PIM-Related Kinases Selectively Regulate Olfactory Sensations in Caenorhabditis elegans

Mammalian PIM kinases and C. elegans PRKs are true orthologs. A, Amino acid sequences of C. elegans PRK-1 (NM_001276777.1) and PRK-2 (CAA84323.2) were aligned with Homo sapiens PIM-1 (NP_001230115.1), PIM-2 (NP_006866.2), and PIM-3 (NP_001001852.2), and Mus musculus PIM-1 (NP_032868.2), PIM-2 (NP_613072.1), and PIM-3 (NP_663453.1), using the Clustal X 2.0 software (Thompson et al., 1997). Identical or similar amino acids present in all or only in some of them are marked with black or gray backgrounds, respectively. Positions of the conserved ATP-binding and catalytically active sites are highlighted with red and blue lines, respectively. B, The amino acid sequences of the C. elegans (PRK-1 and PRK-2), Danio rerio (PIM-1, PIM-2, and PIM-3), Xenopus laevis (PIM-1 and PIM-3), Mus musculus (PIM-1, PIM-2, and PIM-3) and Homo sapiens (PIM-1, PIM-2, and PIM-3) were used to draw a phylogenetic tree of PIM orthologs and paralogs in distinct species (http://www.phylogeny.fr). The scale of 0.4 refers to 40% difference between sequences.

PIM and PRK kinases function similarly in vitro. A, Radioactive kinase assays were performed to analyze the in vitro ability of human PIM-1 and C. elegans PRK-2 to phosphorylate themselves or the PIM substrate NFATC1 (amino acids 1–418) in the absence (−) or presence (+) of 10 μm DHPCC-9. The intensities of phosphorylated proteins are shown at the top and the total amounts of proteins at the bottom. Shown are also the relative levels of phosphorylation of kinases and their substrates in control versus drug-treated samples in this dataset representing three independent experiments. B, PRK-1 autophosphorylation was analyzed in the absence (−) or presence (+) of 10 μm DHPCC-9, AZD-1208, or SGI-1776. The phosphorylated protein was visualized by immunoblotting with the phospho-RXXS*/T* antibody, which has several potential target sites in PRK-1.

PIM inhibitors specifically suppress olfactory sensing via AWC^ON neurons. Synchronized young adult animals were exposed to indicated concentrations of PIM inhibitors DHPCC-9 (A, B), AZD-1208 (C, D), or SGI-1776 (E, F) for 120 min. Chemotaxis assays with odorants were performed for another 120 min. Chemical structures of the PIM inhibitors are shown in between the dose-dependent chemotactic indices of animals in response to AWC odorants (A, C, E) butanone (1:1000), benzaldehyde (1:200), IAA (1:100), and 2,3-pentanedione (1:10000), or AWA odorants (B, D, F) diacetyl (1:1000), pyrazine (10 mg/ml), and TMT (1:1000). Each data point represents the mean of at least six independent experiments with 150–200 animals per assay. Error bars indicate SEM and asterisks statistically significant differences (p < 0.01) between control and drug-exposed animals.

Inhibition of butanone sensing is reversible and dependent on odorant dosage. Animals exposed for 120 min to 200 μm DHPCC-9 (A, C) or 100 μm SGI-1776 (B, D) were assayed for chemotaxis to butanone (1:1000) after indicated intervals of recovery times (A, B), or to indicated dilutions of butanone (C, D). Each data point represents the means and SEM of four independent experiments with 150–200 animals per assay.

PRK-1 expression and activity is required for butanone sensing. A, Reproductive capacities of wild-type (WT; n = 13), prk-1(pk86) (n = 5), and prk-2(ok3069) (n = 9) animals, as determined by the amounts of eggs laid by each parent. Shown in the box plots are mean (cross), median, quartiles (boxes), and range (whiskers). The asterisks indicate statistically significant differences (p < 0.01) between wild-type and mutant animals. B, Wild-type and mutant animals were tested for chemotaxis toward butanone (1:1000) or IAA (1:100). tax-2 was used as a chemotaxis-deficient control strain. For butanone and IAA, the box plot data are derived from five and three independent experiments, respectively, with 100–150 animals per assay. The asterisks indicate statistically significant differences (p < 0.01) between wild-type and mutant animals. C, D, The same strains were also exposed for 120 min to 200 μm DHPCC-9 and then tested as above toward butanone (C) or IAA (D). The asterisks indicate statistically significant differences (p < 0.01) between wild-type and mutant animals.

PRK-1 is essential also for sensing of olfactory repellents via AWB neurons. A, Wild-type animals were treated for 120 min with indicated concentrations of DHPCC-9 or SGI-1776. Aversion assays with 1-octanol (1:100) were performed for another 120 min. Shown are dose-dependent avoidance indices. Each data point represents the mean and SEM of at least four independent experiments with 100–150 animals per assay. The asterisks indicate statistically significant differences (p < 0.01) between control- and drug-exposed animals. B, WT, prk-1(pk86), and prk-2(ok3069) animals were similarly tested for aversion to 1-octanol. Shown are box plot data collected from three independent experiments with 100–150 animals per assay. The asterisks indicate statistically significant differences (p < 0.01) between wild-type and mutant animals.

PIM inhibitors do not affect responses to gustatory attractants or repellents. Wild-type animals were treated for 120 min with 200 μm DHPCC-9 or 100 μm SGI-1776 and assayed for responses to gustatory attractants (A) or repellents (B–D). A, Shown are box plots of chemotactic indices of control- or drug-exposed animals to 2.5 m NaCl or 2.5 m KCl. Data were collected from three independent experiments with 150–200 animals per assay. B–D, Indicated concentrations of CuSO₄ were pipetted across the midlines of the assay plates. Control- or drug-exposed animals were placed on one side and drops of ethanol (B), butanone (C), or IAA (D) to the other side. After 120 min of incubation, chemotactic indices to the odorants were calculated. Shown are box plot data collected from four independent experiments with 150–200 animals per assay. The asterisks indicate statistically significant differences (p < 0.01) between control- and drug-exposed animals.