Multiple ERK substrates execute single biological processes in Caenorhabditis elegans germ-line development

Abstract

RAS-extracellular signal regulated kinase (ERK) signaling governs multiple aspects of cell fate specification, cellular transitions, and growth by regulating downstream substrates through phosphorylation. Understanding how perturbations to the ERK signaling pathway lead to developmental disorders and cancer hinges critically on identification of the substrates. Yet, only a limited number of substrates have been identified that function in vivo to execute ERK-regulated processes. The Caenorhabditis elegans germ line utilizes the well-conserved RAS-ERK signaling pathway in multiple different contexts. Here, we present an integrated functional genomic approach that identified 30 ERK substrates, each of which functions to regulate one or more of seven distinct biological processes during C. elegans germ-line development. Our results provide evidence for three themes that underlie the robustness and specificity of biological outcomes controlled by ERK signaling in C. elegans that are likely relevant to ERK signaling in other organisms: (i) multiple diverse ERK substrates function to control each individual biological process; (ii) different combinations of substrates function to control distinct biological processes; and (iii) regulatory feedback loops between ERK and its substrates help reinforce or attenuate ERK activation. Substrates identified here have conserved orthologs in humans, suggesting that insights from these studies will contribute to our understanding of human diseases involving deregulated ERK activity.

Fig. 1.

MPK-1 signaling controls seven biological processes during C. elegans germ-line development. (A) Schematic of the conserved RAS–ERK signaling pathway in mammals and C. elegans. ERK/MPK-1 activates or inactivates substrate proteins via phosphorylation. (B) MPK-1 controls seven biological processes during germ-line development (11). Fluorescent micrograph of a dissected C. elegans adult hermaphrodite gonad showing chromosome (blue), activated dpMPK-1 (red), and membrane (green) morphology as germ cells progress through oogenesis. The seven MPK-1 controlled germ-line processes examined in this study are indicated (11) (Fig. S1).

Fig. 2.

In silico and genetic enhancer screen identifies 37 candidate MPK-1 substrates that function during germ-line development. (A) Outline of bioinformatics approach and sequences of characterized ERK-docking sites (DS). (B) Genetic logic used in the enhancer screen to identify genes that promote or inhibit MPK-1 pathway function. Wild-type control and sensitized backgrounds contain the rrf-1(pk1417) mutation, which prevents RNAi in the soma but not the germ line (29). At the permissive temperature where the RNAi screen was performed, mpk-1(ga111ts) ERK has lower MPK-1 activation whereas let-60(ga89ts) RAS has elevated/ectopic MPK-1 activation. (C) Pie chart showing predicted molecular functions of 37 genes that enhance one or the other sensitized genetic background (Dataset S2). (D) Fluorescent micrographs illustrating ddx-19 RNAi enhancement of rrf-1; mpk-1(ga111ts). (E) Fluorescent micrographs illustrating gsk-3 attenuated RNAi enhancement of rrf-1;let-60(ga89ts). (Scale bar, 20 μm.)

Fig. 3.

Multiple molecularly diverse substrates function in each MPK-1-dependent process. (A) (Top) Seven MPK-1-dependent processes, shown relative to adult hermaphrodite germ-line development. (Middle and Bottom) Thirty in vitro ERK2 substrates and their functional molecular categories. +, Germ-line process in which a given substrate functions. Analysis with null alleles for 10 substrates (underlined) (Dataset S2), reveals concordant results with RNAi screen. For oocyte growth control, two substrates enhance mpk-1(ga111ts) and produce large oocytes whereas eight enhance let-60(ga89ts) and produce small oocytes (Fig. S1). (B) Nonoverlapping groups of MPK-1 substrates execute pachytene cellular organization (red circle, 7 genes) and germ cell apoptosis (green circle, 4 genes). (C) Partially overlapping groups of substrates regulate pachytene cellular organization (red) and/or oocyte organization and differentiation (blue).

Fig. 4.

DDX-19 and GSK-3 phosphorylation is MPK-1-dependent in vivo. (A and D) ClustalW protein alignment of C. elegans and human orthologs of DDX-19 (A) identifies two conserved phosphoacceptors (S612 and T745; red circles) and two conserved docking sites (red boxes) and GSK-3 (D) identifies four conserved phosphoacceptor sites (T254, T289, T304, and T310; red circles) and two conserved docking sites (one shown here, red boxes) (Fig. S4). Mapping of ERK2 phosphoacceptors for (B) DDX-19 and (E) GSK-3 using in vitro kinase assays and site-directed mutagenesis (SI Methods). (B and D) (Top) ³²P incorporation into indicated proteins. (Middle) Western blot analysis by using anti-HIS antibody for loading control. (Bottom) Kinetic analyses were performed on wild-type and all phosphoacceptor mutant proteins (S/T to A). Values for K_m and V_max (average ± 1 SD of five experiments) were used to calculate the RAR, which was normalized to 1, relative to MBP. LIN-1 is a positive control. (C and F) Western blot analysis of protein extracts from wild-type (WT), mpk-1(ga117)-null, and (C) ddx-19(ok783)-null or (F) gsk-3(nr2047)-null animals probed with antibodies specific for (C) phospho-DDX-19 (pS612 and pT745), total DDX-19, and paramyosin (loading control), (F) phospho-GSK-3 (pT304 and pT310), total GSK-3, and α-tubulin (loading control). The antibodies are specific to the gene product tested because no species was detected in the respective null mutant. DDX-19 is phosphorylated on S612 and/or T745 in WT but not mpk-1 null. *, slower-mobility GSK-3 band detected with total GSK-3 antibody that comigrates with the species detected with the phospho-GSK-3 antibody. In mpk-1-null, this band is not detected on short exposures; however, on longer exposures, a weak band is visible (also weakly detected by total-GSK-3 antibody).

Fig. 5.

MPK-1 substrates function to regulate MPK-1 activation in feedback loops. (A) Ten MPK-1 substrates regulate MPK-1 activation and fall into three categories. (B) Fluorescent micrographs of dissected rrf-1 null gonads, after either gfp RNAi (Upper) or rba-1 RNAi (Lower), stained for dpMPK-1 (red) or DNA (blue). (Scale bar, 20 μm.) (C) Models of possible regulatory feedback loops through which substrates modulate germ-line MPK-1 activation.