heme oxygenase-1 promoter (gt)n polymorphism associates

ARTICLE OPEN ACCESS

Heme oxygenase-1 promoter (GT)npolymorphism associates with HIVneurocognitive impairmentRolando Garza BS Alexander J Gill MD PhD Brandon L Bastien BA Yoelvis Garcia-Mesa PhD

Analise L Gruenewald BS Benjamin B Gelman MD PhD Billy Tsima MD MSCE Robert Gross MD MSCE

Scott L Letendre MD and Dennis L Kolson MD PhD

Neurol Neuroimmunol Neuroinflamm 20207e710 doi101212NXI0000000000000710

Correspondence

Dr Kolson

DennisKolsonuphsupennedu

AbstractObjectiveTo determine whether regulatory variations in the heme oxygenase-1 (HO-1) promoter (GT)ndinucleotide repeat length could identify unique population genetic risks for neurocognitiveimpairment (NCI) in persons living with HIV (PLWH) we genotyped 528 neurocognitivelyassessed PLWH of European American and African American descent and linked genotypes tocognitive status

MethodsIn this cross-sectional study of PLWH (the CNS HIV Antiretroviral Therapy Effect Researchcohort) we determined HO-1 (GT)n repeat lengths in 276 African Americans and 252 Eu-ropean Americans Using validated criteria for HIV-associated NCI (HIV NCI) we foundassociations between allele length genotypes and HIV NCI and between genotypes and plasmamarkers of monocyte activation and inflammation For comparison of HO-1 (GT)n allelefrequencies with another population of African ancestry we determined HO-1 (GT)n allelelengths in African PLWH from Botswana (n = 428)

ResultsPLWH with short HO-1 (GT)n alleles had a lower risk for HIV NCI (OR = 063 95 CI042ndash094) People of African ancestry had a lower prevalence of short alleles and higherprevalence of long alleles compared with European Americans and in subgroup analyses theprotective effect of the short allele was observed in African Americans and not in EuropeanAmericans

ConclusionsOur study identified the short HO-1 (GT)n allele as partially protective against developingHIVNCI It further suggests that this clinical protective effect is particularly relevant in personsof African ancestry where the lower prevalence of short HO-1 (GT)n alleles may limit in-duction of HO-1 expression in response to inflammation and oxidative stress Therapeuticstrategies that enhance HO-1 expressionmay decrease HIV-associated neuroinflammation andlimit HIV NCI

These authors contributed equally to the manuscript

From the Department of Neurology (R Garza AJG BLB YG-M ALG DLK) Perelman School of Medicine University of Pennsylvania Philadelphia Department of Pathology(BBG) University of Texas Medical Branch Galveston Department of Family Medicine amp Public Health (BT) University of Botswana Gaborone Departments of Medicine andBiostatistics (R Gross) Epidemiology and Informatics Perelman School of Medicine University of Pennsylvania Philadelphia and Department of Medicine (SLL) University ofCalifornia San Diego

Go to NeurologyorgNN for full disclosures Funding information is provided at the end of the article

The Article Processing Charge was funded by the authors

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

The syndrome of HIV-associated neurocognitive disorders(HAND) like many CNS disease states is linked to neuro-inflammation and oxidative stress Induction of the antioxi-dant isoenzyme heme oxygenase-1 (HO-1) can reduceoxidative stress and neuroinflammation within the CNS12

HO-1 expression is rapidly induced in response to injury andinduced HO-1 expression is neuroprotective in vitro and invivo in different injury models34

Transcriptional regulation of the HO-1 gene (HMOX1)depends partly on HO-1 promoter region genetic variationsincluding a (GT)n dinucleotide repeat and single nucleotidepolymorphisms (SNPs) Promoters with short HO-1 (GT)nrepeats express higher basal activity and inducibility5ndash8 andcells with these repeats are more resistant to oxidative injury6

An A(-413)T SNP can affect HO-1 promoter transcriptionalactivity and disease outcomes9 albeit less so than the HO-1(GT)n repeat length

10ndash12 Short HO-1 (GT)n repeat allelesassociate with better outcomes in inflammatory and oxidativestressndashassociated diseases7813ndash15 and the prevalence of theseshort alleles differs significantly among different populationsWe recently determined HO-1 (GT)n alleles in individuals inan HIV autopsy cohort (n = 554) and showed that thepresence of 1 or more short HO-1(GT)n alleles associatedwith a lower risk of HIV encephalitis and less brain in-flammation (type I interferon responses and T-cell activa-tion)16 No effect of the A(-413)T SNP was observed

We hypothesized that the HO-1 (GT)n allele genotype alsoassociates with risk for neurocognitive impairment (NCI) inpersons living with HIV (PLWH) and that genotype effectsvary among different populations We genotyped a large clin-ically characterized cohort of PLWH the CNS HIV Anti-retroviral Therapy Effect Research (CHARTER) cohort1718

and associated genotype with neurocognitive diagnosis Wereport significant associations indicating a protective effect ofshort HO-1(GT)n alleles against HIV-associated NCI (HIVNCI) particularly in African Americans

MethodsCohortsFor analysis of HO-1 (GT)n allele genotypes neurocognitivediagnosis and plasma biomarkers 606 PLWH were selectedfrom the CHARTER Genetics Cohort19 based on reportedavailability of neurocognitive diagnosis data and DNA sam-ples and without other inclusionary or exclusionary criteria

to avoid potential bias Of the 606 DNA samples 3 were ofinsufficient quality to obtain reliable (GT)n sequence datatherefore HO-1 (GT)n allele data from 603 individuals werepresented (table 1 and figure 1A) CHARTER is an ongoingobservational study of PLWH enrolled between 2003 and2007 (1561 individuals) from 6HIV treatment centers JohnsHopkins University (Baltimore MD) Mt Sinai School ofMedicine (New York NY) University of California at SanDiego (San Diego CA) University of Texas Medical Branch(Galveston TX) University of Washington (Seattle WA)and Washington University (St Louis MO) Our requiredsample size was estimated at 400 based on our analysis ofthe National NeuroAIDS Tissue Consortium (NNTC) au-topsy cohort which revealed the presence of short HO-1(GT)n allele associated with an increased risk of HIVencephalitis16

Among these 603 individuals 595 had sufficient neuro-psychological (NP) test data for diagnostic HIV NCI assign-ment Therefore NCI data from those 595 individuals wereanalyzed forHIVNCI associations with genotype and subgroupanalyses Among these 595 individuals data from 528 total self-identified African Americans and European Americans wereanalyzed to distinguish differences between these groups

HO-1 (GT)n allele genotyping was also performed on wholebloodndashderived DNA from a Botswana cohort of PLWH onantiretroviral therapy (ART) randomly selected irrespective ofneurocognitive function2021 and under separate IRB approval

Data availabilityAnonymized data are available from the corresponding author(DLK) on reasonable request

Neurocognitive and functional assessmentsIndividuals from CHARTER completed baseline assessmentsincluding NP testing (7 cognitive domains corrected for ageeducation sex and ethnicity) substance abuse history psy-chiatric diagnosis self-reported cognition assessment voca-tional function assessment and assessment of independencewith instrumental activities of daily living (ADLs)17 Specimensand data were obtained with permission of the CHARTERsteering committee Neurocognitive diagnosis was first de-termined by the Frascati criteria for HIV-associated neuro-cognitive disorders22 Under these criteria the diagnoses ofasymptomatic neurocognitive impairment (ANI) and mildneurocognitive disorder (MND) require the presence of mildimpairment in ge2 NP domains that is not attributed to

GlossaryADL = activities of daily livingANI = asymptomatic neurocognitive impairmentART = antiretroviral therapyGT = promoterHAD = HIV-associated dementia HAND = HIV-associated neurocognitive disorders HIV NCI = HIV-associated NCI HO-1 = heme oxygenase-1 MND = mild neurocognitive disorder NCI = neurocognitive impairment NCN = neurocognitivenormal NNTC = National NeuroAIDS Tissue Consortium NP = neuropsychological PLWH = persons living with HIVsHO-1 = soluble HO-1 SNP = single nucleotide polymorphism

