A Chemically Inducible Helper Module for Detecting Protein–Protein

Apr 6, 2017 - In this system, a mutant of FK506-binding protein 12 (FKBPF36 V) is fused with a protein of interest and the intracellular domain of a r...
0 downloads 0 Views 1MB Size
Download PDF
Subscriber access provided by University of Newcastle, Australia

Article

A chemically inducible helper module for detecting protein– protein interactions with tunable sensitivity based on KIPPIS Daiki Kashima, Raiji Kawade, Teruyuki Nagamune, and Masahiro Kawahara Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04063 • Publication Date (Web): 06 Apr 2017 Downloaded from http://pubs.acs.org on April 11, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

A chemically inducible helper module for detecting protein–protein interactions with tunable sensitivity based on KIPPIS Daiki Kashima, Raiji Kawade, Teruyuki Nagamune, Masahiro Kawahara* Department of Chemistry and Biotechnology, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan

ABSTRACT: As protein–protein interactions (PPIs) play essential roles in regulating their functional consequences in cells, methods to detect PPIs in living cells are desired for correct understanding of intracellular PPIs and pharmaceutical development therefrom. Here we demonstrate a c-kit-based PPI screening (KIPPIS) system in combination with a chemically inducible helper module for detecting PPIs in living mammalian cells. In this system, a mutant of FK506-binding protein 12 (FKBPF36V) is fused with a protein of interest and the intracellular domain of a receptor tyrosine kinase c-kit. Constitutive expression of two fusion proteins with interacting proteins of interest in interleukin-3 (IL-3)-dependent cells results in dimerization and subsequent activation of the c-kit intracellular domains, which allows cell proliferation in a culture medium devoid of IL-3. A helper ligand, a small synthetic chemical that homodimerizes FKBP F36V, assists the formation of stable complexes of the fusion proteins and serves as a tuner for sensitivity of the system. Using this system, two model PPIs were successfully detected based on cell proliferation, which was featured by the helper ligand- and PPIdependent phosphorylation of Src family kinases, a hallmark of the c-kit signaling. Notably, the inclusion of the helper module enabled PPI detection with tunable sensitivity. The helper-assisted KIPPIS allows us to configure various affinity thresholds by changing the concentration of the helper ligand, which may be applied to select affinity-matured variants based on cell proliferation advantage.

INTRODUCTION Proteins play important roles in regulating cellular events in vivo.1,2 A marked example is signal transduction, in which a number of signaling molecules are spatiotemporally organized to exactly and efficiently transduce signals for controlling cell fate.3-6 Protein–protein interactions (PPIs) are highly important for proteins to communicate with one another for leading to specific cellular events. Therefore, many researchers have contributed to the development of interactomics.7 To date, surface plasmon resonance (SPR)8-10, mass spectrometry (MS)11-13, isothermal titration calorimetry (ITC)14-16, and nuclear magnetic resonance (NMR)17-22 are widely used for analysing PPIs in vitro. Despite the significance of data analyzed by these methods, we need to carefully evaluate whether the data reflect actual intracellular PPIs because cells are filled with a high-density protein solution, which leads to completely different PPI outcomes due to molecular crowding effects.23,24

In order to detect intracellular PPIs directly in living cells, many investigators have developed artificial detectors in living cells.25-28 For example, fluorescence or bioluminescence resonance energy transfer (FRET29-31 or BRET32,33), bimolecular fluorescence complementation (BiFC)34,35, and protein fragment complementation assay (PCA)36-40 are well-known PPI detection systems. In these systems, functional proteins or their fragments are fused with proteins of interest, whose interaction is probed by fluorescence or enzymatic activity. Moreover, the twohybrid system allows PPI-dependent expression of reporter proteins in the nucleus of eukaryotic cells.41-43 Several systems inspired from signal transduction mechanisms have also been developed and named membrane yeast two-hybrid (MYTH),44-46 mammalian membrane twohybrid (MaMTH),47 kinase substrate sensor (KISS),48 and mammalian protein–protein interaction trap (MAPPIT).4951 In MYTH and MaMTH, proteins of interest are fused with the N- and C-terminal halves of ubiquitin that is linked to a transcription factor. Their interaction induces ubiquitin reconstitution, resulting in cleavage by endogenously expressed deubiquitinating enzymes and

