Paroxysmal nocturnal hemoglobinuria: Unraveling epidemiological patterns, molecular mechanisms, and biotechnological advances
- Autores: Vanda Pinto Peixoto
- Directores de la Tesis: Mónica Almeida Vieira (dir. tes.), Sérgio Filipe Sousa (dir. tes.)
- Lectura: En la Universidade de Vigo ( España ) en 2024
- Idioma: inglés
- Tribunal Calificador de la Tesis: Joao Costa Rodrigues (presid.), Irina Moreira (secret.), Diana Valverde Pérez (voc.)
- Programa de doctorado: Programa de Doctorado en Biotecnología Avanzada por la Universidad de A Coruña y la Universidad de Vigo
- Materias:
- Enlaces
- Tesis en acceso abierto en: Investigo
- Resumen
Paroxysmal Nocturnal Hemoglobinuria (PNH) is a rare and clonal disease that causes debilitating symptoms and is highly lethal if not appropriately treated. It is an acquired, X-linked disorder caused by a somatic mutation in the glycosylphosphatidylinositol (PIG-A) gene, which encodes the A subunit of phosphatidylinositol N-acetylglucosaminyltransferase (GPI-GnT). A single mutation is sufficient to cause the disease. The PIG-A gene loses its function, leading to the absence of glycosylphosphatidylinositol (GPI) anchors, an essential structure that attaches complement regulatory proteins to the membranes of hematopoietic cells.
Proteins like CD55 and CD59 are examples of proteins that no longer remain attached to the cell membrane of red blood cells (RBCs). CD55 regulates complement activation by inhibiting the formation of C3 convertase, an essential enzyme in the complement cascade. Inhibiting this step prevents inappropriate complement activation on the surface of erythrocytes. CD59 acts in preventing the formation of the membrane attack complex (MAC), which is the final stage of complement activation. MAC is a structure that pierces and destroys the cell membrane. The presence of CD59 prevents the incorporation of C9 into the cell membrane, thus avoiding MAC formation and protecting the cell from lysis.
Uncontrolled complement system activation, due to the absence of regulatory proteins, leads to an attack on these cells, resulting in their lysis and triggering hemolytic anemia. Laboratory and clinical analysis of a patient with PNH may reveal DAT-negative intravascular hemolysis (IVH); anemia; elevated reticulocytes, lactate dehydrogenase (LDH), and bilirubin; decreased haptoglobin; and hemoglobinuria. The occurrence of thrombosis in unusual locations is also common. Diagnosis confirmation is typically performed through flow cytometry.
The absence of CD59 on the surface of the RBCs is the main cause of the clinical manifestations of PNH due to the lack of MAC blockade. IVH leads to the release of free hemoglobin (Hb) into the blood, which in turn can result in various effects associated with the reduction of nitric oxide (NO), such as alterations in vascular tone, increased smooth muscle contraction, and contribution to the activation of platelets and monocytes.
Extravascular hemolysis (EVH) also occurs in patients with PNH, due to C3 fragments that are not destroyed by the MAC and can accumulate around negative-CD55 RBCs, especially, but not only, after treatment with C5 inhibitors. These fragments can lead to opsonization of RBCs, causing destruction through the reticuloendothelial system (RES) in the liver and spleen.
The mechanism leading to mutations in the PIG-A gene is unknown, but there is a belief that there may be an association with some degree of aplastic anemia (AA), suggesting some immune attack on hematopoietic stem cells (HSC), conferring a survival advantage to PNH cells. Over the years, attempts have also been made to associate other mutations beyond the PIG-A mutation that may provide a connection conferring a clonal escape advantage to the mutated PIG-A.
The classification of the disease can then encompass three categories: (i) classic PNH, where clinical and laboratory findings include IVH without evidence of any bone marrow deficiency; (ii) PNH associated with some type of bone marrow disorder, such as AA or myelodysplastic syndrome (MDS); (iii) and subclinical PNH, in which patients have a small population of PNH cells without clinical or laboratory evidence of hemolysis or thrombosis, often associated with patients with bone marrow failure (BMF).
