A common functional <i>PIEZO1</i> deletion allele associates with red blood cell density in sickle cell disease patients

PIEZO1 encodes a large mechanosensitive cation channel expressed in multiple cell types, including red blood cells (RBCs). In humans, rare gain-of-function mutations in PIEZO1 cause hereditary xerocytosis (HX), characterized by RBC dehydration and anemia. Recently, Ma et al. identified an in-frame PIEZO1 deletion allele (rs572934641) that is common in individuals of African ancestry.1 In vitro, the deletion increased PIEZO1 inactivation time, mimicking other gain-of-function mutations found in HX patients. RBCs from nine healthy African Americans heterozygotes for rs572934641 were dehydrated when compared to erythrocytes from noncarriers.1 RBC dehydration has been implicated in the clinical variability observed in patients with sickle cell disease (SCD), a multiorgan disorder caused by mutations in the β-globin gene. Increased RBC density, a hallmark of SCD, is independently correlated with hemolysis, priapism, leg ulcer, and renal dysfunction in patients.2 Here, we investigated the association between the common functional PIEZO1 deletion allele and RBC density, hemolytic parameters, estimated glomerular filtration rate (eGFR), and clinical complications in three large SCD cohorts. Our results indicate that common genetic variation in PIEZO1 regulates RBC density in SCD patients, and thus represents one of many factors that influence clinical severity in this heterogenous blood disorder. This project was reviewed and approved by the Montreal Heart Institute Ethics Committee and the different recruiting centers. Informed consent was obtained for all participants in accordance with the Declaration of Helsinki. The GEN-MOD cohort, the Cooperative Study of Sickle Cell Disease (CSSCD), and the Duke University Outcome Modifying Genes (OMG) cohort have been described elsewhere.3, 4 RBC density was measured using the phthalate density distribution technique in GEN-MOD participants.2 DNA genotyping, quality-control, and genotype imputation using haplotypes from phase 3 of the 1000 Genomes Project were described previously.3, 4 We used Nimblegen SeqCap EZ Exome Capture kit to capture exons and we sequenced DNA using the Illumina HiSeq4000 instrument and a paired-ends 2x100 base pairs protocol. Whole-exome sequencing (WES) analysis was performed using the Genome Analysis Toolkit (GATK version 3.4-46). We followed best practices pipeline recommendations for reads quality control and variant calling. All statistical analyses are described in the Supporting Information Methods. rs572934641 maps to a complex DNA sequence region in PIEZO1 with multiple TCC trinucleotide repeats. We inspected high-coverage WES data for 247 SCD patients from GEN-MOD (80% of targeted sequences covered at ≥80X) and identified at least four in-frame alleles. The most frequent allele has two TCC repeats and a frequency of 74.9%. The allele with one TCC repeat, which corresponds to the deletion allele (E756del) characterized recently,1 has a frequency of 22.9%. We also found two rarer alleles with zero (allele frequency = 0.6%) or three (allele frequency = 1.6%) repeats. In 226 SCD patients from GEN-MOD with phenotype and WES-derived genotypes available, the one repeat allele (E756del) was associated with increased RBC density (P = .043, Table 1). To increase our sample size, we imputed this common PIEZO1 deletion allele in 375 GEN-MOD participants (including 149 additional SCD patients without WES data available) using reference haplotypes from phase 3 of the 1000 Genomes Project. The imputation quality metric was excellent (rsq_hat = 0.94) and concordance with genotypes from WES data was high (Supporting Information Figure 1). In this data set of 374 SCD patients, the association between RBC density and imputation-derived genotypes for the PIEZO1 E756del allele was stronger and explained ~2.5% of the phenotypic variation (P = .0039, Table 1). Thus, we provide in vivo evidence that RBCs from SCD patients that carry the PIEZO1-E756del allele are dehydrated.1 We previously reported an association between RBC density in SCD patients and a regulatory DNA variant (rs10751450) within an erythroid enhancer at the ATP2B4 locus.3, 5 ATP2B4 encodes the main RBC calcium pump and erythroid cells with a deletion of the enhancer have increased intracellular Ca2+ concentration.5 Increased intracellular calcium can activate the Gardos channel, leading to potassium efflux and dehydration. Because a gain-of-function mutation in PIEZO1 could similarly result in excess calcium entry into RBCs, we tested if genotypes at ATP2B4-rs10751450 and PIEZO1-rs572934641 interacted to control RBC density in SCD patients. Both variants were independently associated with RBC density (P ATP2B4 = .0016 and P PIEZO1 = .0034 in a multivariate model), but we detected no evidence of interaction on RBC density (Pinteraction = .53). We calculate a polygenic RBC density score by combining alleles from PIEZO1-rs572934641, ATP2B4-rs10751450, as well as rare missense variants in PIEZO1, ATP2B4, and the Gardos channel gene KCNN4 (Supporting Information Table 1). This polygenic score was strongly associated with RBC density (P = 2.4 × 10−4, 6.3% of variance explained, N = 226 SCD patients) (Figure 1). Finally, we asked if the PIEZO1 deletion allele was associated with SCD-related clinical phenotypes that are correlated with RBC density.2 To maximize our sample size, we analyzed phenotype-genotype associations from 402 GEN-MOD participants, as well as 1081 and 552 SCD patients from the CSSCD and OMG study, respectively. Imputation quality for the common PIEZO1 deletion allele in the CSSCD (rsq_hat = 0.90) and OMG study (rsq_hat = 0.93) was sufficiently high for association testing. Given the sample size of our study design, we saw no evidence of association between the PIEZO1 E756del allele and tested SCD clinical phenotypes (Table 1). In conclusion, we provide in vivo evidence that a common PIEZO1 deletion allele is associated with RBC dehydration in SCD patients. Further, a simple polygenic score considering genetic variants at key RBC hydration genes (PIEZO1, ATP2B4, KCNN4) improves the association with RBC density. As for fetal hemoglobin,6 we anticipate that larger studies of RBC density genetics will provide new insights into SCD clinical heterogeneity. We thank all participants for their contribution to this project, and Nassima Djouder and Jugurtha Berkenou for help with the recruitment of GEN-MOD. G.L. is funded by the Canadian Institutes of Health Research (PJT #156248), the Doris Duke Charitable Foundation, and the Canada Research Chair program. Y.I. and G.L. conceived and designed the experiments; Y.I. and M.E.G. performed the experiments; P.B., A.A.K., M.T., C.B., and F.G. contributed DNA samples, clinical information, and expert knowledge; Y.I., M.E.G. and G.L. analyzed the results; Y.I. and G.L. wrote the manuscript with contributions from all authors. The authors declare no competing financial interests. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.