OBJECTIVE

The primary purpose of the current study was to test the hypothesis that the proinsulin–to–C-peptide (PI-to-CP) ratio, as an index of proinsulin secretion, would be higher and associated with indices of β-cell function in African American adults relative to European American adults without type 2 diabetes.

RESEARCH DESIGN AND METHODS

Participants were 114 African American and European American adult men and women. A 2-h oral glucose tolerance test was conducted to measure glucose, insulin, C-peptide, and proinsulin and derive indices of β-cell response to glucose. The Matsuda index was calculated as a measure of insulin sensitivity. The disposition index (DI), the product of insulin sensitivity and β-cell response, was calculated for each phase of β-cell responsivity. Pearson correlations were used to investigate the relationship of the PI-to-CP ratio with each phase of β-cell response (basal, Φb; dynamic, Φd; static, Φs; total, Φtot), disposition indices (DId, DIs, DItot), and insulin sensitivity. Multiple linear regression analysis was used to evaluate independent contributions of race, BMI, and glucose tolerance status on PI-to-CP levels before and after adjustment for insulin sensitivity.

RESULTS

African American participants had higher fasting and 2-h PI-to-CP ratios. The fasting PI-to-CP ratio was positively associated with Φb, and the fasting PI-to-CP ratio and 2-h PI-to-CP ratio were inversely associated with DId and insulin sensitivity only in African American participants.

CONCLUSIONS

The PI-to-CP ratio could be useful in identifying African American individuals at highest risk for β-cell dysfunction and ultimately type 2 diabetes.

The prevalence of type 2 diabetes continues to increase at alarming rates and is rapidly becoming the leading cause of disability and death worldwide (1). In the U.S., African American individuals have the highest prevalence of type 2 diabetes and several of the associated complications (heart disease, stroke, and kidney disease) (2). The reason for this disparity is multifactorial; however, physiological differences in insulin sensitivity and secretion likely play a role. Identification of African American adults who are at the highest risk for type 2 diabetes could allow time for early intervention to slow or prevent disease progression.

Individuals who self-report African American race/ancestry tend to have lower insulin sensitivity (35), higher acute insulin response to glucose (35), and lower hepatic insulin clearance (5,6). Over time, the exaggerated insulin response to glucose observed in African American individuals may lead to β-cell exhaustion, subclinical β-cell dysfunction, and ultimately, type 2 diabetes. It has been suggested that this elevated insulin response contributes to the pathogenesis of type 2 diabetes in this population (7). Given that declining β-cell function is a key component of type 2 diabetes pathophysiology and is responsible for many of the complications and progression of the disease, identification of individuals with early β-cell dysfunction, especially African American individuals, could allow adequate time for interventions.

Proinsulin, the precursor molecule to insulin, may be a practical marker for failing β-cell function (8). In individuals with normal glucose tolerance, little proinsulin is detected in circulation. However, as β-cells lose the ability to effectively process proinsulin to insulin, greater amounts of proinsulin are secreted (9,10). Higher levels of proinsulin relative to insulin are seen in subjects with impaired glucose tolerance (IGT) and type 2 diabetes (1012), and it is well known that proinsulin concentrations predict incident type 2 diabetes (1316). A small, observational study found that proinsulin concentrations not only predicted conversion from IGT to type 2 diabetes but also identified individuals with normal glucose tolerance who converted to type 2 diabetes (17). Therefore, even among individuals with normal glucose tolerance, variance in proinsulin secretion is sufficient to provide predictive information about disease development. Despite elevated concentrations of proinsulin relative to insulin are often recognized as a feature of β-cell dysfunction, how proinsulin concentrations compare with measures of β-cell function remains unclear. The most rigorous means of assessing β-cell function is with the hyperglycemic clamp (18). More feasible estimates can be made using mathematical modeling of the C-peptide response to glucose (19). β-Cell response is commonly expressed relative to insulin sensitivity to account for compensatory hyperinsulinemia. The disposition index (DI; β-cell response × insulin sensitivity) is an index that was developed as a measure of β-cell function. Among normal glucose-tolerant individuals, the DI is similar. However, as β-cell function declines (e.g., with IGT), the DI declines (19,20).

The primary purpose of this study was to determine whether the proportion of proinsulin secreted, reflecting impaired insulin processing by the β-cell is inversely associated with the ability of the β-cell to appropriately respond to a glucose stimulus. Because 50% of insulin secreted from the pancreas is extracted by the liver and never appears in the peripheral circulation, the evaluation of insulin secretion is best measured by assessing circulating C-peptide, which is secreted in equimolar amounts with insulin but does not undergo hepatic extraction (2123). Thus, to examine the proportion of proinsulin secreted, we normalized proinsulin to C-peptide using the proinsulin–to–C-peptide (PI-to-CP) ratio. We sought to test the hypothesis that the fasting and 2-h PI-to-CP ratio would be higher, positively associated with indices of β-cell responsivity, and inversely associated with the DI obtained from mathematical modeling of oral glucose tolerance test (OGTT) results in African American adults relative to European American adults without type 2 diabetes. A secondary purpose was to explore the PI-to-CP ratio and these associations while considering other factors, including insulin sensitivity, BMI, and glucose intolerance.

