Biosynthesis of collagen molecules, as well as collagenous segments of non-collagen proteins, requires extensive post-translational modifications (PTMs) that confer specific functional and structural properties to tissues and organs. Within collagen polypeptide sequences, lysine residues are subject to a series of PTMs, resulting in hydroxylation and subsequent O-linked glycosylation of their side chains. These modifications are catalyzed by two distinct metalloenzyme families, named LH/PLOD and GLT25D/COLGALT, which alternatively process modified lysine side chains yielding 5-hydroxylysine (LH/PLOD), β-(1,O)-galactosyl-5-hydroxylysine (GLT25D/COLGALT), and α-(1,2)-glucosyl-β-(1,O)-galactosyl-5-hydroxylysine (LH/PLOD). In this chapter, we illustrate strategies to recombinantly produce human full-length LH3/PLOD3 as well as GLT25D1/COLGALT1, verify the quality of these recombinant preparations, and perform direct and indirect assays to evaluate their enzymatic activities in vitro on synthetic collagen peptides and gelatin.

During collagen biosynthesis, lysine residues undergo extensive post-translational modifications through the alternate action of two distinct metal ion-dependent enzyme families (i.e., LH/PLODs and GLT25D/COLGALT), ultimately producing the highly conserved α-(1,2)-glucosyl-β-(1,O)-galactosyl-5-hydroxylysine pattern. Malfunctions in these enzymes are linked to developmental pathologies and extracellular matrix alterations associated to enhanced aggressiveness of solid tumors. Here, we characterized human GLT25D1/COLGALT1, revealing an elongated head-to-head homodimeric assembly. Each monomer encompasses two domains (named GT1 and GT2), both unexpectedly capable of binding metal ion cofactors and UDP-α-galactose donor substrates, resulting in four candidate catalytic sites per dimer. We identify the catalytic site in GT2, featuring an unusual Glu-Asp-Asp motif critical for Mn2+ binding, ruling out direct catalytic roles for the GT1 domain, but showing that in this domain the unexpectedly bound Ca2+ and UDP-α-galactose cofactors are critical for folding stability. Dimerization, albeit not essential for GLT25D1/COLGALT1 activity, provides a critical molecular contact site for multi-enzyme assembly interactions with partner multifunctional LH/PLOD lysyl hydroxylase-glycosyltransferase enzymes.

Abstract There is great interest in the identification of biomarkers to guide development of antibody-drug conjugates (ADC). We previously showed that loss of Neurofibromatosis 1 (NF1), a gene frequently mutated across cancers, enhances the activity of DM1, the maytansinoid payload of T-DM1, through a novel function in regulating microtubule (MT) dynamics. Maytansinoids are puzzlingly more effective in cells (in the nanomolar range) vs in vitro (in the micromolar range). Since maytansinoids bind at the interface between tubulin dimers, they are thought to only bind soluble tubulin dimers or MT ends, which would suggest very few binding sites available for pharmacological interaction in vivo, at odds with data. Here we investigated the interaction of DM1 with NF1 and MTs, using cellular and reductionist in vitro systems. To measure in vivo MT dynamics, we transiently transfected the MT end-binding protein EB3-GFP and reconstructed MT trajectories by live-cell imaging. Upon DM1 treatment, KO cells showed a highly significant reduction in MT speed, demonstrating a direct role for NF1 on MT dynamics in cells. In turbidity-based tubulin polymerization assays, recombinant NF1 greatly accelerated polymerization, and completely rescued DM1-induced inhibition. Visual inspection of fluorescent MTs showed that NF1 induced significant MT bundling, a defining feature of many MT-associated proteins, which generates signal indistinguishable from true MT polymerization in turbidity assays. To follow the dynamics of individual microtubules, we applied Total Internal Reflection (TIRF) microscopy on glass-immobilized MTs. As expected, polymerization in the presence of NF1 led to a significant increase in MT dynamics (elongation speed, rescue and catastrophe rate). Expectedly, DM1 led to significant reduction in the fraction of elongating MTs and speed, but these defects were completely or partially rescued by NF1. Importantly, DM1 did not only lead to MT shortening (as proposed by the current model), but also to clear and frequent MT fracturing, indicating that the drug is not only engaging MT ends but also intra-tubular binding sites. This is consistent with recent models of MT formation which incorporate the frequent presence of areas of discontinuity or damage induced by mechanical stress, exposing intra-tubular DM1 binding sites. Interestingly, adding NF1 to DM1-treated MTs generated areas of de novo intra-tubular tubulin insertion, coincident with damaged sites, suggesting an entirely novel role for NF1 in MT repair.In conclusion, we provide evidence for a model in which maytansinoids bind not only to soluble tubulin dimers and MT ends, but also to intra-tubular damaged sites. Thus, the number of binding sites in cells would be proportional to MT damage, suggesting a mechanism for differential efficacy across tumor types and a potential avenue for combinatorial drug development. Citation Format: Eleonora Messuti, Bruno Achutti Duso, Alessia Castiglioni, Giulia Tini, Emanuele Bonetti, Giuseppe Ciossani, Silvia Monzani, Daria Khuntsariya, Zdeněk Lánský, Marcus Braun, Luigi Scietti, Luca Mazzarella. Intra-tubular damage is targeted by maytansinoids and rescued by NF1: Revisiting mechanism and biomarkers of an established ADC payload [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 513.

