Sohye Kang

Oncogenic PI3K deregulates transcription and translation

There have long been indications of a role for PI3K (phosphatidylinositol 3-kinase) in cancer pathogenesis. Experimental data document a requirement for deregulation of both transcription and translation in PI3K-mediated oncogenic transformation. The recent discoveries of cancer-specific mutations in PIK3CA, the gene that encodes the catalytic subunit p110α of PI3K, have heightened the interest in the oncogenic potential of this lipid kinase and have made p110α an ideal drug target.

Cancer-specific mutations in PIK3CA are oncogenic in vivo

The PIK3CA gene, coding for the catalytic subunit p110α of class IA phosphatidylinositol 3-kinases (PI3Ks), is frequently mutated in human cancer. Mutated p110α proteins show a gain of enzymatic function in vitro and are oncogenic in cell culture. Here, we show that three prevalent mutants of p110α, E542K, E545K, and H1047R, are oncogenic in vivo. They induce tumors in the chorioallantoic membrane of the chicken embryo and cause hemangiosarcomas in the animal. These tumors are marked by increased angiogenesis and an activation of the Akt pathway. The target of rapamycin inhibitor RAD001 blocks tumor growth induced by the H1047R p110α mutant. The in vivo oncogenicity of PIK3CA mutants in an avian species strongly suggests a critical role for these mutated proteins in human malignancies.

Oncogenic transformation induced by the p110β, -γ, and -δ isoforms of class I phosphoinositide 3-kinase

Class I phosphoinositide 3-kinase contains four isoforms of the catalytic subunit, p110α, -β, -γ, and -δ. At physiological levels of expression, the wild-type p110α isoform lacks oncogenic potential, but gain-of-function mutations and overexpression of p110α are correlated with oncogenicity. The p110β, -γ, and -δ isoforms induce transformation of cultured cells as wild-type proteins. This oncogenic potential requires kinase activity and can be suppressed by the target of rapamycin inhibitor rapamycin. The p110δ isoform constitutively activates the Akt signaling pathway; p110γ activates Akt only in the presence of serum. The isoforms differ in their requirements for upstream signaling. The transforming activity of the p110γ isoform depends on rat sarcoma viral oncogene homolog (Ras) binding; preliminary data suggest the same for p110β and indicate Ras-independent oncogenic potential of p110δ. The surprising oncogenic potential of the wild-type non-α isoforms of class I phosphoinositide 3-kinase may explain the dearth of cancer-specific mutations in these proteins, because these non-α isoforms could contribute to the oncogenic phenotype of the cell by differential expression.

Oncogenic signaling of class I PI3K isoforms

The catalytic subunits of class I PI3Ks comprise four isoforms: p110α, p110β, p110δ and p110γ. Cancer-specific gain-of-function mutations in p110α have been identified in various malignancies. Cancer-specific mutations in the non-α isoforms of class I PI3K have not yet been identified, however overexpression of either wild-type p110β, p110γ or p110δ is sufficient to induce cellular transformation in chicken embryo fibroblasts. The mechanism whereby these non-α isoforms of class I mediate oncogenic signals is unknown. Here we show that potently transforming class I isoforms signal via Akt/mTOR, inhibit GSK3β and cause degradation of FoxO1. A functional Erk pathway is required for p110γ and p110β transformation but not for transformation by p110δ or the H1047R mutant of p110α. Transformation and signaling by p110γ and p110β are sensitive to loss of interaction with Ras, which acts as a membrane anchor. Mutations in the C2 domain of p110δ reduce transformation, most likely by interfering with membrane association. Several small molecule inhibitors potently and specifically inhibit the oncogenic signaling and transformation of each of the class I PI3K, and, when used in combination with MEK inhibitors, can additively reduce the transformation induced by p110β and p110γ.

