Regulation of collagen deposition in the trout heart during thermal acclimation

Abstract

The passive mechanical properties of the vertebrate heart are controlled in part by the composition of the extracellular matrix (ECM). Changes in the ECM, caused by increased blood pressure, injury or disease can affect the capacity of the heart to fill with blood during diastole. In mammalian species, cardiac fibrosis caused by an increase in collagen in the ECM, leads to a loss of heart function and these changes in composition are considered to be permanent. Recent work has demonstrated that the cardiac ventricle of some fish species have the capacity to both increase and decrease collagen content in response to thermal acclimation. It is thought that these changes in collagen content help maintain ventricle function over seasonal changes in environmental temperatures. This current work reviews the cellular mechanisms responsible for regulating collagen deposition in the mammalian heart and proposes a cellular pathway by which a change in temperature can affect the collagen content of the fish ventricle through mechanotransduction. This work specifically focuses on the role of transforming growth factor β1, MAPK signaling pathways, and biomechanical stretch in regulating collagen content in the fish ventricle. It is hoped that this work increases the appreciation of the use of comparative models to gain insight into phenomenon with biomedical relevance.

Keywords: Cardiac fibroblasts; Cardiac fibrosis; Extracellular matrix; Mechanotransduction; TGF-β; miR-29b.

Fig. 1

Influence of thermal acclimation on the connective tissue content of trout ventricular myocardium quantified using Masson's trichrome staining. (A) The hearts of cold acclimated male trout had significantly more connective tissue in the spongy layer than that of either control or warm acclimated male fish. (B) Cold acclimation of male trout caused an increase in connective tissue content in the compact layer compared to controls while warm acclimation of male trout caused a decrease in connective tissue content, compared to controls. The amount of connective tissue is presented as arbitrary units (A.U.) representing the ratio of connective tissue present in the compartment in relation to muscle. Values are mean ± SEM. Brackets, if present indicate a significant difference between sexes within an acclimation group. Different letters above the bars indicate a significant difference between acclimation groups. Different letters within the bars indicate significant differences between acclimation temperatures when each sex is analyzed separately (p < 0.05). Figure modified from (Klaiman et al., 2011).

Fig. 2

Influence of cold acclimation on collagen composition in the compact and spongy myocardium of zebrafish, Danio rerio. (A) Area, calculated as μm², occupied by each of the four collagen fiber types in the compact myocardium and spongy myocardium within the middle cross-section of hearts from control (27 °C) and cold-acclimated (20 °C) zebrafish. Red denotes the thickest/densest fibers, while green denotes the thinnest. *Values of the same fiber type in the same myocardial layer are significantly different between treatment groups (P < 0.05); n = 9 for all measurements. Figure modified from (Johnson et al., 2014). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Proposed pathway by which a change in environmental temperature leads to changes in collagen deposition in the trout heart. The change in cardiac composition that occurs with thermal acclimation is proposed to result from changes in the biomechanical forces experienced by the heart (A). For example, heart rate decreases at low temperature resulting in an increase in ventricle filling time, and as a result, an increase in stroke volume and a greater extension of the myocardium (Keen et al., 2017; Farrell, 1984). In addition, an acute decrease in environmental temperature causes an increase in blood viscosity, which in turn increases cardiac workload, and deformation of the myocardium cells (Farrell, 1984; Graham et al., 1985). Combined, these effects of low temperature are proposed to lead to an increase in the biomechanical stretch and shear stress experienced by the cardiomyocytes and fibroblasts that compose the myocardium (Keen et al., 2017) (B). Increased stretch of myocytes leads to the release of transforming growth factor β (TGF-β) (Katsumi et al., 2004; MacKenna et al., 1998) (C), while increased stretch of cardiac fibroblasts lead to the activation of mechanically sensitive proteins, and integrin proteins in the membrane (D) (Herum et al., 2017; MacKenna et al., 2000; Manso et al., 2009). This leads to the activation of G-coupled proteins and associated MAPK signaling proteins in the fibroblast (E) (Katsumi et al., 2004; Johnston and Gillis, 2020). Activated component proteins of the MAPK pathway in turn activate transcriptional factors that influence the expression of gene transcripts associated with the regulation of cardiac collagen (F) (Pramod and Shivakumar, 2014; Sinfield et al., 2013). Exposure of cardiac fibroblasts to TGF-β also causes increased activation (phosphorylation) of SMAD2 (G) (Visse and Nagase, 2003; Kolosova et al., 2011; Reed et al., 1994; Johnston and Gillis, 2018), another regulator of gene transcripts associated with collagen deposition, including col1A1. These changes in gene expression result in an increase in the expression and deposition of collagen in the extra cellular matrix (H) (Johnston and Gillis, 2017). Recent work also demonstrates that exposure of trout cardiac fibroblasts to microRNA 29b (miR-29b) leads to a decrease in the expression of col1A3 and in collagen type 1 content of the ECM (Johnston et al., 2019). These results suggest that regulation of collagen turnover by miRs could be involved in the removal of collagen from the heart (Johnston et al., 2019). It is not clear how the expression of miRs may be regulated by a temperature change.