During biosynthesis, collagen lysine residues undergo extensive post-translational modifications essential for the stability and functions of collagen supramolecular assemblies. In the endoplasmic reticulum, two distinct metal ion dependent enzyme families (i.e., multifunctional lysyl hydroxylases-glucosyltransferases LH/PLODs and galactosyltransferases GLT25D/COLGALT) alternatively operate on collagen lysine side chains ultimately generating the highly conserved α-(1,2)-glucosyl-β-(1,O)-galactosyl-5-hydroxylysine pattern. Malfunctions in the collagen lysine post-translational modification machinery is linked to multiple developmental pathologies as well as extracellular matrix alterations causing enhanced cellular proliferation and invasiveness of several solid tumors, prompting for an in-depth characterization of LH/PLOD and GLT25D/COLGALT enzyme families. Here, we present an integrative molecular study of GLT25D1/COLGALT, highlighting an elongated head-to-head homodimeric assembly characterized by an N-terminal segment of each monomer wrapping around its dimerization partner. Each monomer encompasses two Rossman fold-type domains (GT1 and GT2) separated by an extended linker. Both domains were found capable of binding Mn2+ cofactors and UDP-α-galactose donor substrates, resulting in four candidate catalytic sites per dimer. Site-directed mutagenesis and biochemical studies identify the C-terminal GT2 domain as the functional GLT25D1/COLGALT1 catalytic site, highlighting an unprecedented Glu-Asp-Asp motif critical for metal ion binding, and suggesting structural roles for the N-terminal GT1 essential for correct quaternary structure assembly. Conversely, dimerization was not a requirement for GLT25D1/COLGALT1 enzymatic activity in vitro , suggesting that the elongated enzyme homodimer assembly, resembling that of LH/PLOD binding partners, could represent a functional hallmark for correct recognition and successful processing of collagen lysine residues. ### Competing Interest Statement The authors have declared no competing interest.

