Background
The HER family of receptor tyrosine kinases (RTKs) includes EGFR/HER1/ErbB1, HER2/ErbB2, HER3/ErbB3, and HER4/ErbB4 [
1,
2]. Except for HER4, the aberrant activation of HER receptor kinase activity contributes to the tumorigenesis and progression of breast cancer [
3‐
11]. Overexpression of EGFR, HER2 and HER3 occurs in 30–40%, 20–30% and ~ 20% of breast cancer cases, respectively [
4,
11‐
16]. Targeting HER2 has proven to be an effective therapeutic strategy for HER2-positive breast cancer [
17,
18]. Since its approval by FDA in 1998, trastuzumab, an antibody against HER2, has changed the paradigm for the treatment of HER2-positive breast cancer [
18,
19]. However, after the initial success, acquired resistance to trastuzumab has gradually developed, which posts a challenge that needs to be overcome [
18,
20,
21].
The activation of HER receptors are induced by homo- or hetero-dimerization [
2,
22,
23]. Among HER receptors, HER2 is an orphan receptor without a direct ligand and HER3 has impaired kinase activity. The heterodimerization among various HER receptors is an important mechanism to activate all HER receptors in response to ligand stimulation [
2,
15,
24,
25]. The HER2 extracellular domain is always in the extended conformation and ready to be dimerized. Therefore, HER2 is the preferred heterodimeric partner for other HER receptors [
2,
26‐
28]. Overexpression of HER2 in cancers leads to the homodimerization and the constitutive activation of HER2 [
15]. Each HER receptor displays different binding affinities for various downstream signaling proteins. While EGFR and HER2 preferentially activate the Ras-ERK pathway leading to cell proliferation HER3 preferentially activates the PI3K-AKT pathway leading to cell survival [
15,
29]. The heterodimerization among various HER receptors allows them to play a flexible and complex roles in cell signaling [
2,
23‐
25,
29‐
39].
HER2 has been a therapeutic target for treating breast cancer due to its overexpression in 20–30% of breast cancer patients [
6,
8,
11,
40]. Trastuzumab is a recombinant humanized monoclonal antibody that binds to the juxtamembrane region of HER2 [
27,
41,
42]. Trastuzumab is the first HER2-targetted therapy approved by FDA for metastatic breast cancer treatment. It showed strong antitumor effects in both mouse model and HER2-positive breast cancer patients [
6,
8].
While many mechanisms have been proposed for the antitumor activity of trastuzumab, including both extracellular and intracellular actions [
6,
8,
43], the exact mechanisms are not known. The extracellular action is through immune-mediated response. When bound to the target cells, the Fc portion of trastuzumab will be recognized and attacked by Fc receptor on immune effector cells, principally natural-killer (NK) cells. In vitro, this process is called antibody-dependent cellular cytotoxicity (ADCC). There are solid evidence to support ADCC as a major mechanism for trastuzumab action [
44‐
51].
On the other hand, the data regarding the intracellular mechanisms are either controversial at the beginning or challenged by the recent data [
52]. Intracellular action could be through the following mechanisms: inhibition of intracellular signal transduction, stimulation of HER2 internalization and degradation, inhibition of DNA repair, inhibition of proteolytic cleavage of the HER2 extracellular domain, and inhibition of angiogenesis [
6,
8,
43]. While many recent publications claim that early studies support the role of trastuzumab in inhibiting HER2 phosphorylation [
6,
52,
53], many data indicate that trastuzumab either has no effect or stimulates HER2 phosphorylation [
52‐
56]. The data regarding the effects of trastuzumab on the dimerization of HER2, activation of major signaling pathways including AKT and ERK [
6,
8,
43,
57,
58], and HER2 endocytosis/downregulation [
56,
59‐
61] are all controversial. The data regarding the role of trastuzumab on DNA repair [
62], proteolytic cleavage of HER2 extracellular domain [
63], and angiogenesis [
64,
65] are very limited [
6].
The most controversial mechanism regarding trastuzumab function is its effect on the inhibition of HER2 activation. A major reason behind this controversy is the different cellular background of various breast cancer cells lines used in those studies. Each breast cancer cell line has a unique expression profile of various HER receptors, which could significantly affect the effects of trastuzumab. To overcome this problem, in this research we adopted a cell model that allow us to specifically examine the effects of trastuzumab on a single HER receptor without the influence of other HER receptors. We aim to conclusively determine if trastuzumab specifically binds only to HER2, and blocks HER2 homodimerization and activation. To achieve our objective, we adopted a CHO cell model. Besides the parental CHO cells that do not express any detectable HER receptors, three stable CHO cell lines that stably express only a single HER receptor including EGFR (CHO-EGFR), HER2 (CHO-K6), and HER3 (CHO-ErbB3) were employed in this research. We demonstrate that overexpression of HER2 in CHO cells resulted in the homodimerization of HER2 and the phosphorylation of HER2 at all major pY residues. Trastuzumab bound to HER2 specifically and with high affinity. Trastuzumab neither inhibited the homodimerization of HER2, nor inhibited the phosphorylation of HER2 at most phosphotyrosine residues. Moreover, trastuzumab did not inhibit the phosphorylation of ERK and AKT in CHO-K6 cells, and did not inhibit the proliferation of CHO-K6 cells. However, trastuzumab induced strong ADCC in CHO cells overexpressing HER2.
