Dataset selection
The GEO database search using the specified descriptors provided the identification of ten datasets; however, only one of them, GSE53441, satisfied the established inclusion criteria. Similarly, two data sets were also identified that examined variation in gene expression between healthy individuals and patients with SCA, but without fully emphasizing pro-inflammatory pathways (GSE35007 and GSE72999).
GSE53441 was based on individuals diagnosed with SCA with 18 years of age or older and have undergone chromatographic diagnosis. The study only included outpatients who were in good health. Subsequent evaluations were performed on patients who had experienced a VOC in the previous two weeks or an incident of acute chest syndrome in the previous four weeks. The objective of this study was to examine the role of the pro-inflammatory profile in SCA by evaluating the transcriptome of peripheral blood mononuclear cells (PBMCs) in a cohort of 24 adult SCA patients and 10 individuals without the condition (healthy controls). This was achieved by isolating RNA from PBMCs and plasma, and subsequently hybridizing them onto Affymetrix' Human Genome U133 Plus 2.0 microarray in a single batch (Van Beers et al. 2015).
Dataset validation
The selected data collection was validated through the analysis of intra-group repeatability under the development of Pearson’s correlation analysis, which was represented by a heat map (Fig. S1). A strong positive correlation between the expression data of dataset’s samples were found. On the other hand, based on the PCA, the intra-group data repeatability was adequate; the distances among samples in the control and SCA groups were close in the dimension of the PCA1, except for five samples in the SCA group, which were close in the PCA2 (Fig. S1b).
Identification of differentially expressed genes (DEGs)
DEGs were identified by comparing gene expression patterns in samples from the control group (healthy subjects) with those of SCA patients through the interactive online program GEO2R. The findings are shown in Fig. 1A. Each node of the volcano plot (dot) represents a potential DEGs, which were assessed and classified as significant if they fulfilled the threshold criteria [P 0.05, Log2 (fold change) values > 1.5 and − 1.5, respectively]. The volcano plot depicted up-regulated DEGs in red, whereas down-regulated DEGs were shown in blue.
A total of 411 DEGs were identified based on log2 fold change values; 195 were down-regulated and 216 were up-regulated (P < 0.05) (supplementary Table S1 – Fig. 1). The heat map in Fig. 1B illustrates the log2 fold change values from the identified DEGs. Based on the findings of the functional enrichment analysis in DAVID database, DEGs exhibited a notable enrichment in biological processes associated with the type I interferon signaling pathway and viral response; furthermore, protein tyrosine kinase activity, oxygen binding, and oxygen carrier activity were their principal molecular functions (Fig. 1C).
Hub genes identification, validation, and analysis
The hub genes were identified from the selected DEGs (down- and up-regulated) by constructing a PPI-Net in the STRING online database; as a result, a network composed of 374 nodes and 809 interactions was obtained (Fig. 2A). Next, the major clusters (highly linked areas) that comprised the PPI-net were discovered using Cytoscape’s Molecular Complex Complement Detection Tool (MCODE; version 1.5.1). The MCODE analysis revealed nine clusters, selecting the one with the highest score (24.333) as the most important in the PPI-net, and its 25 nodes were the hub genes, which had scores greater than 10 (Fig. 2B and C) (Xu et al. 2019).
Identified hub genes were validated through CytoHubba (Cytoscape complement) applying the five most reported computational methods in the scientific literature: Degree (Deg), Maximum Neighborhood Component (MNC), Maximal Clique Centrality (MCC), EcCentricity (EC) and Closeness (Liu et al. 2018). The supplementary Table S1 lists the scores of the top 10 hub genes identified by the algorithms of the evaluated methods. Next, we performed an intersection analysis of these genes on the website: http://www.ehbio.com/ImageGP/index.php/Home/Index/VennDiagram.html, in order to find the central hub genes (CHGs), which are shown in Fig. 2D-F and described in Table 1. These included: (i) Mx1 interferon-induced GTP-binding protein (MX1), (ii) Interferon-induced protein with tetratricopeptide repeats 1 and 3 (IFIT1, IFIT3), (iii) Signal transducer and activator of transcription 1 (STAT1), (iv) ISG15 ubiquitin-like protein (ISG15), (v) Guanylate-binding protein 1 (GBP1), (vi) 2'-5'-oligoadenylate synthase 1, 2, and 3 (OAS1, OAS2, OAS3, respectively), and (vii) S-adenosyl methionine radical domain-containing protein 2 (RSAD2).
