Acquired Mutations in the Mitochondrial Genome As Prognostic Markers in Sickle Cell Disease

Introduction: Acquired mutations in the mitochondrial genome, referred to as mtDNA heteroplasmy, and alteration in mitochondria DNA (mtDNA) copy number (mtDNA-CN), are emerging markers of aging and inflammation. Given that the pathology in sickle cell disease (SCD) is driven by inflammation leading to premature organ damage, in a process often referred to as “inflammaging”, we hypothesized that these mtDNA dysfunction readouts could have prognostic value in SCD. Here, we investigated if mtDNA heteroplasmy and mtDNA-CN correlate with SCD genotype severity, age and survival in humans. Additionally, we explored the differential tissue distribution of mtDNA heteroplasmy in SCD mice.

Methodology: Human subjects included three adult cohorts (≥18 years old) of African descent enrolled under IRB approved protocols NCT00011648, NCT00081523, and NCT03685721. Details of cohorts: Cohort-1 [total 673 SCD patients: 538 HbSS + 16 HbS-Beta-thalassemia⁰ (554 SCA), 91 HbSC, 25 HbS-Beta-thalassemia⁺ (SB+), 3 other SCD genotypes]; Cohort-2 [total 171: 113 SCA, 30 HbAS, 16 HbAA, 12 other SCD genotypes]; Cohort-3 [total 199: 96 SCA, 47 HbAS, 44 HbAA, 12 other SCD genotypes]. Whole genome sequence (WGS) was performed on the human DNA samples. Humanized Townes SCD mice comprised of 18 littermates [6 HbSS, 6 HbAS, 6 HbAA; 3 males and 3 females in each genotype]. mtDNA was specifically amplified from genomic DNA from 9 organs (heart, lung, brain, kidney, liver, muscle, blood, bone marrow, and spleen; total 162 samples) at age 16 weeks and subjected to deep sequencing. Sequencing reads mapped to mtDNA genome (mito-genome) were utilized for mtDNA-CN estimation and variant calling (LoFreq). Allele frequency (AF) cutoffs of 1%, 5%, 1%, and 4% were applied for Cohorts-1, 2, 3, and mice, respectively, based on their respective mito-genome coverage. Survival data in Cohort-1 were ascertained by proxy interview, medical records, National Death Index, and Social Security Death Index search. Cox proportional hazards model was applied to analyze the relationship between mtDNA heteroplasmy with all-cause mortality.

Results: Overall, considering all 3 human cohorts, mtDNA heteroplasmy burden increased according to severity of SCD genotypes: HbAA < HbAS < SB+ < HBSC < SCA. mtDNA-CN was significantly higher in SCA genotype (p<0.001) compared to all other genotypes across all 3 human cohorts. mtDNA heteroplasmy also increased with age among all SCD patients in cohort-1; mean heteroplasmies: 0.39, 0.39, 0.47, and 0.58 in age (year) groups <40, 41-50, 51-60, and >60, respectively. In mice, a similar pattern of increasing mtDNA heteroplasmy was also observed: HbAA < HbAS < HbSS. Additionally, mtDNA heteroplasmy varied across different tissues within and across genotypes in the mice; with spleen having the highest burden and liver, the lowest. We observed a pattern across the mito-genome with D-loop, RNR1, ND1, COX1, ATP6, ND5, CYT-B regions exhibiting higher heteroplasmic load consistently across HbAA, HbAS and HbSS/SCA genotypes in both humans and mice. In Cohort-1(n=673) where we could ascertain survival data, we observed a positive trend between mtDNA heteroplasmy and all cause-mortality [H.R with 95% CI: Heteroplasmy continuous = 1.01 (0.853 – 1.197); 1 heteroplasmy vs 0 = 1.173 (0.873 – 1.576); 2 hetroplasmy vs 0 = 1.3 (0.806 – 2.096)] that was sustained after correction for age. Seven variants were significantly associated with mortality with HR ranging from 22.09 to 153.98 (P<0.05); of which 6 variants originated from the Complex-I, IV, V genes.

Conclusion: mtDNA variants have prognostic value in SCD. mtDNA heteroplasmy burden increased progressively (HbAA<HbAS < HbSB+ < HbSC < HbSS & HbSβ⁰) with genotypic groups that are associated with increasing phenotypic severity, in both humans and the SCD mouse model. mtDNA heteroplasmic burden also increased with age. The pattern of heteroplasmy indicates that there could be potential hotspots in the mito-genome which are likely to be more susceptible to acquire mtDNA mutations. Finally, specific mtDNA variant(s) appear to impact overall mortality to a much greater extent than the quantity of mtDNA heteroplasmies. Future plans include investigating the rate of accumulation of mtDNA mutations with age among the different genotypes and if there is a threshold above which mtDNA heteroplasmy increases exponentially, in longitudinal studies.