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 9

subsequent release of the transcription factor. This method allows detection of stimuli- and phosphorylationdependent PPIs based on transcriptional activity of a reporter gene. In KISS and MAPPIT, which are good mimics of the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway, prey fused with the STAT3 recruitment sites and bait fused to JAK or a JAKbinding scaffold are designed. In both systems, PPIs between bait and prey induce the phosphorylation of STAT3, which translocate into the nucleus and enhance transcriptional activity of a reporter gene. While these methods well reflect PPIs in mammalian cells in principle, the fusion constructs are transiently overexpressed in the cells by transfection in order to gain sufficiently intense signals, which involves potentially high backgrounds in readout. We previously reported a mammalian PPI detection system based on chimeric proteins in which the intracellular domain of a receptor tyrosine kinase c-kit is fused with proteins of interest.52 Cells were established by retroviral transduction that allows stable expression of the chimeras. In this c-kit-based protein–protein interaction screening (KIPPIS) system, the c-kit intracellular domain dimerizes depending on interaction of proteins of interest and triggers a cell growth signal, thereby functioning as a converter. While a pDI peptide–MDM2 interaction (Kd = 20 nM)53 was successfully detected based on cell proliferation, a relatively weak p53 peptide–MDM2 interaction (Kd = 140 nM),53 which is a common model for endogenous protein–protein interactions within cells, was under the detection limit. In this study, we aimed to improve the sensitivity of the KIPPIS system. We postulated that unstable interaction between the proteins of interest prevented the c-kit intracellular domain dimer from efficient activation and subsequent signaling. In order to improve the sensitivity of KIPPIS, here we introduce a “helper module” for supporting the interaction between the proteins of interest. The introduction of a constitutive helper interaction (using the WW domain and its target peptide) into FRET probes has already proved to be effective.54 While the report elegantly demonstrated the improvement of sensitivity, an affinity threshold for selection cannot be set in principle due to the irreversible nature of the constitutive helper interaction. Here we propose a small chemical-dependent homodimerizing protein as a novel helper module that mediates enhanced sensitivity. We demonstrate that the chemically inducible helper module is an enabling building block to dramatically enhance the sensitivity of KIPPIS.

MATERIALS AND METHODS Constructs Experimental procedures to obtain template plasmids (pFB-FKBP-(G4S)2-kit-IN, pMK-FRBT2098L-(G4S)2-kit-IP, pFB-MDM2-(G4S)2-kit-IN, and pMK-p53pep-(G4S)2-kit-IP) were described in a previous report.52

Figure 1. The construction of FRBT2098L- and FKBP-fused chimeras. (a) The chemical structures for synthetic ligands AP20187 (helper ligand) and a rapamycin analog AP21967. (b) The constructs of chimeras with a helper module (FKBPF36V), protein of interest (FRBT2098L or FKBP), and the ckit intracellular domain (c-kit ICD). In the presence of both AP21967 and the helper ligand, the chimeras form stable complexes. The abbreviated chimera names are shown in red. (c) The constructs of chimeras without the helper module. Without support of the helper interaction, the c-kit intracellular domain dimer would be too unstable to trigger a strong growth signal (gray arrows). The abbreviated chimera names are shown in red.

Firstly, pFB-FKBP-(G4S)2-kit-IN and pFB-MDM2-(G4S)2kit-IN were digested with EcoRI and BamHI and ligated with a pMK-backbone plasmid to obtain pMK-FKBP(G4S)2-kit-IN and pMK-MDM2-(G4S)2-kit-IN. To insert a helper module FKBPF36V at the N-terminus of proteins of interest, pMK-FKBP-(G4S)2-kit-IN, pMKFRBT2098L-(G4S)2-kit-IP, pMK-MDM2-(G4S)2-kit-IN, and pMK-p53pep-(G4S)2-kit-IP were linearized by PCR using primers listed in Supplementary Table 1. As inserts, the FKBPF36V gene was amplified using primers summarized in Supplementary Table 1. These DNA fragments were fused by an In-Fusion HD (Takara Bio, Shiga, Japan) enzyme to obtain pMK-FKBPF36V-(G4S)3-FKBP-(G4S)2-kit-IN, pMK-FKBPF36V-(G4S)5-FKBP-(G4S)2-kit-IN, pMK-FKBPF36V(G4S)3-FRBT2098L-(G4S)2-kit-IP, pMK-FKBPF36V-(G4S)5FRBT2098L-(G4S)2-kit-IP, pMK-FKBPF36V-(G4S)3-MDM2(G4S)2-kit-IN, pMK-FKBPF36V-(G4S)5-MDM2-(G4S)2-kit-IN,

ACS Paragon Plus Environment

Page 3 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

pMK-FKBPF36V-(G4S)3-p53pep-(G4S)2-kit-IP, and pMKFKBPF36V-(G4S)5-p53pep-(G4S)2-kit-IP. The correspondence between the plasmid names and abbreviated construct names is summarized in Supplementary Table 2. To replace the p53 peptide (SQETFSDLWKLLPEN) with its 3A mutant (SQETASDLAKLAPEN), pMK-FKBPF36V(G4S)3-p53pep-(G4S)2-kit-IP and pMK-FKBPF36V-(G4S)5p53pep-(G4S)2-kit-IP were amplified by PCR using primers (3Ap53-f/G3-3Ap53-r and 3Ap53-f/G5-3Ap53-r, respectively) based on PrimeSTAR® Mutagenesis Basal Kit (Takara Bio), resulting in pMK-FKBPF36V-(G4S)3-p53pep(3A)(G4S)2-kit-IP and pMK-FKBPF36V-(G4S)5-p53pep(3A)(G4S)2-kit-IP.

a

b

Chemicals AP20187 (helper ligand) and AP21967 were purchased from Takara Bio (Otsu, Japan). Cell culture, proliferation assay and western blotting Cell lines used and experimental procedures were similar as described previously.52 The detailed experimental conditions were described in Supporting Information.

