HER2 Missense Mutations Have Distinct Effects on Oncogenic

HER2 missense mutations have distinct effects on PNAS PLUS oncogenic signaling and migration

Daniel J. Zabranskya, Christopher L. Yankaskasb, Rory L. Cochrana, Hong Yuen Wonga, Sarah Croessmanna, David Chua, Shyam M. Kavuric,1, Monica Red Brewerd, D. Marc Rosena, W. Brian Daltona, Ashley Cimino-Mathewsa,e, Karen Craveroa, Berry Buttona, Kelly Kyker-Snowmana, Justin Cidadoa,2, Bracha Erlangera, Heather A. Parsonsa, Kristen M. Mantob, Ron Bosec,f, Josh Lauringa, Carlos L. Arteagad, Konstantinos Konstantopoulosb,g, and Ben Ho Parka,b,3

aSidney Kimmel Comprehensive Cancer Center, The Johns Hopkins University School of Medicine, Baltimore, MD 21287; bDepartment of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD 21218; cDivision of Oncology, Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110; dDepartment of Medicine: Breast Cancer Research Program, Vanderbilt-Ingram Cancer Center, Vanderbilt University, Nashville, TN 37232; eDepartment of Pathology, The Johns Hopkins Hospital and School of Medicine, Baltimore, MD 21287; fSiteman Cancer Center, Washington University School of Medicine, St. Louis, MO 63141; and gJohns Hopkins Institute for NanoBioTechnology, Johns Hopkins University, Baltimore, MD 21218

Edited by Peter K. Vogt, The Scripps Research Institute, La Jolla, CA, and approved September 29, 2015 (received for review August 24, 2015) Recurrent human epidermal growth factor receptor 2 (HER2) L755S, has been described to be associated with lapatinib re- missense mutations have been reported in human cancers. These sistance when overexpressed (4, 10). mutations occur primarily in the absence of HER2 gene amplification Past studies comparing overexpression of a mutant cDNA such that most HER2-mutant tumors are classified as “negative” by to single nucleotide knockin of mutant oncogenes have shown FISH or immunohistochemistry assays. It remains unclear whether dramatic differences in signaling and transformed phenotypes nonamplified HER2 missense mutations are oncogenic and whether (11–13). Additionally, we have shown KRAS G12V mutations, as they are targets for HER2-directed therapies that are currently ap- single copies, have relatively little effect in breast epithelial cells, proved for the treatment of HER2 gene-amplified breast cancers. Here but have a profound effect on cancerous phenotypes and tu- we functionally characterize HER2 kinase and extracellular domain morigenicity when coupled with a PIK3CA oncogenic hotspot mutations through gene editing of the endogenous loci in HER2 non- mutation (14). Given these caveats, we used two HER2 nonamplified human breast epithelial cells. In in vitro and in vivo assays, amplified human breast epithelial cell lines to create an isogenic MEDICAL SCIENCES HER2 themajorityof missense mutations do not impart detectable panel of HER2 missense knockin mutants, to study the effects of HER2 oncogenic changes. However, the V777L mutation increased single-copy HER2 mutations under gene expression levels com- PIK3CA biochemical pathway activation and, in the context of a parable to what has been observed in HER2-mutant cancers. mutation, enhanced migratory features in vitro. However, the V777L mutation did not alter in vivo tumorigenicity or sensitivity to Significance HER2-directed therapies in proliferation assays. Our results suggest the oncogenicity and potential targeting of HER2 missense mutations should be considered in the context of cooperating genetic The discovery of human epidermal growth factor receptor 2 HER2 alterations and provide previously unidentified insights into func- ( ) missense mutations in breast and other cancers poten- tional analysis of HER2 mutations and strategies to target them. tially make such tumors susceptible to current and future HER2-targeted therapies. However, the majority of HER2 HER2 missense mutations | breast cancer | mutant oncogenes | mutations occur in HER2 nonamplified cancers, and whether targeted therapies these mutations will predict for sensitivity to HER2-directed therapies remains unknown. Using genome editing, the data presented here suggest that HER2 missense mutations are great success in the treatment of breast cancer has come functionally distinct and require additional oncogenic input to Afrom the identification of human epidermal growth factor HER2 HER2 /Neu ERBB2 impart cancerous phenotypes. These results suggest that receptor 2 ( ) ( ) amplification/overexpression as missense mutations by themselves may not be reliable pre- ∼ a targetable driver in 20% of breast cancers (1). HER2 is a dictors of response to HER2-targeted therapies, a hypothesis member of the ErbB family of transmembrane receptor tyrosine currently being tested in genomically driven clinical trials. kinases, which includes the epidermal growth factor receptor (EGFR/ErbB1), HER3 (ErbB3), and HER4 (ErbB4) (2). Acti- Author contributions: D.J.Z. and B.H.P. designed research; D.J.Z., C.L.Y., R.L.C., H.Y.W., vation of ErbB signaling causes receptor tyrosine autophosphor- S.C., D.C., S.M.K., M.R.B., D.M.R., W.B.D., A.C.-M., K.C., B.B., K.K.-S., J.C., B.E., and K.M.M. performed research; D.J.Z. and C.L.Y. contributed new reagents/analytic tools; D.J.Z., ylation and induces interactions with cytoplasmic signal transduction C.L.Y., R.L.C., H.Y.W., S.C., D.C., S.M.K., M.R.B., D.M.R., W.B.D., A.C.-M., K.C., B.B., K.K.-S., partners that promote a wide variety of cellular processes including J.C., B.E., H.A.P., K.M.M., R.B., J.L., C.L.A., K.K., and B.H.P. analyzed data; and D.J.Z., C.L.Y., proliferation, motility, and escape from apoptosis. In addition to R.L.C., H.Y.W., S.C., D.C., S.M.K., M.R.B., D.M.R., W.B.D., A.C.-M., K.C., B.B., K.K.-S., J.C., B.E., their key role in normal cellular growth and maintenance, the H.A.P., K.M.M., R.B., J.L., C.L.A., K.K., and B.H.P. wrote the paper. dysregulation of ErbB receptors has been extensively implicated in Conflict of interest statement: B.H.P. is a paid consultant for Novartis and is a member of the scientific advisory boards of Horizon Discovery, Ltd and Loxo Oncology, and has re- the development of numerous cancers (1). search contracts with Genomic Health, Inc. and Foundation Medicine. Under separate Whereas overexpression or amplification of HER2 has been licensing agreements between Horizon Discovery, Ltd. and The Johns Hopkins University, well described to deregulate ErbB signaling, cancer genome se- B.H.P. is entitled to a share of royalties received by the university on sales of products. The terms of this arrangement are being managed by Johns Hopkins University, in accordance quencing studies have demonstrated that somatic point muta- with its conflict of interest policies. All other authors declare no potential conflicts. HER2 tions in the gene occur in a number of cancers, including This article is a PNAS Direct Submission. – – 2 4% of breast cancers (3 6). Importantly, these mutations are 1Present address: Lester and Sue Smith Breast Cancer Center, Baylor College of Medicine, most often found in patients as single copies without amplifica- Houston, TX 77019. tion/overexpression of HER2 (HER2-“negative” breast cancers), 2Present address: Oncology iMED, AstraZeneca, Waltham, MA 02451. though HER2 protein expression is often still present. Over- 3To whom correspondence should be addressed. Email: [email protected]. expression studies have implicated a number of these mutations This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. as activating and oncogenic (4, 7–9). Additionally, one mutation, 1073/pnas.1516853112/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1516853112 PNAS Early Edition | 1of10 Downloaded by guest on September 26, 2021 Results MCF-10A cells can be transformed by the introduction of onco- Targeted Knockin of Single Copy, Heterozygous HER2 Missense genic mutations both in vitro and in vivo (14). HER2 Mutations in HER2 Nonamplified Human Breast Epithelial Cell Lines. We found that MCF-10A V777L cells showed in- To model HER2 missense mutations as found in human cancers, creases in phosphorylation of HER2, EGFR, and ERK com- we used adeno-associated virus (AAV)-mediated gene targeting pared with control cell lines in 0.2 ng/mL EGF (physiologic) A SI Appendix to create an isogenic panel of HER2 mutant knockin MCF-10A (Fig. 2 ) and EGF-free ( ,Fig.S2) conditions. No and MCF7 human breast epithelial cells. These two cell lines do difference in AKT phosphorylation was noted in gene targeted A SI Appendix not overexpress or have amplification of HER2, thus heterozy- clones (Fig. 2 and , Fig. S2), indicating that activation of the HER2 receptor by this mutation preferentially activates gous knockin cell lines contain one wild-type and one mutant HER2 copy of HER2 and express HER2 protein at levels consistent ERK signaling. Interestingly, other mutations did not with HER2 nonamplified tumors (SI Appendix, Fig. S1) (15). consistently activate HER2 signaling in MCF-10A cells. These results demonstrate that HER2 mutations, even those occurring in Three AAV gene-targeting vectors were used as backbone vec- the same domain of the protein, have distinct effects on signaling tors to introduce each of seven previously reported mutations pathway activation. into the extracellular or kinase domains of HER2 (Fig. 1). Two HER2 heterozygous -mutant clones were generated for each HER2 Mutant MCF-10A Cells Do Not Exhibit Oncogenic Properties in HER2 mutation. We also generated wild-type control clones Vitro. Prior overexpression studies identified that HER2 muta- using wild-type AAV vector backbones for both MCF-10A and tions can lead to transformative changes, including increases in MCF7. Cell lines were verified to have a single integrated copy of anchorage independent growth and aberrant morphology in 3D the desired mutation and equivalent expression of the mutant culture, even without detectable increases in HER2 phosphory- and wild-type alleles using PCR and RT-PCR followed by Sanger lation (4, 7). Therefore, we performed a series of assays to identify sequencing (SI Appendix, Fig. S1). See SI Appendix, Table S1 for transformed properties of our HER2 mutant cell line panel. a list of cell lines generated and used in this study. Whereas parental MCF-10A cells require EGF supplementation, EGF independence has been demonstrated to be a feature HER2 V777L Mutation Increases HER2 Signaling Pathway Activation in of transformation in MCF-10A cells (17). However, HER2 mu- Nontransformed MCF-10A Cells. Overexpression studies have iden- tations did not confer EGF independence (SI Appendix, Fig. S2), tified that HER2 mutations, including G309A, S310F, L755S, and there was little difference in proliferation rates between V777L, and R896C activate HER2 signaling and promote trans- control and HER2 mutant cells in assay media containing either formation. To assess whether these effects could be recapitulated 0.2 ng/mL (physiologic) or 20 ng/mL (maintenance dose) EGF in genome edited clones, we initially performed Western blotting (Fig. 2B and SI Appendix, Fig. S2). analysis on our MCF-10A isogenic panel. MCF-10A is a non- We next tested the ability of HER2 mutations to promote transformed human breast epithelial cell line with a mostly diploid anchorage independent growth in MCF-10A cells. Knockin of karyotype that requires EGF supplementation for proliferation in HER2 V777L was not sufficient for colony formation in soft agar, culture (16). Using genome editing, we have demonstrated that although MCF-10A cells that overexpress HER2 V777L mutant

