[Skip to Content]
[Skip to Content Landing]
Figure 1.
Determination of Haptoglobin (Hp) Genotype by Nondenaturing Polyacrylamide Gel Electrophoresis
Determination of Haptoglobin (Hp) Genotype by Nondenaturing Polyacrylamide Gel Electrophoresis

Demonstration of the 4 possible Hp genotypes by polyacrylamide gel electrophoresis. Each of the 4 Hp genotypes has a signature banding pattern resulting from differences in stoichiometry of the polymeric composition of the different Hp genotypes.13 In lane 1, Hp 1-1 has a single band corresponding to the Hp 1-1 dimer (see the arrowhead pointing to the right). In lane 5, Hp 2-1m has a band present in Hp 1-1 and 2 additional bands corresponding to Hp 2-1 polymers (trimer and quatermer [see the arrowheads pointing to the left]). In lanes 2 and 4, Hp 2-1 has the bands present in Hp 2-1, as well as higher-order Hp 2-1 polymers. In lane 3, Hp 2-2 has only Hp 2-2 trimers or higher-order Hp 2-2 polymers.

Figure 2.
Mini-Mental State Examination (MMSE) Score Change Over 40 Months by Age at Randomization for Each Haptoglobin (Hp) Genotype
Mini-Mental State Examination (MMSE) Score Change Over 40 Months by Age at Randomization for Each Haptoglobin (Hp) Genotype

A-D, The 4 Hp genotypes are shown. LOESS indicates locally weighted smoothing.

