Abstract Diffuse midline glioma (DMG) is a universally fatal pediatric high-grade glioma characterized by a cold tumor immune microenvironment (TIME). While the extracellular matrix (ECM) is increasingly recognized as a key regulator of tumor–immune interactions and therapeutic responses in glioma, DMG-associated ECM proteins remain poorly defined. Here, we characterize tumor-derived components of the DMG ECM to identify regulators of immunosuppression and therapeutic resistance. Quantitative membrane-enriched proteomics was performed on three patient-derived DMG cell lines (SU-DIPG-IV, SU-DIPG-VI, and SU-DIPG-XII) and compared with the blood–brain barrier endothelial control cell line hCMEC/D3. Here, we report the identification of the ECM constituent termed Protein-J4 (P-J4) as the most significantly upregulated ECM protein in DMG (log2FC = 3.95, p<0.0001), exceeding the differential expression of established therapeutic membrane and extracellular targets including B7-H3, IL-13Rα2, and PDGFRA. RNA-sequencing data from the Children’s Brain Tumor Network and St Jude PeCan databases detected increased P-J4 expression across multiple pediatric brain and solid malignancies, with high expression in pediatric high-grade gliomas correlating with significantly worse overall survival (Kaplan–Meier, p=0.0379). Consistent with its role in fetal tissue development and ECM-associated growth factor bioavailability, including pro–TGF-β, expression of P-J4 may drive immune exclusion and negatively influence patient outcomes in DMG. To functionally interrogate these findings, we are establishing a P-J4 knockout DMG in vitro model and will conduct murine in vivo studies alongside spatial transcriptomics and ATAC-sequencing to define immune suppression and treatment resistance pathways. In parallel, we are currently characterizing a cohort of pediatric brain tumors (n = 120) and relevant healthy tissue controls for P-J4 expression and correlating with immune infiltration and immunosuppression signatures using our novel anti-P-J4 humanized antibodies. In conclusion, we have identified P-J4 as a translationally relevant and targetable ECM-constituent in DMG, with a potential role in diffuse growth, infiltration, immunosuppression and therapeutic resistance.

AMACO (VWA2 protein), secreted by epithelial cells, is strongly expressed at basement membranes when budding or invagination occurs in embryos. In skin, AMACO associates with proteins of the Fraser complex, which form anchoring cords. These, during development, temporally stabilize the dermal–epidermal junction, pending the formation of collagen VII-containing anchoring fibrils. Fraser syndrome in humans results if any of the core members of the Fraser complex (Fras1, Frem1, Frem2) are mutated. Fraser syndrome is characterized by subepidermal blistering, cryptophthalmos, and syndactyly. In an attempt to determine AMACO function, we generated and characterized AMACO-deficient mice. In contrast to Fraser complex mutant mice, AMACO-deficient animals lack an obvious phenotype. The mutually interdependent basement membrane deposition of the Fraser complex proteins, and the formation of anchoring cords, are not affected. Furthermore, hair follicle development in newborn AMACO-deficient mice showed no gross aberration. Surprisingly, it appears that, while AMACO is a component of the anchoring cords, it is not essential for their formation or function.

COMP (Cartilage Oligomeric Matrix Protein), also named thrombospondin-5, is a member of the thrombospondin family of extracellular matrix proteins. It is of clinical relevance, as in humans mutations in COMP lead to chondrodysplasias. The gene encoding zebrafish Comp is located on chromosome 11 in synteny with its mammalian orthologs. Zebrafish Comp has a domain structure identical to that of tetrapod COMP and shares 74% sequence similarity with murine COMP. Zebrafish comp is expressed from 5 hours post fertilization (hpf) on, while the protein is first detectable in somites of 11 hpf embryos. During development and in adults comp is strongly expressed in myosepta, craniofacial tendon and ligaments, around ribs and vertebra, but not in its name-giving tissue cartilage. As in mammals, zebrafish Comp forms pentamers. It is easily extracted from 5 days post fertilization (dpf) whole zebrafish. The lack of Comp expression in zebrafish cartilage implies that its cartilage function evolved recently in tetrapods. The expression in tendon and myosepta may indicate a more fundamental function, as in evolutionary distant Drosophila muscle-specific adhesion to tendon cells requires thrombospondin. A sequence encoding a calcium binding motif within the first TSP type-3 repeat of zebrafish Comp was targeted by CRISPR-Cas. The heterozygous and homozygous mutant Comp zebrafish displayed a patchy irregular Comp staining in 3 dpf myosepta, indicating a dominant phenotype. Electron microscopy revealed that the endoplasmic reticulum of myosepta fibroblasts is not affected in homozygous fish. The disorganized extracellular matrix may indicate that this mutation rather interferes with extracellular matrix assembly, similar to what is seen in a subgroup of chondrodysplasia patients. The early expression and easy detection of mutant Comp in zebrafish points to the potential of using the zebrafish model for large scale screening of small molecules that can improve secretion or function of disease-associated COMP mutants.