A Vector 3 Vector 1 Vector 2

ECD TM JM Kinase Tail

B C

Reported Alteration Domain Base Change Endogenous Exon Cases Allele

G309A ECD c.926G>C 1 5’HA Targeting 3’HA Exon* SEPT S310F ECD c.929C>T 30 Construct

L755S Kinase c.2264T>C 30 Targeted Exon* SEPT I767M Kinase c.2301C>G 5 Pre-Cre

V777L Kinase c.2329G>T 11 Targeted Exon* Y835F Kinase c.2504A>T 1 Post-Cre

R896C Kinase c.2686C>T 3

Fig. 1. Generation of an isogenic HER2 cell line panel. (A) Previously identified HER2 mutations included in the isogenic panel. Three distinct gene targeting vectors were used to introduce the mutations. ECD, extracellular domain; TM, transmembrane region; JM, juxtamembrane region. (B) Genomic locus and prevalence as reported in the COSMIC database as of January 2015 for each introduced HER2 mutation. (C) Representative rAAV-mediated gene targeting strategy for a targeting vector carrying an exonic mutation (*) within the 5′ homology arm (HA). rAAV transduction leads to integration of the targeting vector via homologous recombination of the 5′ and 3′ HAs. After neomycin selection and clone isolation, loxP (triangles) flanked SEPT cassette is excised using Cre recombinase.

2of10 | www.pnas.org/cgi/doi/10.1073/pnas.1516853112 Zabransky et al. Downloaded by guest on September 26, 2021 Targeted1 WT Targeted2 WT Targeted3 WT B PNAS PLUS A MCF-10A G309A I767M R896C V777L Y835F S310F L755S 64 32 P-HER2 16 Total HER2 8 P-ERK 4 MCF-10A Total ERK 2 Targeted WT V777L P-EGFR 1 Fold Growth Change in 0.5 Total EGFR 2 5 7 9 P-AKT Day Total AKT D Targeted GAPDH MCF-10A WT G309A S310F L755S I767M

C Targeted V777L G309A L755S V777L MCF-10A WT V777L O/E V777L Y835F R896C O/E O/E O/E

Fig. 2. HER2 V777L mutation increases signaling pathway activation but does not promote transformed phenotypes in MCF-10A cells. (A) MCF-10A HER2 mutant and control cell lines were grown in physiologic (0.2 ng/mL) EGF supplemented assay media and subjected to Western blotting analysis with the indicated antibodies. White lines represent lanes that have been removed from blot. (B) Relative mean (±SEM) proliferation data for V777L clones and MCF- 10A controls in 0.2 ng/mL EGF supplemented assay media (n ≥ 6 per cell line, two independent experiments were performed). (C) Representative images depicting soft agar colony formation for MCF-10A V777L, controls, and an MCF-10A cell line overexpressing the HER2 V777L cDNA. (Scale bar, 200 μm.) (D)3D acinar morphogenesis assay for MCF-10A HER2 mutant isogenic cell lines and mutant HER2 overexpression cell lines. (Scale bar, 50 μm.) MEDICAL SCIENCES

cDNA formed robust colonies in soft agar as previously reported media (SI Appendix, Fig. S3). Furthermore, the L755S mutation (4) (Fig. 2C). Anchorage independent growth in semisolid me- increased activation of HER2 signaling pathway proteins in dium was not observed in other MCF-10A HER2 mutant cell MCF7 (Fig. 3A). In addition, MCF7 cells have greater basal lines (SI Appendix, Fig. S2). Next, we examined acinar mor- expression of HER3 than MCF-10A cells, and MCF7 HER2 phology of MCF-10A HER2 mutant knockin cells in 3D culture. V777L and L755S cell lines exhibited an increase in HER3 All HER2 mutant knockin cell lines formed spherical structures phosphorylation (Fig. 3A and SI Appendix, Fig. S3). Notably, with normal polarization similar to controls (Fig. 2D). This was AKT activation did not differ significantly among MCF7 pa- distinct from phenotypes observed with overexpression of HER2 rental, control, and HER2-mutant cell lines. Despite this path- mutant cDNAs, which led to larger, irregular, spiculated struc- way activation, HER2 mutations did not affect proliferation in tures with abnormal protrusions (Fig. 2D). MCF7 cells (Fig. 3B and SI Appendix, Fig. S3).

