Coordinate Phosphorylation of Multiple Residues on Single AKT1 and AKT2 Molecules

ORIGINAL ARTICLE Coordinate phosphorylation of multiple residues on single AKT1 and AKT2 molecules

H Guo1,6, M Gao1,6,YLu1, J Liang1, PL Lorenzi2, S Bai3, DH Hawke4,JLi1, T Dogruluk5, KL Scott5, E Jonasch3, GB Mills1 and Z Ding1

Aberrant AKT activation is prevalent across multiple human cancer lineages providing an important new target for therapy. Twenty- two independent phosphorylation sites have been identified on specific AKT isoforms likely contributing to differential isoform regulation. However, the mechanisms regulating phosphorylation of individual AKT isoform molecules have not been elucidated because of the lack of robust approaches able to assess phosphorylation of multiple sites on a single AKT molecule. Using a nanofluidic proteomic immunoassay (NIA), consisting of isoelectric focusing followed by sensitive chemiluminescence detection, we demonstrate that under basal and ligand-induced conditions that the pattern of phosphorylation events is markedly different between AKT1 and AKT2. Indeed, there are at least 12 AKT1 peaks and at least 5 AKT2 peaks consistent with complex combinations of phosphorylation of different sites on individual AKT molecules. Following insulin stimulation, AKT1 was phosphorylated at Thr308 in the T-loop and Ser473 in the hydrophobic domain. In contrast, AKT2 was only phosphorylated at the equivalent sites (Thr309 and Ser474) at low levels. Further, Thr308 and Ser473 phosphorylation occurred predominantly on the same AKT1 molecules, whereas Thr309 and Ser474 were phosphorylated primarily on different AKT2 molecules. Although basal AKT2 phosphorylation was sensitive to inhibition of phosphatidylinositol 3-kinase (PI3K), basal AKT1 phosphorylation was essentially resistant. PI3K inhibition decreased pThr451 on AKT2 but not pThr450 on AKT1. Thus, NIA technology provides an ability to characterize coordinate phosphorylation of individual AKT molecules providing important information about AKT isoform-specific phosphorylation, which is required for optimal development and implementation of drugs targeting aberrant AKT activation.

Oncogene (2014) 33, 3463–3472; doi:10.1038/onc.2013.301; published online 5 August 2013 Keywords: AKT; phosphorylation; NIA

INTRODUCTION between its pleckstrin homology (PH) domain and PtdIns(3,4,5)P3, The serine/threonine kinase AKT (also known as protein kinase B, where AKT is phosphorylated at two critical residues, Thr308/309 PKB), comprising a group of three isoforms (AKT1, AKT2 and AKT3), in the activation T-loop and Ser473/474 in the hydrophobic has critical roles in many aspects of cancer pathophysiology domain of AKT1/2 (unless designated otherwise, phosphorylation including cell survival, growth, metabolism and metastasis.1–3 AKT sites are based on the AKT1 amino acid sequence). 3-Phosphoi- is a bona fide oncogene that is frequently activated in cancer nositide-dependent protein kinase 1 (PDK1) (ref. 7) phosphorylates 8 through a variety of mechanisms including amplification of AKT at Thr308, and mTORC2 as well as other potential PDK2 9 growth factor receptors (e.g., HER2/neu, epidermal growth factor phosphorylate AKT at Ser473. Activated AKT then translocates receptor), amplification or mutation of phosphatidylinositol 3- from the cell membrane to other cell compartments to kinase (PI3K), amplification or mutation of AKT isoforms, and phosphorylate its downstream substrates transducing membrane inactivation of phosphatase and tensin homolog or inositol signals to appropriate functional outcomes.10 Phosphorylation of polyphosphate-phosphatase type II (INPP4B).3 Different AKT Thr308 and Ser473 has been proposed to be required for full isoforms appear to mediate critical non-redundant functions in activation of AKT kinase activity.11 However, whether Thr308 and cancer pathophysiology.4–6 For example, AKT1 has been Ser473 phosphorylation is sufficient for full activity or the multiple implicated as a major contributor to tumor initiation, whereas other phosphorylation sites in AKT isoforms are required for AKT2 appears to primarily increase tumor metastasis.5,6 Therefore, processive phosphorylation or modulate the stability, substrate elucidation of the mechanisms regulating AKT activation, especi- access or activity of AKT has not been elucidated.12–14 Further ally AKT isoform-specific activation, will facilitate therapeutic selective phosphorylation of Thr308 and Ser473 alters the approaches to targeting AKT signaling. substrate selectivity of AKT. Thus, an improved understanding of In the canonical AKT activation model, growth factors or other the role of the multiple phosphorylation sites in AKT is required to stimuli activate class I PI3K at the cell membrane to phosphorylate fully elicit the functional regulation of AKT. PtdIns(4,5)P2 to form PtdIns(3,4,5)P3 on the inner cell membrane. In addition to Thr308 and Ser473, currently 20 other residues AKT is then recruited to the cell membrane through interaction of AKT1 have been experimentally validated as sites for

