Regulation of dynein-driven microtubule sliding by the axonemal protein kinase CK1 in Chlamydomonas flagella

Abstract

Experimental analysis of isolated ciliary/flagellar axonemes has implicated the protein kinase casein kinase I (CK1) in regulation of dynein. To test this hypothesis, we developed a novel in vitro reconstitution approach using purified recombinant Chlamydomonas reinhardtii CK1, together with CK1-depleted axonemes from the paralyzed flagellar mutant pf17, which is defective in radial spokes and impaired in dynein-driven microtubule sliding. The CK1 inhibitors (DRB and CK1-7) and solubilization of CK1 restored microtubule sliding in pf17 axonemes, which is consistent with an inhibitory role for CK1. The phosphatase inhibitor microcystin-LR blocked rescue of microtubule sliding, indicating that the axonemal phosphatases, required for rescue, were retained in the CK1-depleted axonemes. Reconstitution of depleted axonemes with purified, recombinant CK1 restored inhibition of microtubule sliding in a DRB- and CK1-7-sensitive manner. In contrast, a purified "kinase-dead" CK1 failed to restore inhibition. These results firmly establish that an axonemal CK1 regulates dynein activity and flagellar motility.

Figure 1.

Model for regulation of I1 dynein and the CK1 protein. (A) Analysis of wild-type and mutant axonemes has revealed that microtubule sliding activity is regulated by phosphorylation of the I1 dynein subunit IC138 (Wirschell et al., 2007). The data predicts that IC138 is phosphorylated by the axonemal kinase CK1, and that phosphorylation inhibits dynein-driven microtubule sliding activity. The model also indicates that axonemal phosphatase PP2A is required to rescue microtubule sliding activity (Yang and Sale, 2000). (B) C. reinhardtii CK1 is highly conserved and contains characteristic CK1 domains including the N-terminal ATP and substrate-binding domains, the kinesin homology domain (KHD), the catalytic triad, and the nuclear localization signal (NLS). To generate rCK1-KD, K 40, shown to be required for kinase activity (Gao et al., 2002), was replaced by R. A CK1-specific antibody was made to the polypeptide at the C terminus.

Figure 2.

CK1 is an axonemal protein located along the length of the axoneme and is extractable in 0.3 M NaCl buffers. (A) Flagella (Fla) were fractionated into a Nonidet P40 soluble membrane/matrix fraction (NP-40 Ext) and axonemal fraction (Axo), then analyzed by an immunoblot probed with the anti-CK1 antibody (top left). Axonemal CK1 was extracted with 0.3 M NaCl (top right). The bottom panels show Coomassie-stained proteins of the same samples and those used as a loading control. Positions of molecular mass standards are indicated in kD. (B) CK1 localizes along the length of the axoneme (top, “Intact Axonemes”). Biochemical depletion of CK1 removes all detectable CK1 (bottom, “Extracted Axonemes”). Bar, 5 µm.

Figure 3.

Biochemical depletion of CK1 rescues microtubule sliding in isolated pf17 axonemes, and this rescue requires I1 dynein. (A) Experimental strategy to test the role of CK1 in microtubules. (B) ATP-induced microtubule sliding velocity was measured in isolated axonemes, CK1-depleted axonemes, or CK1-depleted axonemes reconstituted with purified rCK1 (Okagaki and Kamiya, 1986; Wirschell et al., 2009). The effect of DRB/CK1-7 and the phosphatase inhibitor microcystin-LR (MC) was also examined. The bars represent: (1) wild-type (WT) axonemes; (2) WT axonemes depleted of CK1 (note that there is no change in velocity); (3) pf17 axonemes (note the slow, baseline sliding velocity); (4) pf17 axonemes plus DRB; (5) pf17 axonemes plus CK1-7; (6) pf17 axonemes plus DRB and microcystin-LR; (7) pf17 axonemes plus CK1-7 and microcystin-LR; (8) pf17 axonemes depleted of CK1 (note the rescue of microtubule sliding); (9) pf17 axonemes depleted of CK1 plus microcystin-LR; (10) ida1pf17 axonemes; (11) ida1pf17 axonemes plus DRB; (12) ida1pf17 axonemes plus CK1-7; and (13) ida1pf17 axonemes depleted of CK1 (note the failure in rescue of sliding). Microtubule sliding velocity is expressed as µm/s, and means and standard deviations (error bars) were calculated from at least three independent experiments with a minimum sample size of 75 axonemes.

Figure 4.

CK1 kinase activity is required for regulation of microtubule sliding. (A) Addition of rCK1 restores inhibition in CK1-depleted pf17 axonemes (bars 3 and 4) and rebinds to CK1-depleted pf17 axonemes (top). DRB and CK1-7 block restoration of inhibition (bars 5 and 6), and microcystin-LR blocks the effects of the kinase inhibitors (bars 7 and 8). (B) rCK1-KD rebinds to CK1-depleted pf17 axonemes (top) but fails to restore inhibition of microtubule sliding velocity (compare bars 2 and 3). Treatment with CK1 inhibitors had no effect (bars 4 and 5). P, pellet; S, supernatant. (C) rCK1 and rCK1-KD were reconstituted at different ratios to CK1-depleted pf17 axonemes (illustrated below bars 3–7). With increasing amounts of rCK1-KD, there is a corresponding increase in microtubule sliding velocity, indicating that rCK1-KD competes for binding and inhibits the effect of rCK1 on microtubule sliding. Microtubule sliding velocity is expressed as µm/s, and averages and standard deviations (error bars) were calculated from at least three independent experiments with a minimum sample size of 75 axonemes.