HER2 V777L and L755S Mutations Increase HER2 Pathway Signaling HER2 Missense Mutations Cooperate with Mutant PIK3CA for Aug- Activation in MCF7 Cells. Whereas biochemical signaling differ- mented Pathway Activation in MCF7 Cells. Because MCF7 HER2 L755S ences were modest in MCF-10A HER2 knockin cell lines, we and V777L cells showed greater signaling activation than their hypothesized that HER2 missense mutations may be highly MCF-10A counterparts, we hypothesized that mutant PIK3CA context dependent and require cooperating genetic alterations E545K in MCF7 may cooperate with HER2 mutations to increase found in cancer cells to promote additional transformative fea- pathway activation and other transformed phenotypes. Numerous tures. Therefore, we created an isogenic panel of HER2 mutant studies have implicated PIK3CA mutations in transformation and MCF7 cell lines containing single copies of HER2 mutations. activation of the PI3 kinase/AKT pathway (17, 20). To test this The parental MCF7 cell line is derived from a metastatic pleural hypothesis, we used genome editing to introduce the HER2 V777L effusion in a patient with ER-positive breast cancer and does not mutation into MCF7 cells that had previously undergone gene overexpresses HER2 protein. Two HER2 kinase domain muta- targeting to restore the PIK3CA alleles to wild type (referred to tions, L755S and V777L, and one extracellular domain mutation, as MCF7 corrected) (19). The resulting cells have three wild- G309A, were chosen for genome editing in MCF7 cells because type copies of PIK3CA and a heterozygous HER2 V777L mutation these mutations have been characterized via overexpression (referred to as MCF7 corrected + V777L). studies to result in increased pathway activation, oncogenic MCF7-corrected + V777L cells displayed slightly lower levels phenotypes, and, in the case of L755S, resistance to lapatinib (4, of phosphorylated HER2 and HER3 compared with MCF7 18). It should be noted that MCF7 contains two copies of an HER2 V777L cells, but relatively more than both MCF7 and activating E545K PIK3CA mutation and one wild-type copy of MCF7-corrected cells (Fig. 3C). These results suggest that the PIK3CA (19), a known oncogene involved in PI3 kinase and HER2 V777L mutation activates the MAP kinase pathway in MAP kinase pathway signaling (17). MCF7 cells and that mutant PIK3CA may augment this signal- We initially tested the effects of HER2 mutations in MCF7 ing, similar to our past studies (17). Interestingly, both MCF7- cells by examining signaling pathway activation. Consistent with corrected and MCF7-corrected + V777L cells had dramatically the results in the MCF-10A background, MCF7 HER2 V777L reduced AKT phosphorylation. Although it has been shown that cells showed increases in phosphorylation of HER2, EGFR, and the absence of PIK3CA mutations in MCF7 cells lowers activa- ERK compared with parental MCF7 and control cells in serum- tion of the PI3 kinase/AKT pathway (19), it was unexpected that starved conditions (Fig. 3A) and HER2 in serum-supplemented isolated activation of HER2 signaling via the V777L mutation

Zabransky et al. PNAS Early Edition | 3of10 Downloaded by guest on September 26, 2021 MCF7 Corrected + V777L 1 MCF7 Corrected + V777L 2 Targeted WT 1 A Targeted WT 2 C lines, though there were no significant differences between HER2 knockin clones and controls (SI Appendix,Fig.S4). G309A 1 G309A 2 L755S 1 L755S 2 V777L 1 V777L 2 MCF7-corrected + V777L mutant cells did not exhibit trans- MCF7 MCF7 Corrected formed phenotypes in in vitro assays compared with controls