RESULTS The helper interaction enhances c-kit intracellular domain-derived signals in FRBT2098L-and FKBP-fused chimeras To evaluate KIPPIS with the chemically inducible helper module, we firstly focused on FK506-binding protein 12 (FKBP) and a mutant of the FKBP–rapamycin-binding domain (FRBT2098L), the latter of which is derived from FKBP–rapamycin-associated protein (FRAP), as proteins of interest.55 The heterodimer formation of FRBT2098L and FKBP can be induced by a rapamycin analog AP21967 (Figure 1a).56 On the other hand, the wild-type FKBP was mutated to create FKBPF36V so that the AP20187 (hereafter called “helper ligand”) conditionally induces homointeraction of FKBPF36V.57 The helper ligand has much higher affinity (Kd = 0.094 nM) against FKBPF36V than rapamycin and its analogues (including AP21967), but not against endogenous wild-type FKBP, which facilitates orthogonal control of FKBPF36V homodimerization within cells. Therefore, the cross-reactivity between the helper ligand and AP21967 is considered to be modest also in our current study. FKBPF36V was genetically fused with the N-terminus of each protein of interest as a helper module (designated as “H” in the chimera names) (Figure 1b). Additionally, the intracellular domain of c-kit58 was fused with the Cterminus of each protein of interest. To improve the conformational flexibility of the fusion proteins, either (GlyGly-Gly-Gly-Ser (G4S))3 or (G4S)5 flexible linker (designated as “3” or “5” in the chimera names) was used for connecting the helper module with the protein of interest, whereas a (G4S)2 flexible linker was used for connecting the protein of interest with the c-kit intracellular domain. A pair of fusion proteins without the helper module was

Figure 2. The helper interaction enhances c-kit intracellular domain-derived signals in FRBT2098L- and FKBP-fused chimeras. The chimeras expressed in the Ba/F3 transductants are shown as the abbreviated names. (-) indicates the parental Ba/F3 cell. (a) Western blotting for detecting PPI-dependent tyrosine phosphorylation of Src family kinases (SFK). Cells were stimulated for 30 min with 100 nM of the helper ligand and with/without 50 nM AP21967 and lysed. Western blotting was performed with the following antibodies; anti-phospho-tyrosine of SFK (p-SFK), anti-SFK, anti-V5 tag to detect the expression level of R chimeras, anti-HA tag to detect the expression level of K chimeras, and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a loading control. (b) Western blotting for analyzing helper ligand- and AP21967-dependent phosphorylation of SFK. Cells were stimulated for 30 min with/without 100 nM of the helper ligand and with/without 50 nM AP21967 and lysed. Western blotting was performed with the following antibodies; anti- p-SFK, anti-SFK, and anti-GAPDH as a loading control. also constructed as a negative control (Figure 1c). V5 and hemagglutinin (HA) epitope tags were appended to the N-terminus of the FRBT2098L (R) and FKBP (K) chimeras, respectively. Overall, the chimera names are designated as HnR or HnK, where each abbreviated character is aligned to represent the domains linked from N- to Cterminus and n means the repeat number of the flexible linker (3 or 5) (Supplementary Table 2). The chimeras were stably expressed in murine interleukin-3 (IL-3)dependent Ba/F3 cells through retrovirus-mediated gene transduction and subsequent drug resistance selection.

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 9

major signaling molecule downstream of receptor tyrosine kinases.59-61 Surprisingly, western blot analysis revealed that the AP21967-dependent heterodimer formation induced remarkable SFK phosphorylation only in the cells coexpressing H5R and H5K constructs (H5R+H5K). In contrast, the (G4S)3-linked chimeras (H3R+H3K) and no-helper control (R+K) induced far lower SFK phosphorylation levels than H5R+H5K. Of note, helper dimerization did not induce the SFK activation in the cells expressing the single chimeras (H3R, H3K, H5R, and H5K). Taken together, the results indicate that the helper interaction boosts the interaction between the R and K chimeras to form stable complexes that are competent for signaling, but does not cause any background signaling without the interaction between the proteins of interest. To demonstrate that the helper interaction assists the activation of signaling, western blotting was performed to compare SFK phosphorylation levels in the cells with and without helper ligand stimulation. As a result, H5R+H5K remarkably phosphorylated SFK only in the presence of the helper ligand (Figure 2b). On the other hand, H3R+H3K and R+K exhibited similar levels of phosphorylation to the control expressing no chimera (-).

Figure 3. The helper interaction enhances the sensitivity of proliferation signals. The chimeras expressed in the Ba/F3 transductants are shown as the abbreviated names, and (-) indicates the parental Ba/F3 cell. (a) Proliferation assay for analyzing AP21967 dependency of cell growth. Cells expressing chimera(s) were cultured with the helper ligand at 100 nM and AP21967 at indicated concentrations (0–50 nM). The initial cell density was 1×105 cells/mL. The viable cell densities on day 3 are indicated as mean ± SD (n=3, biological replicates). (b) Proliferation assay for analyzing helper ligand dependency of cell growth. Cells expressing chimeras (H5R+H5K and R+K) were cultured with (+) /without (-) 50 nM AP21967 and the helper ligand at indicated concentrations (0–100 nM). The initial cell density was 1×105 cells/mL. The viable cell densities on day 3 are indicated as mean ± SD (n=3, biological replicates). To examine signaling events induced by the chimeras, the IL-3-cultured transductants were washed, cultured in an IL-3-deprived medium, and stimulated with or without AP21967 (50 nM) in the presence of the helper ligand (100 nM) abundant enough for helper dimerization. The cell lysates were prepared for subsequent western blot analyses. We first evaluated the expression levels of the chimeras in the transductants using the antibodies against the epitope tags (Figure 2a, blots with V5 and HA tags).