Table 1.  
Baseline Characteristics of the Sample and by the Hp Genotype
Baseline Characteristics of the Sample and by the Hp Genotype
Table 2.  
Baseline MMSE Scores by Hp Genotype Using the Least Squares Methoda
Baseline MMSE Scores by Hp Genotype Using the Least Squares Methoda
Table 3.  
Decline in MMSE Score Over 40 Months by Hp Genotype Using the Least Squares Methoda
Decline in MMSE Score Over 40 Months by Hp Genotype Using the Least Squares Methoda
1.
Katz  MJ, Lipton  RB, Hall  CB,  et al.  Age-specific and sex-specific prevalence and incidence of mild cognitive impairment, dementia, and Alzheimer dementia in blacks and whites: a report from the Einstein Aging Study.  Alzheimer Dis Assoc Disord. 2012;26(4):335-343. doi:10.1097/WAD.0b013e31823dbcfcPubMedGoogle ScholarCrossref
2.
Demirovic  J, Prineas  R, Loewenstein  D,  et al.  Prevalence of dementia in three ethnic groups: the South Florida Program on Aging and Health.  Ann Epidemiol. 2003;13(6):472-478. doi:10.1016/S1047-2797(02)00437-4PubMedGoogle ScholarCrossref
3.
Krishnan  LL, Petersen  NJ, Snow  AL,  et al.  Prevalence of dementia among Veterans Affairs medical care system users.  Dement Geriatr Cogn Disord. 2005;20(4):245-253. doi:10.1159/000087345PubMedGoogle ScholarCrossref
4.
Carson  JA, Michalsky  L, Latson  B,  et al.  The cardiovascular health of urban African Americans: diet-related results from the Genes, Nutrition, Exercise, Wellness, and Spiritual Growth (GoodNEWS) trial.  J Acad Nutr Diet. 2012;112(11):1852-1858. doi:10.1016/j.jand.2012.06.357PubMedGoogle ScholarCrossref
5.
Lin  A, Lacy  ME, Eaton  C, Correa  A, Wu  WC.  Inflammatory obesity phenotypes, gender effects, and subclinical atherosclerosis in African Americans: the Jackson Heart Study.  Arterioscler Thromb Vasc Biol. 2016;36(12):2431-2438. doi:10.1161/ATVBAHA.116.307728PubMedGoogle ScholarCrossref
6.
Cushman  M, Cantrell  RA, McClure  LA,  et al.  Estimated 10-year stroke risk by region and race in the United States: geographic and racial differences in stroke risk.  Ann Neurol. 2008;64(5):507-513. doi:10.1002/ana.21493PubMedGoogle ScholarCrossref
7.
Ferdinand  KC, Nasser  SA.  Racial/ethnic disparities in prevalence and care of patients with type 2 diabetes mellitus.  Curr Med Res Opin. 2015;31(5):913-923. doi:10.1185/03007995.2015.1029894PubMedGoogle ScholarCrossref
8.
Giblett  ER.  Haptoglobin types in American Negroes.  Nature. 1959;183(4655):192-193. doi:10.1038/183192a0PubMedGoogle ScholarCrossref
9.
Maeda  N.  DNA polymorphisms in the controlling region of the human haptoglobin genes: a molecular explanation for the haptoglobin 2-1 modified phenotype.  Am J Hum Genet. 1991;49(1):158-166.PubMedGoogle Scholar
10.
Asleh  R, Briasoulis  A, Berinstein  EM,  et al.  Meta-analysis of the association of the haptoglobin genotype with cardiovascular outcomes and the pharmacogenomic interactions with vitamin E supplementation.  Pharmgenomics Pers Med. 2018;11:71-82. doi:10.2147/PGPM.S159454PubMedGoogle Scholar
11.
Costacou  T, Levy  AP.  Haptoglobin genotype and its role in diabetic cardiovascular disease.  J Cardiovasc Transl Res. 2012;5(4):423-435. doi:10.1007/s12265-012-9361-zPubMedGoogle ScholarCrossref
12.
Levy  AP, Asleh  R, Blum  S,  et al.  Haptoglobin: basic and clinical aspects.  Antioxid Redox Signal. 2010;12(2):293-304. doi:10.1089/ars.2009.2793PubMedGoogle ScholarCrossref
13.
Costacou  T, Rosano  C, Aizenstein  H,  et al.  The haptoglobin 1 allele correlates with white matter hyperintensities in middle-aged adults with type 1 diabetes.  Diabetes. 2015;64(2):654-659. doi:10.2337/db14-0723PubMedGoogle ScholarCrossref
14.
Costacou  T, Secrest  AM, Ferrell  RE, Orchard  TJ.  Haptoglobin genotype and cerebrovascular disease incidence in type 1 diabetes.  Diab Vasc Dis Res. 2014;11(5):335-342. doi:10.1177/1479164114539713PubMedGoogle ScholarCrossref
15.
Staals  J, Henskens  LH, Delanghe  JR,  et al.  Haptoglobin phenotype correlates with the extent of cerebral deep white matter lesions in hypertensive patients.  Curr Neurovasc Res. 2010;7(1):1-5. doi:10.2174/156720210790820163PubMedGoogle ScholarCrossref
16.
Staals  J, Pieters  BM, Knottnerus  IL,  et al.  Haptoglobin polymorphism and lacunar stroke.  Curr Neurovasc Res. 2008;5(3):153-158. doi:10.2174/156720208785425675PubMedGoogle ScholarCrossref
17.
Ravona-Springer  R, Heymann  A, Schmeidler  J,  et al.  Haptoglobin 1-1 genotype is associated with poorer cognitive functioning in the elderly with type 2 diabetes.  Diabetes Care. 2013;36(10):3139-3145. doi:10.2337/dc12-2250PubMedGoogle ScholarCrossref
18.
Langlois  MR, Delanghe  JR.  Biological and clinical significance of haptoglobin polymorphism in humans.  Clin Chem. 1996;42(10):1589-1600.PubMedGoogle Scholar
19.
Goff  DC  Jr, Gerstein  HC, Ginsberg  HN,  et al; ACCORD Study Group.  Prevention of cardiovascular disease in persons with type 2 diabetes mellitus: current knowledge and rationale for the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial.  Am J Cardiol. 2007;99(12A):4i-20i. doi:10.1016/j.amjcard.2007.03.002PubMedGoogle ScholarCrossref
20.
Williamson  JD, Miller  ME, Bryan  RN,  et al; ACCORD Study Group.  The Action to Control Cardiovascular Risk in Diabetes–Memory in Diabetes study (ACCORD-MIND): rationale, design, and methods.  Am J Cardiol. 2007;99(12A):112i-122i. doi:10.1016/j.amjcard.2007.03.029PubMedGoogle ScholarCrossref
21.
National Heart, Lung, and Blood Institute. Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC). https://biolincc.nhlbi.nih.gov/home. Accessed January 27, 2017.
22.
Hochberg  I, Roguin  A, Nikolsky  E, Chanderashekhar  PV, Cohen  S, Levy  AP.  Haptoglobin phenotype and coronary artery collaterals in diabetic patients.  Atherosclerosis. 2002;161(2):441-446. doi:10.1016/S0021-9150(01)00657-8PubMedGoogle ScholarCrossref
23.
Levy  NS, Vardi  M, Blum  S,  et al.  An enzyme linked immunosorbent assay (ELISA) for the determination of the human haptoglobin phenotype.  Clin Chem Lab Med. 2013;51(8):1615-1622. doi:10.1515/cclm-2013-0018PubMedGoogle Scholar
24.
Koch  W, Latz  W, Eichinger  M,  et al.  Genotyping of the common haptoglobin Hp 1/2 polymorphism based on PCR.  Clin Chem. 2002;48(9):1377-1382.PubMedGoogle Scholar
25.
Beeri  MS, Ravona-Springer  R, Moshier  E,  et al.  The Israel Diabetes and Cognitive Decline (IDCD) study: design and baseline characteristics.  Alzheimers Dement. 2014;10(6):769-778. doi:10.1016/j.jalz.2014.06.002PubMedGoogle ScholarCrossref
26.
Moreno  PR, Purushothaman  KR, Purushothaman  M,  et al.  Haptoglobin genotype is a major determinant of the amount of iron in the human atherosclerotic plaque.  J Am Coll Cardiol. 2008;52(13):1049-1051. doi:10.1016/j.jacc.2008.06.029PubMedGoogle ScholarCrossref
27.
Zang  S, Chen  J, Song  Y,  et al; Chinese NAFLD Clinical Research Network (CNAFLD CRN).  Haptoglobin genotype and vitamin E versus placebo for the treatment of nondiabetic patients with nonalcoholic steatohepatitis in China: a multicenter, randomized, placebo-controlled trial design.  Adv Ther. 2018;35(2):218-231. doi:10.1007/s12325-018-0670-8PubMedGoogle ScholarCrossref
28.
Burghardt  KJ, Masri  DE, Dass  SE, Shikwana  SS, Jaber  LA.  Association between haptoglobin gene and insulin resistance in Arab-Americans.  Biomark Med. 2017;11(11):937-945. doi:10.2217/bmm-2017-0094PubMedGoogle ScholarCrossref
29.
Folstein  MF, Folstein  SE, McHugh  PR.  “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician.  J Psychiatr Res. 1975;12(3):189-198. doi:10.1016/0022-3956(75)90026-6PubMedGoogle ScholarCrossref
30.
Crum  RM, Anthony  JC, Bassett  SS, Folstein  MF.  Population-based norms for the Mini-Mental State Examination by age and educational level.  JAMA. 1993;269(18):2386-2391. doi:10.1001/jama.1993.03500180078038PubMedGoogle ScholarCrossref
31.
Staals  J, Bons  JA, van Oostenbrugge  RJ,  et al.  A SELDI-TOF-MS study in lacunar stroke with subsequent haptoglobin phenotyping.  Curr Neurovasc Res. 2008;5(2):93-98. doi:10.2174/156720208784310187PubMedGoogle ScholarCrossref
32.
Rouhl  RP, van Oostenbrugge  RJ, Damoiseaux  JG,  et al.  Haptoglobin phenotype may alter endothelial progenitor cell cluster formation in cerebral small vessel disease.  Curr Neurovasc Res. 2009;6(1):32-41. doi:10.2174/156720209787466082PubMedGoogle ScholarCrossref
33.
Li  W, Prakash  R, Kelly-Cobbs  AI,  et al.  Adaptive cerebral neovascularization in a model of type 2 diabetes: relevance to focal cerebral ischemia.  Diabetes. 2010;59(1):228-235. doi:10.2337/db09-0902PubMedGoogle ScholarCrossref
34.
Prakash  R, Somanath  PR, El-Remessy  AB,  et al.  Enhanced cerebral but not peripheral angiogenesis in the Goto-Kakizaki model of type 2 diabetes involves VEGF and peroxynitrite signaling.  Diabetes. 2012;61(6):1533-1542. doi:10.2337/db11-1528PubMedGoogle ScholarCrossref
35.
Livny  A, Ravona-Springer  R, Heymann  A,  et al.  Haptoglobin 1-1 genotype modulates the association of glycemic control with hippocampal volume in elderly individuals with type 2 diabetes.  Diabetes. 2017;66(11):2927-2932. doi:10.2337/db16-0987PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 632
    Original Investigation
    Neurology
    November 9, 2018

