A Quantitative Clinical Pharmacology-Based Framework For Model-Informed Vaccine Development.

Historically, vaccine development and dose optimization have followed mostly empirical approaches without clinical pharmacology and model-informed approaches playing a major role, in contrast to conventional drug development. This is attributed to the complex cascade of immunobiological mechanisms associated with vaccines and a lack of quantitative frameworks for extracting dose-exposure-efficacy-toxicity relationships. However, the Covid-19 pandemic highlighted the lack of sufficient immunogenicity due to suboptimal vaccine dosing regimens and the need for well-designed, model-informed clinical trials which enhance the probability of selection of optimal vaccine dosing regimens. In this perspective, we attempt to develop a quantitative clinical pharmacology-based approach that integrates vaccine dose-efficacy-toxicity across various stages of vaccine development into a unified framework that we term as model-informed vaccine dose-optimization and development (MIVD). We highlight scenarios where the adoption of MIVD approaches may have a strategic advantage compared to conventional practices for vaccines.

Vaccine models predict rules for updating vaccines against evolving pathogens such as SARS-CoV-2 and influenza in the context of pre-existing immunity.

Currently, vaccines for SARS-CoV-2 and influenza viruses are updated if the new vaccine induces higher antibody-titers to circulating variants than current vaccines. This approach does not account for complex dynamics of how prior immunity skews recall responses to the updated vaccine. We: (i) use computational models to mechanistically dissect how prior immunity influences recall responses; (ii) explore how this affects the rules for evaluating and deploying updated vaccines; and (iii) apply this to SARS-CoV-2. Our analysis of existing data suggests that there is a strong benefit to updating the current SARS-CoV-2 vaccines to match the currently circulating variants. We propose a general two-dose strategy for determining if vaccines need updating as well as for vaccinating high-risk individuals. Finally, we directly validate our model by reanalysis of earlier human H5N1 influenza vaccine studies.

Modeling suggests that multiple immunizations or infections will reveal the benefits of updating SARS-CoV-2 vaccines.

When should vaccines to evolving pathogens such as SARS-CoV-2 be updated? Our computational models address this focusing on updating SARS-CoV-2 vaccines to the currently circulating Omicron variant. Current studies typically compare the antibody titers to the new variant following a single dose of the original-vaccine versus the updated-vaccine in previously immunized individuals. These studies find that the updated-vaccine does not induce higher titers to the vaccine-variant compared with the original-vaccine, suggesting that updating may not be needed. Our models recapitulate this observation but suggest that vaccination with the updated-vaccine generates qualitatively different humoral immunity, a small fraction of which is specific for unique epitopes to the new variant. Our simulations suggest that these new variant-specific responses could dominate following subsequent vaccination or infection with either the currently circulating or future variants. We suggest a two-dose strategy for determining if the vaccine needs updating and for vaccinating high-risk individuals.

Mechanistic Models of COVID-19: Insights into Disease Progression, Vaccines, and Therapeutics.

The COVID-19 pandemic has severely impacted health systems and economies worldwide. Significant global efforts are therefore ongoing to improve vaccine efficacies, optimize vaccine deployment, and develop new antiviral therapies to combat the pandemic. Mechanistic viral dynamics and quantitative systems pharmacology models of SARS-CoV-2 infection, vaccines, immunomodulatory agents, and antiviral therapeutics have played a key role in advancing our understanding of SARS-CoV-2 pathogenesis and transmission, the interplay between innate and adaptive immunity to influence the outcomes of infection, effectiveness of treatments, mechanisms and performance of COVID-19 vaccines, and the impact of emerging SARS-CoV-2 variants. Here, we review some of the critical insights provided by these models and discuss the challenges ahead.

Modeling how antibody responses may determine the efficacy of COVID-19 vaccines.