1Department of Systems Biology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 2Department of Bioinformatics and Computational Biology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 3Department of Genitourinary Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 4Department of Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA and 5Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA. Correspondence: Dr GB Mills or Dr Z Ding, Department of Systems Biology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA. E-mail: [email protected] or [email protected] 6These authors contributed equally to this work. Received 22 February 2013; revised 4 June 2013; accepted 11 June 2013; published online 5 August 2013 Isoform-specific phosphorylation of AKT H Guo et al 3464 phosphorylation using mass spectrometry or site-specific dominant and one minor peak (Figure 1b). The AKT peaks in approaches, including eight serine residues (122, 124, 126, 129, HCT116 wt could be reconstituted, both in pattern and relative 137, 246, 475 and 477), seven threonine residues (34, 72, 146, 305, magnitude, by combining peaks from AKT2 À / À and AKT1 À / À 312, 450 and 479) and five tyrosine residues (176, 315, 326, 437 cells, validating the use of knockout cells to characterize AKT and 474; http://www.phosphosite.org).12–15 Similarly, AKT2 and isoform-specific phosphorylation. The AKT1 and AKT2 isoform- AKT3 have 22 and 18 validated phosphorylation sites, respectively. specific antibodies also recognized the peaks predicted by the Additional AKT isoform-specific phosphorylation sites may remain studies of HCT116 knockout cells (Supplementary Figures S1A and to be identified. The regulation and importance of phosphoryl- B). The only overlap between the AKT1 and AKT2 protein ation of sites other than Thr308 and Ser473 is only beginning to isoelectronic point (pI) pattern was a peak located at P5.75. be elucidated. For example, phosphorylation at Thr305, Thr312 On the basis of the knockout cell lines and isoform-specific and Tyr474 was shown to contribute to optimal AKT activation. antibody data, it was possible to unambiguously assign the Thr72 and Ser246 have been proposed to be autophosphorylated identity of each pI peak identified with the pan (total) AKT in trans, whereas Thr34, Thr450 and Tyr176 phosphorylation antibody to a specific AKT isoform with the exception of the P5.75 appears to be mediated by upstream kinases including atypical peak, which is present in both AKT1 and AKT2. The ability to assess protein kinase C, c-Jun N-terminal kinases and Ack1.12–14 However, both AKT isoforms with a single antibody greatly aids in the regulation and function of phosphorylation sites other than quantification of relative amounts of each AKT1 or AKT2 peak. Thr308 and Ser473 has not been well studied because of the lack Therefore, unless otherwise indicated, we utilized the pan (total)- of reagents able to assess site-specific phosphorylation. The gap in AKT antibody to establish relative changes in AKT1 and AKT2 knowledge is particularly acute in terms of relative roles of migration. coordinate phosphorylation of sites on individual AKT1, AKT2 and AKT3 molecules. A newly emerging technology, nanofluidic proteomic immuno- Phosphorylation determines AKT pI values assay (NIA), has the potential to characterize protein phosphoryla- As 22 different AKT1 phosphorylation sites have been indepen- tion of multiple different sites and, in particular, coordinate dently identified by either mass spectrometry (19 out of 22; http:// phosphorylation of single AKT molecules including segregating www.phosphosite.org) or alternative approaches (13 out of 22)15 effects on different AKT isoforms.16 NIA combines isoelectric and each of first 10 phosphorylation events will change pI with an focusing of proteins with sensitive chemiluminescence detection expected change of 0.04–0.07 pH units (http://web.expasy.org/ with highly specific antibodies, providing a sensitive and compute_pi/)18 because of the charge added by phosphate, quantitative approach to analyzing protein phosphorylation.16 migration of specific isoforms of AKT on NIA should reflect the The NIA approach demonstrated marked differences in coordinate effects of phosphorylation of single and multiple sites on a single phosphorylation of multiple sites in AKT1 and AKT2 under both AKT molecule. As AKT2 had a simpler pattern of peaks than AKT1, basal and insulin-stimulated conditions. we initially used AKT1 À / À cells to characterize AKT2 isoform- specific PTM. Phosphatase treatment shifted AKT2 into a single P6.03 peak, indicating a pI value of unphosphorylated AKT2 of RESULTS 6.03, which is compatible with the predicted pI value (http:// NIA can characterize coordinate phosphorylation on individual web.expasy.org/compute_pi/).18 This suggested that under the AKT molecules conditions assessed, phosphorylation is the dominant PTM in We used wild-type (wt) and AKT2 À / À and AKT1 À / À knockout AKT2 explaining the P5.88 and P5.75 peaks observed under serum HCT116 colon cancer cells17 to characterize AKT isoform-specific starvation conditions. Insulin treatment for 30 min shifted a phosphorylation. HCT116 cells express AKT1 and AKT2 in the fraction of AKT2 into three previously undetected peaks of lower absence of AKT3.17 We tested several phospho-specific, isoform- pI of P5.62, P5.53 and P5.42 indicative of additional modifications specific and total AKT antibodies and identified a panel of of single AKT2 molecules (Figure 2a). Phosphatase sensitivity once antibodies applicable for characterization of AKT residue- and again indicated that the peak shifts induced by insulin were due to isoform-specific phosphorylation (see Materials and methods phosphorylation (Figure 2a). The overlap of peaks under basal and section). The AKT1-specific antibody used did not cross-react insulin-stimulated conditions supported the concept of discrete with AKT2 in AKT1 À / À cells, and the AKT2-specific antibody phosphorylation events accounting for the multiple AKT2 peaks used did not cross-react with AKT1 in AKT2 À / À cells, validating observed under both basal and insulin-stimulated conditions the specificity of the antibodies (Figure 1a). In contrast, a pan (Figure 2b). (total)-AKT antibody recognized both AKT isoforms with AKT1 In AKT2 À / À cells, under serum starvation conditions AKT1 was being the predominant AKT isoform. AKT2 was present at about present in four major peaks, P5.29, P5.40, P5.50 and P5.60 and five 1/3 the level of AKT1 (Figure 1a), but was readily detectable by the minor peaks, P5.22, P5.35, P5.44, P5.53 and P5.75 (Figures 1b and pan AKT and AKT2-specific antibodies. 2c). Phosphatase treatment shifted the majority of AKT1 into a We used Thr308, Ser473 and Thr450 phospho-specific anti- single peak at P5.75 (Figure 2c), indicating that the P5.75 peak in bodies to characterize residue-specific phosphorylation of AKT. In resting cells represents unphosphorylated AKT1, which is consis- HCT116 wt, AKT2 À / À , and AKT1 À / À cells, Thr308 or Ser473 tent with the predicted pI (http://web.expasy.org/compute_pi/).18 phosphorylation was undetectable under serum starvation condi- Two minor peaks P5.66 and P5.60 remained after phosphatase tions (Figure 1a). In contrast, AKT Thr450 was constitutively treatment, potentially representing phosphorylation events phosphorylated. Insulin treatment for 30 min induced substantial resistant to the phosphatase conditions used or alternatively increases in AKT1 pThr308 and pSer473, but not pThr450, in other PTM such as ubiquitination19 or acetylation.20 It may also AKT2 À / À cells, and only modest increases in AKT2 pThr309 and represent splicing variants or degradation of AKT1. Insulin pSer474 in AKT1 À / À cells potentially because of lower levels of treatment increased the number of AKT1 peaks from 9 to 15 total AKT2 in the HCT116 cells (Figure 1a). (Figure 2c). Again, phosphatase treatment shifted AKT1 into the In HCT116 wt cells as assessed with a pan (total) AKT antibody major P5.75 peak and the two minor peaks noted above, in NIA, AKT migrated as eight major peaks (Figure 1b). The consistent with phosphorylation being the primary cause of the multiple peaks likely represent both AKT1 and AKT2 as well as peak shift induced by insulin. Furthermore, there was an overlap differences in post-translational modification (PTM) of each of the peaks identified under basal and insulin-stimulated molecule. Consistent with this contention, in AKT2 À / À cells, conditions (Figure 2d) compatible with discrete phosphorylation AKT1 ran as six peaks. In AKT1 À / À cells, AKT2 consisted of two events on single AKT1 molecules explaining the mobility shift.

