Reciprocal epigenetic Sox2 regulation by SMAD1-SMAD3 is critical for anoikis resistance and metastasis in cancer

Growth factors in the tumor environment are key regulators of cell survival and anoikis resistance during metastasis. Here we reveal significant dichotomy between TGF-β superfamily growth factors BMP and TGF-β/activin and their downstream SMAD effectors in regulation of anchorage-independent tumor cell survival in ovarian cancer. Gene expression profiling uncovered the transcription factor Sox2 as a key signaling node regulated in an opposing manner by anoikis-promoting BMP2 4 and 9 and anoikis-suppressing TGF-β and activin A. Mechanistically, repression of Sox2 by BMPs is mediated by type I receptors ALK2 and ALK3 induced SMAD1 activation, leading to SMAD1-dependent histone H3K27me3 recruitment and DNA methylation at SOX2s promoter. Conversely, TGF-β and activin A promote Sox2 expression directly by ALK5-mediated SMAD3 activation and histone H3K4me3 recruitment. Increased Sox2 expression promotes anoikis resistance, while decreasing Sox2 levels conversely reduces anoikis resistance and activates cell death pathways. Additionally, administrating BMP9 as a strategy to reduce Sox2 robustly inhibits intraperitoneal tumor burden and increases survival in multiple ovarian cancer xenograft models. Importantly, BMP-driven SMAD1 signaling can override the effects of TGF-β and activin on Sox2 regulation, which has potential clinical significance as we find high TGF-β levels in patient ascites. Our findings highlight the contrasting regulation of anoikis by distinct SMAD signaling pathways that are dependent on a novel dichotomous regulation of Sox2 and implicate the use of a subset of BMPs as a therapeutic strategy in cancer

Activation of TRESK background potassium channels by cloxyquin exerts protective effects against excitotoxic-induced brain injury and neuroinflammation in neonatal rats

The excitotoxicity is a common pathological mechanism of perinatal brain injuries (PBI), however neuroinflammation resulting in PBI is both a cause and a consequence of excitotoxicity. TRESK background potassium channels are an important regulator of neuronal excitability. We therefore investigated effects of activation of TRESK channels by selective activator cloxyquin on excitotoxic-induced brain injury and neuroinflammation involving brain mast cells and inflammatory cytokines in neonatal rats. An excitotoxic model mimicking human perinatal brain lesions was established via intracerebral injection of the glutamatergic agonist ibotenate to into newborn rats. P5 rat pups were intraperitoneally pretreated with vehicle, three different doses of cloxyquin (0.2, 1 and 5 mg/kg), or NMDA receptor antagonist MK-801 (positive control) 30 minutes prior to intracerebral injection of 10 µg ibotenate. Rat pups were sacrificed one or five days after the injury. Coronal brain sections were stained with cresyl-violet for histopathological examinations, and with toluidine-blue for brain mast cells assessments. Concentrations of activin A, IL-1β, IL-6 and IL-10 in brain homogenates were measured using ELISA. Cloxyquin dose-dependently ameliorated ibotenate-induced impairments in the cortical and white matter, and suppressed ibotenate-induced activation and number of brain mast cells. Moreover, cloxyquin dose-dependently reduced concentrations of activin A, IL-1β and IL-6 in the brain tissue induced by ibotenate while it elevated IL-10 level. Our findings reveal for the first time that cloxyquin, a selective activator of TRESK channels, dose-dependently exerted protective effects against excitotoxic-induced neonatal brain injury and neuroinflammation. TRESK channels may be a promising new target for the treatment of PBIs.