Targeted WT (Fig. 4 and SI Appendix, Fig. S4). These data suggest that, whereas PIK3CA mutations may cooperate with HER2 muta- P-HER2 V777L MCF7 tions to accentuate signaling activation in MCF7 cells, this Total HER2 cooperativity is not sufficient to increase anchorage independent growth or promote abnormal acinar morphology. P-HER2 P-ERK Total HER2 Cooperativity Between HER2 and PIK3CA Mutations Leads to In- Total ERK P-AKT creased Signaling Activation in MCF-10A Cells. Because the effects P-EGFR Total AKT of HER2 missense mutations as single copies were greater in Total EGFR P-ERK MCF7 cells compared with MCF7 corrected cells, we wanted to Total ERK P-HER3 confirm that activating E545K PIK3CA mutations in MCF7 Total HER3 P-HER3 HER2 mutant cells contributed to the observed effects. There- Total HER3 fore, we created PIK3CA and HER2 double mutant cell lines by P-AKT GAPDH knockin of the PIK3CA E545K mutation into our MCF-10A Total AKT HER2 L755S and V777L cell lines. These resultant double GAPDH knockin (DKI) cell lines (referred to as L755S DKI and V777L DKI) are heterozygous for their respective HER2 mutation and for the PIK3CA E545K mutation. B D We first analyzed activated signaling pathways via Western MCF7 blot in DKI and control cell lines in media with 0.2 ng/mL or 1000 256 Targeted WT 128 G309A without EGF. It has been previously demonstrated that MCF- 100 64 ** L755S 32 10A cells with knockin of PIK3CA E545K (referred to as MCF- V777L 16 + 10 8 10A E545K) activated both the PI3 kinase and MAP kinase 4 MCF7 pathways compared with MCF-10A parental cells (14, 17). MCF- 1 2 MCF7 Corrected 1 MCF7 Corrected + V777L 10A HER2 V777L cells had increased levels of HER2 phos- Fold Change in Growth in Change Fold Fold Change in Growth in Change Fold 0.1 0.5 phorylation compared with MCF-10A + E545K cells but had 1 3 5 7 1 3 5 7 Day Day similar levels of ERK phosphorylation (Fig. 5). Additionally, both L755S DKI and V777L DKI cell lines showed increases in Fig. 3. HER2 and PIK3CA mutations cooperate to increase signaling path- HER2, EGFR, and ERK phosphorylation compared with MCF- way activation in MCF7 cells. (A) MCF7 HER2 mutant and control cell lines 10A + E545K cells and in ERK activation compared with their were grown in serum starved conditions and subjected to Western blotting ± respective single HER2 mutation knockin cell lines in media analysis with the indicated antibodies. (B) Relative mean ( SEM) proliferation A for MCF7 HER2 mutants in serum supplemented media (n ≥ 6, two independent supplemented with 0.2 ng/mL EGF (Fig. 5 ). Differences in experiments were performed). (C) Western blotting analysis of MCF7 and EGFR phosphorylation were less apparent in EGF-free media MCF7-corrected cell lines and HER2 mutant derivatives as in A. White lines conditions (Fig. 5B). The DKI cell lines also showed slight in- represent lanes that have been removed from blot. (D)Mean(±SEM) relative creases in AKT phosphorylation compared with PIK3CA single proliferation for MCF7, MCF7-corrected, and MCF7-corrected + V777L cells as in knockin cells in EGF-free conditions (Fig. 5B), but not in the ≥ ≤ B (n 5, two independent experiments were performed, **P 0.01, two-way presence of 0.2 ng/mL EGF (Fig. 5A). Interestingly, there was an ANOVA followed by Bonferroni multiple comparison test). increase in the ratio of phosphorylated HER2 to total HER2 in L755S DKI and V777L DKI cells compared with their respective did not lead to any discernible increase in AKT phosphorylation. single knockin cell lines in EGF-free conditions, which was also This result, along with the concurrent increases in ERK phos- present in both L755S DKI and one of two V777L DKI cell lines C phorylation, again suggests that V777L HER2 preferentially ac- in 0.2 ng/mL EGF conditions (Fig. 5 ). Although HER2 is tivates the MAP kinase signaling pathway rather than the PI3 classically considered upstream of PI3K, these results indicate kinase/AKT signaling pathway. MCF7 corrected and its deriva- that under certain conditions, interactions between mutant PIK3CA HER2 “ ” tives also grew similarly to each other, but more slowly than and mutant may lead to rewiring and dysre- MCF7 parental cells in serum-supplemented conditions (Fig. 3D gulation of signaling pathways. Similar pathway dysregulation and SI Appendix, Fig. S3). has been reported in other studies (17, 21, 22). Because PIK3CA mutations have been shown to confer EGF- Acinar Morphology and Anchorage Independent Growth in MCF7 independent growth properties to MCF-10A cells, we tested if HER2 Mutant Cell Lines. To determine any morphologic effects of HER2 mutations could augment this phenotype. L755S and HER2 mutations in MCF7 cells, we conducted Matrigel assays. V777L DKI cells proliferate in EGF-free media, but this growth All cell lines appeared similar to parental and control cells. was not increased compared with MCF-10A + E545K cells (SI When cells were seeded in Matrigel along with the EGFR/HER2 Appendix,Fig.S5A). Furthermore, the combination of HER2 inhibitor lapatinib, there was little to no effect on the size or and PIK3CA mutations did not confer anchorage independent morphology of the acini (Fig. 4A and SI Appendix, Fig. S4). growth to MCF-10A cells, and proliferating colonies were not We then performed soft agar colony formation assays to test observed in soft agar assays (SI Appendix,Fig.S5B). We next the ability of HER2 mutations to promote anchorage inde- seeded the L755S and V777L DKI cell lines in Matrigel to ex- pendent growth in MCF7 cells. Whereas MCF7 V777L clones amine their morphology. Both L755S DKI and V777L DKI cells showed a trend toward increased colony number, this was grew with a morphology similar to controls that was not affected modest relative to controls. Similarly, MCF7 HER2 G309A and by treatment with lapatinib (SI Appendix, Fig. S5C). Taken to- L755S did not differ significantly from controls (Fig. 4 B and C gether, our results suggest that, whereas signaling pathway acti- and SI Appendix, Fig. S4). Treatment with lapatinib had a strong vation was increased in DKI cell lines, this is not sufficient to inhibitory effect on colony size and formation in all tested cell promote transformation in these assays.

4of10 | www.pnas.org/cgi/doi/10.1073/pnas.1516853112 Zabransky et al. Downloaded by guest on September 26, 2021 Targeted PNAS PLUS A MCF7 WT G309A L755S V777L

Vehicle

Lapatinib

2.0 B Targeted C MCF7 WT G309A 1.5

1.0

0.5 Relative colony count V777L V777L 0.0 L755S Clone 1 Clone 2 7 1 2 1 2 2 2 1 F L C A A cted7L 9 5S 77 e M d WT 309 75 7 rr 77 777L 2 te G30G L755SL 1V777LV 1o V e C + V g 7 d + ted te Tar Targeted WT c MCF re orrecor C C 7 F C M MCF7

Fig. 4. Three-dimensional morphology and anchorage independent growth for MCF7 HER2 mutant cell lines. (A) MCF7 control and HER2 mutant cells were seeded into 3D Matrigel culture in the presence of DMSO vehicle control or lapatinib (5 μM) and photographed on day 12. (Scale bar, 100 μm.) (B) Repre- sentative images depicting soft agar colony formation in MCF7 HER2 knockin cell lines and controls. (Scale bar, 200 μm.) (C) Relative mean (±SEM) number of MEDICAL SCIENCES colonies formed in soft agar assays as counted visually via phase contrast microscopy (n ≥ 5 at least two independent experiments were performed).

HER2 V777L and PIK3CA E545K Double Mutant Cell Lines Have In- and L755S DKI cell lines. Treatment with lapatinib significantly creased Migratory Capacity in Vitro. Because HER2 overexpres- reduced the migration of V777L DKI cells (Fig. 6 A and B). L755S sion and activation of HER2 signaling have been described to DKI cells exhibited slower wound closure than the MCF-10A + lead to increases in cell migration, we next tested the migratory E545K cells, indicating that the ability to promote migration may capacity of HER2 mutant cell lines in scratch wound healing and be unique to the V777L mutation. Single V777L knockin cells did chemotactic microchannel migration assays. In scratch wound not migrate more quickly than MCF-10A parental or control cells experiments, MCF-10A V777L DKI cells exhibited significantly (SI Appendix,Fig.S6). In accord with these results, the MCF7 increased wound closure compared with both MCF-10A + E545K HER2 V777L cell line also showed an increase in wound closure MCF-10A + E545K ABMCF-10A + E545K C Targeted WT Targeted V777L DKI 1 V777L DKI 2 V777L DKI L755S DKI 1 L755S DKI 2 L755S DKI Targeted WT Targeted 10A Parent V777L DKI 1 V777L DKI 2 V777L DKI L755S DKI 1 L755S DKI 2 L755S DKI

10A Parent 25

V777L 20 L755S V777L L755S 15

10 P-HER2 P-HER2 5 0 Total HER2 Relative pHER2:HER2 ratio Total HER2 L 1 2 K I 0A 77 1 55S 7 45 K 7 5 DKI 1 DKI D ted WTL V E L P-ERK MCF- e 7L P-ERK A + arg 0 77 T 1 L755SL755SV777 DKI V2 Total ERK Total ERK 8 P-AKT 6 P-AKT Total AKT 4 Total AKT P-EGFR 2 P-EGFR Total EGFR 0 Relative pHER2:HER2 ratio pHER2:HER2 Relative Total EGFR 2 S 7L I I 2 GAPDH 0A WT 5 7 -1 7 K KI 1 F d DKI 1D D DK L75 V E545K L GAPDH C ete + S 7 7L M g 55 ar A T 10 L755SL7 V77 V77

Fig. 5. Knockin of PIK3CA E545K mutation in HER2 missense cells increases signaling activation in MCF-10A. MCF-10A HER2 mutant, L755S DKI, V777L DKI, and control cell lines were grown in physiologic 0.2 ng/mL EGF supplemented (A) or EGF free (B) assay media and subjected to Western blotting analysis with the indicated antibodies. White lines represent lanes that have been removed from blot. (C) Bar graphs represent the ratio of pHER2 to total HER2 Western blot band quantification relative to the MCF-10A parental cell line in 0.2 ng/mL (Top) and EGF-free (Bottom) conditions as measured with ImageJ software.