The helper interaction enhances the sensitivity of proliferation signals To assess whether the helper interaction enhances PPIdependent cell proliferation, the cells were cultured in the IL-3-deprived medium with a constant helper ligand concentration (100 nM) and with serial AP21967 concentrations (0.05–50 nM) for controlling the content of FRBT2098L–FKBP dimers. We judged that the transductants “proliferated” in the case that the final cell density after 3-days culture exceeded the initial cell density (1.0 × 105 cells/ml). Consequently, AP21967-dependent cell growth was confirmed for the cells expressing R+K, H3R+H3K, and H5R+H5K (Figure 3a). On the other hand, no proliferation was observed for the cells expressing the single chimeras (H3R, H3K, H5R, and H5K) and no-chimera control (-). In particular, the cells expressing H5R+H5K proliferated even at 5 nM AP21967, at which no-helper control (R+K) hardly proliferated. Furthermore, the maximal cell density induced by H5R+H5K reached up to ~5fold compared to that induced by R+K. These results demonstrate that the helper interaction enhances the formation of the signaling-competent chimera complexes even under a low concentration of AP21967 which mimics weak interactions between proteins of interest. Thus, the sensitivity of KIPPIS was improved by introducing the helper module.

While all R chimeras have equivalent expression levels, the addition of the helper module considerably decreased the expression levels of the K chimeras. To examine the activation of the c-kit intracellular domain, we focused on Src family kinases (SFK), which are well-known as a

ACS Paragon Plus Environment

Page 5 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry clear helper ligand-dependent cell proliferation. Remarkably, the maximal cell density induced by H5R+H5K increased up to ~6-fold as much as that induced by R+K. These results suggest that helper interaction supports the activation and subsequent signal transduction of the c-kit intracellular domain.

a

b Helper ligand

Figure 4. The construction of p53 pep- and MDM2-fused chimeras. (a) Illustration for p53 pep–MDM2 complex (PDB:1YCR). (i) p53 pep (rainbow) fits into a pocket of MDM2 (gray). (ii) The three residues (F19, W23, and L26) of p53 pep fill in a cavity of MDM2. The figure shows a cross section sliced by a red broken line (ii) in the panel (i). (iii) All three residues are periodically coded and positioned to form a line on the helix of p53 pep. Figure shows a cross section sliced by a red broken line (iii) in the panel (i). (b) The constructs of chimeras with the helper module (FKBPF36V), protein of interest (p53 pep or MDM2N), and c-kit intracellular domain (c-kit ICD). In the presence of the helper ligand, the chimeras form stable complexes. (c) The constructs of chimeras without the helper module and with alanine-mutated variants of p53 pep. Without support of the helper interaction or p53–MDM2 interaction, the c-kit intracellular domain dimer would be too unstable to trigger a strong growth signal. The abbreviated chimera names are shown in red.

To examine whether the helper ligand-inducible dimerization of the helper module affects cell proliferation levels, cells were cultured with serial concentrations of the helper ligand (0.01–100 nM) with the AP21967 concentration fixed at 50 nM to induce stable FRBT2098L–FKBP complex formation. Under this condition, cell proliferation was expected to be observed even in the constructs without the helper module. Indeed, the cells expressing nohelper constructs (R+K) proliferated, with the proliferation levels not depending on the concentration of the helper ligand (Figure 3b). In contrast, H5R+H5K showed

Figure 5. The helper interaction enables activation of c-kit intracellular domain in p53 pep- and MDM2-fused chimeras. The chimeras expressed in the Ba/F3 transductants are shown as the abbreviated names. (-) indicates the parental Ba/F3 cell. (a) Western blotting for detecting p53 pep–MDM2 interaction-dependent tyrosine phosphorylation of SFK. Cells were stimulated for 30 min with 100 nM of the helper ligand and lysed. Western blotting was performed with the following antibodies; anti-phospho tyrosine of SFK (p-SFK), antiSFK, anti-V5 tag to detect the expression level of P chimeras, anti-HA tag to detect the expression level of M chimeras, and anti-GAPDH as a loading control. (b) Western blotting for analyzing helper ligand dependent phosphorylation of SFKs. Cells were stimulated for 30 min with/without 100 nM of the helper ligand and lysed. Western blotting was performed with antibodies; anti-p-SFK, anti-SFK, and anti-GAPDH as a loading control.

