Alpelisib Exhibits a Synergistic Effect with Pyrotinib and Reverses Acquired Pyrotinib Resistance in HER2+ Breast Cancer


 Background: Human epidermal growth factor receptor 2 (HER2) plays a vital role in breast cancer progression in patients who overexpress HER2, thus promoting the rapid progress of targeted drugs development and therapy strategies advancement targeting this gene. Pyrotinib, approved in clinical by the Chinese State Drug Administration, is a novel pan-ErbB kinase inhibitor and exhibits better efficacy than lapatinib. Alpelisib is a selective PI3K p110α inhibitor approved for application in HR+, HER2-, PIK3CA mutated breast cancers. We assumed that combining pyrotinib with alpelisib worked better than single-drug treatment.Methods: We performed the drug combination assay to evaluate the combination index (CI) of pyrotinib and alpelisib in HER2+ breast cancer cell lines. Cell functional assays, RT-qPCR (Real Time-Quantitative Polymerase Chain Reaction) and western blotting were performed to elucidate the combined efficacy of two drugs and explore the underlying mechanism. Then we established the acquired pyrotinib resistant HER2+ breast cancer cell lines and evaluate the combined efficacy of two drugs in pyrotinib resistant cells and explore the potential mechanisms.Results: Our data exhibited that a combination of alpelisib and pyrotinib showed a synergistic effect in HER2+ breast cancer by enhancing cell proliferation and migration decrease, G0-G1 cell cycle arrest, and apoptosis rate increase. Additionally, alpelisib combined with pyrotinib showed a tremendous synergistic effect in acquired pyrotinib resistant cells.Conclusions: Our results provided the preclinical evidence that a combination of pyrotinib and alpelisib as an effective therapeutic strategy in treating HER2+ breast cancer, whether patients were sensitive or resistant to pyrotinib treatment.


Background
Breast cancer is the most diagnosed cancer and causes signi cant mortality in women 1,2 . Approximately 15-20% of patients with breast cancer overexpress HER2 (human epidermal growth factor receptor 2) 3-6 , which is relevant to more aggressive progression and worse prognosis. HER2 plays an essential role in regulating cancer cells' proliferation, differentiation, and apoptosis 7 . The relationship between HER2 expression and the aggressive phenotype of HER2+ breast cancer makes this gene appealing for targeted therapies, with an increasing number of drugs being developed to target HER2. Several anti-HER2 drugs are currently available to treat HER2+ breast cancer in clinical. Five targeted drugs, including trastuzumab, pertuzumab, TDM-1, lapatinib, and neratinib, have been approved by the U.S. Food and Drug Administration (FDA) to treat HER2+ breast cancer [8][9][10][11][12] . Pyrotinib, a novel drug that targets the pan-ErbB receptors, exhibits an excellent anti-tumor activity against HER2+ breast cancer and has been approved by the Chinese State Drug Administration for clinical use recently [13][14][15] .
Pyrotinib is a novel oral, irreversible pan-ErbB receptor tyrosine kinase inhibitor (TKI), which shows inhibitory activity against epidermal growth factor receptor (EGFR)/HER1, HER2, and HER4 13 . Clinical trial data revealed the excellent anti-tumor activity of pyrotinib in treating HER2+ breast cancer. In a randomized, phase II trial, the overall response rate in tested patients was 78.5% when treated with pyrotinib compared to 57.1% treated with lapatinib. The median progression-free survival (PFS) was 18.1 months with pyrotinib compared to 7.0 months with lapatinib. Besides, patients showed better tolerance to adverse reactions when treated with pyrotinib 16 .

