The nematode homologue of Mediator complex subunit 28, F28F8.5, is a critical regulator of C. elegans development

Abstract

The evolutionarily conserved Mediator complex is a critical player in regulating transcription. Comprised of approximately two dozen proteins, the Mediator integrates diverse regulatory signals through direct protein-protein interactions that, in turn, modulate the influence of Mediator on RNA Polymerase II activity. One Mediator subunit, MED28, is known to interact with cytoplasmic structural proteins, providing a potential direct link between cytoplasmic dynamics and the control of gene transcription. Although identified in many animals and plants, MED28 is not present in yeast; no bona fide MED28 has been described previously in Caenorhabditis elegans. Here, we identify bioinformatically F28F8.5, an uncharacterized predicted protein, as the nematode homologue of MED28. As in other Metazoa, F28F8.5 has dual nuclear and cytoplasmic localization and plays critical roles in the regulation of development. F28F8.5 is a vital gene and its null mutants have severely malformed gonads and do not reproduce. F28F8.5 interacts on the protein level with the Mediator subunits MDT-6 and MDT-30. Our results indicate that F28F8.5 is an orthologue of MED28 and suggest that the potential to link cytoplasmic and nuclear events is conserved between MED28 vertebrate and nematode orthologues.

Keywords: Caenorhabditis elegans; Development; Evolution; Gene expression regulation; MED28; Mediator complex.

Figure 1. A multiple sequence alignment of selected metazoan homologues of MED28 compared with F28F8.5.

Aligned with PROMALS (http://prodata.swmed.edu/promals/promals.php), variable C- and N-termini not shown, amino acid residue types colored according to Clustal scheme in Jalview, red bars indicate consensus positions of predicted α-helices. Sequences from top to bottom (organism, identifier): Caenorhabditis elegans, O18692; Trichinella spiralis, E5RZQ1; Wuchereria bancrofti, EJW84794.1; Pristionchus pacificus, translated contig of CN657719.1 FG102945.1 CN657262.1 CN656622.1; Ancylostoma ceylanicum, A0A016SKV7; Globodera pallida, translated CV578368.1; Panagrolaimus davidi, translated JZ658977.1; Ixodes scapularis, B7PAW5; Drosophila melanogaster, MED28_DROME; Homo sapiens, MED28_HUMAN. Readers with specific color preferences may download the compared sequences from (File S1) and create the Clustal scheme with different color specifications using the Jalview program (http://www.jalview.org/).

Figure 2. Expression pattern of GFP::F28F8.5 in homozygous animals with edited F28F8.5 gene.

GFP tagged to F28F8.5 at its N-terminus using CRISPR/Cas9 technology visualized the expression of F28F8.5 in the gonads (Panels B and D, arrows) in mitotic nuclei and continues throughout the embryonic development (Panels F, H, J, and L). The wide and likely ubiquitous expression of GFP::F28F8.5 continues during larval stages (larvae L3 and L4 are shown in panels M, N and O, P, respectively) as well as in adults (panels Q and R). Expression of the edited gene in the nuclei of the developing vulva is indicated by the arrows in panels M and N. Panels A, C, E, G, I, K, M, O, and Q show larvae in Nomarski optics and panels B, D, F, H, J, L, N, P, and R in GFP fluorescence. Bars represent 50 μm.

Figure 3. Analysis of GFP::F28F8.5 expression in homozygous animals with edited F28F8.5 gene by confocal microscopy and fluorescence lifetime imaging microscopy (FLIM).

All confocal images of GFP fluorescence are recorded in Channel 1 (495–525 nm). FLIM images (panels A to E, H, J, L, M, O to S, U, and V) are calculated from merged recordings in Channel 1 and Channel 2 (525–585 nm). Panels F, K, and N are images in Nomarski optics at the same optical focus as in corresponding confocal images of GFP fluorescence (panels G, I, L, O). Panels A and B show control images of WT embryos in bean and comma stages, respectively. No specific signal is detected in control embryos by FLIM. Panel C shows an embryo in bean stage expressing GFP::F28F8.5 from the edited gene. Two focal planes of an embryo expressing GFP::F28F8.5 in twofold stage are shown in panels D and E. FLIM detects GFP::F28F8.5 in most or all nuclei of developing embryos. Panels F to J show the distal part of a young adult control hermaphrodite animal and a control embryo in late bean stage. FLIM images in panels H and J show mostly short lifetime fluorescence in the cytoplasm of embryonic cells as well as cells and subcellular structures in the adult control animal (visualized by blue color). Arrowheads pointing at the embryo in panels G and H indicate weak autofluorescence in the cytoplasm of embryonic cells. Arrows in panels F, G and H indicate the turn of gonad and arrowheads indicate nuclei of an enterocyte which is devoid of almost all fluorescence (panels G and H). Panels K to S and U and V show animals with edited F28F8.5 (gfp::F28F8.5). Panel M shows the distal part of an adult hermaphrodite animal expressing GFP::F28F8.5 from the edited gene at recording settings identical with that used in the control sample shown in panels A, B, H, and J. FLIM analysis shows a long lifetime fluorescence in nuclei and in the cytoplasm of most cells that contrasts with the low level of fluorescence seen in the control sample. Arrowheads indicate nuclei of enterocytes in panels L and M. Panels N to V show images of an adult animal and two L1 larvae with edited F28F8.5. Panels P to S, and U show selected focal planes in FLIM. Panel T shows the calibration table for FLIM in the range of 1–3 ns used in all panels presenting FLIM analysis. Blue areas shown in FLIM pictures represent short lifetime fluorescence presumably corresponding to autofluorescence (blue arrows in panels O, S, and U). Arrowheads in panels O, P, R, and S indicate nuclei of enterocytes and in panel U nuclei of early embryos with long lifetime fluorescence characteristic for GFP. Panel V shows the distal part of a male expressing GFP:F28F8.5 in male specific structures, in nuclei as well as in rays (marked by arrowheads) indicating that GFP::F28F8.5 is expressed not only in cell nuclei but also in the cytoplasmic structures. Bars represent 30 μm in panels A to E and 50 μm in panels F to S and U and V.