Methods
Cell culture and treatment
All cells were cultured at 37 °C with 5% CO2 atmosphere. Chinese Hamster Ovary (CHO) cell was purchased from ATCC (ATCC® CCL-61™, Manassas, VA). CHO cell stably expressing human EGFR (CHO-EGFR) was previously generated [
66]. CHO cell stably expressing HER2 (CHO-K6) [
67], and HER3 (CHO-HER3) [
68] were gifts from other labs. Dulbecco’s modified Eagle’s medium (DMEM) containing 10% FBS were used for cell culture. The cells were maintained at 37 °C in a 5% CO
2 atmosphere. To maintain the selection pressure G418 (500 mg/ml) were added to the culture medium. For treatment, cells were starved overnight at DMEM containing 1% FBS and then treated in this starvation medium. EGF, trastuzumab, normal IgG, CP-714724, or vinorelbine was added at indicated concentration for indicated time periods.
Chemicals and antibodies
CP-724714 HER2 inhibitor was purchased from Selleckchem (Houston, TX, USA). Lapatinib and isotype control human IgG were purchased from Sigma-Aldrich (St. Louis, MO, USA). Trastuzumab was purchased from Roche (Basel, Switzerland). Mouse monoclonal anti-human HER2 (9G6) and (A-2), anti-human EGFR (A-10), and anti-human HER3 (RTJ.2) antibodies were purchased from Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Rabbit polyclonal anti-human phospho-HER2 Y-1005, Y-1112, Y-1127, Y-1139, Y-1196, and Y-1248 antibodies were purchased from FroggaBio (Toronto, ON, Canada). All other chemicals were purchased from Sigma-Aldrich.
Cell proliferation assay by MTT
Cell proliferation was determined by MTT assay Vybrant MTT Cell Proliferation Assay Kit from Invitrogen (Grand Island, NY). The detailed protocol of the assay was described previously [
69]. The cells were treated with various agents including EGF, trastuzumab, normal IgG, CP-714724, or vinorelbine for 24 or 48 h. Each value is the average of at least three independent experiments.
Cell lysates and immunoblotting
The lysis of the cells and the immunoblotting were described previously [
66].
Cross-linking assay
Cross-linking assay was employed to determine the dimerization of the receptors. CHO cells were cultured to subconfluency in 60 mm dishes. Following the treatment with EGF, trastuzumab, and normal human IgG of indicated concentrations for 1 h at 37 °C, the cells were collected and suspended in 0.2~ 0.5 ml PBS. BS3 [bis(sulfosuccinimidyl)suberate] was then added to a final concentration of 1.0~ 2.5 mM. The cross-linking reaction was conducted on ice for 2 h. To stop the reaction the quench solution (1 M Tris, pH 7.5, 1:100 dilution) was added and incubated for 15 min on ice. The final concentration of the quench solution was 10 mM. Afterwards, the cells were lysed with 1% NP-40 on ice for 1 h. The dimerization was analyzed by SDS-PAGE followed by immunoblotting. A gel of 5% was run to better separate the dimers from the monomers.
Immunofluorescence staining assay
Cells were cultured on the immunofluorescence slides 48 h before treatment starts. After treatment period, the slides were rinsed in tris-buffered saline (TBS; 6% tris, 8.8% NaCl, 85.2% dH2O, PH = 7.6) and the cells were fixed by cold methanol for 4 min. Blocking was done with incubation of slides in 1% BSA in TBS for an hour. The slides were then treated with 1 μg/ml indicated primary antibodies in TBS for 1 h. Following rinsing in TBS for three times, the slides were incubated with 1 μg/ml FITC- and/or TRITC-conjugated secondary antibody in TBS for an hour in the dark. Thereafter, the slides were washed completely in TBS and incubated in 1 μg/ml DAPI in TBS for 5 min at room temperature in the dark. The slides were observed using a DeltaVision fluorescence microscopy system (Applied Precision Inc., Mississauga, ON, Canada).