Table 1
Features of central hub genes identified in the cluster 1 by each Cytohubba algorithm
Hub genes | Log2FC | Uniprot accession code |
Symbol | Name |
MX1 | Interferon-induced GTP-binding protein Mx1 | 0.187 | P20591 |
IFIT3 | Interferon-induced protein with tetratricopeptide repeats 3 | 0.273 | O14879 |
IFIT1 | Interferon-induced protein with tetratricopeptide repeats 1 | 0.296 | P09914 |
STAT1 | Signal transducer and activator of transcription 1-alpha/beta | 0.079 | P42224 |
ISG15 | Ubiquitin-like protein ISG15 | 0.281 | P05161 |
OAS3 | 2'-5'-oligoadenylate synthase 3 | 0.184 | Q9Y6K5 |
OAS2 | 2'-5'-oligoadenylate synthase 2 | 0.134 | P29728 |
OAS1 | 2'-5'-oligoadenylate synthase 1 | 0.111 | Q05961 |
GBP1 | Guanylate-binding protein 1 | 0.126 | P32455 |
RSAD2 | S-adenosylmethionine-dependent nucleotide dehydratase RSAD2 | 0.347 | Q8WXG1 |
All CHGs were up regulated in patients with SCA, according to the differential expression analysis performed in GEO2R (supplementary Fig. S2). Furthermore, ROC curves (acronym for Receiver Operating Characteristic) were performed to evaluate the magnitude of the variation of their expression between other datasets such as GSE35007, to validate their capability to sensitively predict the diagnosis and/or inflammatory complications in patients with SCA. The results showed a significant association between the variables evaluated (0.554 < AUC < 1; P ≤ 0.05) (supplementary Fig. S3).
A gene ontology analysis of CHGs through text mining in the PubMed search engine was carried out; results evidenced that all were closely related to the type I interferon signaling pathway (IFN-α, IFN-β), which agrees with the functional enrichment analysis performed in DAVID (Fig. 2); similarly, a high degree of co-expression of these genes was found in the ProteomeHD database (https://www.proteomehd.net/proteomehd) (Supplementary Fig. S4).
Type I interferons (IFN-I), such as IFN-α, IFN-β, IFN-ε, IFN-k, and IFN-ω, are polypeptide members of innate immunity that are released in response to different autoimmune disorders and viral infections in humans (Hervas-Stubbs et al. 2011; Peng et al. 2015). All IFN-I mediate their physiological effects through the interferon alpha receptor (IFNAR), which has the capacity to activate the Janus kinase/signal transducer and activator of transcription (JAK/STAT) signaling pathway (Seif et al. 2017), well recognized as one of the primary communication nodes involved in cell function. (Hu et al. 2021)
In summary, the IFNAR is constitutively associated with tyrosine kinase 2 (TYK2) and JAK1, proteins that are trans-phosphorylated when said receptor binds to its ligands; this in turn activates them to phosphorylate tyrosine residues conserved in the cytoplasmic tails of the IFNAR, which serve as docking sites for the recruitment of signaling molecules with the SH2 domain (Src Homology 2), such as STAT1 and STAT2, which once phosphorylated dimerize and move to the nucleus, where they associate with the factor interferon regulator (IRF9), thus forming the interferon-stimulated gene factor 3 (ISGF3), which binds to the IFN-I response elements, positively regulating the expression of genes such as ISGs (interferon-stimulated genes, ISG15); interferons themselves; interferon regulatory factors (IRFs); signal transducers and activators of transcription (STATs); oligoadenylate synthetase (OAS) enzyme genes; guanylate binding proteins (GBPs); nitric oxide synthase 2 (NOS2); interferon-induced transmembrane proteins (IFITMs), tripartite motif proteins (TRIMs); interferon-induced GTP-binding protein Mx1 (MX1), among others (supplementary Fig. S5) (Hervas-Stubbs et al. 2011; Peng et al. 2015; Mesev et al. 2019; Zanin et al. 2021).