Generally, the increase in hemolytic states and the size of the PNH clone are related to the severity of the disease. Symptoms associated with PNH can be diverse, and the manifestation of the disease is quite heterogeneous. Some symptoms include fatigue, headaches, shortness of breath, abdominal pain, chest pain, erectile dysfunction, and dysphagia. Many of these symptoms are related to the fact that free Hb captures NO, leading to conditions such as esophageal spasms, renal failure, thrombosis, and pulmonary hypertension (PH). All these manifestations can be debilitating, making the journey of a patient with PNH exhausting and causing a significant impact in the quality of life (QoL).
The exacerbation of hemolysis can occur because of events such as infections, surgeries, and even transfusions. The major risk associated with the disease is the possibility of occurrence of serious thrombotic events.
Once the pathophysiology of the disease is understood, a question that remains is whether complement activation products or cytolysis itself, are the main culprits for thrombotic complications in PNH. The current understanding of the complement system goes beyond its initial role in combating infections. Currently, its functions include involvement in maintaining balance of the body, supporting tissue regeneration, and contributing to the pathophysiology of various diseases. The complement and coagulation systems are intricately interconnected and play essential roles in preserving homeostasis. The imbalance in hemostasis caused by complement system activation may be closely linked to imbalances in the coagulation system. The existing crosstalk between the two systems may eventually provide a clear answer to the high thrombotic propensity in PNH.
Initially, the management of this pathology was carried out through supportive therapies: RBC transfusions, the use of corticosteroids, supplementation with folic acid and, eventually, iron, as well as anticoagulant therapy. This type of supportive or traditional treatment was, for many years, the only available treatment for these patients, often proving ineffective, leading to approximately 40% of patients developing thromboembolic episodes, one of the leading causes of death in the disease. The overall survival of 10 years after diagnosis was established at 70% with only supportive therapies. The diagnosis of thrombosis in patients unaware of being PNH-positive was associated with a relatively low survival rate of 40%.
In 2007, the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) approved the first targeted treatment for PNH, eculizumab, a monoclonal antibody (mAb) which inhibits the C5 complement protein. Eculizumab became a revolution and a change in perspective for these patients. Meanwhile, numerous studies on anti-complement therapies emerged, targeting various molecules that led to improvements in Hb values and a drastic reduction in mortality and morbidity.
The last two decades, we have witnessed exponential growth in data volume and the introduction of new and advanced bioinformatic tools. This is a dynamic and ever-evolving field, and the availability of web-based public platforms, plays a crucial role in storing and sharing these updated data. By applying these tools, we can effectively address specific scientific questions. By combining theory and practice, keeping pace with the rapid development and continuous updates in bioinformatics, we were able to create a robust research project capable of providing some contributions regarding PNH.
The Chapter I of this dissertation explored the previous knowledge about PNH, with a general introduction of the disease and introduced the bioinformatic studies of this thesis.
The prospect of a potential cure for PNH is a fascinating topic because we know that a single mutation in a pair of bases of the PIG-A gene can generate the PNH clone. Since we live in an era of personalized genomics and considering that this type of pathology originates from a genetic mutation, prioritizing single-nucleotide genetic variants (SNVs) has become essential.
In Chapter II of this dissertation, we applied bioinformatic studies where we analyzed some of the known SNVs, cataloged in databases such as OMIM and UNIPROT, related to the PIG-A gene and PNH. To perform these analyses, different bioinformatic software, based on sequences and structures, was used. The MutationTaster, SIFT, PROVEAN, Align-GVGD, and PolyPhen-2 programs make predictions and assess the level of pathogenicity related to PIG-A, i.e., the functional impact of mutations on proteins originated from the mutated gene. Meanwhile, structure-based predictors like I-Mutant 2.0 evaluate the protein structure for specific mechanistic changes, examining the mutation's effect on the stability of the new protein.
9 SNVs were analyzed, with MutationTaster predicting 2 benign mutations and 7 deleterious mutations. The SIFT software predicted 2 neutral mutations and 7 deleterious mutations. PROVEAN predicted 2 mutations as neutral and 7 mutations as deleterious. Align-GVGD predicted 8 mutations as more deleterious and 1 mutation as less likely to interfere with protein function. PolyPhen-2 predicted 7 mutations as probably damaging and 2 mutations were predicted as benign. For I-Mutant2.0, all G calculations were below 0, predicting that the 9 mutations cause a decrease in protein stability.
