Research ArticleNEUROSCIENCE

PRMT8 as a phospholipase regulates Purkinje cell dendritic arborization and motor coordination

See allHide authors and affiliations

Science Advances  04 Dec 2015:
Vol. 1, no. 11, e1500615
DOI: 10.1126/sciadv.1500615
  • Fig. 1 PRMT8 knockout mice display abnormal Purkinje cell dendrites and aberrant behavior.

    (A) Gene targeting strategy for the prmt8 locus: schematic representation of the prmt8+/+ [wild-type (WT)], prmt8flox (floxed), and prmt8−/− (knockout) alleles. Arrows indicate the positions of the P1, P2, and P3 genotyping primers. (B) Allele-specific PCR analysis using tail genomic DNA. Products of 504 and 252 base pairs (bp) were generated from prmt8+/+ and prmt8−/− mice, respectively. (C) prmt8 mRNA levels in 12-week-old mice were examined in total cerebellar RNA by qPCR. Results are shown as fold change of prmt8 mRNA expression relative to the mean WT value. Data were obtained using the ΔΔCt method with normalization to the reference glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA (mean ± SEM, n = 4 per genotype; **P < 0.05, ***P < 0.0001). (D) Top: Representative confocal images of cerebellar sections from 12- to 14-week-old animals (genotyped as indicated) stained for immunofluorescence detection of calbindin (red) to indicate Purkinje cells and Hoechst 33258 (blue) to label DNA (scale bars, 20 μm). Bottom: Enlarged images of the boxed areas with the Purkinje cell layer (scale bars, 20 μm). (E) Limb-clasping reflex in 10- to 14-week-old mice suspended by the tail as monitored by video recording. The percentage indicates the limb-clasping reflex of mice (genotyped as indicated). (F) Locomotor activity of prmt8−/− mice. The total number of squares crossed by the insect during 30 s is indicated. Data are shown as means ± SEM; ***P < 0.0001 or NS (not significant) as compared to WT animals.

  • Fig. 2 Acetylcholine and choline are decreased, whereas PC levels are increased, in the cerebellum of prmt8−/− mice.

    (A to F) Specific detection of acetylcholine, choline, and PC by MALDI-QIT-TOF/MS. Positive-ion mode mass spectrum of acetylcholine (m/z 146 [M]+) with α-cyano-4-hydroxycinnamic acid (CHCA), choline (m/z 104 [M]+) with 9-aminoacridine hemihydrate (9-AA), or DPPC (m/z 757 [M + Na]+) with 9-AA as matrix. The m/z ratios among the peaks are indicated. Among the three mouse genotypes (WT, heterozygous, and homozygous; n = 4 per genotype) examined, experimental ratios were determined by summing the ion peak areas. Data are shown as means ± SEM; *P < 0.05, **P < 0.01, ***P < 0.0001, or NS as compared to WT mice. (G to I) The mRNA expression of ChAT, PLD1, and PLD2 was analyzed by qPCR. Results are shown as fold change of mRNA expression relative to the mean value of the WT sample. Data were obtained using the ΔΔCt method normalized to the reference GAPDH mRNA (mean ± SEM, n = 4 per genotype; data were not significantly different).

  • Fig. 3 Lysine 107–dependent PRMT8 induces neurite branching in NGF-stimulated PC12 cells.

    (A) Alignment of the amino acids in the HKD motifs from human PRMT8 and PLD2 and the mitochondrial PLD isoform MitoPLD. The amino acids that were mutated in the study are marked with a red asterisk. (B) Schematic representation of the K107R mutation generated in human PRMT8. (C) Translocation of GFP-PABD (Spo2051–91), a PA sensor, to the plasma membrane upon PRMT8-mediated PA generation. PC12 cells were cotransfected with GFP-PABD and pmCherry vector (control), WT PRMT8-mCherry, or the PRMT8 K107R mutant. Controls were stimulated with or without NGF (100 ng ml−1) for 5 min (left panel). Scale bar, 50 μm. Neurite development of NGF-differentiated PC12 cells. (D) PC12 cells were coexpressed with the pVenus vector and pmCherry vector (control), WT PRMT8-mCherry, or the PRMT8 K107R mutant, and stimulated with NGF (100 ng ml−1) for 24 hours. Scale bars, 30 μm. (E to H) Cells bearing neurites that were twofold longer than their cell body lengths were scored in terms of (E) total neurite length, (F) the number of branching points per cell, (G) the number of neurite roots, and (H) the number of extremities. Results are shown as mean ± SEM of five independent experiments with at least 50 cells scored in each experiment; *P < 0.05, **P < 0.01, ***P < 0.0001, or NS as compared to the control (CON).