All CHGs found in this research have a strong relationship with genes involved in the JAK/STAT pathway and were concordant with those obtained by Hermand-Tournamille et al. (2018), who performed a proteomic analysis by label-free mass spectrometry on polymorphonuclear neutrophils (PMNs) from four steady-state SCA patients and four healthy subjects. In this research, 50 up-expressed genes were found in patients with the pathology, including the STAT family (STAT1 and STAT2), OAS 1, 2, and 3, and many INF-I signaling genes (IFIT1, IFIT2, IFIT3, ISG15, ISG20, GBP2, IFI35, MX1, and MX2), whose concentrations in PMNs from SCA patients were on average 10 to 100 times higher than in healthy subjects (Hermand-Tournamille et al. 2018).
Several studies have shown that individuals with SCA in steady-state have increased levels of IFN-I in their plasma as well as high expression of its inducible genes (Hounkpe et al. 2015; Hermand-Tournamille et al. 2018; Hermand et al. 2020). Liu et al. (2021) report that in the SCA is generated a significant activation of classical monocytes (CMo) linked to intravascular hemolysis processes. In fact, monocytosis is frequent in SCA and is associated with hemolysis indicators and adversely with hemoglobin levels. (Allali et al. 2020) The total plasma heme levels and circulating IFN- α in SCA patients were discovered to be positively correlated by these researchers. Additionally, they observed an elevation of IFN-I-inducible genes in genes in sort-purified SCA patients’ CMo by transcriptome analysis. (Liu et al. 2021)
IFNs promotes chemokine synthesis, including chemokine ligand 2 (CCL2), also known as monocyte chemoattractant protein, which leads to the migration of bone marrow-derived classical monocytes into the circulation, exacerbating the inflammatory state in patients with SCA. (Lee et al. 2009; Lehmann et al. 2016; She et al. 2022) Furthermore, CMo from SCA patients display an activated profile, with increased expression of CD11b on their surface and increased production of proinflammatory cytokines including interleukin (IL)-1β and TNF-α as compared with healthy control monocytes (Allali et al. 2020), increasing the inflammatory status, and with it, the aggravation of the hemolytic process in the patient with SCA.
The mononuclear phagocyte system, according to Liu et al. (2023), is critical to SCA pathogenesis. They investigated the effect of hemolysis on circulating monocyte trajectories in SCA mice and discovered that hemolysis increased CSF-1 synthesis, which was mediated in part by endothelial cells via Nrf2, enhancing CMo differentiation into blood patrolling monocytes (PMo) in SCA mice. Hemolysis, on the other hand, increased CCL-2 expression via IFN-I, promoting CMo transmigration and differentiation into tissue monocyte-derived macrophages (Liu et al. 2023).
PMNs, on the other hand, have also been recognized as significant participants in the pathophysiology of SCA; nevertheless, the signaling mechanisms driving their activation and encouragement of a persistent inflammatory state remain unknown. In steady-state SCA patients, the absolute neutrophil count is greater than in ethnicity-matched healthy controls and positively linked with SCA severity (Allali et al. 2020). Activated PMNs play an important role in the initiation and propagation of VOC in patients with SCA and different investigations have shown that IFN-I is able to regulate the differentiation and production of cytokines in these cells (Hermand-Tournamille et al. 2018).