Oncogene (2014) 3463 – 3472 & 2014 Macmillan Publishers Limited Isoform-speciﬁc phosphorylation of AKT H Guo et al 3465 WT AKT2-/- AKT1-/- Anti-total AKT, Serum starved Insulin -+-+-+ AKT1 P5.60 P5.22 P5.29 P5.75 P6.03 AKT2 P5.40 P5.50 P5.88 Total AKT 1200 HCT116 pT308 900 wt pS473 600 pT450 300 0 GAPDH 1.0 1000 AKT2-/- 0.5 750 500 P5.75 0 Total AKT 1.0 250 0 0.5

pT308 0 1000 AKT1-/- 1.0 750 0.5 P5.75 500 P6.03 pS473 0 250 1.0 0 0.5 5.0 5.2 5.4 5.6 5.8 6.0

pT450 0 pI Figure 1. Characterization of phosphorylation of AKT isoforms by western blotting and NIA. (a) AKT isoforms and phosphorylation characterized by western blotting. Wt, AKT2 À / À and AKT1 À / À HCT116 colon cancer cells were serum-starved overnight and treated (or not treated) with insulin (20 mg/ml) for 30 min. Cells were lysed in RIPA buffer with protease inhibitors and phosphatase inhibitors. Lysates (50 mg per lane) were resolved in 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Antibodies for each blot are listed to the left. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) immunoblotting shows equivalent loading. Scanning densitometric values for western blots were obtained using the ImageJ software (version 1.46r; National Institutes of Health, Bethesda, MD, USA). The total AKT, pT308, pS473 and pT450 levels were normalized to the loading control and presented as relative conversion to values in parental cells. (b) AKT isoforms characterized by NIA. Wt, AKT2 À / À and AKT1 À / À HCT116 cells were serum-starved overnight and lysed in NIA RIPA buffer. Samples were analyzed with anti-total AKT antibody. pI is shown on x axis and chemiluminescence on y axis. Only major AKT peaks are shown for clarity and simpliﬁcation of interpretation. Experiments were repeated at least three times and representative data are presented.

Formation of protein complexes could alter the migration of degree, on Ser473 phosphorylation. With T450A and S124A, peaks AKT on NIA. To assess this possibility, we determined the motility still shifted to lower pI indicating these sites are not required for of AKT1 and AKT2 in the presence of 2 M urea and 40 mM phosphorylation on other sites in AKT1. dithiothreitol treatment (Supplementary Figure S2) to disrupt The low protein level of T450A (Supplementary Figure S3C) is 21,22 protein complexes. In all, 2 M urea and 40 mM dithiothreitol did not consistent with this site being required for AKT stability. In alter the motility of AKT1 or AKT2 consistent with the contention contrast, S124A, T308A and S473A were expressed to comparable that the motility shift of AKT is due primarily to PTM. levels to wt AKT1 (Supplementary Figure S3C). Interestingly, each To identify phosphorylation events contributing to the discrete of the mutant forms of AKT1 was downregulated more efficiently peaks of AKT1 observed in NIA, we created non-phophorylatable than wt AKT1 on insulin treatment (Supplementary Figure S3C). mutants of Thr308 (T308A), Ser473 (S473A) Thr450 (T450A) Under basal conditions, phosphorylation of PRAS40, an AKT and Ser124 (S124A). Wt and mutants were stably expressed in substrate, was decreased by S124, T308 and S473 mutations. AKT1 À / À and AKT2 À / À HCT116 cells and detected with total However, most strikingly, the T308 but not other mutations AKT antibody (Figure 2e). As predicted, the T308A and S473A essentially abrogated insulin induced PRAS40 phosphorylation mutations did not significantly alter the migration of AKT1 under implicating T308 in the ability of AKT1 to phosphorylate PRAS40 basal conditions (Figure 2e). In contrast, the T450A and S124A (Supplementary Figure S3C). mutations shifted several low AKT1 pI peaks to a higher pI We further examined effects of the mutations on cell invasion (Figure 2e). The shift to higher pI seen with S124A AKT1 was more as an indication of functional outcomes of altered signaling. pronounced than would be predicted by loss of a single Stable expression of AKT1-wt in HCT116 AKT1 À / À and phosphorylation site. This could be explained by the S124A AKT2 À / À cells dramatically increased invasion (Figure 2f). mutation causing a structural change that altered phosphorylation S124A and T450A were modestly less able to induce cell invasion of other sites in AKT1 or alternatively by Ser124 phosphorylation compared with AKT1-wt (Figure 2f), consistent with S124 being required for processive phosphorylation of additional sites phosphorylation being required for optimal cell invasion. As in AKT1. T450A levels are lower because of decreased stability Under insulin stimulation conditions, there was a significantly (Supplementary Figure S3C), the decrease in invasion induced reduced peak shift with T308A compared with wt suggesting that by the T450A may be secondary to changes in stability rather than phosphorylation of Thr308 contributes to the peak shift. to a functional change in the AKT1 molecule. In contrast, T308A Unexpectedly, insulin stimulation failed to induce a detectable and S473A constructs induced only a modest increase in invasion shift in S473A mobility suggesting that phosphorylation of Thr308 (Figure 2f), indicating that phosphorylation of both T308 and S473 and other sites induced by insulin is dependent, to a major are critical for cell invasion. Expression of AKT1-wt or mutants

& 2014 Macmillan Publishers Limited Oncogene (2014) 3463 – 3472 Isoform-speciﬁc phosphorylation of AKT H Guo et al 3466

acAKT1-/-, Anti-total AKT AKT2-/-, Anti-total AKT e Serum Starved, Anti-total AKT Insulin 30 min, Anti-total AKT

2400 3000 Serum 4000 4000 P5.35 P5.75 P5.44 1800 Serum 2250 P5.53 Starved P5.60 P5.22 P5.75 P6.03 3000 3000 1200 Starved 1500 - PPase AKT1 AKT1 P5.22 600 - PPase 750 2000 wt P5.75 2000 wt 0 0 1000 1000 2000 2000 0 0 Serum 1500 Serum 1500 Starved P5.60 Starved 2000 2000

1000 1000 P5.66 P5.60 P5.60 + PPase P5.88 + PPase 1500 AKT1 1500 AKT1 500 500 T308A T308A 0 0 1000 1000 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 500 500 0 0 1400 1500 Insulin Insulin

1050 P5.07 2000 2000 1125 P5.50 30 mins P5.60 30 mins P5.75 P5.62

P5.53 AKT1

P4.95 AKT1 700 750 - PPase 1500 P5.60

P5.42 1500 - PPase S473A S473A 350 375 1000 1000 0 0 500 500 1400 1800 0 0 Insulin 1050 1350 Insulin 30 mins 30 mins 1200 700 900 1200 + PPase + PPase AKT1 AKT1 350 450 900 900

T450A T450A P5.60

0 0 600 600 P5.75 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 300 300 0 bdAKT1-/-, Anti-total AKT AKT2-/-, Anti-total AKT 0 6000 6000 2000 3000 Serum AKT1 AKT1 Serum 4500 4500 1500 Starved 2250 Starved S124A S124A 1000 1500 3000 P5.75 3000 Insulin Insulin 500 30 mins 750 30 mins 1500 1500 0 0 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

f HCT116 AKT1-/- AKT2-/- expressing AKT1 Parental - wt S124A T450A T308A S473A HCT116

