Actin cytoskeleton remodeling drives epithelial-mesenchymal transition for hepatoma invasion and metastasis in mice.

High invasiveness is a hallmark of human hepatocellular carcinoma (HCC). Large tumors predict invasion and metastasis. Epithelial-mesenchymal transition (EMT) is crucial for cancer invasion and metastasis. However, the mechanisms whereby large tumors tend to undergo EMT remain unclear. We conducted a subgenome-wide screen and identified KLHL23 as an HCC invasion suppressor by inhibiting EMT. KLHL23 binds to actin and suppresses actin polymerization. KLHL23 silencing induced filopodium and lamellipodium formation. Moreover, EMT was suppressed by KLHL23 through its action on actin dynamics. Traditionally, actin cytoskeleton remodeling is downstream of EMT reprogramming. It is therefore intriguing to ask why and how KLHL23 inversely regulates EMT. Activation of actin cytoskeleton remodeling by either KLHL23 silencing or treatment with actin cytoskeleton modulators augmented cellular hypoxic responses in a cell-density-dependent manner, resulting in hypoxia-inducible factor (HIF) and Notch signals and subsequent EMT. Environmental hypoxia did not induce EMT unless actin cytoskeleton remodeling was simultaneously activated and only when cells were at high density. The resulting EMT was reversed by either adenosine 5'-triphosphate supplementation or actin polymerization inhibitors. Down-regulation of KLHL23 was associated with invasion, metastasis, and poor prognosis of HCC and pancreatic cancer. Correlations of tumor size with EMT and inverse association of expression of KLHL23 with HIF/Notch signals were further validated in patient-derived xenograft HCCs in mice. CONCLUSION: Simultaneously activation of actin cytoskeleton remodeling by intrinsic (such as KLHL23 down-regulation) or microenvironment cues is crucial for cell-density-dependent and hypoxia-mediated EMT, providing a mechanistic link between large tumor size and invasion/metastasis. Our findings provide a means of developing the prevention and treatment strategies for tumor invasion and metastasis. (Hepatology 2018;67:2226-2243).

Synergistic induction of insulin resistance by endothelin-1 and cAMP in 3T3-L1 adipocytes.

Both endothelin-1 (ET-1) and cAMP are implicated for inducing insulin resistance. Since we have shown previously that there is a crosstalk between ET-1 and cAMP signaling pathways in regulating glucose uptake in 3T3-L1 adipocytes, we extended our investigation in this study on whether they may have a synergistic effect on inducing insulin resistance. Our results showed that it was indeed the case. Insulin-stimulated glucose uptake, phosphorylation of PKB, IRS-1-associated PI3K, and IRS-1 tyrosine phosphorylation were all inhibited by ET-1 and 8-bromo cAMP in a synergistic manner. IRS-1 protein levels were similarly decreased by ET-1 and 8-bromo cAMP, attributable to suppressed mRNA expression. In addition, after correction for the loss in IRS-1 protein, the inhibition of insulin-stimulated IRS-1 tyrosine phosphorylation or IRS-1-associated PI3K was mainly caused by cAMP. Moreover, whereas IRS-2 protein levels were increased by cAMP or ET-1 plus cAMP, insulin-stimulated IRS-2-associated PI3K activities were abolished by both treatments. Furthermore, ET-1 and ß-adrenergic agonists had similar synergistic inhibition on insulin-stimulated glucose uptake. In conclusion, we have shown that ET-1 and cAMP may synergistically induce insulin resistance in adipocytes via inhibiting IRS-1 expression as well as insulin-stimulated IRS-1/IRS-2 activities.

Depletion of SUMO ligase hMMS21 impairs G1 to S transition in MCF-7 breast cancer cells.

