SMILE: Cost-sensitive multi-task learning for nuclear segmentation and classification with imbalanced annotations.

High throughput nuclear segmentation and classification of whole slide images (WSIs) is crucial to biological analysis, clinical diagnosis and precision medicine. With the advances of CNN algorithms and the continuously growing datasets, considerable progress has been made in nuclear segmentation and classification. However, few works consider how to reasonably deal with nuclear heterogeneity in the following two aspects: imbalanced data distribution and diversified morphology characteristics. The minority classes might be dominated by the majority classes due to the imbalanced data distribution and the diversified morphology characteristics may lead to fragile segmentation results. In this study, a cost-Sensitive MultI-task LEarning (SMILE) framework is conducted to tackle the data heterogeneity problem. Based on the most popular multi-task learning backbone in nuclei segmentation and classification, we propose a multi-task correlation attention (MTCA) to perform feature interaction of multiple high relevant tasks to learn better feature representation. A cost-sensitive learning strategy is proposed to solve the imbalanced data distribution by increasing the penalization for the error classification of the minority classes. Furthermore, we propose a novel post-processing step based on the coarse-to-fine marker-controlled watershed scheme to alleviate fragile segmentation when nuclei are with large size and unclear contour. Extensive experiments show that the proposed method achieves state-of-the-art performances on CoNSeP and MoNuSAC 2020 datasets. The code is available at: https://github.com/panxipeng/nuclear_segandcls.

Development of selective mono or dual PROTAC degrader probe of CDK isoforms.

Cyclin-dependent kinase (CDK) family members are promising molecular targets in discovering potent inhibitors in disease settings, they function differentially. CDK2, CDK4 and CDK6, directly regulate the cell cycle, while CDK9 primarily modulates the transcription regulation. In discovering inhibitors of these CDKs, toxicity associated with off-target effect on other CDK homologs often posts as a clinical issue and hinders their further therapeutic development. To improve efficacy and reduce toxicity, here, using the Proteolysis Targeted Chimeras (PROTACs) approach, we design and further optimize small molecule degraders targeting multiple CDKs. We showed that heterobifunctional compound A9 selectively degraded CDK2. We also identified a dual-degrader, compound F3, which potently induced degradation of both CDK2 (DC50: 62 nM) and CDK9 (DC50: 33 nM). In human prostate cancer PC-3 cells, compound F3 potently inhibits cell proliferation by effectively blocking the cell cycle in S and G2/M phases. Our preliminary data suggests that PROTAC-oriented CDK2/9 degradation is potentially an effective therapeutic approach.

MKL1-actin pathway restricts chromatin accessibility and prevents mature pluripotency activation.

Actin cytoskeleton is well-known for providing structural/mechanical support, but whether and how it regulates chromatin and cell fate reprogramming is far less clear. Here, we report that MKL1, the key transcriptional co-activator of many actin cytoskeletal genes, regulates genomic accessibility and cell fate reprogramming. The MKL1-actin pathway weakens during somatic cell reprogramming by pluripotency transcription factors. Cells that reprogram efficiently display low endogenous MKL1 and inhibition of actin polymerization promotes mature pluripotency activation. Sustained MKL1 expression at a level seen in typical fibroblasts yields excessive actin cytoskeleton, decreases nuclear volume and reduces global chromatin accessibility, stalling cells on their trajectory toward mature pluripotency. In addition, the MKL1-actin imposed block of pluripotency can be bypassed, at least partially, when the Sun2-containing linker of the nucleoskeleton and cytoskeleton (LINC) complex is inhibited. Thus, we unveil a previously unappreciated aspect of control on chromatin and cell fate reprogramming exerted by the MKL1-actin pathway.

Single-cell microRNA-mRNA co-sequencing reveals non-genetic heterogeneity and mechanisms of microRNA regulation.