The helper interaction enables activation of the ckit intracellular domain in p53 pep- and MDM2-fused chimeras p53, a superior guardian of the genome, induces apoptotic cell death and growth arrest in response to severe DNA damage.62,63 MDM2, a negative regulator of p53, binds to the transactivation domain of p53 to form a stacked complex.62,63 Here, we utilized 15–29 amino acid residues located in the transactivation domain of p53 (p53 pep) and 21–113 amino acid residues containing the p53binding N-terminal domain of MDM2 (MDM2N) as proteins of interest. It was previously reported that the p53– MDM2 interaction relies on three key amino acid residues (F19, W23, and L26) in p53 pep (Figure 4a).64 Thus, these

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

three residues were mutated with alanine residues to make a non-interacting control. p53 pep and MDM2N were genetically incorporated into the R and K chimeras instead of FRBT2098L and FKBP, resulting in the chimeras containing p53 pep (P chimeras; V5-tagged), alaninemutated p53 pep (AP chimeras; V5-tagged), and MDM2N (M chimeras; HA-tagged) (Figure 4b and 4c). Again, the chimera names are designated as HnP, HnAP, or HnM (Supplementary Table 2). These chimeras were stably expressed in Ba/F3 cells through retrovirus-mediated gene transduction and subsequent drug resistance selection. To investigate signaling events in the transductants, the cells were precultured devoid of IL-3 and stimulated in the presence of 100 nM of the helper ligand, followed by preparation of cell lysates for western blot analyses. We explored the expression level of each construct in the transductants using anti-V5 tag and anti-HA tag antibodies (Figure 5a). The expression levels of the chimeras were almost equivalent for the different constructs in the P or M chimeras. Next, SFK phosphorylation was detected to analyze the activation of the c-kit intracellular domain fused with the C-terminus of p53 pep and MDM2N. While H5P+H5M ((G4S)5-linked constructs) induced SFK phosphorylation at a high level, H3P+H3M ((G4S)3-linked constructs), P+M, and the alanine-mutated variants (H3AP+H3M and H5AP+H5M) induced SFK phosphorylation comparable to no-chimera control (-). Again, the single chimeras (H3P, H3M, H5P, H5M, P, and M) did not induce the activation of SFK. To investigate the helper ligand dependency of SFK phosphorylation, the cells were stimulated in the presence or absence of the helper ligand, and their lysates were analyzed by western blotting. Similarly to FRBT2098L+FKBP constructs, helper ligand-dependent phosphorylation was detected in (G4S)5-linked constructs (H5P+H5M), but not in the non-interacting control (H5AP+H5M) (Figure 5b). These data strongly suggest that the dimer formation of the helper modules stabilized the p53 pep–MDM2N interaction and strengthened the ckit intracellular domain signaling.

Detection of p53 pep–MDM2N interaction based on cell proliferation We finally examined whether the p53 pep–MDM2N interaction could be detected based on cell proliferation. To evaluate the contribution of the helper interaction, the helper ligand was added at various concentrations (0.01– 100 nM). Consequently, H3P+H3M and especially H5P+H5M showed helper ligand-dependent cell proliferation (Figure 6). On the contrary, no proliferation was observed for the cells expressing the single chimeras (H3P, H3AP, H3M, H5P, H5AP, and H5M), no-helper controls (P, M, and P+M), and non-interacting controls (H3AP+H3M and H5AP+H5M). Consistent with the previous report,52 constructs without the helper module (P+M) were not able to detect the interaction between p53 pep and

Page 6 of 9

MDM2N. Notably, H5P+H5M successfully detected the relatively weak interaction of p53 pep–MDM2N, which cannot be detected in our previous helper-less format.52 These results demonstrate that the helper dimerization stabilized the p53 pep–MDM2N complex to boost the signaling of the chimeras.

Figure 6. Detection of the p53 pep–MDM2N interaction based on cell proliferation. The chimeras expressed in the Ba/F3 transductants are shown as the abbreviated names. Cells expressing chimeras were cultured with the helper ligand at indicated concentrations (0–100 nM) and without IL-3. 5 The initial cell density was 1×10 cells/mL. The viable cell densities on day 3 are indicated as mean ± SD (n=3, biological replicates).

DISCUSSION In this study, we demonstrated a chemically inducible helper module facilitates the detection of intracellular PPIs based on the KIPPIS system. The inclusion of the helper module enabled us to detect the p53 pep–MDM2 interaction that was not able to be detected without the helper module. It should be noted that sensitivity is easily configured in this system. The SFK phosphorylation and cell proliferation assays suggest that the helper interaction assists the association of the proteins of interest and promote the activation of the c-kit intracellular domain. In fact, using FRBT2098L and FKBP as model interacting proteins, the sensitivity of KIPPIS was improved when compared to our previously developed helper-less KIPPIS. Moreover, the p53–MDM2 interaction, which was not detected at all in the helperless KIPPIS, was successfully detected based on cell proliferation, which clearly indicates the superior feature of the helper module. The cells expressing H5R+H5K proliferated only in helper ligand concentrations of 1–100 nM, which exceeds the Kd value of the interaction between the helper and its ligand (0.094 nM) (Figure 3b).57 This may be due to the low expression level of the H5K chimera, whose stability was lowered presumably by fusing two similar FKBPderived domains. Surprisingly, H5R+H5K induced faster cell proliferation than the no-helper control (R+K) in spite of the low expression level of the H5K chimera. In