This in silico method concluded that SNVs have the potential to cause relevant changes in protein conformation and structure. The proposed theoretical and computational approach allowed a better understanding of the atomic system of functional SNVs associated with PNH.
This type of analysis may become more intricate in the future, attempting to interconnect various types of genes that may eventually be connected to PNH clonal expansion or even to the propensity for thrombotic events and/or bone marrow dysfunctions. This task will be increasingly facilitated by the creation of servers that can automate the process of these multiple information.
Case studies are a unique documentation method, resulting from clinical observation that can provide valuable information regarding diagnosis, prophylaxis, and treatment, especially when related to rare diseases such as PNH.
The significance of the case study conducted in Chapter III is directly related to the ability to analyze a specific clinical situation given the high individual variability of this pathology in terms of symptoms, clinical and biochemical data, to correlate analytical results, treatments, disease severity, and progression with the patient's QoL.
One of the relevant aspects of the analysis of this case study is the connection between a potential trigger for the development of the disease, which may be related, for example, to an infection affecting the immune system, leading to an escape of the previously existing clone. In the case of the patient under study, it is quite interesting that the onset and exacerbation of symptoms occurred shortly after a recent cytomegalovirus (CMV) infection, making it possible to inquire about the causal relationship regarding the clonal expansion that may have resulted from it.
Another important aspect to note is that the vaccination schedule carried out prior to the treatment with eculizumab may have been a potential enhancer of a vaccine-mediated thrombosis. To initiate treatment with eculizumab, a vaccination protocol covering various infectious diseases, including those providing protection against Neisseria meningitidis, must be started. Although the lack of literature linking vaccines and thrombotic phenomena, it is impossible not to make this causal connection, as acute symptoms occurred shortly after vaccination against meningococcal disease, and a thrombosis of the splenic vein was diagnosed. Some relevant studies associating vaccine-induced thrombosis have begun to emerge due to the high number of thrombotic events associated with vaccination for COVID-19. However, there is not much scientific evidence published associating these events with meningococcal vaccines or even vaccines of another type and the propensity for blood clot formation, particularly in PNH patients.
Regarding the location of thrombosis, we can try to infer a justification through the pathways of elimination of lysed RBC products. Hb released due to IVH is filtered by the kidneys, resulting in hemoglobinuria, and can also be captured by haptoglobin in the plasma, forming complexes that are removed by the reticuloendothelial system (RES), especially in the liver. EVH mainly occurs in the spleen and liver. Broken RBCs are phagocytosed by macrophages present in these organs. The predilection of thromboses for the abdominal area and complaints of abdominal pain may be related to the significant increase in inflammatory mediators, along with extensive destruction of RBCs in this body region, which can create a pro-inflammatory and pro-coagulant imbalance, creating a conducive environment for thrombotic complications.
The subsequent introduction of eculizumab treatment significantly improved the patient's symptoms, and along with anticoagulant therapy, managed to resolve the thrombus, proving to be an effective prophylactic measure for thrombotic events in PNH. Simultaneously, it allowed the patient to be free from blood transfusions, stabilized Hb levels, although they never increased above the threshold of 10 g/dL. The patient was subsequently infected with SARS-CoV-2, and the fact that he was undergoing treatment with eculizumab may have been helpful in preventing the progression of the characteristic inflammatory activity of the severe and unpredictable COVID-19, especially in more susceptible population.
The advent of eculizumab and anti-complement therapies has saved the lives of patients with the disease, reducing hemolysis, decreasing thrombosis rates, and improving symptoms, proving to be a true "elixir of life." The truth is that, despite being quite effective for most patients, this treatment does not always show complete efficacy.
As seen earlier, SNVs can influence how a gene is expressed or the function of the protein produced. In the context of C5 and eculizumab, the presence of certain genetic single nucleotide polymorphisms (SNPs) in the C5 gene can influence the patient's response to treatment, such as the mutations c.2654G>A and c.2653C>T. Individuals with PNH and carriers of these SNPs present the alteration of a single amino acid: p.Arg885His or p.Arg885Cys, and a weak therapeutic response to eculizumab has been demonstrated in individuals carrying this mutation.
The application of computational studies employing molecular dynamics (MD) in the assessment of genetic polymorphisms of the C5 gene can be relevant for studying the efficacy of therapies in this type of population. In this case, it aims to understand the variability in response to eculizumab and optimize treatment at an individual level.
