A stress-responsive p38 signaling axis in choanoflagellates

Abstract

Animal kinases regulate cellular responses to environmental stimuli, including cell differentiation, migration, survival, and response to stress, but the ancestry of these functions is poorly understood. Choanoflagellates, the closest living relatives of animals, encode homologs of diverse animal kinases and have emerged as model organisms for reconstructing animal origins. However, efforts to identify key kinase regulators in choanoflagellates have been constrained by the limitations of currently available genetic tools. Here, we report on a framework that combines small molecule-driven kinase discovery with targeted genetics to reveal kinase function in choanoflagellates. To study the physiological roles of choanoflagellate kinases, we established two high-throughput platforms to screen the model choanoflagellate Salpingoeca rosetta with a curated library of human kinase inhibitors. We identified 95 diverse kinase inhibitors that disrupt S. rosetta cell proliferation. By focusing on one inhibitor, sorafenib, we identified a p38 kinase as a regulator of the heat shock response in S. rosetta. This finding reveals a conserved p38 function between choanoflagellates, animals, and fungi. Moreover, this study demonstrates that existing kinase inhibitors can serve as powerful tools to examine the ancestral roles of kinases that regulate modern animal development.

Fig. 1. High-throughput screening of a small molecule library revealed inhibitors of S. rosetta cell proliferation. (A) Treatment of S. rosetta cultures with 1255 different small molecules (see Table S1, ESI†) resulted in a distribution of cell counts, assessed by flow cytometry, at the 24-hour endpoint. S. rosetta cell counts were normalized to the average of DMSO controls within the same plate (dark grey). Compounds determined to significantly inhibit S. rosetta cell proliferation by flow cytometry (based on two-tailed p-value <0.05 calculated from z-score), fall below the dotted line and are indicated in red. Compounds that were not detected as significant inhibitors by flow cytometry but were identified by imaging (based on two-tailed p-value <0.05 calculated from z-score) are in blue. Compounds that were not significant inhibitors for either screen are indicated in light grey. Sorafenib (SO), a focus of this study, is labeled. (B) The range of normalized cell counts measured by flow cytometry for compounds that significantly inhibited S. rosetta cell proliferation. Compounds that were the focus of further study – genistein (GE), glesatinib (GL), PP121, masitinib (MA), sotrastaurin (SOT) – are labeled. (C) Comparison of normalized values of compounds that inhibited S. rosetta cell proliferation, assessed by flow cytometry and the corresponding normalized values determined by imaging. Compounds determined to significantly inhibit S. rosetta cell proliferation (based on two-tailed p-value <0.05 calculated from z-score) by flow cytometry fall below the dotted line on the y-axis and by imaging, to the left of the dotted line on the x-axis.

Fig. 2. Glesatinib and sorafenib, two multi-target tyrosine kinase inhibitors, disrupt S. rosetta cell proliferation and tyrosine phosphosignaling. (A) Treatment of S. rosetta cultures with 1 μM sorafenib and glesatinib led to a complete block of cell proliferation, while treatment with 1 μM masitinib or PP121 led to a partial reduction in cell proliferation relative to DMSO-treated cultures. Two biological replicates were conducted per treatment, and each point represents the mean of three measurements from each biological replicate. For timepoints at 40, 60, and 85 hours, cell densities of inhibitor-treated cultures were significantly different from vehicle (DMSO) (p-value <0.01). Significance was determined by a two-way ANOVA multiple comparisons test. (B) S. rosetta cultures treated with 1 μM or 10 μM sorafenib, glesatinib, or PP121 for 24 hours had reduced normalized cell density, whereas masitinib only had reduced normalized cell density at 10 μM. Normalized cell densities were determined to be reduced if differences between treatments and vehicle (DMSO) were significant (p-value <0.01) Significance was determined by determined by a two-way ANOVA multiple comparisons test. Movies show S. rosetta cells treated with 10 μM glesatinib that undergo cell lysis (Movie S1, ESI†) and sorafenib, that have cell body deformation (Movies S2 and S3, ESI†), in comparison to DMSO control (Movie S4, ESI†). (C) Western blot analysis of S. rosetta cultures treated with 1 μM sorafenib and glesatinib for 1 hour showed a decrease in tyrosine phosphorylation of proteins at ∼60 kDa, ∼45 kDa, and ∼35 kDa (indicated by arrows and detected with pY1000 anti-phosphotyrosine antibody) compared to vehicle (DMSO) control. Masitinib and PP121 did not reduce the phosphotyrosine signal.

