Encyclopedia
Please note this is a comparison between Version 1 by Jun-Li Liu and Version 2 by Jason Zhu.
CCN5/WISP2 is a matricellular protein, the expression of which is under the regulation of Wnt signaling and IGF-1.
- matricellular proteins
- knockout mice
- transgenic overexpression
- diet-induced obesity
1. General Overview of CCN/WISP Family of Matricellular Proteins
The CCN/WISP family, which consists of six matricellular proteins, regulates development, cell adhesion and proliferation, extracellular matrix (ECM) remodeling, inflammation, and tumorigenesis. Most of these proteins contain four functional domains: the IGF-BP domain, which has a sequence homology similar to the six classic IGF-BPs, but confers only <1% of affinity to insulin-like growth factors (IGFs) [1][6]; the von Willebrand factor type C repeat (VWC) associated with the ECM; the thrombospondin type I repeat (TSP-1) involved in attachment to integrins; and the cysteine-rich C-terminal repeat (CT, or heparin-binding domain) associated with ligand dimerization and receptor binding [2][3][4][5][6][7,8,9,10,11]. Overall, they share ~50% of amino acid sequences, including 38 cysteines and an N-terminal signal peptide destined for exocytotic secretion, and are believed to form multimeric complexes. Acting either as matricellular proteins [7][12] or, as rwesearchers and others have proposed, growth factors through their interactions with cognate receptors [7][8][9][12,13,14], CCNs regulate the expression and activities of growth factors, cytokines, and matrix metalloproteinases (MMPs) [10][15].
While searching for novel growth factors to promote the proliferation and survival of pancreatic islets, rwesearchers revealed the expression of CCN5/WISP2 (cellular communication network factor 5, Wnt inducible signaling protein 2) in resting pancreatic β-cells and its robust inductions by IGF-1, and further reported that recombinant human protein (rhCCN5) promotes mouse β-cell proliferation and survival in vitro [11][12][1,2]. In a previous article, rwesearchers have reviewed the role of CCN5 in the regulation of pancreatic islet function and metabolic activities in general [13][3].
Before assessing its normal physiology, it is necessary to know where (i.e., in which tissues and cells) CCN5 is normally expressed and how the expression might be regulated (e.g., by obesity and diabetes). It has been reported that CCN5 is expressed throughout murine embryonic development in most organs and tissues [14][15][17,18]. According to the Human Protein Atlas (www.proteinatlas.org/ENSG00000064205-WISP2/tissue, accessed on 9 December 2021), CCN5 is a secreted plasma protein that is highly expressed in the proximal gastrointestinal tract, male and female reproductive tissues (testis, uterine), and adipose tissue. Perhaps more accurately, Genotype-Tissue Expression (GTEx) project is an ongoing effort to build a comprehensive resource to study tissue-specific gene expression and regulation. From there, an expression profile of human CCN5/WISP2 can be retrieved, which highlights CCN5 expression in blood vessels and subcutaneous adipose tissues. However, these results have not been studied in detail and peer reviewed; the results are not all consistent to each other.
Immunohistochemical (IHC) analysis of mouse heart tissue revealed CCN5 expression throughout the cytoplasm of the ventricular myocardium, in the atria and in the valves, and in most nuclei in the myocardium. In the lung, CCN5 was observed in the cytoplasm of alveolar and bronchiolar epithelial cells, endothelial cells, and smooth muscle cells, and also in the nuclei of many cell types [14][17]. In both cases, however, the image resolution was insufficient to clearly define a nuclear localization. All layers of mouse stomach wall and duodenum also displayed CCN5 staining in the cytoplasm. In mice, using IHC, CCN5 is expressed in the cytoplasm of acinar cells through the pancreas; but there was no nuclear staining [14][17]. In humans, analysis of adipose protein secretome highlights CCN5 as a key regulator of obesity and ECM interactions [16][19]. Indeed, CCN5 is one of the only five genes up regulated in obese women. As a novel adipokine, CCN5 expression was significantly increased in human obesity and insulin resistance [17][20].
Furthermore, CCN5 is highly expressed in the testis, with very high expression in peritubular cells and Leydig (testosterone-producing) cells. Numerous genes are expressed in mouse testis, yet most of them have not been studied for their involvement in spermatogenesis and sperm function. Analyzing the phenotype of CCN5-knockout mice might provide insight into the function of CCN5 in male reproduction. In rat and mouse ovaries, CCN5 is expressed in all cell types including stromal cells, thecal cells, granulosa cells, and oocytes [14][17]. Previous studies have revealed that CCN5 is normally expressed in the rat uterus in the smooth muscle, glandular epithelium, and the endometrium [18][21], and rat arterial smooth muscle and endothelium [14][17]. In the human fetus at 5 months, low levels of CCN5 staining were detected in testicular Leydig cells, the uterus, ovarian stroma, fallopian tube, and epididymis [15][18]. In addition to these early observations, a detailed examination of CCN5 expression in various systems can also be found in Grunberg et al. and Twigg et al.’s reviews on the topic [19][20][22,23].