Measuring multiple omics profiles from the same single cell opens up the opportunity to decode molecular regulation that underlies intercellular heterogeneity in development and disease. Here, we present co-sequencing of microRNAs and mRNAs in the same single cell using a half-cell genomics approach. This method demonstrates good robustness (~95% success rate) and reproducibility (R2 = 0.93 for both microRNAs and mRNAs), yielding paired half-cell microRNA and mRNA profiles, which we can independently validate. By linking the level of microRNAs to the expression of predicted target mRNAs across 19 single cells that are phenotypically identical, we observe that the predicted targets are significantly anti-correlated with the variation of abundantly expressed microRNAs. This suggests that microRNA expression variability alone may lead to non-genetic cell-to-cell heterogeneity. Genome-scale analysis of paired microRNA-mRNA co-profiles further allows us to derive and validate regulatory relationships of cellular pathways controlling microRNA expression and intercellular variability.

Characterization of drug responses of mini patient-derived xenografts in mice for predicting cancer patient clinical therapeutic response.

BACKGROUND: Patient-derived organoids and xenografts (PDXs) have emerged as powerful models in functional diagnostics with high predictive power for anticancer drug response. However, limitations such as engraftment failure and time-consuming for establishing and expanding PDX models followed by testing drug efficacy, and inability to subject to systemic drug administration for ex vivo organoid culture hinder realistic and fast decision-making in selecting the right therapeutics in the clinic. The present study aimed to develop an advanced PDX model, namely MiniPDX, for rapidly testing drug efficacy to strengthen its value in personalized cancer treatment. METHODS: We developed a rapid in vivo drug sensitivity assay, OncoVee® MiniPDX, for screening clinically relevant regimens for cancer. In this model, patient-derived tumor cells were arrayed within hollow fiber capsules, implanted subcutaneously into mice and cultured for 7 days. The cellular activity morphology and pharmacokinetics were systematically evaluated. MiniPDX performance (sensitivity, specificity, positive and negative predictive values) was examined using PDX as the reference. Drug responses were examined by tumor cell growth inhibition rate and tumor growth inhibition rate in PDX models and MiniPDX assays respectively. The results from MiniPDX were also used to evaluate its predictive power for clinical outcomes. RESULTS: Morphological and histopathological features of tumor cells within the MiniPDX capsules matched those both in PDX models and in original tumors. Drug responses in the PDX tumor graft assays correlated well with those in the corresponding MiniPDX assays using 26 PDX models generated from patients, including 14 gastric cancer, 10 lung cancer and 2 pancreatic cancer. The positive predictive value of MiniPDX was 92%, and the negative predictive value was 81% with a sensitivity of 80% and a specificity of 93%. Through expanding to clinical tumor samples, MiniPDX assay showed potential of wide clinical application. CONCLUSIONS: Fast in vivo MiniPDX assay based on capsule implantation was developed-to assess drug responses of both PDX tumor grafts and clinical cancer specimens. The high correlation between drug responses of paired MiniPDX and PDX tumor graft assay, as well as translational data suggest that MiniPDX assay is an advanced tool for personalized cancer treatment.

ZEB1, ZEB2, and the miR-200 family form a counterregulatory network to regulate CD8+ T cell fates.

Long-term immunity depends partly on the establishment of memory CD8+ T cells. We identified a counterregulatory network between the homologous transcription factors ZEB1 and ZEB2 and the miR-200 microRNA family, which modulates effector CD8+ T cell fates. Unexpectedly, Zeb1 and Zeb2 had reciprocal expression patterns and were functionally uncoupled in CD8+ T cells. ZEB2 promoted terminal differentiation, whereas ZEB1 was critical for memory T cell survival and function. Interestingly, the transforming growth factor ß (TGF-ß) and miR-200 family members, which counterregulate the coordinated expression of Zeb1 and Zeb2 during the epithelial-to-mesenchymal transition, inversely regulated Zeb1 and Zeb2 expression in CD8+ T cells. TGF-ß induced and sustained Zeb1 expression in maturing memory CD8+ T cells. Meanwhile, both TGF-ß and miR-200 family members selectively inhibited Zeb2. Additionally, the miR-200 family was necessary for optimal memory CD8+ T cell formation. These data outline a previously unknown genetic pathway in CD8+ T cells that controls effector and memory cell fate decisions.

Dense sgRNA Library Construction Using a Molecular Chipper Approach.