    Association of the Haptoglobin Gene Polymorphism With Cognitive Function and Decline in Elderly African American Adults With Type 2 DiabetesFindings From the Action to Control Cardiovascular Risk in Diabetes–Memory in Diabetes (ACCORD-MIND) Study

    Author Affiliations
    • 1Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York
    • 2The Joseph Sagol Neuroscience Center, Sheba Medical Center, Ramat Gan, Israel
    • 3Department of Environmental Medicine and Public Health, Icahn School of Medicine at Mount Sinai, New York, New York
    • 4The Memory and Psychogeriatric Clinic, Sheba Medical Center, Tel Hashomer, Israel
    • 5Division of Geriatrics and Gerontology, Department of Medicine, School of Medicine and Public Health, University of Wisconsin–Madison
    • 6Department of Psychology, Lehman College, The City University of New York, Bronx
    • 7Laboratory of Epidemiology and Population Science, National Institute on Aging, National Institutes of Health, Bethesda, Maryland
    • 8currently a student at Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
    • 9Department of Medicine, Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
    JAMA Netw Open. 2018;1(7):e184458. doi:10.1001/jamanetworkopen.2018.4458
    Key Points

    Question  Is the haptoglobin genotype associated with cognition in African American adults with type 2 diabetes?

    Findings  In this cohort study of 466 African American adults with type 2 diabetes, haptoglobin 1-1 carriers had the poorest cognitive function, while haptoglobin 2-1m carriers had the highest cognitive function. Older haptoglobin 1-1 carriers had greater cognitive decline over time compared with the other genotypes.

    Meaning  The haptoglobin gene polymorphism may contribute to cognitive impairment in African American adults who have type 2 diabetes.

    Abstract

    Importance  African American individuals have higher dementia risk than individuals of white race/ethnicity. They also have higher rates of type 2 diabetes, which may contribute to this elevated risk. This study examined the association of the following 2 classes of alleles at the haptoglobin (Hp) locus that are associated with poor cognition, cardiovascular disease, and mortality: Hp 1-1 (associated with poor cognition and cerebrovascular disease) and Hp 2-1 and Hp 2-2 (associated with greater risk of myocardial infarction and mortality). An additional polymorphism in the promoter region of the Hp 2 allele, restricted to individuals of African descent, yields a fourth genotype, Hp 2-1m. African American adults have a higher prevalence of Hp 1-1 (approximately 30%) compared with individuals of white race/ethnicity (approximately 14%), but the potential role of the Hp genotype in cognition among elderly African American individuals with type 2 diabetes is unknown.

    Objective  To assess the association of the Hp genotypes with cognitive function and decline in elderly African American adults with type 2 diabetes.

    Design, Setting, and Participants  This cohort study used publicly available data and specimens from the Action to Control Cardiovascular Risk in Diabetes–Memory in Diabetes (ACCORD-MIND) study to investigate the association of the Hp genotypes with cognitive function and decline in 466 elderly African American participants with type 2 diabetes. The hypothesis was that the Hp 1-1 genotype compared with the other genotypes would be associated with more cognitive impairment and faster cognitive decline in elderly African American adults with type 2 diabetes. The initial ACCORD trial was performed from October 28, 1999, to September 15, 2014. This was a multicenter clinical study performed in an academic setting.

    Exposures  The Hp genotypes were determined from serum samples by polyacrylamide gel electrophoresis and by enzyme-linked immunosorbent assay.

    Main Outcomes and Measures  The Mini-Mental State Examination (MMSE) was used to measure cognitive function and change after 40 months. The MMSE score ranges from 0 to 30 points; higher scores represent better cognition. Associations were examined with analysis of covariance and linear regression, adjusting for age, sex, education, baseline glycated hemoglobin level, systolic blood pressure, diastolic blood pressure, cholesterol level, creatinine level, and treatment arm (intensive vs standard). The cognitive change model adjusted also for the baseline MMSE score.

    Results  Among 466 African American study participants (mean [SD] age, 62.3 [5.7] years), 64.8% were women, and the genotype prevalences were 29.4% (n = 137) for Hp 1-1, 36.1% (n = 168) for Hp 2-1, 10.9% (n = 51) for Hp 2-1m, and 23.6% (n = 110) for Hp 2-2. The groups differed in their baseline MMSE scores (P = .006): Hp 1-1 had the lowest MMSE score (mean [SE], 25.68 [0.23]), and Hp 2-1m had the highest MMSE score (mean [SE], 27.15 [0.36]). Using the least squares method, the 40-month decline was significant for Hp 1-1 (mean [SE], −0.41 [0.19]; P = .04) and for Hp 2-2 (mean [SE], −0.68 [0.21]; P = .001). However, the overall comparison across the 4 groups did not reach statistical significance for the fully adjusted model. The interaction of age with the Hp 1-1 genotype on MMSE score decline estimate per year change was significant (mean [SE], −0.87 [0.37]; P = .005), whereas it was not significant for Hp 2-1 (mean [SE], 0.06 [0.37]; P = .85), Hp 2-1m (mean [SE], −0.06 [0.51]; P = .89), and Hp 2-2 (mean [SE], −0.44 [0.41]; P = .29), indicating that cognitive decline in Hp 1-1 carriers was accentuated in older ages, whereas it was not significant for the other Hp genotypes.

    Conclusions and Relevance  In this study, the Hp 1-1 genotype, which is 2-fold (approximately 30%) more prevalent among African American individuals than among individuals of white race/ethnicity, was associated with poorer cognitive function and greater cognitive decline than the other Hp genotypes. The Hp gene polymorphism may explain the elevated dementia risk in African American adults. The neuropathological substrates and mechanisms for these associations merit further investigation.

