Three sensors that recognize the Y-αCTT.

(A) Schematic of tubulin states within the microtubule lattice. Tubulin dimers add to the end of a microtubule in a GTP-bound and expanded state, resulting in a stabilizing GTP cap. In the microtubule lattice, β-tubulin undergoes GTP hydrolysis, resulting in a GDP lattice and compaction of α-tubulin. (B) Schematic of three sensors (YL1/2, A1aY1, and 4xCAP-Gly) generated to detect the accessibility of the Y-αCTT along the microtubule lattice. (C) Validation of the specificity of the YL1/2 antibody-based probes for the Y-αCTT. GST-tagged αCTT sequences were probed by western blotting with the commercial YL1/2 monoclonal antibody, recombinant YL1/2 antibody fused to GFP (rMAb-YL1/2-GFP), or a Fab fragment of rMAb-YL1/2-GFP (YL1/2Fab-GFP). Each blot was also probed for GST protein as a loading control. Y: full-length and tyrosinated αCTT sequence; ΔY: detyrosinated αCTT sequence (lacking the C-terminal tyrosine); ΔC2: αCTT sequence lacking the C-terminal two amino acids. (D,E) Representative images of (D) YL1/2Fab-GFP or (E) 4xCAPGly-GFP proteins binding to tyrosinated (Y-MT) or detyrosinated (ΔY-MT) microtubules. HeLa tubulin was polymerized into microtubules and Taxol-stabilized. The microtubules were used directly (Y-MTs) or detyosinated by VASH/SVBP-containing lysate. Scale bar: 5 µm.

Y-αCTT probes bind to the microtubule lattice after Taxol treatment.

(A-D) Live-cell imaging of A1aY1 probe. (A) Representative images of the stable sTagRFP-A1aY1 HeLa cell line before and 15 minutes after treatment with (top) 0.3% DMSO control or (bottom) 10 µM Taxol. Insets (boxes) show magnified views of A1aY1 probe. Scale bars, 10 µm. (B,C) Quantification of probe binding to microtubules. Paired data plots display the localization of A1aY1 probe in individual cells before and after treatment with (B) DMSO or (C) Taxol. (D) Mean difference plot showing the fold change in A1aY1 probe binding in DMSO-vs Taxol-treated cells. DMSO=29 cells across 3 experiments; Taxol=59 cells across 7 experiments. (E-H) Live-cell imaging of 4xCAPGly probe. (E) Representative images of the stable 4xCAPGly-mEGFP HeLa cell line before and 15 minutes after treatment with (top) 0.3% DMSO control or (bottom) 10 µM Taxol. Insets (boxes) show magnified views of CAPGly probe. Scale bars, 10 µm. (F,G) Quantification of probe binding to microtubules. Paired data plots display the localization of 4xCAPGly probe in individual cells before and after treatment with (F) DMSO or (G) Taxol. (H) Mean difference plot showing the fold change in 4xCAPGly probe binding in DMSO- or Taxol-treated cells. DMSO=28 cells across 3 experiments; Taxol=17 cells across 5 experiments. Error bars in (D) and (H) indicate SD. ***: p<0.0002; ****: P<0.0001; ns: not significant [Student’s t test (B,C,F,G: two-tailed; paired), (D,G: unpaired).

MAPs that expand the microtubule lattice increase Y-αCTT probe binding.

(A,B) Representative live-cell images of stable (A) A1aY1) or (B) 4xCAPGly HeLa cell lines transiently expressing the indicated GFP-tagged tubulin or MAP constructs. Cell boundaries are indicated by blue dotted lines. Scale bars, 20 µm. (C,D) Quantification of (C) A1aY1 or (D) CAPGly probe colocalization with GFP-tagged tubulin or MAP constructs. The threshold overlap score (TOS) was measured on a per-cell basis where 1.0 indicates perfect colocalization, −1.0 indicates perfect anti-colocalization, and values near 0 indicate no relationship. Data from three independent experiments are presented as Tukey box plots. The box encompasses the 25th to 75th percentiles, with a line at the median. Whiskers show the last data point within 1.5 times the interquartile range. Outliers are plotted as individual points. *: p < 0.1; **: p < 0.001; ****: p < 0.0001; ns: not significant (Kruskal-Wallis test followed by post-hoc Dunn’s multiple pairwise comparisons with TubA1A as the control). Number of cells analyzed (n) in (C): TubA1A = 69, Tau = 57, MAP2 = 66, Kif5Crigor = 70, CAMSAP2 = 50, CAMSAP3 = 56, and MAP7 = 55 and in (D): TubA1A = 62, Tau = 61, MAP2 = 61, Kif5C rigor = 76, CAMSAP2 = 62, CAMSAP3 = 71, and MAP7 = 57. (E) Schematic model depicting how expander and compactor MAPs regulate microtubule lattice conformation, influencing Y-αCTT accessibility.

