The formin Fhod3 is required for sarcomere reassembly after mitosis in postnatal cardiomyocytes

Abstract

Fhod3, a member of the formin family of proteins that regulate actin polymerization, is essential for the organization of cardiac sarcomeres during embryonic cardiogenesis. We previously demonstrated that the perinatal deletion of Fhod3 in mice results in lethality with disorganized sarcomeres, whereas the deletion in the adult period is non-lethal and does not affect sarcomere morphology. To clarify the underlying mechanism of this developmental stage-dependent outcome, we examined the effects of Fhod3 depletion in primary cultured cardiomyocytes derived from Fhod3^flox neonates. Long-range time-lapse imaging throughout the entire culture period revealed that Fhod3 depletion around birth using muscle creatine kinase-Cre caused sarcomere disruption only in cells that had undergone mitosis and not in cells that had not undergone mitosis, suggesting that sarcomere breakdown by Fhod3 deletion occurs after mitosis. To verify this, we varied the timing of Fhod3 deletion using a tamoxifen-inducible αMHC-MerCreMer line. Cells treated with tamoxifen from day 1 in vitro (DIV1) exhibited a disruption of sarcomeres, similar to that observed in the Fhod3-depleted cells by MCK-Cre around birth. In contrast, cells treated from DIV9, when postnatal mitosis was almost complete, showed organized sarcomere despite Fhod3 depletion. Thus, in the presence of Fhod3, sarcomere reassembly after mitosis was successful, whereas it failed in the absence of Fhod3, indicating the indispensable role of Fhod3 in sarcomere reorganization after mitosis.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00018-026-06340-9.

Introduction

The formin family proteins are a group of actin nucleation factors and play pivotal roles in controlling actin polymerization via the conserved formin homology (FH) domains 1 and 2 [1–3]. Through cooperation of the FH1 and FH2 domains, formins direct formation of straight actin filaments [4, 5], thereby modulating the actin dynamics in various actin structures, such as lamellipodia, filopodia, and contractile fibers [6]. Among 15 members of mammalian formins, Fhod3 has recently been identified as a newly emerging causative gene of human hypertrophic cardiomyopathy [7, 8]. However, its pathogenic mechanism remains unclear.

Fhod3 (formerly designated as Fhos2) is abundantly expressed in the heart [9], and plays an essential role on actin assembly in myofibrils of cultured cardiomyocytes [10–12]. We previously showed that genetic conventional deletion of Fhod3 in mice results in embryonic lethality due to defects in cardiogenesis [13]. Perinatal conditional knockout (KO) of Fhod3 in the heart also caused juvenile lethality at around day 10 after birth with enlarged hearts composed of severely impaired myofibrils, indicating that Fhod3 is crucial for postnatal heart development [14]. However, the tamoxifen-induced conditional KO of Fhod3 in the adult heart neither led to lethal effects nor had any substantial effect on sarcomere structure and localization of sarcomere components [14]. Thus, there appears to be a discrepancy in Fhod3 dependence between neonatal and adult cardiomyocytes; however, the underlying mechanism is not yet elucidated.

Developing cardiomyocytes in the embryonic stage acquire contractile activity by producing myofibrils with a regular array of sarcomeres. During this de novo assembly of sarcomeres, precursors are produced at first in area close to plasma membrane and maturation of sarcomeres occur in a contractility-dependent manner [15–17]. Despite the presence of some differences [18], fundamental principles seem to be shared between this de novo assembly in situ and the sarcomere assembly of cultured cardiomyocytes in vitro [15]; cultured cardiomyocytes exhibit the similar sequence of events involved in the assembly of sarcomeric proteins into myofibrils following plating on culture dishes. Therefore, a significant body of research using cultured cardiomyocytes prepared from embryos and neonates [19–21], as well as cardiomyocytes derived from induced pluripotent stem cells [22, 23], has elucidated the intricacies of the sarcomere assembly process. The actin polymerization factor Fhod3 is essential for both de novo assembly in developing hearts [13] and the sarcomere reassembly after plating of cultured cardiomyocytes[10, 22], supporting the idea that the de novo assembly in developing hearts and the sarcomere reassembly after plating of cultured cardiomyocytes share common mechanisms including actin polymerization factors.

Subsequent to the de novo formation of sarcomeres within the developing heart, sarcomeres need to maintain their structural integrity to ensure the continued function of blood pumping. At the same time, along with the proliferation of cardiomyocytes, sarcomeres repeat a cycle of disassembly and reassembly during mitosis. It remains to be elucidated whether Fhod3 plays a role in both of these two seemingly contradictory processes. In the case of the former, Fhod3 would be involved in maintaining the structural integrity of sarcomeres. If the latter is the case, Fhod3 would be involved in sarcomere reassembly after mitosis of proliferating cardiomyocytes. In contrast to the commonality of the sarcomere assembly of cultured cardiomyocytes after plating with the de novo assembly, there seems to be some differences between these processes and post-mitotic sarcomere reassembly. For example, the process of sarcomere reassembly after mitosis is completed in an extremely short time compared to de novo assembly [24, 25], suggesting that the reconstruction process during mitosis differs from the process of de novo formation. However, the details have yet to be fully elucidated.

In the present study, we deleted the Fhod3 gene in primary culture of cardiomyocytes prepared from Fhod3^flox neonatal mice and examined the effect of Fhod3 depletion on sarcomere structures. Effects of different timings of Fhod3 gene deletion, i.e. perinatal deletion using muscle creatine kinase (MCK)-Cre and drug-induced deletion using α-myosin heavy chain (αMHC)-MerCreMer, on sarcomeric structures were then analyzed. The combination of long-range time-lapse imaging over 10 days and immunostaining revealed a previously unobserved significant relationship between Fhod3 deletion and failure of sarcomere reassembly following mitosis. The current findings highlight the indispensable role of Fhod3 in the process of sarcomere reorganization, which is a requisite phase of cell cycle progression in cardiomyocytes.

