CodonBERT: a BERT-based architecture tailored for codon optimization using the cross-attention mechanism.

MOTIVATION: Due to the varying delivery methods of mRNA vaccines, codon optimization plays a critical role in vaccine design to improve the stability and expression of proteins in specific tissues. Considering the many-to-one relationship between synonymous codons and amino acids, the number of mRNA sequences encoding the same amino acid sequence could be enormous. Finding stable and highly expressed mRNA sequences from the vast sequence space using in silico methods can generally be viewed as a path-search problem or a machine translation problem. However, current deep learning-based methods inspired by machine translation may have some limitations, such as recurrent neural networks, which have a weak ability to capture the long-term dependencies of codon preferences. RESULTS: We develop a BERT-based architecture that uses the cross-attention mechanism for codon optimization. In CodonBERT, the codon sequence is randomly masked with each codon serving as a key and a value. In the meantime, the amino acid sequence is used as the query. CodonBERT was trained on high-expression transcripts from Human Protein Atlas mixed with different proportions of high codon adaptation index codon sequences. The result showed that CodonBERT can effectively capture the long-term dependencies between codons and amino acids, suggesting that it can be used as a customized training framework for specific optimization targets. AVAILABILITY AND IMPLEMENTATION: CodonBERT is freely available on https://github.com/FPPGroup/CodonBERT.

Increased endogenous PKG I activity attenuates EGF-induced proliferation and migration of epithelial ovarian cancer via the MAPK/ERK pathway.

The type I cGMP-dependent protein kinase (PKG I) is recognized as a tumor suppressor, but its role in EGFR regulated epithelial ovarian cancer (EOC) progression remains unclear. We evaluated the in vivo and in vitro effects of activated PKG I in EGF-induced EOC cell proliferation, migration, and invasion. The expressions of EGFR and PKG I were elevated, but the activated PKG I was decreased in EOC tissues of patients and cells lines. The addition of 8-Br-cGMP, a specific PKG I activator, attenuated the EGF-induced EOC cell proliferation, migration, and invasion in vitro. Similarly, activated PKG I also attenuated EOC progression in vivo using an EOC xenograft nude mouse model. The activated PKG I interacted with EGFR, causing increased threonine (693) phosphorylation and decreased tyrosine (1068) phosphorylation of EGFR, which resulted in disrupted EGFR-SOS1-Grb2 combination. Subsequently, the cytoplasmic phosphorylation of downstream proteins (c-Raf, MEK1/2, and ERK1/2) were declined, impeding the phosphorylated ERK1/2's nucleus translocation, and this reduction of phosphorylated tyrosine (1068) EGFR and ERK1/2 were also abolished by Rp-8-Br-cGMPS. Our results suggest that the activation of PKG I attenuates EGF-induced EOC progression, and the 8-Br-cGMP-PKG I-EGFR/MEK/ERK axis might be a potential target for EOC therapy.

Tbc1d10c is a selective, constitutive suppressor of the CD8 T-cell anti-tumor response.

Cancer immunotherapy approaches target signaling pathways that are highly synonymous between CD4 and CD8 T-cell subsets and, therefore, often stimulate nonspecific lymphocyte activation, resulting in cytotoxicity to otherwise healthy tissue. The goal of our study was to identify intrinsic modulators of basic T lymphocyte activation pathways that could discriminately bolster CD8 anti-tumor effector responses. Using a Tbc1d10c null mouse, we observed marked resistance to a range of tumor types conferred by Tbc1d10c deficiency. Moreover, tumor-bearing Tbc1d10c null mice receiving PD-1 or CTLA-4 monotherapy exhibited a 33% or 90% cure rate, respectively. While Tbc1d10c was not expressed in solid tumor cells, Tbc1d10c disruption selectively augmented CD8 T-cell activation and cytotoxic effector responses and adoptive transfer of CD8 T cells alone was sufficient to recapitulate Tbc1d10c null tumor resistance. Mechanistically, Tbc1d10c suppressed CD8 T-cell activation and anti-tumor function by intersecting canonical NF-κB pathway activation via regulation of Map3k3-mediated IKKß phosphorylation. Strikingly, none of these cellular or molecular perturbations in the NF-κB pathway were featured in Tbc1d10c null CD4 T cells. Our findings identify a Tbc1d10c-Map3k3-NF-κB signaling axis as a viable therapeutic target to promote CD8 T-cell anti-tumor immunity while circumventing CD4 T cell-associated cytotoxicity and NF-κB activation in tumor cells.