MAPs that expand the MT lattice increase detyrosination of the Y-αCTT.

(A) Representative images of HeLa cells transiently expressing the indicated GFP-tagged MAPs and then fixed and stained with antibodies against detyrosinated microtubules (ΔY-tubulin) and total microtubules (MTs). Images are shown in inverted grayscale. The nuclei are represented by blue pseudocolor in the bottom panels. Blue dotted lines: boundaries of cells expressing the corresponding MAPs. Scale bar, 20 µm. (B) Quantification of the colocalization of MAPs and detyrosinated microtubules. The threshold overlap score (TOS) was measured on a per-cell basis. Data from three independent experiments are presented as Tukey box plots. The box encompasses the 25th to 75th percentiles, with a line at the median. Whiskers show the last data point within 1.5 times the interquartile range. Outliers are plotted as individual points. ****: p < 0.0001; ns: not significant (Kruskal-Wallis test followed by post-hoc Dunn’s multiple pairwise comparisons with tau as the control). Number of cells analyzed (n): Tau = 70, MAP2 = 72, Kif5Crigor = 76, CAMSAP2 = 66, CAMSAP3 = 69, and MAP7 = 67. (C) Quantification of the intensity of detyrosination on MAP-bound microtubules. The fluorescence intensity of detyrosination was measured on MAP-decorated microtubules and normalized against the total microtubule intensity of MAP-decorated microtubules. Data from three independent experiments are presented as Tukey box plots. ****: p < 0.0001; ns: not significant (Kruskal-Wallis test followed by post-hoc Dunn’s multiple pairwise comparisons with tau as the control). Number of cells analyzed (n): Tau = 67, MAP2 = 70, Kif5Crigor = 75, CAMSAP2 = 67, CAMSAP3 = 70, and MAP7 = 67.

GTP-like tubulin state increases Y-αCTT accessibility and detyrosination.

(A-D) Live-cell imaging of Y-αCTT probes. (A,C) Representative images of stable (A) A1aY1 or (C) 4xCAPGly HeLa cell lines expressing PA-tagged WT or E254A α-tubulin with an IRES-driven GFP protein as a reporter of transfected cells. Cell boundaries are indicated by blue dotted lines. Scale bars, 20 µm. (B,D) Quantification of (B) A1aY1 or (D) 4xCAPGly probe binding to microtubules. The density was measured as the ratio of the skeletonized probe-decorated microtubule length to the total cell area. Data from three independent experiments are presented as Tukey box plots. The box encompasses the 25th to 75th percentiles, with a line at the median. Whiskers show the last data point within 1.5 times the interquartile range. Outliers are plotted as individual points. **: p < 0.01 (Mann-Whitney U test). Number of cells analyzed (n) in (B): WT = 33, E254A = 29 and in (D): WT = 42, E254A = 50. (E,F) Detyrosinated microtubules. (E) Representative images of HeLa cells transiently expressing PA-tagged WT or E254A α-tubulin and then fixed and stained with antibodies against the PA tag, detyrosinated microtubules (ΔY-tubulin), and total microtubules (MTs). Images are shown in inverted grayscale. The nuclei are represented by blue pseudocolor in the bottom panels. Scale bar, 20 µm. (F) Quantification of the intensity of detyrosination in cells expressing PA-tagged WT or E254A α-tubulin. The fluorescence intensity of detyrosination was measured on a per-cell basis and normalized against the total microtubule intensity. Data from three independent experiments are presented as Tukey box plots. ****: p < 0.0001; ns: not significant (Kruskal-Wallis test followed by post-hoc Dunn’s multiple pairwise comparisons with the untransfected sample (untr.) as the control). Number of cells analyzed (n): untransfected = 105, WT = 84, E254A = 94.

MAPs but not nucleotide state promote Y-αCTT exposure in vitro.