Abstract There is great interest in the identification of biomarkers to guide development of antibody-drug conjugates (ADC). Most research has focused on target expression, but key predictors of payload efficacy have not been indeitifed. NF1 is a tumor suppressor classically considered as an inhibitor of RAS signaling, and often mutated in metastatic HER2+ breast cancer (BC). We screened multiple approved drugs for differential sensitivityin CRISPR-engineeredf NF1 KO cells. HER2-targeted agents (small molecules or antibodies) were found to be less effective upon NF1 loss; surprisingly, we identified increased sensitivity to the approved ADC T-DM1, but not to Trastuzumab Deruxtecan (T-Dxd). We then elucidated the underlying molecular cause employing in vivo, in vitro and in vitro reductionist systems. To measure in vivo MT dynamics, we transiently transfected the MT end-binding protein EB3- GFP and reconstructed MT trajectories by live-cell imaging. Upon DM1 treatment, KO cells showed a highly significant reduction in MT speed, demonstrating a direct role for NF1 on MT dynamics in cells. In tubulin polymerization assays, recombinant NF1 greatly accelerated polymerization and completely rescued DM1-induced inhibition. NF1 induced significant MT bundling, a defining feature of many MT-associated proteins, which generates signal indistinguishable from true MT polymerization in turbidity assays. To follow the dynamics of individual microtubules, we applied Total Internal Reflection (TIRF) microscopy on glass-immobilized MTs. As expected, polymerization in the presence of NF1 led to a dose-dependent significant increase in MT dynamics (fraction of elongating MTs, elongation speed, catastrophe rate). Expectedly, DM1 led to significant reduction in the fraction of elongating MTs and speed, but these defects were completely or partially rescued by NF1. Importantly, DM1 did not only lead to MT shortening (as proposed by the current model), but also to clear and frequent MT fracturing, indicating that the drug is not only engaging MT ends but also intra-tubular binding sites. This is consistent with recent models of MT formation which incorporate the frequent presence of areas of discontinuity or damage induced by mechanical stress, exposing intratubular DM1 binding sites. Interestingly, adding NF1 to DM1-treated MTs generated areas of de novo intra-tubular tubulin insertion, coincident with damaged sites, suggesting an entirely novel role for NF1 in MT repair. In conclusion, we provide evidence for a model in which maytansinoids bind not only to soluble tubulin dimers and MT ends, but also to intratubular damaged sites. Thus, the number of binding sites in cells would be proportional to MT damage, suggesting a mechanism for differential efficacy across tumor types and a potential avenue for combinatorial drug development. These results prompt the use of NF1 as a biomarker to select patients for ADC treatment. Funding: FIEO fellowship 2023, AIRC (n25791), Italian MoH-Ricerca Corrente di Rete (ACCORD) 2022, Next Generation EU – PNRR M6C2 – PNRR-MAD-2022-12376934 Citation Format: Eleonora Messuti, Bruno Achutti Duso, Alessia Castiglioni, Giulia Tini, Emanuele Bonetti, Giuseppe Ciossani, Silvia Monzani, Daria Khuntsariya, Marcus Braun, Zdenek Lansky, Luigi Scietti, Luca Mazzarella. NF1 loss is syntetic lethal with Trastuzumab emtansine [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Expanding and Translating Cancer Synthetic Vulnerabilities; 2024 Jun 10-13; Montreal, Quebec, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(6 Suppl):Abstract nr A013.

Hydroxylysine glycosylations are post-translational modifications (PTMs) essential for the maturation and homeostasis of fibrillar and non-fibrillar collagen molecules. The multifunctional collagen lysyl hydroxylase 3 (LH3/PLOD3) and the collagen galactosyltransferase GLT25D1 are the human enzymes that have been identified as being responsible for the glycosylation of collagen lysines, although a precise description of the contribution of each enzyme to these essential PTMs has not yet been provided in the literature. LH3/PLOD3 is thought to be capable of performing two chemically distinct collagen glycosyltransferase reactions using the same catalytic site: an inverting beta-1,O-galactosylation of hydroxylysines (Gal-T) and a retaining alpha-1,2-glucosylation of galactosyl hydroxylysines (Glc-T). In this work, we have combined indirect luminescence-based assays with direct mass spectrometry-based assays and molecular structure studies to demonstrate that LH3/PLOD3 only has Glc-T activity and that GLT25D1 only has Gal-T activity. Structure-guided mutagenesis confirmed that the Glc-T activity is defined by key residues in the first-shell environment of the glycosyltransferase catalytic site as well as by long-range contributions from residues within the same glycosyltransferase (GT) domain. By solving the molecular structures and characterizing the interactions and solving the molecular structures of human LH3/PLOD3 in complex with different UDP-sugar analogs, we show how these studies could provide insights for LH3/PLOD3 glycosyltransferase inhibitor development. Collectively, our data provide new tools for the direct investigation of collagen hydroxylysine PTMs and a comprehensive overview of the complex network of shapes, charges, and interactions that enable LH3/PLOD3 glycosyltransferase activities, expanding the molecular framework and facilitating an improved understanding and manipulation of glycosyltransferase functions in biomedical applications.

