Sickle-cell disease (SCD) is a debilitating genetic disorder that affects millions of people worldwide. The disorder is the result of a point mutation on chromosome 11 in codon 6 of the beta-globin gene locus that changes a GAG codon to a GTG. This results in a single amino acid substitution from a hydrophilic glutamate to a hydrophobic valine. The resulting, mutated beta-globin protein (βS
) in the setting of hypoxia undergoes polymerization with alpha-globin proteins (α) to form aberrant hemoglobin polymers (HbS). These polymers retain different biochemical properties than their normal adult hemoglobin (HbA) counterparts that lead to a cascade of negative effects on the patient’s pathophysiology of tissue circulation and oxygenation [1
]. When deoxygenated, HbS polyermizes into long chains that deform the red cell’s shape into the disease’s characteristic sickle namesake. Once sickled, affected RBCs are prone to binding to each other and the endothelium, which in turn blocks blood flow and causes severe pain as a direct consequence of ischemic injury to the surrounding tissue and organs. This event is known as a vaso-occlusive crisis (VOC) and results in a systemic response in which the patient remains in a state of persistent, diffuse vascular inflammation [2
]. This systemic response further upregulates inflammatory cytokine signaling and promotes increased adhesion molecule expression on the endothelium, thus making the patient prone to additional VOCs. In addition, sickle RBCs have increased fragility and a decreased lifespan, making them prone to lysis. Increased lysis of RBCs causes a systemic increase in circulating free hemoglobin (Hb) and reactive oxygen species (ROS), both of which further promote inflammation and contribute to damage in multiple organs. Accordingly, sickle-cell patients enter a vicious cycle of VOCs and inflammation that drive further chronic vascular injury. Despite the severity of the disease and these symptoms, SCD has had very few treatment options to date.
Until recently, hydroxyurea represented the only pharmacologic intervention strategy to treat SCD. Initially approved in 1997 due to its ability to reduce VOC and transfusion frequency, it was later found to attack SCD pathology at multiple fronts through its ability to simultaneously increase fetal hemoglobin (HbF) and decrease neutrophils [4
]. Increases in circulating HbF content are known to ameliorate the SCD phenotype through the ability of the gamma-globin subunit to disrupt HbS polymer chaining when present, thus protecting against sickling and VOCs. No new pharmacologic agents were approved until 2017, when L-glutamine was approved due to its ability to protect against pro-inflammatory reactive oxygen species (ROS) and reduce yearly VOCs [5
]. Following the success of L-glutamine, crizanlizumab and voxelotor were both approved for use in SCD in the fall of 2019 [6
]. Crizanlizumab is a P-selectin inhibitor that acts to block sickle-cell adhesion to the endothelium, thus preventing the incidence of VOCs by 45%. Voxelotor, formerly called GBT440, acts a hemoglobin modulator that is able to directly bind to the alpha-globin chain of HbS polymers and allosterically stabilize the oxygenated state of the molecule, thus inhibiting HbS chain polymerization, reducing hemolysis, and increasing patient hemoglobin levels. More recently, hopes for the disease include genetic editing through clustered regularly interspaced short palindromic repeats (CRISPR)-based systems [10
]. While preliminary results are encouraging, the use of genetic editing in humans to treat diseases is still in its infancy and the long-term side-effects of such efforts remains to be seen. Additionally, the economic costs of recent therapeutic efforts can be a significant barrier for some (>$
2000 per month) [8
]. Furthermore, the economic challenge of bringing these new treatments to other nations of the world where SCD is more prevalent represents a significant barrier on its own.
At present there is still a great need for readily available, economically relevant SCD therapies that can be administered worldwide with acceptable toxicity.
Clinical research in sickle-cell anemia has focused on several areas including the pharmacological re-expression of a key hemoglobin polymer named fetal hemoglobin (HbF) [13
]. Fetal hemoglobin is a naturally occurring hemoglobin molecule that is present throughout fetal development. In most children, hemoglobin F decreases and is replaced with hemoglobin A during the first few months of life except in the presence of a rare benign asymptomatic genetic disorder, hereditary persistence of fetal hemoglobin (HPFH). The result of HPFH is an asymptomatic, modest elevation of HbF. Patients with SCD who possess the HPFH hereditary trait tend to have a higher quality of life and experience less frequent hospitalizations compared to HPFH negative SCD patients [15
]. This occurs due to the protective effects of HbF against HbS. Specifically, HbF blocks HbS from dimerizing, thereby preventing hemoglobin polymer chaining and decreasing the probability of a vaso-occlusion [1
]. Hydroxyurea, thalidomide, sodium butyrate, and decitabine, a DNMT1 targeting agent, have been previously shown to work through induction of HbF and subsequent protection against VOCs [5
Given the long-term success of hydroxyurea in treating SCD, we focused efforts on developing an in situ HbF intracellular immunoblot assay coupled to a high-throughput drug screen to identify drugs that can be repurposed for treatment of SCD. This is a powerful, yet previously under-utilized methodology to screen for modulation of HbF from a library of drugs that have already been FDA approved for other indications. Through this process, known as drug repurposing, we have identified existing candidate drugs for use in SCD, thus drastically reducing the cost of development for new drugs to market and allowing for fast track potential of old drugs for new indications [23
]. In this report, we present nine new drugs for repurposing in SCD that warrant future clinical investigation.