    Introduction

    African American adults have a significantly higher risk of developing dementia and Alzheimer disease than other racial/ethnic groups,1-3 even after adjusting for age, sex, and education.1 In addition to nongenetic factors, such as poor nutrition4 and obesity,5 higher rates of cerebrovascular disease and cardiovascular risk factors6 (especially type 2 diabetes7) in African American adults suggest that the genetics of vascular disease may underlie this increased risk of dementia.

    Haptoglobin (Hp) is a hemoglobin-binding protein synthesized in the liver and the brain. Two classes of alleles at the Hp locus yield 3 genotypes, Hp 1-1, Hp 2-1, and Hp 2-2. An additional polymorphism in the promoter region of the Hp 2 allele, restricted to individuals of African descent, yields a fourth genotype, Hp 2-1m.8,9 The biophysical properties of the Hp polymers formed in these 4 Hp genotypes are distinct, but little is known about the Hp 2-1m biology. The associations of Hp with health—studied almost solely in individuals of white race/ethnicity—are primarily in patients with type 2 diabetes.10 Hp 2-2 is associated with increased risk of myocardial infarction and mortality.11,12 Hp 1-1 has been linked to cerebral small vessel disease (eg, lacunar infarcts and white matter lesions).13-16 Consistent with the latter, it has been shown in a cohort of elderly white patients with type 2 diabetes that Hp 1-1 carriers have poorer cognitive function compared with Hp 2-1 and Hp 2-2 carriers.17 A study18 examining the prevalence of the Hp genotypes across different populations reported that African American adults have a higher prevalence of Hp 1-1 (approximately 30%) compared with individuals of white race/ethnicity (approximately 14%), but the potential role of the Hp genotype in poorer cognition and cognitive decline in elderly African American adults with type 2 diabetes has not been studied.

    We took advantage of publicly available data and specimens from the Action to Control Cardiovascular Risk in Diabetes–Memory in Diabetes (ACCORD-MIND) study to investigate the association of the Hp genotypes with cognitive function and decline in 466 elderly African American participants. We hypothesized that the Hp 1-1 genotype compared with the other genotypes would be associated with more cognitive impairment and faster cognitive decline in elderly African American adults with type 2 diabetes.

    Methods
    ACCORD Study Design

    The rationale, design, eligibility criteria, and glycemia intervention of the initial ACCORD trial have been reported previously.19 The ACCORD trial was performed from October 28, 1999, to September 15, 2014. Briefly, this was a multicenter clinical study in an academic setting that recruited volunteers with type 2 diabetes and a glycated hemoglobin level of 7.5% or more who were between the ages of 40 and 79 years and had cardiovascular disease. All 10 251 patients were randomly assigned to receive comprehensive intensive therapy targeting a glycated hemoglobin level of less than 6.0% or to receive standard therapy targeting a glycated hemoglobin level of 7.0% to 7.9%. The study protocol was approved by the institutional review boards of its 77 participating sites, as well as a review panel at the National Heart, Lung, and Blood Institute. All patients provided written informed consent. This study followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

    As a substudy, the ACCORD-MIND study20 aimed to examine the association of intensive vs standard glycemic control with cognitive function and decline. In total, 2977 ACCORD trial participants were included in this substudy, 466 of whom were African American individuals who had complete baseline information on the Mini-Mental State Examination (MMSE) score and relevant covariates. Cognition was measured at baseline and at the 20-month and 40-month visits.

    Hp Genotyping

    Through a request for data and serum samples to the Biologic Specimen and Data Repository Information Coordinating Center (BIOLINCC) system of the National Heart, Lung, and Blood Institute,21 specimens were sent to the Technion–Israel Institute of Technology (Haifa) and were genotyped for Hp by polyacrylamide gel electrophoresis (PAGE)22 and by enzyme-linked immunosorbent assay (ELISA),23 with no knowledge of any participant-related information. The ELISA assay provides a spectrophotometric readout that has been validated as having greater than 99% accuracy for Hp 1-1, Hp 2-1, and Hp 2-2.23,24 However, to our knowledge, the present study is first in using the ELISA assay for assessing Hp 2-1m, whose reading was (as expected) intermediate between Hp 1-1 and Hp 2-1. The PAGE method uses hemoglobin enrichment of serum22 and provides a signature banding pattern for Hp 1-1, Hp 2-1, Hp 2-1m, and Hp 2-2 (Figure 1). Hp 2-1m is found only in African American individuals and notably was not seen in any of the samples analyzed from individuals of white race/ethnicity (eg, the Israel Diabetes and Cognitive Decline study25), consistent with previous studies among white individuals26 and nonwhite individuals,27,28 and testifies to the accuracy of the Hp typing.23,24

    Hp Protein Quantitation

    To investigate the potential role of Hp protein levels on cognition and in the association of the Hp genotype with cognition, we quantified Hp protein levels. A 96-well microtiter dish (96-Well Microtiter Microplates; Nunc) was coated with 100 µL of rabbit anti-Hp in phosphate-buffered saline (PBS) (antibody titer 1:3000; Dako) and incubated overnight. The wells were rinsed with wash buffer (PBS with 0.05% TWEEN 20) and then incubated with blocking buffer (PBS with 5% bovine serum albumin and 0.05% TWEEN 20) for 2 hours at room temperature. The wells were then rinsed with wash buffer; serum samples diluted 1:100 000 or Hp standards (0.1-1.0 ng) in blocking buffer were added, and the plate was incubated overnight. The next day, the wells were washed, and mouse anti-Hp (1:10 000; Sigma) was added in blocking buffer for 1 hour. Thereafter, the wells were washed, and horseradish peroxidase anti-mouse antibody (1:20 000; Jackson ImmunoResearch Laboratories) in blocking buffer was added for 30 minutes. The wells were washed again, and substrate was added for 20 minutes, followed by stop solution (1N sulfuric acid). The plate was read at 450 nm in an ELISA reader.