The gene neurofibromatosis 1 (NF1) is increasingly recognized as a key somatic driver of cancerogenesis, in addition to its well-known role as the germline determinant of the onco-developmental syndrome Neurofibromatosis. NF1 is best characterized as a negative regulator of RAS activation, but several lines of evidence suggest that it may have additional, poorly characterized functions.In breast cancer, NF1 loss is known to be associated with resistance to endocrine and HER2-targeted therapy. Here, using HER2+ breast cancer cell lines engineered to ablate NF1 by CRISPR-Cas9 and in vitro reductionist models, we show that NF1 is a bona fide Microtubule-Associated Protein (MAP) with a novel, RAS-independent role in regulating dynamic instability and intra-lattice repair of microtubules. We show that loss of NF1 results in multiple mitotic defects (enlarged mitotic plate, delayed mitotic exit, supranumerary centrosomes and chromosome misalignment) that generate a low-grade aneuploidy that can also be measured as increased aneuploidy score in breast cancer patients bearing pathogenic NF1 mutations. Loss of NF1 leads to increased sensitivity to the approved Antibody-Drug Conjugate T-DM1 and in particular to its payload of the maytansin family, thus representing the first payload-specific predictive biomarker.These findings highlight a novel function for an established tumor suppressor and support the assessment of NF1 status to guide tailored treatment decision in breast cancer### Competing Interest StatementThe authors have declared no competing interest.

The recent advances in structural biology, combined with continuously increasing computational capabilities and development of advanced softwares, have drastically simplified the workflow for protein homology modeling. Modeling of individual proteins is nowadays quick and straightforward for a large variety of protein targets, thanks to guided pipelines relying on advanced computational tools and user-friendly interfaces, which have extended and promoted the use of modeling also to scientists not focusing on molecular structures of proteins. Nevertheless, construction of models of multi-protein complexes remains quite challenging for the non-experts, often due to the usage of specific procedures depending on the system under investigation and the need for experimental validation approaches to strengthen the generated output.In this chapter, we provide a brief overview of the approaches enabling generation of multi-protein complex models starting from homology models of individual protein components. Using real-life examples, we include two examples to guide the reader in the generation of homomeric and heteromeric protein models.

The tumor suppressor NF1 is classically considered a negative RAS regulator, but sparse evidence suggests additional RAS-independent roles. Early studies suggested an interaction with tubulin, which remains poorly characterized to date but may be of particular therapeutic interest as NF1 is somatically mutated across multiple tumor types. We showed that multiple CRISPR-Cas9-engineeerd NF1 KO HER2+ breast cancer cells (BT-474, SK-BR3, HCC1954) become exquisitely sensitive to the Antibody-Drug Conjugate (ADC) Trastuzumab emtansine (T-DM1); we here investigate the underlying mechanism.TDM1 hypersensitivy was specific to the maytansin microtubule-targeting payload, since it was i) replicated by the naked payload but not the naked antibody; ii) absent with other ADCs (T-DxD); iii) not accompanied by increased TDM1 uptake; iv) associated with increased tubulin-maytansin binding in KO cells, as per Cellular Thermal Shift Assay. The mechanism is likely RAS-independent, as KRAS G12V-overexpression did not alter TDM1 sensitivity. RNAseq revealed that KO cells deregulated genes associated with microtubular dynamics and G2-M transition more strongly upon TDM1-treatement. Multiple KO cells investigated by static and live imaging exhibited marked signs of altered mitosis, with longer G2/M, supernumerary centrosomes, chromosome misalignment and frequent aneuploidy, which could also be inferred in multiple public sequencing datasets (TCGA, MSK IMPACT, AACR GENIE). Based on this, we explored NF1 role on microtubule dynamics. We found several lines of evidence for a direct and mitosis-selective interaction between NF1 and tubulin: i) in Immunofluorescence (IF), NF1 was upregulated in mitosis and colocalized with the mitotic spindle; ii) NF1 co-Immunoprecipitated with tubulin in mitosis-enriched but not asynchronous cells; iii) purified NF1 co-eluted with tubulin in size-selection chromatography; iv) in silico modeling with AlphaFold2 predicted an interaction between NF1 central domains and the alpha-beta tubulin dimer. Crucially, KO cells exhibited severe microtubule hypodynamism in cold-induced depolymerization-repolymerization assays, replicated in multiple cell types and accompanied by imbalanced levels of plus/minus-end microtubule-associated proteins. By IF on cocultured WT/KO live cells, KO cells showed significantly higher GTP-tubulin, known to cause microtubular hyperstability, suggesting the intriguing possibility that NF1 may directly regulate tubulin intrinsic GTP-hydrolyzing activity, similar to its role on RAS. In conclusion, we provide extensive mechanistic evidence for a direct and previously underappreciated role of NF1 in microtubular dynamics, which reshapes our understanding of its tumor-suppressive activity and provides a rationale for pharmacological targeting of NF1-mutated tumors. Citation Format: Bruno A. Duso, Eleonora Messuti, Emanuele Bonetti, Giulia Tini, Alessia Castiglioni, Giuseppe Ciossani, Silvia Monzani, Chiara Soriani, Simona Rodighiero, Luigi Scietti, Costantino Jemos, Luca Mazzarella. NF1 (neurofibromatosis 1) controls microtubule dynamics and dictates sensitivity to maytansinoids. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4896.