HCT116 AKT1-/- AKT2-/- expressing

4 (Arbitrary units) Invasion capacity 0

al -wt

AKT1 AKT1-S124AAKT1-T450AAKT1-T308A Vector control AKT1-S473A HCT116 parent Figure 2. Characterization of AKT2 and AKT1 PTM by NIA. (a) pI values of unphosphorylated and phosphorylated AKT2. AKT1 À / À HCT116 cells were serum-starved overnight and treated (or not treated) with insulin (20 mg/ml) for 30 min. Cells were lysed in RIPA buffer then treated (or not treated) with l phosphatase. Samples were analyzed with anti-total AKT antibody. Experiments were repeated at least three times and representative data are presented. (b) Overlap of AKT2 peaks at basal and insulin-stimulated conditions. (c) pI values of unphosphorylated and phosphorylated AKT1. AKT2 À / À HCT116 cells were prepared and analyzed as in (a). Experiments were repeated at least three times and representative data are presented. (d) Overlap of AKT1 peaks at basal and insulin-stimulated conditions. (e) Phosphorylation of AKT1 determines the discrete peaks in NIA. AKT1 À / À and AKT2 À / À double knockout HCT116 cells were infected to stably express AKT1-wt and indicated mutations. Cells were serum-starved overnight with/without insulin stimulation for 30 min and lysed in RIPA. Samples were analyzed with anti-total AKT antibody. Only major AKT peaks are shown for clarity and simpliﬁcation of interpretation. (f) Effects of AKT1 mutations on cell invasion. AKT1 À / À AKT2 À / À double knockout HCT116 cells were infected to stably express AKT1-wt and indicated mutations. Cell invasion was measured by BD Matrigel Invasion Chamber according to the manufacturer’s protocol. Representing images are shown. Image quantiﬁcations were obtained using the ImageJ software (version 1.46r). Invasion levels were presented as relative conversion to values in HCT116 double knockout cells. The error bars represent s.d. of three repeats.

did not alter the proliferation of HCT116 double knockout cells in AKT1 residue-speciﬁc phosphorylation at Thr308, Ser473 and normal culture conditions indicating that AKT1 is not required for Thr450 proliferation of HCT116 cells at least under the conditions We then examined the localization of different phosphorylated assessed (Supplementary Figure S3D). residues in particular pI peak using phospho-speciﬁc antibodies.

Oncogene (2014) 3463 – 3472 & 2014 Macmillan Publishers Limited Isoform-speciﬁc phosphorylation of AKT H Guo et al 3467 AKT2-/-, Anti-pAKT450

160 Serum 2500 160 Total AKT 120 Starved 1875 120 pAKT450 80 1250 80 40 625 40 0 0

160 1500 100 Total AKT Insulin 120 1125 30 mins 75 pAKT450 80 750 50 40 375 25 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

AKT2-/-, pAKT308 AKT2-/-, pAKT473

1000 1000 750 Serum 750 Serum P5.12 P5.20 P5.33 Starved P5.03 P5.20 P5.33 P5.03 500 Starved 500 P5.12 250 250 0 0

1000 1000 750 Insulin 750 Insulin 500 30mins 500 30 mins 250 250 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

1500 Total AKT 1500 Total AKT P5.03 1125 900 pAKT308 1125 600 pAKT473 750 600 750 400 375 300 375 200 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

AKT2-/-, Insulin 30 mins AKT2-/-, Insulin 30 mins pAKT308 pAKT308 1000 800 1000 100 pAKT450 750 600 pAKT473 750 75 500 400 500 50 250 200 250 25 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 Figure 3. AKT1 phosphorylation at Thr450, Thr308 and S473. (a–c) AKT1 phosphorylation at Thr450, Thr308 and S473. AKT2 À / À HCT116 cells were serum-starved overnight and treated (or not treated) with insulin (20 mg/ml) for 30 min. Samples were analyzed with anti-pThr450 (a), anti-pThr308 (b) and anti-pSer473 (c) antibodies. (d) Overlap of AKT1 pThr308 and pSer473 peaks with insulin stimulation. (e) Overlap of AKT1 pThr308 and pThr450 peaks with insulin stimulation.

Phosphorylation at Thr450 in the turn motif has been proposed to detected under basal conditions indicating that the discrete pI occur as a cotranslational event required for optimal protein peaks under basal conditions are due to phosphorylation of sites folding and protein stability21,22 and thus for optimal AKT other than Thr308 or Ser473 (Figures 3b and c). As expected, activity.11 Based on this concept, Thr450 phosphorylation would insulin induced a marked increase in the AKT1 Thr308 and Ser473 be expected on the majority if not all AKT molecules. As predicted, phosphorylation signals (Figures 3b and c). pThr308 and pSer473 the pattern of peaks detected with the pThr450-specific antibody phosphorylation demonstrated a high degree of overlap (Figure 3a, arrow heads designate potential nonspecific peaks not (Figure 3d) indicating that in most cases the two residues were seen in other assays) overlapped to a high degree with peaks coordinately phosphorylated, which is consistent with previous detected with the total AKT antibody in the presence or absence findings.11 However, several minor pThr308 peaks did not overlap of insulin (see Figure 2c and d). The only exception is the minor with pSer473 peaks (Figure 3d, see arrows) indicating pThr308 and P5.75 peak that likely represents unphosphorylated AKT1 that is pSer473 phosphorylation was uncoupled in a fraction of AKT1 not recognized by the pThr450 antibody (Figure 3a, see arrow). molecules. In contrast, as predicted by Thr450 being phosphory- Phosphorylation at Thr308 on the T-loop and Ser473 in the lated in most AKT1 molecules, pThr308 only identified a subset of hydrophobic motif is required for optimal AKT activation.11 In the peaks detected with the pThr450 antibody (Figure 3e). Overlay AKT2 À / À cells, AKT1 Thr308 or Ser473 phosphorylation was not of peaks identified by the total AKT1 and the pThr308 and