Predicting the efficacy of COVID-19 vaccines would aid vaccine development and usage strategies, which is of importance given their limited supplies. Here we develop a multiscale mathematical model that proposes mechanistic links between COVID-19 vaccine efficacies and the neutralizing antibody (NAb) responses they elicit. We hypothesized that the collection of all NAbs would constitute a shape space and that responses of individuals are random samples from this space. We constructed the shape space by analyzing reported in vitro dose-response curves of ~80 NAbs. Sampling NAb subsets from the space, we recapitulated the responses of convalescent patients. We assumed that vaccination would elicit similar NAb responses. We developed a model of within-host SARS-CoV-2 dynamics, applied it to virtual patient populations and, invoking the NAb responses above, predicted vaccine efficacies. Our predictions quantitatively captured the efficacies from clinical trials. Our study thus suggests plausible mechanistic underpinnings of COVID-19 vaccines and generates testable hypotheses for establishing them.

Increased B Cell Selection Stringency In Germinal Centers Can Explain Improved COVID-19 Vaccine Efficacies With Low Dose Prime or Delayed Boost.

The efficacy of COVID-19 vaccines appears to depend in complex ways on the vaccine dosage and the interval between the prime and boost doses. Unexpectedly, lower dose prime and longer prime-boost intervals have yielded higher efficacies in clinical trials. To elucidate the origins of these effects, we developed a stochastic simulation model of the germinal center (GC) reaction and predicted the antibody responses elicited by different vaccination protocols. The simulations predicted that a lower dose prime could increase the selection stringency in GCs due to reduced antigen availability, resulting in the selection of GC B cells with higher affinities for the target antigen. The boost could relax this selection stringency and allow the expansion of the higher affinity GC B cells selected, improving the overall response. With a longer dosing interval, the decay in the antigen with time following the prime could further increase the selection stringency, amplifying this effect. The effect remained in our simulations even when new GCs following the boost had to be seeded by memory B cells formed following the prime. These predictions offer a plausible explanation of the observed paradoxical effects of dosage and dosing interval on vaccine efficacy. Tuning the selection stringency in the GCs using prime-boost dosages and dosing intervals as handles may help improve vaccine efficacies.

Application of quantitative systems pharmacology to guide the optimal dosing of COVID-19 vaccines.

Optimal use and distribution of coronavirus disease 2019 (COVID-19) vaccines involves adjustments of dosing. Due to the rapidly evolving pandemic, such adjustments often need to be introduced before full efficacy data are available. As demonstrated in other areas of drug development, quantitative systems pharmacology (QSP) is well placed to guide such extrapolation in a rational and timely manner. Here, we propose for the first time how QSP can be applied in the context of COVID-19 vaccine development.

Evaluating Coarse-Grained MARTINI Force-Fields for Capturing the Ripple Phase of Lipid Membranes.

Phospholipids, which are an integral component of cell membranes, exhibit a rich variety of lamellar phases modulated by temperature and composition. Molecular dynamics (MD) simulations have greatly enhanced our understanding of phospholipid membranes by capturing experimentally observed phases and phase transitions at molecular resolution. However, the ripple (Pß') membrane phase, observed as an intermediate phase below the main gel-to-liquid crystalline transition with some lipids, has been challenging to capture with MD simulations, both at all-atom and coarse-grained (CG) resolutions. Here, with an aggregate ∼2.5 µs all-atom and ∼122 µs CGMD simulations, we systematically assess the ability of six CG MARTINI 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) lipid and water force-field (FF) variants, parametrized to capture the DPPC gel and fluid phases, for their ability to capture the Pß' phase, and compared observations with those from an all-atom FF. Upon cooling from the fluid phase to below the phase transition temperature with smaller (380-lipid) and larger (>2200-lipid) MARTINI and all-atom (CHARMM36 FF) DPPC lipid bilayers, we observed that smaller bilayers with both all-atom and MARTINI FFs sampled interdigitated Pß' and ripple-like states, respectively. However, while all-atom simulations of the larger DPPC membranes exhibited the formation of the Pß' phase, MARTINI membranes did not sample interdigitated ripple-like states at larger system sizes. We then demonstrated that the ripple-like states in smaller MARTINI membranes were kinetically trapped structures caused by finite size effects rather than being representative of true Pß' phases. We showed that a MARTINI FF variant that could capture the tilted Lß' gel phase, a prerequisite for stabilizing the Pß' phase, was unable to capture the rippled phase upon cooling. Our study reveals that the current MARTINI FFs (including MARTINI3) may require specific reparametrization of the interaction potentials to stabilize lipid interdigitation, a characteristic of the ripple phase.