Zabransky et al. PNAS Early Edition | 5of10 Downloaded by guest on September 26, 2021 A B 0 Hours 16 Hours + LapaƟnib 1.5 Vehicle Lapatinib MCF-10A + 1.0 *** *** E545K *** 0.5 Wound Closed

Relative Fraction of Relative 0.0 L755S DKI K I 5 K 4 DKI D L E5 77 + 7 L755S V 0A 1 F- C M V777L DKI

C Vehicle 20 Hours Lapatinib D 0 Hours 20 Hours + LapaƟnib 0.8 * * 0.6 MCF7 0.4 0.2 Wound Closed

Relative Fraction of Relative 0.0 S V777L 09A 55 3 MCF7 ed WT G L7 V777L et rg a T

E F MCF7 10A + E545K MCF7 10A + E545K V777L V777L DKI 30 V777L 25 0.5 ** V777L DKI * 0.8 20 *** * 0.4 *** ** 20 0.6 15 0.3 0.4 10 0.2 10 Persistence Persistence 0.2 5 0.1 0 0.0 0 0.0 0 0 0 0 0 0 5 2 50 20 5 2 5 2

Fig. 6. Cells harboring both HER2 V777L and PIK3CA E54K mutations have increased migratory capacity in vitro and increased interaction between HER3 and p85. (A) Relative fraction of scratch wound closure (±SEM) as measured after 16 h. Cells were grown to near confluent monolayers, a wound was introduced, and assay media without EGF ± lapatinib (1 μM) was added back to wells (n ≥ 14, at least three independent experiments were performed, ***P ≤ 0.001, one- way ANOVA followed by Bonferroni multiple comparison test). (B) Representative phase contrast images of wound closure. (C) Relative fraction of scratch wound closure (±SEM) for MCF7 and HER2 mutant derivatives in serum-supplemented media ± lapatinib (5 μM) after 20 h (n ≥ 12, *P ≤ 0.05, at least two independent experiments were performed, one-way ANOVA followed by Bonferroni multiple comparison test). (D) Representative images from MCF7 cell line scratch wound closure experiments. (Scale bar, 500 μminB and D.) Mean (±SEM) cell velocity (Left) and persistence (net cell displacement to total distance traveled ratio) (Right) in microchannel migration experiments of MCF-10A + E545K and V777L double knockin cell lines (E) and MCF7 and MCF7 V777L cell lines (F)(n ≥ 30 tracked cells, at least three independent experiments were performed, ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, unpaired t test).

compared with parental control, which was decreased with lapati- MCF-10A V777L DKI cells displayed higher migration nib treatment and not present in MCF7-corrected + V777L cells velocity and persistence (defined as the ratio of net cell dis- (Fig. 6 C and D and SI Appendix,Fig.S6). placement to total distance traveled) inside the 50-μm We next examined the migratory capacity of cells with and channels compared with MCF-10A + E545K cells (Fig. 6E and HER2 – without V777L mutation in collagen-I coated micro- Movie S1). Similarly, MCF7 V777L cells exhibited an increased channels using a microfluidic device constructed of poly- migratory propensity compared with parental MCF7 (Fig. 6F dimethylsiloxane (PDMS) (23–26). This device allows for the and Movie S2). Although both MCF-10A V777L DKI and characterization of single-cell migration, rather than the col- lective migration observed during wound closure. Individual MCF7 V777L migrated faster than their corresponding controls μ cells migrate up a chemotactic gradient inside 10-μm-tall channels in 20- m wide channels, no significant difference was detected E F of prescribed widths. Because all tested cells do not efficiently in persistence (Fig. 6 and and Movies S1 and S2). Taken migrate through narrow channels (10 μm wide or less) we fo- together, these studies demonstrate that HER2 V777L coop- cused our quantitative analysis on cells migrating in 50-μm and erates with the PIK3CA E545K to confer increased migratory 20-μm wide channels (24). potential in breast epithelial cells.

6of10 | www.pnas.org/cgi/doi/10.1073/pnas.1516853112 Zabransky et al. Downloaded by guest on September 26, 2021 HER2 Missense Mutations Do Not Lead to Increased Tumor Growth or HER2 and PIK3CA Mutations Cooperate to Increase Interaction Between PNAS PLUS Invasion in Vivo. The ability of HER2 mutations to increase tumor HER3 and p85. We examined potential interactions in the HER2 formation and xenograft growth was tested. MCF7 HER2 mutant signaling pathway that may be responsible for the cooperative ef- cell lines did not differ significantly in their growth compared fects of HER2 and PIK3CA mutations. Previous work has shown HER2 PIK3CA with controls in the presence or absence of estrogen supple- that overexpression of in cells with mutations en- mentation (Fig. 7 A and B). Additionally, the L755S and V777L hance HER2-mediated signaling through HER3 and p85 (29). DKI MCF-10A–derived cell lines did not form tumors in vivo Using the MCF7 cell line panel, we performed immunoprecipitation assays to examine the effect of HER2 and PIK3CA mutations (Fig. 7C), in contrast to our prior results with mutant KRAS and on HER3/p85 dimerization. MCF7 HER2 L755S and V777L cells PIK3CA double knockin cell lines (14). showed an increase in HER3–p85 interaction relative to HER2 and We performed a tail vein injection assay to test the in vivo PIK3CA wild-type cells (Fig. 7E). Because HER3 levels are sig- invasiveness of MCF7 HER2 mutant cell lines. MCF7 has low nificantly lower in MCF-10A cells, this precluded our ability to metastatic capability and infrequently forms disease sites in the perform immunoprecipitation assays in our MCF-10A cell line HER2 lungs after tail vein injection (27, 28). mutant cells did not panel. Therefore, we overexpressed mutant or wild-type HER2 and form sites of disease in the lungs of nude mice as determined PIK3CA, along with wild-type HER3 in HEK293T cells to examine by gross and microscopic inspection (Fig. 7D and SI Appendix, the consequences of their interactions in another cellular back- Fig. S7). These results suggest additional oncogenic alterations ground. Similar to our results with MCF7 cells, the presence of are required to impart invasive phenotypes to HER2/PIK3CA HER3–p85 dimers increased in cells transfected with both V777L mutant cells. HER2 and E545K PIK3CA compared with wild-type controls

A 1.5 B 0.20 ) ) 3

3 MCF7 MCF7 Targeted WT Targeted WT 0.15 1.0 G309A L755S L755S V777L V777L 0.10 0.5 MEDICAL SCIENCES 0.05 Tumor Volume (cm TumorVolume (cm 0.0 0.00 7 7 5 14 21 28 35 42 14 21 28 3 42 Day Day Tumors Mice per C Cell Line formed group MCF-10A + E545K 0 5 MCF7 Corrected + V777L 1 MCF7 Corrected + V777L 2 10A + L755S DKI 0 10 10A + V777L DKI 0 10 MCF7 Corrected

D E Targeted WT 0.4

V777L 1 V777L 2 0.3 L755S MCF7 MCF7 0.2 WB: 0.1 IP: IP p85:HER3 ratio 0.0 HER3 p85