& 2014 Macmillan Publishers Limited Oncogene (2014) 3463 – 3472 Isoform-specific phosphorylation of AKT H Guo et al 3468 pSer473-specific antibodies showed that the majority of pThr308 but absent from AKT2 À / À cells. AKT2 pSer474 only accounted (B70%) and pSer473 (B67%) was in peaks with pI value of 5.20 or for 7.5% of total hydrophobic pSer473/474 phosphorylation in lower (compare Figures 3b and c with Figures 2c and d). HCT116 wt cells (see Supplementary Table S2 for the identification Comparing total and phospho-specific peaks, the AKT1 P5.03 and quantification of each peak in Figure 4b) suggesting that peak accounted for 7.94±1.14% of total AKT1, 16.8±2.6% of total AKT1 accounts for 92.5% of Ser473/474 phosphorylation. In pThr308 and 15.4±3.2% of total pSer473 (% data represent the contrast to AKT1, AKT2 phosphorylation at pThr309 and pSer474 mean±s.d. of three independent experiments). Assuming that the were largely uncoupled. For example, the AKT2 P5.77 peak p5.03 peak is quantitatively phosphorylated at Thr308 and Ser473, contained pThr309 but not pSer474, and the AKT2 P5.53 and P5.70 AKT1 was extensively phosphorylated at Thr308 (47%) and Ser473 peaks contained pSer474 but not pThr309. Only AKT2 in the minor (52%) after insulin treatment (Figures 3b and c). P5.42 peak likely contained both pThr309 and pSer474 phosphor- We further examined the effects of T308A, S473A, T450A and ylation (first panels in Figures 4a and b). Whether differential S124A mutations in AKT1 on insulin-induced phosphorylation of phosphorylation on Thr308 and Ser473 between AKT1 and AKT2 is Thr308 and Ser473. Interestingly, while T308A mutation abolished cell line or stimulus-specific or kinetic-dependent warrants further pThr308 and significantly decreased pSer473, S473A mutation studies. abolished both pSer473 and pThr308, indicating Thr308 phos- If the pan (total) AKT antibody has similar affinities for both AKT phorylation is completely dependent on pSer473 phosphorylation. isoforms, AKT2 represents 26% of total AKT (Figure 1b; see Thr308 and Ser473 were phosphorylated in T450A and S124A Supplementary Table S3 for the identification and quantification constructs indicating that Thr308 and Ser473 phosphorylation is of each peak in Figure 1b), but only accounted for much smaller not dependent on Thr450 or Ser124 phosphorylation percentage of T-loop or hydrophobic domain phosphorylation (Supplementary Figure S3). However, low pI peaks were shifted (3.8% or 7.5%, respectively). Thus, insulin mainly induces AKT1 but to a higher pI indicative the loss of pI shift because of phosphates not AKT2 phosphorylation and activation in HCT116 wt cells. on Thr450 or Ser124. Effect of PI3K inhibition on AKT isoform-specific phosphorylation AKT2 residue-specific phosphorylation at Thr309 and Ser474 The PI3K pathway is a rich therapeutic target across multiple In AKT1 À / À cells, AKT2 phosphorylation at Thr309 or Ser474 was lineages with several PI3K inhibitors in clinical trials. However, AKT not detected under basal conditions. With insulin treatment, isoform-specific response to these inhibitors has not been pThr309 was detected at low levels in both AKT1 À / À and elucidated. Pre-treatment with GDC0941 (PI3K and mTORC1 dual HCT116 wt cells (Figure 4a). The two minor pThr309 peaks present inhibitor with preference for PI3K) for 3 h, but not rapamycin in AKT1 À / À cells were detected in HCT116 wt cells but not (mTORC1 allosteric inhibitor), abolished insulin-induced AKT AKT2 À / À cells indicating that they likely represent AKT2 phosphorylation on all peaks containing Thr308 and Ser473 in phosphosites (Figure 4a). If the phospho-specific antibody has HCT116 wt cells (Figures 5a and b). Thus, PI3K activity is necessary similar affinity for both AKT isoforms, AKT2 pThr309 only accounts for insulin-induced phosphorylation of all peaks containing Thr308 for 3.8% while AKT1 pThr308 accounting for 96.2% of total T-loop and Ser473 in HCT116 wt cells. pThr308/309 phosphorylation in HCT116 wt cells (see Treatment with GDC0941 completely blocked peak shifts of AKT Supplementary Table S1 for the identification and quantification induced by insulin as assessed with pan-AKT antibody (compare of each peak in Figure 4a). Figure 5c with Figure 1b), indicating that inhibition of PI3K Similar to AKT2 Thr309 phosphorylation, insulin only induced blocked all insulin-induced phosphorylation independent of which modest changes in AKT2 pSer474 in AKT1 À / À cells (Figure 4b). sites were responsible for the mobility shift. Rapamycin did not Three pSer474 containing peaks at P5.42, P5.53 and P5.70 alter the effect of insulin on mobility of AKT pI peaks (Figure 5 and detected in AKT1 À / À cells were also present in HCT116 wt cells Supplementary Figures S4B and C).

Anti-pAKT308, Insulin 30 mins Anti-pAKT473, Insulin 30 mins

AKT1-/- AKT1-/- 1000 1000 750 750 P5.53 P5.42 P5.42 500 P5.77 500 P5.70 250 250 0 0 AKT2-/- AKT2-/- 1000 1000 750 750 P5.39 P5.51 P5.39 500 500 P5.46 250 250 0 0 HCT116 wt HCT116 wt 1000 1000

750 P5.53 750 P5.39 P5.42 P5.39 P5.70 P5.42 P5.51 500 P5.77 500 250 250 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 Figure 4. T-loop and hydrophobic domain phosphorylation on AKT1 and AKT2 with insulin stimulation. (a, b) Comparison of AKT2 pThr309 and pSer474 with AKT1 pThr308 and pSer473. AKT1 À / À , AKT2 À / À and HCT116 wt cells were serum-starved overnight and treated with insulin (20 mg/ml) for 30 min. Samples were analyzed with anti-pThr308 (309) (a) and anti-pSer473 (474) (b) antibodies. Experiments were repeated at least three times and representative data are presented.

Oncogene (2014) 3463 – 3472 & 2014 Macmillan Publishers Limited Isoform-speciﬁc phosphorylation of AKT H Guo et al 3469 HCT116 wt, insulin 30 mins HCT116 wt, insulin 30 mins HCT116 wt, insulin 30 mins Anti-pAKT308 Anti-pAKT473 Anti-total AKT

DMSO DMSO 2000 1400 2200 DMSO 1500 1050 1650 P5.88 1000 700 1100 500 350 550 0 0 0 GDC0941 GDC0941 1400 2000 GDC0941 2800 10µM µ µ 10 M 1050 10 M P6.03 P5.75 1500 2100 P5.88 1000 700 1400 500 350 700 0 0 0 Rapamycin Rapamycin Rapamycin 2000 1400 2400 10µM 10µM 10µM 1050 1500 1800 P5.88 1000 700 1200 500 350 600 0 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

HCT116 wt, serum starved Anti-total AKT

2400 DMSO

1800 P5.75

1200 P6.03 DMSO -+-- 600 GDC0941 --+- 0 Rapamycin ---+ GDC0941 Insulin -+++ 2400 10µM 1800 pAKT308 P5.88 1200 pAKT473 600 pAKT450 0 Total AKT Rapamycin GAPDH 2400 10µM 1800 1200 600 0 5.0 5.2 5.4 5.6 5.8 6.0 Figure 5. Effect of PI3K inhibition on AKT isoform-speciﬁc phosphorylation. (a, b) Effect of PI3K inhibition on AKT phosphorylation at Thr308 or Ser473. Wt HCT116 cells were serum-starved overnight and treated with dimethylsulphoxide (DMSO) control or indicated inhibitors (10 mg/ml) for 3 h, then treated with insulin (20 mg/ml) for 30 min. Samples were analyzed with anti-pThr308 (a) and anti-pSer473 (b) antibodies. (c, d) Effect of PI3K inhibition on overall AKT phosphorylation. Wt HCT116 cells were serum-starved overnight and treated with DMSO control or indicated inhibitors (10 mg/ml) for 3 h, then treated with insulin (20 mg/ml) for 30 min (c) or without insulin treatment (d). Samples were analyzed with anti-total AKT antibody. Experiments were repeated at least three times and representative data are presented. (e) Effect of PI3K inhibition on AKT phosphorylation characterized by western blotting. Samples from (a) were analyzed by western blotting. Antibodies for each blot are listed to the left. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) immunoblotting shows equivalent loading.

