRC 2008, 2009), Pierre Fabre Research Institute (manufacturer of the omega-3 supplement), ExonHit Therapeutics SA, and Avid Radiopharmaceuticals Inc. The promotion of this study was supported by the University Hospital Center of Toulouse. The data sharing activity was supported by the Association Monegasque pour la Recherche sur la maladie d’Alzheimer (AMPA) and the INSERM-University of Toulouse III UMR 1295 (CERPOP) Research Unit. The Mona study was by Région Occitanie Recherche en Transfert Clinique et Normalisation et Innovation Interdisciplinaire en Biosanté [Metabolism, Obesity, Nutrition, and Age (MONA)] and Investissements d’Avenir F-CRIN/FORCE Network.

Authors’ contributions

BV, SG, CT, DL, PSB, BC and LOM conceived and designed the research; JPDS, MC, TD and GC performed the experiments and laboratory analyses; JPDS, MW, JR, MC, AG, BP, CM, NV, YR, PSB, BC and LOM analyzed the data, performed statistical analyses and interpreted results; JPDS and LOM drafted manuscript; JPDS, MW, JR, MC, TD, GC, AG, BP, BV, SG, CT, DL, CM, NV, YR, PSB, BC and LOM edited and revised manuscript. All authors read and approved final version of manuscript.

Acknowledgements

We are grateful to the subjects for their participation in the IT‐DIAB, MAPT and MONA studies. We thank the Biological Resource Center for Biobanking (CHU Nantes, Hôtel Dieu, Centre de ressources biologiques (CRB), Nantes, F‐44093, France (BRIF: BB‐0033‐00040) and the Toulouse university hospital biobank (CRB-TBR) for providing the samples.

MAPT Study Group: Principal investigator: Bruno Vellas (Toulouse); Coordination: Sophie Guyonnet ; Project leader: Isabelle Carrié ; CRA: Lauréane Brigitte ; Investigators: Catherine Faisant, Françoise Lala, Julien Delrieu, Hélène Villars ; Psychologists: Emeline Combrouze, Carole Badufle, Audrey Zueras ; Methodology, statistical analysis and data management: Sandrine Andrieu, Christelle Cantet, Christophe Morin; Multidomain group: Gabor Abellan Van Kan, Charlotte Dupuy, Yves Rolland (physical and nutritional components), Céline Caillaud, Pierre-Jean Ousset (cognitive component), Françoise Lala (preventive consultation). The cognitive component was designed in collaboration with Sherry Willis from the University of Seattle, and Sylvie Belleville, Brigitte 1Gilbert and Francine Fontaine from the University of Montreal. Co-Investigators in associated centres : Jean-François Dartigues, Isabelle Marcet, Fleur Delva, Alexandra Foubert, Sandrine Cerda (Bordeaux); Marie-Noëlle-Cuffi, Corinne Costes (Castres); Olivier Rouaud, Patrick Manckoundia, Valérie Quipourt, Sophie Marilier, Evelyne Franon (Dijon); Lawrence Bories, Marie-Laure Pader, Marie-France Basset, Bruno Lapoujade, Valérie Faure, Michael Li Yung Tong, Christine Malick-Loiseau, Evelyne Cazaban-Campistron (Foix); Françoise Desclaux, Colette Blatge (Lavaur); Thierry Dantoine, Cécile Laubarie-Mouret, Isabelle Saulnier, Jean-Pierre Clément, Marie-Agnès Picat, Laurence Bernard-Bourzeix, Stéphanie Willebois, Iléana Désormais, Noëlle Cardinaud (Limoges); Marc Bonnefoy, Pierre Livet, Pascale Rebaudet, Claire Gédéon, Catherine Burdet, Flavien Terracol (Lyon), Alain Pesce, Stéphanie Roth, Sylvie Chaillou, Sandrine Louchart (Monaco); Kristel Sudres, Nicolas Lebrun, Nadège Barro-Belaygues (Montauban); Jacques Touchon, Karim Bennys, Audrey Gabelle, Aurélia Romano, Lynda Touati, Cécilia Marelli, Cécile Pays (Montpellier); Philippe Robert, Franck Le Duff, Claire Gervais, Sébastien Gonfrier (Nice); Yannick Gasnier and Serge Bordes, Danièle Begorre, Christian Carpuat, Khaled Khales, Jean-François Lefebvre, Samira Misbah El Idrissi, Pierre Skolil, Jean-Pierre Salles (Tarbes). MRI group: Carole Dufouil (Bordeaux), Stéphane Lehéricy, Marie Chupin, Jean-François Mangin, Ali Bouhayia (Paris); Michèle Allard (Bordeaux); Frédéric Ricolfi (Dijon); Dominique Dubois (Foix); Marie Paule Bonceour Martel (Limoges); François Cotton (Lyon); Alain Bonafé (Montpellier); Stéphane Chanalet (Nice); Françoise Hugon (Tarbes); Fabrice Bonneville, Christophe Cognard, François Chollet (Toulouse). PET scans group: Pierre Payoux, Thierry Voisin, Julien Delrieu, Sophie Peiffer, Anne Hitzel, (Toulouse); Michèle Allard (Bordeaux); Michel Zanca (Montpellier); Jacques Monteil (Limoges); Jacques Darcourt (Nice). Medico-economics group: Laurent Molinier, Hélène Derumeaux, Nadège Costa (Toulouse).