Lipid-independent activation of a muscle-specific PKCα splicing variant.

Protein kinase C-α (PKCα) plays a major role in a diverse range of cellular processes. Studies to date have defined the regulatory controls and function of PKCα entirely based upon the previously annotated ubiquitously expressed prototypical isoform. From RNA-seq-based transcriptome analysis in murine heart, we identified a previously unannotated PKCα variant produced by alternative RNA splicing. This PKCα transcript variant, which we named PKCα-novel exon (PKCα-NE), contains an extra exon between exon 16 and exon 17, and is specifically detected in adult mouse cardiac and skeletal muscle, but not other tissues; it is also detected in human hearts. This transcript variant yields a PKCα isoform with additional 16 amino acids inserted in its COOH-terminal variable region. Although the canonical PKCα enzyme is a lipid-dependent kinase, in vitro kinase assays show that PKCα-NE displays a high level of basal lipid-independent catalytic activity. Our unbiased proteomic analysis identified a specific interaction between PKCα-NE and eukaryotic elongation factor-1α (eEF1A1). Studies in cardiomyocytes link PKCα-NE expression to an increase in eEF1A1 phosphorylation and elevated protein synthesis. In summary, we have identified a previously uncharacterized muscle-specific PKCα splicing variant, PKCα-NE, with distinct biochemical properties that plays a unique role in the control of the protein synthesis machinery in cardiomyocytes.NEW & NOTEWORTHY PKCα is an important signaling molecule extensively studied in many cellular processes. However, no isoforms have been reported for PKCα except one prototypic isoform. Alternative mRNA splicing of Prkca gene was detected for the first time in rodent and human cardiac tissue, which can produce a previously unknown PKCα-novel exon (NE) isoform. The biochemistry and molecular effects of PKCα-NE are markedly different from PKCα wild type, suggesting potential functional diversity of PKCα signaling in muscle.

Telangiectasia-ectodermal dysplasia-brachydactyly-cardiac anomaly syndrome is caused by de novo mutations in protein kinase D1.

BACKGROUND: We describe two unrelated patients who display similar clinical features including telangiectasia, ectodermal dysplasia, brachydactyly and congenital heart disease. METHODS: We performed trio whole exome sequencing and functional analysis using in vitro kinase assays with recombinant proteins. RESULTS: We identified two different de novo mutations in protein kinase D1 (PRKD1, NM_002742.2): c.1774G>C, p.(Gly592Arg) and c.1808G>A, p.(Arg603His), one in each patient. PRKD1 (PKD1, HGNC:9407) encodes a kinase that is a member of the protein kinase D (PKD) family of serine/threonine protein kinases involved in diverse cellular processes such as cell differentiation and proliferation and cell migration as well as vesicle transport and angiogenesis. Functional analysis using in vitro kinase assays with recombinant proteins showed that the mutation c.1808G>A, p.(Arg603His) represents a gain-of-function mutation encoding an enzyme with a constitutive, lipid-independent catalytic activity. The mutation c.1774G>C, p.(Gly592Arg) in contrast shows a defect in substrate phosphorylation representing a loss-of-function mutation. CONCLUSION: The present cases represent a syndrome, which associates symptoms from several different organ systems: skin, teeth, bones and heart, caused by heterozygous de novo mutations in PRKD1 and expands the clinical spectrum of PRKD1 mutations, which have hitherto been linked to syndromic congenital heart disease and limb abnormalities.

Cleavage Alters the Molecular Determinants of Protein Kinase C-δ Catalytic Activity.