2. The Effects of a Systemic CCN5/WISP2 Gene Deficiency
As expected from previous reports, CCN5/WISP2 gene deficiency has been associated with mild obesity, insulin resistance, hyperglycemia, and lipotoxic cardiomyopathy [21][4]. In a separate report using this model, CCN5 expression is not required for normal bone formation [22][24]. Although these findings support a role for endogenous CCN5 in the regulation of metabolic homeostasis, the CCN5-mediated mechanism of action, including in pancreatic islets, has yet to be elucidated.
2.1. Adipocyte Hypertrophy, Increased Adipogenesis, and Mild Obesity
First, CCN5-knockout mice fed the normal chow diet (NCD) have been shown to exhibit mild obesity in comparison to wild-type littermates, despite no significant change in food intake [21][4]. When fed a high-fat diet (HFD), they did exhibit increased food intake together with obesity vs. wild-type littermates [21][4]. These changes are consistent with the notion that CCN5 is thought to partially prevent obesity by inhibiting the TGF-β signaling pathway, of which Smad3 is a downstream mediator. Indeed, Smad3-deficient mice are resistant to HFD-induced obesity and diabetes [23][25]. The latter was previously confirmed in Smad3-deficient mice fed a HFD, which exhibited increased whole-body glucose uptake, improved insulin sensitivity, and decreased fasting insulin and glucose levels [24][26]. HFD-fed, Smad3-deficient mice also gained less weight than their wild-type littermates and were protected from ectopic lipid accumulation in the liver [24][26]. Activation of the TGF-β/Smad3 pathway is reported to promote a leaner phenotype via Smad3′s regulation of PPARγ coactivator-1α (PGC-1α), which regulates metabolic genes [24][26]. Indeed, TGF-β represses PGC-1α in a Smad3-dependent manner, and the deletion of Smad3 gene relieves this inhibition [24][26]. Thus, the lack of CCN5 signaling may result in mild obesity, possibly through elevated TGF-β/Smad3 signaling activities. Future experiments will be required to directly establish this mechanism.
Alongside obesity, the mass of subcutaneous and perirenal white adipose tissues (sWAT and pWAT) and of the heart was significantly increased in NCD-fed CCN5-knockout vs. wild-type mice [21][4]. The increase in the fat mass may be accounted for by the notable increases in the expression of adipogenic genes and transcription factors, such as sterol regulatory element-binding protein (SREBP1), CCAAT/enhancer-binding protein alpha (C/EBPα), peroxisome proliferator-activated receptor-gamma (PPARγ), and activating protein 2 (aP2) [21][4]. When activated, transcription factors C/EBPα and PPARγ activate the expression of genes that induce an adipocytic phenotype [25][27]. In the meantime, CCN5 knockout may have removed the inhibition of adipogenic differentiation in mesenchymal precursor cells (MPCs), since CCN5 has been shown to maintain MPCs in an undifferentiated state through activation of the canonical WNT signaling pathway [21][4]. With the rise in adipogenesis, CCN5-knockout mice also exhibited adipocyte hypertrophy in comparison to their wild-type littermates [21][4]. These results further suggest the physiological suppression of adipogenesis by normal CCN5 expression. Together, CCN5 knockout would result in increased adipocyte differentiation, activated TGF-β/Smad3 pathway, and the expression of adipogenic genes, which all contribute to mild obesity.
Cellular hypertrophy was also exhibited in the heart of NCD-fed CCN5-knockout mice, with an increase in the mRNA expression of the hypertrophy-associated genes myosin heavy chain beta (β-MHC) and skeletal actin [21][4]. HFD-fed knockout mice exhibited further hypertrophy [21][4]. rWesearchers speculate that a similar molecular mechanism may have caused cellular hypertrophy in both adipocytes and cardiomyocytes. In fact, CCN5 staining was detected in cardiomyocytes [14][17]; its expression in fibroblasts of the left ventricle was induced by myocardial infraction in mice [26][28]. In future studies, a cardiomyocyte- vs. adipocyte-specific knockout of CCN5 could help to determine whether the same hypertrophy mechanism applies to both adipocytes and cardiomyocytes.