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 NeurologyorgNN

comorbid conditions In addition MND requires functionaldecline by 2 measures of functioning in ADLs HIV-associateddementia (HAD) requires functional decline with greater se-verity than that ofMND Persons having no or only 1 abnormalNP domain test performance are classified as neurocognitivenormal (NCN) We grouped MND and HAD individuals asfunctional HIV NCI individuals whereby functional impair-ment of ADL was required for this grouping

We further classified HIV NCI by distinguishing the presenceor absence of comorbidity factors that were deemed severeenough to contribute to NCI in conjunction with HIV in-fection according to CHARTER guidelines1722 AccordinglyPLWH were stratified by the severity by which these factors(eg depression traumatic brain injury developmental dis-ability substance abuse opportunistic CNS infections nonndashHIV-related neurologic conditions systemic disease and hep-atitis C virus coinfection) are thought to affect NCI as adju-dicated by an experienced neuropsychologist NCI without

such contributing factors is considered incidental HIV NCI(ie incidental to HIV infection alone) HIV NCI withcomorbidities contributing a minor level of impairment is de-fined as HIV NCI with contributing factors172223 Individualswith comorbidities contributing a major level of impairment(confounding factors) were excluded Thus HIV NCI withinthe incidental group is considered to be a consequence of HIVinfection itself whereas HIV NCI within the contributinggroup is considered to be a consequence of HIV infection witha minor additional component due to contributing factors23

To include an appropriate subgroup analysis of HIV NCI vsno NCI we grouped individuals without NCI in each of 2ways (1) NCN and (2) no functional NCI (NCN and ANI)because impairment of ADL is not a criterion for the diagnosisof ANI All procedures were approved by the Human SubjectsProtection Committees of the respective academic centerswhere patients were enrolled Informed consent was obtainedfrom all individuals

Table 1 Cohort demographics and HIV disease parameters

Characteristic

Cohort group HO-1 (GT)n allele genotype

p ValueAll Presence of Short Allele (SS SM SL) Absence of Short Allele (MM ML LL)

No of individuals 606 294 309 mdash

Age (y mean plusmn SD) 438 plusmn 83 434 plusmn 84 442 plusmn 81 0227a

Duration of infection 92 plusmn 64 90 plusmn 66 94 plusmn 63 0493a

Sex ()

Male 800 806 793 0685b

Race ()

Non-Hispanic European American 416 466 366 0013b

African American 455 388 521 0001b

Otherunknown 129 146 113 0228b

Education (y) 127 plusmn 25 128 plusmn 24 127 plusmn 25 0401a

Antiretroviral treated () 726 747 702 0215b

HIV RNA lt50 copiesmL 417 405 415 0798b

Disease parameters (mean plusmn SD)

Log plasma HIV copiesmL 28 plusmn 13 28 plusmn 13 28 plusmn 13 0833a

Log CSF HIV copiesmL 21 plusmn 08 21 plusmn 08 22 plusmn 08 0474a

CD4+ T lymphocytes (cellsμL) 458 plusmn 267 464 plusmn 245 455 plusmn 288 0694a



CD4CD8 T lymphocyte ratio 06 plusmn 08 06 plusmn 12 06 plusmn 04 0447a

CD4+ T lymphocyte nadir (cellsμL) 207 plusmn 190 212 plusmn 181 204 plusmn 198 0578a

Abbreviations L = long allele M = medium allele S = short alleleSample size within demographic and disease parameters may be less than the overall sample size due tomissing data () in categories (duration HIV 265antiretroviral treated 02 plasma HIV 06 CSF HIV 58 CD4+ 06 CD8+ 27 CD3+ 86 CD4CD8 25)a p Values represent comparisons between (SS SM SL) and (MM ML LL) HO-1 (GT)n genotype groups by the Student t testb p Values represent comparisons between (SS SM SL) and (MM ML LL) HO-1 (GT)n genotype groups by the χ2 test

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 7 Number 3 | May 2020 3

Quantification of viral loads andlymphocyte countsHIV infection determined by ELISA was confirmed byWestern blot Clinical chemistry panels complete bloodcounts and CD4+ T cells (flow cytometry) were performedat the respective enrollment sitersquos clinical or equivalentmedical center laboratory Plasma and CSF HIV RNA wasquantified by reverse transcriptase PCR (Roche Amplicor

v 15 lower limit of quantitation 50 copiesmL) in a centrallaboratory17

HO-1 (GT)n repeat and HO-1 A(-413)TSNP sequencingFor the CHARTER and Botswana cohorts genomic DNAisolated from whole blood (DNA Extraction Kit AgilentTechnologies) Detailed protocols used for amplification and

Figure 1 HO-1 (GT)n promoter repeat allele length and A(-413)T SNP allele frequencies differ significantly between AfricanAmericans and European Americans

HO-1 (GT)npromoter repeats andA(-413)T SNPswereassessed in603 and 601PLWH respectively in theCHARTER cohort (A) DistributionofHO-1 (GT)npromoterallele lengths among all cohort individuals alleles were defined as short ldquoSrdquo (lt27 (GT)n repeats) medium ldquoMrdquo (27ndash34 repeats) and long ldquoLrdquo (gt34 repeats) (B) HO-1(GT)n allele length frequencies in European American and African American subgroups compared by the Kolmogorov-Smirnov test (C) Allele frequencies and (D)genotype frequencies of the HO-1 promoter (GT)n alleles and (E) allele frequencies and (F) genotype frequencies of the HO-1 promoter A(-413)T SNP in the totalcohort and in African American and European American subpopulations Allelic and genotypic frequency distributionswere compared by the χ2 test CHARTER =CNS HIV Antiretroviral Therapy Effect Research HO-1 = heme oxygenase-1 PLWH = persons living with HIV SNP = single nucleotide polymorphism


determination of genotype were previously described by ourgroup16 Briefly HO-1 (GT)n repeat lengths were determinedby PCR of the (GT)n repeat regionwith a 6-fluorescein amiditeslabeled forward primer followed by fragment size determinationon a capillary sequencer In a subset of samples PCR productswere run on a 2 agarose gel to assess amplification of thetarget sequence predicted sizes (homozygotes and hetero-zygotes) All samples were run at least twice from independentPCR reactions to ensure accurate and reproducible sizing andwere determined to be accurate within plusmn 1 GT repeats In rare(lt4) cases in which identical GT repeat lengths were notobtained in duplicate samples were run additional times toconfirm accurate repeat lengths A(-413)T SNP (rs2071746)genotyping was performed using TaqMan SNP GenotypingAssay (ThermoFisher Assay ID C__15869717_10) per themanufacturerrsquos instructions

Determination of plasma sCD163 and sHO-1We selected sCD163 as a macrophage-specific associativebiomarker for HIV NCIHAND groups inclusive of MNDand HAD2425 Also CD163 receptor engagement (scaveng-ing hemoglobinhaptoglobin complexes) is a stimulus forproduction of HO-126 Plasma soluble HO-1 (sHO-1) servesas a marker of inflammation and oxidative stress2728 Each wasquantified in duplicate by commercial ELISA (1) sCD163 (Ramp D Systems DY1607 LOD 156 pgmL) and (2) sHO-1(ENZO Life Sciences ADI-960-800 LOD 49 pgmL) instored frozen (minus80degC) plasma samples

Statistical analysesFor comparisons between groups differences are expressed asmedian plusmn 95 CI Comparisons of distributions of categoricalvariables were determined by χ2 and method for OR analysisComparisons of continuous variables between groups wereperformed using the Student t test or 2-way analysis of vari-ance with the multiple comparisons test Confounding andeffect modification was assessed by including individual vari-ables associated with the outcome with a p value lt 010 usingthe Mantel-Haenszel procedure Analyses of linear trendswere performed by Spearman rank correlation Comparisonsof allele distributions were analyzed by the Kolmogorov-Smirnov test