Decker et al., (2011) demonstrated that IFN-α can increase the recruitment of PMNs and the differentiation of monocytes into dendritic cells, which present antigens from dying cells to CD4+ lymphocytes, resulting in their proliferation (Decker 2011). On the other hand, Zimmermann et al. (2016) evidenced that IFN-I prolongs neutrophil survival through the expression of CXC motif chemokine ligand 10 (CXCL10) mRNA and can potentiate IL-6 production in neutrophils stimulated with R848 (Resquimod), a TLR8 agonist (Zimmermann et al. 2016). In congruence with the latter results, Glennon-Alty et al. (2021) measured the effects of IFN-α on the functions of healthy neutrophils incubated in vitro in the absence and presence of proinflammatory cytokines, such as TNF-α and granulocyte-macrophage colony-stimulating factor (GM-CSF), and found that IFN-α significantly increased CXCL10 expression, which was enhanced in the presence of TNF-α. These investigators identified the important role of IFN-α in the regulation of chemokine production by neutrophils (Glennon-Alty et al. 2021).
IFN-α was found to significantly decrease the expression of the chemokine genes CXCL1, CXCL2, CXCL3, CCL3, and CCL4 in response to GM-CSF and TNF-α, but enhanced the expression of CXCL10, which has important implications for neutrophil-mediated regulation of the immune response in vivo. The chemokines CXCL1, CXCL2, and CXCL3, and CCL3 and CCL4 behave as chemoattractant of neutrophils, while CXCL10 is mainly a chemoattractant of T cells, monocytes, NK cells, and dendritic cells (Glennon-Alty et al. 2021).
The above-described effects exerted by IFN-I on CMo and PMN could explain the effects of this cytokine on the inflammatory status and VOC events in patients with SCA. In fact, inappropriate neutrophil activation can induce tissue damage (Decker 2011), and being related to the generation of inflammatory diseases such as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE), where type I IFNs play an important role (Decker 2011; Zimmermann et al. 2016; Glennon-Alty et al. 2021).
According to what was described above, IFN-I is proposed as a new target for drug design, based on the inhibition and/or reduction of its immunomodulatory effects in patients with SCA, given the effects derived from IFN-I on the pathophysiology of SCA and the large number of identified CHGs associated with it signaling pathway. At present, pharmaceutical approaches for IFN-I pathway antagonism have relied on monoclonal antibodies such as anifrolumab, a RIFNAR1 antagonist (Peng et al. 2015), or rontalizumab, an anti-IFN- antibody utilized in lupus clinical trials (Maurer et al. 2015).
By analyzing identified CHGs, the signal transducer and activator of transcription 1 (STAT1) was chosen as a biological target for molecular docking assays, due to its central role as modulator of INF-I pathways.
Identification of polyphenols exhibiting therapeutic potential for SCA
More than 100 bioactive phenolic compounds that inhibit inflammatory processes were found by searching the databases of natural products under examination. Text mining in search engines (PubMed and Science Direct) was used to reduce these into 32 based on their anti-inflammatory activity in experimental models (in vitro and in vivo). Supplementary Table S2 lists the classification of all chosen phenolic compounds as flavonoids, including anthocyanins, flavones, isoflavones, and flavonols. A total of 15 flavonoids were selected for molecular docking investigations out of 32 (Supplementary Table S3), owing to their significant affinity towards biological targets that are implicated in chronic inflammatory processes. These targets comprise TNFα, IL1β, IL6, TLR4, CXCL8, and STAT 3 (unpublished study). To compare the binding properties of flavonoids with STAT1, a reference inhibitor (STOCK-1N-69677; ((2S,3S,4S,4S,5R)-3,4,5,6-tetrahydroxyoxane-2-carboxylic acid, or D-glucopyranuronic acid) was extracted from the literature, based on its high activity with STAT1 (Raj et al. 2016).