Protein kinase C-δ (PKCδ) is an allosterically activated enzyme that acts much like other PKC isoforms to transduce growth factor-dependent signaling responses. However, PKCδ is unique in that activation loop (Thr507) phosphorylation is not required for catalytic activity. Since PKCδ can be proteolytically cleaved by caspase-3 during apoptosis, the prevailing assumption has been that the kinase domain fragment (δKD) freed from autoinhibitory constraints imposed by the regulatory domain is catalytically competent and that Thr507 phosphorylation is not required for δKD activity. This study provides a counternarrative showing that δKD activity is regulated through Thr507 phosphorylation. We show that Thr507-phosphorylated δKD is catalytically active and not phosphorylated at Ser359 in its ATP-positioning G-loop. In contrast, a δKD fragment that is not phosphorylated at Thr507 (which accumulates in doxorubicin-treated cardiomyocytes) displays decreased C-terminal tail priming-site phosphorylation, increased G-loop Ser359 phosphorylation, and defective kinase activity. δKD is not a substrate for Src, but Src phosphorylates δKD-T507A at Tyr334 (in the newly exposed δKD N terminus), and this (or an S359A substitution) rescues δKD-T507A catalytic activity. These results expose a unique role for δKD-Thr507 phosphorylation (that does not apply to full-length PKCδ) in structurally organizing diverse elements within the enzyme that critically regulate catalytic activity.

A novel phosphorylation site at Ser130 adjacent to the pseudosubstrate domain contributes to the activation of protein kinase C-δ.

Protein kinase C-δ (PKCδ) is a signalling kinase that regulates many cellular responses. Although most studies focus on allosteric mechanisms that activate PKCδ at membranes, PKCδ also is controlled via multi-site phosphorylation [Gong et al. (2015) Mol. Cell. Biol. 35: , 1727-1740]. The present study uses MS-based methods to identify PKCδ phosphorylation at Thr(50) and Ser(645) (in resting and PMA-treated cardiomyocytes) as well as Thr(37), Thr(38), Ser(130), Thr(164), Thr(211), Thr(215), Ser(218), Thr(295), Ser(299) and Thr(656) (as sites that increase with PMA). We focused on the consequences of phosphorylation at Ser(130) and Thr(141) (sites just N-terminal to the pseudosubstrate domain). We show that S130D and T141E substitutions co-operate to increase PKCδ's basal lipid-independent activity and that Ser(130)/Thr(141) di-phosphorylation influences PKCδ's substrate specificity. We recently reported that PKCδ preferentially phosphorylates substrates with a phosphoacceptor serine residue and that this is due to constitutive phosphorylation at Ser(357), an ATP-positioning G-loop site that limits PKCδ's threonine kinase activity [Gong et al. (2015) Mol. Cell. Biol. 35: , 1727-1740]. The present study shows that S130D and T141E substitutions increase PKCδ's threonine kinase activity indirectly by decreasing G loop phosphorylation at Ser(357). A S130F substitution [that mimics a S130F single-nt polymorphism (SNP) identified in some human populations] also increases PKCδ's maximal lipid-dependent catalytic activity and confers threonine kinase activity. Finally, we show that Ser(130)/Thr(141) phosphorylations relieve auto-inhibitory constraints that limit PKCδ's activity and substrate specificity in a cell-based context. Since phosphorylation sites map to similar positions relative to the pseudosubstrate domains of other PKCs, our results suggest that phosphorylation in this region of the enzyme may constitute a general mechanism to control PKC isoform activity.

The C2 Domain and Altered ATP-Binding Loop Phosphorylation at Ser³59 Mediate the Redox-Dependent Increase in Protein Kinase C-δ Activity.

The diverse roles of protein kinase C-δ (PKCδ) in cellular growth, survival, and injury have been attributed to stimulus-specific differences in PKCδ signaling responses. PKCδ exerts membrane-delimited actions in cells activated by agonists that stimulate phosphoinositide hydrolysis. PKCδ is released from membranes as a Tyr(313)-phosphorylated enzyme that displays a high level of lipid-independent activity and altered substrate specificity during oxidative stress. This study identifies an interaction between PKCδ's Tyr(313)-phosphorylated hinge region and its phosphotyrosine-binding C2 domain that controls PKCδ's enzymology indirectly by decreasing phosphorylation in the kinase domain ATP-positioning loop at Ser(359). We show that wild-type (WT) PKCδ displays a strong preference for substrates with serine as the phosphoacceptor residue at the active site when it harbors phosphomimetic or bulky substitutions at Ser(359.) In contrast, PKCδ-S359A displays lipid-independent activity toward substrates with either a serine or threonine as the phosphoacceptor residue. Additional studies in cardiomyocytes show that oxidative stress decreases Ser(359) phosphorylation on native PKCδ and that PKCδ-S359A overexpression increases basal levels of phosphorylation on substrates with both phosphoacceptor site serine and threonine residues. Collectively, these studies identify a C2 domain-pTyr(313) docking interaction that controls ATP-positioning loop phosphorylation as a novel, dynamically regulated, and physiologically relevant structural determinant of PKCδ catalytic activity.