2.2. Increased Lipid Accumulation and Fibrosis in the Heart
While NCD-fed CCN5-knockout mice exhibited lipid accumulation in the heart, they also showed an increase in the mRNA levels of lipid oxidation-associated genes (CPT1β and MCAD) and glycerol-3-phosphate acyltransferase (GPAT), which is associated with triglyceride synthesis [21][4]. Although the increase in lipid oxidation-related genes does not align with lipid accumulation, the latter may be explained by the increase in cell size due to hypertrophy and impaired glucose metabolism, likely leading to increased lipid oxidation and utilization. As such, triglyceride synthesis most likely overrides lipid oxidation, resulting in a net accumulation of fat. Perhaps associated with that, NCD-fed CCN5-knockout mice also exhibited fibrosis in the heart, particularly in interstitial and perivascular areas. This was confirmed through collagen deposition in those areas and an increase in the expression of fibrosis-related genes (TGF-β1, Col1, and α-SMA) and inflammation-related genes (F4/80/CD3 and CD11c) [21][4]. Interestingly, CCN5 has been shown to inhibit TGF-β activation in myofibroblasts and the level of CCN5 was reduced in individuals with fibrotic heart failure [27][29]. The lack of TGF-β inhibition in the face of CCN5 deficiency may thus trigger an immune response through the expression of proinflammatory genes and facilitate fibrosis in the heart. Indeed, obesity is linked to the activation of pro-fibrotic signaling through the renin-angiotensin-aldosterone pathway, TGF-β, and oxidative stress [28][30]. Further, cardiac fibrosis is strongly associated with metabolic dysfunction, including obesity [28][29][30,31].
Hence, CCN5 gene deficiency led to lipid accumulation in the heart, despite an apparent increase in lipid oxidation too. Cardiomyocyte hypertrophy may have occurred through a similar mechanism as in adipose tissues, involving increased differentiation of MPCs. Moreover, CCN5 knockout caused cardiac fibrosis and collagen deposition. The lack of CCN5 gene expression likely allowed the induction of a pro-fibrotic pathway due to elevated activities of the TGF-β/Smad3 signaling. On the other hand, whether this pathway is also involved in the induction of fibrosis in adipose tissues remains to be determined.
To further study the cellular mechanism, cardiomyocytes from NCD-fed CCN5 knockout-mice were isolated which exhibited mild insulin resistance. The latter was confirmed by reduced phosphorylation of AKT1 and GSK-3β following insulin administration relative to cardiomyocytes isolated from wild-type mice [21][4]. This was consistent with the mild hyperglycemia and hyperinsulinemia phenotypes and suggests that cardiomyocytes may have switched to more lipid catabolism for energy, thus increasing the expression of lipid oxidation-associated genes. The combination of lipid accumulation, increased lipid oxidation, mild insulin resistance, and cardiac fibrosis may have all contributed to systolic and diastolic dysfunctions of the heart [21][4]. Systolic dysfunction is associated with an abnormal left ventricular ejection fraction, which has been reported to be increased in obese individuals [29][31]. Diastolic dysfunction is proportionally associated with obesity, too, and is characterized by prolonged left ventricular relaxation, decreased blood flow velocity through the mitral valve and E/A ratio, and elevated ventricle filling pressure [29][31].
2.3. Mild Hyperglycemia, Hyperinsulinemia and Cardiac Dysfunction
Perhaps due to the mild obesity, lipid accumulation, and increased lipid oxidation, CCN5-knockout mice fed an NCD also showed mild hyperinsulinemia, hyperglycemia, and insulin resistance [21][4]. On the other hand, HFD-fed CCN5 knockout mice exhibited increased water intake, consistent with a mild diabetic phenotype, together with higher blood glucose and insulin levels [21][4]. Interestingly, while fasting glucose levels were notably higher in NCD-fed CCN5 knockout mice, HFD-fed CCN5-knockout mice did not exhibit such an increase vs. wild-type mice [21][4].
In summary, a systemic deficiency of CCN5 gene expression caused adipocyte hypertrophy, increased adipogenesis, and lipid accumulation, resulted in insulin resistance and glucose intolerance, which are further exacerbated upon HFD feeding. In addition, CCN5 deficiency caused cardiac fibrosis, cardiomyocyte hypertrophy, and lipid accumulation, leading to significant deficits in cardiac function. With little increase or no change in food intake, reswearchers would expect decreased energy expenditure in these mice, which has not been measured. Based on our reports, researcherswe also expect changes in pancreatic β-cell mass or function. It should be noted that these reports have never specified the sex of the animals [21][4], while our preliminary results clearly differ and show significant sexual dimorphic patterns in these mice (J.L. Liu et al. Unpublished observations). Nevertheless, current observations indicate that normal endogenous expression of CCN5 gene suppresses adipogenesis, but promotes/maintains cellular division, insulin sensitivity, and cardiac function.