Evaluation of the promising activity of polyphenols against SCA
Molecular docking studies were conducted to explore the possible effects of flavonoids via their interaction with the active site of known STAT1 inhibitors. Once the flavonoids 3D structures were optimized, the CHG crystal (STAT1) was selected with PDB code 1YVL, which has been frequently documented in docking studies (Szelag et al. 2015; Raj et al. 2016). According to Raj et al. (2016), STAT1’s active site is made up of the amino acid residues between positions 573 and 670 that made up the SH2 domain (grid dimensions: x: -21.914 Å, y: 1.074 Å, z: 115.493 Å; size: 30 x 30 x 30) (Raj et al. 2016).
Docking studies were developed in triplicate with an exhaustiveness of 1000, determining nine conformations based on effectiveness value, free energy, and RMSD between conformations. In Table 1, the average binding energies are shown; these were in a range between − 6,7 and − 4,5 Kcal/mol for flavonoids, in contrast with the reference ligand (-5.7 K cal/mol).
All assessed flavonoids showed lowest binding energy in comparison to the reference ligand: STOCK-1N-69677. The flavonoids having the lowest binding energy with STAT1 were quercetin, diosmetin, and fisetin, indicating strong binding stability (Fig. 3 – Table 2). Hydrogen bonds with amino acids such as V461, E524, S560, and S583 were one of the main ways that these systems interacted together. π-alkyl interactions, on the other hand, were seen in flavonoids mostly with the amino acids P465, K566, V509, and L562. In addition, flavonoids exerted π-cation and π-anion interaction with R331 and E563, E559, E524 in STAT1, respectively. Finally, only rhamnetin and cyanidin had van der waals and amine - π -Stacked interaction, respectively.
Table 2
Result of molecular docking of flavonoids evaluated against STAT1.
Ligands | Binding energy (Kcal/mol)* | Interactions | | |
H-bond | π-alkyl | π-cation | π-anion | Van der Waals | Amine- π -Stacked |
Acacetin | -7,3 | S462 | P465, K566, V509, L562 | R331 | E563, E559 | - | - |
Cyanidin | -7,1 | V461, S560 | P465 | R559 |
Chrysin | -7,3 | - | L562, K566, V509, P465 | - |
Diosmetin | -7,5 | V461 |
Epicatechin | -7,2 | K566 | L562, P465 |
Fisetin | -7,4 | E524, F581 | R586, I582, M521 | - | E524 |
Formononetin | -7,1 | S583, E524 | I582, R586, M521, K525 | - |
Gallocatechin | -6,7 | Q497 | - | R331 | - |
Kaempferol | -7,3 | - | P465, K566, V509, L562 | E563, E559 |
Luteolin | -7,3 | V461 |
Morin | -7,2 | - | P465, K566, L562 |
Naringenin | -7,3 | - | P465, K566, V509, L562 |
Pelargonidin | -7,3 | V461 | P465 |
Quercetin | -7,5 | - | P465, K566, V509, L562 |
Rhamnetin | -6,9 | V461 | P465, K566, V509 | S560 |
Reference | -5,7 | Q3, Q41, W43 | - | - | - | - |
* Average value of three replicates |
Molecular dynamics assays were developed to validate the docking results. For it, we worked with the flavonoids that presented the best free energy (quercetin, diosmetin, and fisetin). The results obtained were compared with those of the reference ligand. Figure 4 shows the different trajectory analyses employed to assess stability (RMSD), fluctuations (RMSF), area of solvent accessibility (SASA), and compaction with the radius of gyration.
The RMSD values of the evaluated systems ranged from 2.5 to 15 Å in the first 20 ns, except for fisetin, which remained constant throughout the 100 ns of testing, with a value of 2.5 Å, which was the same for the rest of the systems after stabilization (native STAT, reference ligand, quercetin, and diosmetin), thus denoting adequate stability (Fig. 4a). Behavior remained similar in the studies where STAT3 was evaluated against quercetin, luteolin and wogonin, whose RMSD variations were minimal with stabilities in all systems throughout the simulation time and concluding that the molecules anchor to the protein. no notable alteration in the binding site (Li et al. 2023).