ResultsHO-1 promoter (GT)n allele repeat lengths andthe A(-413)T SNP genotype have modaldistributions that suggest higher-riskgenotypes in African Americans comparedwith European AmericansTo define relationships between HIV NCI immune activa-tion and HO-1 (GT)n allele length in PLWH we studiedindividuals (n = 606) in the CHARTER Genetics Cohort(table 1)1929 Of the 606 DNA samples 3 were of insufficientquality to obtain reliable sequence data therefore HO-1(GT)n allele data from 603 individuals were presented (table

1 and figure 1A) The distribution of HO-1 (GT)n allelelengths was trimodal (peaks at 23 30 and 39) with a range of13ndash44 and we assigned HO-1 (GT)n alleles as short ldquoSrdquo(lt27) medium ldquoMrdquo (27ndash34) or long ldquoLrdquo (gt34) (GT)nrepeats (figure 1 A and B)816 HO-1 (GT)n allele and ge-notype distributions (figure 1 C and D) differed significantlybetween self-identified African Americans and EuropeanAmericans African Americans expressed more HO-1 (GT)nldquoLrdquo alleles (333 vs 36 p lt 0001) fewer HO-1 (GT)n ldquoMrdquoalleles (425 vs 618 p lt 0001) and fewer ldquoSrdquo alleles(242 vs 346 p lt 0001) than European Americans

We also genotyped the HO-1 promoter region A(-413)TSNP (rs2071746) in 601 (992) individuals The A(-413)TSNP can affect HO-1 promoter transcriptional activity anddisease outcomes9 albeit less so than the HO-1 (GT)n repeatlength10ndash12 The frequency of the ldquoTrdquo A(-413)T SNP wassignificantly higher in African Americans than in EuropeanAmericans (660 vs 418) whereas the frequency of theldquoArdquo SNP was lower (34 vs 582) (figure 1E) and asexpected for Hardy-Weinberg equilibrium the TT and ATgenotypes were more prevalent in African Americans (figure1F) Also the A(-413)T SNP and the HO-1 (GT)n allelegenotypes were not independently expressed the ldquoArdquo SNPassociated with medium ldquoMrdquoHO-1 (GT)n alleles whereas theldquoTrdquo SNP associated with both short ldquoSrdquo and long ldquoLrdquo HO-1(GT)n alleles (not shown) These findings are consistent withprevious studies of HO-1 (GT)n repeats and the A(-413)TSNP816

Neither the HO-1 (GT)n allele genotype nor theA(-413)T SNP genotype correlates with plasmaor CSF viral load T-cell (CD4+ and CD8+) countsor historical CD4+ nadir in PLWHWe examined associations between HO-1 (GT)n allele ge-notype A(-413)T SNP genotype plasma and CSF viral loadT-cell (CD4+ CD8+) counts and CD4+ nadir and found nosignificant correlations (table 1) We conclude that neitherthe HO-1 (GT)n allele genotype nor the A(-413)T SNP ge-notype affects HIV replication or severity of HIV-driven im-mune suppression

PLWH who have short HO-1 (GT)n alleles havea lower prevalence of HIV NCIWe determined associations between the HO-1 (GT)n allelegenotype and functional HIV NCI (grouped as total MNDand HAD individuals) in comparison to those without func-tional HIV NCI (grouped either as total ANI and NCNindividuals or only those categorized as NCN figure 2)Comparing the no functional HIV NCI groups with thefunctional HIV NCI groups the presence of at least 1 shortHO-1 (GT)n allele associated with a lower functional HIVNCI prevalence (OR = 063 95 CI 042ndash094 figure 2A)The protective effect of the short allele was greater in indi-viduals with functional HIV NCI without other contributingfactors (OR = 048 95 CI 028ndash084 figure 2B) Further-more the group with 2 short alleles ldquoSSrdquo had the lowest


functional HIV NCI prevalence (150 and 102 re-spectively figure 2 C and D) among all HO-1 (GT)n allelegenotypes whereas those with 2 long alleles ldquoLLrdquo had the

highest functional HIV NCI prevalence (313 for both) Noeffect of the A(-413)T SNP genotype was observed in anygroup analysis (not shown)

Figure 2 PLWH with short HO-1 (GT)n alleles have significantly decreased risk of functional HIV NCI

Individuals underwent a comprehensive neuro-cognitive test battery (7 domains) and were assignedfunctional neurocognitive diagnoses by Frascati cri-teria for HIV-associated neurocognitive disorders (A)Within the total cohort (contributing and incidentalrisk(s)) the presence of at least 1 short ldquoSrdquo HO-1 (GT)nallele associated with lower functional HIV NCI(groups MND and HAD) vs no functional HIV NCI (NCNand ANI groups) by an OR of 063 (95 CI 042ndash094)(χ2 test p = 0020) (B) Within the HIV NCI incidentalgroup (without other contributing comorbidities) thepresence of at least 1 short ldquoSrdquo HO-1 (GT)n allele as-sociated with lower functional HIV NCI by an OR of048 (95 CI 028ndash084 χ2 test p = 0007) The prev-alence of functional HIV NCI increased linearly withstratification of genotypes from homozygous shortldquoSSrdquo to homozygous long ldquoLLrdquo within (C) the total co-hort and within (D) the functional HIV NCI groupwithout contributing comorbidities (χ2 tests for lineartrend p = 0020 p = 0002 respectively) Subgroupanalyses were similarly performed by comparingfunction HIV NCI groups MND and HAD with NCNalone (EndashH) with similar results ANI = asymptomaticneurocognitive impairment HAD = HIV-associateddementia HO-1 = heme oxygenase-1 MND = mildneurocognitive disorder NCI = neurocognitive im-pairment NCN = neurocognitive normal PLWH =persons living with HIV


We performed a subgroup analysis by comparing the NCNgroup alone (removing the ANI subgroup) with the MND andHAD group Again the presence of at least 1 short allele as-sociated significantly with a lower prevalence of functional HIVNCI (OR = 059 95 CI 038ndash089 figure 2E) with a lowerOR for functional HIV NCI without contributing factors (OR= 043 95 CI 024ndash075 figure 2F) Similarly individualswith 2 short alleles ldquoSSrdquo had the lowest functional HIV NCIprevalence (239 147 respectively) whereas those with 2long alleles ldquoLLrdquo had the highest prevalence of functional NCI(454 385 respectively) (figure 2 G andH) Again no effectof the A(-413)T SNP genotype was observed (not shown)

The significant risk reduction for HIV NCI byshort HO-1 (GT)n alleles is observed in AfricanAmericans and not in European AmericansWithin the CHARTER cohort non-Hispanic self-identifiedEuropean Americans represented 416 and self-identifiedAfrican Americans represented 455 the remaining 129represented otherunknown racial groups (table 1) AfricanAmericans had lower mean CD4+ T-cell counts CD4+CD8+

T-cell ratio and CD4+ T-cell nadirs compared with European

Americans with no difference in the absolute mean CD8+

T-cell count (table 2)

Because African Americans have a different HO-1 (GT)n alleledistribution with fewer short and more long alleles than Euro-pean Americans we hypothesized that the protective effect ofa short HO-1 (GT)n allele may be more easily detected inAfrican Americans Comparing the no functional HIV NCIgroup with the functional HIV NCI group (figure 3) and fur-ther analyzing those with HIV NCI without other contributingfactors we observed a different effect between African Ameri-cans and European Americans The presence of at least 1 shortallele associated with a strong protective effect in AfricanAmericans (OR = 038 95 CI 016ndash084 figure 3A) but notin European Americans (figure 3B) However the effect dif-ference between the races was not significant (Mantel-Haenszeltest p = 031) Also although the prevalence of functional HIVNCI was similar in both racial subgroups with at least 1 shortallele (117 and 122 respectively) there was no significantdifference in functional HIV NCI prevalence in African Amer-icans without a short allele (260) compared with EuropeanAmericans (165) (p = 011) Within the African American

Table 2 Demographics and parameters

Parameter

Racial group

p ValueEuropean American African American

No of individuals 252 276 mdash

Age (y mean plusmn SD) 438 plusmn 91 443 plusmn 74 0496a

Duration of infection 88 plusmn 67 98 plusmn 61 0120a

Sex ()