Cell proliferation assay and colony formation assay
Cells were seeded in a 96-well plate and cultured overnight. Then cells were added with DMSO, pyrotinib, alpelisib, or combinations. Every 24h, cells were added with CCK-8 and incubated for 3h, then OD450 was measured. The experiment was performed three times, and the cell proliferation curves were drawn.
Cells were cultured in a 6-well plate at a density of 3000 cells/well, then cells were added with different treatments the other day. The culture medium containing corresponding drugs was renewed every three days, and cells were cultured for three weeks. Finally, cells were xed with 4% paraformaldehyde and stained with 0.1% Crystal Violet. The colony (>50 cells/ colony) was counted and compared to the controlled group. The experiment was performed in triplicate, and the data were representative of three separate experiments.

Transwell assay
Cells (10^5) suspended in 300 µL serum-free medium containing different drug treatments were transferred into the upper chamber of 24-well transwell chambers (#3422, Corning, NY, USA), and 600μL medium containing 20% FBS was added to the lower chamber. After incubation for 24 h, cells crossed the membrane were xed with 4% PFA and stained with 0.1% Crystal violet. For each sample, 5 elds of view were obtained, and cell numbers were counted. The experiment was performed in triplicate, and the data were representative of three separate experiments.

Cell cycle assay
Cells were starved for 24h and treated with the corresponding treatments for 72h, and then cells were harvested and xed with 75% ethanal at -20℃ overnight. Fixed cells were washed with PBS, resuspended in 100μL PBS, stained with 20μL 7-AAD in darkness for 15min at room temperature, and nally measured using a ow cytometer (BD Accuri C6 Plus, USA) according to the manufacturer's instruction. The data were analyzed using Flowjo software (v.10.5; TreeStar, CA, USA). The nal data were obtained from three independent experiments.

cell apoptosis assay
Cells were obtained after 72h of treatment and washed twice with PBS. Then cells were re-suspended in 100μL 1× binding buffer (BD Biosciences, NJ, USA), added with 5μL of Annexin V-FITC and 7-AAD and incubated in darkness for 15 minutes at room temperature. Samples were added with another 400μL 1× binding buffer before being measured with a ow cytometer (BD Accuri C6 Plus, USA). The data were analyzed using Flowjo software (v.10.5; TreeStar, CA, USA).
Tunel assays were performed according to the manufacturer's instructions (Beyotime, Shanghai, China). Cells were plated in a 24-well plate and treated with different treatments for 72 hours. Then cells were xed with 4% PFA for 30 minutes and permeabilized with 0.3% Triton-X-100 for 5 minutes. Cells were incubated with terminal deoxyribonucleotidyl transferase (TdT) at 37℃ in darkness for an hour and added with uorescent mounting media (Beyotime, Shanghai, China). Pictures were taken by using DMi8 (Leica, German).
All assays were performed in triplicate, and the data were obtained from three separate experiments.

NA extraction and quantitative real-time PCR (qRT-PCR)
Total RNA was extracted with Trizol reagent (Thermo Fisher Scienti c, Waltham, MA, USA) following the manufacturer's instructions. RNA was reverse-transcribed with ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan

Protein extraction and western blot analysis
Total protein was extracted using Cell lysis buffer for Western and IP (Beyotime, Shanghai, China) supplemented with proteinase and phosphatase inhibitors cocktail tablet (Pierce, Thermo, USA). Protein quanti cation was performed using the BCA reagent (Pierce, Thermo, USA) according to the manufacturer's instructions, and then samples were denatured with 5× loading buffer at 100℃ for 5min.
The extracted protein samples were loaded into 8% -15% SDS-PAGE gel for separation and transferred to 0.45μM polyvinylidene uoride (PVDF) membrane (Millipore, Massachusetts, USA). The membrane was blocked with 5% skim milk or BSA for an hour. Primary antibodies were added, and the membrane was incubated at 4℃ overnight. Subsequently, the membrane was added with uorescent secondary antibodies (LI-COR Biosciences, USA) and incubated in darkness for an hour at room temperature. The immunoreactive bands were obtained and analyzed using the LI-COR Odyssey CLx (LI-COR Biosciences, USA). The primary and secondary antibodies used are enlisted in Supplementary Table 2.