Study Design and Sample

This was a secondary analysis of a cross-sectional, observational study conducted at the University of Alabama at Birmingham (UAB) between 2013 and 2018. Owing to disparities in the risk for type 2 diabetes among African American and European American adults, the study was designed to investigate the relationship between race/ancestry, insulin sensitivity, and adiposity. Additional details regarding the original study can be found elsewhere (24). Participants were healthy African American and European American men and women, aged 19–45 years, who were recruited by public advertisement (flyers and newspaper ads). Race/ancestry was determined by self-report and by genetic admixture analysis, as described below. Recruited individuals were screened for glucose tolerance status with a 2-h 75-g OGTT, and those with 2-h glucose ≥200 mg/dL were excluded from participation. Other exclusion criteria were absence of regular menstrual cycle, pregnant, lactating, or postmenopausal, smoking, not weight stable (change in weight >2.5 kg in the previous 6 months), taking oral contraceptives, use of any medication known to affect carbohydrate or lipid metabolism or energy expenditure, and use of antihypertensive agents that affect glucose tolerance (e.g., thiazide diuretics at doses >25 mg/day, ACE inhibitors). Participants were instructed to maintain their usual activity level, avoid strenuous physical activity the day prior to testing, and avoid all physical activity on the morning of testing. Women were tested 3–7 days after cessation of menstruation, while in the follicular phase of the menstrual cycle. All study assessments were conducted at the core facilities of the Center for Clinical and Translational Science (CCTS), Nutrition Obesity Research Center (NORC), and Diabetes Research Center (DRC). The UAB Institutional Review Board approved the study.

Anthropometrics and Body Composition

Each participant underwent standard anthropometric measurements (weight and height), wearing light clothing and no shoes. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m2). Total body fat and lean mass were measured by DXA (iDXA instrument, GE Healthcare). Participants were scanned in light clothing while laying supine with arms at their side.

OGTT

All participants in the study completed a 2-h 75-g OGTT. Participants arrived in the fasting state, and venous access was obtained. Blood was obtained at −10 min and −5 min relative to glucose load ingestion. These fasting values were later averaged to create the fasting measure used for calculations. The oral glucose load was then administered at time 0, and participants had 5 min to consume it. Blood samples were taken at 10, 20, 30, 60, 90, and 120 min after glucose ingestion. Samples were processed for serum and stored at −85°C until assayed for glucose, insulin, C-peptide, and proinsulin. Proinsulin concentrations were measured at fasting and 120 min.

Normal glucose tolerance was defined as blood glucose <100 mg/dL at fasting and <140 mg/dL at 2 h. Impaired fasting glucose was defined as fasting glucose of 100–125 mg/dL. IGT was defined as 2-h glucose of 140–199 mg/dL. For statistical analyses, these last two categories were combined into the single category of IGT.

Oral C-Peptide Minimal Model and Insulin Sensitivity

Data from the OGTT were used to apply the minimal model of C-peptide secretion and kinetics originally developed for the intravenous glucose graded infusion (25). C-peptide kinetics are described using an established two-compartment model. Pancreatic secretion rate is the sum of glucose concentration and its rate of increase. The model allows for the estimation of indices of β-cell responsivity specific to basal (Φb), dynamic or first-phase (Φd), and static (Φs) insulin secretion. In addition, a single global index of total β-cell responsivity to glucose (Φtot) is calculated by combining the dynamic and static indices. Dynamic sensitivity reflects the stimulatory effect of the rate at which glucose increases upon the secretion of stored insulin. It is defined as the amount of insulin (per unit of C-peptide distribution volume) released in response to the maximum glucose concentration. Static sensitivity reflects the effect of glucose on β-cell secretion and is the ratio between the secretion rate and glucose concentrations above threshold at steady state. Modeling was conducted using MATLAB software (MathWorks, Natick, MA). The Matsuda index was calculated as a measure of insulin sensitivity (SI):

graphic
where fasting glucose (G) and insulin (I) are taken from time 0 of the OGTT, and mean data represent the average glucose and insulin obtained during the entire OGTT (26). The DI was calculated by multiplying each phase of β-cell response by SI: DId = Φd × SI; DIs = Φs × SI; and DItot = Φtot × SI (19).

Assays

Glucose was measured using a Sirrus analyzer (Stanbio, Boerne, TX). Insulin was assayed in duplicate using immunofluorescence technology (Tosoh A1A-900 immunoassay analyzer, Tosoh Corp., South San Francisco, CA), assay sensitivity was 0.5 μU/mL; interassay coefficient of variation (CV) was 3.95%, and intra-assay CV was 1.49%. C-peptide was also measured on the TOSOH analyzer; minimum assay sensitivity was 0.2 ng/ml; interassay CV was 6.81% and intra-assay was 1.67%. The assay for C-peptide had a cross-reactivity with proinsulin of 0.047% and 0.0% with insulin. Proinsulin was measured in duplicate using Millipore (Billerica, MA) Human Total Proinsulin ELISA kits, with 100% cross-reactivity to intact proinsulin and its major processed intermediate, des(31,32) proinsulin, and 81% cross-reactivity to its processed intermediate des(64,65) proinsulin. Minimum assay sensitivity was 0.6 pmol/L, interassay CV was 2.2%, and intraassay CV was 2.5%. Proinsulin cross-reactivity was 0.1% with both C-peptide and insulin. Proinsulin concentrations were normalized for C-peptide by taking the ratio of proinsulin in pmol/L to C-peptide in pmol/L (PI-to-CP).