Programmed cell death protein 1 (PD-1) is an immunoregulatory target which is recognized by different monoclonal antibodies, approved for the therapy of multiple types of cancer. Different anti-PD-1 antibodies display different therapeutic properties and there is a pharmaceutical interest to generate and characterize novel anti-PD-1 antibodies. We screened multiple human antibody phage display libraries to target novel epitopes on the PD-1 surface and we discovered a unique and previously undescribed binding specificity (termed D12) from a new antibody library (termed AMG). The library featured antibody fragments in single-chain fragment variable (scFv) format, based on the IGHV3-23*03 (VH ) and IGKV1-39*01 (Vκ) genes. The D12 antibody was characterized by surface plasmon resonance (SPR), cross-reacted with the Cynomolgus monkey antigen and bound to primary human T cells, as shown by flow cytometry. The antibody blocked the PD-1/PD-L1 interaction in vitro with an EC50 value which was comparable to the one of nivolumab, a clinically approved antibody. The fine details of the interaction between D12 and PD-1 were elucidated by x-ray crystallography of the complex at a 3.5 Å resolution, revealing an unprecedented conformational change at the N-terminus of PD-1 following D12 binding, as well as partial overlap with the binding site for the cognate PD-L1 and PD-L2 ligands which prevents their binding. The results of the study suggest that the expansion of antibody library repertoires may facilitate the discovery of novel binding specificities with unique properties that hold promises for the modulation of PD-1 activity in vitro and in vivo.

Heritable thoracic aortic disease and familial thoracic aortic aneurysm/dissection are important causes of human morbidity/mortality, most without identifiable genetic cause. In a family with familial thoracic aortic aneurysm/dissection, we identified a missense p. (Ser178Arg) variant in PLOD1 segregating with disease, and evaluated PLOD1 enzymatic activity, collagen characteristics and in human aortic vascular smooth muscle cells, studied the effect on function. Comparison with homologous PLOD3 enzyme indicated that the pathogenic variant may affect the N-terminal glycosyltransferase domain, suggesting unprecedented PLOD1 activity. In vitro assays demonstrated that wild-type PLOD1 is capable of processing UDP-glycan donor substrates, and that the variant affects the folding stability of the glycosyltransferase domain and associated enzymatic functions. The PLOD1 substrate lysine was elevated in the proband, however the enzymatic product hydroxylysine and total collagen content was not different, albeit despite collagen fibril narrowing and preservation of collagen turnover. In VSMCs overexpressing wild-type PLOD1, there was upregulation in procollagen gene expression (secretory function) which was attenuated in the variant, consistent with loss-of-function. In comparison, si-PLOD1 cells demonstrated hypercontractility and upregulation of contractile markers, providing evidence for phenotypic switching. Together, the findings suggest that the PLOD1 product is preserved, however newly identified glucosyltransferase activity of PLOD1 appears to be affected by folding stability of the variant, and is associated with compensatory vascular smooth muscle cells phenotypic switching to support collagen production, albeit with less robust fibril girth. Future studies should focus on the impact of PLOD1 folding/variant stability on the tertiary structure of collagen and ECM interactions.