Phosphatidylinositol 3-Kinase: The Oncoprotein

The catalytic and regulatory subunits of class I phosphoinositide 3-kinase (PI3K) have oncogenic potential. The catalytic subunit p110α and the regulatory subunit p85 undergo cancer-specific gain-of-function mutations that lead to enhanced enzymatic activity, ability to signal constitutively, and oncogenicity. The β, γ, and δ isoforms of p110 are cell-transforming as overexpressed wild-type proteins. Class I PI3Ks have the unique ability to generate phosphoinositide 3,4,5 trisphosphate (PIP(3)). Class II and class III PI3Ks lack this ability. Genetic and cell biological evidence suggests that PIP(3) is essential for PI3K-mediated oncogenicity, explaining why class II and class III enzymes have not been linked to cancer. Mutational analysis reveals the existence of at least two distinct molecular mechanisms for the gain of function seen with cancer-specific mutations in p110α; one causing independence from upstream receptor tyrosine kinases, the other inducing independence from Ras. An essential component of the oncogenic signal that is initiated by PI3K is the TOR (target of rapamycin) kinase. TOR is an integrator of growth and of metabolic inputs. In complex with the raptor protein (TORC1), it controls cap-dependent translation, and this function is essential for PI3K-initiated oncogenesis.

Mutated PI 3-kinases : Cancer targets on a silver platter

Dyskeratosis congenita (DC) is a rare multi-system syndrome characterized by naildystrophy, abnormal skin pigmentation and mucosal leukoplakia. The gene mutated in the Xlinkedform of human DC encodes for dyskerin, a nucleolar pseudourydilase that is involved inrRNA maturation. Dyskerin is also involved in telomerase function through its interaction withthe telomerase RNA (hTR). Mutations in dyskerin result in low levels of hTR, decreasedtelomerase activity and telomere shortening. Autosomal dominant DC is characterized bymutations in hTR, supporting the hypothesis that the DC phenotype may be caused by impairedtelomere maintenance. Several mutations have been identified in different regions of hTR inpatients affected by autosomal dominant DC. Recent reports have shown that co-expression ofwild-type hTR with hTR harboring mutations found in the pseudoknot domain does not affecttelomerase activity in vitro. However, these studies did not assess the consequences of mutanthTR expression at the telomeres. Here we provide the first direct in vivo evidence that a mutanthTR carrying the GC to AG double substitution in the pseudoknot at nucleotides 107-108 foundin patients affected by autosomal dominant DC does not behave as a dominant-negative fortelomere maintenance. Rather it reconstitutes a weakly active telomerase enzyme, which isdefective in telomere elongation.

Requirement of Phosphatidylinositol(3,4,5)Trisphosphate in Phosphatidylinositol 3-Kinase-Induced Oncogenic Transformation

Phosphatidylinositol 3-kinases (PI3K) are divided into three classes, which differ in their substrates and products. Class I generates the inositol phospholipids PI(3)P, PI(3,4)P2, and PI(3,4,5)P3 referred as PIP, PIP2, and PIP3, respectively. Class II produces PIP and PIP2, and class III generates only PIP. Substrate and product differences of the three classes are determined by the activation loops of their catalytic domains. Substitution of the class I activation loop with either class II or III activation loop results in a corresponding change of substrate preference and product restriction. We have evaluated such activation loop substitutions to show that oncogenic activity of class I PI3K is linked to the ability to produce PIP3. We further show that reduction of cellular PIP3 levels by the 5′-phosphatase PIPP interferes with PI3K-induced oncogenic transformation. PIPP also attenuates signaling through Akt and target of rapamycin. Class III PI3K fails to induce oncogenic transformation. Likewise, a constitutively membrane-bound class I PI3K mutant retaining only the protein kinase is unable to induce transformation. We conclude that PIP3 is an essential component of PI3K-mediated oncogenesis and that inability to generate PIP3 abolishes oncogenic potential. (Mol Cancer Res 2009;7(7):1132–8)