Antibody-dependent cellular cytotoxicity (ADCC)
ADCC of trastuzumab in CHO cells expressing HER2 or EGFR was determined by using Promega ADCC Bioassay kit according to Manufacturer’s instruction. Cultured cells were plated at the density of 15,000 cells per well in complete culture medium overnight before bioassay. On the day of bioassay, the series of concentrations of trastuzumab were added to the cells, followed by addition of ADCC Bioassay Effector Cells. The E:T ratio was 5:1. After 6 h of induction at 37 °C, Bio-Glo™ Luciferase Assay Reagent was added and luminescence was determined.
Discussion
The most controversial mechanism regarding trastuzumab function is its effect on the inhibition of HER2 activation. A major reason behind this controversy is the different cellular background of various breast cancer cells lines used in those studies. Each breast cancer cell line has a unique expression profile of various HER receptors, which could significantly affect the effects of trastuzumab due to the heterodimerization among HER receptors. In this research we adopted a CHO cell model. Besides the parental CHO cells that do not express any detectable HER receptors, three stable CHO cell lines that stably express only a single HER receptor including EGFR (CHO-EGFR), HER2 (CHO-K6), and HER3 (CHO-ErbB3) were employed in this research. Our cell model system avoided the interference of other HER receptors, and is very suitable to study the effects of trastuzumab on the homodimerization of HER2 and the phosphorylation of HER2 homodimers. We aim to conclusively determine if trastuzumab specifically binds only to HER2, and blocks HER2 homodimerization and activation.
We showed that trastuzumab only bound to HER2 specifically and with high affinity. Trastuzumab did not bind to EGFR and HER3 even at high dosage (10 ng/ml) (Fig.
2). Most HER2-positive breast cancer cells also express EGFR and HER3, our finding suggest that any trastuzumab effects on these cells must be initiated through the interaction between trastuzumab and HER2.
We next examined the effects of trastuzumab on HER2 dimerization. HER2 is an orphan receptor and does not have a ligand. However, HER2 are heterodimerized with EGFR in response to EGF stimulation and heterodimerized with HER3 in response to HRG [
1]. HER2 is also homodimerized when overexpressed in cells. CHO-K6 cells only expresses a single HER receptor HER2, not EGFR, HER3 or HER4. Thus our results are regarding the effects of trastuzumab on the homodimerization of HER2.
We showed that in CHO-K6 cells HER2 was mostly dimerized, likely due to the overexpression (Fig.
3). This is not surprising. As revealed by crystal structures of the HER2 extracellular region, HER2 adopts an extended configuration, which resembles the configuration of EGFR seen in each molecule of an EGFR dimer. Thus, ErbB2 possesses a constitutive, or ligand independent, activated conformation, which allows the HER2 homodimerization when overexpressed [
1,
27,
72].
We also showed that trastuzumab did not block the homodimerization of HER2 (Fig.
3). While it is originally proposed that trastuzumab acts to block HER2 dimerization, so far, no research has been done to determine the effects of trastuzumab on the homodimerization of HER2. Given the fact that trastuzumab binds to the juxtamembrane region of HER2 [
27], which is not essential for HER2 dimerization, our results are not surprising. What surprising is that our data suggest that trastuzumab at high dosage actually enhanced the homodimerization of HER2 (Fig.
3). While we are not certain how trastuzumab stimulates the homodimerization of HER2, it is possible that it functions through HER2 transmembrane domain. Many data support the role of HER2 transmembrane domain in HER2 dimerization and activation [
27]. Parts of juxtamembrane region has also been implicated in HER2 dimerization and activation [
73‐
75]. As trastuzumab binds to the extracellular juxtamembrane region of HER2, it will likely affect the function of HER2 transmembrane domain and juxtamembrane region in terms of HER2 dimerization. It is possible that somehow the specific effects of trastuzumab enhanced the interaction between two HER2 transmembrane domains and thus increased HER2 homodimerization as we observed here.
It has been believed that trastuzumab functions to inhibit HER2 activation/phosphorylation and HER2-mediated cell signaling [
6,
52,
53]. However, our data indicated that trastuzumab only had very limited effects on HER2 phosphorylation. Among six pY residues examined in this research, HER2 had no effects on the phosphorylation of pY1005, pY1112, pY1027, pY1196, and pY1248 (Figs.
5 and
6). While HER2 decreased the phosphorylation of pY1139, which is a much weaker inhibition when compared with CP-724714 (Figs.
5,
6 and
7). In general, this is consistent with our observation regarding the role of trastuzumab in HER2 dimerization. Trastuzumab did not block HER2 dimerization, thus it did not block HER2 phosphorylation. It is not clear how the effects of HER2 transmembrane domain on HER2 dimerization affect the phosphorylation of HER2. Some research indicated the presence of an alternative dimerization mode of HER2. In this mode, HER2 dimerization is mediated by both transmembrane domain and the cytoplasmic juxtamembrane region of HER2. Such a dimerization mode exert inhibiting effects on the HER2 kinase activity [
73‐
75]. Thus, in theory, the enhanced dimerization through the interaction of transmembrane domain and the juxtamembrane region could result in the inhibition of certain HER2 phosphorylation including pY1139. Recently, some researches with various breast cancer cell lines have shown that trastuzumab did not significantly alter HER2 phosphorylation [
53‐
56,
76]. Moreover, there is one research shows the enhanced phosphorylation of pY1248 in response to trastuzumab [
52].