ACS Paragon Plus Environment

Page 7 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the case of H5P+H5M whose expression levels were equivalent to the no-helper control (P+M), the p53 pep– MDM2 interaction was detected even at 0.01 nM of the helper ligand, which is far lower than the Kd value of the helper module–helper ligand interaction. Therefore, the helper interaction has beneficial effects on the sensitivity of our PPI detection system. There was a major difference in signaling activities between the chimeras with the (G4S)3 linker (H3R+H3K and H3P+H3M) and those with the (G4S)5 linker (H5R+H5K and H5P+H5M). We speculate that steric hindrance is one of the factors which explain why the length of the flexible linker connecting the helper module with the protein of interest greatly affected the activation of the c-kit intracellular domain. The crystallographic analyses revealed that the distance between the N-termini of FRB and FKBP is 47.1 Å in the FRB–rapamycin–FKBP complex (PDB: 2RSE), and the distance between the N-termini of p53 pep and MDM2N is 29.0 Å (PDB: 1YCR) (Supplementary Figure 1). These distances were different from the distance between the C-termini of two helper modules in the presence of helper ligand (33.4 Å) (PDB: 1BL4). In addition, the N-termini of both protein-of-interest pairs are directed to opposite sides. To enable the helper modules and proteins of interest to form complex without steric effects, the linkers connecting the helper module to the protein of interest require the sufficient length and flexibility. Interestingly, while the helper interaction strongly supported proliferation of the cells with the interaction between proteins of interest, cell proliferation was not observed in the single chimera-expressing cells in which there is no interaction between proteins of interest in principle. This result indicates that the region flanked by FKBPF36V and the c-kit intracellular domain (i.e. the (G4S)3 or (G4S)5 linker, the protein of interest, and the (G4S)2 linker) prevents the c-kit intracellular domains from taking active conformation even when the helper module is dimerized by the helper ligand. This signaling off-state can be readily switched on with the use of interacting proteins of interest, which brings the N-termini of the ckit intracellular domains into proximity to yield an active dimer. Proteins of interest used in this study have a common structural feature that their C-termini are directed to the same side in the complex. Therefore, the relatively short (G4S)2 linker was enough for the c-kit intracellular domains to take active dimeric conformation. If we use proteins of interest with the C-termini that are directed to the opposite side, optimization of the linker between the proteins of interest and c-kit intracellular domain will be needed for strong signal transduction. To date, investigators have developed PPI detection systems based on cell signaling such as MYTH,44-46 MaMTH,47 KISS,48 and MAPPIT.49-51 While these are all qualified as mammalian PPI detection systems, the processes are untunable because the readout is constitutively represented after the expression of proteins of interest. In contrast, the helper-assisted KIPPIS we reported here is featured not only by the improved sensitivity with the low

background, but also by the capability of tuning sensitivity using a chemical ligand. It means that this system allows us to set various affinity thresholds by changing the concentration of the helper ligand. Thus, the helperassisted KIPPIS could be applied to screen affinitymatured variants that specifically bind to a target protein in mammalian cells.

ASSOCIATED CONTENT Supporting Information. Further information for experimental procedures is described in Supporting Information; a primers list (Supplementary Table 1), correspondence between plasmid names and construct names (Supplementary Table 2), detailed protocols for assays (western blotting and a cell proliferation assay), and illustration for the crystal structures of proteins (Supplementary Figure 1).

AUTHOR INFORMATION Corresponding Author * [email protected]

ACKNOWLEDGMENT We are grateful to T. Miura (The University of Tokyo) for designing constructs. D.K. was supported by Graduate Program for Leaders in Life Innovation (GPLLI) in The University of Tokyo. This work was supported by JSPS KAKENHI Grant Number 15H04190 (to M.K.).

