SeqPredNN: a neural network that generates protein sequences that fold into specified tertiary structures.

BACKGROUND: The relationship between the sequence of a protein, its structure, and the resulting connection between its structure and function, is a foundational principle in biological science. Only recently has the computational prediction of protein structure based only on protein sequence been addressed effectively by AlphaFold, a neural network approach that can predict the majority of protein structures with X-ray crystallographic accuracy. A question that is now of acute relevance is the "inverse protein folding problem": predicting the sequence of a protein that folds into a specified structure. This will be of immense value in protein engineering and biotechnology, and will allow the design and expression of recombinant proteins that can, for instance, fold into specified structures as a scaffold for the attachment of recombinant antigens, or enzymes with modified or novel catalytic activities. Here we describe the development of SeqPredNN, a feed-forward neural network trained with X-ray crystallographic structures from the RCSB Protein Data Bank to predict the identity of amino acids in a protein structure using only the relative positions, orientations, and backbone dihedral angles of nearby residues. RESULTS: We predict the sequence of a protein expected to fold into a specified structure and assess the accuracy of the prediction using both AlphaFold and RoseTTAFold to computationally generate the fold of the derived sequence. We show that the sequences predicted by SeqPredNN fold into a structure with a median TM-score of 0.638 when compared to the crystal structure according to AlphaFold predictions, yet these sequences are unique and only 28.4% identical to the sequence of the crystallized protein. CONCLUSIONS: We propose that SeqPredNN will be a valuable tool to generate proteins of defined structure for the design of novel biomaterials, pharmaceuticals, catalysts, and reporter systems. The low sequence identity of its predictions compared to the native sequence could prove useful for developing proteins with modified physical properties, such as water solubility and thermal stability. The speed and ease of use of SeqPredNN offers a significant advantage over physics-based protein design methods.

Fat Grafts Show Higher Hypoxia, Angiogenesis, Adipocyte Proliferation, and Macrophage Infiltration than Flaps in a Pilot Mouse Study.

BACKGROUND: Over 137,000 breast reconstructions are performed annually by American Society of Plastic Surgeons (ASPS) members. Vascularized flaps and avascular lipofilling each account for over 33,000 autologous reconstructions. Although clinical and experimental observations suggest biologic differences with diverging effects on locoregional tumor control, comparative animal models are lacking. The authors standardized existing techniques in immunocompetent mice, laying the foundation for in vivo models of autologous breast reconstruction combinable with orthotopic tumor implantations. METHODS: Twenty-five groin flaps and 39 fat grafts were transferred in female BALB/c-mice. Adipocytes were tracked via Hoechst-Calcein-DiI staining ( n = 2 per group), and postoperative volume retentions were compared via magnetic resonance imaging ( n = 3 per group) on days 1, 11, 21, and 31. Proliferation indices, microvessel densities, tissue hypoxia, and macrophage infiltrates were compared via Ki67, CD31, pimonidazole, and hematoxylin-eosin staining on days 5, 10, 15, 20, and 30 ( n = 4 per group). RESULTS: Viable adipocytes were present in both groups. Graft volumes plateaued at 42.7 ± 1.2% versus 81.8 ± 4.0% of flaps ( P < 0.001). Initially, grafts contained more hypoxic cells (day 5: 15.192 ± 1.249 versus 1.157 ± 192; P < 0.001), followed by higher proliferation (day 15: 25.2 ± 1.0% versus 0.0 ± 0.0%; P < 0.001), higher microvessel numbers (day 30: 307.0 ± 13.2 versus 178.0 ± 10.6; P < 0.001), and more pronounced macrophage infiltrates (graded 3 versus 2; P < 0.01). CONCLUSION: This comparative murine pilot study of vascularized flaps versus avascular lipofilling suggests differences in volume retention, proliferation, angiogenesis, hypoxia, and inflammation. CLINICAL RELEVANCE STATEMENT: The biological differences of fat grafting versus flap transfer are not fully understood because no single comparative experimental model has been established to date. The authors present the first comparative small animal model of both techniques, which will allow the gaining of deeper insights into their biological effects.

Sulfated hyaluronic acid inhibits the hyaluronidase CEMIP and regulates the HA metabolism, proliferation and differentiation of fibroblasts.