Activin A and ALK4 Identified as Novel Regulators of Epithelial to Mesenchymal Transition (EMT) in Human Epicardial Cells

The epicardium, the mesothelial layer covering the heart, is a crucial cell source for cardiac development and repair. It provides cells and biochemical signals to the heart to facilitate vascularization and myocardial growth. An essential element of epicardial behavior is epicardial epithelial to mesenchymal transition (epiMT), which is the initial step for epicardial cells to become motile and invade the myocardium. To identify targets to optimize epicardium-driven repair of the heart, it is vital to understand which pathways are involved in the regulation of epiMT. Therefore, we established a cell culture model for human primary adult and fetal epiMT, which allows for parallel testing of inhibitors and stimulants of specific pathways. Using this approach, we reveal Activin A and ALK4 signaling as novel regulators of epiMT, independent of the commonly accepted EMT inducer TGFβ. Importantly, Activin A was able to induce epicardial invasion in cultured embryonic mouse hearts. Our results identify Activin A/ALK4 signaling as a modulator of epicardial plasticity which may be exploitable in cardiac regenerative medicine.

Activin A Promotes Osteoblastic Differentiation of Human Preosteoblasts through the ALK1-Smad1/5/9 Pathway

Activin A, a member of transforming growth factor-β superfamily, is involved in the regulation of cellular differentiation and promotes tissue healing. Previously, we reported that expression of activin A was upregulated around the damaged periodontal tissue including periodontal ligament (PDL) tissue and alveolar bone, and activin A promoted PDL-related gene expression of human PDL cells (HPDLCs). However, little is known about the biological function of activin A in alveolar bone. Thus, this study analyzed activin A-induced biological functions in preosteoblasts (Saos2 cells). Activin A promoted osteoblastic differentiation of Saos2 cells. Activin receptor-like kinase (ALK) 1, an activin type I receptor, was more strongly expressed in Saos2 cells than in HPDLCs, and knockdown of ALK1 inhibited activin A-induced osteoblastic differentiation of Saos2 cells. Expression of ALK1 was upregulated in alveolar bone around damaged periodontal tissue when compared with a nondamaged site. Furthermore, activin A promoted phosphorylation of Smad1/5/9 during osteoblastic differentiation of Saos2 cells and knockdown of ALK1 inhibited activin A-induced phosphorylation of Smad1/5/9 in Saos2 cells. Collectively, these findings suggest that activin A promotes osteoblastic differentiation of preosteoblasts through the ALK1-Smad1/5/9 pathway and could be used as a therapeutic product for the healing of alveolar bone as well as PDL tissue.