LTBP1 is a large extracellular matrix protein and an associated ligand of fibrillinmicrofibrils. Knowledge of LTBP1 functions is largely limited to its role in targeting and sequestering TGFβ growth factors within the extracellular matrix, thereby regulating their bioavailability. However, the recent description of a wide spectrum of phenotypes in multiple tissues in patients harboring LTBP1 pathogenic variants suggests a multifaceted role of the protein in the homeostasis of connective tissues. To better understand the human pathology caused by LTBP1 deficiency it is important to investigate its functional role in extracellular matrix formation. In this study, we show that LTBP1 coordinates the incorporation of fibrillin-1 and −2 into the extracellular matrix in vitro . We also demonstrate that this function is differentially exerted by the two isoforms, the short and long forms of LTBP1. Thereby our findings uncover a novel TGFβ-independent LTBP1 function potentially contributing to the development of connective tissue disorders.

AMACO (VWA2 protein) is a basement membrane-associated protein secreted by epithelial cells. It is strongly expressed when invagination or budding occurs during development. AMACO associates with the Fraser complex, which when mutated causes Fraser syndrome, characterized by subepidermal blistering, cryptoph-thalmos, and syndactyly. The core Fraser complex proteins FRAS1, FREM1, and FREM2 localize at the dermal-epidermal junction and mediate adhesion to the underlying dermis during embryonic development. Earlier transmission electron microscopy studies of adult mouse skin showed clustered AMACO deposition below the lamina densa. In this study, we report a distinct cord-like suprastructure in the neonate dermis to which AMACO- and Fraser complex-associated proteins contribute. We propose anchoring cords to designate the suprastructure. Anchoring cords have a diameter of 60 nm when immunolabeled, originate from the basement membrane, and extend several microns into the dermis. In normal skin, they are evident after immunogold electron microscopy and are strikingly appreciated in thicker sections. In recessive dystrophic epidermolysis bullosa skin, they are directly visible where collagen VII anchoring fibrils are ablated. Immunofluorescence and coimmunoprecipitation of skin extracts identify a direct interaction of FREM2 and AMACO.

Diffuse midline glioma (DMG) is a high-grade glioma with a median overall survival of 9-11 months. Radiotherapy is the only recognized treatment. The DMG tumor microenvironment (TME) contains few, if any, tumor infiltrating lymphocytes (TILs) or inflammatory cytokines, thus is distinctive of an ‘immunologically cold’ tumor/TME.1 DMG lack the expression of immunosuppressive immune checkpoint proteins, likely explaining the failure of immune checkpoint inhibitors (ICI) tested under clinical trials for DMG patients, and suggestive of an alternative mechanism underpinning the cold TME. 1 Glioblastomas also harbor a cold TME, which can be somewhat explained by T cell lymphopenia, influenced by the sequestration of T cells in the bone marrow (through Beta-arrestin-induced internalization of Sphingosine-1-phosphate receptor 1 [S1PR1]). 2 Dopaminergic activation of Beta-arrestin and hence S1PR1 internalization, is potentially regulated through dopaminergic peripheral nerves in primary and secondary lymphoid organs, regulated by the Dopamine receptor D2 (DRD2), that is highly expressed on T cells. ONC201 is a potent DRD2 antagonist, currently in phase I-III clinical trials for DMG patients, alone and in combination with radiotherapy and the PI3K/AKT inhibitor paxalisib (NCT05009992). Proteomic profiling of DMG patient-derived cells +/-ONC201 showed increased expression of several antigen presenting pathway proteins, including Beta-2-microglobulin (B2M) and HLA class I histocompatibility antigen, A alpha chain (HLA-A). This was confirmed in vivo using SU-DIPG-VI patient-derived xenograft mouse model tissues +/-ONC201 alone, and together with paxalisib. Excitingly, this combination (given orally) promoted the recruitment of TILs to the tumor, revealing novel immunomodulatory effects. In vivo, ONC201 promoted the expression of EMILIN-3, a TGF-β antagonist that is known to inhibit HLA-A/B2M expression, possibly explaining the increased MHC-I activity. This study uncovers a novel link between treatment of DMG with ONC201 and paxalisib and the role dopaminergic peripheral nerves signaling may play on the sequestration of T cells within lymphoid organs and lymphopenia.

The monoclonal antibody ER-TR7 was used in a great number of studies for detecting reticular fibroblasts and the ECM of lymphoid and non-lymphoid organs even if the protein recognized by the ER-TR7 antibody was not known. We have now identified native collagen VI microfibrils as its tissue antigen.

