Effective identification of Akt interacting proteins by two-step chemical crosslinking, co-immunoprecipitation and mass spectrometry

Abstract

Akt is a critical protein for cell survival and known to interact with various proteins. However, Akt binding partners that modulate or regulate Akt activation have not been fully elucidated. Identification of Akt-interacting proteins has been customarily achieved by co-immunoprecipitation combined with western blot and/or MS analysis. An intrinsic problem of the method is loss of interacting proteins during procedures to remove non-specific proteins. Moreover, antibody contamination often interferes with the detection of less abundant proteins. Here, we developed a novel two-step chemical crosslinking strategy to overcome these problems which resulted in a dramatic improvement in identifying Akt interacting partners. Akt antibody was first immobilized on protein A/G beads using disuccinimidyl suberate and allowed to bind to cellular Akt along with its interacting proteins. Subsequently, dithiobis[succinimidylpropionate], a cleavable crosslinker, was introduced to produce stable complexes between Akt and binding partners prior to the SDS-PAGE and nanoLC-MS/MS analysis. This approach enabled identification of ten Akt partners from cell lysates containing as low as 1.5 mg proteins, including two new potential Akt interacting partners. None of these but one protein was detectable without crosslinking procedures. The present method provides a sensitive and effective tool to probe Akt-interacting proteins. This strategy should also prove useful for other protein interactions, particularly those involving less abundant or weakly associating partners.

Figure 1. Schematic diagram of a two-step crosslinking approach for identification of Akt binding partners.

Akt antibody was immobilized on protein A/G beads via DSS (succinimidyl suberate) cross-linking. Following the incubation of the cell lysate with the immobilized Akt antibody, DSP (dithiobis[succinimidylpropionate]), a cleavable crosslinker, was used to produce stable complexes between Akt and its binding partners. The co-immunoprecipitated complexes were eluted with Laemmli buffer containing 5% β-mercaptoethanol. The liberated proteins were subjected to SDS-PAGE, tryptic digestion, and MS-based analysis followed by western blot validation.

Figure 2. Western blot analysis of Akt and ERK1/2 monitored during co-IP procedures.

a. Chemical structures of the crosslinkers used in the study. b. Antibody contamination was eliminated by DSS-crosslinking of Akt antibody to protein A/G agarose beads. c. ERK1/2 co-immunoprecipitation with Akt was significantly improved by DSP-crosslinking. d. DSS-crosslinking did not alter antibody-antigen binding. nc, negative control using IgG instead of Akt antibody at a similar protein level.

Figure 3. Representative MS/MS spectra of Akt and ERK1/2 peptides obtained from the co-IP complex.

MS/MS analysis of the doubly charged ion at m/z 912.46 or 798.90 originated from the peptide segment Y–K of Akt (a), or A[193–205]K of ERK1/2 (b), respectively. The sequence of the peptide was assigned with a single letter abbreviation based on the fragment ions observed for the peptide segments. N-terminal b ions and C-terminal y ions resulting from the amide bond cleavage are labeled. * denotes the addition of 145.0198 Dalton due DSP-crosslinking.

Figure 4. Identification of Nudc and TFII-I as potential Akt binding partners.

Representative MS/MS spectra of Nudc (a) and TFII-I (b) peptides identified from the co-IP products. The sequence of peptides was assigned with a single letter abbreviation based on the fragment ions observed for the peptide segments. N-terminal b ions and C-terminal y ions resulting from the amide bond cleavage are labeled. * denotes the addition of 145.0198 Dalton due DSP-crosslinking. Western blotting indicated that Nudc co-immunoprecipitated with Akt in Neuro 2A cells (a). nc, negative control using IgG instead of Akt antibody at a similar protein level.

Figure 5. IGF-dependent interaction of Akt with GSK-3β revealed by the crosslinking/co-IP approach.

a, Western blot analysis indicating that active Akt is associated with GSK-3β more than inactive Akt. The active Akt status was indicated by phosphorylation of T308. b, Western blot analysis of the co-IP complex with or without DSP-crosslinking following the incubation of the cell lysate with Akt antibody immobilized on the beads by DSS. GSK-3β was detected in either control or IGF-stimulated samples only when DSP-crosslinking was used. c, Representative MS/MS spectrum of a peptide corresponding to GSK-3β identified only with DSP-crosslinking approach. The sequence of the peptide was assigned with a single letter abbreviation based on the fragment ions observed for the peptide segments. N-terminal b ions and C-terminal y ions resulting from the amide bond cleavage are labeled. Con, control sample without IGF stimulation. IGF, IGF-stimulated samples obtained by treating Neuro 2A cells with IGF at 10 ng/mL for 10 min. nc, negative control using IgG instead of Akt antibody at a similar protein level. * denotes the addition of 145.0198 Dalton due DSP-crosslinking.

Figure 6. IGF-dependent interaction of Akt with EBP1.

a, Western blot analysis indicating that active Akt is associated with EBP1 more than inactive Akt. The active Akt status was indicated by phosphorylation of T308. b, Western blot analysis of the co-IP complex with or without DSP-crosslinking following the incubation of the IGF-stimulated cell lysate with Akt antibody immobilized on the beads by DSS. EBP1 was negligibly detected in the IP product without the use of DSP-crosslinking. c, Representative MS/MS spectrum of a peptide corresponding to EBP1 identified only with DSP-crosslinking approach. The sequence of the peptide was assigned with a single letter abbreviation based on the fragment ions observed for the peptide segments. N-terminal b ions and C-terminal y ions resulting from the amide bond cleavage are labeled. Con, control sample without IGF stimulation. IGF, IGF-stimulated samples obtained by treating Neuro 2A cells with IGF at 10 ng/mL for 10 min. nc, negative control using IgG instead of Akt antibody at a similar protein level. * denotes the addition of 145.0198 Dalton due DSP-crosslinking.

Figure 7. IGF-independent interaction of Akt with ERK1/2 or Nudc revealed by western blot analysis.

The active Akt status was indicated by phosphorylation of T308. Con, control sample without IGF stimulation. IGF-stimulated samples were obtained by treating Neuro 2A cells with IGF at 10 ng/mL for 10 min. nc, negative control using IgG instead of Akt antibody at a similar protein level.