Results

MCK-Cre-mediated Fhod3 deletion in primary culture of cardiomyocytes

To examine the effect of Fhod3 depletion on sarcomeres in cultured cardiomyocytes, we utilized MCK-Cre mice, which express Cre recombinase in the heart from the late-embryonic stage to adulthood [26]. By crossing Fhod3^flox/flox mice with Fhod3^+/–; MCK-Cre^Tg mice, we obtained Fhod3 conditional knockout (cKO) neonates (Fhod3^flox/–; MCK-Cre^Tg) and control littermates (Fhod3^flox/+) (Fig. 1A). The Fhod3 cKO mice were born in the expected Mendelian ratio and were indistinguishable in appearance from control littermates at birth but died by around postnatal day 15 (P15) [14]. We prepared primary cultures of cardiomyocytes from P0 neonates. As expected, the Cre-mediated recombination occurred under culture conditions (Fig. 1B), consistent with previous observations in vivo [14], and the Fhod3 protein exhibited a decrease at DIV10 (Fig. 1C). An immunofluorescent analysis revealed that cKO cells actually showed reduced Fhod3 immunosignals and disorganized sarcomeres (Fig. 1D). The characteristic C-zone pattern of Fhod3 localization [10], which was observed in control cells, was disappeared in cKO cells. However, residual signals of Fhod3 were still observed in the perinuclear region of Fhod3-depleted rounded cardiomyocytes (Fig. 1D). This perinuclear pattern has also been previously observed in Fhod3-depleted rat cardiomyocytes [10]. However, in the HA-tagged Fhod3 knock-in mouse model, the perinuclear pattern was only observed when the anti-Fhod3 antibody was used; it was not observed with the anti-HA antibody (Fig. S1), indicating that this perinuclear staining is non-specific background with the anti-Fhod3 antibody. Therefore, in this study, we evaluated the positivity of Fhod3 based on the presence or absence of the Fhod3-specific clear C-zone pattern (doublet between Z-lines) in Fhod3 immunostaining.

Fig. 1.

MCK-Cre mediated deletion of Fhod3 in primary culture of cardiomyocytes. A Preparation of primary culture from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) and control littermate (Fhod3^flox/+) P0 neonates. Partially created with BioRender.com (https://www.biorender.com). B PCR analyses of genomic DNAs prepared from cultured cardiomyocytes at DIV8. Small black arrows “a”– “d” indicate primers used. The PCR product from the wild-type allele (Fhod3⁺) or the null allele (Fhod3^–) was 329 bp in length (#3). The PCR product from the floxed allele (Fhod3^flox) was 387 bp in length (#2). The 549-bp fragment was produced by Cre-mediated recombination (#1). C Detection of the Fhod3 protein by immunoblot analysis. Proteins prepared from cultured cardiomyocytes of Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) and control littermate (Fhod3^flox/+) P0 neonates at DIV10 were analyzed by immunoblot with the anti-Fhod3-(873–974). The loading amounts were verified by immunoblot with anti-GAPDH antibody. D Confocal fluorescence micrographs of cultured cardiomyocytes from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) and control littermate (Fhod3^flox/+) neonates. Cardiomyocytes at DIV8 were stained with the anti-α-actinin (red) and anti-Fhod3-(650–802) (green) antibodies and Hoechst (blue). Scale bars, 20 μm

In Fhod3-depleted cardiomyocytes, the Z-band marker α-actinin showed a punctate pattern, which is in marked contrast with the well-organized striated pattern of α-actinin in control cells (Fig. 1D). Fhod3-depleted cardiomyocytes also showed a rounded cell shape, as observed in siRNA-mediated knocked down cardiomyocytes [10]; the morphological difference was striking compared to control cardiac myocytes, which exhibited a polygonal or star-shaped morphology with aligned myofibrils. The present observed phenotypes in primary cultured mouse cardiomyocytes induced by Cre-mediated Fhod3-depletion were consistent with those observed in rat neonatal cardiomyocytes with siRNA-mediated Fhod3-knockdown [10] and with those observed in cardiac sections of Cre-mediated Fhod3-depleted mice [14]. Thus, Cre-mediated Fhod3 depletion in cultured cardiomyocytes seems to induce sarcomere disorganization and concomitant alternations in cell morphology.

Monitoring of Fhod3-depleted cardiomyocyte culture

To investigate the time course of morphological changes of cardiomyocytes associated with Fhod3 depletion, we monitored cell morphology throughout the entire culture period using an incubation monitoring system (Fig. 2A). At the end of culture on DIV 7–15, cells were fixed and immunostained for Fhod3 and α-actinin to evaluate Fhod3 expression and sarcomere organization. As shown in Fig. 2B, cardiomyocytes from cKO neonates showed significant depletion of Fhod3, which was determined by the presence or absence of the Fhod3-specific clear C-zone pattern as described above, and sarcomere disorganization at the end of the monitoring period. The immunofluorescent images were subsequently aligned with the final frame of monitoring images to identify the correspondence of cells (Fig. 2C and D). In contrast to control cells showing well-organized sarcomeres and periodic localization of Fhod3 (Fig. 2C), Fhod3-depleted cardiomyocytes typically showed disorganized sarcomeres with a flattened and rounded morphology (Fig. 2D middle and lower panels; cells outlined with solid lines). However, some cardiomyocytes showed well-organized sarcomeres despite Fhod3 depletion (Fig. 2D top and middle panels; cells outlined with dotted lines). Thus, depletion of Fhod3 did not always lead to sarcomere disruption.