Statistical analysis
Student's t-tests and Analyses of variance models were used to compare mean values between the tested and controlled groups. Statistical analyses were performed with IBM SPSS version 22 (SPSS, NY, USA) and GraphPad Prism version 7 (GraphPad Software Inc., CA, USA). Results were presented as mean ± SD of three independent tests. The statistical signi cance of the difference between tested and controlled groups was assessed at signi cance thresholds of * (P < 0.05), ** (P < 0.01), and *** (P < 0.001).

Alpelisib showed a synergistic effect when combined with pyrotinib in treating HER2+ breast cancer cells
We rst veri ed the ErbB, HR expression, and PIK3CA mutation situations in four breast cancer cell lines (BT-474, SKBR-3, MCF-7, and T47D). We found that BT-474 and SKBR-3 overexpressed HER2, while BT-474, T47D, and MCF-7 had mutations in PIK3CA (Table 1). We also examined the protein and mRNA expression of EGFR, HER2, and PI3K p110α (Fig. 1A-1C). Then we evaluate the inhibitory effects of pyrotinib and alpelisib on four cell lines. The result indicated that three cell lines were sensitive to alpelisib except for MCF-7 (Fig. 1E). Only BT-474 and SKBR-3 were sensitive to pyrotinib (Fig. 1D). The IC50 values of four breast cancer cells to different treatments were presented in Table 1. To examine the combined effect of pyrotinib and alpelisib in HER2+ breast cancer cells, we exposed BT-474 and SKBR-3 cells to two drugs individually or in combination with different concentrations and calculated the combination index. The result indicated that a combination of pyrotinib and alpelisib showed a synergistic effect in treating HER2+ breast cancer cells, whether cells had mutations in PIK3CA (Fig 1F,  1G). Table 1 The molecular and gene signature of 4 breast cancer cell lines and their IC50 to pyrotinib or alpelisib treatment Cell Line HER2 Protein Expression [1] PIK3CA mutation situation [3] HR Protein Expression IC50 (72h) [ 3.2 Alpelisib combined with pyrotinib enhanced the inhibitory effect on cancer-associated phenotypes signi cantly We rst exposed BT-474 and SKBR-3 cells to increasing concentration of pyrotinib (DMSO, 3nM, 6nM, 9nM) and alpelisib (DMSO, 0.1uM, 0.2uM, 0.3uM) for 24h and detected the protein expression of ErbB/PI3K/AKT pathway through western blotting (Fig. 1H, 1I). We determined 6nM of pyrotinib and 0.2μM of alpelisib for further experiments, then we exposed BT-474 and SKBR-3 cells to two inhibitors, individually or in combination and detected cancer-associated phenotypes. The combination therapy signi cantly decreased the cell proliferation, cell clone formation and invasion, increased cell apoptosis rate, and arrested the cell cycle to the G0-G1 stage (Fig 1J, 1K, Fig 2A -2J).
3.3 The combination of pyrotinib and alpelisib could enhance the downregulation of the PI3K/AKT pathway activation and strengthen the activation of the apoptosis pathway We detected the protein levels of ErbB and downstream PI3K/AKT signaling pathways. We observed that combined pyrotinib with alpelisib signi cantly reduced PI3K/AKT signaling pathway activation (Fig 3A,  3C, 3D). Besides, we detected the protein expression levels related to cell apoptosis. The result showed that the apoptosis-related proteins were signi cantly activated when treating cells with pyrotinib and alpelisib (Fig 3B, 3E, 3F).

