Distinct functional constraints driving conservation of the cofilin N-terminal regulatory tail

Abstract

Cofilin family proteins have essential roles in remodeling the cytoskeleton through filamentous actin depolymerization and severing. The short, unstructured N-terminal region of cofilin is critical for actin binding and harbors the major site of inhibitory phosphorylation. Atypically for a disordered sequence, the N-terminal region is highly conserved, but specific aspects driving this conservation are unclear. Here, we screen a library of 16,000 human cofilin N-terminal sequence variants for their capacity to support growth in S. cerevisiae in the presence or absence of the upstream regulator LIM kinase. Results from the screen and biochemical analysis of individual variants reveal distinct sequence requirements for actin binding and regulation by LIM kinase. LIM kinase recognition only partly explains sequence constraints on phosphoregulation, which are instead driven to a large extent by the capacity for phosphorylation to inactivate cofilin. We find loose sequence requirements for actin binding and phosphoinhibition, but collectively they restrict the N-terminus to sequences found in natural cofilins. Our results illustrate how a phosphorylation site can balance potentially competing sequence requirements for function and regulation.

Fig. 1. A dual-functionality screen of a cofilin-1 N-terminal mutagenesis library.

a Growth of TeTO₇-COF1 yeast exogenously expressing the indicated cofilin in the presence or absence of doxycycline (DOX). Image representative of n = 2. b Yeast harboring the indicated expression plasmids were grown on either glucose or galactose to induce LIMK1 expression in the presence or absence of doxycycline. Image representative of n = 7. c Design of cofilin N-terminal combinatorial mutagenesis library. X indicates any of the 20 natural amino acids. d Schematic of the cofilin library dual-functionality screen. The schematic was made using Biorender. e Graphs show a change in the relative representation of the indicated cofilin variants over time during culture in either glucose (no LIMK1) or galactose (LIMK1 induction). Data are from three replicate screens performed independently. Blue circles indicate timepoints used to identify functional sequences. Orange circles indicate timepoints used to identify sequences inhibited by LIMK1. Time zero is after the derepression of P_GAL-LIMK1 over the course of 1– 2 population doublings in raffinose + DOX media.

Fig. 2. Cofilin library screen results for yeast growth rescue.

a Waterfall plot depicting the enrichment or depletion of all 16,000 cofilin library sequences expressed in TeTO₇-COF1 yeast grown in doxycycline. Sequences are ordered left to right from highest to lowest average enrichment from three independent screens. b Probability logo of the n = 4099 cofilin N-terminal library sequences enriched in the screen (average log₂ fold change from three independent experiments >0). The red line indicates the height threshold for the frequency of a residue in the enriched sequences being significantly different (p = 0.05) from the background frequency as calculated by a binomial probability function with Bonferroni correction. c Heat map of cofilin N-terminal library sequence enrichment or depletion. Color scale indicates the mean fold change of the 800 (positions 2, 4, and 5) or 8000 (position 3) library sequences containing the indicated residue across 3 independent replicates. d Distribution of cofilin sequences by enrichment scores according to the identity of residues at the indicated positions. Each distribution includes n = 800 sequences, with each value being the average from three separate experiments. Box plots indicate the median (middle line), 25th and 75th percentile (box), and 10th and 90th percentile (whiskers). Source data for all graphs is found in Supplementary Data 2.

Fig. 3. Impact of cofilin N-terminal mutations on yeast growth and actin binding.

a, b Growth assay comparing cofilin mutants containing substitutions of residues Ala2, Gly4, or Val5. Images representative of n = 3 independent experiments. DOX, doxycycline. c–e Pyrene quenching assays comparing cofilin^WT to the indicated cofilin mutants. Actin concentration pre-polymerization was 1 µM. Independent biological replicates for binding assays are as follows: cofilin^WT (n = 4), cofilin^A2L (n = 3), cofilin^A2D (n = 4), cofilin^S3D (n = 3), phospho-cofilin^WT (n = 2), cofilin^G4A (n = 3), cofilin^G4F (n = 2), cofilin^G4E (n = 2), cofilin^G4K (n = 3), cofilin^V5M (n = 3), and cofilin^A2L,V5M (n = 2). Source data are provided as a Source Data file.