The functional behavior of a protein is significantly influenced by its dynamic characteristics. MD simulations facilitate the analysis of intricate and dynamic processes that can occur in biological systems, providing detailed insights into the movement of particles over time. MD simulation applications include studies on protein stability, molecular recognition, conformational changes, and movement in a biological context.
Therefore, in Chapter IV, MD simulations were applied to understand the molecular basis associated with the impact of different complement C5 mutations on the therapeutic response to eculizumab: The structure of the C5-eculizumab complex in Homo sapiens corresponds to entry 5I5K in the Protein Data Bank (PDB) and corresponds to its Cryo-EM structure with a resolution of 4.2 Å. This structure was aligned with the complement C5 model available in the recently developed and revolutionary AlphaFold software. A complete model was built from the experimental structure 5I5K, with missing amino acid residues modeled from the AlphaFold structural database model. The protonation state of all amino acid residues at pH 7.0 was predicted using the PlayMolecule web server. System preparation involved the use of the AMBER20 software package and Xleap. The system was described using the ff14SB force field. Using the LEAP module of the AMBER software, charges in the complex system were neutralized by adding counter ions and the system was placed in a TIP3P water model box with a minimum distance of at least 12 Å between the protein surface and the box edge, treated with periodic boundary conditions. Three-dimensional (3D) structures were prepared with the Visual Molecular Dynamics (VMD) program.
Subsequently, 3D models of the PIG-A protein associated with different genetic polymorphisms were created. For the analysis of mutated variants (p.Arg885His and p.Arg885Cys), the wild-type (wt) complex was modeled using the mutagenesis functionality in the Pymol 1.7.2.1 software, incorporating Dunbrack rotamer libraries. Solvation and neutralization protocols were applied to both the wt and mutant systems, and the three systems underwent four consecutive stages of energy minimization to resolve conflicts before MD simulation. They were then subjected to a two-step equilibration protocol for MD.
MD simulations were performed with 3 replicas of 300 ns each for the wt complex and the 2 mutants, with periodic boundary conditions and a 2.0 fs integration step, using the SHAKE algorithm to restrict all covalent bonds involving hydrogen atoms. A 10 Å cutoff for non-bonded interactions was maintained throughout the molecular simulation procedure. Coordinates were saved every 10 ps. The final trajectories were analyzed in terms of root-mean-square deviation (RMSd) and root-mean-square fluctuation (RMSF).
The MM-GBSA method was employed to predict the binding free energy of the C5-eculizumab complex. The contribution of each amino acid residue to these binding free energies was estimated by applying the energy decomposition method. From each MD simulation trajectory, a total of 300 conformations taken from the last 300 ns of the simulation were considered for each MM-GBSA calculation.
Regarding the obtained results, it was observed that the amino acids contributing most to the formation and stability of the C5-eculizumab complex are Arg885, Trp917, Gln854, Phe918, and Glu915. These residues have a free energy contribution ranging from -1.7 ± 1.1 to -8.8 ± 0.9 kcal/mol for the complex binding free energy. Arg885 is the largest contributor to the complexation free energy (-8.8 ± 0.9 kcal/mol), followed by Trp917 (-7.4 ± 0.8 kcal/mol). These results confirm these two amino acid residues as the main participants in the interaction of eculizumab with C5. Gln854, Glu915, and Phe918 have a contribution to the complex free energy of -3.5 ± 1.5, -1.7 ± 1.1, and -1.8 ± 0.3 kcal/mol, respectively.
In the p.Arg885His variant, the change from arginine to histidine reduces the binding affinity of eculizumab, resulting in an increase in Gbind of +17.4 kcal/mol, from -69.8 to -52.4 kcal/mol. For p.Arg885Cys, the difference in Gbind is also significant, with an increase of +9.4 kcal/mol, from -69.8 to -60.4 kcal/mol.
An analysis of the energetic components for the complexation free energy shows that both the nonpolar and polar contributions to the eculizumab binding free energy in the mutated variants represent an unfavorable energy change for the stability of the polymorphic variants.
