Phosphorylation of tropomodulin1 contributes to the regulation of actin filament architecture in cardiac muscle

Abstract

Tropomodulin1 (Tmod1) is an actin-capping protein that plays an important role in actin filament pointed-end dynamics and length in striated muscle. No mechanisms have been identified to explain how Tmod1's functional properties are regulated. The purpose of this investigation was to explore the functional significance of the phosphorylation of Tmod1 at previously identified Thr54. Rat cardiomyocytes were assessed for phosphorylation of Tmod1 using Pro-Q Diamond staining and (32)P labeling. Green fluorescent protein-tagged phosphorylation-mimic (T54E) and phosphorylation-deficient (T54A) versions of Tmod1 were expressed in cultured cardiomyocytes, and the ability of these mutants to assemble and restrict actin lengths was observed. We report for the first time that Tmod1 is phosphorylated endogenously in cardiomyocytes, and phosphorylation at Thr54 causes a significant reduction in the ability of Tmod1 to assemble to the pointed end compared with that of the wild type (WT; 48 vs. 78%, respectively). In addition, overexpression of Tmod1-T54E restricts actin filament lengths by only ∼3%, whereas Tmod1-WT restricts the lengths significantly by ∼8%. Finally, Tmod1-T54E altered the actin filament-capping activity in polymerization assays. Taken together, our data suggest that pointed-end assembly and Tmod1's thin filament length regulatory function are regulated by its phosphorylation state.

Keywords: post-translational modification; sarcomere.

Figure 1.

GFP-Tmod1 is phosphorylated in cardiomyocytes in culture. Phosphorylation of GFP-Tmod1 is detected by Pro-Q Diamond staining. GFP-Tmod1 immunoprecipitated from transduced rat cardiomyocytes treated with okadaic acid (+OA) show an increased level of total phosphorylation compared with that for vehicle (DMSO) control (−OA) and GFP transduced controls. The Coomassie blue–stained gel (left) shows the presence of immunoprecipitated GFP and GFP-Tmod1 (asterisks in respective lanes). The same gel after Pro-Q Diamond staining (right) indicates that GFP-Tmod1 has a detectable level of total phosphorylation when treated with okadaic acid. Phosphorylated ovalbumin (45 kDa) present in the prestained ladder is used as a positive control for the Pro-Q Diamond stain.

Figure 2.

Expression of a phosphorylation-mimic mutation at position 54 (GFP-Tmod1-T54E) within Tmod1 perturbs pointed-end assembly. Neonatal rat cardiomyocytes expressing GFP-Tmod1 (WT and mutant) or GFP alone were stained with anti-GFP antibodies, Texas Red phalloidin, and anti-α-actinin antibodies (to stain the Z discs). Right panels: merged images of the triple staining (GFP, green; F-actin, red; α-actinin, blue). Clear and consistent striated assembly of GFP-Tmod1 was observed at thin filament pointed ends in the majority of cells expressing the GFP-Tmod1 WT and the GFP-Tmod1-T54A mutant. However, faint and inconsistent pointed-end assembly was observed in cells expressing the phosphorylation-mimic mutation at threonine position 54. No significant differences in α-actinin staining were observed in cells expressing any of the Tmod1 proteins. Scale bar = 10 μm.

Figure 3.

Cardiac myocytes expressing GFP-Tmod1-T54E display inconsistent pointed-end assembly. The graph shows the percentage of myocytes demonstrating diffuse and/or inconsistent or consistent thin filament pointed-end assembly. Transfected cells were fixed 4 d after transfection and stained for GFP, with Texas Red-conjugated phalloidin, and for sarcomeric α-actinin. Only GFP-positive cardiomyocytes were analyzed. Data from 3 cultures were analyzed, and the values are presented as percentages (means ± sd from each culture); <50% of cells expressing GFP-Tmod1-T54E display faint, inconsistent pointed-end assembly, whereas >66% of the cells expressing GFP-Tmod1 WT and GFP-Tmod1-T54A have clear, consistent pointed-end assembly. These data indicate that the addition of the T54E mutation in Tmod1 perturbs its assembly at the pointed ends of thin filaments.

Figure 4.

Expression of phosphorylation-mimic GFP-Tmod1-T54E does not shorten thin filament lengths in an way analogous to that of WT Tmod1. Transfected cells were fixed 4 d after transfection and stained with Texas Red-conjugated phalloidin to measure thin filament lengths. Only GFP-positive cardiomyocytes were analyzed. A) Data are from 3 cultures, and the values are presented as percentage thin filament length shortened compared with that of cells expressing GFP alone (means ± sd). It is documented that the expression of WT GFP-Tmod1 will restrict (shorten) thin filament lengths (14). Indeed, expression of GFP-Tmod1 WT and GFP-Tmod1-T54A shortened thin filaments compared with GFP alone (7.7 and 6.8%, respectively). However, cells expressing GFP-Tmod1-T54E had a significant reduction in the ability to shorten thin filament lengths compared with that of controls (3.1%). n.s., not significant. **P ≤ 0.005. B) Histogram of actin length measurements from a representative experiment (y axis, frequency; x axis, thin filament length). A gaussian curve fit of the length measurements shows nearly identical thin filament length distribution and frequency in cells expressing GFP-Tmod1 WT and GFP-Tmod1-T54A, which are shorter than those in cells expressing GFP alone. The length frequency and distribution in cells expressing GFP-Tmod1-T54E are similar to those observed in cells expressing GFP alone.

Figure 5.

Tmod1-T54E displays a modest disruption in actin filament-capping activity in the presence of long muscle α-tropomyosin in vitro. Influence of the T54E mutation on inhibition of the pointed-end elongation of gelsolin-capped actin filaments (5.6 nM) was examined using full-length Tmod1 in the presence of 1 μM stTM. A reduction in actin filament-capping activity with Tmod1-T54E compared with that for WT Tmod1 was observed.

Figure 6.

Loss of actin filament capping activity is not due to loss of tropomyosin-binding activity. Binding of N-terminal Tmod1 fragment Tmod1-T54E to a muscle αtropomyosin peptide (αTM1aZip) was determined by circular dichroism analysis. The temperature dependence of the ellipticity at 222 nm was measured using circular dichroism spectroscopy, and the dissociation constant was calculated from the curve. Complex formation of Tmod1-T54E and αTM1aZip was observed (see increase in the helical content and melting temperature). However, there is no significant difference in the dissociation constant between Tmod1-T54E and WT N-terminal Tmod1; this indicates that T54E mutation did not influence the interaction with the long muscle α-tropomyosin in vitro.