Hyaluronan (HA) is an extracellular matrix component that regulates a variety of physiological and pathological processes. The function of HA depends both on its overall amount and on its size, properties that are controlled by HA synthesizing and degrading enzymes. The lack of inhibitors that can specifically block individual HA degrading enzymes has hampered attempts to understand the contribution of individual hyaluronidases to different physiological and pathological processes. CEMIP is a recently discovered hyaluronidase that cleaves HA through mechanisms and under conditions that are distinct from those of other hyaluronidases such as HYAL1 or HYAL2. The role of its hyaluronidase activity in physiology and disease is poorly understood. Here, we characterized a series of sulfated HA derivatives (sHA) with different sizes and degrees of sulfation for their ability to inhibit specific hyaluronidases. We found that highly sulfated sHA derivatives potently inhibited CEMIP hyaluronidase activity. One of these compounds, designated here as sHA3.7, was characterized further and shown to inhibit CEMIP with considerable selectivity over other hyaluronidases. Inhibition of CEMIP with sHA3.7 in fibroblasts, which are the main producers of HA in the interstitial matrix, increased the cellular levels of total and high molecular weight HA, while decreasing the fraction of low molecular weight HA fragments. Genetic deletion of CEMIP in mouse embryonic fibroblasts (MEFs) produced analogous results and confirmed that the effects of sHA3.7 on HA levels were mediated by CEMIP inhibition. Importantly, both CEMIP deletion and its inhibition by sHA3.7 suppressed fibroblast proliferation, while promoting differentiation into myofibroblasts, as reflected in a lack of CEMIP in myofibroblasts within skin wounds in experimental mice. By contrast, adipogenic and osteogenic differentiation were attenuated upon CEMIP loss or inhibition. Our results demonstrate the importance of CEMIP for the HA metabolism, proliferation and differentiation of fibroblasts, and suggest that inhibition of CEMIP with sulfated HA derivatives such as sHA3.7 has potential utility in pathological conditions that are dependent on CEMIP function.

Loss of ASAP1 in the MMTV-PyMT model of luminal breast cancer activates AKT, accelerates tumorigenesis, and promotes metastasis.

ASAP1 is a multi-domain adaptor protein that regulates cytoskeletal dynamics, receptor recycling and intracellular vesicle trafficking. Its expression is associated with poor prognosis in a variety of cancers, and can promote cell migration, invasion and metastasis. Although amplification and expression of ASAP1 has been associated with poor survival in breast cancer, we found that in the autochthonous MMTV-PyMT model of luminal breast cancer, ablation of ASAP1 resulted in an earlier onset of tumor initiation and increased metastasis. This was due to tumor cell-intrinsic effects of ASAP1 deletion, as ASAP1 deficiency in tumor, but not in stromal cells was sufficient to replicate the enhanced tumorigenicity and metastasis observed in the ASAP1-null MMTV-PyMT mice. Loss of ASAP1 in MMTV-PyMT mice had no effect on proliferation, apoptosis, angiogenesis or immune cell infiltration, but enhanced mammary gland hyperplasia and tumor cell invasion, indicating that ASAP1 can accelerate tumor initiation and promote dissemination. Mechanistically, these effects were associated with a potent activation of AKT. Importantly, lower ASAP1 levels correlated with poor prognosis and enhanced AKT activation in human ER+/luminal breast tumors, validating our findings in the MMTV-PyMT mouse model for this subtype of breast cancer. Taken together, our findings reveal that ASAP1 can have distinct functions in different tumor types and demonstrate a tumor suppressive activity for ASAP1 in luminal breast cancer.

Loss of ASAP1 in mice impairs adipogenic and osteogenic differentiation of mesenchymal progenitor cells through dysregulation of FAK/Src and AKT signaling.

ASAP1 is a multi-domain adaptor protein that regulates cytoskeletal dynamics, receptor recycling and intracellular vesicle trafficking. Its expression is associated with poor prognosis for a variety of cancers, and promotes cell migration, invasion and metastasis. Little is known about its physiological role. In this study, we used mice with a gene-trap inactivated ASAP1 locus to study the functional role of ASAP1 in vivo, and found defects in tissues derived from mesenchymal progenitor cells. Loss of ASAP1 led to growth retardation and delayed ossification typified by enlarged hypertrophic zones in growth plates and disorganized chondro-osseous junctions. Furthermore, loss of ASAP1 led to delayed adipocyte development and reduced fat depot formation. Consistently, deletion of ASAP1 resulted in accelerated chondrogenic differentiation of mesenchymal cells in vitro, but suppressed osteo- and adipogenic differentiation. Mechanistically, we found that FAK/Src and PI3K/AKT signaling is compromised in Asap1GT/GT MEFs, leading to impaired adipogenic differentiation. Dysregulated FAK/Src and PI3K/AKT signaling is also associated with attenuated osteogenic differentiation. Together these observations suggest that ASAP1 plays a decisive role during the differentiation of mesenchymal progenitor cells.

Autochthonous mouse melanoma and mammary tumors do not express the pluripotency genes Oct4 and Nanog.