REFERENCES (1) Pawson, T.; Nash, P. Genes Dev. 2000, 14, 1027-1047. (2) Lienert, F.; Lohmueller, J. J.; Garg, A.; Silver, P. A. Nat. Rev. Mol. Cell Biol. 2014, 15, 95-107. (3) Kolch, W.; Pitt, A. Nat. Rev. Cancer 2010, 10, 618-629. (4) Ronnstrand, L. Cell. Mol. Life Sci. 2004, 61, 2535-2548. (5) Roskoski, R., Jr. Biochem. Biophys. Res. Commun. 2005, 337, 1-13. (6) Sattler, M.; Salgia, R. Leuk. Res. 2004, 28 Suppl 1, S11-20. (7) Zhou, M.; Li, Q.; Wang, R. ChemMedChem 2016, 2016, 201500495. (8) Vernon-Wilson, E. F.; Kee, W. J.; Willis, A. C.; Barclay, A. N.; Simmons, D. L.; Brown, M. H. Eur. J. Immunol. 2000, 30, 21302137. (9) Rogstam, A.; Linse, S.; Lindqvist, A.; James, P.; Wagner, L.; Berggard, T. Biochem. J. 2007, 401, 353-363. (10) Oesterreicher, S.; Blum, W. F.; Schmidt, B.; Braulke, T.; Kubler, B. J. Biol. Chem. 2005, 280, 9994-10000. (11) Gingras, A. C.; Gstaiger, M.; Raught, B.; Aebersold, R. Nat. Rev. Mol. Cell Biol. 2007, 8, 645-654. (12) Watanabe, M.; Heddle, J. G.; Kikuchi, K.; Unzai, S.; Akashi, S.; Park, S. Y.; Tame, J. R. Proc. Natl. Acad. Sci. USA 2009, 106, 2176-2181. (13) Rinner, O.; Mueller, L. N.; Hubalek, M.; Muller, M.; Gstaiger, M.; Aebersold, R. Nat. Biotechnol. 2007, 25, 345-352. (14) Weber, P. C.; Salemme, F. R. Curr. Opin. Struct. Biol. 2003, 13, 115-121. (15) Rudolph, M. G.; Linnemann, T.; Grunewald, P.; Wittinghofer, A.; Vetter, I. R.; Herrmann, C. J. Biol. Chem. 2001, 276, 23914-23921. (16) Jung, H. I.; Bowden, S. J.; Cooper, A.; Perham, R. N. Protein Sci. 2002, 11, 1091-1100. (17) Bonvin, A. M.; Boelens, R.; Kaptein, R. Curr. Opin. Chem. Biol. 2005, 9, 501-508.

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(18) O'Connell, M. R.; Gamsjaeger, R.; Mackay, J. P. Proteomics 2009, 9, 5224-5232. (19) Vaynberg, J.; Fukuda, T.; Chen, K.; Vinogradova, O.; Velyvis, A.; Tu, Y.; Ng, L.; Wu, C.; Qin, J. Mol. Cell 2005, 17, 513-523. (20) Sun, Z. Y.; Kim, S. T.; Kim, I. C.; Fahmy, A.; Reinherz, E. L.; Wagner, G. Proc. Natl. Acad. Sci. USA 2004, 101, 16867-16872. (21) Arunkumar, A. I.; Klimovich, V.; Jiang, X.; Ott, R. D.; Mizoue, L.; Fanning, E.; Chazin, W. J. Nat. Struct. Mol. Biol. 2005, 12, 332-339. (22) Vaynberg, J.; Qin, J. Trends Biotechnol. 2006, 24, 22-27. (23) Ellis, R. J. Trends Biochem. Sci. 2001, 26, 597-604. (24) Ellis, R. J.; Minton, A. P. Nature 2003, 425, 27-28. (25) Petschnigg, J.; Snider, J.; Stagljar, I. Curr. Opin. Biotechnol. 2011, 22, 50-58. (26) Lam, M. H.; Stagljar, I. Proteomics 2012, 12, 1519-1526. (27) Yao, Z.; Petschnigg, J.; Ketteler, R.; Stagljar, I. Nat. Chem. Biol. 2015, 11, 387-397. (28) Wehr, M. C.; Rossner, M. J. Drug Discov. Today 2016, 21, 415-429. (29) Piston, D. W.; Kremers, G. J. Trends Biochem. Sci. 2007, 32, 407-414. (30) Ciruela, F. Curr. Opin. Biotechnol. 2008, 19, 338-343. (31) Padilla-Parra, S.; Tramier, M. BioEssays 2012, 34, 369-376. (32) Pfleger, K. D.; Eidne, K. A. Nat. Methods 2006, 3, 165-174. (33) De, A.; Jasani, A.; Arora, R.; Gambhir, S. S. Front. Endocrinol. 2013, 4, 131. (34) Kerppola, T. K. Annu. Rev. Biophys. 2008, 37, 465-487. (35) Kerppola, T. K. Chem. Soc. Rev. 2009, 38, 2876-2886. (36) Ozawa, T.; Kaihara, A.; Sato, M.; Tachihara, K.; Umezawa, Y. Anal. Chem. 2001, 73, 2516-2521. (37) Wehr, M. C.; Laage, R.; Bolz, U.; Fischer, T. M.; Grunewald, S.; Scheek, S.; Bach, A.; Nave, K. A.; Rossner, M. J. Nat. Methods 2006, 3, 985-993. (38) Stefan, E.; Aquin, S.; Berger, N.; Landry, C. R.; Nyfeler, B.; Bouvier, M.; Michnick, S. W. Proc. Natl. Acad. Sci. USA 2007, 104, 16916-16921. (39) Shekhawat, S. S.; Ghosh, I. Curr. Opin. Chem. Biol. 2011, 15, 789-797. (40) Tchekanda, E.; Sivanesan, D.; Michnick, S. W. Nat. Methods 2014, 11, 641-644. (41) Luo, Y.; Batalao, A.; Zhou, H.; Zhu, L. BioTechniques 1997, 22, 350-352. (42) Rajagopala, S. V.; Sikorski, P.; Caufield, J. H.; Tovchigrechko, A.; Uetz, P. Methods 2012, 58, 392-399. (43) Snider, J.; Kittanakom, S.; Damjanovic, D.; Curak, J.; Wong, V.; Stagljar, I. Nat. Protoc. 2010, 5, 1281-1293. (44) Stagljar, I.; Korostensky, C.; Johnsson, N.; te Heesen, S. Proc. Natl. Acad. Sci. USA 1998, 95, 5187-5192. (45) Gisler, S. M.; Kittanakom, S.; Fuster, D.; Wong, V.; Bertic, M.; Radanovic, T.; Hall, R. A.; Murer, H.; Biber, J.; Markovich, D.; Moe, O. W.; Stagljar, I. Mol. Cell Proteomics 2008, 7, 1362-1377. (46) Deribe, Y. L.; Wild, P.; Chandrashaker, A.; Curak, J.; Schmidt, M. H.; Kalaidzidis, Y.; Milutinovic, N.; Kratchmarova, I.; Buerkle, L.; Fetchko, M. J.; Schmidt, P.; Kittanakom, S.; Brown, K. R.; Jurisica, I.; Blagoev, B.; Zerial, M.; Stagljar, I.; Dikic, I. Sci. Signal. 2009, 2, ra84. (47) Petschnigg, J.; Groisman, B.; Kotlyar, M.; Taipale, M.; Zheng, Y.; Kurat, C. F.; Sayad, A.; Sierra, J. R.; Mattiazzi Usaj, M.; Snider, J.; Nachman, A.; Krykbaeva, I.; Tsao, M. S.; Moffat, J.; Pawson, T.; Lindquist, S.; Jurisica, I.; Stagljar, I. Nat. Methods 2014, 11, 585592. (48) Lievens, S.; Gerlo, S.; Lemmens, I.; De Clercq, D. J.; Risseeuw, M. D.; Vanderroost, N.; De Smet, A. S.; Ruyssinck, E.; Chevet, E.; Van Calenbergh, S.; Tavernier, J. Mol. Cell Proteomics 2014, 13, 3332-3342.