The monoclonal antibody ER-TR7 was used in a great number of studies for detecting reticular fibroblasts and the ECM of lymphoid and non-lymphoid organs even if the protein recognized by the ER-TR7 antibody was not known. We have now identified native collagen VI microfibrils as its tissue antigen.

Prolyl 3-hydroxylation is a rare collagen type I post translational modification in fibrillar collagens. The primary 3Hyp substrate sites in type I collagen are targeted by an endoplasmic reticulum (ER) complex composed by cartilage associated protein (CRTAP), prolyl 3-hydroxylase 1 (P3H1) and prolyl cis/trans isomerase B, whose mutations cause recessive forms of osteogenesis imperfecta with impaired levels of α1(I)3Hyp986. The absence of collagen type I 3Hyp in wild type zebrafish provides the unique opportunity to clarify the role of the complex in vertebrate. Zebrafish knock outs for crtap and p3h1 were generated by CRISPR/Cas9. Mutant fish have the typical OI patients' reduced size, body disproportion and altered mineralization. Vertebral body fusions, deformities and fractures are accompanied to reduced size, thickness and bone volume. Intracellularly, collagen type I is overmodified, and partially retained causing enlarged ER cisternae. In the extracellular matrix the abnormal collagen type I assembles in disorganized fibers characterized by altered diameter. The data support the defective chaperone role of the 3-hydroxylation complex as the primary cause of the skeletal phenotype.

Congenital anomalies of the kidney and urinary tract (CAKUT) are the most common cause (40-50%) of chronic kidney disease (CKD) in children. About 40 monogenic causes of CAKUT have so far been discovered. To date less than 20% of CAKUT cases can be explained by mutations in these 40 genes. To identify additional monogenic causes of CAKUT, we performed whole exome sequencing (WES) and homozygosity mapping (HM) in a patient with CAKUT from Indian origin and consanguineous descent. We identified a homozygous missense mutation (c.1336C>T, p.Arg446Cys) in the gene Von Willebrand factor A domain containing 2 (VWA2). With immunohistochemistry studies on kidneys of newborn (P1) mice, we show that Vwa2 and Fraser extracellular matrix complex subunit 1 (Fras1) co-localize in the nephrogenic zone of the renal cortex. We identified a pronounced expression of Vwa2 in the basement membrane of the ureteric bud (UB) and derivatives of the metanephric mesenchyme (MM). By applying in vitro assays, we demonstrate that the Arg446Cys mutation decreases translocation of monomeric VWA2 protein and increases translocation of aggregated VWA2 protein into the extracellular space. This is potentially due to the additional, unpaired cysteine residue in the mutated protein that is used for intermolecular disulfide bond formation. VWA2 is a known, direct interactor of FRAS1 of the Fraser-Complex (FC). FC-encoding genes and interacting proteins have previously been implicated in the pathogenesis of syndromic and/or isolated CAKUT phenotypes in humans. VWA2 therefore constitutes a very strong candidate in the search for novel CAKUT-causing genes. Our results from in vitro experiments indicate a dose-dependent neomorphic effect of the Arg446Cys homozygous mutation in VWA2.

Searchable abstracts of presentations at key conferences on calcified tissues ISSN 2052-1219 (online)

Collagen XXVIII is the last discovered member of the collagen superfamily and thus has been only sparsely investigated. We studied collagen XXVIII in zebrafish to gain insight into its structure, evolution and expression. In contrast to human and mouse, the zebrafish genome contains four collagen XXVIII genes, col28a1a and -b, and col28a2a and -b. Genomic context and phylogenetic analysis revealed that the a2 branch was lost during evolution of mammals, whereas the duplication of the a1 and a2 branches results from the whole genome duplication in the teleost lineage. Sequence analysis revealed conservation of domain structure and the unique imperfections in the triple helical domain. Two major forms of collagen XXVIII were identified, Col28a1b in adult and Col28a2a in 3–5dpf zebrafish. Composite agarose/polyacrylamide gel electrophoresis revealed that both these chains mainly form dimers of trimers, although Col28a1b appears to be more polydisperse. Homodimers are abundant, although it is possible that complexes consisting of Col28a2a and Col28a1a or -a2b occur. Peptide mass fingerprint analysis revealed that the C-terminal Kunitz domain is often proteolytically processed. In contrast to murine collagen XXVIII, the zebrafish orthologs are widely expressed and not only present in the nervous system. They are differentially expressed in the liver, thymus, muscle, intestine and skin. Altogether our results point to a unique nature of collagen XXVIII within the collagen family.