Fig. 3. S. rosetta p38 binds to sorafenib. The ActivX ATP probe was used to pull down kinases from S. rosetta lysates that were pretreated with either DMSO or the ATP-competitive inhibitor sorafenib. We found that pretreatment with sorafenib reduced the level of p38 recovered using the ActivX ATP probe, indicating that sorafenib and p38 interact and outcompete ActivX ATP probe binding. Kinases plotted are only those that were identified in both vehicle and sorafenib pre-treatments. For full kinase enrichment list, see Table S2 (ESI†), and for alignment of S. rosetta p38 with those from animals and fungi, see Fig. S9A (ESI†).

Fig. 4. S. rosetta p38 phosphorylation is induced by environmental stressors. (A) Strategy for generating Sr-p38 and Sr-JNK knockout cell lines. The Sr-p38 and Sr-JNK loci were targeted by a guide RNA complexed with Cas9 that anneals before the kinase domain and directs Cas9 to introduce a double-strand break downstream of codon 15 (codon 15 is serine in Sr-p38 and alanine in Sr-JNK), indicated by (*). The Cas9-guide RNA complex was coupled with a double-stranded homology-directed repair to introduce a palindromic premature termination stop sequence and a puromycin resistance cassette. The resulting truncated proteins, Sr-p38^1–15 and Sr-JNK,^1–15 lack the kinase domain and phosphorylation sites (indicted by the extended circle) recognized by both phospho-p38 antibodies used in this study. Protein diagrams were created with IBS 2.0. (B) Heat shock induces Sr-p38 phosphorylation in wild-type cells and the phospho-p38 signal is recognized by the anti-ACTIVE® p38 antibody (Promega #V1211). This phospho-p38 signal is decreased in Sr-p38^1–15 knockout cell lines but not Sr-JNK^1–15 knockout lines, indicating that the Anti-ACTIVE® p38 antibody (Promega #V1211) detects Sr-p38 and that Sr-p38, but not Sr-JNK, responds to heat shock. Three biological replicates of wild-type cells, ten clones of Sr-p38^1–15 and five clones of Sr-JNK^1–15 strains were incubated at 37 °C for one hour. Lysates from the treated cultures were analyzed by western blot with the Anti-ACTIVE® p38 antibody and quantified by densitometry to identify if any changes in Sr-p38 phosphorylation occurred. Significance was determined by a one-way ANOVA multiple comparisons test between wild-type cells and Sr-p38^1–15 or Sr-JNK^1–15. (C) Similar to (B), the phospho-p38 signal recognized by the p38 MAPK pThr180/pTyr182 (Biorad #AHP905) antibody in heat shocked wild-type cells is decreased in Sr-p38^1–15 knockout cells but not Sr-JNK^1–15 knockout cells. (D) S. rosetta cells, normally cultured at 22 °C were incubated at 37 °C to induce heat shock. Lysates from the treated cultures were analyzed by western blot with the Anti-ACTIVE® p38 antibody (Promega #V1211) to identify if any changes in Sr-p38 phosphorylation occurred. 30 minutes of heat shock was sufficient to induce Sr-p38 phosphorylation. (E) S. rosetta cells were treated with hydrogen peroxide, a form of oxidative stress for 10 min and 30 min 10 min of treatment with 0.5 M H₂O₂ at 22 °C was sufficient to induce Sr-p38 phosphorylation detected by the anti-ACTIVE® p38 antibody (Promega #V1211) (F) Sr-p38^1–15 and Sr-JNK^1–15 strains grow similarly to wild-type. Four wild-type cultures and four randomly selected Sr-p38^1–15 and Sr-JNK^1–15 clones were grown in 24-well plates over a 96-hour growth course and showed similar growth. Significance was determined by a two-way ANOVA multiple comparisons test. (G) The induction of Sr-p38 phosphorylation by heat shock was kinase-dependent. S. rosetta cultures pretreated with 10 μM or 1 μM sorafenib for 30 minutes followed by 30 minutes of heat shock at 37 °C and probed with the Anti-ACTIVE® p38 antibody (Promega #V1211) had decreased Sr-p38 phosphorylation. APS6-46 treated cultures were not different from vehicle (DMSO) control. In (D), (E) and (G), 4–12% bis-Tris SDS-PAGE gels were used to resolve the bands observed.

Fig. 5. Sr-p38 regulates the heat shock response in S. rosetta. Proposed mechanism for regulation of the stress-responsive Sr-p38 axis. Sr-p38 is phosphorylated by upstream kinases in response to heat shock. Sorafenib, a multi-kinase inhibitor, targets kinases upstream of Sr-p38 and disrupts Sr-p38 signaling. Separately, sorafenib and 94 other small molecules inhibit S. rosetta cell proliferation by targeting an unknown kinase that regulates S. rosetta cell proliferation.