USP8 and Hsp70 regulate endoreplication by synergistically promoting Fzr deubiquitination and stabilization

Abstract

Endoreplication is characterized by multiple rounds of DNA replication without cell division and determines the growth and final size of endoreplicating cells and tissues in eukaryotes. The cyclic ubiquitination and degradation of several cell cycle regulators are required for endoreplication progression. However, the deubiquitinase that deubiquitinates and stabilizes key factors to modulate endoreplication remains unknown. Here, we found in the endoreplicating Drosophila salivary gland and Bombyx silk gland that the depletion of ubiquitin-specific peptidase 8 (USP8) led to endoreplication arrest and a decrease in gland size. Mechanistically, we showed that USP8 interacted with the Fizzy-related (Fzr) protein, a conserved master regulator of endoreplication, thereby deubiquitinating and stabilizing Fzr to modulate endoreplication. Moreover, the molecular chaperone heat shock protein 70 (Hsp70) mediated proper folding of Fzr and increased the interaction between Fzr and USP8, thereby promoting the deubiquitination and stabilization of Fzr. Together, our study demonstrates that USP8 and Hsp70 regulate endoreplication by synergistically maintaining Fzr stability though deubiquitination.

Fig. 1.. Depletion of the deubiquitinase USP8 in the Drosophila salivary glands and Bombyx PSGs blocks endoreplication progression.

(A) RNAi screening of deubiquitinases in the Drosophila salivary gland revealed that the knockdown of only USP8 decreased gland size. Salivary glands were dissected at 120 hours AEL. Scale bar, 200 μm. (B) Overexpression of dominant-negative USP8^C572A in the salivary gland decreased gland size at 120 hours AEL. Scale bar, 200 μm. (C and D) Salivary gland–specific USP8 knockdown reduced the C value and DNA content in the salivary gland at 120 hours AEL. (E and F) EdU staining analyses of the effects of either USP8 knockdown (E) or USP8^C572A overexpression (F) on DNA replication in the salivary gland at 96 hours AEL. Scale bar, 50 μm. (G) CRISPR-Cas9–mediated USP8 mutation in the Bombyx PSGs resulted in a decrease in PSG size at just wandering and thin cocoon with low silk production. Scale bars, 1 cm. (H and I) PSG-specific USP8 mutation decreased the C value and DNA content in PSG cells at just wandering. (J) EdU staining revealed that PSG-specific USP8 mutation blocked DNA replication in PSG cells at the second day of the fourth larval instar. Scale bar, 50 μm. i, RNAi; AEL, after egg laying; WT, wild type. The data are presented as the mean ± SE (error bars) of three independent biological replicates. For the significance test: **P < 0.01 and ***P < 0.001 versus the control.

Fig. 2.. USP8 stabilizes Fzr through deubiquitination.

(A and B) USP8 overexpression in Bombyx BmE cells and Drosophila S2 cells increased Fzr protein levels. (C and D) USP8 depletion in the Bombyx PSGs and Drosophila salivary glands decreased Fzr protein levels. (E) Fzr protein levels in the PSGs and salivary glands were reduced by CHX treatment, and this reduction was reversed by simultaneous MG132 treatment. (F and G) USP8 depletion in the PSGs and salivary glands increased Fzr ubiquitination. (H to J) Fzr overexpression in the salivary gland reversed the effects of USP8 knockdown on the gland size, the C value, and DNA replication during endoreplication. Salivary glands were dissected at 120 hours AEL for size and C value measurement and were dissected at 96 hours AEL for EdU staining and RT-qPCR. Scale bars, 200 μm (H) and 50 μm (J). OE, overexpression; Bm, Bombyx; Dm, Drosophila; WCL, whole cell lysate; h, hours. The data are presented as the mean ± SE (error bars) of three independent biological replicates. For the significance test: *P < 0.05 and **P < 0.01 versus the control.

Fig. 3.. USP8 interacts with Fzr.