Role of SMAD2 and SMAD3 on Adipose Tissue Development and Function

Introduction: Obesity and its associated metabolic syndrome are major medical problems worldwide including United States. Adipose tissue is the primary site of energy storage, playing important roles in health. Adipose tissue also has other critical functions, producing adipocytokines and contributing to normal nutrient metabolism, which in turn play important roles in satiety, inflammation, and total energy homeostasis. Activin A and activin B play important roles in maintaining body composition and energy homeostasis. This dissertation highlights the role of activin/SMADs signaling in adipose tissue development, function, and maintenance. SMAD2/3 proteins are downstream mediators of transforming growth factor-β (TGFβ) family signaling, including activins, which regulate critical preadipocyte and mature adipocyte functions. Previous studies have demonstrated that Smad2 global knockout mice exhibit embryonic lethality, whereas global loss of Smad3 protects mice against diet-induced obesity and the direct contributions of Smad2 and Smad3 in adipose tissues individually or in combination and the responses of these tissues to activin signaling are unknown. Additionally, our lab demonstrated that the combined loss of activin A and activin B have reduced adiposity in mice and appearance of brown-like cells in visceral white adipose tissue. However, the cell-autonomous role of activins on cell proliferation and differentiation remained unknown in vitro. My hypothesis was that activin signaling regulate adipocyte differentiation and functions via SMAD2/3-mediated mechanism(s) and that the individual or combined adipose-specific deletion of SMAD2/SMAD3 would result in reduced adiposity similar to activin deficient mice. Objective: Here, we sought to determine the primary effects of adipocyte-selective reduction of Smad2 or Smad3 individually and in combination, on diet-induced adiposity and to establish whether preadipocytes isolated from subcutaneous and visceral white adipose tissues differ in their differentiation capacity. We also assessed the role of activins on cell proliferation and differentiation using an in vitro model. Research Design: To assess the adipose-selective requirements of Smad2, Smad3 and Smad2/3, we generated three lines of adipose-selective conditional knockout (cKO) mice including Smad2cKO, Smad3cKO, and Smad2/3 double cKO mice using Smad2 and/or Smad3 “floxed” mice intercrossed with Adiponectin-Cre mice. Additionally, we isolated preadipocytes and examined adipogenic activity of visceral and subcutaneous preadipocyte and the effects of activin on preadipocyte proliferation and differentiation in vitro. Furthermore, we used mouse embryonic fibroblasts (MEFs) from wild type mice and activin double knockout mice to study the cell autonomous role of activin on differentiation and cell fate. Results: Our results demonstrated that subcutaneous preadipocytes differentiate uniformly and almost all wildtype subcutaneous preadipocytes differentiated into mature adipocytes. In contrast, visceral preadipocytes differentiated poorly. Exogenous activin A promoted proliferation and suppressed differentiation of subcutaneous preadipocytes more robustly given that visceral adipocytes differentiate poorly at baseline. Additionally, global knockout of activin A and B promoted differentiation and browning in differentiated MEFs in vitro consistent with in vivo studies. Furthermore, we showed that Smad2cKO mice did not exhibit significant effects on weight gain, irrespective of diet, whereas Smad3cKO male mice displayed a trend of reduced body weight on high fat diet. On both (LFD and HFD) diets, Smad3cKO male mice displayed an adipose depot-selective phenotype, with significant reduction in subcutaneous fat mass but not visceral fat mass. Smad2/3cKO male mice did not show any difference in body weight or fat mass compared to control mice. Female mice with adipose-selective combined deletion of Smad2/3, displayed reduced body weight and reduction of fat mass in both visceral and subcutaneous depot with higher metabolic rate on HFD compared to control littermates. Conclusions: Our study demonstrated that Smad3 is an important contributor to the development and/or maintenance of subcutaneous white adipose tissue in a sex-selective fashion. Combined reduction of Smad2/3 protects female mice from diet induced obesity and is important for visceral and subcutaneous depots in a sex-selective fashion. These findings have implications for understanding SMAD-mediated, depot selective regulation of adipocyte growth and differentiation. Activin treatment promoted proliferation of preadipocytes, while activin deficiency promoted differentiation and altered the phenotypic characteristics of White adipocytes to brown-like cells in vitro consistent with in vivo.

Profiling Activins and Follistatin in Colorectal Cancer According to Clinical Stage, Tumour Sidedness and Smad4 Status

This study explored the roles of activins and follistatin in colorectal cancers. Paired malignant and normal colonic tissues were collected from archived paraffin-embedded (n = 90 patients) alongside fresh (n = 40 patients) specimen cohorts. Activin β-subunits, follistatin and Smad4 mRNAs and proteins were measured by real-time PCR and immunohistochemistry (IHC). Mature activin-A, -B, -AB and follistatin proteins were measured by ELISA. Cancer tissues having ≤ the 20th percentile of the Smad4 IHC score were considered as low (L-S4) group. The Smad4-intact SW480 and Smad4-null HT29 colon cancer cell lines were treated with activins and follistatin, and cell cycle was analysed by flow cytometry. The cell cycle inducing (CCND1/CCND3) and inhibitory (p21/p27) proteins alongside the survival (survivin/BCL2) and pro-apoptosis (Casp-8/Casp-3) markers were measured by immunofluorescence. Thirty-nine patients had right-sided cancers (30%) and showed higher rates of L-S4 tumours (n = 17; 13.1%) alongside worse clinicopathological characteristics relative to left-sided cancers. The βA-subunit and activin-A increased, whilst βB-subunit and activin-AB decreased, in malignant sites and the late-stage cancers revealed the greatest abnormalities. Interestingly, follistatin declined markedly in early-stage malignant tissues, whilst increased significantly in the advanced stages. All activin molecules were comparable between the early stage right- and left-sided tumours, whereas the late-stage right-sided cancers and L-S4 tumours showed more profound deregulations. In vitro, activin-A increased the numbers of the SW480 cells in sub-G1 and G0/G1-phases, whereas reduced the HT29 cell numbers in the sub-G1 phase with simultaneous increases in the G0/G1 and S phases. The p21/p27/Casp-8/Casp-3 proteins escalated, whilst CCND1/CCND3/BCL2/survivin declined in the SW480 cells following activin-A, whereas activin-A only promoted p21 and p27 alongside reduced CCND3 in the HT29 cells. By contrast, activin-AB increased the numbers of SW480 and HT29 cells in Sub-G1 and G0/G1-phases and promoted the anti-cancer and reduced the oncogenic proteins in both cell lines. In conclusion, activins and follistatin displayed stage-dependent dysregulations and were markedly altered during the advanced stages of right-sided and L-S4 cancers. Moreover, the activin-A actions in CRC could be Smad4-dependent, whereas activin-AB may act as a Smad4-independent tumour suppressor protein.