Determination of Genetic Admixture

Admixture analysis was performed on study participants with available DNA samples (n = 104). Filtering, quality control, and merging of genetic data were performed using PLINK (version 1.9) and the gaston package in R software (2729). Study participants were genotyped with the Infinium Global Screening Array version 3.0 (Illumina, San Diego, CA) at the Genomics Center at the University of Minnesota. CEU (Northern Europeans from Utah) and YRI (Yoruba in Ibadan, Nigeria) reference samples from 1000 Genomes Project Phase 3 were used to estimate European and African ancestry (respectively), while Colombian, Pima, Maya, Surui, and Karitiana reference samples from the Human Genome Diversity Project were used to estimate Native American ancestry (30,31). Prior to analysis, quality control was performed both separately for the reference and study data sets and on the merged data sets. Any individuals and variants meeting the following criteria were removed: 1) nonautosomal single nucleotide polymorphisms (SNPs); 2) SNPs or samples with a call rate of ≤90%; 3) SNPs deviating from Hardy-Weinberg equilibrium (P < 10−7); 4) SNPs with minor allele frequency of ≤0.01; and 5) first-degree relatives. Supervised admixture analysis was performed with ADMIXTURE version 1.3.0 (32). The analysis was conducted with K = 3 clusters to infer ancestry fractions for individuals in the study data set via comparison with African, European, and Native American reference populations. Since DNA data were missing for 14 individuals, models were adjusted for self-reported race rather than admixture to maximize sample size. In those participants with genetic admixture data, results were confirmed to be similar regardless of whether self-reported race or admixture was used as a covariate.

Statistical Analyses

Data are presented as mean ± SD or median (95% CI). Comparisons of descriptive characteristics by self-reported race were made by χ2 for categorical variables and independent samples t tests for continuous variables. Differences in the fasting and 2-h PI-to-CP ratio according to race group were made using ANCOVA, while adjusting for BMI. Partial correlations were used to investigate the relationship of the PI-to-CP ratio with β-cell responsivity to glucose (Φb, Φd, Φs, Φtot), disposition indices (DId, DIs, DItot), and insulin sensitivity, while adjusting for BMI. Multiple linear regression analysis was used to evaluate independent contributions of race, BMI, and IGT status on PI-to-CP levels while adjusting for age and sex. Associations were evaluated before and after adjustment for insulin sensitivity. Linear relationships between the outcomes and continuous variables were confirmed with scatterplots. Residual normality was verified with histograms and quantile-quantile plots, while residual plots were examined to confirm homoscedasticity. No evidence of multicollinearity was observed in any of the models (all variance inflation factors <5). Assumptions for all other statistical tests (data normality and equal variances) were verified prior to analyses. Any nonnormally distributed variables were transformed to achieve a normal distribution. Analyses were performed with RStudio Statistical Software (R Core Team, 2022, version 4.1.3). Statistical tests were two-tailed with significance set at P < 0.05.

The analysis included 114 individuals. Sample characteristics are shown in Table 1 according to self-reported race. The study sample was 47% men, with a mean age of 29.36 ± 7.9 years and mean BMI of 27.26 ± 5.68 kg/m2. The mean fasting PI-to-CP ratio for the entire sample was 0.022 ± 0.0086, with a range of 0.0086 to 0.053. The mean 2-h PI-to-CP ratio for the entire sample was 0.03 ± 0.011, with a range of 0.011 to 0.08. African American participants had greater BMI and Φd and were less insulin sensitive than European American participants.

Table 1

Descriptive characteristics by self-reported race

African American participantsEuropean American participants
n = 54n = 60P value*
Age, years 30.3 ± 8.07 28.5 ± 7.71 0.23 
Sex, n   0.33 
 Male 23 31  
 Female 31 29  
African admixture, % 84.5 ± 6.6 0.6 ± 1.2 <0.0001 
European admixture, % 14.5 ± 6.5 98.1 ± 5.9 <0.0001 
Native American admixture, % 1 ± 0.8 1.3 ± 5.2 0.69 
BMI, kg/m2 29.1 ± 6.14 25.6 ± 4.68 <0.001 
Total fat mass, kg 28.6 ± 12.6 24 ± 10.17 0.04 
Total lean mass, kg 53.05 ± 11.9 48.61 ± 9.19 0.03 
IGT, n 11 0.28 
Fasting insulin, μIU/mL 6.86 (5.93, 7.94) 5.68 (4.9, 6.58) 0.07 
Fasting CP, nmol/L 0.49 (0.44, 0.54) 0.48 (0.43, 0.54) 0.97 
2-h CP, nmol/L 2.24 (2.02, 2.48) 2.3 (2.05, 2.59) 0.64 
Fasting PI, pmol/L 11.08 (9.29, 13.12) 8.62 (7.39, 10.04) 0.03 
2-h PI, pmol/L 69.3 (59.77, 80.39) 59.9 (50.89, 70.6) 0.19 
Fasting PI-to-CP ratio 0.023 (0.02, 0.025) 0.018 (0.016, 0.02) <0.001 
2-h PI-to-CP ratio 0.031 (0.028, 0.034) 0.026 (0.024, 0.029) 0.01 
HOMA-insulin resistance 1.53 (1.31, 1.78) 1.23 (1.04, 1.45) 0.09 
Insulin sensitivity 4.96 (4.29, 5.73) 6.3 (5.41, 7.34) 0.03 
Φb, 10−9 min−1 6.13 (5.56, 6.76) 5.98 (5.38, 6.64) 0.72 
Φd, 10−9 1,060 (908, 1,238) 689.96 (596.59, 797.95) <0.0001 
Φs, 10−9 min−1 83.19 (65.36, 105.87) 76.59 (64.1, 91.51) 0.66 
Φtot, 10−9 min−1 103.61 (84.51, 127.03) 91.77 (79.15, 106.41) 0.34 
African American participantsEuropean American participants
n = 54n = 60P value*
Age, years 30.3 ± 8.07 28.5 ± 7.71 0.23 
Sex, n   0.33 
 Male 23 31  
 Female 31 29  
African admixture, % 84.5 ± 6.6 0.6 ± 1.2 <0.0001 
European admixture, % 14.5 ± 6.5 98.1 ± 5.9 <0.0001 
Native American admixture, % 1 ± 0.8 1.3 ± 5.2 0.69 
BMI, kg/m2 29.1 ± 6.14 25.6 ± 4.68 <0.001 
Total fat mass, kg 28.6 ± 12.6 24 ± 10.17 0.04 
Total lean mass, kg 53.05 ± 11.9 48.61 ± 9.19 0.03 
IGT, n 11 0.28 
Fasting insulin, μIU/mL 6.86 (5.93, 7.94) 5.68 (4.9, 6.58) 0.07 
Fasting CP, nmol/L 0.49 (0.44, 0.54) 0.48 (0.43, 0.54) 0.97 
2-h CP, nmol/L 2.24 (2.02, 2.48) 2.3 (2.05, 2.59) 0.64 
Fasting PI, pmol/L 11.08 (9.29, 13.12) 8.62 (7.39, 10.04) 0.03 
2-h PI, pmol/L 69.3 (59.77, 80.39) 59.9 (50.89, 70.6) 0.19 
Fasting PI-to-CP ratio 0.023 (0.02, 0.025) 0.018 (0.016, 0.02) <0.001 
2-h PI-to-CP ratio 0.031 (0.028, 0.034) 0.026 (0.024, 0.029) 0.01 
HOMA-insulin resistance 1.53 (1.31, 1.78) 1.23 (1.04, 1.45) 0.09 
Insulin sensitivity 4.96 (4.29, 5.73) 6.3 (5.41, 7.34) 0.03 
Φb, 10−9 min−1 6.13 (5.56, 6.76) 5.98 (5.38, 6.64) 0.72 
Φd, 10−9 1,060 (908, 1,238) 689.96 (596.59, 797.95) <0.0001 
Φs, 10−9 min−1 83.19 (65.36, 105.87) 76.59 (64.1, 91.51) 0.66 
Φtot, 10−9 min−1 103.61 (84.51, 127.03) 91.77 (79.15, 106.41) 0.34 