Page 8 of 9

(49) Lemmens, I.; Eyckerman, S.; Zabeau, L.; Catteeuw, D.; Vertenten, E.; Verschueren, K.; Huylebroeck, D.; Vandekerckhove, J.; Tavernier, J. Nucleic Acids Res. 2003, 31, e75. (50) Lemmens, I.; Lievens, S.; Eyckerman, S.; Tavernier, J. Nat. Protoc. 2006, 1, 92-97. (51) Lievens, S.; Peelman, F.; De Bosscher, K.; Lemmens, I.; Tavernier, J. Cytokine Growth Factor Rev. 2011, 22, 321-329. (52) Mabe, S.; Nagamune, T.; Kawahara, M. Sci. Rep. 2014, 4, 6127. (53) Pazgier, M.; Liu, M.; Zou, G.; Yuan, W.; Li, C.; Li, J.; Monbo, J.; Zella, D.; Tarasov, S. G.; Lu, W. Proc. Natl. Acad. Sci. USA 2009, 106, 4665-4670. (54) Grunberg, R.; Burnier, J. V.; Ferrar, T.; Beltran-Sastre, V.; Stricher, F.; van der Sloot, A. M.; Garcia-Olivas, R.; Mallabiabarrena, A.; Sanjuan, X.; Zimmermann, T.; Serrano, L. Nat. Methods 2013, 10, 1021-1027. (55) Putyrski, M.; Schultz, C. FEBS Lett. 2012, 586, 2097-2105. (56) Grunberg, R.; Ferrar, T. S.; van der Sloot, A. M.; Constante, M.; Serrano, L. Nucleic Acids Res. 2010, 38, 2645-2662. (57) Clackson, T.; Yang, W.; Rozamus, L. W.; Hatada, M.; Amara, J. F.; Rollins, C. T.; Stevenson, L. F.; Magari, S. R.; Wood, S. A.; Courage, N. L.; Lu, X.; Cerasoli, F., Jr.; Gilman, M.; Holt, D. A. Proc. Natl. Acad. Sci. USA 1998, 95, 10437-10442. (58) Gabbianelli, M.; Testa, U. Mediterr. J. Hematol. Infect. Dis. 2009, 1, e2009009. (59) Abram, C. L.; Courtneidge, S. A. Exp. Cell. Res. 2000, 254, 113. (60) Young, M. A.; Gonfloni, S.; Superti-Furga, G.; Roux, B.; Kuriyan, J. Cell 2001, 105, 115-126. (61) Roskoski, R., Jr. Pharmacol. Res. 2015, 94, 9-25. (62) Chene, P. Nat. Rev. Cancer 2003, 3, 102-109. (63) Joerger, A. C.; Fersht, A. R. Cold Spring Harb. Perspect. Biol. 2010, 2, a000919. (64) Lu, F.; Chi, S. W.; Kim, D. H.; Han, K. H.; Kuntz, I. D.; Guy, R. K. J. Comb. Chem. 2006, 8, 315-325.

ACS Paragon Plus Environment

Page 9 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Insert Table of Contents artwork here

Helper module

MDM2N

p53 pep

Helper ligand

c-kit ICD

ACS Paragon Plus Environment

9