BACKGROUND: hMMS21 is a human SUMO ligase required for DNA damage repair and mitotic progression in HeLa cervical cancer cells. Owing to the diversity of cancer, we further investigated the effect of hMMS21-depletion on MCF-7 breast cancer cells. METHODS: hMMS21-depletion was achieved by RNA interference. Cellular hMMS21 and E2F1 mRNA levels were estimated by RT-PCR and real-time PCR. Cell cycle profile was assessed by flow cytometry. Western blot and co-immunoprecipitation were used to determine the protein levels of various factors involved in G1-S transition and CDK2- or CDK4-associated p21 and p27. Kinase activity of cyclin E/CDK2 was measured in anti-cyclin E immunoprecipitate. RESULTS: hMMS21-depletion induced slower cell growth and G1-S transition. While it had no effect on cyclin D1 or phospho-Rb (S807/811) levels, hMMS21-depletion provoked lower E2F1 levels and cyclin E/CDK2 activity. The decreased cyclin E/CDK2 activity correlated with increased cellular p21(CIP1) levels and CDK2-p21 association. Moreover, ectopic expression of Flag-hMMS21 but not its ligase-inactive mutant rescued the decreased growth rates of hMMS21-depletd cells. Thus, depletion of hMMS21 seems to impair G1-S transition due to lowered E2F1 protein levels and cyclin E/CDK2 activity. The decreased cyclin E/CDK2 activity is probably attributable to its greater association with p21 as a result of increased p21 levels. In addition, hMMS21-mediated sumoylation appears to be involved. GENERAL SIGNIFICANCE: This study demonstrates that hMMS21 is required for G1-S transition in breast cancer cells and implies that manipulation of hMMS21-mediated sumoylation may alter the growth rates of breast cancer cells.

Endothelin-1 stimulates interleukin-6 secretion from 3T3-L1 adipocytes.

BACKGROUND: Since both endothelin-1 (ET-1) and interleukin-6 (IL-6) may induce insulin resistance and adipose tissue is a major contributor of circulating IL-6, we examined the effects of ET-1 on IL-6 secretion from 3T3-L1 adipocytes. METHODS: IL-6 release was measured by ELISA. RT-PCR and real-time PCR analyses were used to determine cellular IL-6 mRNA levels. A luciferase reporter driven by promoter (-1310/+198) of mouse IL-6 gene was transfected into 3T3-L1 adipocytes to monitor IL-6 transcription. RESULTS: Treatment of adipocytes with ET-1 dose- and time-dependently increased IL-6 secretion. The stimulatory effect of ET-1 on IL-6 secretion was abolished by actinomycin D and ET-1 induced an increase in IL-6 mRNA levels. ET-1 was able to enhance the IL-6 promoter activity and its stimulatory effect was inhibited by GF109203X, U0126, salicylate, dominant negative CREB and mithramycin A. Thus it appears that ET-1 may stimulate IL-6 secretion mainly through an enhanced IL-6 transcription, by a mechanism involving both protein kinase C and p42/p44 mitogen-activated protein kinase, and probably downstream NF-kappaB, CREB and Sp1 transcription factors. GENERAL SIGNIFICANCE: This study demonstrates that ET-1 is able to increase IL-6 secretion from adipocytes and raises the possibility that ET-1-induced insulin resistance may be mediated by IL-6.

Mechanism of inhibition of insulin-stimulated glucose transport by 4-bromocrotonic acid in 3T3-L1 adipocytes.

We have demonstrated previously that 4-bromocrotonic acid (Br-C4) inhibited insulin-stimulated glucose transport by interfering with GLUT4 translocation. In the present study, we further examined the underlying mechanism involved. Since insulin-induced insulin receptor substrate-1-associated phosphatidylinositol (PI) 3-kinase activity was not altered by Br-C4, we determined and found insulin activation of protein kinase B (PKB) and protein kinase Clambda (PKClambda) were both inhibited. However, time-course studies showed that only the inhibition of PKB activation correlated with the inhibition of insulin-stimulated glucose transport. In concert, insulin-stimulated Ser(473/474) phosphorylation on PKB(alpha/beta) were similarly decreased by Br-C4. The finding that okadaic acid-stimulated glucose transport and PKClambda activity were both inhibited by Br-C4 suggested that the effect of Br-C4 on Ser(473/474) phosphorylation was not mediated by protein phosphatase 2A. Moreover, whereas Br-C4 nearly abolished insulin-stimulated integrin-linked kinase (ILK) activity, it only inhibited insulin-stimulated PKB activity by 20%, implying that ILK was not the major kinase for Ser(473/474) phosphorylation. Taken together, these results support the notion that PKB is involved in insulin-stimulated glucose transport. In addition, Br-C4 seems to inhibit insulin-stimulated glucose transport via inhibiting insulin activation of PKB, probably by interfering with insulin activation of an upstream kinase responsible for the phosphorylation of Ser(473/474) residue.