Fig. 2.

Long-term monitoring experiment of Fhod3-depleted cardiomyocytes with MCK-Cre. A Primary cultures prepared from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) and control littermate (Fhod3^flox/+) P0 neonates were monitored every 3 h with the incubation monitoring system CM30 housed within a standard CO2 incubator for 7–15 days. Partially created with BioRender.com (https://www.biorender.com). B The percentage of cardiomyocytes that are positive for Fhod3 expression (left) and that are positive for well-organized sarcomeres (right) determined from immunofluorescent images at the end of culture on DIV 7–15 (experiment #1, DIV15; experiment #2, DIV7; experiment #3, DIV12). The criteria for positivity of Fhod3 and sarcomere were determined as described under “Materials and Methods.” For each genotype, the ratios of positive cells were calculated from a total of 43–101 cells in a single monitoring experiment, and the mean ± SEM was obtained from three independent monitoring experiments with cardiomyocytes prepared separately. *, p < 0.05. C Representative immunofluorescent images of cultured cardiomyocytes from control littermate (Fhod3^flox/+) P0 neonates at DIV7 (left panels) and the corresponding final frame of monitoring images (right panel). Lower panels show the magnified views of the boxed areas in upper panels. Scale bars, 20 μm. D Representative images of cultured cardiomyocytes from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) P0 neonates at DIV7 (left panels) and the corresponding final frame of monitoring images (right panel). In middle panels, cells with well-organized sarcomeres and cells with disorganized sarcomeres are outlined with dotted and solid lines, respectively. Upper and lower panels show the magnified views of the boxed areas, with dotted and solid lines, respectively. Scale bars, 20 μm

Monitoring of Fhod3-depleted cardiomyocyte revealed the post-mitotic disorganization of sarcomeres

To elucidate the relationship between Fhod3 depletion, sarcomere disruption, and cell morphology, we tracked morphological changes of cardiomyocytes in monitoring images throughout the entire culture period. In monitoring brightfield images, which were acquired under epi-oblique illumination, the nucleus of cardiomyocyte was identified as a rounded convex structure with approximate diameter of 20 μm, and the mitotic event was successfully identified as the process of division of this bulge (Fig. 3A left panels; Movie S1). Following mitosis, in many cells, the resultant two bulges corresponding to two daughter nuclei fused into a single rounded convex structure shortly following mitosis without cytokinesis, as observed in previous study [27]. Immunostaining at the end of the monitoring period revealed that these cells were binucleated but their two nuclei were very close to each other (Fig. 3A right panels). Although the occurrence was rare, some subsets of cells divided into two daughter cells with completion of cytokinesis (data not shown).

Fig. 3.

Time-lapse imaging reveals the temporal relationship between mitosis and sarcomere disruption in Fhod3-depleted cardiomyocytes. A–B Representative time-lapse images of mitotic events in cardiomyocytes prepared from control littermate (Fhod3^flox/+) A and Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) B P0 neonates (left panels) and the final frame of monitoring images and corresponding immunofluorescent images at the end of culture (right panel). Scale bars, 20 μm. Yellow arrowheads indicate nuclear division. C–D The temporal relationship between the timing of mitosis and sarcomere disruption. Cultured cardiomyocytes from control littermate (Fhod3^flox/+) C and Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) D neonates were classified into two groups based on the presence or absence of Fhod3 signals in immunofluorescence images at the end of culture: Fhod3-positive and Fhod3-negative groups (upper and lower). Subsequently, each cardiomyocyte was plotted against the timing of mitotic events (DIV), with color-coded levels of sarcomere organization: deep blue, excellent; sky blue, high; pink, low; red, poor. The criteria for levels of sarcomere organization were determined as described under “Materials and Methods.” Three replicates of an independent monitoring experiment using different sets of mice were represented by distinct dots (circle, triangle and square). The total number of cells analyzed for control and Fhod3 cKO cells from these replicates were 137 and 245, respectively

After mitosis, their polygonal shape of control cardiomyocytes was maintained (Fig. 3A; Movie S1). In contrast, cKO cells showed cell shape changes following mitosis, then became flat and rounded shape (Fig. 3B; Movie S2). To verify that these morphological changes in cKO cells are associated with mitosis, we comprehensively analyzed the timing of mitosis and investigated its correlation with sarcomere disorganization at the fix time point. In control cells (Fig. 3C), sarcomeres were well organized, irrespective of the timing of mitosis or its absence. In contrast, the majority of cKO cells exhibited disorganized sarcomeres, although Fhod3 was not depleted in a small subset of cKO cells, which exhibited organized sarcomeres (Fig. 3D). Importantly, among Fhod3-depleted cardiomyocytes, cells that had not undergone mitosis showed well-organized sarcomeres despite the absence of Fhod3. These findings suggest the possibility that mitosis is involved in sarcomere disorganization induced by Fhod3 depletion. Since sarcomeres are disassembled during mitosis and subsequently reassembled following this cellular event, it is possible that Fhod3 is required for sarcomere reassembly after mitosis.