Neurotrypsin (NT) is a highly specific nervous system multi-domain serine protease best known for its selective processing of the potent synaptic organizer agrin. Its enzymatic activity is thought to influence processes of synaptic plasticity, with its deregulation causing accelerated neuromuscular junction (NMJ) degeneration or contributing to forms of mental retardation. These biological effects are likely to stem from NT-based regulation of agrin signaling. However, dissecting the exact biological implications of NT-agrin interplay is difficult, due to the scarce molecular detail regarding NT activity and NT-agrin interactions. We developed a strategy to reliably produce and purify a catalytically competent engineered variant of NT called "NT-mini" and a library of C-terminal agrin fragments, with which we performed a thorough biochemical and biophysical characterization of NT enzyme functionality. We studied the regulatory effects of calcium ions and heparin, identified NT's heparin-binding domain, and discovered how zinc ions induce modulation of enzymatic activity. Additionally, we investigated myotube differentiation and hippocampal neuron excitability, evidencing a dose-dependent increase in neuronal activity alongside a negative impact on myoblast fusion when using the active NT enzyme. Collectively, our results provide in vitro and cellular foundations to unravel the molecular underpinnings and biological significance of NT-agrin interactions.

Multifunctional human collagen lysyl hydroxylase (LH/PLOD) enzymes catalyze post-translational hydroxylation and subsequent glycosylation of collagens, enabling their maturation and supramolecular organization in the extracellular matrix (ECM). Recently, the overexpression of LH/PLODs in the tumor microenvironment results in abnormal accumulation of these collagen post-translational modifications, which has been correlated with increased metastatic progression of a wide variety of solid tumors. These observations make LH/PLODs excellent candidates for prospective treatment of aggressive cancers. The recent years have witnessed significant research efforts to facilitate drug discovery on LH/PLODs, including molecular structure characterizations and development of reliable high-throughput enzymatic assays. Using a combination of biochemistry and in silico studies, we characterized the dual role of Fe2+ as simultaneous cofactor and inhibitor of lysyl hydroxylase activity and studied the effect of a promiscuous Fe2+ chelating agent, 2,2’-bipyridil, broadly considered a lysyl hydroxylase inhibitor. We found that at low concentrations, 2,2’-bipyridil unexpectedly enhances the LH enzymatic activity by reducing the inhibitory effect of excess Fe2+. Together, our results show a fine balance between Fe2+-dependent enzymatic activity and Fe2+-induced self-inhibited states, highlighting exquisite differences between LH/PLODs and related Fe2+, 2-oxoglutarate dioxygenases and suggesting that conventional structure-based approaches may not be suited for successful inhibitor development. These insights address outstanding questions regarding druggability of LH/PLOD lysyl hydroxylase catalytic site and provide a solid ground for upcoming drug discovery and screening campaigns.

Collagen is a major constituent of the extracellular matrix (ECM) that confers fundamental mechanical properties to tissues. To allow proper folding in triple-helices and organization in quaternary super-structures, collagen molecules require essential post-translational modifications (PTMs), including hydroxylation of proline and lysine residues, and subsequent attachment of glycan moieties (galactose and glucose) to specific hydroxylysine residues on procollagen alpha chains. The resulting galactosyl-hydroxylysine (Gal-Hyl) and less abundant glucosyl-galactosyl-hydroxylysine (Glc-Gal-Hyl) are amongst the simplest glycosylation patterns found in nature and are essential for collagen and ECM homeostasis. These collagen PTMs depend on the activity of specialized glycosyltransferase enzymes. Although their biochemical reactions have been widely studied, several key biological questions about the possible functions of these essential PTMs are still missing. In addition, the lack of three-dimensional structures of collagen glycosyltransferase enzymes hinders our understanding of the catalytic mechanisms producing this modification, as well as the impact of genetic mutations causing severe connective tissue pathologies. In this mini-review, we summarize the current knowledge on the biochemical features of the enzymes involved in the production of collagen glycosylations and the current state-of-the-art methods for the identification and characterization of this important PTM.

SARS-CoV-2 proximal origin is still unclear, limiting the possibility of foreseeing other spillover events with pandemic potential. Here we propose an evolutionary model based on the thorough dissection of SARS-CoV-2 and RaTG13 – the closest bat relative – spike dynamics, kinetics and binding to ACE2. Our results indicate that both spikes share nearly identical, high affinities for Rhinolophus affinis bat and human ACE2, pointing out to negligible species barriers directly related to receptor binding. Also, SARS-CoV-2 spike shows a higher degree of dynamics and kinetics optimization that favors ACE2 engagement. Therefore, we devise an affinity-independent evolutionary process that likely took place in R. affinis bats and limits the eventual involvement of other animal species in initiating the pandemic to the role of vector.