Vav3-rac1 signaling regulates prostate cancer metastasis with elevated Vav3 expression correlating with prostate cancer progression and posttreatment recurrence.

Prostate cancer remains the second leading cause of cancer death in men in the Western world. Yet current therapies do not significantly improve the long-term survival of patients with distant metastasis. In this study, we investigated the role of the guanine nucleotide exchange factor Vav3 in prostate cancer progression and metastasis and found that Vav3 expression correlated positively with prostate cancer cell migration and invasion. Stimulation of the receptor tyrosine kinase EphA2 by ephrinA1 resulted in recruitment and tyrosine phosphorylation of Vav3, leading to Rac1 activation as well as increased migration and invasion in vitro. Reduction of Vav3 resulted in fewer para-aortic lymph nodes and bone metastasis in vivo. Clinically, expression of Vav3 and EphA2 was elevated in late-stage and metastatic prostate cancers. Among patients with stage IIB or earlier prostate cancer, higher Vav3 expression correlated with lower cumulative biochemical failure-free survival, suggesting that Vav3 may represent a prognostic marker for posttreatment recurrence of prostate cancer. Together, our findings provide evidence that the Vav3-mediated signaling pathway may serve as a therapeutic target for prostate cancer metastasis.

Two protein kinase C isoforms, δ and ε, regulate energy homeostasis in mitochondria by transmitting opposing signals to the pyruvate dehydrogenase complex.

Energy production in mitochondria is a multistep process that requires coordination of several subsystems. While reversible phosphorylation is emerging as the principal tool, it is still unclear how this signal network senses the workloads of processes as different as fuel procurement, catabolism in the Krebs cycle, and stepwise oxidation of reducing equivalents in the electron transfer chain. We previously proposed that mitochondria use oxidized cytochrome c in concert with retinol to activate protein kinase Cδ, thereby linking a prominent kinase network to the redox balance of the ETC. Here, we show that activation of PKCε in mitochondria also requires retinol as a cofactor, implying a redox-mechanism. Whereas activated PKCδ transmits a stimulatory signal to the pyruvate dehdyrogenase complex (PDHC), PKCε opposes this signal and inhibits the PDHC. Our results suggest that the balance between PKCδ and ε is of paramount importance not only for flux of fuel entering the Krebs cycle but for overall energy homeostasis. We observed that the synthetic retinoid fenretinide substituted for the retinol cofactor function but, on chronic use, distorted this signal balance, leading to predominance of PKCε over PKCδ. The suppression of the PDHC might explain the proapoptotic effect of fenretinide on tumor cells, as well as the diminished adiposity observed in experimental animals and humans. Furthermore, a disturbed balance between PKCδ and PKCε might underlie the injury inflicted on the ischemic myocardium during reperfusion. dehydrogenase complex.

Regulation of intermediary metabolism by the PKCdelta signalosome in mitochondria.

