Research ArticleCELL BIOLOGY

The Huntingtin-interacting protein SETD2/HYPB is an actin lysine methyltransferase

See allHide authors and affiliations

Science Advances  02 Oct 2020:
Vol. 6, no. 40, eabb7854
DOI: 10.1126/sciadv.abb7854
  • Fig. 1 SETD2 binds actin.

    (A) Immunoblot analysis showing localization of SETD2 in whole-cell extracts (WCE) and both nuclear and cytoplasmic (Cyto) compartments of 786-0 and HEK293T cells. Lamin A/C and lactate dehydrogenase (LDH) are used as controls for the nuclear and cytoplasmic fractions, respectively. (B) Immunoblot analysis showing coimmunoprecipitation of endogenous SETD2 and endogenous actin in 786-0 cells using SETD2 antibodies from two different sources. (C) Immunoblot analysis showing that actin is methylated in SETD2-proficient but not in SETD2-deficient 786-0 cells by immunoprecipitation (IP) of endogenous actin using two different antibodies directed against the SETD2 trimethyl-lysine epitope. Data in (A to C) are representative of experiments repeated at least three times with similar results. (D) Deconvolution microscopy imaging of filamentous (F)–actin using phalloidin (green) and SETD2 methyl epitope–specific antibody (magenta) showing colocalization of the methyl mark with F-actin in 786-0 cells. Scale bars, 10 μm. (E) Representative intensity profiles of the staining observed with the methyl-specific antibody and phalloidin. Merged images from (D) are shown on the left to indicate the position for line profiles. (F) Quantification of intensity profiles seen in (E). y axis represents the ratio of methyl antibody to phalloidin intensity. Each small circle represents a single cell. Large circles represent means from 20 cells for each independent biological replicate; data are means ± SEM from these values (n = 3).

  • Fig. 2 SETD2 methylates actin.

    (A) Autoradiography showing in vitro methylation of actin using tritiated S-adenosylmethionine (3H-SAM) as methyl group donor and recombinant glutathione-S-transferase (GST)–tagged SETD2 catalytic SET domain (amino acids 1418 to 1714). Film shows automethylation of SETD2 and histone methylation as positive control. Data are representative of experiments repeated at least three times, with similar results. (B) Fluorescence-based in vitro methylation using recombinant tSETD2 (amino acids 1418 to 2564) with purified cardiac muscle (red), smooth muscle (green), skeletal muscle (purple) actins, and recombinant actin (rActin) from HEK293T cells (blue) or E. coli (orange). Data are means ± SEM (n = 2). (C) Fluorescence-based in vitro methylation of actin using recombinant tSETD2 (amino acids 1418 to 2564) following addition of Latrunculin A (LatA) at indicated concentrations. Methyltransferase activity calculated by slope of linear regression from fluorescence trace (a.u./min), against LatA concentration on x axis. Data are means ± SEM (n = 3). y axis in (B) and (C) plotted after subtracting automethylation from samples with SETD2 alone. (D and E) Immunoblot analysis showing recognition of actin proteins by SETD2 methyl epitope antibodies anti-Me3K40 (D) and anti-Me3K36 (E) following in vitro methylation with recombinant GST-tagged SETD2 (amino acids 1418 to 1714). Data are representative of experiments repeated at least three times with similar results.

  • Fig. 3 SETD2 methylates lysine-68 on actin.

    (A) Amino acid sequence showing conserved KxP SETD2 recognition motif present in all actin isoforms. Position of the lysine residue varies depending on actin isoform; reference to this site as “ActK68” is based on its position in β-actin (ACTB). Histone H3 sequence containing the KxP motif is shown below for reference. (B) Representative tandem mass spectrometry (MS/MS) spectrum of trimethylated ActK68 peptides recovered from SETD2-proficient 786-0 cells. m/z, mass/charge ratio. (C) Fluorescence-based in vitro methylation assay showing in vitro methylation of biotin-labeled K68-containing actin peptides (amino acids 62 to 78) with recombinant tSETD2 (amino acids 1418 to 2564). Sequence for the peptides used is shown in (A). Data are means ± SEM (n = 4). (D) Immunoblot analysis showing dependency of the ActK68me3 mark on SETD2 by IP of endogenous actin from whole-cell extracts of SETD2-proficient and SETD2-deficient 786-0 cells using the anti-Me3K68 antibody. Data are representative of experiments repeated at least three times with similar results.

  • Fig. 4 SETD2 regulates actin polymerization in cells.

