A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories.

Pancreatic cancer is an aggressive disease that typically presents late with poor outcomes, indicating a pronounced need for early detection. In this study, we applied artificial intelligence methods to clinical data from 6 million patients (24,000 pancreatic cancer cases) in Denmark (Danish National Patient Registry (DNPR)) and from 3 million patients (3,900 cases) in the United States (US Veterans Affairs (US-VA)). We trained machine learning models on the sequence of disease codes in clinical histories and tested prediction of cancer occurrence within incremental time windows (CancerRiskNet). For cancer occurrence within 36 months, the performance of the best DNPR model has area under the receiver operating characteristic (AUROC) curve = 0.88 and decreases to AUROC (3m) = 0.83 when disease events within 3 months before cancer diagnosis are excluded from training, with an estimated relative risk of 59 for 1,000 highest-risk patients older than age 50 years. Cross-application of the Danish model to US-VA data had lower performance (AUROC = 0.71), and retraining was needed to improve performance (AUROC = 0.78, AUROC (3m) = 0.76). These results improve the ability to design realistic surveillance programs for patients at elevated risk, potentially benefiting lifespan and quality of life by early detection of this aggressive cancer.

Molecular response to PARP1 inhibition in ovarian cancer cells as determined by mass spectrometry based proteomics.

BACKGROUND: Poly (ADP)-ribose polymerase (PARP) inhibitors have entered routine clinical practice for the treatment of high-grade serous ovarian cancer (HGSOC), yet the molecular mechanisms underlying treatment response to PARP1 inhibition (PARP1i) are not fully understood. METHODS: Here, we used unbiased mass spectrometry based proteomics with data-driven protein network analysis to systematically characterize how HGSOC cells respond to PARP1i treatment. RESULTS: We found that PARP1i leads to pronounced proteomic changes in a diverse set of cellular processes in HGSOC cancer cells, consistent with transcript changes in an independent perturbation dataset. We interpret decreases in the levels of the pro-proliferative transcription factors SP1 and ß-catenin and in growth factor signaling as reflecting the anti-proliferative effect of PARP1i; and the strong activation of pro-survival processes NF-κB signaling and lipid metabolism as PARPi-induced adaptive resistance mechanisms. Based on these observations, we nominate several protein targets for therapeutic inhibition in combination with PARP1i. When tested experimentally, the combination of PARPi with an inhibitor of fatty acid synthase (TVB-2640) has a 3-fold synergistic effect and is therefore of particular pre-clinical interest. CONCLUSION: Our study improves the current understanding of PARP1 function, highlights the potential that the anti-tumor efficacy of PARP1i may not only rely on DNA damage repair mechanisms and informs on the rational design of PARP1i combination therapies in ovarian cancer.

Siglec-15 recognition of sialoglycans on tumor cell lines can occur independently of sialyl Tn antigen expression.

Siglec-15 is a conserved sialic acid-binding Ig-like lectin expressed on osteoclast progenitors, which plays an important role in osteoclast development and function. It is also expressed by tumor-associated macrophages and by some tumors, where it is thought to contribute to the immunosuppressive microenvironment. It was shown previously that engagement of macrophage-expressed Siglec-15 with tumor cells expressing its ligand, sialyl Tn (sTn), triggered production of TGF-ß. In the present study, we have further investigated the interaction between Siglec-15 and sTn on tumor cells and its functional consequences. Based on binding assays with lung and breast cancer cell lines and glycan-modified cells, we failed to see evidence for recognition of sTn by Siglec-15. However, using a microarray of diverse, structurally defined glycans, we show that Siglec-15 binds with higher avidity to sialylated glycans other than sTn or related antigen sequences. In addition, we were unable to demonstrate enhanced TGF-ß secretion following co-culture of Siglec-15-expressing monocytic cell lines with tumor cells expressing sTn or following Siglec-15 cross-linking with monoclonal antibodies. However, we did observe activation of the SYK/MAPK signaling pathway following antibody cross-linking of Siglec-15 that may modulate the functional activity of macrophages.

Towards general network architecture design criteria for negative gas adsorption transitions in ultraporous frameworks.

Switchable metal-organic frameworks (MOFs) have been proposed for various energy-related storage and separation applications, but the mechanistic understanding of adsorption-induced switching transitions is still at an early stage. Here we report critical design criteria for negative gas adsorption (NGA), a counterintuitive feature of pressure amplifying materials, hitherto uniquely observed in a highly porous framework compound (DUT-49). These criteria are derived by analysing the physical effects of micromechanics, pore size, interpenetration, adsorption enthalpies, and the pore filling mechanism using advanced in situ X-ray and neutron diffraction, NMR spectroscopy, and calorimetric techniques parallelised to adsorption for a series of six isoreticular networks. Aided by computational modelling, we identify DUT-50 as a new pressure amplifying material featuring distinct NGA transitions upon methane and argon adsorption. In situ neutron diffraction analysis of the methane (CD4) adsorption sites at 111 K supported by grand canonical Monte Carlo simulations reveals a sudden population of the largest mesopore to be the critical filling step initiating structural contraction and NGA. In contrast, interpenetration leads to framework stiffening and specific pore volume reduction, both factors effectively suppressing NGA transitions.