Genetic screens using single-guide-RNA (sgRNA) libraries and CRISPR technology have been powerful to identify genetic regulators for both coding and noncoding regions of the genome. Interrogating functional elements in noncoding regions requires sgRNA libraries that are densely covering, and ideally inexpensive, easy to implement and flexible for customization. We present a Molecular Chipper protocol for generating dense sgRNA libraries from genomic regions of interest. This approach utilizes a combination of random fragmentation and a Type III restriction enzyme to derive a dense coverage of sgRNA library from input DNA.

The DNA Methylcytosine Dioxygenase Tet2 Sustains Immunosuppressive Function of Tumor-Infiltrating Myeloid Cells to Promote Melanoma Progression.

Ten-Eleven-Translocation-2 (Tet2) is a DNA methylcytosine dioxygenase that functions as a tumor suppressor in hematopoietic malignancies. We examined the role of Tet2 in tumor-tissue myeloid cells and found that Tet2 sustains the immunosuppressive function of these cells. We found that Tet2 expression is increased in intratumoral myeloid cells both in mouse models of melanoma and in melanoma patients and that this increased expression is dependent on an IL-1R-MyD88 pathway. Ablation of Tet2 in myeloid cells suppressed melanoma growth in vivo and shifted the immunosuppressive gene expression program in tumor-associated macrophages to a proinflammatory one, with a concomitant reduction of the immunosuppressive function. This resulted in increased numbers of effector T cells in the tumor, and T cell depletion abolished the reduced tumor growth observed upon myeloid-specific deletion of Tet2. Our findings reveal a non-cell-intrinsic, tumor-promoting function for Tet2 and suggest that Tet2 may present a therapeutic target for the treatment of non-hematologic malignancies.

Capture, amplification, and global profiling of microRNAs from low quantities of whole cell lysate.

MicroRNAs (miRNAs) are small non-coding RNAs that control gene expression at the post-transcriptional level via a complex regulatory network that requires genome-wide miRNA profiling to dissect. The patterns of miRNA expression at the genome scale are rich in diagnostic and prognostic information for human diseases such as cancers. This analysis, however, requires multi-step purification of RNAs from large quantities of cells, which is not only time consuming and costly but also challenging in situations where cell numbers are limited. In this study, we report direct capture, amplification, and library preparation of miRNAs from whole cell lysate without the need of pre-purification. As a result, it enables genome-wide miRNA profiling reproducibly with low quantity of cell samples (∼500 hematopoietic cells). Specifically, we conducted a systematic investigation of two key steps - cell lysis for miRNA release and 3' adaptor ligation required for direct miRNA capture and amplification. The obtained expression profile not only distinguishes cell types but also detects individual miRNA alterations in closely related isogenic cell lines. This approach, which is substantially simple as compared to the standard methods because of elimination of the need for RNA purification, is advantageous for the measurement of low quantity samples.

Novel determinants of mammalian primary microRNA processing revealed by systematic evaluation of hairpin-containing transcripts and human genetic variation.

Mature microRNAs (miRNAs) are processed from hairpin-containing primary miRNAs (pri-miRNAs). However, rules that distinguish pri-miRNAs from other hairpin-containing transcripts in the genome are incompletely understood. By developing a computational pipeline to systematically evaluate 30 structural and sequence features of mammalian RNA hairpins, we report several new rules that are preferentially utilized in miRNA hairpins and govern efficient pri-miRNA processing. We propose that a hairpin stem length of 36 ± 3 nt is optimal for pri-miRNA processing. We identify two bulge-depleted regions on the miRNA stem, located ∼16-21 nt and ∼28-32 nt from the base of the stem, that are less tolerant of unpaired bases. We further show that the CNNC primary sequence motif selectively enhances the processing of optimal-length hairpins. We predict that a small but significant fraction of human single-nucleotide polymorphisms (SNPs) alter pri-miRNA processing, and confirm several predictions experimentally including a disease-causing mutation. Our study enhances the rules governing mammalian pri-miRNA processing and suggests a diverse impact of human genetic variation on miRNA biogenesis.

miR-125b promotes MLL-AF9-driven murine acute myeloid leukemia involving a VEGFA-mediated non-cell-intrinsic mechanism.

