Exploring kinase family inhibitors and their moiety preferences using deep SHapley additive exPlanations.

BACKGROUND: While it has been known that human protein kinases mediate most signal transductions in cells and their dysfunction can result in inflammatory diseases and cancers, it remains a challenge to find effective kinase inhibitor as drugs for these diseases. One major challenge is the compensatory upregulation of related kinases following some critical kinase inhibition. To circumvent the compensatory effect, it is desirable to have inhibitors that inhibit all the kinases belonging to the same family, instead of targeting only a few kinases. However, finding inhibitors that target a whole kinase family is laborious and time consuming in wet lab. RESULTS: In this paper, we present a computational approach taking advantage of interpretable deep learning models to address this challenge. Specifically, we firstly collected 9,037 inhibitor bioassay results (with 3991 active and 5046 inactive pairs) for eight kinase families (including EGFR, Jak, GSK, CLK, PIM, PKD, Akt and PKG) from the ChEMBL25 Database and the Metz Kinase Profiling Data. We generated 238 binary moiety features for each inhibitor, and used the features as input to train eight deep neural networks (DNN) models to predict whether an inhibitor is active for each kinase family. We then employed the SHapley Additive exPlanations (SHAP) to analyze the importance of each moiety feature in each classification model, identifying moieties that are in the common kinase hinge sites across the eight kinase families, as well as moieties that are specific to some kinase families. We finally validated these identified moieties using experimental crystal structures to reveal their functional importance in kinase inhibition. CONCLUSION: With the SHAP methodology, we identified two common moieties for eight kinase families, 9 EGFR-specific moieties, and 6 Akt-specific moieties, that bear functional importance in kinase inhibition. Our result suggests that SHAP has the potential to help finding effective pan-kinase family inhibitors.

Identification of pan-kinase-family inhibitors using graph convolutional networks to reveal family-sensitive pre-moieties.

BACKGROUND: Human protein kinases, the key players in phosphoryl signal transduction, have been actively investigated as drug targets for complex diseases such as cancer, immune disorders, and Alzheimer's disease, with more than 60 successful drugs developed in the past 30 years. However, many of these single-kinase inhibitors show low efficacy and drug resistance has become an issue. Owing to the occurrence of highly conserved catalytic sites and shared signaling pathways within a kinase family, multi-target kinase inhibitors have attracted attention. RESULTS: To design and identify such pan-kinase family inhibitors (PKFIs), we proposed PKFI sets for eight families using 200,000 experimental bioactivity data points and applied a graph convolutional network (GCN) to build classification models. Furthermore, we identified and extracted family-sensitive (only present in a family) pre-moieties (parts of complete moieties) by utilizing a visualized explanation (i.e., where the model focuses on each input) method for deep learning, gradient-weighted class activation mapping (Grad-CAM). CONCLUSIONS: This study is the first to propose the PKFI sets, and our results point out and validate the power of GCN models in understanding the pre-moieties of PKFIs within and across different kinase families. Moreover, we highlight the discoverability of family-sensitive pre-moieties in PKFI identification and drug design.

Rosa1, a Transposable Element-Like Insertion, Produces Red Petal Coloration in Rose Through Altering RcMYB114 Transcription.

Rose (Rosa sp.) flowers have a rich diversity of colors resulting from the differential accumulation of anthocyanins, flavonols, and carotenoids. However, the genetic and molecular determinants of the red-petal trait in roses remains poorly understood. Here we report that a transposable element-like insertion (Rosa1) into RcMYB114, a R2R3-MYB transcription factor's promoter region causes its transcription, resulting in red petals. In red-petal varieties, RcMYB114 is expressed specifically in flower organs, but is absent from non-red varieties. Sequencing, yeast two-hybrid, transient transformation, and promoter activity assays of RcMYB114 independently confirmed the role of Rosa1 in altering RcMYB114's transcription and downstream effects on flower color. Genetic and molecular evidence confirmed that the Rosa1 transposable element-like insertion, which is a previously unknown DNA transposable element, is different from those in other plants and is a reliable molecular marker to screen red-petal roses.

Discovery of moiety preference by Shapley value in protein kinase family using random forest models.

BACKGROUND: Human protein kinases play important roles in cancers, are highly co-regulated by kinase families rather than a single kinase, and complementarily regulate signaling pathways. Even though there are > 100,000 protein kinase inhibitors, only 67 kinase drugs are currently approved by the Food and Drug Administration (FDA). RESULTS: In this study, we used "merged moiety-based interpretable features (MMIFs)," which merged four moiety-based compound features, including Checkmol fingerprint, PubChem fingerprint, rings in drugs, and in-house moieties as the input features for building random forest (RF) models. By using > 200,000 bioactivity test data, we classified inhibitors as kinase family inhibitors or non-inhibitors in the machine learning. The results showed that our RF models achieved good accuracy (> 0.8) for the 10 kinase families. In addition, we found kinase common and specific moieties across families using the Shapley Additive exPlanations (SHAP) approach. We also verified our results using protein kinase complex structures containing important interactions of the hinges, DFGs, or P-loops in the ATP pocket of active sites. CONCLUSIONS: In summary, we not only constructed highly accurate prediction models for predicting inhibitors of kinase families but also discovered common and specific inhibitor moieties between different kinase families, providing new opportunities for designing protein kinase inhibitors.

The RhHB1/RhLOX4 module affects the dehydration tolerance of rose flowers (Rosa hybrida) by fine-tuning jasmonic acid levels.