HER3 777L MCF7 L755SV rected V777L 1% MCF7 Targeted WT Input p85 ected + V777L MCF7 Cor GAPDH 7 Corr F MC

Fig. 7. HER2 mutations do not enhance tumor growth or invasion in vivo. (A)Mean(±SEM) in vivo tumor growth of MCF7 HER2 mutants in mice supplemented with estrogen pellets (n ≥ 5 animals per group, two independent experiments were performed). (B)Mean(±SEM) in vivo tumor growth for MCF7 HER2 mutants in mice without estrogen pellet supplementation. n ≥ 5 animals per group. (C)MCF-10A+ E545K, L755S, and V777L DKI cells do not form tumors as xenografts in nude mice. (D) Representative images of H&E stained lung sections from tail vein injection assays showing benign lung parenchyma and lack of proliferating, multicellular sites of carcinoma. n ≥ 5 animals per group. (Scale bar, 100 μm.) (E) MCF7 parental, MCF7-corrected, and HER2 mutant cell lysates were immunoprecipitated with a HER3 antibody. Antibody pulldowns were then subjected to Western blotting analysis with the indicated primary antibodies. One percent of total protein was used as a control (Left). Bar graphs represent the ratio of p85 after immunoprecipitation to HER3 input control Western blot band as measured with ImageJ software (Right). Bars for MCF7 V777L and MCF7 corrected + V777L represent the average of two clones.

Zabransky et al. PNAS Early Edition | 7of10 Downloaded by guest on September 26, 2021 (SI Appendix,Fig.S7B and C). This increase in HER3–p85 in- mutations in clinical samples are not amplified or overexpressed, teraction in V777L mutant cells may represent a potential signaling our approach using somatic cell gene targeting has led to pre- pathway rewiring leading to increased MAP kinase activation via viously unidentified insights into this issue. However, we recog- HER3/PI3 kinase interaction as we have previously described nize that limitations of our models, including the inability to (30). These results suggest that the V777L HER2 missense mu- analyze immunologic effects of HER2-directed therapies, varying tation in combination with PIK3CA mutations increase the for- protein expression levels, and tumor microenvironment. None- mation of HER3–p85 dimers as a potential mechanism of theless, our isogenic cell line models provide useful tools for increased MAP kinase pathway signaling. understanding the functional consequences of HER2 mutations in isolation and in combination with other oncogenes, as well as HER2 Mutations and Proliferative Response to HER2 Targeted Therapies. testing responses to targeted therapies. Given that overexpression models of HER2 mutations revealed Our study illustrates that different HER2 mutations impart differential sensitivities to HER2 targeted drugs, we measured the distinct phenotypes depending on the individual mutation and IC50 values of two HER2/EGFR tyrosine kinase inhibitors, lapa- the presence of other genetic alterations. This is consistent with tinib and neratinib, on our isogenic HER2 knockin cell line panel recent work demonstrating similar findings for AKT1 and HER3 (SI Appendix, Tables S3 and S4 and Fig. S8) (4, 18). mutations (32, 33) and has implications for predicting response For MCF-10A cells, ∼650–1,000 nM lapatinib was necessary to to targeted therapies. For example, in our study, HER2 muta- inhibit cell proliferation for cells regardless of HER2 and/or tions did not overtly promote cell proliferation. Therefore, it may PIK3CA mutation status (SI Appendix, Table S3). Whereas MCF7 be expected that HER2 targeted therapies would not show dif- cells were relatively insensitive to lapatinib, HER2 mutations again ferential sensitivity to cells harboring these mutations in terms of did not alter the response to lapatinib. Interestingly, cells har- reducing cell number in standard growth assays. On the other boring a HER2 L755S mutation did not exhibit resistance to hand, the V777L HER2 mutation in the appropriate context led lapatinib in terms of proliferation or biochemical pathway acti- to an increase in cellular migration, potentially resulting in in- vation (SI Appendix,Fig.S8). Similar to lapatinib, HER2 muta- creased metastatic potential that was affected by the HER2 in- tions did not predict for increased sensitivity or resistance to hibitor lapatinib. HER2-directed therapies in this regard may neratinib in our models. However, neratinib was slightly more possibly afford clinical benefit. These results underscore the potent in L755S DKI cell lines compared with controls (SI Ap- need for examining multiple aspects of cancer phenotypes when pendix, Fig. S8 and Table S4), though not so for MCF7 HER2 assessing response to targeted therapies. L755S cells. Moreover, all cell lines were relatively insensitive to The variable effects of individual HER2 mutations coupled trastuzumab. Additionally, cell lines with both a PIK3CA and a with a requisite need for oncogene cooperativity may prove HER2 mutation were not more sensitive to the PI3K p110α-spe- challenging for the use of HER2 mutation status to guide therapy cific inhibitor BYL-719 than cells with a single PIK3CA mutation, for patients with breast cancer. Our studies have implicated though cells without a PIK3CA mutation were relatively resistant oncogenic mutant PIK3CA as a potential partner to HER2 mu- to BYL-719 (SI Appendix, Fig. S8 and Table S4). tations, and, indeed, HER2 kinase domain mutations and PIK3CA mutations have been reported concurrently in breast cancer se- Discussion quencing efforts (34). We tested the effects of PIK3CA mutations HER2 mutations in nonamplified/nonoverexpressed breast can- in conjunction with HER2 mutations due to their close relation- cers represent a phenomenon that can potentially be exploited ship as members of the same signaling pathway; however, it is therapeutically. Whereas previous overexpression studies of likely that there are other genes, which when mutated, cooperate mutant HER2 cDNAs have suggested a number of HER2 mu- with HER2 missense mutations to impart transformative effects. tations are activating and promote transformation, our study Indeed, the relatively low frequency of HER2 missense mutations found only the V777L mutation to be of functional significance. may be explained by the need for concurrent oncogenic mutations Additionally, the effects of the HER2 V777L mutation appear to to achieve a selective advantage in HER2-mutant tumors. Addi- be accentuated by a PIK3CA mutation. Consistently across two tionally, it may explain why many HER2 mutations in our isogenic different human breast cell line models, HER2 V777L combined models did not have a detectable phenotype, even though they are with PIK3CA E545K imparted features of increased signaling recurrently found in human cancers. Further work in this regard is pathway activation and migration as noted by changes in West- ongoing and may help elucidate whether mechanisms of cooper- ern blotting, scratch wound healing, and microchannel migration ativity are shared or unique among altered oncogenes and tumor assays. Notably, the HER2 V777L kinase domain mutation led to suppressors and HER2 mutations. an increased interaction between p85 and HER3 in the presence In conclusion, using an isogenic panel of nonamplified/non- of a PIK3CA E545K mutation. These results suggest that HER2 overexpressed HER2-mutant cell lines, we have determined that missense mutations require additional genetic alterations to the majority of HER2 mutations alone are not sufficient to promote features of transformation, leading to an increased in- promote transformative phenotypes in breast cell lines. How- teraction of known signaling partners. In addition, the HER2 ever, HER2 kinase domain mutations, notably V777L, can ex- mutation L755S, which previous overexpression studies have hibit cooperativity with the activating PIK3CA E545K mutation. shown imparts resistance to lapatinib (4, 10, 18, 31) did not show Cells with both HER2 V777L and PIK3CA E545K mutations obvious resistance phenotypes in our genome edited cell lines. exhibited key features of transformation in vitro, including on- Endogenous expression of L755S mutant HER2 in our models cogenic signaling pathway activation and increased migratory may not be sufficient to produce resistance to lapatinib due to potential, but not increased tumorigenicity in vivo. Although the lower levels of the mutant protein. This finding may be of clinical majority of our genome-edited cell lines did not yield an overt importance, because patients that do not have amplification/ phenotype, this has clinically impactful consequences given that overexpression or high expression of L755S mutant HER2 may most cancers with HER2 missense mutations do not have con- still be sensitive to lapatinib, an FDA approved therapy for HER2- current overexpression/amplification of HER2. Indeed, given the amplified breast cancers. current landscape evaluating HER2 targeted therapies in can- Trials are currently ongoing to evaluate the efficacy of tar- cers harboring these mutations, the positive aspect of our find- geting HER2 mutations in HER2 negative cancers with FDA ings and their importance to clinical oncology warrants emphasis. approved therapies. However, given the rarity and variety of Specifically, these findings provide a previously unidentified HER2 mutations, such trials may be underpowered to de- context for the study and clinical significance of HER2 muta- finitively address this question. Because the majority of HER2 tions, in that detailed analysis and study of coexisting mutations