The hematopoietic stem cell-enriched miR-125 family microRNAs (miRNAs) are critical regulators of hematopoiesis. Overexpression of miR-125a or miR-125b is frequent in human acute myeloid leukemia (AML), and the overexpression of these miRNAs in mice leads to expansion of hematopoietic stem cells accompanied by perturbed hematopoiesis with mostly myeloproliferative phenotypes. However, whether and how miR-125 family miRNAs cooperate with known AML oncogenes in vivo, and how the resultant leukemia is dependent on miR-125 overexpression, are not well understood. We modeled the frequent co-occurrence of miR-125b overexpression and MLL translocations by examining functional cooperation between miR-125b and MLL-AF9 By generating a knock-in mouse model in which miR-125b overexpression is controlled by doxycycline induction, we demonstrated that miR-125b significantly enhances MLL-AF9-driven AML in vivo, and the resultant leukemia is partially dependent on continued overexpression of miR-125b Surprisingly, miR-125b promotes AML cell expansion and suppresses apoptosis involving a non-cell-intrinsic mechanism. MiR-125b expression enhances VEGFA expression and production from leukemia cells, in part by suppressing TET2 Recombinant VEGFA recapitulates the leukemia-promoting effects of miR-125b, whereas knockdown of VEGFA or inhibition of VEGF receptor 2 abolishes the effects of miR-125b In addition, significant correlation between miR-125b and VEGFA expression is observed in human AMLs. Our data reveal cooperative and dependent relationships between miR-125b and the MLL oncogene in AML leukemogenesis, and demonstrate a miR-125b-TET2-VEGFA pathway in mediating non-cell-intrinsic leukemia-promoting effects by an oncogenic miRNA.

Tissue factor is an angiogenic-specific receptor for factor VII-targeted immunotherapy and photodynamic therapy.

Identification of target molecules specific for angiogenic vascular endothelial cells (VEC), the inner layer of pathological neovasculature, is critical for discovery and development of neovascular-targeting therapy for angiogenesis-dependent human diseases, notably cancer, macular degeneration and endometriosis, in which vascular endothelial growth factor (VEGF) plays a central pathophysiological role. Using VEGF-stimulated vascular endothelial cells (VECs) isolated from microvessels, venous and arterial blood vessels as in vitro angiogenic models and unstimulated VECs as a quiescent VEC model, we examined the expression of tissue factor (TF), a membrane-bound receptor on the angiogenic VEC models compared with quiescent VEC controls. We found that TF is specifically expressed on angiogenic VECs in a time-dependent manner in microvessels, venous and arterial vessels. TF-targeted therapeutic agents, including factor VII (fVII)-IgG1 Fc and fVII-conjugated photosensitizer, can selectively bind angiogenic VECs, but not the quiescent VECs. Moreover, fVII-targeted photodynamic therapy can selectively and completely eradicate angiogenic VECs. We conclude that TF is an angiogenic-specific receptor and the target molecule for fVII-targeted therapeutics. This study supports clinical trials of TF-targeted therapeutics for the treatment of angiogenesis-dependent diseases such as cancer, macular degeneration and endometriosis.

Targeting tissue factor as a novel therapeutic oncotarget for eradication of cancer stem cells isolated from tumor cell lines, tumor xenografts and patients of breast, lung and ovarian cancer.

Targeting cancer stem cell (CSC) represents a promising therapeutic approach as it can potentially fight cancer at its root. The challenge is to identify a surface therapeutic oncotarget on CSC. Tissue factor (TF) is known as a common yet specific surface target for cancer cells and tumor neovasculature in several solid cancers. However, it is unknown if TF is expressed by CSCs. Here we demonstrate that TF is constitutively expressed on CD133 positive (CD133+) or CD24-CD44+ CSCs isolated from human cancer cell lines, tumor xenografts from mice and breast tumor tissues from patients. TF-targeted agents, i.e., a factor VII (fVII)-conjugated photosensitizer (fVII-PS for targeted photodynamic therapy) and fVII-IgG1Fc (Immunoconjugate or ICON for immunotherapy), can eradicate CSC via the induction of apoptosis and necrosis and via antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity, respectively. In conclusion, these results demonstrate that TF is a novel surface therapeutic oncotarget for CSC, in addition to cancer cell TF and tumor angiogenic vascular endothelial TF. Moreover, this research highlights that TF-targeting therapeutics can effectively eradicate CSCs, without drug resistance, isolated from breast, lung and ovarian cancer with potential to translate into other most commonly diagnosed solid cancer, in which TF is also highly expressed.