Using multiscale molecular dynamics simulations to obtain insights into pore forming toxin mechanisms.

Pore forming toxins (PFTs) are virulent proteins released by several species, including many strains of bacteria, to attack and kill host cells. In this article, we focus on the utility of molecular dynamics (MD) simulations and the molecular insights gleaned from these techniques on the pore forming pathways of PFTs. In addition to all-atom simulations which are widely used, coarse-grained MARTINI models and structure-based models have also been used to study PFTs. Here, the emphasis is on methods and techniques involved while setting up, monitoring, and evaluating properties from MD simulations of PFTs in a membrane environment. We draw from several case studies to illustrate how MD simulations have provided molecular insights into protein-protein and protein-lipid interactions, lipid dynamics, conformational transitions and structures of both the oligomeric intermediates and assembled pore structures.

Physical 'strength' of the multi-protein chain connecting immune cells: Does the weakest link limit antibody affinity maturation?: The weakest link in the multi-protein chain facilitating antigen acquisition by B cells in germinal centres limits antibody affinity maturation.

The affinities of antibodies (Abs) for their target antigens (Ags) gradually increase in vivo following an infection or vaccination, but reach saturation at values well below those realisable in vitro. This 'affinity ceiling' could in many cases restrict our ability to fight infections and compromise vaccines. What determines the affinity ceiling has been an unresolved question for decades. Here, we argue that it arises from the strength of the chain of protein complexes that is pulled by B cells during the process of Ag acquisition. The affinity ceiling is determined by the strength of the weakest link in the chain. We identify the weakest link and show that the resulting affinity ceiling can explain the Ab affinities realized in vivo, providing a conceptual understanding of Ab affinity maturation. We explore plausible evolutionary underpinnings of the affinity ceiling, examine supporting evidence and alternative hypotheses and discuss implications for vaccination strategies.

Predicting interfacial hot-spot residues that stabilize protein-protein interfaces in oligomeric membrane-toxin pores through hydrogen bonds and salt bridges.

Pore forming toxins (PFTs) are proteins which form unregulated oligomeric pores on target plasma membranes to cause ion leakage and cell death and represent the largest class of bacterial virulence factors. With increasing antibiotic-resistant bacterial strains, alternate therapies are being developed to target toxin pore formation rather than the bacteria themselves. One strategy is to undermine the stability of these multimeric pores, whose subunits are held together by complex amino acid interaction networks, by identifying key residues in these networks which could be plausible drug or mutagenesis targets. However, this requires a quantitative assessment of per residue contributions towards pore stability, which cannot be reliably gleaned from static crystal/cryo-EM pore structures. In this study, we overcome this limitation by developing a computational screening algorithm that employs fully atomistic molecular dynamics simulations coupled with energy-based screening that can predict 'hot-spot' residues which engage in persistent and stabilizing hydrogen bonds or salt bridges across protein-protein interfaces. Application of this algorithm to prototypical α-PFT (cytolysin A) and ß-PFT (α-hemolysin) membrane-inserted pores yielded a small predicted set of highly interacting residues, blocking of which could destabilize pore complexes. Previous mutagenesis studies validate some of our in silico predictions. The algorithm could be applied to all pores with known structures to generate a database of destabilizing mutations, which could then serve as a basis for experimental validation and rational structure-based inhibitor design.Communicated by Ramaswamy H. Sarma.

Targeting TMPRSS2 and Cathepsin B/L together may be synergistic against SARS-CoV-2 infection.

The entry of SARS-CoV-2 into target cells requires the activation of its surface spike protein, S, by host proteases. The host serine protease TMPRSS2 and cysteine proteases Cathepsin B/L can activate S, making two independent entry pathways accessible to SARS-CoV-2. Blocking the proteases prevents SARS-CoV-2 entry in vitro. This blockade may be achieved in vivo through 'repurposing' drugs, a potential treatment option for COVID-19 that is now in clinical trials. Here, we found, surprisingly, that drugs targeting the two pathways, although independent, could display strong synergy in blocking virus entry. We predicted this synergy first using a mathematical model of SARS-CoV-2 entry and dynamics in vitro. The model considered the two pathways explicitly, let the entry efficiency through a pathway depend on the corresponding protease expression level, which varied across cells, and let inhibitors compromise the efficiency in a dose-dependent manner. The synergy predicted was novel and arose from effects of the drugs at both the single cell and the cell population levels. Validating our predictions, available in vitro data on SARS-CoV-2 and SARS-CoV entry displayed this synergy. Further, analysing the data using our model, we estimated the relative usage of the two pathways and found it to vary widely across cell lines, suggesting that targeting both pathways in vivo may be important and synergistic given the broad tissue tropism of SARS-CoV-2. Our findings provide insights into SARS-CoV-2 entry into target cells and may help improve the deployability of drug combinations targeting host proteases required for the entry.