The RMSF results show that all complexes maintained a similar behavior according to the fluctuation of residues compared to the native protein, starting at residue 125. However, the STAT1-fisetin complex presents lower mobility and flexibility in the residues between positions. 1–125 with RMSF values between 2 and 4 Å (Fig. 4b).
The solvent accessible area (SASA) of each assessed complex is shown in Fig. 4c. The obtained SASA values ranged from 34000 to 39000 Å. They decreased throughout time and settled into a steady range after 30 ns of simulation. However, the degree of compaction of the complexes in relation to the native receptor is determined by their radius of gyration, as seen in Fig. 4d. The complexes including STAT1-diosmetin and STAT1-reference ligand showed values that were quite comparable to those of native STAT1, the rest of the complex exhibited highest values regarding STAT1 native, varying between 36 and 38 Å.
Free energy binding analysis was conducted utilizing 1755 complex frames from the trajectory in a 100 ns simulation, employing MMGBSA calculations. Table 3 reveals STAT1-quercetin, STAT1-diosmetin, STAT1-fisetin, and STAT1-reference ligand were − 30.1605, -20.9186, -28.6466, and − 7.7690 Kcal/mol, respectively. The free energies of the complexes between STAT1 and flavonoids show a greater energy affinity compared to the reference ligand of the protein, validating the results obtained by molecular docking.
Table 3
Binding free energy components (kcal/mol) of STAT1 against ligands and control using the GBSA method.
Energy Component | Quercetin | Diosmetin | Fisetin | Reference Ligand |
VDWAALS | -28.6940 | -24.3291 | -34.4278 | -9.9308 |
EEL | -100.7583 | -55.3592 | -15.1823 | -124.6605 |
EGB | 101.5923 | 61.8841 | 24.8570 | 129.0466 |
ESURF | -4.3005 | -3.1143 | -3.8935 | -2.2244 |
Δ Ggas | -129.4523 | -79.6883 | -49.6101 | -134.5912 |
Δ Gsolv | 97.2918 | 58.7698 | 20.9635 | 126.8222 |
Δ Total ± SD | -32.1605 | -20.9186 | -28.6466 | -7.7690 |
The anti-inflammatory properties of the flavonoids discussed in this article, as well as their possible therapeutic use in the treatment of SCA, are supported by many studies. The main mechanism associated with the protective effect of these flavonoids is the capture of free radicals and reactive oxygen species (Nur et al. 2011; do Nascimento et al. 2013).
Thangaswamy et al. (2021) evaluated the therapeutic benefits of quercetin in a SCD transgenic sickle mouse model (NY1DD) subjected to 18 h of hypoxia followed by 3 h of reoxygenation. The administration of this flavonoids in a dose of 200 mg/kg to NY1DD sickle mice, attenuates hypoxia-induced pathophysiology and brings parameters close to wild type control mice during reoxygenation (Thangaswamy et al. 2021).
On the other hand, using nuclear magnetic resonance and high-resolution mass spectroscopy, Adeniyi et al. (2022) showed that quercitrin (quercetin 3-rhamnoside), a glycoside derivative of quercetin is the active ingredient in Alchornea cordifolia Müll. Arg., also known as Christmas Bush, a plant traditionally used in Africa to treat SCA. In their study Methanol extracts of A.cordifolia leaves and its sub-fractions showed > 70% suppression of HbSS erythrocyte sickling in blood samples treated with sodium metabisulphite or through incubation in 100% N2. Moreover, at 0.4 mg/mL, the purified quercitrin showed 93.1 ± 2.69% erythrocyte sickling inhibition and 87.2 ± 2.39% substantial anti-sickling action, which prevented the polymerization of isolated HbS and stabilized sickle erythrocyte membranes. The comparison of blood samples metabolome by flow-infusion electrospray-high-resolution mass spectrometry showed that quercitrin could change the metabolomes of HbSS erythrocytes to resemble those of HbAA by pathways associated with the alteration of arachidonic acid metabolism, anaerobic bioenergy, and antioxidants process (Adeniyi et al. 2022).