Values are means ± SD, median (95% CI), or as indicated otherwise.

*

Significance for main effect of self-reported race. Bold values are statistically significant (P < 0.05).

Significant differences between European American and African American participants were observed in the PI-to-CP ratio, with African American individuals having higher levels of fasting and 2-h PI-to-CP ratio (Table 1). These differences remained significant upon adjustment for BMI (Fig. 1). Upon exclusion of those with IGT, the fasting and 2-h PI-to-CP ratios remained significantly higher in African American participants (P = 0.006 for fasting and P = 0.04 for 2-h; data not shown).

Figure 1

Fasting and 2-h PI-to-CP ratios by self-reported race. Data are adjusted for BMI and expressed as adjusted means ± SE. African American participants, n = 54; European American participants, n = 60. †P < 0.01, *P < 0.05.

Figure 1

Fasting and 2-h PI-to-CP ratios by self-reported race. Data are adjusted for BMI and expressed as adjusted means ± SE. African American participants, n = 54; European American participants, n = 60. †P < 0.01, *P < 0.05.

Close modal

Associations between the fasting PI-to-CP and 2-h PI-to-CP ratios with indices of β-cell responsivity, DIs, and insulin sensitivity are presented by self-reported race in Table 2. The fasting PI-to-CP ratio was positively associated with Φb only in African American participants. Additionally, the fasting PI-to-CP and 2-h PI-to-CP ratios were each inversely associated with DId only in African American participants (Supplementary Fig. 1). The fasting and 2-h PI-to-CP ratios were inversely associated with insulin sensitivity in African American participants, but not European American participants.

Table 2

Partial correlation coefficients for fasting and 2-h PI-to-CP ratios with indices of β-cell response, DIs, and insulin sensitivity

Fasting PI-to-CP ratio2-h PI-to-CP ratio
r, P valuer, P value
African American participantsEuropean American participantsAfrican American participantsEuropean American participants
β-Cell response     
 Φb (10−9 min−10.4, 0.01 0.02, 0.88 0.3, 0.05 −0.01, 0.95 
 Φd (10−9−0.22, 0.13 0.03, 0.82 −0.14, 0.33 0.07, 0.64 
 Φs (10−9 min−10.23, 0.1 0.08, 0.55 0.24, 0.09 0.1, 0.48 
 Φtot (10−9 min−10.16, 0.25 0.01, 0.92 0.12, 0.39 0.05, 0.69 
DIs     
 DId −0.39, 0.005 −0.04, 0.79 −0.35, 0.01 −0.04, 0.74 
 DIs 0.04, 0.79 −0.01, 0.94 0.08, 0.59 −0.03, 0.85 
 DItot −0.1, 0.53 −0.07, 0.59 −0.13, 0.4 −0.09, 0.48 
Insulin sensitivity −0.46, 0.003 −0.14, 0.34 −0.46, 0.002 −0.23, 0.13 
Fasting PI-to-CP ratio2-h PI-to-CP ratio
r, P valuer, P value
African American participantsEuropean American participantsAfrican American participantsEuropean American participants
β-Cell response     
 Φb (10−9 min−10.4, 0.01 0.02, 0.88 0.3, 0.05 −0.01, 0.95 
 Φd (10−9−0.22, 0.13 0.03, 0.82 −0.14, 0.33 0.07, 0.64 
 Φs (10−9 min−10.23, 0.1 0.08, 0.55 0.24, 0.09 0.1, 0.48 
 Φtot (10−9 min−10.16, 0.25 0.01, 0.92 0.12, 0.39 0.05, 0.69 
DIs     
 DId −0.39, 0.005 −0.04, 0.79 −0.35, 0.01 −0.04, 0.74 
 DIs 0.04, 0.79 −0.01, 0.94 0.08, 0.59 −0.03, 0.85 
 DItot −0.1, 0.53 −0.07, 0.59 −0.13, 0.4 −0.09, 0.48 
Insulin sensitivity −0.46, 0.003 −0.14, 0.34 −0.46, 0.002 −0.23, 0.13 

Data are adjusted for BMI. Bold values are statistically significant (P < 0.05).