HER2+ breast cancer cells up-regulated ErbB ligand expression to counter alpelisib treatment
To investigate the potential mechanism behind the synergistic effect between pyrotinib and alpelisib, we treated BT-474 and SKBR-3 cells with elevated concentrations of alpelisib (DMSO, 0.05uM, 0.5uM, 5uM). We detected the expression of total EGFR, HER2, and their phosphorylation forms through western blotting. We observed that cancer cells expressed higher activated HER2 and EGFR when treated with alpelisib than controlled groups (Fig 4A-4D). Some underlying mechanisms exist to activate the ErbB receptors to counter alpelisib treatment and cells are more sensitive to pyrotinib. Receptors are activated when combined with their ligands. We hypothesized that the expression of ErbB ligands or proteins that activate these ligands might be up-regulated when cells were treated with alpelisib. Therefore, we detected the mRNA expression levels of two Metallopeptidase (ADAM10 and ADAM17) and seven ErbB ligands (EREG, AREG, EPGN, TGF-α, BTC, EGF, HB-EGF) expression levels through qPCR. When treated with alpelisib, BT-474 cells expressed signi cantly higher levels of HB-EGF, AREG, EPGN, and BTC (EREG, TGF-α, and BTC in SKBR-3 cells) (Fig 4E, 4F). The result could explain the higher levels of activated HER2 and EGFR when cells were treated with alpelisib. The schema chart of potential mechanism was shown in Fig 6A. 3.5 HER2+ breast cancer cells acquired pyrotinib resistance by down-regulating phosphorylation of ErbB proteins and escaped target by pyrotinib To investigate the potential resistance mechanism of pyrotinib in HER2+ breast cancer cells, we constructed a pyrotinib-resistant cell line, SKBR-3-PR. The IC50 of SKBR-3-PR to pyrotinib was 104.97 ± 24.20 nM compared with 4.83 ± 0.88 nM in SKBR-3 cells (Fig. 5a). Then we detected the mRNA and protein expression of HER2, EGFR. At the mRNA level, the expression of ErbB2 was not signi cantly changed, and the expression of EGFR was up-regulated in SKBR-3-PR cells compared to SKBR-3 cells ( Fig   5D). The protein expression of HER2 and EGFR were not changed; however, their phosphorylated forms were signi cantly down-regulated in SKBR-3-PR cells relative to SKBR-3 cells (Fig 5E, 5F). The results meant the acquired pyrotinib resistant cells were less dependent on ErbB receptors activation and counted more on other pathways and induced cells resistant to pyrotinib treatment.
Though less effective than inhibiting p110α, alpelisib showed an inhibitory effect on p110δ. We combined pyrotinib with alpelisib to treat SKBR-3-PR cells and observed that these two drugs exhibited a synergistic effect in pyrotinib resistant cancer cells and reduced cell proliferation, enhanced cell apoptosis, and arrested the cell cycle to the G0-G1 stage (Fig. 5C, K-Q).
In summary, SKBR-3-PR cells up-regulated the expression of p110δ to compensate for p110α downregulation and activate the downstream AKT signaling pathway. Alpelisib could inhibit p100δ in high concentrations. Thus, alpelisib and pyrotinib showed a synergistic effect in treating acquired pyrotinib resistant breast cancer cells. The schema chart of potential mechanism was shown in Fig 6B.