Male 714 889 lt00001b

Education (y) 134 plusmn 25 121 plusmn 22 lt00001a

Antiretroviral treated ()

HIV RNA lt50 copiesmL 743 706 0380b

HAND () 392 431 0376b

MND and HAD 221 205 0670b


Log plasma HIV copiesmL 28 plusmn 13 28 plusmn 13 0690a

Log CSF HIV copiesmL 21 plusmn 08 21 plusmn 08 0881a

CD4+ T lymphocytes (cellsμL) 504 plusmn 272 419 plusmn 261 0000a



CD4CD8 T lymphocyte ratio 07 plusmn 12 05 plusmn 04 0025a

CD4+ T lymphocyte nadir (cellsμL) 237 plusmn 199 183 plusmn 172 0001a

Abbreviations HAD = HIV-associated dementia HAND = HIV-associated neurocognitive disorders MND = mild neurocognitive disorderSample size within demographic and disease parameters may be less than the overall sample size due tomissing data () in categories (duration HIV 286HAND 19 plasma HIV 06 CSF HIV 61 CD4+ 06 CD8+ 28 CD3+ 89 CD4CD8 27)a p Values represent comparisons between the Caucasian and African American groups by the Student t testb p Values represent comparisons between the Caucasian and African American groups by the χ2 test


Figure 3 African American PLWHwith 1 ormore short HO-1 (GT)n alleles have significantly decreased risk of functional HIVNCI in the absence of contributing comorbidity factors

Within the group of African Americans (A) but notthe European American group (B) with functional HIVNCI without contributing comorbidities the pres-ence of at least 1 short ldquoSrdquo HO-1 (GT)n allele associ-atedwith lower functional HIVNCI (MNDandHAD) vsno functional HIV NCI (NCN and ANI) by an OR of038 (95 CI 016ndash084) (χ2 test p = 0018) (C)Functional NCI frequency increased linearly withstratification of genotypes from homozygous shortldquoSSrdquo to homozygous long ldquoLLrdquo within the AfricanAmerican group (χ2 test for linear trend p = 0020)but not in the European American group (notshown) (D) African Americans have a lower preva-lence of HO-1 (GT)n genotypes containing a short ldquoSrdquoallele and a higher prevalence of genotypes con-taining a long ldquoLrdquo allele than European Americans (χ2

test p lt 00001) Subgroup analyses were similarlyperformed by comparing function HIV NCI groupsMND and HAD with NCN alone (EndashH) with similarresults ANI = asymptomatic neurocognitive impair-ment HAD = HIV-associated dementia HO-1 = hemeoxygenase-1 MND = mild neurocognitive disorderNCI = neurocognitive impairment NCN = neuro-cognitive normal PLWH = persons living with HIV


group there was a significant linear trend of increased preva-lence of functional HIV NCI with genotypes expressing allelesof increasingHO-1 (GT)n repeats from ldquoSSrdquo to ldquoLLrdquo (p= 0020figure 3C) which was not observed in the European Americancohort (not shown) We note that no European Americans inour cohort had the ldquoLLrdquo genotype (figure 3D)

In a subgroup analysis of functional HIV NCI without othercontributing factors we compared the NCN group with thefunctional NCI (MND and HAD) group and again observedsimilar effects The presence of at least 1 short allele associatedwith a lower prevalence of NCI (OR = 034 95 CI014ndash081) in African Americans but not in EuropeanAmericans (figure 3 E and F) Again within the AfricanAmerican group there was a significant linear trend of in-creased prevalence of functional HIV NCI with genotypesexpressing alleles of increasing HO-1 (GT)n repeats fromldquoSSrdquo to ldquoLLrdquo (p = 0011 figure 3G) No significant trend wasobserved in European Americans (figure 3H) No effect of theA(-413)T SNP genotype was observed in comparisonsamong any groups (not shown)

Genetic ancestry analysis confirms self-identified racial assignments and ancestry-specific associations between the HO-1 (GT)ngenotype and functional HIV NCIPreviously ancestry-informative nuclear DNA SNP variantanalysis of the CHARTER PLWH cohort defined categoriesof ancestry assignment by genetic ancestry clusters usingprincipal component ancestry-based stratifications (Euro-pean African or admixed Hispanic)30 In the CHARTERcohort 981 of self-identified African Americans and Euro-pean Americans were assigned to a genetic ancestry clusterresulting in 977 concordance with African and Europeanancestry respectively No self-identifying African Americanswere classified as European ancestry and no self-identifyingEuropean Americans were classified as African ancestry The10 discordant self-identified European Americans and 5 dis-cordant self-identified African Americans were classified asadmixed Hispanic ancestry In addition 7 PLWHwho did notself-identify as European American or African Americanwere classified as European or African ancestry (4 and 3respectively) When the racial subgroup associations betweenHO-1 (GT)n allele genotype and functional HIV NCI wererepeated using genetically assigned ancestry clusters theresults and statistical significance were nearly identical to theanalyses using self-identification (figure e-1 linkslwwcomNXIA231) Specifically in African Americans the presenceof at least 1 short allele associated with a lower prevalence offunctional HIV NCI in both NCN and ANI (OR = 031 95CI 012ndash078) and NCN-only subgroup analyses (OR = 02895 CI 011ndash069) whereas no statistical significance wasobserved in these associations in the European ancestry group(both p gt 05) Although the ORs were lower in the geneti-cally assigned African ancestry group compared with the self-identified African American group no significant change inORs was observed (p gt 05 Mantel-Haenszel test)

Short HO-1 (GT)n alleles associate with lowerplasma sHO-1 in virally suppressed AfricanAmerican PLWHA previous report associated short HO-1 (GT)n repeat lengthswith lower plasma soluble CD14 (sCD14) levels in ART-treated African American PLWH but not in European Amer-ican PLWH which suggests that short HO-1 (GT)n alleleshave a downmodulating effect against HIV-associated in-flammation in African Americans8 We determined relation-ships between HO-1 (GT)n genotype plasma sCD163(monocyte activation)26 and plasma sHO-1 (inflammationand oxidative stress)2728 in ART-treated virally suppressedindividuals within the cohort We did not observe significantdifferences in the total cohort (not shown) However when westratified groups according to current CD4+ T-cell count(above and below 350 cellsμL) as an indicator of relative levelof immunosuppression3132 we observed significant differ-ences In African Americans (figure 4A) but not in EuropeanAmericans (figure 4B) with current CD4 T-cell counts ge350cellsμL the presence of 1 or more short alleles associated withlower plasma sHO-1 This effect was independent of HIV NCIdiagnosis An observed difference in plasma sCD163 in AfricanAmericans (figure 4C) was not statistically significant (p =0076) and no sCD163 difference was observed in EuropeanAmericans (not shown) Furthermore sCD163 associatedpositively with sHO-1 in African Americans (figure 4D) butnot in European Americans (not shown)

Africans from Botswana have a higherprevalence of long HO-1 (GT)n alleles thanAfrican Americans and European AmericansTo confirm the apparent association between the trimodalHO-1 (GT)n allele length distribution genotype prevalencedistribution and African ancestry we genotyped a cohort(n = 428) of PLWH from previous observational studies ofAfricans in Botswana2021 The allele distributions differedsignificantly from that observed in African Americans andEuropean Americans genotyped in our CHARTER andNNTC cohorts (figure 5A)16 Individuals in the Botswanacohort had the highest prevalence of long ldquoLrdquo alleles (384)compared with African Americans (p lt 001) and EuropeanAmericans (324 47 p lt 001 respectively figure 5 B andC) Significantly different HO-1 (GT)n allele genotype fre-quencies were also observed with a higher prevalence of ldquoLLrdquogenotypes in the Botswana cohort (168 107 and 07respectively) confirming a higher prevalence of long HO-1(GT)n alleles in individuals of African ancestry (comparedwith European Americans)