A single mutation at position 885 does not induce a significant variation in the stability of the C5 protein (average C5 RMSd from 4.2 for wt to 4.4 Å for mutants). However, the analysis of eculizumab reveals a significantly more drastic change in the average RMSd (6.0 for wt to 7.9 and 6.7 Å for His and Cys variants, respectively). These results suggest that the studied SNPs interfere with C5's ability to stabilize eculizumab, which adopts a less stable bound conformation in the mutants.
Comparison with wt shows that polymorphic variants result in an increase in amino acid flexibility, as indicated by the average RMSF (3.6 for wt to 4.0 and 4.4 Å for His and Cys variants, respectively).
The mutation at position 885 causes significant conformational changes in the structure of C5 and drastic alterations in the interaction of the complex. Overall, all residues in C5 show higher flexibility when compared to wt. The C5-eculizumab interface, due to histidine/cysteine alterations, loses affinity.
Our analysis suggests that the lack of eculizumab efficacy in patients with PNH and Arg885His/Cys polymorphisms may be attributed to a disruption in the eculizumab blocking mechanism, resulting in an interruption of the conventional process. These findings offer the opportunity to develop modifications to eculizumab, making it less susceptible to changes in the C5 position 885. This may contribute to the creation of a more comprehensive mAb, capable of being effective in patients with various genetic variants in this position.
C5 polymorphisms are one cause of the weak response to eculizumab, but there are some other reasons, like bone marrow failure and/or opsonization of PNH erythrocytes coated with C3 due to deficient CD55 expression, causing persistent EVH during treatment. The poor response leads to the non-abolition of blood transfusions, for example.
The emergence of eculizumab resolved the potentially fatal aspects of PNH, reducing the thrombotic risk of the disease. The focus can now be directed towards finding new therapies that may respond to patients who did not achieve the expected response with eculizumab.
In Chapter V, an attempt was made to compile all the important data regarding clinical trials associated with the treatment of PNH.
The review carried out is a summary of clinical trial records for the treatment of PNH using the ClinicalTrials.gov database. Various approved treatments and new approaches under study are described, such as inhibitors of the terminal and/or proximal pathways of the complement system.
A relevant research interest focused on the proximal pathway of the complement, as EVH may persist even with terminal complement inhibition, using anti-C5 agents. In this case, inhibiting the proximal pathway can address both IVH and EVH, improving the patient's clinical situation. The risks associated with this global blockade need to be appropriately measured, as the complement system is a defense mechanism against pathogens that can cause severe infections.
Some new proximal complement inhibitors that specifically inhibit the alternative pathway (AP) at the level of Factor B (FB) or Factor D (FD) are currently at the forefront of research because they do not represent a blockade of the classical complement pathway (CP), which is crucial for the immune system. Moreover, these drugs are designed to prevent both IVH and EVH, can be administered orally as monotherapy, and demonstrate improvements in hematological biomarkers and patients QoL.
It is possible to conclude with great satisfaction that recent years have seen significant developments in clinical and biochemical research related to finding the most effective therapy for PNH with fewer associated side effects. This has led to the approval of the anti-C5 ravulizumab by the FDA and EMA. The age of approvals for proximal complement inhibitors, such as pegcetacoplan (C3 inhibitor), has also begun. More recently (December 2023) the innovative first oral monotherapy drug, iptacopan, which blocks the complement system's FB, was approved for the treatment of PNH.
The range of possibilities in therapeutic intervention is expanding, and it is essential to make these treatments available to patients. Despite technical and scientific developments, in many countries, alternatives remain limited. If the medications are genuinely available, the patient's lifestyle and opinion, along with all the specificities of the disease for each individual, should be considered. The exhaustive study of all mechanisms, side effects, modes of administration, and administration frequency can lead to a preference for one therapy over another. The complex and individualized nature of this decision in managing PNH requires open communication between the patient and the healthcare professional.
However, current anti-complement treatments do not offer a definitive cure. Future studies should focus on correlating the variability in the clinical picture, especially observing the pattern of expansion, regression, or stabilization of the PNH clone, with the specific mutational profile of each PNH patient and its relationship to changes in the bone marrow. This approach can provide valuable insights to guide more effective and personalized therapeutic strategies for individuals affected by PNH.
The development and enhancement of the studies conducted have allowed a scientific expansion in the state of the art of PNH, with an emphasis on the epidemiological, molecular, and biotechnological approach, relating genetic mutations, disease specificities, and therapeutic complexity.