Biological sample collection: Bertrand Perret, Claire Vinel, Sylvie Caspar-Bauguil (Toulouse).

Safety management : Pascale Olivier-Abbal

DSA Group: Sandrine Andrieu, Christelle Cantet, Nicola Coley, Sophie Guyonnet

Author’s information

Not applicable

Montgomery MK, Turner N. Mitochondrial dysfunction and insulin resistance: An update. Endocrine Connections. 2015;4:R1–15.
Rovira-Llopis S, Bañuls C, Diaz-Morales N, Hernandez-Mijares A, Rocha M, Victor VM. Mitochondrial dynamics in type 2 diabetes: Pathophysiological implications. Redox Biology. 2017;2016:637–45.
Antoun G, McMurray F, Thrush AB, Patten DA, Peixoto AC, Slack RS, et al. Impaired mitochondrial oxidative phosphorylation and supercomplex assembly in rectus abdominis muscle of diabetic obese individuals. Diabetologia. 2015;58:2861–6.
Sivitz WI, Yorek MA. Mitochondrial dysfunction in diabetes: From molecular mechanisms to functional significance and therapeutic opportunities. Antioxidants and Redox Signaling. 2010;12:537–77.
Petersen KF, Befroy D, Dufour S, Dziura J, Ariyan C, Rothman DL, et al. Mitochondrial dysfunction in the elderly: Possible role in insulin resistance. Science (80-). 2003;300:1140–2.
Petersen KF, Dufour S, Befroy D, Garcia R, Shulman GI. Impaired Mitochondrial Activity in the Insulin-Resistant Offspring of Patients with Type 2 Diabetes. N Engl J Med. 2004;350:664–71.
Kelley DE, He J, Menshikova E V, Ritov VB. Dysfunction of mitochondria in human skeletal muscle in type 2 diabetes. Diabetes. 2002;51:2944–50.
Maechler P, Wollheim CB. Mitochondrial function in normal and diabetic β-cells. Nature. 2001;414:807–12.
Gore E, Duparc T, Genoux A, Perret B, Najib S, Martinez LO. The Multifaceted ATPase Inhibitory Factor 1 (IF1) in Energy Metabolism Reprogramming and Mitochondrial Dysfunction: A New Player in Age-Associated Disorders? Antioxid Redox Signal. 2022.
Genoux A, Pons V, Radojkovic C, Roux-Dalvai F, Combes G, Rolland C, et al. Mitochondrial Inhibitory Factor 1 (IF1) Is Present in Human Serum and Is Positively Correlated with HDL-Cholesterol. PLoS One. 2011;6:e23949.
Genoux A, Ruidavets J-B, Ferrières J, Combes G, Lichtenstein L, Pons V, et al. Serum IF1 concentration is independently associated to HDL levels and to coronary heart disease: the GENES study. J Lipid Res. 2013;54:2550–8.
Genoux A, Lichtenstein L, Ferrières J, Duparc T, Bongard V, Vervueren P-L, et al. Serum levels of mitochondrial inhibitory factor 1 are independently associated with long-term prognosis in coronary artery disease: the GENES Study. BMC Med. 2016;14:125.
Wargny M, Smati S, Pichelin M, Bigot-Corbel E, Authier C, Dierry V, et al. Fatty liver index is a strong predictor of changes in glycemic status in people with prediabetes: The IT-DIAB study. PLoS One. 2019;14:e0221524.
Ramin-Mangata S, Wargny M, Pichelin M, Le May C, Thédrez A, Blanchard V, et al. Circulating PCSK9 levels are not associated with the conversion to type 2 diabetes. Atherosclerosis. 2020;293:49–56.
Croyal M, Wargny M, Chemello K, Chevalier C, Blanchard V, Bigot-Corbel E, et al. Plasma apolipoprotein concentrations and incident diabetes in subjects with prediabetes. Cardiovasc Diabetol. 2022;21:21.
Vion J, Sramkova V, Montastier E, Marquès MA, Caspar-Bauguil S, Duparc T, et al. Metabolic and cardiovascular adaptations to an 8-wk lifestyle weight loss intervention in younger and older obese men. Am J Physiol - Endocrinol Metab. 2021;321:E325–37.