Multiple linear regression analyses for the fasting and 2-h PI-to-CP ratios are shown in Table 3. BMI and race were both associated with the fasting PI-to-CP ratio, with a higher BMI associated with a higher fasting PI-to-CP ratio, and European American participants having a lower fasting PI-to-CP ratio compared with African American participants. Upon the inclusion of insulin sensitivity, race was the only variable associated with the fasting PI-to-CP ratio. Race was the only variable associated with the 2-h PI-to-CP ratio, with European American participants having a lower 2-h PI-to-CP ratio compared with African American participants. Neither glucose tolerance status nor BMI was associated with the 2-h PI-to-CP ratio. When insulin sensitivity was included in the model, race remained associated with the 2-h PI-to-CP ratio, with European American participants having lower 2-h PI-to-CP ratios compared with African American participants and lower insulin sensitivity trending toward an association with a higher 2-h PI-to-CP ratio (P = 0.05).

Table 3

Multiple linear regression models for fasting and 2-h PI-to-CP ratio

β (SE)P value
Dependent variable: fasting PI-to-CP ratio   
 Model 1, R2 = 0.27   
  European American race −0.31 (0.07) 0.0006 
  IGT (NGT is reference) 0.17 (0.1) 0.06 
  BMI 0.2 (0.01) 0.04 
 Model 2, R2 = 0.27   
  European American race −0.3 (0.07) 0.001 
  IGT (NGT is reference) 0.15 (0.1) 0.12 
  BMI 0.17 (0.01) 0.09 
  Insulin sensitivity −0.07 (0.01) 0.48 
Dependent variable: 2-h PI-to-CP ratio   
 Model 1, R2 = 0.16   
  European American race −0.26 (0.07) 0.008 
  IGT (NGT is reference) 0.14 (0.1) 0.14 
  BMI 0.1 (0.01) 0.32 
 Model 2, R2 = 0.19   
  European American race −0.22 (0.07) 0.03 
  IGT (NGT is reference) 0.09 (0.11) 0.39 
  BMI 0.03 (0.01) 0.76 
  Insulin sensitivity −0.2 (0.01) 0.05 
β (SE)P value
Dependent variable: fasting PI-to-CP ratio   
 Model 1, R2 = 0.27   
  European American race −0.31 (0.07) 0.0006 
  IGT (NGT is reference) 0.17 (0.1) 0.06 
  BMI 0.2 (0.01) 0.04 
 Model 2, R2 = 0.27   
  European American race −0.3 (0.07) 0.001 
  IGT (NGT is reference) 0.15 (0.1) 0.12 
  BMI 0.17 (0.01) 0.09 
  Insulin sensitivity −0.07 (0.01) 0.48 
Dependent variable: 2-h PI-to-CP ratio   
 Model 1, R2 = 0.16   
  European American race −0.26 (0.07) 0.008 
  IGT (NGT is reference) 0.14 (0.1) 0.14 
  BMI 0.1 (0.01) 0.32 
 Model 2, R2 = 0.19   
  European American race −0.22 (0.07) 0.03 
  IGT (NGT is reference) 0.09 (0.11) 0.39 
  BMI 0.03 (0.01) 0.76 
  Insulin sensitivity −0.2 (0.01) 0.05 

Standardized β coefficients are shown. All models are adjusted for age and sex. African American race is the reference group. NGT, normal glucose tolerance.

The purpose of this study was to determine whether proinsulin secretion, as reflected in the PI-to-CP ratio, is associated with indices of β-cell function in a sample of African American and European American adults without type 2 diabetes. The results of this study indicate that African American adults have higher fasting and 2-h PI-to-CP ratios than European American adults, and that fasting and 2-h PI-to-CP ratios are inversely associated with DId only in African American adults. These findings suggest that subclinical β-cell dysfunction may be more prevalent in African American adults than in European American adults. Whether this observed β-cell dysfunction is the result of life-long exposure to an “excessive” first-phase insulin response to glucose is not clear at this time. Future work is needed to investigate the usefulness of the PI-to-CP ratio for identifying individuals at elevated risk for β-cell decline and ultimately type 2 diabetes.

Proinsulin levels are well known to increase during type 2 diabetes disease progression, and elevated levels are considered a feature of β-cell dysfunction (9,12,33). It has been suggested that this is the result of impaired proinsulin processing and premature release of immature secretory granules, possibly due to an increased demand on the β-cells (9). However, how the PI-to-CP ratio compares with actual measures of β-cell response and function remains unclear. A small study by Fritsche et al. (34) evaluated the proinsulin-to-insulin (PI-to-I) ratio with measures of β-cell function derived from the hyperglycemic clamp in a group of young and old adults without type 2 diabetes. Results demonstrated the PI-to-I ratio was inversely associated with first-phase insulin secretion from the clamp (r = −0.6, P < 0.001; 2.5–5 min after the initial glucose bolus). We did not observe an association between the fasting or 2-h PI-to-CP ratios with Φd. Possibly if we had collected proinsulin data at 30 min (analogous to a first phase during the OGTT), we might have observed an association between the 30-min PI-to-CP ratio and Φd. Additionally, our sample was young and healthy, with a mean age of 29 years, and only a few individuals with IGT. The effect of age on insulin secretion has been shown to be partly attributable to impaired proinsulin processing (34); therefore, the association between the PI-to-CP ratio and phases of β-cell responsivity may possibly become more apparent with increasing age and/or deteriorating β-cell function.

