JCS039933 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Curve fitting of recovery data from FRAP experiments with PRMT1-EGFP. Experimental data were fitted with a one-phase exponential recovery function and the Weibull function. Differences between experimental data and calculated curves (residuals) show that the Weibull function (lower panel, left) provides better fitting than the simple exponential function (lower panel, right).

• Supplemental Figure S2 -

  Fig. S2. PRMT5-GFP forms large cytoplasmic aggregates in vivo. Left panel, cells stably expressing PRMT5-GFP analyzed by confocal microscopy as described in the legend of Fig. 3. Note the granular pattern and the large aggregates (arrows) in the cytoplasm. Scale bar: 20 μm. Right panel, extracts from stably transfected cells were separated on glycerol gradients as described in Fig. 2. and proteins were detected by western blotting using anti-GFP antibodies or a commercially available antibody to PRMT5. Note that the GFP fusion protein and the endogenous protein do not sediment identically.

• Supplemental Figure S3 -

  Fig. S3. Characterization of PRMT2. (A) HEK293 cells stably expressing PRMT2 were investigated by confocal microscopy, showing that PRMT2 is preferentially localized in the nucleus where it is enriched in a speckled pattern. (B) PRMT2 can interact with PRMT8. HEK293 cells stably expressing PRMT2 were transiently transfected with PRMT1-8. Upper panel, expression of transfected PRMTs in total cell extracts; cell extracts were separated on SDS-PAGE gel and analyzed by western blotting with antibodies against GFP. Lower panel, total cell extracts of transfected HEK293 cells were immunoprecipitated with antibodies against PRMT2, washed and separated on SDS-PAGE gel; western blotting analysis was performed with GFP-specific antibodies. (C) PRMT2 has weak enzymatic activity. PRMT2 (and PRMT1 as a control) was immunoprecipitated from HEK293 cells, and subjected to a radioactive methylation assay, using hypomethylated whole-cell extract as a substrate. Under standard conditions, only PRMT1 shows detectable activity. When the assay was repeated under more alkaline conditions, some activity was seen for PRMT2. Note that the activity of PRMT2 was very low compared with PRMT1.

• Supplemental Figure S4 -

  Fig. S4. CARM1/PRMT4 shuttles between nucleus and cytoplasm during the cell cycle. HEK293 cells stably expressing CARM1/PRMT4 were investigated by confocal microscopy. (A) Cells were blocked at the G1-S transition by incubation with hydroxyurea or aphidicolin for 20 hours, or left untreated (asynchronous). (B) Cells were blocked at the G1-S transition with aphidicolin as in A, and then released from the block by washing thoroughly in PBS and further cultivation in fresh medium for the indicated times. Cells underwent S phase and G2 before they entered mitosis after 15 hours. Note that CARM1/PRMT4 remains nuclear during S and G2, but is strongly reduced in G1 nuclei. Identical results were obtained by a double thymidine block. (C) Cells undergoing different phases of mitosis were investigated, showing that CARM1/PRMT4 is predominantly nuclear up to prophase, but is absent from condensed chromosomes in later phases of mitosis, and is predominantly cytoplasmic in early G1.

• Supplemental Figure S5 -

  Fig. S5. Activity assay of GFP-PRMT fusion proteins. Longer exposure of a gel identical with the one shown in Fig. 1C. Note that bands in the control lane (without added PRMT) are caused by low activity of endogenous PRMT1 that survived the heat inactivation.