(A and B) Co-IP assays with an anti-Flag antibody revealed that exogenously overexpressed HA-tagged USP8 interacted with Flag-tagged Fzr in both Bombyx BmE cells and Drosophila S2 cells. (C and D) Co-IP assays with an anti-Fzr antibody revealed that endogenous USP8 in cultured cells and glands interacted with endogenous Fzr. (E and F) Co-IP assays with an anti-Flag antibody confirmed that the deletion of the dimerization domain of USP8 (USP8-C), not the UCH domain (USP8-N), blocked the interaction of USP8 with Fzr in cultured cells. (G and H) Truncation experiment followed by co-IP assay with an anti-HA antibody revealed that several WD40 domains of Fzr were essential for the interaction of Fzr with USP8. The number indicated the deletion of corresponding WD40 domain of Fzr. IgG, immunoglobulin G.

Fig. 4.. Hsp70 interacts with Fzr.

(A and B) Co-IP assays with an anti-Flag antibody revealed that exogenously overexpressed V5-tagged Hsp70 interacted with Flag-tagged Fzr in both Bombyx BmE cells and Drosophila S2 cells. (C and D) Co-IP assays with an anti-Fzr antibody revealed that endogenous Hsp70 interacted with endogenous Fzr in cultured cells, the Bombyx PSGs, and Drosophila salivary glands. (E) Co-IP assays with an anti-Fzr antibody confirmed that the treatment with the Hsp70 inhibitor PES attenuated endogenous Hsp70-Fzr interaction in the PSGs and salivary glands. (F) Co-IP assays with an anti-Flag antibody confirmed the deletion of the SBD domain of Hsp70, not NBD domain, blocked the interaction of Hsp70 with Fzr in cultured cells. Hsp70 and Fzr from both Bombyx and Drosophila were separately analyzed in corresponding cells. SBD, substrate-binding domain. NBD, nucleotide-binding domain. (G and H) Truncation experiment followed by co-IP assay with an anti-V5 antibody revealed that several WD40 domains of Fzr were essential for the interaction of Fzr with Hsp70. The number indicated the deletion of corresponding WD40 domain of Fzr.

Fig. 5.. Hsp70 promotes the interaction between Fzr and USP8 by mediating Fzr folding.

(A and B) Exogenously overexpressed Hsp70 promoted the interaction between exogenously overexpressed USP8 and Fzr in BmE cells and S2 cells. (C and D) GST pull-down assay showed that supplementation with recombinant Hsp70 increased the USP8-Fzr interaction in vitro. (E to H) The interaction between endogenous USP8 and Fzr in the cultured glands and cells was attenuated following a treatment with the Hsp70 interacting activity inhibitor PES for 2 hours. (I and J) The treatment with the Hsp70 inhibitors PES or VER155008 decreased the luciferase enzyme activity of luciferase-fused Fzr (Fzr-Luc) in cultured cells, suggesting that Hsp70 promoted proper folding of Fzr. (K and L) Overexpression of intact Hsp70 increased the luciferase enzyme activity of Fzr-Luc in cultured cells, but truncated Hsp70 with the deletion of SBD domain or NBD domain lost this promotion. The data are presented as the mean ± SE (error bars) of three independent biological replicates. For the significance test: *P < 0.05 and **P < 0.01 versus the control.

Fig. 6.. Hsp70 regulates endoreplication by maintaining Fzr stability.

(A) CRISPR-Cas9–mediated Hsp70 mutation in the Bombyx PSG decreased PSG size at just wandering and affected cocoon formation. Scale bar, 1 cm. (B) Hsp70 knockdown in the Drosophila salivary gland reduced gland size at 120 hours AEL. Scale bar, 200 μm. (C and D) Hsp70 depletion blocked DNA replication in the PSGs at the second day of the fourth larval instar and salivary glands at 96 hours AEL. Scale bars, 50 μm. (E) Hsp70 depletion in the PSGs and salivary glands decreased the Fzr protein level in two glands. (F) Exogenously overexpressed Hsp70 in cultured cells increased the protein level of exogenously overexpressed Fzr. Hsp70 and Fzr from both Bombyx and Drosophila were separately analyzed in corresponding cells. (G) Hsp70 depletion elevated Fzr ubiquitination in the PSGs and salivary glands. (H and I) Fzr overexpression in the salivary gland reversed the effects of Hsp70 knockdown on gland size at 120 hours AEL and DNA replication at 96 hours AEL. Scale bars, 200 μm (H) and 50 μm (I).