Vellas B, Carrie I, Gillette-Guyonnet S, Touchon J, Dantoine T, Dartigues JF, et al. MAPT STUDY: A MULTIDOMAIN APPROACH FOR PREVENTING ALZHEIMER’S DISEASE: DESIGN AND BASELINE DATA. J Prev Alzheimer’s Dis. 2014;1:13–22.
Raffin J, Angioni D, Giudici K V, Valet P, Aggarwal G, Nguyen AD, et al. Physical Activity, Body Mass Index, and Blood Progranulin in Older Adults: Cross-Sectional Associations in the MAPT Study. Journals Gerontol Ser A. 2022.
Andrieu S, Guyonnet S, Coley N, Cantet C, Bonnefoy M, Bordes S, et al. Effect of long-term omega 3 polyunsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): a randomised, placebo-controlled trial. Lancet Neurol. 2017;16:377–89.
Genoux A, Duparc T, Ruidavets JB, Ingueneau C, Najib S, Ferrières J, et al. A reference measurement of circulating ATPase inhibitory factor 1 (IF1) in humans by LC-MS/MS: Comparison with conventional ELISA. Talanta. 2020;219:121300.
Blanchard V, Garçon D, Jaunet C, Chemello K, Billon-Crossouard SS, Aguesse A, et al. A high-throughput mass spectrometry–based assay for large-scale profiling of circulating human apolipoproteins. J Lipid Res. 2020;61:jlr.D120000835.
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412–9.
RCoreTeam. A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna (2016). 2016;:2013. https://www.r-project.org/index.html.
Cabou C, Honorato P, Briceño L, Ghezali L, Duparc T, León M, et al. Pharmacological inhibition of the F1-ATPase/P2Y1 pathway suppresses the effect of apolipoprotein A1 on endothelial nitric oxide synthesis and vasorelaxation. Acta Physiol. 2019;226:e13268.
Martinez LO, Jacquet S, Esteve JP, Rolland C, Cabezon E, Champagne E, et al. Ectopic beta-chain of ATP synthase is an apolipoprotein A-I receptor in hepatic HDL endocytosis. Nature. 2003;421:75–9.
Castaing-Berthou A, Malet N, Radojkovic C, Cabou C, Gayral S, Martinez LO, et al. PI3Kβ Plays a Key Role in Apolipoprotein A-I-Induced Endothelial Cell Proliferation Through Activation of the Ecto-F1-ATPase/P2Y1 Receptors. Cell Physiol Biochem. 2017;42:579–93.
Cardouat G, Duparc T, Fried S, Perret B, Najib S, Martinez LO. Ectopic adenine nucleotide translocase activity controls extracellular ADP levels and regulates the F1-ATPase-mediated HDL endocytosis pathway on hepatocytes. Biochim Biophys Acta. 2017;1862:832–41.
Radojkovic C, Genoux A, Pons V, Combes G, De Jonge H, Champagne E, et al. Stimulation of Cell Surface F1-ATPase Activity by Apolipoprotein A-I Inhibits Endothelial Cell Apoptosis and Promotes Proliferation. Arter Thromb Vasc Biol. 2009;29:1125–30.
Cavelier C, Ohnsorg PM, Rohrer L, von Eckardstein A. The beta-Chain of Cell Surface F0F1 ATPase Modulates ApoA-I and HDL Transcytosis Through Aortic Endothelial Cells. Arter Thromb Vasc Biol. 2012;32:131–9.
Föger B, Ritsch A, Doblinger A, Wessels H, Patsch JR. Relationship of plasma cholesteryl ester transfer protein to HDL cholesterol: Studies in normotriglyceridemia and moderate hypertriglyceridemia. Arterioscler Thromb Vasc Biol. 1996;16:1430–6.
Hayek T, Azrolan N, Verdery RB, Walsh A, Chajek-Shaul T, Agellon LB, et al. Hypertriglyceridemia and cholesteryl ester transfer protein interact to dramatically alter high density lipoprotein levels, particle sizes, and metabolism: Studies in transgenic mice. J Clin Invest. 1993;92:1143–52.
Feller S, Boeing H, Pischon T. Body mass index, waist circumference, and the risk of type 2 diabetes mellitus: implications for routine clinical practice. Dtsch Arztebl Int. 2010;107:470–6.
Maeda N, Funahashi T, Matsuzawa Y, Shimomura I. Adiponectin, a unique adipocyte-derived factor beyond hormones. Atherosclerosis. 2020;292:1–9.