Fig. 4. Cofilin library screen results for yeast growth rescue with and without LIMK1.

a Scatter plot depicting the change in representation of Ser3 cofilin library sequences with and without LIMK1 expression. Data points show average enrichment in glucose and galactose across three independently performed replicates. b Logo of sequences supporting growth in yeast but are selectively inhibited by LIMK1 (lower right quadrant in panel a) on the background of all sequences that support yeast growth in the absence of LIMK1 (upper and lower right quadrants in panel a). c Distribution of cofilin sequences by enrichment scores with and without LIMK1 expression according to the identity of the phosphoacceptor residue. Each distribution includes n = 8000 sequences, and values are the average from three independent screens. Box plots indicate the median (middle line), 25th and 75th percentile (box), and 10th and 90th percentile (whiskers). Source data is found in Supplementary Data 2.

Fig. 5. Effects of cofilin Ser3 substitutions on phosphorylation by LIMK1.

a TeTO₇-COF1 yeast expressing the indicated human cofilin-1 mutants were grown in the presence of doxycycline to repress endogenous Cof1 expression and/or galactose to induce LIMK1 expression. Images are representative of n = 3 independent experiments. DOX, doxycycline. b Immunoblots with the indicated antibodies of lysates corresponding to yeast plated in (a). Samples in the cofilin blot were separated on Phos-tag SDS-PAGE to resolve phosphorylated (upper band) from unphosphorylated protein (lower band). Kss1 serves as a loading control. The chart shows the fraction of cofilin phosphorylated, calculated from quantified intensities of the upper and lower bands. Data are presented as mean values ± SD for n = 3 independent experiments. c In vitro radiolabel kinase assay comparing phosphorylation of purified cofilin (2 µM) by LIMK1^CAT (2 nM) for 10 min at 30 °C. The catalytically inactive LIMK1^D460N kinase domain serves as a negative control. The chart shows the quantification of phosphorylation rates of cofilin mutants normalized to cofilin^WT. Data are presented as mean values ± SD for n = 4 independent experiments. d N-terminal sequences of mammalian and yeast isoforms of cofilin and twinfilin. e In vitro radiolabel kinase assay showing phosphorylation of the indicated forms of cofilin-1 or twinfilin (2 µM) by LIMK1 kinase domain (2 nM) for 10 min at 30 °C. LIMK1^D460N kinase domain serves as a negative control. The graph shows the quantification of phosphorylation rates normalized to cofilin^WT. Data are presented as mean values ± SD for n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 6. Effects of cofilin Glycine 4 substitutions on LIMK1 substrate suitability.

a Yeast growth assay with the TeTO₇-COF1 strain expressing the indicated cofilin variants with or without LIMK1 expression. Representative of n = 3 independent experiments. DOX, doxycycline. b Immunoblots with the indicated antibodies of lysates corresponding to yeast plated in panel A separated by standard or Phos-tag PAGE. EV, empty vector. c Quantification of immunoblots measuring the fraction of total cofilin phosphorylated by LIMK1. Error bars show SD for n = 3 independent experiments. d In vitro radiolabel kinase assay comparing phosphorylation of purified 2 µM cofilin by 2 nM LIMK1 kinase domain for 10 min at 30 °C. The catalytically inactive LIMK1^D460N kinase domain serves as a negative control. e Quantification of phosphorylation rates of cofilin mutants normalized to cofilin^WT. Error bars show SD for n = 6 independent experiments. f In vitro radiolabel kinase assay comparing phosphorylation of cofilin^WT and cofilin^G4F by 2 nM LIMK1 kinase domain across a range of concentrations. Reactions performed for 10 min at 30 °C. g Quantification of phosphorylation rates of cofilin mutants. Error bars show SD for n = 3 independent experiments. Kinetic parameters include SE for k_cat and K_m. Source data are provided as a Source Data file.

Fig. 7. Cofilin Gly4 mutations counter the inhibitory effects of phosphomimetic S3D substitution.

a Yeast growth assays with TeTO₇-COF1 strain expressing the indicated cofilin variants in the presence or absence of doxycycline (DOX). Immunoblots of cofilin protein levels are shown below, with Kss1 as a loading control. Images are representative of n = 3 independent experiments. b Pyrene quenching assays comparing cofilin^WT to cofilin variants with a phosphomimetic substitution or phosphorylation modification at residue Ser3. Actin concentration pre-polymerization was 1 µM. WT, S3D, and G4F single mutant data are the same as shown in Fig. 3 and are included here for comparison. Binding parameters are listed in Table 1. Independent biological replicates for binding assays are as follows: cofilin^WT (n = 4), cofilin^S3D (n = 3), phospho-cofilin^WT (n = 2), cofilin^G4F (n = 2), cofilin^S3D,G4F (n = 3), cofilin^G4L (n = 6), and cofilin^S3D,G4L (n = 3). c Overlapping constraints promoting conservation of the cofilin phosphosite motif sequence. Source data are provided as a Source Data file.