DiscussionThe global burden of NCI from HIV infection remains highdespite the effectiveness of ART in suppressing HIV replica-tion increasing life expectancy and reducing the severity ofHIV NCI in PLWH171833 The spectrum of HAND hasevolved with the use of ART Despite the decreased prevalence


of its most severe form HAD from 20 to 2 its lesssevere but nonetheless disabling form MND affects15 ofART-suppressed PLWH Thus ART provides incompleteprotection from developing functionally disabling HIV NCIand this challenging therapeutic gap may widen as the pop-ulation of PLWH ages34ndash37 Closing this gap depends onidentifying and managing those pathologic processes that driveHIV NCI in ART-suppressed PLWH and identifying individ-uals who are highly vulnerable to those processes

Persistent oxidative stress and inflammation are risk factorsfor the development of HIV NCI in ART-suppressedPLWH38ndash41 and we have identified the cytoprotective en-zymeHO-1 as a potential therapeutic target1642ndash44 Our studyidentifies the HO-1 promoter (GT)n repeat as a potential riskmodulator for the development of HIV NCI in such indi-viduals We emphasize this effect in individuals with func-tional HIV NCI (ie impaired ADLs subgroups MND andHAD) in comparison to PLWHwithout functional HIV NCIour conclusions were confirmed by comparisons with eitherNCN or combined NCN and ANI group subgroup analysesWe specifically demonstrate that the presence of short HO-1

(GT)n alleles which are known to have higher HO-1 tran-scriptional activity associates with a decreased risk for HIVNCI Moreover our subgroup analyses demonstrate that thisHO-1 (GT)n allele risk reduction is relatively specific for NCIdue to the presence of HIV infection alone and not commoncomorbid cognitive risk factors These findings suggesta therapeutic opportunity for further HIV NCI risk reductionthrough targeting HO-1 expression function and associatedfactors in PLWH who are receiving ART

Although we observed that a protective effect of the shortallele associated with NCI due to the presence of HIV in-fection alone and not contributing comorbidity factors wecannot rule out protective effects against other causes of NCIThe definition of contributing factors as defined by CHAR-TER criteria is that these factors are felt to affect NCI toa minor degree as assigned by the CHARTER group (adju-dicating Neurologists) Within these individuals we speculatethat the contribution of neuroinflammation and CNS oxida-tive stress which are the downstream targets of HO-1 enzy-matic products to NCI is relatively low compared with thatwhich is attributable to HIV infection itself

Figure 4 Short HO-1 (GT)n alleles associatewith lower oxidative stress (sHO-1) in virally suppressedAfrican American PLWH

Virally suppressed individuals with current CD4+ T cell counts greater than 350 cellsμL were examined African Americans (A) but not European Americans(B) with at least 1 short ldquoSrdquoHO-1 (GT)n allele had lowermean plasma sHO-1 (A) statistically insignificant (p = 0076) difference in favor of lower plasma sCD163was observed in African Americans (C) but not in European Americans (p = 0676 not shown) Plasma sCD163 associated with plasma sHO-1 in AfricanAmericans (D) but not in European Americans by Spearman rank correlation (not shown) HO-1 = heme oxygenase-1 sHO-1 = soluble HO-1 PLWH = personsliving with HIV


Previous studies suggested that HO-1 may inhibit HIV in-fection andor replication45 suggesting a mechanism bywhich differential expression of HO-1 could affect HIVpathogenesis both in the periphery and in the CNSHoweverour previous studies did not confirm an effect of HO-1 in-duction or suppression on HIV-1 replication in humanmonocyte-derived macrophages the primary HIV reservoirwithin the CNS43 These in vitro findings are consistent withthe lack of significant associations between either the HO-1(GT)n allele genotype or the A(-413)T SNP and plasma HIVload or CSF HIV load16 We also did not observe associationsbetween either the HO-1 (GT)n allele genotype or the A(-413)T SNP and blood CD4 T-cell counts Thus our studiesindicate that HO-1 does not directly modulate HIV replica-tion and associated immune suppression

We therefore propose an indirect role for HO-1 in modifyingHIV neuropathogenesis mechanisms that could modify therisk for HIV NCI Through our previous analysis of autop-sied brain specimens from PLWH with advanced HIV

disease along with HIV-negative controls (n = 554) weshowed that the presence of 1 or more short HO-1 (GT)nalleles was associated with lower risk of HIV encephalitis(OR = 062) and lower brain expression of type I interferonresponses and T-cell activation markers16 In a subset ofthese autopsy cases (n = 156) we further demonstrated thatbrain HO-1 expression is reduced in PLWH with HIV NCIand that this HO-1 deficiency associates with brain HIVload markers of type I interferon responses and expressionof immunoproteasome subunits4346 Together these au-topsy studies suggest that higher HO-1 expression may limitneuroinflammation providing an indirect mechanism bywhich HO-1 modifies the risk for HIV NCI In in vitrostudies we also demonstrated that HIV infection of mac-rophages the primary CNS target of HIV infection results inloss of HO-1 expression and that this is associated with therelease of neurotoxic levels of glutamate4344 Correction ofthis HO-1 deficiency prevents this neurotoxic effect in vitroThus a state of reduced HO-1 expression within the HIV-infected brain could promote both neuroinflammation and

Figure 5Africans fromBotswana have a higher prevalence of longHO-1 (GT)n alleles than African Americans and EuropeanAmericans

HO-1 (GT)n allele lengths were determined in African PLWH in Botswana (n = 428) who were previously enrolled in cohort studies2021 and allele distributionswere compared with those determined in individuals in our CHARTER and NNTC cohorts (592 European Americans and 446 African Americans)16 (A)Distributions of HO-1 (GT)n allele lengths within European American African American and Botswana cohort groups compared by the Kolmogorov-Smirnovtest Short ldquoSrdquomedium ldquoMrdquo and long ldquoLrdquoHO-1 (GT)n (B) allele and (C) genotype frequencieswithin EuropeanAmerican African American andBotswana cohortgroups compared by the χ2 test CHARTER = CNS HIV Antiretroviral Therapy Effect Research HO-1 = heme oxygenase-1 NNTC = National NeuroAIDS TissueConsortium PLWH = persons living with HIV


neuronal injury through excitotoxic mechanisms and ther-apeutic induction of HO-1 expression might therefore pre-vent or suppress these processes

Collectively our studies strongly support a role for HO-1dysregulation in the neuropathogenesis of HIV infectionthrough regulation of both neuroimmune activation andexcitotoxic injury and they further identify HO-1 inductionas a potential treatment approach for HIV NCI16434446

This HO-1ndashtargeting therapeutic opportunity is likely dis-tinct from therapeutic opportunities that focus on risk re-duction by targeting contributing comorbidity factors thisthus emphasizes the need for specific pharmacologic thera-pies that are adjuncts to suppressive ART22 Our current andprevious studies offer a rationale for targeting of HO-1 ex-pression particularly early after HIV infection as one suchadjunctive approach to address the gap in therapeutics forHIV NCI reduction by ART

Our study highlights HO-1 induction as a therapeutic appli-cation to HIV NCI that might be particularly important inindividuals of African ancestry African Americans have a lowerprevalence of protective short HO-1 (GT)n alleles and a higherprevalence of higher-risk long alleles compared with EuropeanAmericans Because short HO-1 (GT)n repeats produce higherbasal HO-1 promoter activity and inducibility the lowerprevalence of short alleles and higher prevalence of long allelesin African Americans suggests greater vulnerability to oxidativestress and oxidative stressndashinduced injury in individuals of Af-rican ancestry We note that racial assignment within theCHARTER cohort was defined by self-identification and weconfirmed the high concordance (977) with previously ge-netically assigned racial ancestry clusters based on sequencingof ancestry-informative nuclear DNA SNP variants30 Wefound statistically significant associations between the presenceof short alleles and lower OR for HIV NCI when individualswere classified by either genetic ancestry assignment or byself-identification We thus report our observations primarily inself-identified groups based on our original study design theincreased clinical and research utility of self-identification andwith confidence in the applicability to this patient cohort