    Hp Standards

    To evaluate the role of Hp protein levels in cognition and in the association of the Hp genotypes with cognition, Hp protein was obtained by immunoaffinity purification on an anti-Hp column. The Hp protein levels of the standards were determined using the following extinction coefficients (for solutions of 1 mg/mL), in ascending order: 1.18 for Hp 2-2, 1.21 for Hp 2-1, 1.23 for Hp 2-1m, and 1.24 for Hp 1-1.

    Mini-Mental State Examination

    The MMSE29 is a global cognitive screening instrument that assesses orientation, memory, attention/concentration, language, and visual construction. The MMSE score ranges from 0 to 30 points; higher scores represent better cognition.

    Statistical Analysis

    Descriptive data are reported as number (percentage) or mean (SD), as appropriate. Unadjusted bivariate analyses included 1-way analysis of variance or χ2 test, as appropriate, to compare Hp genotype differences relative to sociodemographic and clinical variables. The sociodemographic variables included baseline age, sex, and education (high school or less, some college or technical school, high school graduate or general equivalency diploma, and college graduate or more). The clinical variables included baseline glycated hemoglobin level, systolic blood pressure, diastolic blood pressure, cholesterol level, creatinine level, and cardiovascular disease (which includes stroke and myocardial infarction). An indicator variable for treatment arm (intensive vs standard) was also included.

    The primary analysis used the MMSE to assess the association of the Hp genotypes with MMSE score and measure cognitive function and change after 40 months. eTable 1 in the Supplement lists the distribution of the MMSE scores in the 4 genotypes. Two sets of linear regression models were fit. The primary model adjusted for baseline age, sex, and education. In a second model, we further adjusted for baseline clinical factors (glycated hemoglobin level, systolic blood pressure, diastolic blood pressure, cholesterol level, creatinine level, and cardiovascular disease) and the glycemic control study arm.

    The association between Hp genotypes and MMSE score was examined both cross-sectionally and longitudinally using the same primary and secondary sets of adjustment models as described in the previous paragraph. The cross-sectional analyses compared the MMSE score differences measured at baseline across the 4 Hp genotypes, while the longitudinal analysis using the mixed model with random participants focused on the change in MMSE score from baseline to the 20-month and 40-month follow-ups, with additional adjustment for the baseline MMSE score and use of blood pressure and oral glycemic control medications at the time of follow-up visits. For the change models, we assumed that missing data and dropouts were at random (ie, the unobserved data could be anticipated using the covariates considered in the model through the expectation-maximization algorithm).

    The ACCORD-MIND study participants were young (mean age, 62 years) in the context of a study on cognitive decline, while the MMSE shows greater variability in older individuals.30 Therefore, we performed secondary analyses to examine the interaction of age with Hp genotypes on the MMSE score change. We hypothesized that, at older ages, there would be greater cognitive decline in those with the Hp 1-1 genotype. To investigate this, we first plotted the MMSE score change against the baseline age by genotype using nonparametric local linear regression (locally weighted smoothing 80% smoothed with 95% CI). We then repeated the aforementioned longitudinal models (in the previous paragraph) but allowed each genotype to have its specific age association (ie, replacing the baseline age variable with the interaction term of baseline age with each individual genotype). Statistical analysis was performed using a software program (SAS, version 9.4; SAS Institute Inc). Two-sided P < .05 was defined as the significance level in all statistical tests.

    Results

    Table 1 lists the sociodemographic and clinical characteristics of the participants by Hp genotype. Among 466 African American study participants (mean [SD] age, 62.3 [5.7] years), 64.8% were women, and the genotype prevalences were 29.4% (n = 137) for Hp 1-1, 36.1% (n = 168) for Hp 2-1, 10.9% (n = 51) for Hp 2-1m, and 23.6% (n = 110) for Hp 2-2. The Hp groups differed in baseline glycated hemoglobin levels, with the highest level for the Hp 2-2 genotype (8.8%) and the lowest level for the Hp 2-1m genotype (8.2%). Otherwise, there were no significant differences among the 4 groups.

    Hp Genotype and Baseline MMSE Scores

    In the overall analysis of covariance controlling for the sociodemographic variables, the groups differed in their baseline MMSE scores (Table 2): Hp 1-1 had the lowest MMSE score (mean [SE], 25.68 [0.23]) and Hp 2-1m had the highest MMSE score (mean [SE], 27.15 [0.36]). Post hoc comparisons among pairs of the Hp genotypes (eTable 2 in the Supplement), adjusting for multiple comparisons, demonstrated that those with the Hp 2-1m genotype had better scores than those with the Hp 1-1 genotype (mean [SE] difference, 1.46 [0.30]); P = .001) and those with the Hp 2-2 genotype (mean [SE] difference, 1.18 [0.44]; P = .04). In model 2, the comparison of Hp 1-1 with Hp 2-1m withstood adjustment for multiple comparisons (mean [SE] difference, 1.39 [0.30]; P = .007). All other comparisons of pairs of genotypes showed no differences in the baseline MMSE scores.

    Hp Genotype and Longitudinal MMSE Score

    Table 3 summarizes model 1 and model 2 for longitudinal decline in MMSE score by Hp genotype. The MMSE score declined over time among those with the Hp 1-1, Hp 2-1, and Hp 2-2 genotypes; in the Hp 2-1m group, there was marginal improvement. Using the least squares method, the 40-month decline was significant for Hp 1-1 (mean [SE], −0.41 [0.19]; P = .04) and for Hp 2-2 (mean [SE], −0.68 [0.21]; P = .001). However, the overall comparison across the 4 groups did not reach statistical significance for model 1 or the fully adjusted model 2 (Table 3 and eFigure 1 in the Supplement).

    The Role of Age in the Association of Hp With Cognitive Decline

    As shown in Figure 2, the MMSE score change was subtle for all Hp genotypes in the younger ages (eFigure 2 in the Supplement shows the age distribution of the cohort). The line became significantly steeper for the Hp 1-1 genotype near age 70 years, indicating more decline in older Hp 1-1 carriers. However, in the older age strata, the lines for the other 3 genotypes remained flat. This observation was confirmed by the parametric regression models. The interaction of age with the Hp 1-1 genotype on MMSE score decline estimate per year change was significant (mean [SE], −0.87 [0.37]; P = .005), whereas it was not significant for Hp 2-1 (mean [SE], 0.06 [0.37]; P = .85), Hp 2-1m (mean [SE], −0.06 [0.51]; P = .89), and Hp 2-2 (mean [SE], −0.44 [0.41]; P = .29).