Phytohormones are key factors in plant responsiveness to abiotic and biotic stresses, and maintaining hormone homeostasis is critically important during stress responses. Cut rose (Rosa hybrida) flowers experience dehydration stress during postharvest handling, and jasmonic acid (JA) levels change as a result of this stress. However, how JA is involved in dehydration tolerance remains unclear. We investigated the functions of the JA- and dehydration-induced RhHB1 gene, which encodes a homeodomain-leucine zipper I γ-clade transcription factor, in rose flowers. Silencing RhHB1 decreased petal dehydration tolerance and resulted in a persistent increase in JA-Ile content and reduced dehydration tolerance. An elevated JA-Ile level had a detrimental effect on rose petal dehydration tolerance. RhHB1 was shown to lower the transient induction of JA-Ile accumulation in response to dehydration. In addition to transcriptomic data, we obtained evidence that RhHB1 suppresses the expression of the lipoxygenase 4 (RhLOX4) gene by directly binding to its promoter both in vivo and in vitro. We propose that increased JA-Ile levels weaken the capacity for osmotic adjustment in petal cells, resulting in reduced dehydration tolerance. In conclusion, a JA feedback loop mediated by an RhHB1/RhLOX4 regulatory module provides dehydration tolerance by fine-tuning bioactive JA levels in dehydrated flowers.

Molecular Evidences for the Interactions of Auxin, Gibberellin, and Cytokinin in Bent Peduncle Phenomenon in Rose (Rosa sp.).

In roses (Rosa sp.), peduncle morphology is an important ornamental feature. The common physiological abnormality known as the bent peduncle phenomenon (BPP) seriously decreases the quality of rose flowers and thus the commercial value. Because the molecular mechanisms underlying this condition are poorly understood, we analysed the transcriptional profiles and cellular structures of bent rose peduncles. Numerous differentially expressed genes involved in the auxin, cytokinin, and gibberellin signaling pathways were shown to be associated with bent peduncle. Paraffin sections showed that the cell number on the upper sides of bent peduncles was increased, while the cells on the lower sides were larger than those in normal peduncles. We also investigated the large, deformed sepals that usually accompany BPP and found increased expression level of some auxin-responsive genes and decreased expression level of genes that are involved in cytokinin and gibberellin synthesis in these sepals. Furthermore, removal of the deformed sepals partially relieved BPP. In summary, our findings suggest that auxin, cytokinin, and gibberellin all influence the development of BPP by regulating cell division and expansion. To effectively reduce BPP in roses, more efforts need to be devoted to the molecular regulation of gibberellins and cytokinins in addition to that of auxin.

Proteome and Ubiquitome Changes during Rose Petal Senescence.

Petal senescence involves numerous programmed changes in biological and biochemical processes. Ubiquitination plays a critical role in protein degradation, a hallmark of organ senescence. Therefore, we investigated changes in the proteome and ubiquitome of senescing rose (Rosa hybrida) petals to better understand their involvement in petal senescence. Of 3859 proteins quantified in senescing petals, 1198 were upregulated, and 726 were downregulated during senescence. We identified 2208 ubiquitinated sites, including 384 with increased ubiquitination in 298 proteins and 1035 with decreased ubiquitination in 674 proteins. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses revealed that proteins related to peptidases in proteolysis and autophagy pathways were enriched in the proteome, suggesting that protein degradation and autophagy play important roles in petal senescence. In addition, many transporter proteins accumulated in senescing petals, and several transport processes were enriched in the ubiquitome, indicating that transport of substances is associated with petal senescence and regulated by ubiquitination. Moreover, several components of the brassinosteroid (BR) biosynthesis and signaling pathways were significantly altered at the protein and ubiquitination levels, implying that BR plays an important role in petal senescence. Our data provide a comprehensive view of rose petal senescence at the posttranslational level.

A RhABF2/Ferritin module affects rose (Rosa hybrida) petal dehydration tolerance and senescence by modulating iron levels.

Plants often develop the capacity to tolerate moderate and reversible environmental stresses, such as drought, and to re-establish normal development once the stress has been removed. An example of this phenomenon is provided by cut rose (Rosa hybrida) flowers, which experience typical reversible dehydration stresses during post-harvest handling after harvesting at the bud stages. The molecular mechanisms involved in rose flower dehydration tolerance are not known, however. Here, we characterized a dehydration- and abscisic acid (ABA)-induced ferritin gene (RhFer1). Dehydration-induced free ferrous iron (Fe2+ ) is preferentially sequestered by RhFer1 and not transported outside of the petal cells, to restrict oxidative stresses during dehydration. Free Fe2+ accumulation resulted in more serious oxidative stresses and the induction of genes encoding antioxidant enzyme in RhFer1-silenced petals, and poorer dehydration tolerance was observed compared with tobacco rattle virus (TRV) controls. We also determined that RhABF2, an AREB/ABF transcription factor involved in the ABA signaling pathway, can activate RhFer1 expression by directly binding to its promoter. The silencing of RhABF2 decreased dehydration tolerance and disrupted Fe homeostasis in rose petals during dehydration, as did the silencing of RhFer1. Although both RhFer1 and Fe transporter genes are induced during flower natural senescence in plants, the silencing of RhABF2 or RhFer1 accelerates the petal senescence processes. These results suggest that the regulatory module RhABF2/RhFer1 contributes to the maintenance of Fe levels and enhances dehydration tolerance through the action of RhFer1 locally sequestering free Fe2+ under dehydration conditions, and plays synergistic roles with transporter genes during flower senescence.