Our study suggests that individuals of African ancestry may beat a higher risk for developing HIV NCI and indeed somestudies have found a higher incidence of HIV NCI in AfricanAmerican PLWH compared with European AmericanPLWH even after correcting for comorbidity factors47 Thepositive effect of a short HO-1 (GT)n allele in our cohort wasobserved in African Americans and not in European Ameri-cans perhaps because of the higher prevalence of long HO-1(GT)n alleles in African Americans which may contribute toa higher risk for inflammation oxidative stress and HIV NCIThe protective effect of the short HO-1 (GT)n allele appearsto be driven by African ancestry although the test for effect ofrace as a modifier was not statistically significant (p = 011)and our studymight be underpowered to detect such an effectThe question of race itself as a contributing risk factor

warrants further investigation Our data do support a previousreport suggesting increased HIV disease progression risk inAfrican American PLWH because of their higher prevalenceof long HO-1 (GT)n repeats

8 In that study increased HIVdisease risk was inferred from the detection of higher plasmasoluble CD14 levels (suggested as an indicator of monocytetransition to an activated state48) in PLWH on suppressiveART These associations were observed in African AmericanPLWH and not in European American PLWH

Further supporting a role for the HO-1 (GT)n allele genotypein HIV pathogenesis in individuals of African ancestry wefound associations between allele genotype inflammatory andoxidative stress biomarkers and HIV disease status Weassessed plasma soluble CD163 as a marker of inflammationand monocyte activation26 plasma sHO-1 as a marker ofinflammation and oxidative stress2728 and current CD4+

T-cell count (cutoff of 350 cellsμL) as a marker of HIVimmune status3132 We demonstrated that the presence of 1or more short HO-1 (GT)n alleles associates with lowerplasma sHO-1 in virally suppressed African American PLWHbut not in European American PLWH with CD4+ T-cellcounts greater than 350 cellsμL However we did not ob-serve differences in plasma sHO-1 in European Americans orin African Americans with CD4+ T-cell counts less than 350cellsμL These data suggest that the HO-1 (GT)n allele ge-notype may modulate inflammation oxidative stress and riskfor HIV NCI in African American PLWH and that theseprotective effects are particularly expressed in individualswithout severe immunosuppression

We believe that the association between lower plasma sHO-1levels and the presence of ge1 short HO-1 (GT)n alleles inAfrican American PLWH reflects an indirect effect of the shortallele associating with lower levels of inflammation andor oxi-dative stress which in turn associate with less induction of HO-1 expression Thus there may be less HO-1 available forcleavage and release from itsmembrane-bound state within cellsWe do not have direct proof of this hypothesis but we do findsome evidence supporting this in the published literature272849

Several reports of increased plasma sHO-1 levels in humandisease states associated with systemic inflammation providesuch evidence acute HIV infection49 Strongyloides stercoralisinfection28 and coronary artery disease27

A major implication of our study is that individuals of Africanancestry might be more vulnerable to HIV NCI than indi-viduals of European ancestry because of differences in HO-1(GT)n allele genotype prevalence However determining thiswill require additional prospective cohort studies The pro-tective effect of a short HO-1 (GT)n allele may be detected inAfrican American PLWH and not in European AmericanPLWH because of the markedly lower prevalence of longalleles in European Americans which may contribute toa lower risk for inflammation oxidative stress and HIV NCIFurthermore the observation that Africans from Botswanaexpress an even higher prevalence of long HO-1 (GT)n


alleles than African Americans suggests that the effects ofallele genotype on HIV NCI risk may be even more pro-nounced in native African PLWH Thus our studies supportthe need for studies of sufficiently large populations of PLWHworldwide to confirm risk-modifying effects of the HO-1(GT)n allele genotype in different racial groups and to moreeffectively define HIV NCI risk worldwide50 We suggest thatadjunctive HO-1ndashinducing therapeutics may reduce HIVNCI in virally suppressed PLWH and that such therapies maybe particularly beneficial to individuals of African ancestrywhere endogenous HO-1 expression may be limited

AcknowledgmentThe authors acknowledge the support and technical assistanceof the Genomics Analysis Core (Jennifer Rosado ResearchSpecialist) at the University of Pennsylvania Perelman Schoolof Medicine They also acknowledge the support of theNational NeuroAIDS Tissue Consortium and the CHARTERgroup for providing all DNA samples and cohort clinical andparaclinical data and the Botswana Ministry of Health andWellness for ethics oversight and approval

Study fundingThis work was funded by the National Institute of MentalHealth (R01 MH104134 DLK R01 MH111389 DLK R01MH107345 SLL K24 MH097673 SLL R01 MH080701 RGross R01 MH101017 BBG and F30 MH102120 AJG)the National Institute of General Medical Sciences (R25GM071745 R Garza) the Fogarty International Center(D43 TW009781 R Gross) and support from the PennCenter for AIDS Research (P30 AI45008) the Penn MentalHealth AIDS Research Center (P30 MH 097488) andthe UCSD HIV Neurobehavioral Research Center (P30MH62512)

DisclosureR Garza AJ Gill BL Bastien Y Garcia-Mesa AL Grue-newald BB Gelman B Tsima R Gross SL Letendreand DL Kolson is Associate Editor for Neurology Neuro-immunology amp Neuroinflammation Go to NeurologyorgNNfor full disclosures

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationOctober 30 2019 Accepted in final form February 21 2020

References1 Gozzelino R Jeney V Soares MP Mechanisms of cell protection by heme oxygenase-

1 Annu Rev Pharmacol Toxicol 201050323ndash3542 Lee H Choi YK Regenerative effects of heme oxygenase metabolites on neuro-

inflammatory diseases Int J Mol Sci 201820E783 Wagener FA Volk HD Willis D et al Different faces of the heme-heme oxygenase

system in inflammation Pharmacol Rev 200355551ndash5714 Schipper HM Song W A heme oxygenase-1 transducer model of degenerative and

developmental brain disorders Int J Mol Sci 2015165400ndash54195 Chen YH Lin SJ Lin MW et al Microsatellite polymorphism in promoter of heme

oxygenase-1 gene is associated with susceptibility to coronary artery disease in type 2diabetic patients Hum Genet 20021111ndash8

6 Hirai H Kubo H Yamaya M et al Microsatellite polymorphism in heme oxygenase-1gene promoter is associated with susceptibility to oxidant-induced apoptosis inlymphoblastoid cell lines Blood 20031021619ndash1621

7 Rueda B Oliver J Robledo G et al HO-1 promoter polymorphism associated withrheumatoid arthritis Arthritis Rheum 2007563953ndash3958

8 Seu L Burt TD Witte JS Martin JN Deeks SG McCune JM Variations in the hemeoxygenase-1 microsatellite polymorphism are associated with plasma CD14 and viralload in HIV-infected African-Americans Genes Immun 201213258ndash267

9 Ono K Goto Y Takagi S et al A promoter variant of the heme oxygenase-1 gene mayreduce the incidence of ischemic heart disease in Japanese Atherosclerosis 2004173315ndash319

10 Cao L Zhang Z Cai B et al Association of heme oxygenase-1 gene rs2071746polymorphism with vascular outcomes in patients with atherosclerotic strokeJ Neurol Sci 2014344154ndash157

Appendix Authors

Name Location Contribution

RolandoGarza BS

University ofPennsylvaniaPhiladelphia

Conducted genotypingexperiments analyzed andinterpreted genotype data anddrafted the original manuscript

Alexander JGill MD PhD


Conceptualized the study anddesigned experimentsanalyzed and interpretedgenotype data and drafted theoriginal manuscript

Appendix (continued)


Brandon LBastien BA


Conducted genotypingexperiments analyzed andinterpreted genotype data andcritically reviewed edited andapproved the final version ofthe manuscript

YoelvisGarcia-MesaPhD


Analyzed and interpretedgenotype data and criticallyreviewed edited and approvedthe final version of themanuscript

Analise LGruenewaldBS



Benjamin BGelman MDPhD

University of TexasMedical BranchGalveston TX

Curated NNTC cohort samplesand contributed to analysis ofdata and critically reviewededited and approved the finalversion of the manuscript