  • Fig. 4 PRMT8 is an HKD motif–dependent phospholipase.

    (A) BiFC experiments indicate dimerization of WT PRMT8 and the PRMT8 K107R mutant at the plasma membrane of HEK293T cells. All cells were cotransfected with BiFC pair vectors and mCherry vector as a transfection control. Scale bars, 20 μm. (B) Purified WT PRMT8 and the catalytically inactive PRMT8 K107R mutant were produced through baculoviral expression followed by Flag tag purification; proteins were visualized by Coomassie blue staining. (C and D) MS analysis of the reaction products generated by incubation of DPPC with either WT PRMT8-Flag or PRMT8 K107R mutant proteins. (C) Positive ion mode; (D) negative ion mode. Upper panels: Standard choline or DPPA with buffers but no enzyme. Middle panels: DPPC substrate with buffers but no enzyme. Bottom panels: DPPC substrate incubated with either WT PRMT8-Flag or PRMT8 K107R mutant proteins, resulting in the conversion of a portion of DPPC to the expected hydrolyzed products. Insets display the MS2 spectra of the following DPPA peaks at m/z 647.6: palmitic acid, m/z 256.0 [M − H]; 16:0 lyso-PA, m/z 391.9 [M − H2O]; 16:0 lyso-PA, m/z 409.8 [M − H]. (E) Proposed working model for PRMT8-mediated PC metabolism.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/1/11/e1500615/DC1

    Materials and Methods

    Fig. S1. Functional analysis of prmt8 using zebrafish.

    Fig. S2. Quantitative analysis of the gait abnormalities in the prmt8−/− mice.

    Fig. S3. Interaction of PRMT8-Flag and mammalian PLDs.

    Fig. S4. Organizations of the catalytic domains in PLD superfamily.

    Fig. S5. Dimerization of PRMT8-Flag wild-type (WT) and K107R and G121A mutants.

    Fig. S6. Measuring vertebrate PRMT8 paralog-derived products from PC by MALDI-QIT-TOF/MS.

    Fig. S7. Measuring PRMT8-derived [3H]choline from [3H]DPPC using TLC method.

    Fig. S8. Methyltransferase and PC phospholipase activities of PRMT8.

    Fig. S9. Loss of catalytic activities of PRMT8 by replacing serine 120 to alanine.

    Table S1. Primers used for qPCR quantification of mRNA.

    Movie S1. Behavior of mice in a novel environment.

    References (4750)

  • Supplementary Materials

    This PDF file includes:

    • Materials and Methods
    • Fig. S1. Functional analysis of prmt8 using zebrafish.
    • Fig. S2. Quantitative analysis of the gait abnormalities in the prmt8−/− mice.
    • Fig. S3. Interaction of PRMT8-Flag and mammalian PLDs.
    • Fig. S4. Organizations of the catalytic domains in PLD superfamily.
    • Fig. S5. Dimerization of PRMT8-Flag wild-type (WT) and K107R and G121A mutants.
    • Fig. S6. Measuring vertebrate PRMT8 paralog-derived products from PC by MALDI-QIT-TOF/MS.
    • Fig. S7. Measuring PRMT8-derived 3Hcholine from 3HDPPC using TLC method.
    • Fig. S8. Methyltransferase and PC phospholipase activities of PRMT8.
    • Fig. S9. Loss of catalytic activities of PRMT8 by replacing serine 120 to alanine.
    • Table S1. Primers used for qPCR quantification of mRNA.
    • Legend for movie S1
    • References (47–50)

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). Behavior of mice in a novel environment.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article