(A-D) Nucleotide state does not determine probe binding. (A,C) Representative images of (A) 4xCAPGly-mEGFP or (C) YL1/2Fab-GFP binding to a mixture containing both AlexaFluor-568 labeled GMPCPP-stabilized microtubules and AlexaFluor-647 labeled GDP microtubules. (B,D) Quantification of the fluorescence intensity of (B) 4xCAPGly-mEGFP or (D) YL1/2Fab-GFP probe binding per length of microtubule. Data are presented as scatter plots with data from three independent experiments in different shades of gray. ns: not significant (two-tailed, Student’s t test). (E-G) Stepping KIF5C can increase YL1/2Fab probe binding. (E) Flowchart of the in vitro reconstitution assay examining the effect of KIF5C in different nucleotide states on YL1/2Fab binding. (F) Representative images of YL1/2Fab-GFP probe binding to GDP-MTs in the absence or presence of KIF5C in different nucleotide states. (G) Quantification of the mean fluorescence intensity of YL1/2Fab-GFP probe along GDP-MTs under the conditions shown in (F). Each spot indicates the probe binding on an individual microtubule. Number of microtubules (n) = 74-105 from three independent experiments. n.s., not significant, ****p<0.0001 (two-tailed, t-test).

The nucleotide state alters Y-αCTT interactions with the microtubule body.

(A) Representative image from MD simulations identifying four distinct sites where the Y-αCTT interacts with the body of tubulin subunits in the microtubule: sites 1 (green) and 2 (cyan) are on the adjacent β-tubulin along a protofilament (i.e. next tubulin towards the microtubule minus end) whereas sites 3 (purple) and 4 (salmon) are cis-interactions with α-tubulin itself. The tubulin body is shown in cartoon and colored gray. The Y-αCTT is shown in stick and colored yellow with the glutamate sidechains in red. (B) Interaction rate of the glutamate residues in the Y-αCTT with the four tubulin body sites for microtubules in the (gold) GDP or (blue) GTP states. (C) The fraction of Y-αCTTs that are inaccessible as a function of time for microtubules in the (gold) GDP or (blue) GTP state where inaccessibility is defined as one or more salt bridges formed between glutamates E445-E450 and the interaction sites in the microtubule body.

Recombinant YL1/2 antibody and purified probes.

(A) Schematic of a typical mammalian IgG molecule containing two heavy (H) and two light (L) chains. Light chains are comprised of one variable (VL, light orange) and one constant (CL, dark orange) region. Heavy chains are comprised of one variable (VH, light gray) and three constant (CH1– 3, dark gray) regions. Red dots: complementarity determining regions (CDRs). Purple lines: disulfide bonds. rMAb-EGFP: recombinant monoclonal antibody (rMAb) with EGFP fused to the C-terminus of the light chain. Fab-EGFP: Fragment antibody binding (Fab) produced from rMAb-EGFP by papain cleavage. (B) Experimentally-determined YL1/2 protein sequence. The deduced amino acid sequences of YL1/2 IgG heavy and light chains are shown. The red text indicates the CDRs involved in antigen recognition. Asterisks demarcate every 10 aa. (C) Coomassie-stained SDS-PAGE gel of purified proteins.

A1aY1 sensor must be imaged in live cells.

COS-7 cells expressing sTagRFP-A1aY1 and mEGFP were (A) imaged live, (B) fixed and mounted, or (C) fixed and stained for total tubulin (microtubules). Scale bar: 10 μm. Magnified views of the red and purple boxed regions are shown to the right.

4xCAPGly sensor must be imaged in live cells.

COS-7 cells expressing 4xCAPGly-mSc3 and mEGFP were (A) imaged live, (B) fixed and mounted, or (C) fixed and stained for total tubulin (microtubules). Scale bar: 10 μm. Magnified views of the red and purple boxed regions are shown to the right.

Transiently-expressed probes bind to the microtubule lattice after Taxol expansion.

Representative images of (A) 4xCAPGly-mEGFP or (B) sTagRFP-A1aY1 probes imaged live in (top) HeLa or (bottom) COS-7 cells before or after addition of 10 μM Taxol. Scale bars: 10 μm.

Western blot of HeLa cells expressing internal PA-tagged tubulin.

Whole cell lysates were prepared from (A) 4xCAPGly-mSc3 stable HeLa cells, (B) sTagRFP-A1aY1 stable HeLa cells, or (C) HeLa cells. In all cases, the cells were untransfected (untr.) or transfected with plasmids for expressing PA-tagged WT or E254A α-tubulin (TubA1A). Asterisks denote upshifted PA-tagged tubulin bands.

The Y-αCTT primarily contacts four sites along a GDP microtubule lattice.

Jaccard index plot indicating the frequency of two residues simultaneously forming salt bridges with the Y-αCTT based on MD simulations of a GDP microtubule lattice. The scale represents the Jaccard index where an index of 0 indicates that the two residues are never interacting with the Y-αCTT at the same time and an index of 1 indicates that when one of the two residues is interacting with the Y-αCTT, the other is also interacting with the Y-αCTT.

Primers used in cloning.