Billy TsimaMD MSCE

University ofBotswanaGaborone Botswana

Curated the Botswana samplesand critically reviewed editedand approved the final versionof the manuscript

Robert GrossMD MSCE



Scott LLetendreMD

University ofCalifornia San DiegoSan Diego

Curated CHARTER cohortsamples and provided supportfor analysis of neurocognitivedata and critically reviewededited and approved the finalversion of the manuscript

Dennis LKolson MDPhD


Conceptualized the study anddesigned experiments draftedthe original manuscript andcritically reviewed edited andapproved the final version ofthe manuscript


11 Mateo I Sanchez-Juan P Rodriguez-Rodriguez E et al Synergistic effect of hemeoxygenase-1 and tau genetic variants on Alzheimerrsquos disease risk Dement GeriatrCogn Disord 200826339ndash342

12 Sponholz C Huse K Kramer M et al Gene polymorphisms in the heme degradationpathway and outcome of severe human sepsis Shock 201238459ndash465

13 Chen M Zhou L Ding H et al Short (GT) (n) repeats in heme oxygenase-1 genepromoter are associated with lower risk of coronary heart disease in subjects with highlevels of oxidative stress Cell Stress Chaperones 201217329ndash338

14 Bai CH Chen JR Chiu HC Chou CC Chau LY Pan WH Shorter GT repeatpolymorphism in the heme oxygenase-1 gene promoter has protective effect onischemic stroke in dyslipidemia patients J Biomed Sci 20101712

15 Vazquez-Armenta G Gonzalez-Leal N JV-dlT M et al Short (GT)n microsatelliterepeats in the heme oxygenase-1 gene promoter are associated with antioxidant andanti-inflammatory status in Mexican pediatric patients with sepsis Tohoku J Exp Med2013231201ndash209

16 Gill AJ Garza R Ambegaokar SS Gelman BB Kolson DL Heme oxygenase-1 pro-moter region (GT)n polymorphism associates with increased neuroimmune activa-tion and risk for encephalitis in HIV infection J Neuroinflammation 20181570

17 Heaton RK Clifford DB Franklin DR Jr et al HIV-associated neurocognitivedisorders persist in the era of potent antiretroviral therapy CHARTER StudyNeurology 2010752087ndash2096

18 Heaton RK Franklin DR Jr Deutsch R et al Neurocognitive change in the era ofHIV combination antiretroviral therapy the longitudinal CHARTER study ClinInfect Dis 201560473ndash480

19 Jia P Zhao Z Hulgan T et al Genome-wide association study of HIV-associatedneurocognitive disorder (HAND) a CHARTER group study Am J Med Genet BNeuropsychiatr Genet 2017174413ndash426

20 Gross R Bellamy SL Ratshaa B et al CYP2B6 genotypes and early efavirenz-basedHIV treatment outcomes in Botswana AIDS 2017312107ndash2113

21 Vujkovic M Bellamy SL Zuppa AF et al Brief report CYP2B6 516GgtT minor alleleprotective of late virologic failure in efavirenz-treated HIV-infected patients in Bot-swana J Acquir Immune Defic Syndr 201775488ndash491

22 Antinori A Arendt G Becker JT et al Updated research nosology for HIV-associatedneurocognitive disorders Neurology 2007691789ndash1799

23 Robertson K Yosief S Neurocognitive assessment in the diagnosis of HIV-associatedneurocognitive disorders Semin Neurol 20143421ndash26

24 Burdo TW Weiffenbach A Woods SP Letendre S Ellis RJ Williams KC ElevatedsCD163 is a marker of neurocognitive impairment in HIV-infected individuals oneffective ART AIDS 2013271387ndash1395

25 Imp BM Rubin LH Tien PC et al Monocyte activation is associated with worsecognitive performance in HIV-infected women with virologic suppression J Infect Dis2017215114ndash121

26 Etzerodt A Moestrup SK CD163 and inflammation biological diagnostic andtherapeutic aspects Antioxid Redox signaling 2013182352ndash2363

27 Idriss NK Lip GY Balakrishnan B Jaumdally R Boos CJ Blann AD Plasmahaemoxygenase-1 in coronary artery disease A comparison with angiogenin matrixmetalloproteinase-9 tissue inhibitor of metalloproteinase-1 and vascular endothelialgrowth factor Thromb Haemost 20101041029ndash1037

28 Rajamanickam A Munisankar S Bhootra Y Dolla C Nutman TB Babu S Microbialtranslocation associated with an acute-phase response and elevations in MMP-1 HO-1 and proinflammatory cytokines in Strongyloides stercoralis infection Infect Immun201785e00772

29 Grant I Franklin DR Jr Deutsch R et al Asymptomatic HIV-associated neuro-cognitive impairment increases risk for symptomatic decline Neurology 2014822055ndash2062

30 Hulgan T Samuels DC Bush W et al Mitochondrial DNA haplogroups and neu-rocognitive impairment during HIV infection Clin Infect Dis 2015611476ndash1484

31 Ford N Meintjes G Vitoria M Greene G Chiller T The evolving role of CD4 cellcounts in HIV care Curr Opin HIV AIDS 201712123ndash128

32 May MT Gompels M Delpech V et al Impact on life expectancy of HIV-1 positiveindividuals of CD4+ cell count and viral load response to antiretroviral therapy AIDS2014281193ndash1202

33 Heaton RK Franklin DR Ellis RJ et al HIV-associated neurocognitive disordersbefore and during the era of combination antiretroviral therapy differences in ratesnature and predictors J Neurovirol 2011173ndash16

34 Scott JC Woods SP Carey CL et al Neurocognitive consequences of HIV infectionin older adults an evaluation of the ldquocorticalrdquo hypothesis AIDS Behav 2011151187ndash1196

35 Mothobi NZ Brew BJ Neurocognitive dysfunction in the highly active antiretroviraltherapy era Curr Opin Infect Dis 2012254ndash9

36 Chan P Brew BJ HIV associated neurocognitive disorders in the modern antiviraltreatment era prevalence characteristics biomarkers and effects of treatment CurrHIVAIDS Rep 201411317ndash324

37 Saylor D Dickens AM Sacktor N et al HIV-associated neurocognitive disorder -pathogenesis and prospects for treatment Nat Rev Neurol 201612309

38 Mollace V Nottet HS Clayette P et al Oxidative stress and neuroAIDS triggersmodulators and novel antioxidants Trends Neurosciences 200124411ndash416

39 Hulgan T Morrow J DrsquoAquila RT et al Oxidant stress is increased during treatmentof human immunodeficiency virus infection Clin Infect Dis 2003371711ndash1717

40 Gill AJ KolsonDL Chronic inflammation and the role for cofactors (hepatitis C drugabuse antiretroviral drug toxicity aging) in HAND persistence Curr HIVAIDS Rep201411325ndash335

41 Akay C Cooper M Odeleye A et al Antiretroviral drugs induce oxidative stress andneuronal damage in the central nervous system J Neurovirol 20142039ndash53

42 Cross SA Cook DR Chi AW et al Dimethyl fumarate an immune modulator andinducer of the antioxidant response suppresses HIV replication and macrophage-mediated neurotoxicity a novel candidate for HIV neuroprotection J Immunol 20111875015ndash5025

43 Gill AJ Kovacsics CE Cross SA et al Heme oxygenase-1 deficiency accompaniesneuropathogenesis of HIV-associated neurocognitive disorders J Clin Invest 20141244459ndash4472

44 Gill AJ Kovacsics CE Vance PJ Collman RG Kolson DL Induction of hemeoxygenase-1 deficiency and associated glutamate-mediated neurotoxicity is a highlyconserved HIV phenotype of chronic macrophage infection that is resistant to anti-retroviral therapy J Virol 20158910656ndash10667

45 Devadas K Dhawan S Hemin activation ameliorates HIV-1 infection via hemeoxygenase-1 induction J Immunol 20061764252ndash4257

46 Kovacsics CE Gill AJ Ambegaokar SS Gelman BB Kolson DL Degradation of hemeoxygenase-1 by the immunoproteasome in astrocytes a potential interferon-gamma-dependentmechanism contributing toHIV neuropathogenesis Glia 2017651264ndash1277

