Conjugated linoleic acid improves endothelial Ca2+ signaling by blocking growth factor and cytokine-mediated Cx43 phosphorylation

Highlights

• GF/Cyt, alone or in combination, promote endothelial dysfunction and are rescued by the nutraceutical CLA.

• GF/Cyt coupled directly to Src mediate this damage, and is rescued by t10, c12 CLA or a 1:1 mix of c9, t11 and t10, c12 CLA.

• Interleukins are less damaging than GFs, and are more resistant to CLA rescue.

Abstract

Sustained Ca²⁺ burst signaling is crucial for endothelial vasodilator production and is disrupted by growth factors and cytokines. Conjugated linoleic acid (CLA), a Src inhibitor in certain preparations, is generally regarded as safe during pregnancy by the FDA. Multiple CLA preparations; t10, c12 or c9, t11 CLA, or a 1:1 mixture of the two were administered before growth factor or cytokine treatment. Growth factors and cytokines caused a significant decrease in Ca²⁺ burst numbers in response to ATP stimulation. Both t10, c12 CLA and the 1:1 mixture rescued VEGF₁₆₅ or TNFα inhibited Ca²⁺ bursts and correlated with Src-specific phosphorylation of connexin 43. VEGF₁₆₅, TNFα, and IL-6 in combination at physiologic concentrations revealed IL-6 amplified the inhibitory effects of lower dose of VEGF₁₆₅ and TNFα. Again, the 1:1 CLA mixture was most effective at rescue of function. Therefore, CLA formulations may be a promising treatment for endothelial dysfunction in diseases such as preeclampsia.

Introduction

The vascular endothelium is responsible for sensing blood borne signals and maintenance of homeostasis in the cardiovascular system. Endothelial dysfunction is characterized in part by a reduction in vasodilator production and a shift towards a pro-inflammatory profile, and can be caused by local and systemic changes in the growth factor and cytokine environment (Visser et al., 2007; Saito et al., 2007). There are a variety of diseases in which endothelial dysfunction plays a role, including hypertension, atherosclerosis, chronic kidney failure, and preeclampsia (PE) (Rajendran et al., 2013; Boeldt and Bird, 2017). Despite the widespread acknowledgement of the important role endothelial dysfunction plays in such disorders, therapeutic targeting of the endothelium remains underutilized.

Endothelial intracellular free Ca²⁺ concentration ([Ca²⁺]_i) is an important determinant of vasodilator production (Tran et al., 2009). Such elevations in [Ca²⁺]_i often take the form of periodic Ca²⁺ bursts, which correlate directly with vasodilator production (Yi et al., 2010; Yi et al., 2005; Krupp et al., 2013). Ca²⁺ bursting is dependent on functionally coupled Cx43 gap junctions (Yi et al., 2010; Boeldt et al., 2015; Boeldt et al., 2017). Cx43 gap junctions facilitate communication between adjacent cells through the passage of small molecules and ions, such as IP3 and likely Ca²⁺ (Bird et al., 2013; Solan and Lampe, 2009). The function of Cx43 gap junctions is highly influenced by growth factor and cytokine mediated kinase activation. Src activation results in direct phosphorylation of Cx43 at the inhibitory Y265 residue (Solan and Lampe, 2014). Phosphorylation of Y265 promotes downregulation of Cx43-mediated communication, most likely by destabilizing the Cx43 plaques and increasing the rate of protein turnover (Solan and Lampe, 2014). Because sustained endothelial Ca²⁺ signaling depends on functionally coupled Cx43 gap junctions, the net result of increased Y265 phosphorylation is reduced Ca²⁺ signaling capacity (Boeldt et al., 2015; Ampey et al., 2019).

Previous studies have shown that exposure of intact endothelium to 10 ng/ml Vascular Endothelial Growth Factor (VEGF)₁₆₅ reduces ATP-stimulated Ca²⁺/nitric oxide (NO) production to a level that is not compensated for by the Ca²⁺/NO production attributable to VEGF₁₆₅ itself (Boeldt et al., 2017; Yi et al., 2011). Many of the growth factors and cytokines that may mediate endothelial dysfunction can be sorted into groups/classes based on how their receptors couple directly (VEGF₁₆₅, basic Fibroblast Growth Factor (bFGF), Epidermal Growth Factor (EGF), Tumor Necrosis Factor (TNF)α) (D'Angelo et al., 1995; Deo et al., 2002; Eliceiri et al., 1999; Huang et al., 2003; Nwariaku et al., 2002; Parenti et al., 1998; Tang et al., 2007; Waltenberger et al., 1994) or indirectly (Interleukin (IL)-6, IL-8, IL-1β) (Huang et al., 2013; Liu et al., 2016) to the same common post receptor kinase pathways. However, the literature has widely failed to address the effects of these growth factors and cytokines when used in combination, as they exist in pathophysiologic context, instead of individually. Therefore, there is a need to construct “cocktails” utilizing multiple growth factors and cytokines to bridge the gap from single agonist experiments to experimental conditions that are more representative of the in-vivo environment of endothelial dysfunction. Co-stimulation with multiple growth factors and cytokines may be dependent on an unappreciated signaling hierarchy, and because these kinases may present a common pathway for endocrine-mediated endothelial dysfunction, they may be a prime target for therapeutic intervention (Bird et al., 2013).

In this study, we use human umbilical vein endothelial cells (HUVECs) to model the combined actions of these growth factors and cytokines on endothelial cell function. We use PE as a model system to construct our cocktails as it has been described as a condition of cytokine-mediated endothelial dysfunction (Redman et al., 1999). As appropriate levels of growth factors and cytokines are necessary to support normal pregnancy, we also use a cocktail representative of normal pregnancy as a control. For this study we include TNFα, IL-6 and VEGF in our cocktails, using doses attributed to ‘normal’ and PE pregnancies for TNFα and IL-6 (Lau et al., 2013; Tosun et al., 2010; Xie et al., 2011), and varied doses of VEGF₁₆₅ to mimic systemic and local concentrations. While PE is used as the model for which we construct our dysfunctional phenotype, the resulting findings will likely apply to a broad range of diseases that have an endothelial dysfunction component.

The pharmacological Src inhibitor PP2 has already proven successful in rescuing VEGF₁₆₅ and TNFα pretreated endothelium from a dysfunctional Ca²⁺ signaling phenotype to one more congruent with normal function (Boeldt et al., 2015; Ampey et al., 2019). A pharmacological tool like PP2 is not used in humans due to significant toxicity and concerns of teratogenicity in pregnancy, and therefore offers no therapeutic value. A pair of isomers of conjugated linoleic acid (CLA) used at FDA recommended doses for treatment of inflammatory conditions including rheumatoid arthritis, Crohn's disease and hypertension may hold promise as a safe method for Src inhibition, even in pregnancy. The t10, c12 CLA isomer has been shown to be a Src inhibitor in cancer cells (Shahzad et al., 2018). When used in ovine endothelial cells, VEGF₁₆₅-mediated inhibition of Ca²⁺ responses were rescued with as little as 5 μm t10, c12 CLA, and 50 μM t10, c12 CLA rescued TNFα-mediated Ca²⁺ response inhibition (Boeldt et al., 2015; Ampey et al., 2019). Herein, we examine the efficacy of individual c9, t11 and t10, c12 isomers, as well as a 1:1 mix to rescue Cx43-dependent Ca²⁺ signaling in HUVEC pretreated with individual growth factors or cytokines. We also examine the effect of combinations of a selection of growth factors and cytokines on Ca²⁺ signaling and the ability of CLA isomers alone or as a 1:1 mix, to restore function to the normal phenotype.