African American individuals are known to exhibit a higher acute insulin response to glucose independent of insulin sensitivity (3,7,35,36). This has been observed in healthy, normoglycemic, African American children and adults (7,35,36) and has been attributed to both greater insulin secretion and lower hepatic insulin extraction (3,5,6). Over time, chronic upregulation of the β-cell response may contribute to β-cell exhaustion and subsequent dysfunction, potentially placing African American individuals at higher risk for type 2 diabetes. Similar to previous studies, African American participants in our study had higher Φd compared with European American participants (7,35,36). Further, our novel observation that African American adults have higher fasting PI-to-CP ratios, but not higher fasting insulin or C-peptide, could indicate a declining ability of the β-cells to efficiently process proinsulin to insulin. This is likely also reflected in the positive association between the fasting PI-to-CP ratio and Φb observed in African American participants.

One mechanism for interpreting β-cell function considering insulin sensitivity is using the DI, which can be calculated for each phase of β-cell responsivity (19). In individuals with normal glucose tolerance, insulin is released in a characteristic biphasic response, with the first phase defined as the initial burst of insulin from β-cells. The second phase then follows with a steady, longer lasting secretion of insulin (20). Early in type 2 diabetes disease progression, first-phase insulin secretion declines and is virtually absent in overt type 2 diabetes (37). We observed the fasting and 2-h PI-to-CP ratios to be inversely associated with the DId in African American participants, which could be reflective of failing β-cell compensation for insulin resistance. Further, the fasting and 2-h PI-to-CP ratios were inversely associated with insulin sensitivity in African American participants, but not European American participants, suggesting that insulin resistance may contribute to an increased workload on the β-cells and subsequent dysfunction.

Proinsulin concentrations differ with obesity and glucose tolerance status (33,38). However, proinsulin concentrations do not differ by obesity status among individuals of the same glucose tolerance status (33,39), suggesting that glucose tolerance, not obesity, affects β-cell function. Similarly, we observed that BMI was not associated with the PI-to-CP ratio when accounting for glucose tolerance status and insulin sensitivity (Table 3). However, obesity is associated with elevated β-cell response independent of insulin sensitivity, presumably as an adaptation to the higher proportion of body fat rather than a compensatory response to insulin resistance (40). Taken together, these observations indicate that although obesity is associated with a greater β-cell response to glucose, obesity per se does not impair β-cell function. The implication of these observations is that elevated β-cell response alone does not lead to β-cell dysfunction; presumably, a “second hit” is required. Further research is needed to understand the factors, both physiological and genetic, that precipitate β-cell dysfunction.

Strengths of the current study include rigorous measures of β-cell responsivity and function, insulin sensitivity, genetic admixture, and the diverse sample. Another strength of this study is the use of C-peptide to normalize for proinsulin concentrations rather than insulin, which is necessary to evaluate the proportion of processed insulin that is released. Approximately 50% of insulin secreted from the pancreas is extracted by the liver on the first pass and never appears in peripheral circulation. Therefore, the evaluation of β-cell secretory capacity can only be accomplished with the measurement of C-peptide, which is secreted in equimolar amounts with insulin but is not extracted by the liver (2123).

Limitations include the secondary analysis, cross-sectional study design, and sample size. Additionally, because of the secondary analysis, we were limited to the sample available to us, including few individuals with IGT and a younger age range. Another limitation is the lack of sociocultural and environmental factors, which may influence race disparities in chronic disease. The basis for the differences here is likely due in part to the plethora of sociocultural and environmental factors that contribute to health disparities.

Additionally, these differences could be genetic or even epigenetic (potentially from sociocultural/environmental factors). It is possible that our findings may lead to the hypothesis that a certain combination of genetic polymorphisms, passed through genetic disequilibrium to members of admixed populations, may render some individuals (regardless of self-reported ancestry) to be prone to β-cell dysfunction (and elevated proinsulin concentrations). Findings from such hypothesis testing could provide great insight into the understanding of genetic predisposition (rather than genetic determinism) in type 2 diabetes. Further, such results would provide objective scientific data that would serve to eliminate stereotypical social ideas of physiological inferiority and would inform potential therapies to prevent or treat β-cell dysfunction or type 2 diabetes among all people. Tackling the complex issue of the etiology of β-cell dysfunction depends on understanding the contributions of all relevant factors, including genetic, epigenetic, physiological, and sociocultural. Future work will be vital in identifying and understanding how the multitude of factors that play a role in health disparities interact to contribute to β-cell dysfunction, variability in proinsulin concentrations, and ultimately, type 2 diabetes.

In conclusion, the proportion of proinsulin secreted is higher in African American adults, perhaps reflecting subclinical β-cell dysfunction. Further, in African American adults, the proportion of proinsulin secreted is inversely associated with Φd response to glucose, possibly suggesting usefulness as a marker for the initial decline in β-cell function that precedes and leads to IGT and ultimately type 2 diabetes. These findings suggest that differences in proinsulin and β-cell function among healthy African American and European American adults may have implications with respect to disparities in metabolic outcomes. Large scale studies will need to confirm these findings before proinsulin can be validated as a potential tool for assessing β-cell dysfunction, particularly in high-risk populations, such as African American adults.

This article contains supplementary material online at https://doi.org/10.2337/figshare.21922860.

Acknowledgments. The authors thank the volunteers for their participation in this study. The authors would also like to gratefully acknowledge the UAB Metabolism Core Laboratory (Nutrition Obesity Research Center, Diabetes Research Center, and Center for Clinical and Translational Science) for their involvement in this study, the Genomics Center at the University of Minnesota, and Michael Crowley, PhD, at the UAB Genomic Core Laboratory, for DNA sample processing.