8of10 | www.pnas.org/cgi/doi/10.1073/pnas.1516853112 Zabransky et al. Downloaded by guest on September 26, 2021 in patients may be required to make meaningful patient man- Immunoprecipitation. Cells were grown in serum-starved media conditions PNAS PLUS agement decisions regarding HER2-directed therapies. Taken and were then washed with ice-cold PBS, scraped, and lysed on ice in lysis μ together, our results support that HER2 missense mutations may buffer. A total of 1 mg of protein extract was incubated with 1 g of anti- HER3 antibody (Millipore 05–390) overnight at 4 °C, then with Dynabeads be targetable in the appropriate genetic context. Protein G for immunoprecipitation (Life Technologies) for 4 h at 4 °C. See SI Materials and Methods for details. Materials and Methods Cell Culture. MCF-10A, MCF7, and 293T cell lines were maintained as pre- Scratch Wound Healing Assays. Cells were plated in six-well plates and grown viously described (14). Please see SI Materials and Methods for complete to near confluent monolayers. Scratch wounds were introduced in a cross details. Parental cell lines were authenticated via short tandem repeat pattern with a 200-μL pipette tip. For assays with inhibitors, compounds were profiling analysis at the Johns Hopkins Genetic Resources Core Facility. added after the scratch wound was introduced. Phase contrast images were taken after media was replaced and wounds were followed over time. See SI Gene Targeting and Generation of HER2 Missense Mutation Cell Lines. Gene Materials and Methods for details. targeting was carried out using recombinant AAV vectors as previously described (17). Vectors were generated with primers listed in SI Appendix,TableS2.SeeSI Microchannel Migration Assays. Standard photolithography and replica Materials and Methods for details specific to HER2 gene targeting. molding were used to create the polydimethylsiloxane (PDMS) microfluidic device as previously described (23–26). See SI Materials and Methods for details. Overexpression of HER2, HER3, and PIK3CA cDNAs in Human Cells. HEK-293T cells were transfected using the Fugene 6 system (Promega). pCFG5 plasmids Flow Cytometry and Extracellular Staining. The 1 × 106 cells were treated with containing wild-type, G309A, L755S, or V777L HER2 cDNAs, and LXSN plas- Human TruStain FcX Fc receptor blocking solution (BioLegend) before staining. mids containing wild-type or E545K mutant PIK3CA were derived in previous Data were collected using a FACSCalibur II and analyzed using FACSComp studies (4, 14). A wild-type HER3 expression vector was created by subclon- software (BD Biosciences). Gates and quadrants were set using isotype control ing HER3 cDNA from the pCMV6-XL4-ERBB3 plasmid (SC118918; Origene) staining. See SI Materials and Methods for a list of antibodies. into a pIRES-neo3 backbone. HEK-293T cells were used in assays 48 h after transfection and overexpression was confirmed by Western blot. In Vivo Assays. The 8- to 10-wk-old female athymic nude mice (Harlan Lab- oratories) were used for in vivo assays. For xenografts, mice with or without Cell Proliferation Assays. Assays were performed as previously described (14). estrogen pellet supplementation were injected s.c. in either flank with 200 μL 6 See SI Materials and Methods for more detail. mixture containing 2 × 10 cells in 20% PBS and 80% growth factor reduced Matrigel (BD Biosciences). Tumor volumes were calculated by multiplying Drug Inhibitor Assays. Lapatinib, neratinib, and BYL-719 were obtained from length, width, and height for each individual tumor. For tail vein injection assays, 1 × 105 cells in 200 μL of PBS were injected into Selleck and were dissolved in dimethyl sulfoxide (DMSO). Trastuzumab was MEDICAL SCIENCES obtained from the Johns Hopkins Research Pharmacy and was dissolved in the tail vein of 8- to 10-wk-old female athymic nude mice. After 5 wk, animals bacteriostatic water. The 1–2 × 103 cells per well were plated into 96-well plates were euthanized and their lungs were examined grossly and were then fixed in on day 0. MCF-10A derivatives were grown in assay media containing 20 ng/μL 10% (vol/vol) formalin, paraffin embedded, sectioned, and stained with hema- toxylin and eosin. Five stained sections were examined for evidence of pro- EGF, MCF-10A derivatives with PIK3CA E545K mutation were grown in assay liferating disease per experimental group under a phase contrast microscope. media without EGF, whereas MCF7 and MCF7-corrected derivatives were grown All animal experiments were performed in accordance with institutional in assay media containing 5% (vol/vol) FBS. See SI Materials and Methods and The National Institutes of Health Guide for the Care and Use of Labo- for details. ratory Animals guidelines (36).

Colony Formation Assay in Semisolid Medium. The 5 × 103 exponentially Statistics. All statistical analyses were performed using GraphPad Prism 5 soft- growing cells were plated and experiments were performed as previously ware (GraphPad Software). Significance levels are indicated using one or more described (14). See SI Materials and Methods for details. asterisks: *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001. Error bars represent ± SEM.

Acinar Morphogenesis Assay. Morphogenesis assays were conducted in ACKNOWLEDGMENTS. This work was supported by the Avon Foundation growth factor reduced Matrigel (BD Biosciences) as previously described (35). (B.H.P. and J.L.), NIH CA009071 (to H.A.P., K.C., and B.H.P.), GM007309 (to For assays with inhibitors or DMSO vehicle controls (0.5%), compounds were D.J.Z.), CA168180 (to R.L.C.), CA167939 (to S.C.), CA183804 (to K.K.-S.), added at seeding and media were refreshed every 4 d. Photographs were Conquer Cancer Foundation Young Investigator Award 117451 (to H.A.P.), taken under phase contrast microscopy (Nikon) after 12 d. NIH Grant R01 CA080195 (to C.L.A.), Breast Cancer Specialized Program of Research Excellence (SPORE) P50 CA098131, and Vanderbilt-Ingram Cancer Center Support Grant P30 CA68485, and by the support of NIH P30 CA006973, Immunoblotting. Cells were washed and seeded in the indicated media. After the Sandy Garcia Charitable Foundation, the Commonwealth Foundation, the 48 h, protein lysates were harvested, and immunoblotting was conducted as Santa Fe Foundation, the Breast Cancer Research Foundation, the Health previously described (11). See SI Materials and Methods for details and a list Network Foundation, the Marcie Ellen Foundation, the Helen Golde Trust, the of primary antibodies. Augustine Fellowship (to W.B.D.), and the Robin Page/Lebor Foundation.

