Enhancing antibody affinity through experimental sampling of non-deleterious CDR mutations predicted by machine learning.

The application of machine learning (ML) models to optimize antibody affinity to an antigen is gaining prominence. Unfortunately, the small and biased nature of the publicly available antibody-antigen interaction datasets makes it challenging to build an ML model that can accurately predict binding affinity changes due to mutations (ΔΔG). Recognizing these inherent limitations, we reformulated the problem to ask whether an ML model capable of classifying deleterious vs non-deleterious mutations can guide antibody affinity maturation in a practical setting. To test this hypothesis, we developed a Random Forest classifier (Antibody Random Forest Classifier or AbRFC) with expert-guided features and integrated it into a computational-experimental workflow. AbRFC effectively predicted non-deleterious mutations on an in-house validation dataset that is free of biases seen in the publicly available training datasets. Furthermore, experimental screening of a limited number of predictions from the model (<10^2 designs) identified affinity-enhancing mutations in two unrelated SARS-CoV-2 antibodies, resulting in constructs with up to 1000-fold increased binding to the SARS-COV-2 RBD. Our findings indicate that accurate prediction and screening of non-deleterious mutations using machine learning offers a powerful approach to improving antibody affinity.

Decreased serotonin transporter activity in the mitral valve contributes to progression of degenerative mitral regurgitation.

Degenerative mitral valve (MV) regurgitation (MR) is a highly prevalent heart disease that requires surgery in severe cases. Here, we show that a decrease in the activity of the serotonin transporter (SERT) accelerates MV remodeling and progression to MR. Through studies of a population of patients with MR, we show that selective serotonin reuptake inhibitor (SSRI) use and SERT promoter polymorphism 5-HTTLPR LL genotype were associated with MV surgery at younger age. Functional characterization of 122 human MV samples, in conjunction with in vivo studies in SERT-/- mice and wild-type mice treated with the SSRI fluoxetine, showed that diminished SERT activity in MV interstitial cells (MVICs) contributed to the pathophysiology of MR through enhanced serotonin receptor (HTR) signaling. SERT activity was decreased in LL MVICs partially because of diminished membrane localization of SERT. In mice, fluoxetine treatment or SERT knockdown resulted in thickened MV leaflets. Similarly, silencing of SERT in normal human MVICs led to up-regulation of transforming growth factor ß1 (TGFß1) and collagen (COL1A1) in the presence of serotonin. In addition, treatment of MVICs with fluoxetine not only directly inhibited SERT activity but also decreased SERT expression and increased HTR2B expression. Fluoxetine treatment and LL genotype were also associated with increased COL1A1 expression in the presence of serotonin in MVICs, and these effects were attenuated by HTR2B inhibition. These results suggest that assessment of both 5-HTTLPR genotype and SERT-inhibiting treatments may be useful tools to risk-stratify patients with MV disease to estimate the likelihood of rapid disease progression.

Optimizing Antibody Affinity and Developability Using a Framework-CDR Shuffling Approach-Application to an Anti-SARS-CoV-2 Antibody.

The computational methods used for engineering antibodies for clinical development have undergone a transformation from three-dimensional structure-guided approaches to artificial-intelligence- and machine-learning-based approaches that leverage the large sequence data space of hundreds of millions of antibodies generated by next-generation sequencing (NGS) studies. Building on the wealth of available sequence data, we implemented a computational shuffling approach to antibody components, using the complementarity-determining region (CDR) and the framework region (FWR) to optimize an antibody for improved affinity and developability. This approach uses a set of rules to suitably combine the CDRs and FWRs derived from naturally occurring antibody sequences to engineer an antibody with high affinity and specificity. To illustrate this approach, we selected a representative SARS-CoV-2-neutralizing antibody, H4, which was identified and isolated previously based on the predominant germlines that were employed in a human host to target the SARS-CoV-2-human ACE2 receptor interaction. Compared to screening vast CDR libraries for affinity enhancements, our approach identified fewer than 100 antibody framework-CDR combinations, from which we screened and selected an antibody (CB79) that showed a reduced dissociation rate and improved affinity against the SARS-CoV-2 spike protein (7-fold) when compared to H4. The improved affinity also translated into improved neutralization (>75-fold improvement) of SARS-CoV-2. Our rapid and robust approach for optimizing antibodies from parts without the need for tedious structure-guided CDR optimization will have broad utility for biotechnological applications.

Mitigation of Blood Borne Cell Attachment to Metal Implants through CD47-Derived Peptide Immobilization.