Differential expression of activin A receptor type 1C in cancers of the breast.

Breast cancer affects women at relatively high frequency (1). We mined published microarray datasets (2, 3) to determine in an unbiased fashion and at the systems level genes most differentially expressed in the primary tumors of patients with breast cancer. We report here significant differential expression of the gene encoding activin A receptor type 1C, ACVR1C, when comparing primary tumors of the breast to the tissue of origin, the normal breast. ACVR1C was also differentially expressed in the tumor cells of patients with triple negative breast cancer. ACVR1C mRNA was present at significantly lower quantities in tumors of the breast as compared to normal breast tissue. Analysis of human survival data revealed that expression of ACVR1C in primary tumors of the breast was correlated with overall survival in patients with luminal A subtype cancer, demonstrating a relationship between primary tumor expression of a differentially expressed gene and patient survival outcomes influenced by PAM50 molecular subtype. ACVR1C may be of relevance to initiation, maintenance or progression of cancers of the female breast.

Activin A and Follistatin Serum Concentrations in Breast Augmentation Patients

Background: Capsular contracture is caused by an excessive fibrotic reaction similar as observed in other progressive fibrotic disorders. For their pathogenesis, several studies confirmed the importance of activins and follistatin. The aim of this study was to determine and analyze serum levels of Activin A and follistatin in patients with capsular contracture after aesthetic breast augmentation. Methods: The study included 361 female patients who underwent primary aesthetic breast augmentation, came for control examination after breast augmentation or for revision operation because of capsular contracture. Blood samples were taken and using a specific ELISA to determine the serum concentration levels of Activin A and Follistatin. Results: Ninety-six patients (n = 96), who developed a capsular contracture Baker ≥°III and underwent revision surgery were collected (capsular fibrosis group). One-hundred and fourteen patients (n = 114) were asymptomatic for capsular fibrosis Baker ≥°III after primary breast augmentation and 33 (n = 33) of them had developed no capsular fibrosis after more than 10 years (long-term group). For control group, blood samples were taken from 167 patients (n = 167) before primary aesthetic breast augmentation. Serum Activin A levels were significantly higher in the long-term Group compared with those in the capsular fibrosis- and the control groups. Follistatin levels were significantly lower in the capsular fibrosis group compared to the control- and the long-term groups. A small amount of control group patients (n = 16) developed a capsular fibrosis within 2 years after primary breast augmentation with significant lower follistatin levels. Retrospectively, they showed significantly lower serum follistatin levels than the control group even before the onset of capsular contracture. Conclusions: Capsular fibrosis has no effect on Activin A serum levels. In contrast, follistatin serum levels are lower in patients with capsular fibrosis. These results show that besides many other factors, a dysregulation of the Activin–follistatin axis may have importance on the pathogenesis of capsular contracture.