Early exposure to broadly neutralizing antibodies may trigger a dynamical switch from progressive disease to lasting control of SHIV infection.

Antiretroviral therapy (ART) for HIV-1 infection is life-long. Stopping therapy typically leads to the reignition of infection and progressive disease. In a major breakthrough, recent studies have shown that early initiation of ART can lead to sustained post-treatment control of viremia, raising hopes of long-term HIV-1 remission. ART, however, elicits post-treatment control in a small fraction of individuals treated. Strikingly, passive immunization with broadly neutralizing antibodies (bNAbs) of HIV-1 early in infection was found recently to elicit long-term control in a majority of SHIV-infected macaques, suggesting that HIV-1 remission may be more widely achievable. The mechanisms underlying the control elicited by bNAb therapy, however, remain unclear. Untreated infection typically leads to progressive disease. We hypothesized that viremic control represents an alternative but rarely realized outcome of the infection and that early bNAb therapy triggers a dynamical switch to this outcome. To test this hypothesis, we constructed a model of viral dynamics with bNAb therapy and applied it to analyse clinical data. The model fit quantitatively the complex longitudinal viral load data from macaques that achieved lasting control. The model predicted, consistently with our hypothesis, that the underlying system exhibited bistability, indicating two potential outcomes of infection. The first had high viremia, weak cytotoxic effector responses, and high effector exhaustion, marking progressive disease. The second had low viremia, strong effector responses, and low effector exhaustion, indicating lasting viremic control. Further, model predictions suggest that early bNAb therapy elicited lasting control via pleiotropic effects. bNAb therapy lowers viremia, which would also limit immune exhaustion. Simultaneously, it can improve effector stimulation via cross-presentation. Consequently, viremia may resurge post-therapy, but would encounter a primed effector population and eventually get controlled. ART suppresses viremia but does not enhance effector stimulation, explaining its limited ability to elicit post-treatment control relative to bNAb therapy.

Assessing the extent of the structural and dynamic modulation of membrane lipids due to pore forming toxins: insights from molecular dynamics simulations.

Infections caused by many virulent bacterial strains are triggered by the release of pore forming toxins (PFTs), which form oligomeric transmembrane pore complexes on the target plasma membrane. The spatial extent of the perturbation to the surrounding lipids during pore formation is relatively unexplored. Using all-atom molecular dynamics simulations, we investigate the changes in the structure and dynamics of lipids in a 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) lipid bilayer in the presence of contrasting PFTs. Cytolysin A (ClyA), an α toxin with its inserted wedge shaped bundle of inserted α helices, induces significant asymmetry across the membrane leaflets in comparison with α hemolysin (AHL), a ß toxin. Despite the differences in hydrophobic mismatch and uniquely different topologies of the two oligomers, perturbations to lipid order as reflected in the tilt angle and order parameters and membrane thinning are short ranged, lying within ∼2.5 nm from the periphery of either pore complex, and commensurate with distances typically associated with van der Waals forces. In contrast, the spatial extent of perturbations to the lipid dynamics extends outward to at least 4 nm for both proteins, and the continuous survival probabilities reveal the presence of a tightly bound shell of lipids in this region. Displacement probability distributions show long tails and the distinctly non-Gaussian features reflect the induced dynamic heterogeneity. A detailed profiling of the protein-lipid contacts with tyrosine, tryptophan, lysine and arginine residues shows increased non-polar contacts in the cytoplasmic leaflet for both PFTs, with a higher number of atomic contacts in the case of AHL in the extracellular leaflet due to the mushroom-like topology of the pore complex. The short ranged nature of the perturbations observed in this simple one component membrane suggests inherent plasticity of membrane lipids enabling the recovery of the structure and membrane fluidity even in the presence of these large oligomeric transmembrane protein assemblies. This observation has implications in membrane repair processes such as budding or vesicle fusion events used to mitigate PFT virulence, where the underlying lipid dynamics and fluidity in the vicinity of the pore complex are expected to play an important role.