Intriguingly, GDC0941 did not alter the pattern of basal AKT1 Figures S4B and C). Indeed, under basal conditions with peaks, indicating that basal AKT1 phosphorylation is not GDC0941 treatment for 3 h, AKT2 pThr451 decreased by at least dependent on ongoing PI3K activity (Figure 5d). In contrast, 50% (Supplementary Figure S4C). The mechanism by which GDC0941 markedly altered the pattern of basal AKT2 phosphor- GDC0941 selectively decreases Thr451 phosphorylation in AKT2 ylation resulting in a decrease in the p5.88 peak and an increase in remains to be determined. the P6.03 (unphosphorylated AKT2) peak, consistent with basal As assessed by western blotting, GDC0941, but not rapamycin, AKT2 phosphorylation being dependent on ongoing PI3K activity abolished AKT pThr308 and pSer473 and modestly decreased (Figure 5d). Indeed using a pan AKT antibody to directly compare pThr450 consistent with the NIA results (Figure 5e). AKT1 and AKT2, GDC0941 induced a dose-dependent increase in the P6.03 unphosphorylated AKT2 peak without significantly altering the AKT1 peaks present (Supplementary Figure S4A). The AKT isoform-specific phosphorylation in other cell lines effect of GDC0941 on basal AKT2 phosphorylation was confirmed We examined AKT phosphorylation profiles in multiple cell lines using an AKT2-specific antibody (Supplementary Figure S4B). with different genetic backgrounds and cell lineages to determine Consistent with these results, GDC0941 markedly decreased the generality of phosphorylation patterns. In H358 lung cancer pThr451 in AKT2 but not pThr450 in AKT1 (Supplementary cells, the AKT peaks were similar to those in HCT116 wt cells under

& 2014 Macmillan Publishers Limited Oncogene (2014) 3463 – 3472 Isoform-speciﬁc phosphorylation of AKT H Guo et al 3470 H358, Anti-total AKT A549, Anti-total AKT

6000 Serum 6000 Serum Starved Starved 4500 P5.88 4500 P5.75 P5.75 3000 3000 P6.03 1500 1500 0 0

6000 Insulin 6000 Insulin 4500 30 mins 4500 30 mins P5.88 3000 3000 1500 1500 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0

HEK293, Anti-total AKT SKOV3, Anti-total AKT

4000 Serum 6000 Serum Starved Starved

3000 4500 P5.75 P5.75 2000 3000 1000 1500 0 0 Insulin Insulin 4000 6000 30 mins 30 mins 3000 4500 P5.60 2000 3000 1000 1500 0 0 5.0 5.2 5.4 5.6 5.8 6.0 5.0 5.2 5.4 5.6 5.8 6.0 Figure 6. AKT isoform-speciﬁc phosphorylation in other cell lines. H358 (a) and A549 (b) lung cancer cells, HEK293 embryonic kidney cells (c) and SKOV3 ovarian cancer cells (d) were serum-starved overnight and treated (or not treated) with insulin (20 mg/ml) for 30 min. Samples were analyzed with anti-total AKT antibody.

both serum starvation and insulin stimulation conditions consis- samples from patients undergoing therapies targeting the AKT tent with similar regulatory mechanisms (Figure 6a and pathway. Supplementary Figures S5A and B). However, in the A549 lung NIA is dependent on both the quality and validation of available cancer cell line, although the pattern of peaks was similar, the antibodies. In order to be fully quantitative, different isoforms AKT2 P5.88 peak was significantly higher than the AKT1 peaks and PTM must be identified with an antibody with equal consistent with higher AKT2 levels in A549 cells (Figure 6b and affinity for each isoform analyzed and also where the affinity is Supplementary Figure S5I). not altered by PTM. The identification of peaks representing In HEK293 cells (Figure 6c and Supplementary Figures S5E and specific isoforms and PTMs can also prove problematic. The F) and SKOV3 cells (Figure 6d and Supplementary Figures S5G and approach herein of using knockout isogenic cells, mutant H), multiple sites were phosphorylated, as indicated by low pI constructs and multiple isoform and phospho-specific antibodies peaks under serum starvation conditions. Insulin did not alter the can alleviate these challenges. Indeed, the validated total, location of AKT peaks in SKOV3 cells (Figure 6d and phospho-specific and isoform-specific antibodies demonstrate a Supplementary Figures S5G and H), potentially because of high marked concordance in terms of the relative amounts of AKT1 and basal AKT phosphorylation likely as a consequence of HER2 AKT2 and of the differentially phosphorylated molecules in amplification and the presence of a PIK3CA mutation in this line. parental and knockout cells. The AKT2 P5.70 peak was detected by pSer473/474 but not In this study, we developed and implemented an approach pThr308/309 antibodies in H358, A549, HEK293 and SKOV3 based on NIA technology to characterize total and residue-specific (Supplementary Figures S5A–H), verifying uncoupled AKT2 phos- phosphorylation profiles of specific AKT1 and AKT2 molecules. By phorylation on Ser474 and Thr309 is generalizable across cell lines combining data from both basal and insulin-treated conditions, and cell lineages. we were able to identify at least 12 and as many as 16 independent AKT1 peaks and at least 5 and up to 6 independent AKT2 peaks, indicative of complex patterns of phosphorylation of DISCUSSION different residues on individual AKT molecules. To a major degree NIA provides a rapid, quantitative, sensitive and specific approach using knockout cells, mutant constructs and phospho-specific to characterize PTM of proteins in tissue culture and patient antibodies, we were able to define which AKT isoforms and which samples.16 The approach is fast, with the ability to analyze dozens phosphorylation sites are found in a particular peak. For example, of samples with multiple antibodies within 3–4 h. Protein can be under serum starvation conditions (Figure 1b), the P6.03 detected at picogram levels16 allowing analysis of PTM changes in peak contains AKT2 with no phosphorylation; the P5.88 peak samples where limited protein is available. The predicted pI values contains AKT2 with pThr451; the P5.75 peak contains AKT2 with of 5.75 and 5.96 for unphosphorylated AKT1 and AKT2 (http:// pThr451 and AKT1 with no phosphorylation. Under basal web.expasy.org/compute_pi/)18 are suitable for the isoelectric conditions, the lower pI peaks consist of AKT1 with pThr450 focusing approach for NIA. In addition to exploring underlying phosphorylation without pThr308 or pSer473 phosphorylation mechanisms regulating AKT phosphorylation, NIA could be used sites (Supplementary Table S4). Identification of the phosphoryla- to monitor AKT isoform-specific activation and inhibition in tion sites constituting the peaks present following insulin