Funding. This study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases at the National Institutes of Health (R01DK096388), UAB Nutrition Obesity Research Center (P30DK56336), UAB Diabetes Research Center (P30DK079626), and UAB Center for Clinical & Translational Science Pilot Grant. C.A.C. was supported by National Heart, Lung, and Blood Institute of the National Institutes of Health award number T32HL105349.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. C.A.C. and B.A.G. were responsible for the analysis and interpretation of data and drafting of the manuscript. F.P. conducted mathematical modeling for β-cell responsivity. L.A.F. assisted with statistical analyses and conducted genetic admixture analysis. W.T.G. and B.A.G. obtained funding. B.A.G. designed the study. All authors reviewed and edited the manuscript and approved the final version. C.A.C. and B.A.G. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 81st Scientific Sessions of the American Diabetes Association, virtual meeting, 25–29 June 2021.

1.
Saeedi
P
,
Petersohn
I
,
Salpea
P
, et al.;
IDF Diabetes Atlas Committee
.
Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the International Diabetes Federation Diabetes Atlas, 9th edition
.
Diabetes Res Clin Pract
2019
;
157
:
107843
2.
Centers for Disease Control and Prevention
.
National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2020
.
Accessed 9 September 2022. Available from https://www.cdc.gov/diabetes/data/statistics-report/index.html
3.
Ellis
AC
,
Alvarez
JA
,
Granger
WM
,
Ovalle
F
,
Gower
BA
.
Ethnic differences in glucose disposal, hepatic insulin sensitivity, and endogenous glucose production among African American and European American women
.
Metabolism
2012
;
61
:
634
640
4.
Haffner
SM
,
D’Agostino
R
,
Saad
MF
, et al
.
Increased insulin resistance and insulin secretion in nondiabetic African-Americans and Hispanics compared with non-Hispanic whites. The Insulin Resistance Atherosclerosis Study
.
Diabetes
1996
;
45
:
742
748
5.
Osei
K
,
Schuster
DP
.
Ethnic differences in secretion, sensitivity, and hepatic extraction of insulin in black and white Americans
.
Diabet Med
1994
;
11
:
755
762
6.
Gower
BA
,
Granger
WM
,
Franklin
F
,
Shewchuk
RM
,
Goran
MI
.
Contribution of insulin secretion and clearance to glucose-induced insulin concentration in African-American and Caucasian children
.
J Clin Endocrinol Metab
2002
;
87
:
2218
2224
7.
Armiyaw
L
,
Sarcone
C
,
Fosam
A
,
Muniyappa
R
.
Increased β-cell responsivity independent of insulin sensitivity in healthy African American Adults
.
J Clin Endocrinol Metab
2020
;
105
:
e2429
e2438
8.
Given
BD
,
Cohen
RM
,
Shoelson
SE
,
Frank
BH
,
Rubenstein
AH
,
Tager
HS
.
Biochemical and clinical implications of proinsulin conversion intermediates
.
J Clin Invest
1985
;
76
:
1398
1405
9.
Kahn
SE
,
Halban
PA
.
Release of incompletely processed proinsulin is the cause of the disproportionate proinsulinemia of NIDDM
.
Diabetes
1997
;
46
:
1725
1732
10.
Saad
MF
,
Kahn
SE
,
Nelson
RG
, et al
.
Disproportionately elevated proinsulin in Pima Indians with noninsulin-dependent diabetes mellitus
.
J Clin Endocrinol Metab
1990
;
70
:
1247
1253
11.
Davies
MJ
,
Rayman
G
,
Gray
IP
,
Day
JL
,
Hales
CN
.
Insulin deficiency and increased plasma concentration of intact and 32/33 split proinsulin in subjects with impaired glucose tolerance
.
Diabet Med
1993
;
10
:
313
320
12.
Røder
ME
,
Porte
D
Jr
,
Schwartz
RS
,
Kahn
SE
.
Disproportionately elevated proinsulin levels reflect the degree of impaired B cell secretory capacity in patients with noninsulin-dependent diabetes mellitus
.
J Clin Endocrinol Metab
1998
;
83
:
604
608
13.
Loopstra-Masters
RC
,
Haffner
SM
,
Lorenzo
C
,
Wagenknecht
LE
,
Hanley
AJ
.
Proinsulin-to-C-peptide ratio versus proinsulin-to-insulin ratio in the prediction of incident diabetes: the Insulin Resistance Atherosclerosis Study (IRAS)
.
Diabetologia
2011
;
54
:
3047
3054
14.
Hanley
AJ
,
D’Agostino
R
Jr
,
Wagenknecht
LE
, et al.;
Insluin Resistance Atrherosclerosis Study
.
Increased proinsulin levels and decreased acute insulin response independently predict the incidence of type 2 diabetes in the insulin resistance atherosclerosis study
.
Diabetes
2002
;
51
:
1263
1270
15.
Nijpels
G
,
Popp-Snijders
C
,
Kostense
PJ
,
Bouter
LM
,
Heine
RJ
.
Fasting proinsulin and 2-h post-load glucose levels predict the conversion to NIDDM in subjects with impaired glucose tolerance: the Hoorn Study
.
Diabetologia
1996
;
39
:
113
118
16.
Zethelius
B
,
Byberg
L
,
Hales
CN
,
Lithell
H
,
Berne
C
.
Proinsulin and acute insulin response independently predict Type 2 diabetes mellitus in men--report from 27 years of follow-up study
