Research ArticleCELL BIOLOGY

Dose-dependent functions of SWI/SNF BAF in permitting and inhibiting cell proliferation in vivo

See allHide authors and affiliations

Science Advances  20 May 2020:
Vol. 6, no. 21, eaay3823
DOI: 10.1126/sciadv.aay3823
  • Fig. 1 The SWI/SNF BAF complex promotes cell cycle exit.

    (A) Schematic representation of the two conserved SWI/SNF subcomplexes (using C. elegans nomenclature). (B) Table of C. elegans and mammalian homolog names for SWI/SNF subunits. The SWI/SNF complex consists of core subunits (green), accessory (blue), and BAF- (purple) and PBAF-specific (orange) signature subunits. (C) Lineage of the C. elegans mesoblast (M). The M cell is born during early embryogenesis and initiates proliferation halfway through the first larval stage (L1), forming 14 striated muscle cells (BWM), two scavenger cells [coelomocytes (CC)], and two ventral muscle precursor cells [sex myoblasts (SM)]. The SMs remain quiescent and migrate anteriorly to the vulva, resuming proliferation late in the third larval stage (L3), and differentiate to form 16 muscle cells required for egg laying. (D) Design of the lineage-tracing reporter, single-copy integrated into the C. elegans genome. A universal promoter (Peft-3) drives expression of tagBFP2 flanked by two LoxP sites and followed by the let-858 untranslated region (UTR). Excision of tagBFP2 leads to mCherry expression, providing a visible switch from blue-to-red fluorescence in cells where CRE is expressed and all daughter cells. (E) Representative image of mesoblast lineage descendants marked by the lineage tracing construct in an L4 larva (lateral view, ventral down; arrowheads point to BWM, brackets indicate egg-laying muscle precursors). (F) Representative images of the vulva region of RNAi-treated larvae. Anterior to the left, ventral down; scale bars, 10 μm in all images. (G) Quantification of mesoblast lineage descendants per animal at the L4 stage following RNAi by feeding of synchronized L1 larvae for the indicated genes, in wild-type or fzr-1 mutant backgrounds. Twenty to 30 animals were scored for each condition.

  • Fig. 2 Knockouts of endogenous SWI/SNF genes induce cell division phenotypes that differ greatly from the corresponding RNAi phenotypes.

    (A) Schematic of Lox sites (yellow diamonds) integrated into the endogenous SWI/SNF genes indicated. The swsn-4 ATPase-dead lysine-to-alanine mutation (K to A) is shown as a red block. Transcriptional start sites are indicated with arrows, introns are shown as black lines, and exons as colored blocks. (B) Representative images of mesoblast lineage descendants in wild-type and indicated SWI/SNF gene knockout animals. Arrowheads point at body-wall muscle (BWM), and brackets indicate egg-laying muscle precursors. Scale bar, 10 μm. (C) Quantifications of mesoblast lineage descendants for the indicated genotypes, in the tail area (early dividing body wall muscles), and around the vulva (late-dividing egg-laying muscles). Note that, in contrast to RNAi, SWI/SNF gene knockouts lead to overproliferation of the early dividing BWM precursors and cell division arrest of the late-dividing egg-laying muscle precursor cells. Twenty-six to 35 animals were scored per condition.

  • Fig. 3 Opposite and acute cell division phenotypes depending on the dosage of residual SWI/SNF activity.

    (A) Quantification of total number of mesoblast descendants for the indicated genotypes. *P ≤ 0.05, ****P ≤ 0.0001. (B) Quantification of mesoblast descendants for the indicated genotypes in the tail (early divisions) and around the vulva (late divisions). (C) Expression of SWSN-1::GFP in the M cell at 4 and 7 hours of larval development following gene knockout or gene knockout combined with protein degradation (+Prot. Deg.). Scale bars, 10 μm. Arrows indicate mesoblast cells, outlined in zoom images (scale bars, 1 μm). (D) Quantification of SWSN-1::GFP by fluorescence intensity in M for indicated times and genotypes (normalized to wild-type levels). (E) Quantifications of total numbers of mesoblast descendants for the indicated genotypes. (F) Representative images of strong overproliferation following SWSN-1 protein degradation and lin-23 RNAi (top), and of the one-cell arrest (arrow) after swsn-1 gene knockout together with protein degradation in lin-23 RNAi (bottom). Scale bars, 10 μm. (G) Quantifications of mesoblast descendants for indicated genotypes, in the tail (early) and around the vulva (late). (H) Quantification of SWSN-1::GFP by fluorescence intensity in M descendants for indicated times and genotypes (normalized to wild-type levels). (I) Quantifications of total numbers of mesoblast descendants for the indicated genotypes and times. n numbers of worms scored for all panels in fig. S4D. A.U., arbitrary units.