Discussion
Adverse reactions are the main reasons for drug withdrawal, which is often relevant to high doses. To lower the dose patients needed to reduce the possibility of adverse reaction but with no e cacy reduction, patients are often treated with several drug combinations. Besides, drug combinations are used to delay or reverse the acquired drug resistance. It is crucial to combine pyrotinib with other anti-tumor drugs to extend the validity of pyrotinib when treating patients.
It has been reported that the PI3K-AKT pathway is one of the most crucial downstream signaling pathways of the ErbB receptor. In theory, simultaneously inhibition of upstream signaling molecular ErbB and the downstream signaling molecular p110α overlap their anti-tumor activities and may not show a synergistic effect when treating HER2+ breast cancer. ADAM (a disintegrin and metalloprotease) proteins are a class of membrane-anchored metalloproteases that process the ectodomains of membraneanchored growth factors, cytokines and receptors. Among many ADAM proteins, ADAM10 and ADAM17 were reported to show proteolysis function and release ErbB ligands anchored to the membrane 26,27 . EGF, HB-EGF, AREG, EREG, EPGN, TFG-α, BTC are the main ErbB ligands that activate HER2 and EGFR 28-31 . Our data showed that HER2+ breast cancer cells exposed to alpelisib treatment alone up-regulated the expression of ErbB ligands, which had a feedback activation of the ErbB receptors. Thus HER2+ breast cancer cells treated with alpelisib were more sensitive to ErbB inhibitor, which can explain the synergistic effect of alpelisib and pyrotinib.
HER2+ breast cancer acquires resistance to anti-HER2 therapy through several different mechanisms. HER2 mutation can induce continuous self-phosphorylation of HER2 or escape distinguished by targeted inhibitors, thus activating the downstream signaling pathway. In some cases, PIK3CA mutation or alternative upstream pathway activation induces continuous activation of the PI3K/AKT signaling pathway. Besides, tumor microenvironment and cell epigenetics change can induce cancer resistance to anti-HER2 therapy [32][33][34][35][36][37] . To investigate the potential mechanisms that contribute to pyrotinib resistance and explore the treatment strategies, we compared the differences in acquired pyrotinib resistant cells and their parent cells. In our study, pyrotinib resistance HER2+ breast cancer cells reduced the activation of HER2 and EGFR signi cantly, which could escape the anti-ErbB therapy and account for the low sensitivity of cells to pyrotinib treatment. P110δ, the isoforms of P110α, can also bind to PI3K regulatory subunit P85α to activate the downstream AKT. P110δ was reported to be predominantly expressed in white blood cells 38,39 . However, evidence revealed that p110δ is also expressed in some cancer cell lines and human tissues such as breast cancer cells and it plays a vital role in breast cancer 40,41 . In our study, though acquired pyrotinib resistant cells expressed lower levels of p110α, no signi cant reduction of the activation of downstream AKT was observed, which meant there might be some alternative molecular exist to compensate for the down-regulation of p110α. We observed that p110δ was signi cantly upregulated in pyrotinib resistant cells. Besides, we observed a signi cant mTOR activation in pyrotinib resistant cells compared to the sensitive ones, which meant there might be other mechanisms upstream of the mTOR that were strengthened to activate mTOR. As a result, activation of AKT/mTOR signaling via the P110δ up-regulation and other signaling pathways activation enhanced the proliferation and survival of pyrotinib resistant cells.
This study has limitations. Though our ndings are compelling, our research is based on in-vitro models.
In-vivo assays should be done to verify the synergistic effects observed in cell experiments. Besides, more different pyrotinib resistant breast cancer cell lines need to be established and clinical studies need to be conducted to verify the results and the mechanism observed in this study. Furthermore, clinical trials are required to establish the safety and e cacy of drug combinations in patients.

Conclusions
In conclusion, our study revealed that pyrotinib exhibited synergistic anti-tumor effects with alpelisib in HER2+ breast cancer. Besides, in acquired pyrotinib resistant HER2+ breast cancer, alpelisib also showed signi cant synergistic antitumor effects with pyrotinib and could reverse drug resistance. These results suggest that the combination of pyrotinib and alpelisib is an effective treatment strategy that can be applied for HER2+ breast cancer patients, whether sensitive or resistant to pyrotinib treatment.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All data generated for this study are included in the article.

Competing interests
No author has nancial or other contractual agreements that might cause con icts of interest.

Author contributions
Hao Chen performed study concept, design and writing; Yuhao Si, Jialiang Wen and Chunlei Hu provided acquisition, analysis and interpretation of data, and statistical analysis; Erjie Xia, Yinghao Wang, and Jizhao Niu provided technical and material support; Ouchen Wang performed the development of methodology, review, and revision of the paper.      Supplementary Files