The key complications associated with bare metal stents and drug eluting stents are in-stent restenosis and late stent thrombosis, respectively. Thus, improving the biocompatibility of metal stents remains a significant challenge. The goal of this protocol is to describe a robust technique of metal surface modification by biologically active peptides to increase biocompatibility of blood contacting medical implants, including endovascular stents. CD47 is an immunological species-specific marker of self and has anti-inflammatory properties. Studies have shown that a 22 amino acid peptide corresponding to the Ig domain of CD47 in the extracellular region (pepCD47), has anti-inflammatory properties like the full-length protein. In vivo studies in rats, and ex vivo studies in rabbit and human blood experimental systems from our lab have demonstrated that pepCD47 immobilization on metals improves their biocompatibility by preventing inflammatory cell attachment and activation. This paper describes the step-by step protocol for the functionalization of metal surfaces and peptide attachment. The metal surfaces are modified using polyallylamine bisphosphate with latent thiol groups (PABT) followed by deprotection of thiols and amplification of thiol-reactive sites via reaction with polyethyleneimine installed with pyridyldithio groups (PEI-PDT). Finally, pepCD47, incorporating terminal cysteine residues connected to the core peptide sequence through a dual 8-amino-3,6-dioxa-octanoyl spacer, are attached to the metal surface via disulfide bonds. This methodology of peptide attachment to metal surface is efficient and relatively inexpensive and thus can be applied to improve biocompatibility of several metallic biomaterials.

Stability and bioactivity of pepCD47 attachment on stainless steel surfaces.

In-stent restenosis (ISR) and late stent thrombosis are the major complications associated with the use of metal stents and drug eluting stents respectively. Our lab previously investigated the use of peptide CD47 in improving biocompatibility of bare metal stents in a rat carotid stent model and our results demonstrated a significant reduction in platelet deposition and ISR. However, this study did not characterize the stability of the pepCD47 on metal surfaces post storage, sterilization and deployment. Thus, the objective of the present study was 1) to test the stability of the peptide post - storage, sterilization, exposure to shear and mechanical stress and 2) to begin to expand our current knowledge of pepCD47 coated metal surfaces into the preclinical large animal rabbit model. Our results show that the maximum immobilization density of pepCD47 on metal surfaces is approximately 350 ng/cm2. 100% of the pepCD47 was retained on the metal surface post 24 weeks of storage at 4 °C, exposure to physiological shear stress, and mechanical stress of stent expansion. The bioactivity of the pepCD47 was found to be intact post 24 weeks of storage and ethylene oxide sterilization. Finally our ex vivo studies demonstrated that compared to bare metal the rabbit pepCD47 coated surfaces showed - 45% reduced platelet adhesion, a 10-fold decrease in platelet activation, and 93% endothelial cell retention. Thus, our data suggests that pepCD47 coating on metal surfaces is stable and rabbit pepCD47 shows promising preliminary results in preventing thrombosis and not inhibiting the growth of endothelial cells. STATEMENT OF SIGNIFICANCE: Biocompatibility of bare metal stents is a major challenge owing to the significantly high rates of in-stent restenosis. Previously we demonstrated that peptide CD47 functionalization improves the biocompatibility of bare metal stents in rat model. A similar trend was observed in our ex vivo studies where rabbit blood was perfused over the rabbit pepCD47 functionalized surfaces. These results provide valuable proof of concept data for future in vivo rabbit model studies. In addition, we investigated stability of the pepCD47 on metal surface and observed that pepCD47 coating is stable over time and resistant to industrially relevant pragmatic challenges.

Cell death atlas of the postnatal mouse ventral forebrain and hypothalamus: effects of age and sex.

Naturally occurring cell death is essential to the development of the mammalian nervous system. Although the importance of developmental cell death has been appreciated for decades, there is no comprehensive account of cell death across brain areas in the mouse. Moreover, several regional sex differences in cell death have been described for the ventral forebrain and hypothalamus, but it is not known how widespread the phenomenon is. We used immunohistochemical detection of activated caspase-3 to identify dying cells in the brains of male and female mice from postnatal day (P) 1 to P11. Cell death density, total number of dying cells, and regional volume were determined in 16 regions of the hypothalamus and ventral forebrain (the anterior hypothalamus, arcuate nucleus, anteroventral periventricular nucleus, medial preoptic nucleus, paraventricular nucleus, suprachiasmatic nucleus, and ventromedial nucleus of the hypothalamus; the basolateral, central, and medial amygdala; the lateral and principal nuclei of the bed nuclei of the stria terminalis; the caudate-putamen; the globus pallidus; the lateral septum; and the islands of Calleja). All regions showed a significant effect of age on cell death. The timing of peak cell death varied between P1 to P7, and the average rate of cell death varied tenfold among regions. Several significant sex differences in cell death and/or regional volume were detected. These data address large gaps in the developmental literature and suggest interesting region-specific differences in the prevalence and timing of cell death in the hypothalamus and ventral forebrain.