650 Target the activin receptor 1c on CD4+ T cells to achieve anti-tumor therapeutic effects

BackgroundActivins, members of the transforming growth factor-ß (TGF-ß) superfamily, were isolated and identified in endocrine system, and have been widely studied in endocrine-related cancers,1 2 but not substantially in the context of immune system and endocrine-unrelated cancers.3–5 It has been reported that upon binding to the receptors, activins cause the intracellular recruitment and phosphorylation of smad proteins, which mediate the expression of Foxp3.6–9 Therefore, we hypothesized that the blockade of the interaction of activins and their receptors will inhibit the activins-mediated Foxp3 induction in CD4+ T cells, thus modify the immune suppressive tumor microenvironment and achieve the goal of cancer immunotherapy.MethodsELISA (enzyme-linked immunosorbent assay) has been performed to determine the plasma level of Activin A in tumor-bearing mice and cancer patients. In vitro iTreg (induced regulatory T cells) differentiation has been done to naïve CD4+ cells isolated from wild type mice in the presence or absence of Activin A, and the percentage of Foxp3+ cells was demonstrated by flow cytometric analysis. qRT-PCR analysis has been conducted to determine the mRNA level of activin receptor isotypes in the immune subpopulations sorted from Foxp3-YFP mice. In the end, in vivo subcutaneous transplanted tumor studies have been done to evaluate the anti-tumor therapeutic effects of activin-receptor 1c blockade.ResultsWe show here that tumor-bearing mice had elevated Activin A levels, which correlated directly with tumor burden. Likewise, cancer patients had elevated plasma Activin A compared to healthy controls. Importantly, our in vitro studies suggested that Activin A promoted differentiation of conventional CD4+ cells into Foxp3-expressing induced Tregs, especially when TGF-ß was limited. Database and qRT-PCR analysis of sorted major immune cell subsets in mice revealed that activin receptor 1C (Acvr1c) was uniquely expressed on Tregs and was highly upregulated during iTreg differentiation. Mice deficient in Acvr1c were more resistant to cancer progression compared to wild type mice. This phenotype correlated with reduced expression of the FoxP3 transcription factor in CD4+ cells. Similar phenomena were observed when we treated the mice with anti-Acvr1c antibody after tumor inoculation. This anti-tumor therapeutic effect was more significant when anti-Acvr1c antibody was administrated in combination with anti-PD-1 antibody.ConclusionsBlocking Activin A signaling through its receptor 1c is a promising and disease-specific strategy for preventing the accumulation of immunosuppressive iTregs in cancer. Hence it represents a potential candidate for cancer immunotherapy.AcknowledgementsThis research is supported by the Bloomberg-Kimmel Institute (Immunometabolism Program & Immune Modulation Program), the Melanoma Research Alliance, the NIH (RO1AI099300, RO1AI089830, and R01AI137046), and The DoD (PC130767).ReferencesRisbridger GP, Schmitt JF, Robertson DM. Activins and inhibins in endocrine and other tumors. Endocr Rev 2001;22(6):836–858.Cui X, et al. Perspectives of small molecule inhibitors of activin receptor-like kinase in anti-tumor treatment and stem cell differentiation (Review). Mol Med Rep 2019;19(6):5053–5062.Michael IP, et al. ALK7 signaling manifests a homeostatic tissue barrier that is abrogated during tumorigenesis and metastasis. Dev Cell 2019;49(3):409–424.Wu B, et al. The TGF-ß superfamily cytokine Activin-A is induced during autoimmune neuroinflammation and drives pathogenic Th17 cell differentiation. Immunity 2021;54(2):308–323.Antsiferova M, et al. Activin promotes skin carcinogenesis by attraction and reprogramming of macrophages. MBO Mol Med 2017;9(1):27–45.Tsuchida K, et al. Activin isoforms signal through type I receptor serine/threonine kinase ALK7. Mol Cell Endocrinol 2004;220(1–2):59–65.Khalil AM, et al. Differential binding activity of TGF-ß family proteins to select TGF-ß receptors. J Pharmacol Exp Ther 2016;358(3):423–430.Huber S, et al. Activin a promotes the TGF-beta-induced conversion of CD4+CD25- T cells into Foxp3+ induced regulatory T cells. J Immunol 2009;182(8):4633–4640.Iizuka-Koga M, et al. Induction and maintenance of regulatory T cells by transcription factors and epigenetic modifications. J Autoimmun 2017;83:113–121.Ethics ApprovalAll animal experiments were performed under protocols approved by the Johns Hopkins University Institutional Animal Care and Use Committee (IACUC).