Fraser syndrome (FS) is a phenotypically variable, autosomal recessive disorder characterized by cryptophthalmus, cutaneous syndactyly, and other malformations resulting from mutations in FRAS1, FREM2, and GRIP1. Transient embryonic epidermal blistering causes the characteristic defects of the disorder. Fras1, Frem1, and Frem2 form the extracellular Fraser complex, which is believed to stabilize the basement membrane. However, several cases of FS could not be attributed to mutations in FRAS1, FREM2, or GRIP1, and FS displays high clinical variability, suggesting that there is an additional genetic, possibly modifying contribution to this disorder. An extracellular matrix protein containing VWA-like domains related to those in matrilins and collagens (AMACO), encoded by the VWA2 gene, has a very similar tissue distribution to the Fraser complex proteins in both mouse and zebrafish. Here, we show that AMACO deposition is lost in Fras1-deficient zebrafish and mice and that Fras1 and AMACO interact directly via their chondroitin sulfate proteoglycan (CSPG) and P2 domains. Knockdown of vwa2, which alone causes no phenotype, enhances the phenotype of hypomorphic Fras1 mutant zebrafish. Together, our data suggest that AMACO represents a member of the Fraser complex.

ObjectiveWe previously demonstrated the ability of matrilin-3 to modulate the gene expression profile of primary human chondrocytes (PHCs) toward a state favoring cartilage catabolism. The structure within matrilin-3 responsible for the induction of these catabolic genes is unknown. Here, we investigated the potential of matrilin-3 (MATN3) and truncated matrilin-3 proteins, in both monomeric and oligomeric form, to stimulate interleukin (IL)-6 release in PHCs.MethodsWe expressed full-length matrilin-3 oligomers, matrilin-3 von Willebrand factor A (VWA) domain oligomers, matrilin-3 four epidermal growth factor (EGF) domain oligomers, matrilin-3 monomers without oligomerization domains, matrilin-3 VWA domain monomers, and matrilin-3 4EGF monomers. We then incubated PHCs in the absence or presence of full-length matrilin-3 or one of the truncated matrilin-3 proteins and finally determined the release of IL-6 in cell-culture supernatants.ResultsThe addition of full-length matrilin-3 oligomers, matrilin-3 VWA domain oligomers, and, less pronounced, matrilin-3 monomers without oligomerization domains, and matrilin-3 4EGF-oligomers to the cell-culture medium led to a significant induction of IL-6 in PHCs.DiscussionBased on recombinant expression of different matrilin-3 domains in both monomeric and oligomeric form, this work demonstrated that the VWA1 domain of matrilin-3 is primarily responsible for the induction of IL-6 release and that the oligomerization of the VWA1 domain markedly promotes its activity.

In a recent study the DVWA gene located on human chromosome 3p24.3 was identified as a susceptibility locus for knee osteoarthritis in Japanese and Chinese patients (Miyamoto, Y., Shi, D., Nakajima, M., Ozaki, K., Sudo, A., Kotani, A., Uchida, A., Tanaka, T., Fukui, N., Tsunoda, T., Takahashi, A., Nakamura, Y., Jiang, Q., Ikegawa, S., 2008. Common variants in DVWA on chromosome 3p24.3 are associated with susceptibility to knee osteoarthritis. Nat. Genet. 40, 994–998). The authors concluded that DVWA codes for a novel protein containing two von Willebrand factor A (VWA) domains without a signal peptide sequence. The experimental data provided in this interesting study led to the suggestion of a mechanism for the etiology of the disease, based on an interaction between DVWA protein and β-tubulin. More recently, no significant association between DVWA and osteoarthritis was found in UK patient samples (Valdes, A.M., Spector, T.D., Doherty, S., Wheeler, M., Hart, D.J., Doherty, M., 2008. Association of the DVWA and GDF5 polymorphisms with osteoarthritis in UK populations. Ann. Rheum. Dis. Dec 3. [Epub ahead of print]), but a meta-analyses with data from individuals of white European descent from the Netherlands, the UK, Spain and Greece and the original Japanese and Chinese cohort provided evidence for a global association of one of the polymorphisms, a cysteine to tyrosine exchange (rs7639618) (Meulenbelt, I., Chapman, K., Dieguez-Gonzalez, R., Shi, D., Tsezou, A., Dai, J., Malizos, K.N., Kloppenburg, M., Carr, A., Nakajima, M., van der Breggen, R., Lakenberg, N., Gomez-Reino, J.J., Jiang, Q., Ikegawa, S., Gonzalez, A., Loughlin, J., Slagboom, E.P., 2009. Large replication study and meta-analyses of DVWA as an osteoarthritis susceptibility locus in European and Asian populations. Hum. Mol. Genet. 8, 1518–1523). However, there was no independent association with knee osteoarthritis in Europeans. Here we present information that the newly identified DVWA represents the human gene coding for the collagen VI α4 chain, which could point to a more complex disease mechanism.