.
Diabetologia
2003
;
46
:
20
26
17.
Pfützner
A
,
Hermanns
I
,
Ramljak
S
, et al
.
Elevated intact proinsulin levels during an oral glucose challenge indicate progressive β-cell dysfunction and may be predictive for development of type 2 diabetes
.
J Diabetes Sci Technol
2015
;
9
:
1307
1312
18.
Hannon
TS
,
Kahn
SE
,
Utzschneider
KM
, et al.;
RISE Consortium
.
Review of methods for measuring β-cell function: design considerations from the Restoring Insulin Secretion (RISE) Consortium
.
Diabetes Obes Metab
2018
;
20
:
14
24
19.
Cobelli
C
,
Toffolo
GM
,
Dalla Man
C
, et al
.
Assessment of beta-cell function in humans, simultaneously with insulin sensitivity and hepatic extraction, from intravenous and oral glucose tests
.
Am J Physiol Endocrinol Metab
2007
;
293
:
E1
E15
20.
Bergman
RN
,
Finegood
DT
,
Kahn
SE
.
The evolution of beta-cell dysfunction and insulin resistance in type 2 diabetes
.
Eur J Clin Invest
2002
;
32
(
Suppl. 3
):
35
45
21.
Faber
OK
,
Hagen
C
,
Binder
C
, et al
.
Kinetics of human connecting peptide in normal and diabetic subjects
.
J Clin Invest
1978
;
62
:
197
203
22.
Rubenstein
AH
,
Block
MB
,
Starr
J
,
Melani
F
,
Steiner
DF
.
Proinsulin and C-peptide in blood
.
Diabetes
1972
;
21
(
Suppl.
):
661
672
23.
Horwitz
DL
,
Starr
JI
,
Mako
ME
,
Blackard
WG
,
Rubenstein
AH
.
Proinsulin, insulin, and C-peptide concentrations in human portal and peripheral blood
.
J Clin Invest
1975
;
55
:
1278
1283
24.
Tay
J
,
Goss
AM
,
Garvey
WT
, et al
.
Race affects the association of obesity measures with insulin sensitivity
.
Am J Clin Nutr
2020
;
111
:
515
525
25.
Breda
E
,
Cavaghan
MK
,
Toffolo
G
,
Polonsky
KS
,
Cobelli
C
.
Oral glucose tolerance test minimal model indexes of beta-cell function and insulin sensitivity
.
Diabetes
2001
;
50
:
150
158
26.
Matsuda
M
,
DeFronzo
RA
.
Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp
.
Diabetes Care
1999
;
22
:
1462
1470
27.
Chang
CC
,
Chow
CC
,
Tellier
LC
,
Vattikuti
S
,
Purcell
SM
,
Lee
JJ
.
Second-generation PLINK: rising to the challenge of larger and richer datasets
.
Gigascience
2015
;
4
:
7
28.
The R Foundation
.
The R Project for Statistical Computing
.
2021
.
Accessed 9 September 2022. Available from https://www.R-project.org/
29.
The Comprehensive R Archive Network (CRAN)
.
Package ‘gaston’
.
Genetic Data Handling (QC, GRM, LD, PCA) & Linear Mixed Models. 2020. Accessed 10 September 2021. Available from https://CRAN.R-project.org/package=gaston
30.
Auton
A
,
Brooks
LD
,
Durbin
RM
, et al.;
1000 Genomes Project Consortium
.
A global reference for human genetic variation
.
Nature
2015
;
526
:
68
74
31.
Cavalli-Sforza
LL
.
The Human Genome Diversity Project: past, present and future
.
Nat Rev Genet
2005
;
6
:
333
340
32.
Alexander
DH
,
Novembre
J
,
Lange
K
.
Fast model-based estimation of ancestry in unrelated individuals
.
Genome Res
2009
;
19
:
1655
1664
33.
Shiraishi
I
,
Iwamoto
Y
,
Kuzuya
T
,
Matsuda
A
,
Kumakura
S
.
Hyperinsulinaemia in obesity is not accompanied by an increase in serum proinsulin/insulin ratio in groups of human subjects with and without glucose intolerance
.
Diabetologia
1991
;
34
:
737
741
34.
Fritsche
A
,
Madaus
A
,
Stefan
N
, et al
.
Relationships among age, proinsulin conversion, and beta-cell function in nondiabetic humans
.
Diabetes
2002
;
51
(
Suppl. 1
):
S234
S239
35.
Hannon
TS
,
Bacha
F
,
Lin
Y
,
Arslanian
SA
.
Hyperinsulinemia in African-American adolescents compared with their American white peers despite similar insulin sensitivity: a reflection of upregulated beta-cell function?
Diabetes Care
2008
;
31
:
1445
1447
36.
Chandler-Laney
PC
,
Phadke
RP
,
Granger
WM
, et al
.
Adiposity and β-cell function: relationships differ with ethnicity and age
.
Obesity (Silver Spring)
2010
;
18
:
2086
2092
37.
Pfeifer
MA
,
Halter
JB
,
Porte
D
Jr
.
Insulin secretion in diabetes mellitus
.
Am J Med
1981
;
70
:
579
588
38.
Koivisto
VA
,
Yki-Järvinen
H
,
Hartling
SG
,
Pelkonen
R
.
The effect of exogenous hyperinsulinemia on proinsulin secretion in normal man, obese subjects, and patients with insulinoma
.
J Clin Endocrinol Metab
1986
;
63
:
1117
1120
39.
Røder
ME
,
Dinesen
B
,
Hartling
SG
, et al
.
Intact proinsulin and beta-cell function in lean and obese subjects with and without type 2 diabetes
.
Diabetes Care
1999
;
22
:
609
614
40.
van Vliet
S
,
Koh
HE
,
Patterson
BW
, et al
.
Obesity is associated with increased basal and postprandial β-cell insulin secretion even in the absence of insulin resistance
.
Diabetes
2020
;
69
:
2112
2119
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.