2. Experimental Section
2.1. Cell Culture
Cell lines K562 (isolated from 53 y.o. female, CML), THP-1 (isolated from 1 y.o. male, AML), and HEL92.1.7 (isolated from 30 y.o. male, erythroleukemia) suspension cell lines were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in Roswell Park Memorial Institute (RPMI) media (Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal bovine serum, 10,000 units of penicillin, 10 mg of streptomycin and 200 mM of glutamate. All cells were kept in a 37 °C, 5% CO2 incubator.
Blood was obtained from healthy volunteers and sickle cell donors following written informed consent under a protocol approved by the Institutional Review Board (IRB) of The Ohio State University (OSU; Columbus, OH, USA) (IRB Protocol Number: 1997C0194) in accordance with the Declaration of Helsinki. Blood was used from across 42 consenting sickle-cell donor volunteers to develop and perform the experiments used across this study. Peripheral blood mononuclear cells (PBMCs) were obtained using ficoll centrifugation. PBMCs were cultured under a consecutive two-phase culture system: (I) RPMI supplemented with 10% FBS, Stem Cell Factor (10 ng/uL, Sigma-Aldrich #H8416), Erythropoietin (1.5 U/uL, Sigma-Aldrich #H5166), Interleukin-3 (1 ng/uL, Sigma-Aldrich #H7166), and Dexamethasone (1 uM, Selleck Chem #S4028) for one week and (II) MethoCultTM Optimum (#4034) following standard colony plating assay conditions for an additional week.
2.2. In-Cell Western Blot
Cell lines (K562, THP-1, and HEL92.1.7) and in vitro differentiated RBCs were cultured in black 96-well or 384-well plates in RPMI media (Invitrogen, Grand Island, NY, USA) supplemented with 10% FBS, 10,000 units of penicillin, 10 mg of streptomycin and 200 mM of glutamate. All cells were kept in a 37 °C, 5% CO2 incubator. After incubation, plates were spun down (1500 rpm, 5 min) and supernatant was removed without disturbing the cell layer. Cells were then fixed in-plate using 3.7% formaldehyde for 20 min. After fixation, standard in-cell Western assays were performed as previously published by LI-COR Biosciences.
2.3. National Institute of Health (NIH) Clinical Collection (NCC) Screen
The NIH Clinical Collection (NCC) is a library containing 725 small molecules and inhibitors that was made public via Common Fund support as a part of the Molecular Libraries and Imaging program (http://commonfund.nih.gov/molecularlibraries/tools
). K562 and in vitro differentiated RBC cells were plated in black 384-well plates using a MultiDrop Combi Reagent Dispenser (ThermoFisher Scientific, Waltham, MA, USA, 5) at 5 e3
and 2.5 e4
cells per well respectively. The NCC library was dispensed across the plates by a BioMek Automated Workstation. All cells were kept in a 37 °C, 5% CO2
incubator. After incubation, plates were spun down (1500 rpm, 5 min) and supernatant was removed without disturbing the cell layer. Cells were then fixed in-plate using 3.7% formaldehyde for 20 min. After fixation, standard in-cell Western assays were performed as previously published by LICOR.
2.4. MTS Viability Assays
K562 and HEL92.1.7 cell lines were cultured in 96-well plates at 50,000 cells per well in RPMI media (Invitrogen, Grand Island, NY, USA) supplemented with 10% FBS, 10,000 units of penicillin, 10 mg of streptomycin and 200 mM of glutamate. After incubation, a solution of MTS tetrazolium (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium)) Promega, G1111) and phenazine methosulfate (Sigma-Aldrich, St. Louis, MO, P9625) was added to all plate wells. Plates were then incubated for an additional 3 h and then measured via plate reader at 490 nm absorbance.
Cells were lysed in standard lysis buffer (10 mM Tris pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% deoxycholic acid, 0.1% SDS, 5 mM ethylenediamine tetraacetic acid supplemented with phosphatase inhibitor cocktail II (Sigma-Aldrich, Madison, WI, USA, P2850), phosphatase inhibitor cocktail III (Sigma-Aldrich, P5726), protease inhibitor cocktail Σ (Sigma-Aldrich, P-8340), and phenylmethyl sulfonyl fluoride (Sigma-Aldrich, P-7626). Protein concentrations were determined using bicinchoninic acid assay (BCA). Lysates were run on sodium dodecyl sulfate-polyacrylamide gels, transferred onto nitrocellulose membranes and incubated with antibody. Membranes were developed using SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermofisher Scientific, 34580).