Oncogene (2014) 3463 – 3472 & 2014 Macmillan Publishers Limited Isoform-specific phosphorylation of AKT H Guo et al 3471 stimulation is more complex because of the number of peaks, American Type Culture Collection (Manassas, VA, USA). Cells were cultured however, the combination of mutant constructs and phospho- in Dulbecco’s modified Eagle’s medium (Invitrogen, Carlsbad, CA, USA; specific antibodies indicate that pThr450/451 is present in all HeLa and HEK293 cells), McCoy’s 5A (Invitrogen; HCT116 wt and knockout peaks and that pThr308 and pSer473 contribute to the mobility cells), or RPMI 1640 medium (Invitrogen; NCI-H358, A549 and SKOV3 cells) shifts following insulin treatment resulting in accumulation of AKT supplemented with 10% (v/v) fetal calf serum (Invitrogen). Cell line identify was routinely confirmed by STR profiling in the MDACC CCSG core. forms with low pI and complex phosphorylation patterns (Supplementary Table S5). We identified unexpected isoform-specific regulation of AKT Reagents and antibodies phosphorylation by insulin stimulation as well as by PI3K Rapamycin was from Cell Signaling Technology (Beverly, MA, USA). The PI3K inhibition, which had not been appreciated using other and mTORC1 inhibitor GDC-0941 was from Axon Medchem (Groningen, The approaches. In particular, under basal conditions AKT1 molecules Netherlands). Antibodies against pan (total) AKT (CS-4691), phospho-AKT were phosphorylated on multiple sites, but not on the two Thr308 (CS-2965), phospho-AKT Thr450 (CS-9267), phospho-AKT Ser473 (CS- classical Thr308 and Ser473 sites. Insulin treatment markedly 9271) and AKT2 (CS-3063) were from Cell Signaling Technology. AKT1 increased AKT1 phosphorylation and in particular increased the antibody (BD-610861) was from BD Biosciences (San Jose, CA, USA). An anti-glyceraldehyde 3-phosphate dehydrogenase antibody (AM4300) was complexity of phosphorylation on each AKT1 molecule. Phosphor- obtained from Ambion (Austin, TX, USA). ylation of both Thr308 and Ser473 as well as all other possible residues phosphorylated in response to insulin were dependent on PI3K activity. AKT2 phosphorylation was much less responsive Constructs to insulin treatment than AKT1. Insulin markedly increased Full-length complementary DNA encoding AKT1 (NM_001014431.1) was phosphorylation of Thr308 or Ser473 in AKT1 and to a lesser from Invitrogen. The indicated mutations were introduced using standard degree AKT2. In AKT1, pThr308 and pSer473, in most cases, site-directed mutagenesis and verified by sequencing. Complementary occurred on the same molecule as previously reported.11 DNAs were cloned into a Gateway compatible pBabe-puro retroviral vector (Addgene, Cambridge, MA, USA, #1764) by Gateway recombination However, there was a sub-population of AKT1 molecules with according to the manufacturer’s protocol (Invitrogen). only pThr308 or pSer473. Intriguingly, Thr308 or Ser473 phosphorylation in AKT2 frequently occurred on independent AKT2 molecules. The specific AKT1 and AKT2 phosphorylation Lysate preparation profiles may contribute to their different substrate specificities. For To prepare cell lysates for western blotting, cells were lysed in RIPA buffer example, GSKb, p21cip, SKP2 and palladin are preferentially (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycho- phosphorylated by AKT1, whereas GSKa, MDM2 and AS160 appear late, 0.1% sodium dodecyl sulfate) supplemented with protease inhibitor to be targeted by AKT2.4,23–26 The site-specific AKT and phosphatase inhibitor cocktails (Pierce, Rockford, IL, USA). Western blotting was performed as described previously.28 To prepare cell lysates phosphorylation profiles also contribute to substrate specificities. for NIA, cells were lysed in NIA-specific RIPA lysis buffer (ProteinSimple Inc., AKT with Thr308 but not Ser473 phosphorylation effectively M M 27 Santa Clara, CA, USA; 20 m HEPES, pH 7.5, 150 m NaCl, 1% NP40). phosphorylates GSK3 and TSC2 but not FOXO1/FOXO3a. Our Alternatively, 0.25% sodium deoxycholate supplemented with protease data show Thr308 phosphorylation is, to a major degree, inhibitor and phosphatase inhibitor cocktails (ProteinSimple Inc.). To dependent on Ser473 phosphorylation. Previous studies11,27 remove phosphorylation from proteins, lysates were treated with l suggested that phosphorylation of these two sites could be phosphatase (Millipore, Billerica, MA, USA) according to the manufacturer’s independent under certain conditions. However, in contrast to protocol. transient expression of tagged AKT in these studies,11 we expressed an untagged AKT by stable infection, to levels similar Nanofluidic proteomic immunoassay to those in wt cells, in AKT knockout cells, which is more likely to Lysates were analyzed using a NanoPro 1000 system (ProteinSimple Inc.) reflect physiological AKT phosphorylation processes. with an optimized protocol. The ampholyte premix was G2 5–8 Basal AKT2 but not AKT1 phosphorylation was sensitive to PI3K (ProteinSimple Inc. #040-973) nested with G2 5–6 (ProteinSimple Inc. inhibition. Using reagents validated with AKT1 and AKT2 knock- #040-971) at a ratio of 1:1. Samples were loaded at a concentration of outs, PI3K inhibition decreased pThr451 on AKT2 but not pThr450 200 mg/ml in capillaries. Focusing conditions were 40 000 mW for 40 min. on AKT1. Whether this represents differential activity of isoform- Ultraviolet immobilization time was 60 s with an exposure time of 240 s. specific pThr450 kinases or phosphatases needs to be determined. Primary antibodies were used at 1:50 dilution for 2 h. Anti-rabbit In summary, we developed and implemented an approach (ProteinSimple Inc. #040-656) and anti-mouse (ProteinSimple Inc. #040- 655) secondary antibodies were used at 1:100 dilution for 1 h. Tertiary based on NIA technology for AKT isoform-specific phosphorylation antibody (ProteinSimple Inc. #041-106) was used at 1:100 dilution for analysis and revealed unexpected and important aspects of 10 min. The pI standards for 4.9, 5.5, 6, 6.4 and 7.0 (ProteinSimple Inc.) were regulation of AKT isoform phosphorylation, which has been used based on predicted pI of AKT isoforms. undetectable by other methods, like mass spectrometry and western blotting. The differential AKT1 and AKT2 phosphorylation under basal and insulin-stimulated conditions as well as the CONFLICT OF INTEREST differential sensitivity to inhibitors suggests different mechanisms The authors declare no conflict of interest. regulating phosphorylation and may contribute to the different functions of AKT1 and AKT2. The approach and the results from this study have the potential to advance our understanding of AKT ACKNOWLEDGEMENTS isoform-specific activation, which will aid in implementation of We thank Dr Bert Vogelstein for providing HCT116 wt and knockout cell lines. The approaches to target aberrant AKT activation in cancer cells. work is supported by National Institutes of Health (NIH) grant 5R21CA126700, a grant from the University of Texas MD Anderson Cancer Center Kidney Cancer Multi- disciplinary Research Program, and research support from AstraZeneca to ZD, and MATERIALS AND METHODS the Clinical Proteomic Tumor Analysis Consortium (CPTAC) grants (U24 CA126477, Cell culture U24 CA126479, and U54 CA112970) to GBM and NCI CCSG grant (P30 CA016672). The HeLa cell line was a Tet-on derivative obtained from BD Clontech (Palo Alto, CA, USA). The wt, AKT1 À / À and AKT2 À / À HCT116 colon cancer cell lines were gifts from Dr Bert Vogelstein (Johns Hopkins University, REFERENCES Baltimore, MD, USA). The lung cancer cell lines NCI-H358 and A549, the 1 Brazil DP, Yang ZZ, Hemmings BA. Advances in protein kinase B signalling: AKTion ovarian cancer cell line SKOV3, and HEK293 were obtained from the on multiple fronts. Trends Biochem Sci 2004; 29: 233–242.

