Engineering the ABA plant stress pathway for regulation of induced proximity

Abstract

Chemically induced proximity (CIP) systems use small molecules and engineered proteins to control and study biological processes. However, small molecule-based systems for controlling protein abundance or activities have been limited by toxicity, instability, cost, and slow clearance of the small molecules in vivo. To address these problems, we modified proteins of the plant abscisic acid (ABA) stress response pathway to control the proximity of cellular proteins and showed that the system could be used to regulate transcription, signal transduction, and subcellular localization of proteins in response to exogenously applied ABA. We also showed that the ABA CIP system can be combined with other CIP systems to simultaneously control multiple processes. We found that, when given to mice, ABA was orally available and had a 4-hour half-life. These properties, along with its lack of toxicity and low cost, suggest that ABA may be well suited for therapeutic applications and as an experimental tool to control diverse cellular activities in vivo.

Fig. 1

The use of ABA-induced proximity for domain reconstitution, detected as induction of gene expression. (A) Regions of PYL_cs and ABI_cs used in our studies. (B) Scheme of ABA-induced luciferase activation and the design of the constructs. (C) Dose response of ABA-induced luciferase activation for 24 hours in TC1 ES cells. (D) Dose response of ABA- or Rap-induced luciferase activation for 24 hours in CHO cells. (E) Time course (0 to 24 hours) of luciferase activation by ABA or Rap in E14 cells. (F) Time course (0 to 3 hours) of luciferase activation by ABA or Rap in E14 cells. (G) Time course of luciferase activity upon drug withdrawal after induction for 24 hours in CHO cells. For (C) through (G), the cells were transfected with the ABA- or Rap-activator cassette and the luciferase reporter for 24 hours before addition of ABA. For all experiments, the induction fold change was calculated relative to the values of noninduced samples. Data are the means ± SEM (n = 3) from experiments of three independent transfections and induction from the same passage of cells. Independent experiments were repeated six or more times.

Fig. 2

The use of ABA-induced proximity to control protein subcellular localization and signal transduction. (A) Left: cytoplasmic localization of GFP-PYL_cs to Numb-ABI_cs induced by ABA in 293T cells. Numb-ABI_cs was detected with an antibody that recognizes the FLAG tag. Right: nuclear localization of Numb-GFP-PYL_cs to ABI_cs-NLS induced by ABA in 293T cells. In both panels, nuclei are stained with DAPI. (B) ABA-induced ERK phosphorylation by SOS localization. Left panel: Cells were transfected with myr-ABI_cs and SOS-PYL_cs (lanes 1 to 5), myr-ABI_cs and SOS-FKBP (lanes 6 and 7), or SOS-PYL_cs only (lanes 8 and 9) for 24 hours, and then 250 μM ABA was added for the indicated period. Right: quantification of ERK phosphorylation in the left panel. The fold change was calculated by a-phos ERK/a-HA (for SOS fusion proteins) and then normalized to lane 1. The results are representative of three independent experiments. a-phos ERK, an antibody that recognizes phosphorylated ERK1/2; a-HA, an antibody that recognizes the HA tag; a-FLAG, an antibody that recognizes the FLAG tag; a-Hsp90, an antibody that recognizes heat shock protein 90; a-ERK, an antibody that recognizes total ERK1/2.

Fig. 3

Independent control of transcription and protein localization by the ABA and Rap systems. (A) Independent induction of luciferase by ABA and Rap for 24 hours in CHO cells. The cells were transfected with the ABA- or Rap-activator cassette and the luciferase reporter for 24 hours before addition of drug. Induction fold change was calculated relative to the values of noninduced samples. Data are the means ± SEM (n = 3) from experiments of three independent transfections and induction from the same passage of cells. Six independent repeats were performed. (B) Independent localization of mCherry-PYL_cs and GFP-Frb by ABA and Rap. 293T cells were transfected with all of the indicated constructs 24 hours before the addition of ABA, Rap, or both at the indicated concentrations for 2 hours.

Fig. 4

ABA-induced proximity can be engineered free of phosphatase activity. (A) Sequence alignment of the PP2C domain from different species. Abbreviations for the amino acids are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr. (B) Phosphatase activity of wild-type and mutant GST-ABI_cs. Phosphatase activity was measured by the ProFluor Ser/Thr PPase Assay Kit (20). Data are the means ± SEM (n = 3) of three independent experiments. (C) Pull-down of GFP-PYL_cs from whole-cell lysates by wild-type (WT) or mutant GST-ABI_cs. GFP-PYL_cs and GST-ABI_cs were detected by antibodies against GFP (a-GFP) or GST (a-GST). Data shown are representative of three independent experiments. (D) Luciferase activation by ABA in CHO cells expressing VP16-PYL_cs and wild-type or mutant Gal4DBD-ABI_cs transgenes for 24 hours. Data are the means ± SEM from experiments of three independent transfections and induction from the same passage of cells.

Fig. 5

ABA is stable and orally available. CHO cells were transfected with the ABA activator cassette and luciferase reporter for 24 hours before addition of ABA-containing serum. (A) ABA stability in cell culture. ABA (100 μM) was incubated with CHO cells for the indicated times and then the medium was used for luciferase activation. (B) Bio-functional assay used to evaluate ABA concentration in serum. (C) ABA stability in human serum. ABA (1 mM) was incubated with fresh human serum for the indicated times and then used [10% (v/v) of culture medium] for luciferase activation. (D) Oral availability of ABA in mice. Serum was collected from mice at the indicated times after gavage of ABA in a mixture of ethanol, Tween 20, and Cremophor (4:3:1), and then 10% (v/v) of culture medium was added and applied to cells for luciferase activation. Induction fold change was calculated on the basis of the values of cells that were not exposed to serum. The data in (A) and (C) are the means ± SEM of triplicates from representative experiments; the data in (D) are the means ± SEM (n = 6 to 8).