Kwak SY, Chung IH, Kang J, Perakakis N, Yoo EH, Lee J, et al. Sex specific effect of ATPase inhibitory factor 1 on body weight: studies in high fat diet induced obese mice and genetic association studies in humans. Metabolism. 2020;105:154171.
Kahancová A, Sklenář F, Ježek P, Dlasková A. Overexpression of native IF1 downregulates glucose-stimulated insulin secretion by pancreatic INS-1E cells. Sci Rep. 2020;10:1551.
Kahancová A, Sklenář F, Ježek P, Dlasková A. Regulation of glucose-stimulated insulin secretion by ATPase Inhibitory Factor 1 (IF1). FEBS Lett. 2018;592:999–1009.
Zhang K, Bao R, Huang F, Yang K, Ding Y, Lauterboeck L, et al. ATP synthase inhibitory factor subunit 1 regulates islet β-cell function via repression of mitochondrial homeostasis. Lab Investig. 2022;102:69–79.
Formentini L, Ryan AJ, Gálvez-Santisteban M, Carter L, Taub P, Lapek JD, et al. Mitochondrial H+-ATP synthase in human skeletal muscle: contribution to dyslipidaemia and insulin resistance. Diabetologia. 2017;60:2052–65.
Conte M, Sabbatinelli J, Chiariello A, Martucci M, Santoro A, Monti D, et al. Disease-specific plasma levels of mitokines FGF21, GDF15, and Humanin in type II diabetes and Alzheimer’s disease in comparison with healthy aging. GeroScience. 2021;43:985–1001.
Lehtonen JM, Auranen M, Darin N, Sofou K, Bindoff L, Hikmat O, et al. Diagnostic value of serum biomarkers FGF21 and GDF15 compared to muscle sample in mitochondrial disease. J Inherit Metab Dis. 2021;44:469–80.
Tsygankova PG, Itkis YS, Krylova TD, Kurkina M V., Bychkov IO, Ilyushkina AA, et al. Plasma FGF-21 and GDF‐15 are elevated in different inherited metabolic diseases and are not diagnostic for mitochondrial disorders. J Inherit Metab Dis. 2019;42:918–33.
Boutari C, Pappas PD, Theodoridis TD, Vavilis D. Humanin and diabetes mellitus: A review of in vitro and in vivo studies. World J Diabetes. 2022;13:213–23.
Mizuno M, Kuno A, Yano T, Miki T, Oshima H, Sato T, et al. Empagliflozin normalizes the size and number of mitochondria and prevents reduction in mitochondrial size after myocardial infarction in diabetic hearts. Physiol Rep. 2018;6:e13741.
de Marañón AM, Canet F, Abad-Jiménez Z, Jover A, Morillas C, Rocha M, et al. Does Metformin Modulate Mitochondrial Dynamics and Function in Type 2 Diabetic Patients? Antioxid Redox Signal. 2021;35:377–85.
Sawicki KT, Ben-Sahra I, McNally EM. Sglt2 inhibition on cardiac mitochondrial function: Searching for a sweet spot. Journal of the American Heart Association. 2021;10:e021949.
Esparza-Moltó PB, Nuevo-Taρioles C, Chamorro M, Nájera L, Torresano L, Santacatterina F, et al. Tissue-specific expression and post‐transcriptional regulation of the ATPase inhibitory factor 1 (IF1) in human and mouse tissues. FASEB J. 2019;33:1836–51.

No competing interests reported.

3.Sup.tablesR1.10.docx

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Plasma level of ATPase inhibitory factor 1 (IF1) is associated with type 2 diabetes risk in humans: a prospective cohort study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study participants

Data collection

Biochemical analyses

Outcome: Plasma IF1

Statistical analyses

Results

Baseline characteristics of the participants

Plasma IF1 level according to diabetes status, NOD conversion, and baseline characteristics

Association between the plasma IF1 level and metabolic and anthropometric parameters

Plasma IF1 level and risk of T2D

Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1