Crystallographic and electronic evolution of lanthanum strontium ferrite (La0.6Sr0.4FeO3-δ) thin film and bulk model systems during iron exsolution.

We study the changes in the crystallographic phases and in the chemical states during the iron exsolution process of lanthanum strontium ferrite (LSF, La0.6Sr0.4FeO3-δ). By using thin films of orthorhombic LSF, grown epitaxially on NaCl(001) and rhombohedral LSF powder, the materials gap is bridged. The orthorhombic material transforms into a fluorite structure after the exsolution has begun, which further hinders this process. For the powder material, by a combination of in situ core level spectroscopy and ex situ neutron diffraction, we could directly highlight differences in the Fe chemical nature between surface and bulk: whereas the bulk contains Fe(iv) in the fully oxidized state, the surface spectra can be described perfectly by the sole presence of Fe(iii). We also present corresponding magnetic and oxygen vacancy concentration data of reduced rhombohedral LSF that did not undergo a phase transformation to the cubic perovskite system based on neutron diffraction data.

Probing the canonicity of the Wnt/Wingless signaling pathway.

The hallmark of canonical Wnt signaling is the transcriptional induction of Wnt target genes by the beta-catenin/TCF complex. Several studies have proposed alternative interaction partners for beta-catenin or TCF, but the relevance of potential bifurcations in the distal Wnt pathway remains unclear. Here we study on a genome-wide scale the requirement for Armadillo (Arm, Drosophila beta-catenin) and Pangolin (Pan, Drosophila TCF) in the Wnt/Wingless(Wg)-induced transcriptional response of Drosophila Kc cells. Using somatic genetics, we demonstrate that both Arm and Pan are absolutely required for mediating activation and repression of target genes. Furthermore, by means of STARR-sequencing we identified Wnt/Wg-responsive enhancer elements and found that all responsive enhancers depend on Pan. Together, our results confirm the dogma of canonical Wnt/Wg signaling and argue against the existence of distal pathway branches in this system.

The bromodomain protein BRD4 regulates splicing during heat shock.

The cellular response to heat stress is an ancient and evolutionarily highly conserved defence mechanism characterised by the transcriptional up-regulation of cyto-protective genes and a partial inhibition of splicing. These features closely resemble the proteotoxic stress response during tumor development. The bromodomain protein BRD4 has been identified as an integral member of the oxidative stress as well as of the inflammatory response, mainly due to its role in the transcriptional regulation process. In addition, there are also several lines of evidence implicating BRD4 in the splicing process. Using RNA-sequencing we found a significant increase in splicing inhibition, in particular intron retentions (IR), following heat treatment in BRD4-depleted cells. This leads to a decrease of mRNA abundancy of the affected transcripts, most likely due to premature termination codons. Subsequent experiments revealed that BRD4 interacts with the heat shock factor 1 (HSF1) such that under heat stress BRD4 is recruited to nuclear stress bodies and non-coding SatIII RNA transcripts are up-regulated. These findings implicate BRD4 as an important regulator of splicing during heat stress. Our data which links BRD4 to the stress induced splicing process may provide novel mechanisms of BRD4 inhibitors in regard to anti-cancer therapies.

The intron-enriched HERV-K(HML-10) family suppresses apoptosis, an indicator of malignant transformation.

BACKGROUND: Human endogenous retroviruses (HERVs) constitute 8% of the human genome and contribute substantially to the transcriptome. HERVs have been shown to generate RNAs that modulate host gene expression. However, experimental evidence for an impact of these regulatory transcripts on the cellular phenotype has been lacking. RESULTS: We characterized the previously little described HERV-K(HML-10) endogenous retrovirus family on a genome-wide scale. HML-10 invaded the ancestral genome of Old World monkeys about 35 Million years ago and is enriched within introns of human genes when compared to other HERV families. We show that long terminal repeats (LTRs) of HML-10 exhibit variable promoter activity in human cancer cell lines. One identified HML-10 LTR-primed RNA was in opposite orientation to the pro-apoptotic Death-associated protein 3 (DAP3). In HeLa cells, experimental inactivation of HML-10 LTR-primed transcripts induced DAP3 expression levels, which led to apoptosis. CONCLUSIONS: Its enrichment within introns suggests that HML-10 may have been evolutionary co-opted for gene regulation more than other HERV families. We demonstrated such a regulatory activity for an HML-10 RNA that suppressed DAP3-mediated apoptosis in HeLa cells. Since HML-10 RNA appears to be upregulated in various tumor cell lines and primary tumor samples, it may contribute to evasion of apoptosis in malignant cells. However, the overall weak expression of HML-10 transcripts described here raises the question whether our result described for HeLa represent a rare event in cancer. A possible function in other cells or tissues requires further investigation.