4-hydroxytamoxifen-induced deletion of Fhod3 in primary culture of cardiomyocytes

To elucidate the relationship between mitosis and sarcomere disruption by Fhod3 depletion, we attempted to vary the timing of Fhod3 deletion in cultured cardiomyocytes. In the present conditions of primary culture, cardiomyocytes typically undergo mitosis and exit the cell cycle by around DIV10 at the latest (Fig. 3C). If Fhod3 is depleted after mitosis, it is expected that Fhod3 depletion will not cause sarcomere disruption. To explore this hypothesis, we adapted the αMHC-MerCreMer/loxP system [28]. Using this system, we had previously effectively deleted Fhod3 from mouse hearts in vivo [14]. In the present study, we prepared primary culture of cardiomyocytes from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg neonates (Fig. 4A). The addition of 0.5 μM of 4-hydroxytamoxifen (4-OHT), an active metabolite of tamoxifen, to cultured cardiomyocytes from DIV1 successfully caused Cre-mediated deletion of the Fhod3 gene (Fig. 4B), and the levels of Fhod3 protein decreased (Fig. 4C). An immunofluorescent analysis revealed that Fhod3 depletion by 4-OHT also resulted in disorganized sarcomeres with a rounded cell shape (Fig. 4D). Recent findings demonstrated that cardiac troponin T accumulates in the submembranous region of rounded mitotic cardiomyocytes in association with sarcomere disassembly [29]. Analogously, in Fhod3-deficient cells, troponin T also accumulated in the submembrane region.

Fig. 4.

4-OHT-induced deletion of Fhod3 in primary cultured cardiomyocytes. A Preparation of primary culture from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates. Partially created with BioRender.com (https://www.biorender.com). B PCR analyses of genomic DNAs prepared at the indicated DIV from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg cardiomyocytes treated with DMSO or 4-OHT. The PCR product from the floxed allele (Fhod3^flox) was 387 bp in length (#2). The 549-bp fragment was produced by Cre-mediated recombination (#1). Primers used are shown in Fig. 1B. C Detection of the Fhod3 protein by immunoblot analysis. Proteins prepared at the indicated DIV from cultured cardiomyocytes of Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates were analyzed by immunoblot with the anti-Fhod3-(873–974). The loading amounts were verified by immunoblot with anti-GAPDH antibody. D Representative immunofluorescent images of cultured cardiomyocytes from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates. Cardiomyocytes were treated with 0.5 μM 4-OHT from DIV1, then fixed at DIV10 and stained with the anti-troponin T (green) and anti-Fhod3-(650–802) (green) antibodies and Hoechst (blue). Scale bars, 20 μm

The effects of different timings of Fhod3 deletion in cultured cardiomyocytes

Using this αMHC-MerCreMer/loxP system, we compared the effects of different timings of Fhod3 deletion in cultured cardiomyocytes. Cardiomyocytes treated with 4-OHT from DIV1 exhibited Fhod3 depletion and sarcomere disorganization when compared with control cells (Fig. 5A), while cardiomyocytes treated with 4-OHT from DIV9 exhibited well-organized sarcomeres similar to control cells despite the Fhod3 depletion (Fig. 5B; Fig. S2A). To verify this observation, cardiomyocytes treated with 4-OHT from DIV1 or DIV9 were fixed and immunostained for Fhod3 and α-actinin at every 4 days (Fig. 5C). Subsequently, the sarcomere structure and Fhod3 localization in all images from all timepoints were evaluated in a double blind manner. As shown in Fig. 5D, 4-OHT treatment from DIV9 caused the same level of Fhod3 deletion with 4-OHT treatment from DIV1; however, the former led to relatively minor disorganization of sarcomeric structures compared to the latter (Fig. 5E), although not to a statistically significant degree (Fig. S2B and C). One potential explanation for the absence of statistical significance could be attributed to the cellular damage resulting from the prolonged culture period. Indeed, even in the control group (DMSO DIV9), the sarcomere positivity rate began to decline after DIV22. Consequently, these findings support our hypothesis that disruption of sarcomere structure was dependent on the timing of Fhod3 depletion.

Fig. 5.

Effect of varying the timing of 4-OHT treatment on sarcomeric structure. (A-B) Representative immunofluorescent images of cultured cardiomyocytes from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates. Cardiomyocytes were treated with 0.5 μM 4-OHT from DIV1 A and from DIV9 B, then fixed at DIV22 and stained with the anti-α-actinin (red) and anti-Fhod3-(650–802) (green) antibodies and Hoechst (blue). Scale bars, 20 μm. C Cardiomyocytes prepared from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates were treated with 0.5 μM 4-OHT from DIV1 or from DIV9, then fixed and immunostained at the indicated DIV. D-E The percentage of cardiomyocytes that are positive for Fhod3 expression D and that are positive for well-organized sarcomeres E determined from immunofluorescent images at the indicated DIV. The evaluation of images was conducted by the researcher blinded to conditions, with the criteria described in “Materials and Methods.” The ratios of positive cells were calculated from a total of 55–197 (119 ± 36) cells in a single experiment for each condition (total 7,154 cells), and the mean ± SEM was obtained from three independent experiments with cardiomyocytes prepared separately

Monitoring of Fhod3-depleted cardiomyocyte treated with 4-OHT at different time points

Next, we examined the temporal relationship between Fhod3 depletion, sarcomere disruption, and mitosis in the αMHC-MerCreMer/loxP system. Cardiomyocytes prepared from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates were treated with 4-OHT or DMSO from DIV1 or DIV9 and were monitored using the incubation monitoring system (Fig. 6A and B). As anticipated, the monitoring movie demonstrated that Fhod3-depleted cardiomyocytes exhibited an alternation of cell shape following mitosis, subsequently adopting a flattened and rounded morphology, while cardiomyocytes without mitosis retained their polygonal morphologies with their thick cell bodies (Movie S3; Table S1). At the end of the culture period, cells were fixed and immunostained to analyze Fhod3 expression and sarcomere organization. As shown in Fig. 6C, cardiomyocytes treated with 4-OHT from DIV9 exhibited a reduction of Fhod3 expression comparable to that observed in cells treated from DIV1; however, although the difference was not statistically significant, the former exhibited a tendency towards higher levels of sarcomere organization than the latter (Fig. 6D), again supporting the hypothesis that disruption of sarcomere structure was dependent on the timing of Fhod3 depletion. Subsequently, immunofluorescent images were referenced with time-lapse images to analyze the relationship between the mitotic events and sarcomere organization. As observed in MCK-Cre-mediated depletion, cells treated with 4-OHT from DIV1 exhibited post-mitotic sarcomere disruption when compared with DMSO-treated control cells (Fig. 6E). In contrast, cells treated with 4-OHT from DIV9 showed a less pronounced disorganization of sarcomeres (Fig. 6F). Notably, among Fhod3-negative cells treated with 4-OHT from DIV9, cells that had undergone mitosis until roughly DIV9 showed organized sarcomeres despite the absence of Fhod3 at the time point of immunostaining, whereas cells that had undergone mitosis after roughly DIV9 showed disorganized sarcomeres. These findings suggest that the successful reassembly of sarcomeres after postnatal mitosis is dependent on Fhod3.