Cell line profiling to improve monoclonal antibody production

Mammalian cell culture performance is influenced by both intrinsic (genetic) and extrinsic (media and process) factors. In this study, intrinsic capacity of various monoclonal antibody‐producing Chinese Hamster Ovary (CHO) cell lines was compared by exposing them to the same culture condition. Microarray‐based transcriptomics and LC–MS/MS shotgun proteomics technologies were utilized to obtain expression landscape of different cell lines. Specific transcripts and proteins correlating with productivity, growth rate and cell size have been identified. The proteomics analysis results showed a strong correlation between the intracellular protein expression levels of the recombinant DHFR and productivity. In contrast, neither the light chain nor the heavy chain of the recombinant monoclonal antibody showed correlation to productivity. Other top ranked proteins which demonstrated positive correlation to productivity included the adaptor protein complex subunits AP3D1and AP2B2, DNA repair protein DDB1 and the ER translocation complex component, SRPR. The subunits of molecular chaperone T‐complex protein 1 and the regulator of mitochondrial one‐carbon metabolism MTHFD2 showed negative correlation to productivity. The transcriptomics analysis has identified the regulators of calcium signaling, Tmem20 and Rcan1, as the top ranked genes displaying positive and negative correlation to productivity, respectively. For the second part of the study, the principal component analysis (PCA) was generated to view the underlying global structure of the expression data. A clear division and expression polarity was observed between the two distinct clusters of cell lines, independent of link to productivity or any other traits examined. The primary component of the PCA generated from either transcriptomics or proteomics data displayed a strong correlation to cell size and doubling time, while none of the main principal components showed correlation to productivity. Our findings suggest that productivity is rather a minor feature in the context of global transcriptional or protein expression space. Biotechnol. Bioeng. 2014;111: 748–760.

PI 3-Kinases: Hidden Potentials Revealed

The potential oncogenicity of PI3-kinases is revealed by two principal mechanisms: mutations causing gain of function and over-expression of wild-type proteins. Cancer-specific mutations in PIK3CA, the gene coding for the catalytic subunit p110a of PI 3-kinase, are oncogenic in the animal. These mutations are therefore significant determinants of the oncogenic cellular phenotype in human tumors and are appropriate and promising targets for small molecule inhibitors. Over-expression of wild-type p110b, g, and d induces oncogenic transformation in cell culture. Although these non-alpha isoforms of PI 3-kinase have not been found mutated in human cancer, deregulated expression could contribute to cellular oncogenic properties and deserves increased attention.

Utilization of tyrosine- and histidine-containing dipeptides to enhance productivity and culture viability.

Adequate supply of nutrients, especially providing a sufficient level of specific amino acids, is essential for cell survival and production. Complex raw materials such as soy hydrolysates or yeast extracts are the source for both free amino acids and peptides. However, typical chemically defined (CD) media provide amino acids only in free form. While most amino acids are highly soluble in media and can be provided at fairly high concentrations, certain amino acids such as tyrosine have poor solubility and thus, only a limited amount can be added as a media component. The limited solubility of amino acids in media can raise the risk of media precipitation and instability, and could contribute to suboptimal culture performance due to insufficient nutrient levels to meet cellular demands. In this study, we examine the use of chemically synthesized dipeptides as an alternative method for delivering amino acids to various monoclonal antibody producing cell lines. In particular, we focus on tyrosine‐containing dipeptides. Due to their substantially higher solubility (up to 250‐fold as compared with free tyrosine), tyrosine‐containing dipeptides can efficiently provide large amounts of tyrosine to cultured cells. When tested in fed‐batch processes, these supplemental dipeptides exerted positive effects, including enhanced culture viability and titer. Moreover, dipeptide‐supplemented cultures displayed improved metabolic profiles including lower lactate and NH  4+ production, and better pH maintenance. In bioreactor studies using two‐sided pH control, a lactate spike occurring on Day 10 and the concomitant high levels of base addition could be prevented with dipeptide supplementation. These beneficial effects could be obtained by one‐time addition of dipeptides during inoculation, and did not require further feeds during the entire 11–15‐day process. Non‐tyrosine‐containing dipeptides, such as His–Gly, also showed improved productivity and viability over control cultures. Biotechnol. Bioeng. 2012;109: 2286–2294.