Figure 4. Downregulation of F28F8.5 by RNAi induces developmental defects.

Animals developed from parents injected with dsRNA specific for F28F8.5 show retention of normal and malformed embryos (panels A and K, respectively), vacuoles (panels A and C, arrows), herniation and burst through the vulva (panels B, C, I, and J, arrowheads) and defective development of the gonad (panels J and M). Panel D shows a male nematode with defects of male specific structures—missing rays and fan and an abnormal distal part of the body (arrow). Panel E shows the magnified distal part of the male nematode in panel D and the defective male specific structures (arrow). Panels F and H show L3 larvae that were found atrophic, with thin enterocytes (arrowheads) and a dilated gut lumen (arrows). The dumpy phenotype with masses of tissue and vacuoles (panel G, arrows) were also common in the progeny of microinjected parents. Other phenotypes seen included molting defects indicated by arrows in panel L and cellular defects (indicated by arrowheads in panels L and M). Animals treated by control RNAi were morphologically normal and representative images are shown in panels N to Q. Panel N shows a L2 larva, panel O shows a young L3 larva with developing germline (arrow). Panel P shows a young L4 Larva with developing vulva marked by an arrow. Panel Q shows a grossly normal adult hermaphrodite animal with few developing embryos and vulva (arrow). Panel R and S show the distal part of the body of a male animal with normal appearance of male specific structures. Arrow marks spicules (in panel R) and normal sensory rays (in panel S). All images are in Nomarski optics. Bars represent 50 μm.

Figure 5. Disruption of F28F8.5 by CRISPR/Cas9 technique.

Animals with disrupted F28F8.5 on both alleles express GFP under the regulation of F28F8.5 promoter. Panels A to J show paired images of animals in Nomarski optics and in GFP fluorescence. Panels A and B show an adult hermaphrodite animal with diffuse fluorescence in cells in the head area including anterior arms of the excretory cell (arrowheads). Panels C and D show a malformed larva probably in L3 stage with a Dpy phenotype and diffuse fluorescence in a malformed gonad (arrows) and the intestine (arrowheads). Panels E and F show an adult hermaphrodite animal with diffuse fluorescence in gut, pharyngeal cells and severely malformed gonad containing irregular structures (arrows). Panels G, H, I, and J show an adult animal with a malformed gonad, Pvul phenotype, dense gut and diffuse GFP fluorescence throughout the body. Panels K and L show the central part of the body of a hermaphrodite with the Pvul phenotype (arrowhead) and malformation of gonad (arrow). Panel M is composed of three consecutive images showing an adult hermaphrodite animal with severely malformed gonad (arrows), and missing uterus and spermathecae. The fluorescence images show that unlike GFP::F28F8.5, GFP alone localizes diffusely in the cytoplasm and is not found in nuclei. Bars represent 50 μm.

Figure 6. Heterozygous animals with one edited disrupted allele of F28F8.5 and one WT allele.

Heterozygous hermaphrodites carrying one edited allele of F28F8.5 and one WT had grossly normal appearance and could be recognized by rol phenotype, presence of embryos, weak mostly cytoplasmic GFP fluorescence and absence of nuclear localization of GFP fluorescence. Panels A (Nomarski optics) and B (GFP fluorescence) show an L3 larva with weak fluorescence (panel B, arrowhead points at the gonad and arrows point at the head and pharynx). Inlets show head area at higher magnification (rotated 90° clockwise). Panels C and D show an adult hermaphrodite animal (C in Nomarski optics and D in GFP fluorescence) with weak cytoplasmic fluorescence in most cells. The arrowhead in panel D points at the nucleus of an enterocyte in focal plane that is devoid of GFP fluorescence. Arrows indicate two embryos with GFP fluorescence accumulated in nuclei which is most likely the result of spontaneous SEC self-excision. Panels E and F show an adult hermaphrodite in Nomarski optics (panel E) and GFP fluorescence (panel F). Arrows indicate the head area with diffuse intracellular fluorescence visible in panel F. Arrowheads point at two nuclei of enterocytes in focal plane that are also devoid of fluorescence. In contrast to the animal shown in panels C and D, the animal shown in the panel E and F contains embryos that have mostly diffuse cytoplasmic expression of GFP. Bars represent 50 μm.

Figure 7. Binding of F28F8.5 to MDT-6 and MDT-30 in vitro.

GST-F28F8.5 or GST alone were expressed in bacteria and purified using glutathione–agarose beads that were incubated with ³⁵S-Methionine-MDT-6 produced using rabbit reticulocyte lysate (A and B) or His₆-MDT-30-FLAG expressed in bacteria and purified using a nickel column (C and D). Panels A and C show fractions bound to glutathione–agarose beads resolved by polyacrylamide gel electrophoresis and visualized by autoradiography (panel A). For quantification, dried gel areas corresponding to proteins detected by radioactivity were excised and the radioactively labeled MDT-6 was determined using scintillation counter (panel B). Panels C and D show the interaction of FLAG-MDT-30 with GST-F28F8.5 or GST alone. FLAG-MDT-30 pulled down by GST or GST-F28F8.5 was determined by Western blot using an anti-FLAG antibody (panel C) and by densitometry (panel D). Both assayed Mediator subunits, MDT-6 and MDT-30 bind GST-F28F8.5 preferentially in comparison to GST only.