We analyzed the correlation between genotype and degree of hemolysis in 23 HS patients utilizing the RBC lifespan detected by Levitt's CO breath test. To our knowledge, this is the first study focusing on genotype-hemolytic phenotype association in HS. It is the first time to measure RBC lifespan by Levitt's CO breath test to assess the degree of hemolysis in HS alone, excluding the effect of bone marrow hematopoietic compensation. In this investigation, we found no clear correction between the genotype and degree of hemolysis.
Clinical studies exploring genotype-phenotype corrections in HS have remained contentious. Tole et al3 reported that children with SLC4A1 mutations had the mildest phenotype, with the highest Hb levels, the lowest reticulocyte counts and unconjugated bilirubin levels. Nevertheless, van Vuren et al2 found that mutations in ANK1 and SPTB generated a more severe clinical phenotype characterized by lower Hb levels and higher reticulocyte counts compared with mutations in SPTA1. However, Hb, as an indicator of HS phenotype, is affected by both the degree of hemolysis and hematopoietic compensation, and hence cannot objectively reflect the effect of genotype on clinical phenotype in HS. We evaluated RBC lifespan by Levitt's CO breath test to detect the degree of hemolysis independently, and we found that the degree of hemolysis in HS was unrelated to membrane skeletal protein mutation genes, mutation types or mutation sites.
Ankyrin-1 plays a central role in stabilizing the erythrocyte membrane by attaching spectrin dimers to the transmembrane domain of band 3. Concomitant spectrin and ankyrin-1 deficiency could be observed in HS patients’ erythrocytes, and it is postulated that aberrant ankyrin-1 expression may contribute to secondary spectrin deficiency15. Previous studies have shown more severe anemia and a higher rate of splenectomy in ANK1-HS than SPTB-HS16,17, thus ANK1-HS was speculated to manifest more severe hemolysis. Tole et al. concluded that children with SPTA1-HS had the most severe clinical phenotype with the lowest Hb levels and almost all patients undergoing splenectomy in early childhood3. In our study, 3 patients with RBC lifespan of more than 30 days carried SPTB or SLC4A1 mutations, and patients with ANK1 or SPTA1 mutations showed a trend toward more severe hemolysis. However, there was no significant difference in RBC lifespan between different mutation genes, probably due to sample size constraints.
The types of mutations in HS could be categorized as missense and non-missense mutations. Missense mutations that interfere with a single interaction within the ankyrin complex have little effect on the membrane skeleton9. Therefore, it is reasonable to speculate that missense mutations may generate a relatively modest hemolytic phenotype. There was no statistical difference in RBC lifespan between missense and non-missense mutations in our investigation. Notably, 2 patients with the RBC lifespan of nearly 40 days both carried missense mutations, suggesting that missense mutations may be associated with milder hemolysis. However, a larger sample size is required to further investigate the association between the degree of hemolysis and mutation types. It was previously assumed that non-missense mutations would mainly result in the premature introduction of stop codons, leading to either expression of truncated protein or non-expression of relevant alleles, rather than impacting protein function9. This seemed to imply that non-missense mutations might induce comparable phenotypes, while the data of van Vuren did not support the hypothesis. They proposed that the incorporation of truncated proteins would disrupt cytoskeleton function and thus non-missense mutations leading to truncated protein expression would result in more severe clinical manifestations compared to allelic deletions that reduce the amount of normally formed protein2. In our cohort, one patient with SPTB splicing mutation had the considerably longer RBC lifespan than other non-missense mutations. We speculated that although the mutation altered the mRNA splicing process, generating the truncated protein expression, the produced protein preserved crucial structural domains such as the spectrin binding domain, and hence had less impact on the membrane skeleton, resulting in much milder hemolysis. This seemed to suggest that whether the incorporated truncated proteins preserved essential structural domains may also affect the degree of hemolysis in HS.
The distribution of mutation regions in membrane proteins was reported to be a potential factor influencing the clinical phenotype. It has previously been demonstrated that the interaction between ankyrin and β-spectrin was critical for erythrocyte deformability and stability. Mutations in the spectrin-binding domain of ANK1 disrupted this interaction, resulting in more severe cytoskeleton assembly or function disruption and thereby leading to a more severe clinical phenotype17. Similarly, van Vuren et al2 proposed that mutations in the ANK1, SPTB, and SPTA1 spectrin-binding domains contributed to more severe phenotypes. In our cohort, we observed that 4 HS patients with mutations in the non-spectrin-binding domain had the longer RBC lifespan, suggesting that mutations in the spectrin-binding domain may lead to a more severe hemolytic phenotype. However, we did not discover a clear relationship between the degree of hemolysis and mutation sites, which could possibly be attributed to the small sample size.
Notably, Wang et al18 reported that 2 HS patients with the same genotype showed distinct levels of MCV and MCHC, suggesting the two individuals had different clinical phenotypes. The methylation level of the ANK1 promoter region was associated with ANK1 expression19. We speculated that the varied ANK1 expression, due to the variance in methylation levels of the ANK1 promoter region, contributed to the heterogeneity of hemolytic phenotype in patients with the same genotype. It was assumed that factors such as the level of methylation in the promoter region altered gene expression levels, accounting for variable degree of hemolysis.
There are several limitations to this study. First, since the majority of the patients included in this cohort were adults, we may underreport cases with the more severe phenotype of HS as children. Second, certain severe mutations are lethal and the absence of extremely severe HS cases might result in an inaccurate conclusion. Finally, the small sample size may have skewed the results. With the expansion of the sample size and the conduction of further research, more detailed and comprehensive results will be obtained.
In conclusion, we found no clear correction between the genotype and degree of hemolysis in HS. Since both hemolysis and bone marrow hematopoietic compensation influenced the severity of clinical manifestations, we hypothesized that the genotype was not the source of the variability of HS clinical presentation, which was consistent with the previous study20.