    Hp Protein Levels

    To explore whether the levels of the protein product of the Hp gene are associated with cognition and cognitive decline and whether they link the Hp genotypes to cognition, we repeated all analyses first using the serum level of Hp protein as the predictor and then including it as a covariate. The Hp protein level was not associated with cognition or cognitive decline, and its inclusion as a covariate in the Hp genotype and cognition models did not alter the results.

    Discussion

    In this cohort study of elderly African American adults having long-standing type 2 diabetes and participating in the ACCORD-MIND study, we found suggestive evidence that the Hp 1-1 genotype was associated with poorer global cognitive function and with significant cognitive decline at the 40-month follow-up, adjusting for age, sex, and education as well as when also adjusting for cardiovascular risk factors and the glycemic control study arm. In addition, among Hp 1-1 carriers, older baseline age may modulate the extent of the decline. The Hp protein level was not associated with cognition and did not alter the Hp genotype and cognition associations, suggesting that the structure and function of the Hp genotypes differentially influence cognition. Our study also confirms a 2-fold (approximately 30%) higher prevalence of the Hp 1-1 genotype among African American adults, as shown in another study,18 compared with individuals of white race/ethnicity (approximately 14%).25 The study by Langlois and Delanghe18 focused on African American adults from the Seattle, Washington, area, and the present study broadens to numerous regions in the United States.

    Although the data suggest that individuals with different Hp genotypes had some difference in the amount of cognitive decline, the main association of Hp genotype with cognitive decline was not statistically significant. However, there was a significant interaction between age and Hp genotype such that older Hp 1-1 carriers showed significantly greater MMSE score decline over time compared with the other genotypes. Poorer cognitive outcomes in Hp 1-1 genotype carriers may be caused by greater cerebrovascular disease. The Hp 1-1 genotype has been shown to correlate with stroke14 and with white matter hyperintensities13 in adults with type 1 diabetes. Similarly, in middle-aged patients with hypertension, the Hp 1-1 genotype correlates with the extent of deep white matter damage,15 and the frequency of the Hp 1 allele was significantly higher in patients with first symptomatic lacunar stroke due to small vessel disease.31 The cluster numbers of endothelial cells, which mitigate cerebral small vessel disease, are lower when Hp 1-1 protein is added to progenitor cell cultures32; cluster formation is poorer in patients with lacunar stroke with the Hp 1-1 phenotype.32 In contrast, the Hp 2 allele potentiates gene expression of proangiogenic factors in endothelial progenitor cells, potentially improving blood perfusion and recovery from ischemic injury.33,34 The association of the Hp 1-1 genotype with cerebrovascular disease and poor cognitive outcomes and the association of the Hp 2-2 genotype with peripheral vascular complications of type 2 diabetes suggest that Hp may serve different functions depending on the neuropathological processes in vascular diseases of different organs. This association merits further investigation because it may have significant clinical implications, suggesting caution against the universal application of therapies targeting vascular disease across all Hp genotypes.

    The Hp 2-1m genotype, pertaining solely to individuals of African descent and investigated herein for the first time (to our knowledge) in the context of cognitive aging, was associated with better cognitive function at baseline despite being the group with the oldest age and lowest education. Longitudinally, it was the only group with a suggestion of cognitive improvement over time. Because these results may suggest neuroprotection of this genotype, further investigation is warranted.

    Strengths and Limitations

    Strengths of the study include the large sample of African American adults with longitudinal cognitive data and a well-validated diagnosis of type 2 diabetes. The generalizability of the results may be limited because the ACCORD-MIND study sample is derived from a clinical trial and includes participants with severe type 2 diabetes, thus not representing the population of aging individuals with type 2 diabetes. Therefore, the study may have a survivor bias. For example, the Hp 2-2 genotype is associated with greater risk of myocardial infarction.11 Because there were no differences herein in the prevalence of myocardial infarction among different genotypes, it is possible that the Hp 2-2 individuals in the ACCORD-MIND study may be those who survived a myocardial infarction. The distribution of the Hp genotypes was similar in those with complete (n = 409) and incomplete data (n = 57) data, suggesting that, in this short-term study, censorship does not depend on Hp genotype. It has recently been shown in older persons of white race/ethnicity with type 2 diabetes that poor glycemic control was strongly associated with hippocampal volume among those with the Hp 1-1 genotype but not among other genotypes.35 This outcome suggests that individuals with the Hp 1-1 genotype may be more vulnerable to the deleterious influences of poor glycemic control.

    Conclusions

    To our knowledge, this is the first study to investigate the role of the Hp genotype in cognitive function and decline among elderly African American adults with type 2 diabetes, who are at particularly high risk of developing dementia1-3 and other type 2 diabetes complications.6,7 We show that the Hp 1-1 genotype, which is 2-fold (approximately 30%) more prevalent among African American adults than among individuals of white race/ethnicity, was associated with poorer cognitive function and greater cognitive decline than the other Hp genotypes. The Hp gene polymorphism may explain the elevated dementia risk in African American adults. Future research directions should address the neurobiological substrates and mechanisms underlying the associations of Hp and cognition in a more representative sample of older African American adults with type 2 diabetes and with a broader and more sensitive cognitive assessment.

    Back to top
    Article Information

    Accepted for Publication: September 14, 2018.

    Published: November 9, 2018. doi:10.1001/jamanetworkopen.2018.4458

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2018 Beeri MS et al. JAMA Network Open.

    Corresponding Author: Michal S. Beeri, PhD, Department of Psychiatry, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, PO Box 1230, New York, NY 10029 (michal.beeri@mssm.edu).

    Author Contributions: Dr Beeri had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Schnaider Beeri, Sano, Ravona-Springer, Gleason, Ehrenberg, Seligman.

    Acquisition, analysis, or interpretation of data: Schnaider Beeri, Lin, Liu, Bendlin, Guerrero-Berroa, Soleimani, Launer, Ehrenberg, Lache, Seligman, Levy.