PKCδ has emerged as a novel regulatory molecule of oxidative phosphorylation by targeting the pyruvate dehydrogenase complex (PDHC). We showed that activation of PKCδ leads to the dephosphorylation of pyruvate dehydrogenase kinase 2 (PDK2), thereby decreasing PDK2 activity and increasing PDH activity, accelerating oxygen consumption, and augmenting ATP synthesis. However, the molecular components that mediate PKCδ signaling in mitochondria have remained elusive so far. Here, we identify for the first time a functional complex, which includes cytochrome c as the upstream driver of PKCδ, and uses the adapter protein p66Shc as a platform with vitamin A (retinol) as a fourth partner. All four components are necessary for the activation of the PKCδ signal chain. Genetic ablation of any one of the three proteins, or retinol depletion, silences signaling. Furthermore, mutations that disrupt the interaction of cytochrome c with p66Shc, of p66Shc with PKCδ, or the deletion of the retinol-binding pocket on PKCδ, attenuate signaling. In cytochrome c-deficient cells, reintroduction of cytochrome c Fe(3+) protein restores PKCδ signaling. Taken together, these results indicate that oxidation of PKCδ is key to the activation of the pathway. The PKCδ/p66Shc/cytochrome c signalosome might have evolved to effect site-directed oxidation of zinc-finger structures of PKCδ, which harbor the activation centers and the vitamin A binding sites. Our findings define the molecular mechanisms underlying the signaling function of PKCδ in mitochondria.

RACK1 and CIS mediate the degradation of BimEL in cancer cells.

RACK1 is a 7-WD motif-containing protein with numerous downstream effectors regulating various cellular functions. Using a yeast two-hybrid screen, we identified dynein light chain 1 as a novel interacting partner of RACK1. Additionally, we demonstrated that RACK1 formed a complex with DLC1 and Bim, specifically BimEL, in the presence of apoptotic agents. Upon paclitaxel treatment, RACK1, DLC1, and CIS mediated the degradation of BimEL through the ElonginB/C-Cullin2-CIS ubiquitin-protein isopeptide ligase complex. We further showed that RACK1 conferred paclitaxel resistance to breast cancer cells in vitro and in vivo. Finally, we observed an inverse correlation between CIS and BimEL levels in both ovarian and breast cancer cell lines and specimens. Our study suggests a role of RACK1 in protecting cancer cells from apoptosis by regulating the degradation of BimEL, which together with CIS could play an important role of drug resistance in chemotherapy.

O3 and UV/O3 oxidation of organic constituents of biotreated municipal wastewater.

Dissolved organic matter (DOM) in the biotreated effluent of a municipal wastewater treatment plant was separated by XAD-8 and XAD-4 resins into four fractions: hydrophobic acids, non-acid hydrophobics, transphilics and hydrophilics. Ozonation with and without ultraviolet (UV) enhancement removed most UV-absorbing substances in the first 30 min achieving 78% and 63% reduction in UV254, respectively; the UV enhancement resulted in a greater reduction in dissolved organic carbon (DOC) (90% vs. 36%). Ozone reacted sequentially with aromatic hydrophobics, transphilics, and then hydrophilics; however, under UV, it reacted with all four organic fractions simultaneously. Low-MW hydrophilics were the most abundant fraction in the ozone-treated effluent.

Data mining techniques for cancer detection using serum proteomic profiling.

OBJECTIVE: Pathological changes in an organ or tissue may be reflected in proteomic patterns in serum. It is possible that unique serum proteomic patterns could be used to discriminate cancer samples from non-cancer ones. Due to the complexity of proteomic profiling, a higher order analysis such as data mining is needed to uncover the differences in complex proteomic patterns. The objectives of this paper are (1) to briefly review the application of data mining techniques in proteomics for cancer detection/diagnosis; (2) to explore a novel analytic method with different feature selection methods; (3) to compare the results obtained on different datasets and that reported by Petricoin et al. in terms of detection performance and selected proteomic patterns. METHODS AND MATERIAL: Three serum SELDI MS data sets were used in this research to identify serum proteomic patterns that distinguish the serum of ovarian cancer cases from non-cancer controls. A support vector machine-based method is applied in this study, in which statistical testing and genetic algorithm-based methods are used for feature selection respectively. Leave-one-out cross validation with receiver operating characteristic (ROC) curve is used for evaluation and comparison of cancer detection performance. RESULTS AND CONCLUSIONS: The results showed that (1) data mining techniques can be successfully applied to ovarian cancer detection with a reasonably high performance; (2) the classification using features selected by the genetic algorithm consistently outperformed those selected by statistical testing in terms of accuracy and robustness; (3) the discriminatory features (proteomic patterns) can be very different from one selection method to another. In other words, the pattern selection and its classification efficiency are highly classifier dependent. Therefore, when using data mining techniques, the discrimination of cancer from normal does not depend solely upon the identity and origination of cancer-related proteins.