The Netrin-1 receptor UNC5B is an axon guidance regulator that is also expressed in endothelial cells (ECs), where it finely controls developmental and tumor angiogenesis. In the absence of Netrin-1, UNC5B induces apoptosis that is blocked upon Netrin-1 binding. Here, we identify an UNC5B splicing isoform (called UNC5B-Δ8) expressed exclusively by ECs and generated through exon skipping by NOVA2, an alternative splicing factor regulating vascular development. We show that UNC5B-Δ8 is a constitutively pro-apoptotic splicing isoform insensitive to Netrin-1 and required for specific blood vessel development in an apoptosis-dependent manner. Like NOVA2, UNC5B-Δ8 is aberrantly expressed in colon cancer vasculature where its expression correlates with tumor angiogenesis and poor patient outcome. Collectively, our data identify a mechanism controlling UNC5B's necessary apoptotic function in ECs and suggest that the NOVA2/UNC5B circuit represents a post-transcriptional pathway regulating angiogenesis.

The use of immunomodulatory agents for the treatment of cancer is gaining a growing biopharmaceutical interest. Antibody-cytokine fusion proteins, namely immunocytokines, represent a promising solution for the regulation of the immune system at the site of disease. The three-dimensional arrangement of these molecules can profoundly influence their biological activity and pharmacokinetic properties. Structural techniques might provide important insight in the 3D arrangement of immunocytokines. Here, we performed structure investigations on clinical grade fusion proteins L19-IL2, IL12-L19L19 and L19L19-IL2 to elucidate their quaternary organization. Crystallographic characterization of the common L19 antibody fragment at a resolution of 2.0-Å was combined with low-resolution studies of the full-length chimeric molecules using small-angle synchrotron X-ray scattering (SAXS) and negative stain electron microscopy. Characterization of the full-length quaternary structures of the immunocytokines in solution by SAXS consistently supported the diabody structure in the L19-IL2 immunocytokine and allowed generation of low-resolution models of the chimeric proteins L19L19-IL2 and IL12-L19L19. Comparison with 3D reconstructions obtained from negative-stain electron microscopy revealed marked flexibility associated to the linker regions connecting the cytokine and the antibody components of the chimeric proteins. Collectively, our results indicate that low-resolution molecular structure characterizations provide useful complementary insights for the quality control of immunocytokines, constituting a powerful tool to guide the design and the subsequent optimization steps towards clinical enhancement of these chimeric protein reagents.

AbstractProcollagen lysyl hydroxylases and glycosyltransferases (LH, also known as procollagen lysyl‐2‐oxoglutarate dioxygenases (PLOD)) are essential biosynthesis enzymes present in all collagen‐containing organisms, from sponges to humans. Higher vertebrates present three separatePLODgenes encoding for distinct enzyme isoforms (LH1, LH2a/b, and LH3), sharing ∼70% amino acid sequence identity. The LH1 and LH2 isoforms exclusively display Fe2+, 2‐oxoglutarate‐dependent lysyl 5‐hydroxylase activity, whereas LH3 is a multifunctional enzyme, able to further catalyze the Mn2+‐dependent β‐(1,O)‐galactosylation and the subsequent α‐(1,2)‐glucosylation of 5‐hydroxylysines. Despite exclusive selectivity for lysine residues within collagenous polypeptides, little is known about the specificity of LH enzymes for different amino acid sequences in different collagen types: LH1 and LH3 isoforms act on collagen triple‐helical regions, whereas the LH2 isoform specifically hydroxylates collagen telopeptides, yet no consensus sequences, nor minimum sequence lengths, have been proposed as requirements for catalysis. Available crystal structures of full‐length human LH3 show an elongated homodimeric quaternary structure, with three aligned domains constituting each enzyme's polypeptide: the N‐terminal glycosyltransferase (GT) domain, a central noncatalytic accessory (AC) domain, and a C‐terminal lysyl hydroxylase (LH) domain. Dimerization occurs in the C‐terminal domain, in proximity to the LH catalytic site. Dimerization is indeed essential for LH activity, but is dispensable for the glycosyltransferase activities of LH3.