Fig. 6.

Long-term monitoring experiment of Fhod3-depleted cardiomyocytes with αMHC-MerCreMer. A-B Primary cultures prepared from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg P0 neonates were monitored every 3 h with the incubation monitoring system CM30 housed within a standard CO₂ incubator for 17–22 days. Partially created with BioRender.com (https://www.biorender.com). C-D The percentage of cardiomyocytes that are positive for Fhod3 expression C and that are positive for well-organized sarcomeres D determined from immunofluorescent images at the end of culture on DIV 17–22 (experiment #1, DIV22; experiment #2, DIV18; experiment #3, DIV17). The criteria for positivity of Fhod3 and sarcomere were determined as described under “Materials and Methods.” For each condition, the ratios of positive cells were calculated from a total of 4–60 cells in a single monitoring experiment, and the mean ± SEM was obtained from three independent monitoring experiments with cardiomyocytes prepared separately. *, p < 0.05; **, p < 0.01; ns, not significant. E–F The temporal relationship between the timing of mitosis and sarcomere disruption in cardiomyocytes treated with DMSO or 0.5 μM 4-OHT from DIV1 E and from DIV9 F. Cultured cardiomyocytes from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg neonates were classified into two groups based on the presence or absence of Fhod3 signals in immunofluorescence images at the end of culture: Fhod3-positive and Fhod3-negative groups. Subsequently, each cardiomyocyte was plotted against the timing of mitotic events (DIV), with color-coded levels of sarcomere organization: deep blue, excellent; sky blue, high; pink, low; red, poor. The criteria for levels of sarcomere organization were determined as described under “Materials and Methods.” Three replicates of an independent monitoring experiment using different sets of mice were represented by distinct dots (circle, triangle and square). The total number of cells analyzed for DMSO from DIV1, 4-OHT from DIV1, DMSO from DIV9, and 4-OHT from DIV9 from these replicates was 95, 131, 64 and 140, respectively

The effects of blocking the cell cycle at the G2/M border on sarcomeric structures in Fhod3-depleted cardiomyocytes

Finally, we examined whether the blocking of the cell cycle before entering into mitosis can prevent the sarcomere disassembly in Fhod3-depleted cardiomyocytes. Cardiomyocytes prepared from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) P0 neonates (Fig. 7A) were treated with or without a selective CDK1 inhibitor, RO3306 [30], for 16 h prior to fixation (Fig. 7B). 16 h treatment of rat neonatal cardiomyocytes with RO3306 has been reported to pause cell cycle at the G2/M checkpoint [25]. As shown in Fig. 7C, Fhod3 cKO cardiomyocytes treated with RO3306 exhibited lower percentage of binucleated cells than those of control cells, indicating the RO3306 treatment successfully prevent cells from entering mitosis. The RO3306 treatment did not affect the ratio of Fhod3 positive cells (Fig. 7D), while the treatment significantly attenuated the ratio of sarcomere disorganization (Fig. 7E). Of note, Fhod3 depleted cardiomyocytes in RO3306-treated group still exhibited well-organized sarcomeres (Fig. 7F and G), although DMSO-treated Fhod3 depleted cardiomyocytes showed disorganized sarcomeres (Fig. 7H). These findings support our hypothesis that Fhod3 depletion does not affect existing sarcomeres; instead, it selectively exerts an effect on the process of sarcomere reassembly after mitosis.

Fig. 7.

Effect of blocking the cell cycle in the G2/M phase on sarcomere disassembly in Fhod3-depleted cardiomyocytes. A Preparation of primary culture from Fhod3 cKO (Fhod3^flox/–; MCK-Cre^Tg) P0 neonates. Partially created with BioRender.com (https://www.biorender.com). B Cardiomyocytes were treated with 10 μM RO3306 or DMSO for 16 h before the fixation at DIV3. C The percentage of binucleated cardiomyocytes determined from immunofluorescent images at the end of culture on DIV3. D–F The percentage of cardiomyocytes that are positive for Fhod3 expression D, that are positive for well-organized sarcomeres E, and that are positive for well-organized sarcomeres in Fhod3 negative cardiomyocytes F, determined from immunofluorescent images at the end of culture on DIV3. The criteria for positivity of Fhod3 and sarcomere were determined as described under “Materials and Methods.” For each condition, the ratios of positive cells were calculated from a total of 67–92 cells in a single monitoring experiment, and the mean ± SEM was obtained from four independent monitoring experiments with cardiomyocytes prepared separately. **, p < 0.01; ns, not significant. G-H Representative immunofluorescent images of cultured cardiomyocytes treated with 10 μM RO3306 G or DMSO H. Cardiomyocytes fixed at DIV3 were stained with the anti-α-actinin (red) and anti-Fhod3-(650–802) (green) antibodies and Hoechst (blue). Scale bars, 20 μm. Arrows and arrowheads indicate Fhod3 negative cells and positive cells, respectively

Discussion

In the present study, we defined the temporal relationship between sarcomeric disruption caused by Fhod3 depletion and mitosis in mouse neonatal cardiomyocytes using a long-range cell culture monitoring system. The majority of cardiomyocytes prepared from P0 neonates completed their final mitosis within 10 days following seeding. The depletion of Fhod3 after the last mitosis appeared to demonstrate minimal effect on sarcomeres, whereas depletion before the last mitosis resulted in a failure of sarcomere reassembly, leading to cellular morphological changes exhibiting a flat, rounded shape (Fig. 8). Thus, Fhod3 depletion appears to exert a selective effect on the process of sarcomere reassembly, without affecting existing sarcomeres .