Fig. 4

The effect of bFGF and TGF-β1 on SMAD2 phosphorylation in cultured trout cardiac fibroblasts. Cardiac fibroblast cultures were treated with medium alone or medium containing 15 ng ml⁻¹ bFGF or 15 ng ml⁻¹ TGF-β1 repetitively for 7 days and total SMAD2 and phosphorylated SMAD2 were quantified using Western blot and densitometry. (B) Mean pSMAD2 and SMAD2 levels measured by densitometry and standardized to total protein. Different letters denote an effect of treatment on phosphorylation level of SMAD2, relative to control (P < 0.05). Note the large break on the y-axis. The n for each experiment is 3, with each n being a protein sample extracted from cultured cells derived from a different trout ventricle. Figure modified from (Johnston and Gillis, 2018).

Fig. 5

The effect of TGF-β1 treatment on the expression of genes involved in ECM regulation in cultured trout cardiac fibroblasts 72 h post-treatment. The amount of transcript in each TGF-β1 group is given relative to the control group, which is set to 1 in each panel. *Significant effect of TGF-β1 treatment on gene expression (P < 0.05). n = 3–5. Figure modified from (Johnston and Gillis, 2018).

Fig. 6

The effect of TGF-β1 on collagen production by cultured trout cardiac fibroblasts. A) Average amount of hydroxyproline, used as a proxy for collagen, produced per cell after 24, 48 and 72 h of TGF-β1 treatment in cell pellets. B) The effect of TGF-β1 or l-ascorbic acid (l-AA) treatment on collagen type 1 deposition in cultured cardiac fibroblasts measured 7 days after treatment using Western blot. In panel A different numbers indicate a significant effect of time on the amount of hydroxyproline produced within control cells (P < 0.05) and different letters indicate a significant effect of time on the amount of hydroxyproline produced per cell in the TGF-β1-treated group (P < 0.05). *Significant effect of TGF-β1 on hydroxyproline concentration between control and TGF-β1-treated cells (P < 0.05). In panel B different numbers indicate a significant differences between values. n = 5 for extracellular matrix (ECM) data, where each n represents a separate cell line established from a single heart from a different fish, and each n contains 8–15 technical replicates. Figure modified from (Johnston and Gillis, 2018).

Fig. 7

Activation of p38 and ERK1/2 pathways in response to stretch. Phosphorylation levels of p38 proteins in stretched and control (unstretched) cells measured after 20 min of stretch (A) and 24 h of stretch (C). Phosphorylation levels of ERK proteins in stretched and control cells measured at 20 min (B) and 24 h (D). Asterisks (*) indicate a significant effect of stretch on MAPK phosphorylation (P < 0.05). Open triangles (Δ) signify individual control (unstretched) data points, and open circles (○) are individual data points from stretched cells. Points with similar numerical values were staggered for better readability. Error bars represent standard error of the mean of each group. N = 3 for each group. The n-value represents a fibroblast line from the same individual fish maintained in separate passages, cryopreserved on a different passage number and day, and thawed for experiments on different days. Figure modified from (Johnston and Gillis, 2020).

Fig. 8

The effect of miR-29b transfection on collagen type I protein levels. Cardiac fibroblast cultures were transfected with 10 nmol l⁻¹ mature zebrafish miR-29b mimic or a nonsense small interfering RNA (siRNA), and re-transfected after 3 days, with sampling occurring at 7 days. Mean collagen type I levels measured by densitometry and standardized to total protein on the same membrane, which was cut in half and blotted separately. Letters represent a significant change relative to control (P < 0.05). n = 3 for each experiment, with each n being a protein sample extracted from cultured cells derived from the same trout ventricle maintained in separate passages and cryovials, and thawed on different days, as previously described (Johnston and Gillis, 2017). Figure modified from (Johnston et al., 2019).