& 2014 Macmillan Publishers Limited Oncogene (2014) 3463 – 3472 Isoform-specific phosphorylation of AKT H Guo et al 3472 2 Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. Cell 2007; 16 Fan AC, Deb-Basu D, Orban MW, Gotlib JR, Natkunam Y, O’Neill R et al. Nanofluidic 129: 1261–1274. proteomic assay for serial analysis of oncoprotein activation in clinical specimens. 3 Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/AKT pathway Nat Med 2009; 15: 566–571. for cancer drug discovery. Nat Rev Drug Discov 2005; 4: 988–1004. 17 Ericson K, Gan C, Cheong I, Rago C, Samuels Y, Velculescu VE et al. Genetic 4 Chin YR, Toker A. The actin-bundling protein palladin is an Akt1-specific substrate inactivation of AKT1, AKT2, and PDPK1 in human colorectal cancer cells that regulates breast cancer cell migration. Mol Cell 2010; 38: 333–344. clarifies their roles in tumor growth regulation. Proc Natl Acad Sci USA 2010; 107: 5 Dillon RL, Marcotte R, Hennessy BT, Woodgett JR, Mills GB, Muller WJ. Akt1 and 2598–2603. akt2 play distinct roles in the initiation and metastatic phases of mammary tumor 18 Bjellqvist B, Hughes GJ, Pasquali C, Paquet N, Ravier F, Sanchez JC et al. The progression. Cancer Res 2009; 69: 5057–5064. focusing positions of polypeptides in immobilized pH gradients can be predicted 6 Endersby R, Zhu X, Hay N, Ellison DW, Baker SJ. Nonredundant functions for Akt from their amino acid sequences. Electrophoresis 1993; 14: 1023–1031. isoforms in astrocyte growth and gliomagenesis in an orthotopic transplantation 19 Yang WL, Wang J, Chan CH, Lee SW, Campos AD, Lamothe B et al. The E3 ligase model. Cancer Res 2011; 71: 4106–4116. TRAF6 regulates Akt ubiquitination and activation. Science 2009; 325: 1134–1138. 7 Alessi DR, James SR, Downes CP, Holmes AB, Gaffney PR, Reese CB et al. 20 Sundaresan NR, Pillai VB, Wolfgeher D, Samant S, Vasudevan P, Parekh V et al. The Characterization of a 3-phosphoinositide-dependent protein kinase which deacetylase SIRT1 promotes membrane localization and activation of Akt and phosphorylates and activates protein kinase Balpha. Curr Biol 1997; 7: 261–269. PDK1 during tumorigenesis and cardiac hypertrophy. Sci Signal 2011; 4: ra46-. 8 Sarbassov DD, Guertin DA, Ali SM, Sabatini DM. Phosphorylation and regulation of 21 Facchinetti V, Ouyang W, Wei H, Soto N, Lazorchak A, Gould C et al. The mam- Akt/PKB by the rictor-mTOR complex. Science 2005; 307: 1098–1101. malian target of rapamycin complex 2 controls folding and stability of Akt and 9 Bozulic L, Surucu B, Hynx D, Hemmings BA. PKB[alpha]/Akt1 acts downstream of protein kinase C. EMBO J 2008; 27: 1932–1943. DNA-PK in the DNA double-strand break response and promotes survival. Mol Cell 22 Oh WJ, Wu CC, Kim SJ, Facchinetti V, Julien LA, Finlan M et al. mTORC2 can 2008; 30: 203–213. associate with ribosomes to promote cotranslational phosphorylation and stabi- 10 Andjelkovic M, Alessi DR, Meier R, Fernandez A, Lamb NJ, Frech M et al. Role of lity of nascent Akt polypeptide. EMBO J 2010; 29: 3939 51. translocation in the activation and function of protein kinase B. J Biol Chem 1997; 23 Bouzakri K, Zachrisson A, Al-Khalili L, Zhang BB, Koistinen HA, Krook A et al. siRNA- 272: 31515–31524. based gene silencing reveals specialized roles of IRS-1/Akt2 and IRS-2/Akt1 11 Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P et al. Mechanism in glucose and lipid metabolism in human skeletal muscle. Cell Metab 2006; of activation of protein kinase B by insulin and IGF-1. EMBO J 1996; 15: 4: 89–96. 6541–6551. 24 Brognard J, Sierecki E, Gao T, Newton AC. PHLPP and a second isoform, PHLPP2, 12 Mao M, Fang X, Lu Y, Lapushin R, Bast Jr. RC, Mills GB. Inhibition of growth-factor- differentially attenuate the amplitude of Akt signaling by regulating distinct Akt induced phosphorylation and activation of protein kinase B/Akt by atypical isoforms. Mol Cell 2007; 25: 917–931. protein kinase C in breast cancer cells. Biochem J 2000; 352(Pt 2): 475–482. 25 Gao D, Inuzuka H, Tseng A, Chin RY, Toker A, Wei W. Phosphorylation by Akt1 13 Powell DJ, Hajduch E, Kular G, Hundal HS. Ceramide disables 3-phosphoinositide promotes cytoplasmic localization of Skp2 and impairs APCCdh1-mediated Skp2 binding to the pleckstrin homology domain of protein kinase B (PKB)/Akt by a destruction. Nat Cell Biol 2009; 11: 397–408. PKCzeta-dependent mechanism. Mol Cell Biol 2003; 23: 7794–7808. 26 Gonzalez E, McGraw TE. Insulin-modulated Akt subcellular localization determines 14 Mahajan K, Coppola D, Challa S, Fang B, Chen YA, Zhu W et al. Ack1 mediated Akt isoform-specific signaling. Proc Natl Acad Sci USA 2009; 106: 7004–7009. AKT/PKB tyrosine 176 phosphorylation regulates its activation. PLoS One 2010; 5: 27 Jacinto E, Facchinetti V, Liu D, Soto N, Wei S, Jung SY et al. SIN1/MIP1 maintains e9646. rictor-mTOR complex integrity and regulates Akt phosphorylation and substrate 15 Hornbeck PV, Kornhauser JM, Tkachev S, Zhang B, Skrzypek E, Murray B et al. specificity. Cell 2006; 127: 125–137. PhosphoSitePlus: a comprehensive resource for investigating the structure and 28 Ding Z, Liang J, Li J, Lu Y, Ariyaratna V, Lu Z et al. Physical association of PDK1 with function of experimentally determined post-translational modifications in man AKT1 is sufficient for pathway activation independent of membrane localization and mouse. Nucleic Acids Res 2012; 40(Database issue): D261–D270. and phosphatidylinositol 3 kinase. PLoS One 2010; 5:e9910.

Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)

Oncogene (2014) 3463 – 3472 & 2014 Macmillan Publishers Limited