Purpose: Studying stress erythropoiesis (SE) in sickle cell anemia (SCA) is difficult because of the rigor in accessing bone marrow samples. However, the characterization of circulating cell populations reflecting SE is feasible and critical, especially in the perspectives of gene therapy. Materials and methods: Clinical and laboratory data were collected from 49 subjects with SCA (SS-genotype) under chronic transfusion (CT, n = 15), hydroxyurea therapy (HC, n = 20), or under both treatments (CT/HC, n = 14) at the Blood Center of Ribeirão Preto - Brazil. The mean age was 35.22 y/o (20-64), and 53.06% were males. Six patients from each group, under steady-state, and five healthy donors (HD group) were selected for downstream analysis of PBMCs. Briefly, 5 mL of blood were collected in EDTA-evacuated tubes and processed immediately with Ficoll-Paque PLUS (Cytiva). 106 live cells were stained with anti-CD235a, -CD123, -CD36, and -CD34 (stemness and erythroid maturation markers) BD Biosciences antibodies and analyzed by FACS in a BD FACSCanto II. FACS analysis was conducted in the FlowJo software v.11, using conventional gating strategies and dimensionality reduction machine learning algorithms. Complete blood and reticulocyte count, liver function, hemoglobin profiling, and FACS data were plotted for statistical comparison of mean ranks using Kruskal-Wallis, Dunn’s multiple comparisons, or Spearman’s rank-order correlation tests. Data analysis was performed in statistical software GraphPad Prism v.6.0 and open-source statistical programming language R v.4.0.2, using packages ‘Hmisc’ and ‘corrplot’ for correlation analysis and visualization of results. Significance levels were considered when p<0.05 in which ‘*’, ‘**’, ‘***’ and ‘****’ indicate p<0.05; p<0.01, p<0.001, and p<0.0001, respectively. Results: Subjects under HC had overall higher median Hb, HCt, MCV, MCH, HBS, and HBF while CT groups presented higher RDW. The number of nucleated RBCs (nRBC) was similar among groups and the ratio of RBCHb/RetHb increased under HC treatment. Also, levels of plasma-free hemoglobin (calculated based on total and cellular Hb) were lower in HC. Reticulocytosis was higher in CT and hemolysis parameters appeared discretely lower in HC (Table 1). FACS analysis revealed that patients had higher circulating CD34+ cell counts. Levels of CD235+ cells inside the PBMC compartment were higher in HC patients. Also, CD34+CD235a+ population counts were significantly higher in HC than in CT and CT/HC, representing more than 15% of the circulating CD34+ compartment in some patients. Up to 60% of the CD34+CD235a+ were also CD36+ (Figure 1 A). In contrast, CD34+CD123+ were increased among CT and CT/HC groups. CD235a+ and CD34+CD235a+ populations positively correlated with HbF levels in the HC group, whilst its dose did not associate with any blood count or hemoglobin parameters (Figure 1 B). Interestingly, the same population counts positively correlated with age among non-HC takers. Conclusion: These results reaffirm the capacity of HC treatment to improve hematological parameters in SCA. Notwithstanding, HC presented increased levels of SE cell populations than the CT groups, reflecting different mechanisms of SE onset between CT and HC treated patients. Financial Support: CAPES, CTC/FAPESP-(2013/08135-2); INCTC-(465539/2014-9).Complete blood count and hemolysis biomarkers of HC and CT / HC groups compared with CT, calculated using Kruskal-Wallis test. ‘*’, ‘**’, ‘***’ and ‘****’ indicate p?0.05; p?0.01, p?0.001, and p?0.0001, respectively.A) Levels of CD34+ and CD34+CD235a+ cell populations. B) Correlogram of stress erythropoiesis cell FACS data versus hematological parameters of hydroxyurea treatment groups. Levels of stress cells are positively correlated with levels of HbF. The authors do not declare any conflict of interest
Read full abstract