    Drafting of the manuscript: Schnaider Beeri, Lin, Liu, Gleason, Ehrenberg, Seligman.

    Critical revision of the manuscript for important intellectual content: Schnaider Beeri, Lin, Sano, Ravona-Springer, Liu, Bendlin, Gleason, Guerrero-Berroa, Soleimani, Launer, Ehrenberg, Lache, Levy.

    Statistical analysis: Schnaider Beeri, Lin, Liu, Bendlin, Ehrenberg, Seligman.

    Obtained funding: Schnaider Beeri, Gleason, Guerrero-Berroa.

    Administrative, technical, or material support: Gleason, Soleimani, Ehrenberg, Lache, Seligman, Levy.

    Supervision: Sano, Ehrenberg.

    Conflict of Interest Disclosures: Dr Bendlin reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Guerrero-Berroa reported receiving grants from the Alzheimer’s Association during the conduct of the study. Dr Levy reported owning a patent claiming that the haptoglobin genotype can predict diabetic vascular complications. No other disclosures were reported.

    Funding/Support: This work was supported by grants R01 AG034087 (Dr Beeri) and P01 AG02219 and P50 AG05138 (Dr Sano) from the National Institute on Aging. This work was supported in part by the Intramural Research Program at the National Institute on Aging.

    Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    References
    1.
    Katz  MJ, Lipton  RB, Hall  CB,  et al.  Age-specific and sex-specific prevalence and incidence of mild cognitive impairment, dementia, and Alzheimer dementia in blacks and whites: a report from the Einstein Aging Study.  Alzheimer Dis Assoc Disord. 2012;26(4):335-343. doi:10.1097/WAD.0b013e31823dbcfcPubMedGoogle ScholarCrossref
    2.
    Demirovic  J, Prineas  R, Loewenstein  D,  et al.  Prevalence of dementia in three ethnic groups: the South Florida Program on Aging and Health.  Ann Epidemiol. 2003;13(6):472-478. doi:10.1016/S1047-2797(02)00437-4PubMedGoogle ScholarCrossref
    3.
    Krishnan  LL, Petersen  NJ, Snow  AL,  et al.  Prevalence of dementia among Veterans Affairs medical care system users.  Dement Geriatr Cogn Disord. 2005;20(4):245-253. doi:10.1159/000087345PubMedGoogle ScholarCrossref
    4.
    Carson  JA, Michalsky  L, Latson  B,  et al.  The cardiovascular health of urban African Americans: diet-related results from the Genes, Nutrition, Exercise, Wellness, and Spiritual Growth (GoodNEWS) trial.  J Acad Nutr Diet. 2012;112(11):1852-1858. doi:10.1016/j.jand.2012.06.357PubMedGoogle ScholarCrossref
    5.
    Lin  A, Lacy  ME, Eaton  C, Correa  A, Wu  WC.  Inflammatory obesity phenotypes, gender effects, and subclinical atherosclerosis in African Americans: the Jackson Heart Study.  Arterioscler Thromb Vasc Biol. 2016;36(12):2431-2438. doi:10.1161/ATVBAHA.116.307728PubMedGoogle ScholarCrossref
    6.
    Cushman  M, Cantrell  RA, McClure  LA,  et al.  Estimated 10-year stroke risk by region and race in the United States: geographic and racial differences in stroke risk.  Ann Neurol. 2008;64(5):507-513. doi:10.1002/ana.21493PubMedGoogle ScholarCrossref
    7.
    Ferdinand  KC, Nasser  SA.  Racial/ethnic disparities in prevalence and care of patients with type 2 diabetes mellitus.  Curr Med Res Opin. 2015;31(5):913-923. doi:10.1185/03007995.2015.1029894PubMedGoogle ScholarCrossref
    8.
    Giblett  ER.  Haptoglobin types in American Negroes.  Nature. 1959;183(4655):192-193. doi:10.1038/183192a0PubMedGoogle ScholarCrossref
    9.
    Maeda  N.  DNA polymorphisms in the controlling region of the human haptoglobin genes: a molecular explanation for the haptoglobin 2-1 modified phenotype.  Am J Hum Genet. 1991;49(1):158-166.PubMedGoogle Scholar
    10.
    Asleh  R, Briasoulis  A, Berinstein  EM,  et al.  Meta-analysis of the association of the haptoglobin genotype with cardiovascular outcomes and the pharmacogenomic interactions with vitamin E supplementation.  Pharmgenomics Pers Med. 2018;11:71-82. doi:10.2147/PGPM.S159454PubMedGoogle Scholar
    11.
    Costacou  T, Levy  AP.  Haptoglobin genotype and its role in diabetic cardiovascular disease.  J Cardiovasc Transl Res. 2012;5(4):423-435. doi:10.1007/s12265-012-9361-zPubMedGoogle ScholarCrossref
    12.
    Levy  AP, Asleh  R, Blum  S,  et al.  Haptoglobin: basic and clinical aspects.  Antioxid Redox Signal. 2010;12(2):293-304. doi:10.1089/ars.2009.2793PubMedGoogle ScholarCrossref
    13.
    Costacou  T, Rosano  C, Aizenstein  H,  et al.  The haptoglobin 1 allele correlates with white matter hyperintensities in middle-aged adults with type 1 diabetes.  Diabetes. 2015;64(2):654-659. doi:10.2337/db14-0723PubMedGoogle ScholarCrossref
    14.
    Costacou  T, Secrest  AM, Ferrell  RE, Orchard  TJ.  Haptoglobin genotype and cerebrovascular disease incidence in type 1 diabetes.  Diab Vasc Dis Res. 2014;11(5):335-342. doi:10.1177/1479164114539713PubMedGoogle ScholarCrossref
    15.
    Staals  J, Henskens  LH, Delanghe  JR,  et al.  Haptoglobin phenotype correlates with the extent of cerebral deep white matter lesions in hypertensive patients.  Curr Neurovasc Res. 2010;7(1):1-5. doi:10.2174/156720210790820163PubMedGoogle ScholarCrossref
    16.