Fig. 8.

Schematic representation of the involvement of Fhod3 in the final mitosis after birth. In the presence of Fhod3, sarcomere reassembly after mitosis is successful. The depletion of Fhod3 after the last mitosis demonstrates minimal effect on sarcomeres, whereas depletion before the last mitosis results in a failure of sarcomere reassembly. Partially created with BioRender.com (https://www.biorender.com)

The present conclusion that Fhod3 is involved in post-mitotic reassembly of sarcomere explains our previous findings in conventional [13] and conditional knockout mice [14]. During the embryonic stage, sarcomeres within cardiomyocytes repeat disassemble and reassembly in every cell proliferation cycle [24]. Therefore, Fhod3 is absolutely required for cardiac development during this period. Conversely, shortly after birth, mouse cardiomyocytes undergo binucleation following a final mitosis without cytokinesis [31]. Consequently, cardiomyocytes in P2 mouse neonates are predominantly mononucleated, whereas approximately 80% of cardiomyocytes become binucleated by P11 [32]. During this binucleation process, sarcomeres are disassembled at least partially [33], indicating that reassembly occurs subsequent to binucleation. It is therefore deduced that cardiomyocytes require Fhod3 until the final exit of cell cycle but no longer require it thereafter.

Discriminating whether Fhod3 functions to maintain the structural integrity of sarcomeres or plays a role in a sole step of post-mitotic reassembly is crucial for understanding the pathophysiology of Fhod3 in human genetic cardiomyopathies. This is because, although genetic mutations in Fhod3 gene have emerged as a new cause of cardiomyopathy [34], the molecular mechanisms are still largely unclear. To the best of our knowledge, this is the first report to demonstrate that Fhod3 is specifically involved in the process of reassembly after mitosis, and it is also the first to identify a critical factor for the post-mitotic reassembly process. The present finding of the critical role of Fhod3 on sarcomere reassembly is expected to accelerate research into the molecular mechanisms underlying Fhod3-associated cardiomyopathies by introducing a new conceptual framework regarding its etiology. This is of considerable significance because the involvement of sarcomere reassembly processes in the pathophysiology of cardiomyopathy has heretofore been unrecognized. Furthermore, the process of disassembly and reassembly of sarcomeres during mitosis of cardiomyocytes seems to be closely associated with the regenerative capacity of these cells [35–37]. Nevertheless, the precise mechanisms underlying these associations to be fully elucidated. At least, the process of disassembly and reassembly during the mitosis differs from de novo formation in several aspects. In the case of post-mitotic reassembly, all essential components seem to be present, although they have yet to be integrated [25]. The elucidation of cell cycle-dependent regulation of sarcomeric structures promises to unveil novel methodologies for fostering regenerative capacity of cardiomyocytes in the forthcoming future.

We have used two types of Cre-loxP systems to delete the Fhod3 gene in cultured cardiomyocytes prepared from P0 neonates. MCK-Cre, which is expressed around birth [26], considerably depleted Fhod3 before the final mitosis occurring around P7. For Fhod3 deletion at an arbitrary time, we used a tamoxifen-inducible MerCreMer mice line [28]. In this system, it has been recognized that high-dose tamoxifen not only induces MerCreMer-mediated recombination but also causes undesired loxP-independent effects due to Cre accumulation into nucleus [38, 39], as well as direct Cre-independent effects on contractility and calcium handling of cardiomyocytes [40]. However, in the present study, Fhod3 depletion and sarcomere disruption were sufficiently induced at 0.5 μM of 4-OHT, a low concentration that does not affect cardiomyocyte function [40], indicating that the direct effects of tamoxifen on cardiomyocytes are not involved in the present results. In addition, to exclude unintended effects due to nuclear accumulation of MerCreMer, we verified the effect of 0.5 μM 4-OHT on cardiomyocytes prepared from Fhod3^+/+; αMHC-MerCreMer^Tg/Tg neonates and confirmed no effects on sarcomere structure (Fig. S3). Thus, in the present experiment with cardiomyocytes prepared from Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg neonates, the differences observed with and without 0.5 μM of 4-OHT treatment were attributed to Fhod3 depletion.

Several limitations associated with the present study warrant mention. Firstly, it is important to note that the present study may have been unable to detect all events of mitosis. In the present long-range time-lapse imaging over 10 days, images were acquired every 3 h, a minimum interval for the autofocus mode of the incubation monitoring system CM30, which was needed to delineate the process of nuclear division. Since the process of nuclear division (mitosis) in cultured mouse neonatal cardiomyocytes requires approximately 3 h to complete [41], it was anticipated that the majority of events of nuclear division would be detected. However, it should be noted that events completed within three hours may not be captured. In addition, following the initial seeding of prepared cardiomyocytes, the process of spreading on substrates requires a duration of 9–12 h. Therefore, mitotic events that occurred during this period were also not identified in the present study. Second, due to the inherent limitation of unlabeled monitoring, we could not directly monitor nuclear division and alternations of sarcomeric structure during mitosis. Instead, we only monitored changes in cell morphology, i.e. the movement of nuclear bulge associated with nuclear division, as well as the flattening and rounding associated with sarcomere disruption. Although we evaluated sarcomeric structure based on the immunofluorescence staining after completing the culture, the process of sarcomere reassembly or the failure of reorganization itself was not directly observed. Furthermore, the resolution limitations of non-labeled monitoring images precluded observation of the intricacies of mitotic events, such as re-fusion of nuclei. Finally, despite the quantitative experimental data exhibiting substantial variability, certain experiments failed to detect statistically significant differences due to the limited number of replicates. To facilitate more detailed observation, future improvements and refinements to the observation system are necessary.

