EPSC-derived hepatocyte-like cells validate functional effects of SERPING1 mutations in hereditary angioedema
Background: Limitations of Conventional Single-Dimensional Screening and the Omic Data Bottleneck of SERPING1 Genetic Variants in Hereditary Angioedema R&D
Conventional single-dimensional mouse knockout models and 2D immortalized cell line screening guidelines have revealed critical limitations in accurately mimicking the complexity of liver tissue-specific microenvironments and human in vivo efficacy. Cellular dissociation-induced structural disruption noise and inter-individual genetic variations have significantly compromised the accuracy of in silico computational models, leading to persistent data barriers that hinder the effective screening of novel drug candidates and the achievement of target-specific prophylactic concentrations. In particular, the diverse array of SERPING1 gene mutations (point mutations, insertions, deletions, and large fragment mutations) observed in patients with hereditary angioedema (HAE) Type 1 has made it challenging to precisely measure the intracellular accumulation and secretory flux of C1 esterase inhibitor (C1INH), resulting in repeated pharmacokinetic mismatches during clinical trials by multinational pharmaceutical companies. Existing static baseline analyses only track protein levels in human plasma, failing to dissect the gene-specific transcriptional delays and post-translational processing bottlenecks, thereby creating an omic data bottleneck.
Discovery: CRISPR/Cas9 Gene Editing EPSC-HLC Scan and Single-Cell Resolution Multidimensional Independent Variable Tensor Synchronization
In this study, we established a high-resolution platform by reprogramming peripheral blood mononuclear cell (PBMC)-derived erythroblasts from HAE patients into expanded potential stem cells (EPSCs) and subsequently differentiating them into hepatocyte-like cells (HLCs) to enable computational omics synchronization. We integrated the transcriptional matrix variation curves and protein secretion behavior for each SERPING1 mutation profile into a multidimensional independent variable tensor, allowing for the precise in silico reconstruction of mutation-specific energy barriers and binding free energies. The EPSC-HLC performance, which significantly surpasses existing simple differentiation models, perfectly recapitulates the genetic defects and secretory deficiencies observed in patients. Furthermore, we completed a sufficiency assay by performing CRISPR/Cas9-mediated genome editing and reverse-transplantation into normal control cells. This allowed us to fully elucidate the topological network variations of the downstream complement and kinin secretion cascades and to demonstrate the molecular biological integrity of computational predictions.
Establishment of a Precision-Stratified Model for SERPING1 Mutation Profile Modulation and Reversible Complement-Contact System Homeostasis
This model implements an architecture that derives SERPING1 mRNA instability coefficients and protein folding energies from individual patient omics matrix data, enabling precision stratification of patient populations. By targeting the upstream rate-limiting steps of the complement activation pathway, we have established a reversible and autonomous down-clamping/up-clamping feedback loop that regulates abnormal bradykinin release flux. From single amino acid substitution mutations to large fragment deletion mutations, we calculate intracellular protein accumulation rate constants using a differential equation-based approach, ensuring a dynamic maintenance of vascular permeability homeostasis. This in silico validated reversible model serves as the backbone for a precision-stratified model that maximizes ex vivo efficacy and stability by inhibiting kallikrein activity in the microenvironment without inducing cytotoxicity, even under external oxidative stress or inflammatory cytokine gradients.
Prospects: Establishing a Programmable Computational Toxicology Standard and Launching a Next-Generation IND Digital Governance System
The establishment of this computational genetics architecture will supersede conventional post-clinical analysis frameworks and reshape omics R&D governance into a molecular mechanism-based, programmable systems biology infrastructure. In the development of in vivo CRISPR therapeutics (e.g., NTLA-2002) and RNAi therapeutics by global biotech and multinational pharmaceutical companies (e.g., Intellia, Ionis, Takeda), it will be possible to link patient-specific cell line reaction genetic gradient correction coefficients at the high-throughput screening stage. This represents a computational moat and a core proprietary asset that eliminates batch effects and inter-production unit variations that occur during cell line establishment and commercial screening. Furthermore, it will serve as a digital data governance standard that not only meets the companion diagnostic (CDx) licensing requirements of global regulatory agencies such as the FDA but also drastically shortens the timeline for pharmacologic safety verification during clinical trial protocol (IND) application and cGMP licensing evaluation.
Hereditary angioedema (HAE) with C1 esterase inhibitor (C1INH) deficiency is caused by pathogenic SERPING1 mutations that disrupt production of the plasma protease inhibitor C1INH. However, the molecular mechanisms and consequences of patient-specific mutations remain poorly understood due to the lack of physiologically relevant human models. Here, we established a personalized, isogenic, stem-cell-derived hepatocyte platform to investigate the underlying mutation-specific mechanisms of HAE. Specifically, peripheral blood mononuclear cell (PBMC)-expanded erythroblasts from four representative HAE-C1INH-Type1 patients containing distinct point, insertion, deletion, or large fragment SERPING1 mutations were reprogrammed into expanded potential stem cells (EPSCs) and further differentiated into hepatocyte-like cells (HLCs). These HLCs exhibited appropriate transcriptional transitions, mature hepatic features, and C1INH secretion comparable to that observed in human plasma. All patient-derived HLCs demonstrated impaired C1INH secretion with mutation-specific differences in both SERPING1 transcription and intracellular accumulation. Moreover, to verify that the mutations directly drive the phenotype, we performed CRISPR/Cas9-mediated genome repair, which restored SERPING1 mRNA expression and C1INH secretion. Conversely, identical patient mutations installed into healthy EPSCs showed the same transcriptional and secretory defects, confirming sufficiency. Collectively, we have established a robust human hepatocyte model that accurately recapitulates key hepatocyte-specific aspects of HAE pathophysiology and provides a scalable foundation for investigation of future precision therapies.
The discovery of the EPSC-HLC platform in this study goes beyond theoretical SERPING1 mechanism exploration and directly translates into the actual global finished pharmaceutical supply chain market and the next generation of precision medicine-based bio-business lines.
First, by immediately scanning SERPING1 mutation C1INH secretion kinetics using a Python algorithm-based cell modeling approach in the clinical setting, it eliminates the temporal noise associated with acute airway closure and edema, and safeguards the hepatocyte homeostasis.
At the same time, by linking to the open-source NCBI dbSNP and Ensembl databases, which contain large-scale transcriptomic datasets, it enables the virtual simulation of confounding variables from heterogeneous allogeneic organs during clinical trial design and the real-time reconstruction of the effective docking concentration of the target C1INH protein, realizing a companion diagnostic (CDx) panel interface.
Furthermore, in the large-scale approval clinical trials of next-generation CRISPR/Cas9 and RNAi target gene therapies by multinational companies, by linking the cell secretion quantity correction coefficients, it eliminates batch-to-batch gene expression variations and maximizes the probability of obtaining clinical trial protocol and cGMP commercial operation licenses from global regulatory agencies, serving as a backbone infrastructure.