    Staals  J, Pieters  BM, Knottnerus  IL,  et al.  Haptoglobin polymorphism and lacunar stroke.  Curr Neurovasc Res. 2008;5(3):153-158. doi:10.2174/156720208785425675PubMedGoogle ScholarCrossref
    17.
    Ravona-Springer  R, Heymann  A, Schmeidler  J,  et al.  Haptoglobin 1-1 genotype is associated with poorer cognitive functioning in the elderly with type 2 diabetes.  Diabetes Care. 2013;36(10):3139-3145. doi:10.2337/dc12-2250PubMedGoogle ScholarCrossref
    18.
    Langlois  MR, Delanghe  JR.  Biological and clinical significance of haptoglobin polymorphism in humans.  Clin Chem. 1996;42(10):1589-1600.PubMedGoogle Scholar
    19.
    Goff  DC  Jr, Gerstein  HC, Ginsberg  HN,  et al; ACCORD Study Group.  Prevention of cardiovascular disease in persons with type 2 diabetes mellitus: current knowledge and rationale for the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial.  Am J Cardiol. 2007;99(12A):4i-20i. doi:10.1016/j.amjcard.2007.03.002PubMedGoogle ScholarCrossref
    20.
    Williamson  JD, Miller  ME, Bryan  RN,  et al; ACCORD Study Group.  The Action to Control Cardiovascular Risk in Diabetes–Memory in Diabetes study (ACCORD-MIND): rationale, design, and methods.  Am J Cardiol. 2007;99(12A):112i-122i. doi:10.1016/j.amjcard.2007.03.029PubMedGoogle ScholarCrossref
    21.
    National Heart, Lung, and Blood Institute. Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC). https://biolincc.nhlbi.nih.gov/home. Accessed January 27, 2017.
    22.
    Hochberg  I, Roguin  A, Nikolsky  E, Chanderashekhar  PV, Cohen  S, Levy  AP.  Haptoglobin phenotype and coronary artery collaterals in diabetic patients.  Atherosclerosis. 2002;161(2):441-446. doi:10.1016/S0021-9150(01)00657-8PubMedGoogle ScholarCrossref
    23.
    Levy  NS, Vardi  M, Blum  S,  et al.  An enzyme linked immunosorbent assay (ELISA) for the determination of the human haptoglobin phenotype.  Clin Chem Lab Med. 2013;51(8):1615-1622. doi:10.1515/cclm-2013-0018PubMedGoogle Scholar
    24.
    Koch  W, Latz  W, Eichinger  M,  et al.  Genotyping of the common haptoglobin Hp 1/2 polymorphism based on PCR.  Clin Chem. 2002;48(9):1377-1382.PubMedGoogle Scholar
    25.
    Beeri  MS, Ravona-Springer  R, Moshier  E,  et al.  The Israel Diabetes and Cognitive Decline (IDCD) study: design and baseline characteristics.  Alzheimers Dement. 2014;10(6):769-778. doi:10.1016/j.jalz.2014.06.002PubMedGoogle ScholarCrossref
    26.
    Moreno  PR, Purushothaman  KR, Purushothaman  M,  et al.  Haptoglobin genotype is a major determinant of the amount of iron in the human atherosclerotic plaque.  J Am Coll Cardiol. 2008;52(13):1049-1051. doi:10.1016/j.jacc.2008.06.029PubMedGoogle ScholarCrossref
    27.
    Zang  S, Chen  J, Song  Y,  et al; Chinese NAFLD Clinical Research Network (CNAFLD CRN).  Haptoglobin genotype and vitamin E versus placebo for the treatment of nondiabetic patients with nonalcoholic steatohepatitis in China: a multicenter, randomized, placebo-controlled trial design.  Adv Ther. 2018;35(2):218-231. doi:10.1007/s12325-018-0670-8PubMedGoogle ScholarCrossref
    28.
    Burghardt  KJ, Masri  DE, Dass  SE, Shikwana  SS, Jaber  LA.  Association between haptoglobin gene and insulin resistance in Arab-Americans.  Biomark Med. 2017;11(11):937-945. doi:10.2217/bmm-2017-0094PubMedGoogle ScholarCrossref
    29.
    Folstein  MF, Folstein  SE, McHugh  PR.  “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician.  J Psychiatr Res. 1975;12(3):189-198. doi:10.1016/0022-3956(75)90026-6PubMedGoogle ScholarCrossref
    30.
    Crum  RM, Anthony  JC, Bassett  SS, Folstein  MF.  Population-based norms for the Mini-Mental State Examination by age and educational level.  JAMA. 1993;269(18):2386-2391. doi:10.1001/jama.1993.03500180078038PubMedGoogle ScholarCrossref
    31.
    Staals  J, Bons  JA, van Oostenbrugge  RJ,  et al.  A SELDI-TOF-MS study in lacunar stroke with subsequent haptoglobin phenotyping.  Curr Neurovasc Res. 2008;5(2):93-98. doi:10.2174/156720208784310187PubMedGoogle ScholarCrossref
    32.
    Rouhl  RP, van Oostenbrugge  RJ, Damoiseaux  JG,  et al.  Haptoglobin phenotype may alter endothelial progenitor cell cluster formation in cerebral small vessel disease.  Curr Neurovasc Res. 2009;6(1):32-41. doi:10.2174/156720209787466082PubMedGoogle ScholarCrossref
    33.
    Li  W, Prakash  R, Kelly-Cobbs  AI,  et al.  Adaptive cerebral neovascularization in a model of type 2 diabetes: relevance to focal cerebral ischemia.  Diabetes. 2010;59(1):228-235. doi:10.2337/db09-0902PubMedGoogle ScholarCrossref
    34.
    Prakash  R, Somanath  PR, El-Remessy  AB,  et al.  Enhanced cerebral but not peripheral angiogenesis in the Goto-Kakizaki model of type 2 diabetes involves VEGF and peroxynitrite signaling.  Diabetes. 2012;61(6):1533-1542. doi:10.2337/db11-1528PubMedGoogle ScholarCrossref
    35.
    Livny  A, Ravona-Springer  R, Heymann  A,  et al.  Haptoglobin 1-1 genotype modulates the association of glycemic control with hippocampal volume in elderly individuals with type 2 diabetes.  Diabetes. 2017;66(11):2927-2932. doi:10.2337/db16-0987PubMedGoogle ScholarCrossref
    ×