Intriguingly, MCK-Cre-mediated cKO cells exhibited a higher proportion of cells that had undergone mitosis compared to control cells (Fig. 3D). This tendency was also observed in Fhod3^flox/flox; αMHC-MerCreMer^Tg/Tg cardiomyocytes treated with 4-OHT from DIV1 (Fig. 6E) but not in those treated with 4-OHT from DIV9 (Fig. 6F). The present findings suggest that the absence of Fhod3 during early perinatal period may promote entry into the cell-cycle. However, as discussed below, the present study did not detect all mitotic events. Therefore, this hypothesis should be carefully explored in a future study.

Fhod3 has been identified as a causative gene of human hypertrophic cardiomyopathy [7, 8]. However, the precise pathogenic mechanism remains unclear. The present conclusion that Fhod3 plays an essential role specifically in the process of sarcomere reassembly following mitosis might provide important insights into this mechanism. In consideration of the preceding findings that Fhod3 is implicated in sarcomere assembly during cardiogenesis [13], it appears compelling that Fhod3 is a pivotal factor in sarcomere assembly, irrespective of whether the assembly is de novo or reassembly. The process of sarcomere assembly occurs not only during cardiogenesis and cell cycle progression but also during physiological and pathological hypertrophic processes, manifesting as sarcomere expansion. The impairment of these events appears to contribute to the progression of cardiomyopathies. Further exploration is warranted to provide a more precise understanding of the mechanisms involved.

Materials and methods

Mice

Mice heterozygous for the constitutive null Fhod3 allele (Fhod3^+/–) (Accession No. CDB0598K: https://large.riken.jp/distribution/mutant-list.html) were generated by replacement of exon 1 with lacZ as previously described [13]. Mice carrying the Fhod3 floxed allele (Accession No. CDB0927K) were generated by introducing loxP sequences flanking exon 18 of Fhod3 gene as previously described [14]. Transgenic mice expressing Cre recombinase under the control of muscle myosin kinase (MCK) promoter (MCK-Cre mice, B6.Cg-Tg(Ckmm-cre)5Khn/J;006475) and transgenic mice expressing Cre recombinase fused to two mutated estrogen receptors under the control of α-mysin heavy chain (αMHC) promoter (αMHC-MerCreMer mice, B6.Cg-Tg(Myh6-cre/Esr1*)1Jmk/J; 005657) were purchased from The Jackson Laboratory and were used for deletion of Fhod3 [14]. The HA-tagged Fhod3 knock-in mouse (Fhod3^HA/HA) expressing three copies of the human influenza hemagglutinin (HA) tag sequence integrated at the N-terminus of Fhod3 was generated using CRISPR/Cas9 editing. A detailed description of this mouse will be provided in a future publication.

All the experimental protocol was approved by the Animal Care and Use Committee of Miyazaki University (Permit Number: 2020-507-7). All mice were housed and maintained in a specific pathogen-free animal facility at University of Miyazaki, and all efforts were made to minimize the number of animals used and their suffering. All experiments were performed in strict accordance with the guidelines for Proper Conduct of Animal Experiments (Science Council of Japan) and the “Guide for the Care and Use of Laboratory Animals” published by the US National Institute of Health.

Primary culture of mouse neonatal cardiomyocytes

Primary cultures of cardiomyocytes were prepared from ventricles of neonatal mice as described previously [42] with minor modifications. Briefly, hearts were isolated from P0 neonates, dissected and then digested with collagenase type II (300 units/ml, Worthington CLS2) for three successive periods of 10 min each at 37°C. Cells were preplated for 70 min into 90-mm culture dishes in DMEM supplemented with 10% FCS to reduce the number of non-myocytes. Unattached cells were plated and cultured in DMEM with 10% FCS. The purity of myocytes in the culture was greater than 90%, which was estimated by immunofluorescence staining with an anti-sarcomeric actin antibody as described previously [10].

The standard culture medium was DMEM with 10% FCS. The medium until DIV 4 also contained 0.1 mM bromodeoxyuridine (BrdU) to prevent proliferation of non-cardiomyocytes [43]. The maintenance medium after DIV5 was DMEM with 5% FCS.

For blocking the cell cycle in the G2/M phase, cardiomyocytes were treated with a CDK1 inhibitor RO3306 (Selleck Biotech) for 16 h prior to fixation.

Analysis of Cre-mediated recombination of Fhod3 gene

The MCK-Cre-mediated recombination was induced by the perinatal expression of MCK-Cre, with no additional treatment. MerCreMer-mediated recombination was induced by the addition of 0.5 μM of 4-hydroxytamoxifen (4-OHT; nacalai tesque) to the culture medium from either DIV1 or from DIV9. The Cre-mediated recombinatorial deletion of exon 18 of the floxed Fhod3 allele was confirmed by PCR analysis of genomic DNA as described previously [14]. Briefly, genomic DNAs were extracted from cardiomyocytes and Cre-mediated recombination was analyzed by PCR using the following primers shown in Fig. 1B: a, 5′-GGGTTGCAGAGAATCACCCAG-3′; c, 5′-CCAGACTCGAGCACCTGTTTG-3′; and d, 5′-TGGTCCCTCACTTGAGGTGAC-3′.

