The recruitment of acetylated and unacetylated tropomyosin to distinct actin polymers permits the discrete regulation of specific myosins in fission yeast

Abstract

Tropomyosin (Tm) is a conserved dimeric coiled-coil protein, which forms polymers that curl around actin filaments in order to regulate actomyosin function. Acetylation of the Tm N-terminal methionine strengthens end-to-end bonds, which enhances actin binding as well as the ability of Tm to regulate myosin motor activity in both muscle and non-muscle cells. In this study we explore the function of each Tm form within fission yeast cells. Electron microscopy and live cell imaging revealed that acetylated and unacetylated Tm associate with distinct actin structures within the cell, and that each form has a profound effect upon the shape and integrity of the polymeric actin filament. We show that, whereas Tm acetylation is required to regulate the in vivo motility of class II myosins, acetylated Tm had no effect on the motility of class I and V myosins. These findings illustrate a novel Tm-acetylation-state-dependent mechanism for regulating specific actomyosin cytoskeletal interactions.

Fig. 1.

Growth and division of naa25Δ S. pombe cells. (A) Plates showing growth of naa25⁺ (left) and naa25Δ (right) cells when incubated at 25°C (top panel) or 30°C (middle panel). (B) Micrographs of DAPI-stained naa25⁺ (left) and naa25Δ (right) cells grown at 25°C. naa25Δ cells frequently fail to complete septation (arrowheads) and these cells often continue to grow at the septum and form branches during subsequent growth cycles (arrows). Scale bars: 10 μm. (C) Phase-contrast images of naa25⁺, naa25Δ, naa25Δ pREP41αSmTm and naa25Δ pREP41αSmTm-AS cells grown on minimal medium at 25°C. Scale bar: 5 μm. (D) Rhodamine-phalloidin staining of actin in naa25Δ cells grown in EMM2 minimal medium at 25°C compared with naa25⁺ cells grown under equivalent conditions (inset). Scale bar: 5 μm.

Fig. 2.

Unmodified Cdc8 associates with and affects the appearance of Cdc8–actin polymers in vivo and in vitro. (A) Cdc8 immunofluorescence staining of the naa25Δ strain using Cdc8 antiserum. Scale bar: 5 μm. (B–D) Negative staining of (B) F-actin–acetylated Cdc8, (C) actin–unmodified Cdc8, or (D) actin–cardiac Tm. Obliquely oriented strands characteristic of tropomyosin (Lehman et al., 1994) are observed in each sample. In contrast to cardiac Tm and acetylated-Cdc8-labelled actin filaments (B,D), breaks (arrowheads) are evident in the unacetylated Cdc8–actin polymers (C), which also often formed bundles (arrows). Samples were prepared using identical methods. Scale bars: 40 nm.

Fig. 3.

Acetylation-state-specific anti-Cdc8 antibodies reveal that only acetylated Cdc8 localises to the cytokinetic actomyosin ring. (A) Purified endogenous and recombinant Cdc8 proteins were used to characterise the anti-Cdc8^ACE antibody and the anti-Cdc8^UNACE antibody. Protein samples were run on a Coomassie Blue-stained gel to demonstrate equal loading. (B) Anti-Cdc8 (upper panel) and anti-Cdc8^ACE (lower panel) antibodies were used to probe naa25Δ and naa25⁺ cell extracts. The anti-Cdc8^ACE antibody only recognises Cdc8 from naa25⁺ cells. (C–G) Immunofluorescence of naa25⁺ (C,D,E) and naa25Δ (F,G) cells using anti-Cdc8 (C), anti-Cdc8^ACE (D,F) or anti-Cdc8^UNACE (E,G) acetylation-specific antibodies. Scale bars: 5 μm.

Fig. 4.

Class I myosin dynamics do not require acetylated Cdc8. (A,B) GFP signal from gfp-myo1 naa25⁺ (A) and gfp-myo1 naa25Δ (B) cells reveal Myo1 is recruited to the cell poles and septum in both strains. (C,D) Kymograph analysis of GFP-Myo1 in vivo dynamics in S. pombe naa25⁺ (C) and naa25Δ (D) strains. Kymographs are maximum projections of 100 five-z-slice stacks at different time points, each 250 mseconds apart. Scale bars: 5 μm.

Fig. 5.

Class V myosin movements do not require acetylated Cdc8. (A,B) GFP signal from gfp-myo52 naa25⁺ (A) and gfp-myo52 naa25Δ (B) cells reveal Myo52 is recruited to motile foci that concentrate at sites of cell growth in each strain. (C,D) Kymograph analysis of GFP-Myo52 in vivo motility in S. pombe naa25⁺ (C) and naa25Δ (D) strains. Kymographs are maximum projections of 100 five-z-slice stacks at different time points, each 250 mseconds apart. Scale bars: 5 μm.

Fig. 6.

Function of both S. pombe class II myosins require acetylated Cdc8. In vivo localisation of Myo2 in S. pombe naa25⁺ (A) and naa25Δ (B) cells. Myo2 was seen localised to abnormal contractile structures (arrowheads) and actin filaments not associated with the CAR in cells lacking Naa25. (C) Myo2 rings constricted within an hour of forming in naa25⁺ cells. (D) Myo2 often localised to contractile rings which failed to form properly or constrict in naa25Δ cells. (E,F) Anti-Myo2 (E) and anti-Myp2 (F) immunofluorescence of each endogenous class II myosin confirmed the live cell imaging data. Although each myosin only localised to distinct CAR structures during mitosis (insets), both proteins were seen associated with disrupted CAR structures (arrows) and interphase actin filaments (arrowheads) in naa25Δ cells. Scale bars: 5 μm.

Fig. 7.

Model comparing the effect that Tm acetylation has upon actin-based motility of class II and class V myosins. (A) Acetylation of Cdc8 permits the actomyosin interaction-driven cooperative movement of the Tm filament from its normal residency in the closed (red lines) into the open position (green lines) on an individual actin strand (grey circles). This facilitates the regulated cooperative binding of multiple motor domains of myosin II heavy chain filaments along the actin strand. (B) The 36 nm step taken by myosin V dimers results in each motor domain interacting with a separate actin strand. As 98% of acetylated Tm filaments occupy the closed or ‘off’ position on actin, each actomyosin interaction would be regulated by Tm and would therefore affect this dimeric motor's motility. (C) By contrast, significantly less (32%) of the of unacetylated Tm filaments (blue lines) occupy the closed position and therefore have negligible regulatory effect on myosin.