Woodfordia fruticosa (L.) Kurz is another plant that is often used in traditional medicine to treat symptoms associated with SCA, particularly in the tribal populations of Amarkantak, Madhya Pradesh. S. Mishra et al. (2022) tested W. fruticosa extracts for phytoconstituents, antioxidants, anti-inflammatory, and anti-sickling activities. The study found that the plant's methanolic extract effectively reversed sickle red blood cells (66 ± 1%) and prevented Hb polymerization. The extract also inhibited lipoxygenase. Similarly, they demonstrated the presence of flavonoids such as kempferol and quercetin along with numerous of their derivatives such as quercetin 3-O-(600-galloyl)-D galactopyranoside, quercetin 3-O-(600-galloyl)-D galactopyranoside, quercetin, quercetin 3-O-D-galactoside, quercetin 3-L-arabinoside, quercetin 3-rhamnoside, quercetin 3-L-arabinoside, quercetin 3-rhamnoside (Mishra et al. 2022).
Muhammad et al. (2019) investigated quercetin’s anti-sickling effects on human sickle erythrocytes using in vitro models that altered deoxyhemoglobin and redox homeostasis. The study indicated that quercetin treatment efficiently avoided sickling at 5.0 µg/mL, reversed it at 7.5 µg/mL, and significantly protected erythrocyte membrane integrity (P < 0.05). The above-mentioned mechanisms are connected to quercetin's capacity to reduce lipid peroxidation and enhance GSH and CAT levels (P < 0.05) (Muhammad et al. 2019).
Balushi et al. (2019) found that antioxidant compounds like quercetin may improve sickle cell membrane permeability. Increased cation permeability in these cells promotes HbS polymerization by dehydrating RBCs and externalizing the prothrombotic aminophospholipid phosphatidylserine (PS). Quercetin inhibited three key cation routes involved in dehydration: deoxygenation-induced cation conductance, Ca2+ and K+ channel, and K+- Cl− cotransporter. It also decreased Ca2+-induced PS exposure and hemolysis (Balushi et al. 2019).
There are also reports in the literature of glycosides derived from diosmetin (another flavonoid with the lowest binding energy with STAT1) with anti-sickle properties like diosmin. Gwozdzinsk et al. (2023) discovered that diosmin slightly reduced total carbonyl levels and slightly increased the total non-enzymatic antioxidant capacity of red blood cells. In turn, diosmin caused a big rise in the amounts of total thiols and glutathione in these cells. Researchers looked at the rheological properties of red blood cells and found that diosmin slightly lowers the internal viscosity of red blood cells. This, along with raising the levels of glutathione and thiol compounds, helps protect red blood cells from oxidative stress and keeps the cell membrane stable, which may be helpful for people with SCA (Gwozdzinski et al. 2023).
Chrysin is yet another flavonoid that has garnered significant attention for its anti-sickling properties. The compound has been linked to many mechanisms, including the modification of deoxy-haemoglobin, 2,3-bisphosphoglycerate mutase, redox homeostasis, and functional chemistry in human sickle erythrocytes. Muhammad et al. (2020) report that chrysin can reverse sickling by 66.5% and 69.6% at 12.5 µg/mL and prevent sickling to the maximum extent possible at 2.5 µg/mL. The observed proportion of hemolysis in comparison to induced erythrocytes indicates that treatment with chrysin considerably restored the integrity of the erythrocyte membrane. Lipid peroxidation was also considerably inhibited and reversed by chyrsin (Muhammad et al. 2020).