Antibodies

Rabbit anti-Fhod3 polyclonal antibodies anti-Fhod3-(650–802) and anti-Fhod3-(873–974) were prepared as described previously [9]. The mouse monoclonal antibody against α-actinin (clone EA-53) was purchased from Sigma; mouse monoclonal antibody against cardiac troponin T (clone 13-11) was from BD Biosciences. The rat monoclonal antibody against HA (clone 3F10) was purchased from Roche. Alexa Fluor 488-conjugated F(ab')₂ fragment of anti-mouse IgG and Alexa Fluor 555-conjugated F(ab')₂ fragment of anti-rabbit IgG were purchased from Cell Signaling Technology. Alexa Fluor 488-conjugated F(ab')₂ fragment of anti-rat IgG was purchased from Jackson ImmunoResearch.

Immunoblot analysis

Immunoblot analysis was performed as described previously [14]. Briefly, the hearts of neonates were homogenized and sonicated at 4 °C in a lysis buffer composed of 10% glycerol, 135 mM NaCl, 5 mM EDTA, and 20 mM Hepes, pH 7.4, containing protease inhibitor mixture (Sigma). The lysates were applied to SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Millipore). Protein transfer was immediately confirmed by fast green staining (loading). The membrane was probed with the antibody, followed by development using ECL-plus (GE Healthcare) for visualization of the antibodies.

Immunofluorescence analysis of cultured cardiomyocytes

Immunofluorescence staining was performed according to the previously described method [10] with minor modifications. Cultured cardiomyocytes were fixed in 3.7% formaldehyde for 15 min, permeabilized with 0.1% Triton X-100 in phosphate-buffered saline for 4 min, and blocked with phosphate-buffered saline (137 mM NaCl, 2.68 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄, pH 7.4) containing 3% bovine serum albumin for 60 min. Cells were labeled overnight at 4 °C with primary antibodies diluted in a dilution buffer (PBS containing 3% bovine serum albumin, 2% goat serum, and 0.1% Triton X-100) and then labeled for 2 h at room temperature with a fluorescein-conjugated secondary antibody mixture in the same buffer. For Fhod3 immunofluorescence staining, anti-Fhod3-(650–802) antibody was used. Nuclei were stained with Hoechst 33342 (Invitrogen). Images were taken with a confocal laser scanning microscope (TCS SP-8; Leica or FV4000; Evident).

Quantification of sarcomere organization was conducted based on the proportion of cell area occupied by well-organized sarcomeres with a width of more than 2 μm of the Z-line visualized by α-actinin staining. Cardiomyocytes were classified into the following four types: excellent, more than 80%; high, 60 ~ 80%; low, 30–60%; poor, less than 30%. In the calculation of sarcomere-positive cells, both excellent and high types were counted as positive cells. The positivity of Fhod3 was determined based on the presence or absence of the Fhod3-specific clear C-zone pattern (doublet between Z-lines) [44] in Fhod3 immunostaining.

Long range time-lapse imaging

Time-lapse imaging was performed using an incubation monitoring system (CM30; Evident) housed within a CO₂ incubator. Images were acquired every 3 hours for 2 to 3 weeks at 37 °C and 5% CO₂ by epi-oblique illumination at 630 nm LED light and captured using CMOS camera.

Statistical analysis

All data presented in the graph were expressed as mean ± SEM (Standard Error of the Mean). Two groups were compared by unpaired Student’s t-test. Multiple groups were compared by analysis of variance followed by post-hoc Tukey-Kramer test. A p value of <0.05 was considered to be statistically significant. JMP 18.2 or JMP Student Editon (JMP Statistical Discovery) was used for all statistical analysis.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Ryu Takeya conceived and designed the overall study. Shuhei Sakaguchi and Yohko Kage designed and performed the experiments and analyses. Eka Adip Pradipta performed experiments and discussed the results. Ayako Miura and Koji Furukawa coordinated the study. Shuhei Sakaguchi and Ryu Takeya wrote and edited the manuscript. All authors read and approved the final manuscript.

Funding

Open Access funding provided by University of Miyazaki. This work was supported in part by JSPS KAKENHI (JP22K19407, JP22H04922(AdAMS), and JP24K10079 to R.T., JP22K11754 to Y.K., and JP21K08183 to A.M.); a bounty for pediatric cardiovascular research from Miyata Cardiac Research Promotion Foundation (to R.T.); a grant from Kawano Masanori Memorial Public Interest Incorporated Foundation for Promotion of Pediatrics (Grant Number 35–17 to R.T.); a grant from Suzuken Memorial Foundation (to A.M.); a grant from Takeda Science Foundation (to A.M.); a grant from The Naito Foundation (to A.M.); a Grant for Clinical Research from University of Miyazaki Hospital (to S.S.); the President's Strategic Priority Budget of the University of Miyazaki (to Y.K, A.M., and R.T.).

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.

Declarations

Ethics approval

All the experimental protocol was approved by the Animal Care and Use Committee of Miyazaki University (Permit Number: 2020–507-7). All mice were housed and maintained in a specific pathogen-free animal facility at University of Miyazaki, and all efforts were made to minimize the number of animals used and their suffering. All experiments were performed in strict accordance with the guidelines for Proper Conduct of Animal Experiments (Science Council of Japan) and the “Guide for the Care and Use of Laboratory Animals” published by the US National Institute of Health.

Competing interest

The authors declare that they have no conflicts of interest with the contents of this article.