Regulation of the DNA Methylation Landscape in Human Somatic Cell Reprogramming by the miR-29 Family.

Reprogramming to pluripotency after overexpression of OCT4, SOX2, KLF4, and MYC is accompanied by global genomic and epigenomic changes. Histone modification and DNA methylation states in induced pluripotent stem cells (iPSCs) have been shown to be highly similar to embryonic stem cells (ESCs). However, epigenetic differences still exist between iPSCs and ESCs. In particular, aberrant DNA methylation states found in iPSCs are a major concern when using iPSCs in a clinical setting. Thus, it is critical to find factors that regulate DNA methylation states in reprogramming. Here, we found that the miR-29 family is an important epigenetic regulator during human somatic cell reprogramming. Our global DNA methylation and hydroxymethylation analysis shows that DNA demethylation is a major event mediated by miR-29a depletion during early reprogramming, and that iPSCs derived from miR-29a depletion are epigenetically closer to ESCs. Our findings uncover an important miRNA-based approach to generate clinically robust iPSCs.

A Molecular Chipper technology for CRISPR sgRNA library generation and functional mapping of noncoding regions.

Clustered regularly-interspaced palindromic repeats (CRISPR)-based genetic screens using single-guide-RNA (sgRNA) libraries have proven powerful to identify genetic regulators. Applying CRISPR screens to interrogate functional elements in noncoding regions requires generating sgRNA libraries that are densely covering, and ideally inexpensive, easy to implement and flexible for customization. Here we present a Molecular Chipper technology for generating dense sgRNA libraries for genomic regions of interest, and a proof-of-principle screen that identifies novel cis-regulatory domains for miR-142 biogenesis. The Molecular Chipper approach utilizes a combination of random fragmentation and a type III restriction enzyme to derive a densely covering sgRNA library from input DNA. Applying this approach to 17 microRNAs and their flanking regions and with a reporter for miR-142 activity, we identify both the pre-miR-142 region and two previously unrecognized cis-domains important for miR-142 biogenesis, with the latter regulating miR-142 processing. This strategy will be useful for identifying functional noncoding elements in mammalian genomes.

Hyperglycemia repression of miR-24 coordinately upregulates endothelial cell expression and secretion of von Willebrand factor.

An elevated level of von Willebrand factor (VWF) in diabetic patients is associated with increased risk of thrombotic cardiovascular events. The underlying mechanism of how VWF expression is upregulated in diabetes mellitus is poorly understood. We now report that hyperglycemia-induced repression of microRNA-24 (miR-24) increases VWF expression and secretion in diabetes mellitus. In diabetic patients and diabetic mouse models (streptozotocin/high-fat diet-induced and db/db mice), miR-24 is reduced in both tissues and plasma. Knockdown of miR-24 in mice leads to increased VWF mRNA and protein levels and enhanced platelet tethering (spontaneous thrombosis). miR-24 tightly controls VWF levels through pleiotropic effects, including direct binding to the 3' untranslated region of VWF and targeting FURIN and the histamine H1 receptor, known regulators of VWF processing and secretion in endothelial cells. We present a novel mechanism for miR-24 downregulation through hyperglycemia-induced activation of aldose reductase, reactive oxygen species, and c-Myc. These findings support a critical role for hyperglycemic repression of miR-24 in VWF-induced pathology. miR-24 represents a novel therapeutic target to prevent adverse thrombotic events in patients with diabetes mellitus.

Pharmacological modulation of the AKT/microRNA-199a-5p/CAV1 pathway ameliorates cystic fibrosis lung hyper-inflammation.