  • Fig. 4 The SWI/SNF complex is continuously required in epidermal precursor cells.

    (A) The lineage of vulval development. The y axis indicates the time (hours) of larval development after hatching; vertical lines represent vulval precursor cells (VPCs), horizontal lines are cell divisions, and hyp7 denotes hypodermal fusion fate. (B) Maximum projection of the vulva after 40 hours of development, at the time when quantifications are carried out. Vulval nuclei express mCherry from the lineage-tracing reporter after Plin-31::CRE activity. Individual nuclei are easily identifiable. Scale bar, 10 μm. See movie S1 for Z-stack. (C) Quantification of the number of vulval nuclei for the indicated genotypes. The lin-35(n745) mutation did not affect the VPC division pattern but was included to increase the efficiency of RNAi, as the neuroblast derived precursor (P) cells are relatively resistant to RNAi. Eleven to 90 animals were scored per condition. **P ≤ 0.01, ****P ≤ 0.0001. ns, not significant.

  • Fig. 5 Simultaneous knockout of Polycomb PRC2 and SWI/SNF subunit genes rescues mesoblast descendant overproliferation but not cell division arrest.

    (A) Schematic of Lox site integrations into the endogenous EZH2-related Polycomb gene mes-2, with Lox sites indicated by yellow diamonds. The transcriptional start site is indicated with an arrow, introns are shown as black lines, and exons as colored blocks. (B) Quantifications of mesoblast lineage descendants in the indicated genotypes, in the tail area (early-dividing BWM), and around the vulva (late-dividing egg-laying muscles). Thirteen to 27 animals were scored per condition. (C) Quantifications of total numbers of mesoblast lineage descendants in the indicated genotypes. Sixteen to 35 animals were scored per condition. *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001.

  • Fig. 6 Transcriptome RNA-seq demonstrates a continuous requirement for SWI/SNF BAF complexes in normal transcription and proliferation control.

    (A) Quantification of M cell DNA content in synchronized L1 larvae after 7 hours of development for the indicated genotypes. Wild-type M cells have undergone the S phase but not yet divided, leading to a 4C DNA content, whereas M cells in swsn-1::gfpLox + Prot. Deg. animals show an approximately 2C DNA content, indicative of cells arresting in the G1/early S phase. DNA was stained with propidium iodide, and DNA content normalized to that of differentiated embryonic BWM cells (2C). (B) Quantifications of mesoblast descendants for the indicated genotypes and treatments in L1 larvae. Fourteen to 28 animals were quantified per condition. (C) Principal components analysis (PCA) indicating clustering of replicate RNA-seq libraries, prepared from fluorescence-activated cell sorting (FACS)–sorted 2000-cell samples from wild-type and swsn-1::gfpLox + Prot. Deg. L1 larvae at 5.5 hours of development. Samples “A” and “B” are true biological replicates, with RNA-seq libraries prepared from different starting populations of synchronized worms. Within A and B, duplicate/triplicate RNA-seq libraries were prepared from different 2000-cell populations, isolated from the same starting worm population, and can thus be considered “semibiological” replicates. (D) Volcano plot indicating differentially expressed genes between swsn-1::gfpLox + Prot. Deg. and wild-type isolated mesoblast cells at 5.5 hours of development. (E) Quantifications of the number of mRNA molecules per cell in smFISH experiments for the indicated genes, in synchronized L1 larvae at 6.5 hours of larval development (just before the usual first M division), and in wild-type compared with swsn-1::gfpLox + Prot. Deg. Twenty to 27 animals were scored per condition. **P ≤ 0.05.

Supplementary Materials

  • Supplementary Materials

    Dose-dependent functions of SWI/SNF BAF in permitting and inhibiting cell proliferation in vivo

    Aniek van der Vaart, Molly Godfrey, Vincent Portegijs, Sander van den Heuvel

    Download Supplement

    This PDF file includes:

    • Figs. S1 to S6
    • Tables S1 and S2

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article