1. Arteaga CL, Engelman JA (2014) ERBB receptors: From oncogene discovery to basic 10. Trowe T, et al. (2008) EXEL-7647 inhibits mutant forms of ErbB2 associated with la- science to mechanism-based cancer therapeutics. Cancer Cell 25(3):282–303. patinib resistance and neoplastic transformation. Clin Cancer Res 14(8):2465–2475. 2. Yarden Y, Sliwkowski MX (2001) Untangling the ErbB signalling network. Nat Rev Mol 11. Konishi H, et al. (2007) Knock-in of mutant K-ras in nontumorigenic human epithelial cells Cell Biol 2(2):127–137. as a new model for studying K-ras mediated transformation. Cancer Res 67(18):8460–8467. 3. Ellis MJ, et al. (2012) Whole-genome analysis informs breast cancer response to ar- 12. Lauring J, et al. (2010) Knock in of the AKT1 E17K mutation in human breast epithelial omatase inhibition. Nature 486(7403):353–360. cells does not recapitulate oncogenic PIK3CA mutations. Oncogene 29(16):2337–2345. 4. Bose R, et al. (2013) Activating HER2 mutations in HER2 gene amplification negative 13. Di Nicolantonio F, et al. (2008) Replacement of normal with mutant alleles in the breast cancer. Cancer Discov 3(2):224–237. genome of normal human cells unveils mutation-specific drug responses. Proc Natl 5. Shah SP, et al. (2012) The clonal and mutational evolution spectrum of primary triple- Acad Sci USA 105(52):20864–20869. negative breast cancers. Nature 486(7403):395–399. 14. Wang GM, et al. (2013) Single copies of mutant KRAS and mutant PIK3CA cooperate 6. Lee JW, et al. (2006) Somatic mutations of ERBB2 kinase domain in gastric, colorectal, in immortalized human epithelial cells to induce tumor formation. Cancer Res 73(11): and breast carcinomas. Clin Cancer Res 12(1):57–61. 3248–3261. 7. Greulich H, et al. (2012) Functional analysis of receptor tyrosine kinase mutations in 15. Subik K, et al. (2010) The expression patterns of ER, PR, HER2, CK5/6, EGFR, Ki-67 and lung cancer identifies oncogenic extracellular domain mutations of ERBB2. Proc Natl AR by immunohistochemical analysis in breast cancer cell lines. Breast Cancer (Auckl) Acad Sci USA 109(36):14476–14481. 4:35–41. 8. Rexer BN, et al. (2013) Human breast cancer cells harboring a gatekeeper T798M 16. Soule HD, et al. (1990) Isolation and characterization of a spontaneously immortal- mutation in HER2 overexpress EGFR ligands and are sensitive to dual inhibition of ized human breast epithelial cell line, MCF-10. Cancer Res 50(18):6075–6086. EGFR and HER2. Clin Cancer Res 19(19):5390–5401. 17. Gustin JP, et al. (2009) Knockin of mutant PIK3CA activates multiple oncogenic 9. Wang SE, et al. (2006) HER2 kinase domain mutation results in constitutive phos- pathways. Proc Natl Acad Sci USA 106(8):2835–2840. phorylation and activation of HER2 and EGFR and resistance to EGFR tyrosine kinase 18. Kancha RK, et al. (2011) Differential sensitivity of ERBB2 kinase domain mutations inhibitors. Cancer Cell 10(1):25–38. towards lapatinib. PLoS One 6(10):e26760.

Zabransky et al. PNAS Early Edition | 9of10 Downloaded by guest on September 26, 2021 19. Beaver JA, et al. (2013) PIK3CA and AKT1 mutations have distinct effects on sensitivity 28. Barkan D, et al. (2008) Inhibition of metastatic outgrowth from single dormant tumor to targeted pathway inhibitors in an isogenic luminal breast cancer model system. cells by targeting the cytoskeleton. Cancer Res 68(15):6241–6250. Clin Cancer Res 19(19):5413–5422. 29. Chakrabarty A, et al. (2010) H1047R phosphatidylinositol 3-kinase mutant enhances 20. Gabelli SB, et al. (2014) Activation of PI3Kα by physiological effectors and by onco- HER2-mediated transformation by heregulin production and activation of HER3. – genic mutations: Structural and dynamic effects. Biophys Rev 6(1):89–95. Oncogene 29(37):5193 5203. 21. Masuda H, et al. (2012) Role of epidermal growth factor receptor in breast cancer. 30. Blair BG, et al. (2015) A phosphoproteomic screen demonstrates differential de- Breast Cancer Res Treat 136(2):331–345. pendence on HER3 for MAP kinase pathway activation by distinct PIK3CA mutations. – 22. Hart JR, et al. (2015) The butterfly effect in cancer: A single base mutation can re- Proteomics 15(2-3):318 326. 31. Yang B, Zhang H, Wang H (2015) Atomistic insights into the lung cancer-associated L755P model the cell. Proc Natl Acad Sci USA 112(4):1131–1136. mutation in HER2 resistance to lapatinib: A molecular dynamics study. J Mol Model 21(2):24. 23. Balzer EM, et al. (2012) Physical confinement alters tumor cell adhesion and migration 32. Jaiswal BS, et al. (2013) Oncogenic ERBB3 mutations in human cancers. Cancer Cell phenotypes. FASEB J 26(10):4045–4056. 23(5):603–617. 24. Tong Z, et al. (2012) Chemotaxis of cell populations through confined spaces at single- 33. Yi KH, Axtmayer J, Gustin JP, Rajpurohit A, Lauring J (2013) Functional analysis of cell resolution. PLoS One 7(1):e29211. non-hotspot AKT1 mutants found in human breast cancers identifies novel driver 25. Hung W-C, et al. (2013) Distinct signaling mechanisms regulate migration in un- mutations: Implications for personalized medicine. Oncotarget 4(1):29–34. – confined versus confined spaces. J Cell Biol 202(5):807 824. 34. Network CGA; Cancer Genome Atlas Network (2012) Comprehensive molecular por- 26. Stroka KM, Konstantopoulos K (2014) Physical biology in cancer. 4. Physical cues traits of human breast tumours. Nature 490(7418):61–70. guide tumor cell adhesion and migration. Am J Physiol Cell Physiol 306(2): 35. Debnath J, Muthuswamy SK, Brugge JS (2003) Morphogenesis and oncogenesis of C98–C109. MCF-10A mammary epithelial acini grown in three-dimensional basement membrane 27. Li N, et al. (2011) Siva1 suppresses epithelial-mesenchymal transition and metastasis cultures. Methods 30(3):256–268. of tumor cells by inhibiting stathmin and stabilizing microtubules. Proc Natl Acad Sci 36. Committee on Care and Use of Laboratory Animals (1996) Guide for the Care and Use USA 108(31):12851–12856. of Laboratory Animals (Natl Inst Health, Bethesda), DHHS Publ No (NIH) 85-23.

10 of 10 | www.pnas.org/cgi/doi/10.1073/pnas.1516853112 Zabransky et al. Downloaded by guest on September 26, 2021