In cystic fibrosis (CF) patients, hyper-inflammation is a key factor in lung destruction and disease morbidity. We have previously demonstrated that macrophages drive the lung hyper-inflammatory response to LPS in CF mice, because of reduced levels of the scaffold protein CAV1 with subsequent uncontrolled TLR4 signalling. Here we show that reduced CAV1 and, consequently, increased TLR4 signalling, in human and murine CF macrophages and murine CF lungs, is caused by high microRNA-199a-5p levels, which are PI3K/AKT-dependent. Downregulation of microRNA-199a-5p or increased AKT signalling restores CAV1 expression and reduces hyper-inflammation in CF macrophages. Importantly, the FDA-approved drug celecoxib re-establishes the AKT/miR-199a-5p/CAV1 axis in CF macrophages, and ameliorates lung hyper-inflammation in Cftr-deficient mice. Thus, we identify the AKT/miR-199a-5p/CAV1 pathway as a regulator of innate immunity, which is dysfunctional in CF macrophages contributing to lung hyper-inflammation. In addition, we found that this pathway can be targeted by celecoxib.

Effects of genetic variations on microRNA: target interactions.

Genetic variations within microRNA (miRNA) binding sites can affect miRNA-mediated gene regulation, which may lead to phenotypes and diseases. We perform a transcriptome-scale analysis of genetic variants and miRNA:target interactions identified by CLASH. This analysis reveals that rare variants tend to reside in CDSs, whereas common variants tend to reside in the 3' UTRs. miRNA binding sites are more likely to reside within those targets in the transcriptome with lower variant densities, especially target regions in which nucleotides have low mutation frequencies. Furthermore, an overwhelming majority of genetic variants within or near miRNA binding sites can alter not only the potential of miRNA:target hybridization but also the structural accessibility of the binding sites and flanking regions. These suggest an interpretation for certain associations between genetic variants and diseases, i.e. modulation of miRNA-mediated gene regulation by common or rare variants within or near miRNA binding sites, likely through target structure alterations. Our data will be valuable for discovering new associations among miRNAs, genetic variations and human diseases.

Nonstochastic reprogramming from a privileged somatic cell state.

Reprogramming somatic cells to induced pluripotency by Yamanaka factors is usually slow and inefficient and is thought to be a stochastic process. We identified a privileged somatic cell state, from which acquisition of pluripotency could occur in a nonstochastic manner. Subsets of murine hematopoietic progenitors are privileged whose progeny cells predominantly adopt the pluripotent fate with activation of endogenous Oct4 locus after four to five divisions in reprogramming conditions. Privileged cells display an ultrafast cell cycle of ∼8 hr. In fibroblasts, a subpopulation cycling at a similar ultrafast speed is observed after 6 days of factor expression and is increased by p53 knockdown. This ultrafast cycling population accounts for >99% of the bulk reprogramming activity in wild-type or p53 knockdown fibroblasts. Our data demonstrate that the stochastic nature of reprogramming can be overcome in a privileged somatic cell state and suggest that cell-cycle acceleration toward a critical threshold is an important bottleneck for reprogramming. PAPERCLIP:

An extensive network of TET2-targeting MicroRNAs regulates malignant hematopoiesis.

The Ten-Eleven-Translocation 2 (TET2) gene, which oxidates 5-methylcytosine in DNA to 5-hydroxylmethylcytosine (5hmC), is a key tumor suppressor frequently mutated in hematopoietic malignancies. However, the molecular regulation of TET2 expression is poorly understood. We show that TET2 is under extensive microRNA (miRNA) regulation, and such TET2 targeting is an important pathogenic mechanism in hematopoietic malignancies. Using a high-throughput 3' UTR activity screen, we identify >30 miRNAs that inhibit TET2 expression and cellular 5hmC. Forced expression of TET2-targeting miRNAs in vivo disrupts normal hematopoiesis, leading to hematopoietic expansion and/or myeloid differentiation bias, whereas coexpression of TET2 corrects these phenotypes. Importantly, several TET2-targeting miRNAs, including miR-125b, miR-29b, miR-29c, miR-101, and miR-7, are preferentially overexpressed in TET2-wild-type acute myeloid leukemia. Our results demonstrate the extensive roles of miRNAs in functionally regulating TET2 and cellular 5hmC and reveal miRNAs with previously unrecognized oncogenic potential. Our work suggests that TET2-targeting miRNAs might be exploited in cancer diagnosis.