The homeodomain transcription factors Oct4 and Nanog maintain pluripotency and self-renewal in embryonic stem cells. In somatic cells, inappropriate expression of these genes has been associated with loss of differentiation, malignant transformation, and the acquisition of cancer stem cell-like properties. As cancer stem cells have been suggested to underlie the growth and malignancy of tumors, Oct4 and Nanog may represent therapeutic targets. Their expression could also act as a marker of the cancer stem cell population, permitting its isolation and characterisation. Nevertheless, the existence of multiple pseudogenes and isoforms of these genes has complicated the interpretation of the data that supports a role for Oct4 and Nanog in the cancer context. Here we addressed this issue using knockin mice in which IRES elements are used to allow GFP expression under the control of the endogenous Oct4 or Nanog promoters, while maintaining correct expression of the Oct4 or Nanog gene. These mice were crossed with MT/ret mice that develop melanomas, and with MMTV-PyMT mice and MMTV-Neu mice that develop mammary adenocarcinomas. We analysed the tumors that developed in these compound mice for GFP expression. In this way we could assess transcription of Oct4 and Nanog in autochthonous cancers without the complication of factors such as pseudogene expression, alternative splicing and antibody specificity. Both the Oct4 and Nanog knockin tumor-bearing mice expressed GFP in blastocysts and testes as expected. However, we could find no evidence for expression of the GFP reporter above background levels in tumors using FACS, qPCR and immunohistochemistry. Furthermore, cultivation of Oct4GFP and NanogGFP MMTV-PyMT tumor cells either adherently or as spheroids had no effect on the expression of the GFP reporter. Together these data suggest that Oct4 and Nanog are not expressed in tumor cells that arise in the autochthonous cancer models studied here.

Tumor-initiating properties of breast cancer and melanoma cells in vivo are not invariably reflected by spheroid formation in vitro, but can be increased by long-term culturing as adherent monolayers.

Cancer stem cells (CSCs) have been studied intensively in recent years due to their potential importance for understanding neoplastic disease and the design of antitumor therapies. A number of properties attributed to CSCs have been used to define the CSC population, the most important of which is the ability to initiate reproducibly the growth of tumors in vivo. Other assays such as spheroid formation, expression of particular markers and label retention are also used for defining CSCs, although the degree to which these assays invariably reflect the ability to form tumors in vivo remains to be carefully evaluated. Given the importance of correctly defining and isolating CSCs if valid conclusions about their characteristics are to be made, we used syngeneic animal models to compare these different assays. In standard spheroid assays, cell aggregation rather than spheroid growth from single cell suspensions ensued, but aggregation was circumvented by the inclusion of methylcellulose in the medium. Label-retaining subpopulations did not reliably exhibit an enhanced ability to form spheroids and were enriched for senescent cells. Spheroid-forming ability was found to correspond to expression of established CSC markers, although not invariably. Furthermore, spheroid-forming ability was not always reflected in tumor-initiating properties in vivo. Long-term culture of primary mammary tumor cells as adherent monolayers increased their tumor-initiating ability in vivo. This increase was attenuated when the cells were subsequently cultivated as spheroids. Together these data indicate that assays that are widely used to define CSC subpopulations do not invariably reflect tumor-initiating properties in vivo.

Genes involved in the metastatic cascade of medullary thyroid tumours.

The process of how a benign tumour turns invasive and capable to survive in distant organs remains poorly understood, despite the evidence that metastasis formation is the primary cause of cancer patient mortality. This ignorance is partly due to the lack of appropriate animal models from which to investigate this complex process. The retinoblastoma (Rb) tumour suppressor pathway (pRb/E2F) is mutated in almost all human tumours, and a number of laboratories have now established pRb- or E2F-deficient mouse models. Consistent with the role of mutation in retinoblastoma in cancer biology, Rb heterozygous mice are prone to develop tumours. Among the ensuing tumours, the medullary thyroid carcinomas (MTCs) have a lessened tendency to form secondary cancers and metastases. Intriguingly, if an E2f3 mutation is introduced in this genetic background, more aggressive MTCs develop, which metastasize more frequently. Gene chip microarrays, however, provide an unbiased approach for examining the genome-wide expression levels and enable identification of a large set of metastasis-enriched gene sets. The identified genes may simply represent putative markers of the disease stage. Alternatively, genes may be identified that causally determine a link to the onset of metastasis. We describe the use of gene chip microarrays for identification of putative markers enriched in metastatic mouse MTCs. The chapter details how the most promising candidates are verified using additional methods, such as quantitative real-time PCR. In this case, co-transfection of the E2F-transcription factor using a heterologous reporter gene system is suggestive of E2Fs directly regulating putative metastasis markers.

Discovery of a novel tumour metastasis-promoting gene, NVM-1.

We have previously reported that over-expression of a panel of 119 genes correlates with the metastatic potential of pancreatic carcinoma cells. We sought to identify and functionally characterize candidate tumour metastasis promoting genes among this library using a secondary phenotype-assisted screen. Here we report the discovery of the metastasis-promoting function of a hitherto not characterized gene located on chromosome 14 (ORF138), which we have named 'novel metastasis-promoting gene 1' (NVM-1). The NVM-1 transcript is extensively alternatively spliced, is expressed endogenously in a number of different tissues, and is strongly over-expressed at the protein level in a variety of human tumour types. Importantly, NVM-1 expression stimulates the migratory and invasive behaviour of tumour cells and promotes metastasis formation in experimental animals in vivo. Up-regulation of FMNL2 and MT1E and down-regulation of TIMP4 and MHC-I is observed as a consequence of NVM-1 expression. Together these data identify NVM-1 as a gene that is functionally involved in tumour metastasis, and suggest that NVM-1 may constitute a promising therapeutic target for inhibition of tumour metastasis.