47 Cross S Onen N Gase A Overton ET Ances BM Identifying risk factors for HIV-associated neurocognitive disorders using the international HIV dementia scaleJ Neuroimmune Pharmacol 201381114ndash1122

48 Shive CL Jiang W Anthony DD Lederman MM Soluble CD14 is a nonspecificmarker of monocyte activation AIDS 2015291263ndash1265

49 Angin M Fathi A King M et al Acute HIV-1 infection is associated with increasedplasma levels of heme oxygenase-1 and presence of heme oxygenase-1-specific reg-ulatory T cells AIDS 201731635ndash641

50 Robertson K Liner J Hakim J et al NeuroAIDS in africa J Neurovirol 201016189ndash202


DOI 101212NXI000000000000071020207 Neurol Neuroimmunol Neuroinflamm

Rolando Garza Alexander J Gill Brandon L Bastien et al impairment

polymorphism associates with HIV neurocognitivenHeme oxygenase-1 promoter (GT)

This information is current as of April 10 2020

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm

ServicesUpdated Information amp

httpnnneurologyorgcontent73e710fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent73e710fullhtmlref-list-1

This article cites 50 articles 5 of which you can access for free at

Citations httpnnneurologyorgcontent73e710fullhtmlotherarticles

This article has been cited by 1 HighWire-hosted articles

Subspecialty Collections

httpnnneurologyorgcgicollectionhivHIV

httpnnneurologyorgcgicollectionall_geneticsAll Genetics a

httpnnneurologyorgcgicollectionall_cognitive_disorders_dementiAll Cognitive DisordersDementiafollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online



The syndrome of HIV-associated neurocognitive disorders(HAND) like many CNS disease states is linked to neuro-inflammation and oxidative stress Induction of the antioxi-dant isoenzyme heme oxygenase-1 (HO-1) can reduceoxidative stress and neuroinflammation within the CNS12

HO-1 expression is rapidly induced in response to injury andinduced HO-1 expression is neuroprotective in vitro and invivo in different injury models34

Transcriptional regulation of the HO-1 gene (HMOX1)depends partly on HO-1 promoter region genetic variationsincluding a (GT)n dinucleotide repeat and single nucleotidepolymorphisms (SNPs) Promoters with short HO-1 (GT)nrepeats express higher basal activity and inducibility5ndash8 andcells with these repeats are more resistant to oxidative injury6

An A(-413)T SNP can affect HO-1 promoter transcriptionalactivity and disease outcomes9 albeit less so than the HO-1(GT)n repeat length

10ndash12 Short HO-1 (GT)n repeat allelesassociate with better outcomes in inflammatory and oxidativestressndashassociated diseases7813ndash15 and the prevalence of theseshort alleles differs significantly among different populationsWe recently determined HO-1 (GT)n alleles in individuals inan HIV autopsy cohort (n = 554) and showed that thepresence of 1 or more short HO-1(GT)n alleles associatedwith a lower risk of HIV encephalitis and less brain in-flammation (type I interferon responses and T-cell activa-tion)16 No effect of the A(-413)T SNP was observed

We hypothesized that the HO-1 (GT)n allele genotype alsoassociates with risk for neurocognitive impairment (NCI) inpersons living with HIV (PLWH) and that genotype effectsvary among different populations We genotyped a large clin-ically characterized cohort of PLWH the CNS HIV Anti-retroviral Therapy Effect Research (CHARTER) cohort1718

and associated genotype with neurocognitive diagnosis Wereport significant associations indicating a protective effect ofshort HO-1(GT)n alleles against HIV-associated NCI (HIVNCI) particularly in African Americans

MethodsCohortsFor analysis of HO-1 (GT)n allele genotypes neurocognitivediagnosis and plasma biomarkers 606 PLWH were selectedfrom the CHARTER Genetics Cohort19 based on reportedavailability of neurocognitive diagnosis data and DNA sam-ples and without other inclusionary or exclusionary criteria

to avoid potential bias Of the 606 DNA samples 3 were ofinsufficient quality to obtain reliable (GT)n sequence datatherefore HO-1 (GT)n allele data from 603 individuals werepresented (table 1 and figure 1A) CHARTER is an ongoingobservational study of PLWH enrolled between 2003 and2007 (1561 individuals) from 6HIV treatment centers JohnsHopkins University (Baltimore MD) Mt Sinai School ofMedicine (New York NY) University of California at SanDiego (San Diego CA) University of Texas Medical Branch(Galveston TX) University of Washington (Seattle WA)and Washington University (St Louis MO) Our requiredsample size was estimated at 400 based on our analysis ofthe National NeuroAIDS Tissue Consortium (NNTC) au-topsy cohort which revealed the presence of short HO-1(GT)n allele associated with an increased risk of HIVencephalitis16

Among these 603 individuals 595 had sufficient neuro-psychological (NP) test data for diagnostic HIV NCI assign-ment Therefore NCI data from those 595 individuals wereanalyzed forHIVNCI associations with genotype and subgroupanalyses Among these 595 individuals data from 528 total self-identified African Americans and European Americans wereanalyzed to distinguish differences between these groups

HO-1 (GT)n allele genotyping was also performed on wholebloodndashderived DNA from a Botswana cohort of PLWH onantiretroviral therapy (ART) randomly selected irrespective ofneurocognitive function2021 and under separate IRB approval

Data availabilityAnonymized data are available from the corresponding author(DLK) on reasonable request

Neurocognitive and functional assessmentsIndividuals from CHARTER completed baseline assessmentsincluding NP testing (7 cognitive domains corrected for ageeducation sex and ethnicity) substance abuse history psy-chiatric diagnosis self-reported cognition assessment voca-tional function assessment and assessment of independencewith instrumental activities of daily living (ADLs)17 Specimensand data were obtained with permission of the CHARTERsteering committee Neurocognitive diagnosis was first de-termined by the Frascati criteria for HIV-associated neuro-cognitive disorders22 Under these criteria the diagnoses ofasymptomatic neurocognitive impairment (ANI) and mildneurocognitive disorder (MND) require the presence of mildimpairment in ge2 NP domains that is not attributed to

GlossaryADL = activities of daily livingANI = asymptomatic neurocognitive impairmentART = antiretroviral therapyGT = promoterHAD = HIV-associated dementia HAND = HIV-associated neurocognitive disorders HIV NCI = HIV-associated NCI HO-1 = heme oxygenase-1 MND = mild neurocognitive disorder NCI = neurocognitive impairment NCN = neurocognitivenormal NNTC = National NeuroAIDS Tissue Consortium NP = neuropsychological PLWH = persons living with HIVsHO-1 = soluble HO-1 SNP = single nucleotide polymorphism







Characteristic






Sex ()

Male 800 806 793 0685b

Race ()












































Parameter

Racial group





Sex ()

Male 714 889 lt00001b




HAND () 392 431 0376b

































































Appendix Authors


RolandoGarza BS








Brandon LBastien BA












Billy TsimaMD MSCE



Robert GrossMD MSCE



Scott LLetendreMD


































































Reprints









Characteristic






Sex ()

Male 800 806 793 0685b

Race ()












































Parameter

Racial group





Sex ()

Male 714 889 lt00001b




HAND () 392 431 0376b

































































Appendix Authors


RolandoGarza BS








Brandon LBastien BA












Billy TsimaMD MSCE



Robert GrossMD MSCE



Scott LLetendreMD


































































Reprints































Parameter

Racial group





Sex ()

Male 714 889 lt00001b




HAND () 392 431 0376b

































































Appendix Authors


RolandoGarza BS








Brandon LBastien BA












Billy TsimaMD MSCE



Robert GrossMD MSCE



Scott LLetendreMD


































































Reprints

























































Appendix Authors


RolandoGarza BS








Brandon LBastien BA












Billy TsimaMD MSCE



Robert GrossMD MSCE



Scott LLetendreMD


































































Reprints































































Reprints




heme oxygenase-1 promoter (gt)n polymorphism associates

Documents