Our results suggest that trastuzumab has, if any, limited effects on HER2-mediated intracellular signaling. Indeed, when we examined the effects of trastuzumab on the phosphorylation of ERK and AKT, we showed that trastuzumab did not block the phosphorylation of both ERK and AKT in CHO-K6 cells (Fig.
7). Together, our data indicate that trastuzumab did not significantly alter HER2 activation and HER2 mediated intracellular signaling in the absence of other HER receptors. However, we need to be cautious to apply these findings to breast cancer cells. CHO cell is derived from hamster ovary, thus the expressed human HER2 may not be coupled well with downstream signaling cascades.
We then examined if trastuzumab induces ADCC in CHO-K6 cell. We showed that trastuzumab indeed induces strong ADCC in CHO-K6 cells (Fig.
8). This is specifically due to the expression of HER2 in CHO-K6 cells as there is no ADCC observed in CHO-EGFR cells (Fig.
8). The role of trastuzumab in the induction of ADCC in HER2-positive breast cancer cells have been consistently well supported by many researches [
44‐
51]. Our results confirmed the role of trastuzumab in the induction of ADCC in a simple but specific cell setting.
We also showed that trastuzumab did not affect cell proliferation in CHO-K6 cells (Fig.
9). Some reports indicated that trastuzumab had little effect on proliferation and survival [
58,
77]. However, other reports indicated that trastuzumab inhibited ErbB2 activation, and decreased the activation of ERK and PI3K-AKT pathways, which leads to reduced cell proliferation [
57]. Given that trastuzumab has little effects on the phosphorylation of HER2, it is likely that trastuzumab has no effects on HER2-mediated cell signaling leading to cell proliferation. Although trastuzumab induces ADCC in CHO-K6 cell, under the culture conditions used for the MTT assay, no effector cells were present and no ADCC response are expected. It is interesting to note that our above finding are different from the observation by Ghosh [
78]. Ghosh et al. reported that trastuzumab inhibited HER2 homodimer-mediated ERK phosphorylation and cell growth. The difference could be due to the different model system used in these two studies. The HER2 receptor used in the research by Ghosh et al. is fused with FKBP, and the receptor homodimerization is induced by a chemical linker AP1510 that dimerizes the receptor intracellularly through the fused FKBP.
It should be noted that while we observed strong inhibition of HER2 phosphorylation by CP724714 at 1 μM (Fig.
5b). We only observed the inhibition of CHO-K6 cell proliferation at much higher CP724714 concentration (Fig.
9f). We are not sure what cause this discrepancy, however, there are several possible explanations. Firstly, at 1 μM, CP724714 may not completely inhibit Her2 phosphorylation, we can see weak phosphorylation of Y1139 in Fig.
5b. There could be weak phosphorylation of HER2 at other pY residues that were not examined. A weak HER2 phosphorylation may still be sufficient to support cell growth. Secondly, there could be the existence of kinase-independent effects of HER2 receptors. There are many reports supporting the existence of kinase-independent cell signaling of various receptor tyrosine kinases including EGFR and Insulin receptor [
79‐
82].
Conclusion
Together, in this research we adopted a cell model that allow us to specifically examine the effects of trastuzumab on a single HER receptor without the influence of other HER receptors. Three CHO cell lines stably expressing only human EGFR, HER2, or HER3 were used. These model system allow us to specifically examine the effects of trastuzumab on the homodimerization of HER2 and the phosphorylation of HER2 homodimers. We demonstrate that overexpression of HER2 in CHO cells results in the homodimerization of HER2 and the phosphorylation of HER2 at all major pY residues. Trastuzumab binds to HER2 specifically and with high affinity. Trastuzumab does not inhibit the homodimerization of HER2. Trastuzumab does not inhibit the phosphorylation of HER2 at most phosphotyrosine residues.
We also observed that trastuzumab neither inhibits the phosphorylation of the downstream signaling proteins including ERK and AKT, nor inhibits the proliferation of CHO cells overexpressing HER2. However, caution is needed to apply these findings to breast cancer cells. However, trastuzumab induces strong ADCC in CHO cell overexpressing HER2. We concluded that trastuzumab exerts its antitumor activity through the induction of ADCC, rather than the inhibition of HER2-mediated cell signaling in the absence of other HER receptors.