Preferential Presentation of High-Affinity Immune Complexes in Germinal Centers Can Explain How Passive Immunization Improves the Humoral Response.

Passive immunization (PI) with external antibodies has been used classically for rapid but temporary alleviation of disease. Transcending this role, recent studies have shown PI to induce lasting improvements in natural antibody production, suggesting that PI could become a powerful tool to engineer humoral responses. We propose a mechanism with which PI can alter the humoral response. Antigen-specific B cells evolve and get selected in germinal centers (GCs) on the basis of their ability to acquire antigen from antibody-antigen complexes presented in GCs. When external antibodies of high affinity for antigen are used, they form the majority of the complexes in GCs, letting only B cells with even higher affinities be selected. Using an in silico GC reaction model, we show that this mechanism explains the improved humoral responses following PI. The model also synthesizes several independent experimental observations, indicating the robustness of the mechanism, and proposes tunable handles to optimize PI.

Assessing the Structure and Stability of Transmembrane Oligomeric Intermediates of an α-Helical Toxin.

Protein membrane interactions play an important role in our understanding of diverse phenomena ranging from membrane-assisted protein aggregation to oligomerization and folding. Pore-forming toxins (PFTs) are the primary vehicle for infection by several strains of bacteria. These proteins which are expressed in a water-soluble form (monomers) bind to the target membrane and conformationally transform (protomers) and self-assemble to form a multimer transmembrane pore complex through a process of oligomerization. On the basis of the structure of the transmembrane domains, PFTs are broadly classified into ß or α toxins. In contrast to ß-PFTs, the paucity of available crystal structures coupled with the amphipathic nature of the transmembrane domains has hindered our understanding of α-PFT pore formation. In this article, we use molecular dynamics (MD) simulations to examine the process of pore formation of the bacterial α-PFT, cytolysin A from Escherichia coli (ClyA) in lipid bilayer membranes. Using atomistic MD simulations ranging from 50 to 500 ns, we show that transmembrane oligomeric intermediates or "arcs" form stable proteolipidic complexes consisting of protein arcs with toroidal lipids lining the free edges. By creating initial conditions where the lipids are contained within the arcs, we study the dynamics of spontaneous lipid evacuation and toroidal edge formation. This process occurs on the time scale of tens of nanoseconds, suggesting that once protomers oligomerize, transmembrane arcs are rapidly stabilized to form functional water channels capable of leakage. Using umbrella sampling with a coarse-grained molecular model, we obtain the free energy of insertion of a single protomer into the membrane. A single inserted protomer has a stabilization free energy of -52.9 ± 1.2 kJ/mol and forms a stable transmembrane water channel capable of leakage. Our simulations reveal that arcs are stable and viable intermediates that can occur during the pore-formation pathway for ClyA.

Capturing the Membrane-Triggered Conformational Transition of an α-Helical Pore-Forming Toxin.

Escherichia coli cytolysin A (ClyA) is an α-helical pore-forming toxin (PFT) which lyses target cells by forming membrane permeabilizing pores. The rate-determining step of this process is the conversion of the soluble ClyA monomer into a membrane inserted protomer. We elucidate the mechanism of this conformational transition using molecular dynamics simulations of coarse-grained models of ClyA and a membrane. We find that a membrane is necessary for the conformational conversion because membrane-protein interactions counteract the loss of the many intraprotein hydrophobic interactions that stabilize the membrane-inserting segments in the ClyA monomer. Of the two membrane-inserting segments, the flexible and highly hydrophobic ß-tongue inserts first while the insertion of helix αA1 is membrane assisted. We conclude that the ß-tongue is designed to behave as a quick-response membrane sensor, while helix αA1 improves target selectivity for cholesterol-containing cell membranes by acting as a fidelity check.