2.6. Circular Chromosome Conformation Capture (4C)
K562 and HEL92.1.7 cell lines were grown in culture under treatment for 72 h and lysed for DNA collection. 4C libraries were generated using previously published procedures and used DpnII and Csp6I for subsequent RE digestions. 4C primers were designed for the HBG2 promoter region (FW: CAAAGCACCTGGATGATC, RV: TTGTCTCTAGCTCCAGTGAG). Libraries were sequenced on an Illumina MiSeq system and analyzed using 4Cseq analysis pipeline.
2.7. Statistical Analysis
Data are shown as the mean ± standard error of the mean (SEM). The significance of difference in antibody ratios was evaluated by two-tailed, unpaired Student’s t-test. All statistical procedures were performed using commercial software (GraphPad Prism, version 8.4.2, GraphPad Software Inc., San Diego, CA, USA).
The present study is one of the first to screen a large library of FDA-approved drugs for repurposing to SCD. In our technical testing, we showed the previously underutilized in-cell Western to be a powerful and sensitive technique for screening for modulators of fetal hemoglobin. Sequential, large-scale screens on K562 cells and in vitro differentiated red blood cells revealed an overlap of 9 drugs that can positively modulate HbF expression. Importantly, all nine drug hits have FDA approval for various indications and have promise for repurposing therapeutically toward SCD. Furthermore, individual validation experiments confirmed the ability of these drugs to increase HbF expression with little to no effects on cytotoxicity. Of the drugs validated in this study, imatinib, pyrimethamine, and quinidine show the strongest effects on HbF expression. Additional pre-clinical studies are underway in our laboratory to further characterize these drugs for use in SCD.
The in-cell western methodology used in this study remains a powerful yet underutilized tool for large-scale studies. We developed this assay to detect intracellular HbF and then combined it with a large-scale drug screen utilizing the NIH Clinical Collection (NCC), a plated array of small molecules with a history of use in humans and prior FDA approval. While this collection was originally assembled as part of the Molecular Libraries and Imaging Initiative to promote large screen experiments, it has previously never been used in the field of SCD. With the recent push for new therapeutic options to treat SCD, high powered detection assays such as ours are invaluable in finding new drugs. Furthermore, one main advantage our assay retains over others is the ease of translation to other libraries. While our work focused solely on the NCC, other drug libraries and small molecule collections can just as easily be plated and tested for positive modulators of HbF. Lastly, it is important to note that the K562 cell line used in our initial screen can be chemically stimulated towards erythroid differentiation where accumulation of HbF naturally occurs. Accordingly, future experiments using this methodology should utilize primary cells for increased reliability and analysis power. While genetic editing and other treatments are in development for SCD, our methodology represents a new way to quickly identify new or existing pharmacologic agents that can be quickly fast-tracked to fight the disease.
In our experiments we identified nine drugs as positive modulators of HbF. Of these drug candidates, we showed pyrimethamine, mafenide acetate, and quinidine hydrochloride to have strong effects on HbF expression in our model cell lines. Interestingly, all three of these drugs have historical use in controlling microbial infection. In addition to these screen hits, we investigated and identified imatinib (Gleevec) as an additional, potent modulator of HbF expression and under investigated therapeutic for SCD [34
]. Our data shows imatinib to exert strong increases in HbF expression in BCR-ABL positive K562 and BCR-ABL negative HEL92.1.7 cell lines. While originally designed as a BCR-ABL specific targeted therapy, it is known to inhibit additional receptor tyrosine kinases, namely platelet-derived growth factor receptor (PDGFR), c-Kit, and c-Abl [36
]. Previous work in a Kit-activating mouse model has shown a possible link between continuous Kit activation and partial blockage of erythroid differentiation [31
]. Furthermore, our 4C data demonstrate shifts in frequency and location of HBG2 promoter cis
binding interactions concurrent with an increase in expression of HbF. We hypothesize that imatinib may be inhibiting the c-Kit tyrosine kinase and allowing terminal erythroid differentiation signaling to occur in our model cell lines. Further experimentation is needed, however, to characterize the link between imatinib and c-Kit signaling in these cell models.
Historically there have been very few therapeutic options to treat SCD. Hydroxyurea remained the sole pharmacological agent for almost 20 years before the recent global push for new therapeutics. Given this global effort to find new treatments, we performed a large-scale screen of the NCC to identify new candidates for fast-track use in SCD. Our study is the first to utilize this library and present nine new drugs for consideration for the disease. Furthermore, we suggest pyrimethamine and imatinib as strong candidates for repurposing due to their strong effects on HbF expression, low cytotoxicity, and safe history of use in humans. With further testing and in vivo characterization, these drugs have real potential for combating SCD on a global scale.