    (A) Immunoblot analysis showing decreased F-actin in SETD2-deficient 786-0, HEK293T, and MEF cells. Whole-cell lysate shows absence of SETD2, associated with the expected loss of histone H3K36me3 methylation. (B) Quantitation of F-/G- actin ratio (top) and whole-cell lysate actin (bottom) from the data is shown in (A). Data are means ± SEM (n = 3 for 786-0 and HEK293T; n = 6 for MEF). (C) Immunoblot analysis of F-/G-actin ratio in HEK293T cells expressing wild-type or K68A/R mCherry–β-actin. (D) Immunoblot analysis of whole-cell lysates shows no change in total actin levels or SETD2 histone methylation with expression of K68A/R mCherry–β-actin. (E) Quantitation of F-/G-actin and total actin shown in (C) and (D), respectively. Data are means ± SEM (n = 3).

  • Fig. 5 ActK68me3 localizes to the insoluble F-actin fraction in cells.

    Immunoblot (IB) analysis (A) and quantitation (B) showing changes in actin polymerization (F-actin in the insoluble fraction) following washout after treatment with the actin depolymerizing agent latrunculin A (LatA). Data are means ± SEM (n = 4). (C) IB analysis using anti-Me3K68 antibody showing that ActK68me3 occurs on endogenous actin from the insoluble F-actin fraction of SETD2-proficient 786-0 cells; tubulin (which is also methylated by SETD2) is not found in this fraction. Data are representative of experiments repeated at least three times with similar results. Absence of H3K36me3 in the insoluble fraction is used as a control to confirm loss of SETD2 in (A) and (C). (D) IP of actin using anti-Me3K68 antibody from soluble (G-actin) and insoluble (F-actin) fractions of SETD2-proficient cells treated with LatA for indicated times. Light and dark exposures are shown for contrast in amount of methylated actin between fractions. Data are representative of three experiments performed using anti-Me3K68 and anti-Me3Pan antibodies with similar results. In an additional experiment performed shortly after cells were brought up from cryopreservation, we saw a less marked loss of methylated insoluble actin with LatA, but saw no increase in methylated soluble actin, consistent with data shown here.

  • Fig. 6 SETD2 regulates cell migration.

    (A) In vitro scratch assay at 0 and 24 hours after wound inflection, illustrating that SETD2-deficient cells migrate slower than SETD2-proficient 786-0 cells. Scale bars, 1000 μm. (B) Quantification of scratch assays seen in (A). Small circles represent each independent measurement across all experiments. Large circles represent mean from a minimum of nine individual measurements for each independent biological replicate (n = 4).

  • Fig. 7 SETD2 methylates actin in association with HTT.

    (A) Schematic diagram showing interaction between the SETD2 C-terminal and HTT N-terminal proline-rich region (PRR), as reported in (2, 24). (B) Immunoblot (IB) analysis showing coimmunoprecipitation of mCherry-tagged SETD2 C-terminal [SETD2-(C)-mCherry] and endogenous HTT in HEK293T cells. (C) IB analysis showing that actin methylation by SETD2 is dependent on both HTT and HIP1R using a SETD2 methyl-epitope antibody (anti-Me3K36) to immunoprecipitate actin from 786-0 cells after siRNA-mediated knockdown of HTT or HIP1R. Input lysates shown to confirm knockdown of HTT and HIP1R. (D and E) IB analysis (D) and quantitation (E) of decreased insoluble F-actin in 786-0 cells after siRNA-mediated knockdown of HTT or HIP1R. Data are means ± SEM (n = 3). (F) Quantitation of migration data after siRNA-mediated knockdown of HTT or HIP1R in SETD2-proficient versus SETD2-deficient 786-0 cells. Small circles each represent an independent measurement from all biological replicates. Large circles represent mean from 12 measurements for each independent biological replicate (n = 4). (G) IB analysis showing decreased actin methylation after expression of a mutant HTT protein by IP of actin using a SETD2 methyl-epitope antibody (anti-Me3K36) from HEK293T cells expressing a doxycycline (Doxy)–inducible CFP-tagged N-terminal HTT construct containing 94 polyglutamine repeats (CFP-HTT94Q) and a retrotransactivator (rtTA3). Input lysate confirms expression of CFP-tagged mutant protein and loss of SETD2 methylation activity (H3K36me3) without change in endogenous SETD2 expression. Data in (B), (C), and (G) are representative of experiments repeated three times with similar results.

Supplementary Materials

  • Supplementary Materials

    The Huntingtin-interacting protein SETD2/HYPB is an actin lysine methyltransferase

    Riyad N. H. Seervai, Rahul K. Jangid, Menuka Karki, Durga Nand Tripathi, Sung Yun Jung, Sarah E. Kearns, Kristen J. Verhey, Michael A. Cianfrocco, Bryan A. Millis, Matthew J. Tyska, Frank M. Mason, W. Kimryn Rathmell, In Young Park, Ruhee Dere, Cheryl Lyn Walker

    Download Supplement

    The PDF file includes:

    • Figs. S1 to S10
    • Legends for movies S1 to S3

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article