Adhesion-based simple capture and recovery of circulating tumor cells using a blood-compatible and thermo-responsive polymer-coated substrate

Abstract

Circulating tumor cells (CTCs) have been a focus of study for metastatic cancer diagnostics, in in vitro anti-cancer drug screening to decide the chemotherapeutic course, and cancer biology research. For these purposes, there have been efforts made to collect CTCs from the peripheral blood of cancer patients. Here, we explore the possibility of collecting CTCs using blood-compatible and thermo-responsive poly(2-(2-ethoxyethoxy) ethyl acrylate-co-2-(2-methoxyethoxy) ethyl methacrylate) (P(Et2A-Me2MA)) through adhesion and detachment by incubation under a lower critical solution temperature of P(Et2A-Me2MA). A P(Et2A-Me2MA)-coated substrate is dissolved by incubation under 15 °C. A P(Et2A-Me2MA)-coated substrate can suppress platelet adhesion whereas it allows the cancer cells to adhere by an epithelial cell adhesion molecule (EpCAM) expression-independent mechanism. These results suggest that cancer cells can specifically adhere to a P(Et2A-Me2MA)-coated substrate, which can be used to isolate CTCs from peripheral blood. Moreover, approximately 90% of the adherent cells can be detached by incubation at 10 and 15 °C for 30 and 90 min, respectively. The collected cells can be cultured healthily in the presence of dissolved P(Et2A-Me2MA), suggesting that the cytotoxicity of P(Et2A-Me2MA) is low. In conclusion, P(Et2A-Me2MA) is suitable for the development of devices that collect intact CTCs via an adhesion-based method.

---

Introduction

Cancer is one of the leading causes of death worldwide, particularly in developed nations. Over 90% of cancer deaths are in patients with metastatic cancer.¹ The early diagnosis and treatment of metastatic cancer is crucial for decreasing cancer mortality. Once the cancer becomes metastatic, metastatic cancer cells invade the peripheral blood. These cells are known as circulating tumor cells (CTCs).² CTCs have been a focus in the diagnosis and prognosis of metastatic cancer and many efforts have been made to detect CTCs in the peripheral blood.^2,3 In addition to their role in the diagnosis of metastatic cancer, CTCs may assist in deciding the chemotherapeutic course because CTCs provide extensive information about the tumor, including its genetic mutations and chemosensitivities. In particular, growth-cultured CTCs provide information about chemosensitivity by allowing for the in vitro screening of anti-cancer drugs, which can serve as a basis for evidence-based personalized medicine.⁴ Moreover, CTCs are valuable for cancer biology research. To serve these purposes, intact CTCs should be isolated in a way that maintains their chemosensitivity or other cancer cell properties throughout the isolation process.

There are two main methods for the detection CTCs in peripheral blood: an antibody-based method,^5,6 a microchip technology-based method,^7–9 and a combination of the two.^10,11 In particular, the CellSearch System is the only technique approved by FDA for the diagnosis and prognosis of metastatic breast, prostate and colorectal cancers.^5,6 The CellSearch System detects CTCs with an antibody against the epithelial cell adhesion molecule (EpCAM).⁵ However, due to its reliance on the recognition of EpCAM, the CellSearch System cannot detect CTCs that do not possess EpCAM on their cell surface, thus leading to false-negative diagnoses of cancer metastasis.^7–9 From the vantage point of anti-cancer drug screening, antibody-based methods for isolating CTCs should be avoided because the antibodies used for cell labeling may alter crucial cancer cell phenotypes, such as chemosensitivity. Furthermore, antibody-based methods often require special apparatuses and well-trained operators. To overcome these limitations, microchip technology-based methods have been developed.^7–9 Microchip technology-based methods separate between CTCs and blood cells based on the differences in their physical properties (e.g., size and deformability) to isolate CTCs from peripheral blood. However, there are several technical limitations to microchip technology-based methods; including contamination by leukocytes, the difficulties of large-scale processing¹² and potential changes in gene expression that could lead to inaccurate evaluation of chemosensitivity after strong shear stress.¹³ For these reasons, a new method for isolating intact CTCs in peripheral blood is required.

We have previously reported that poly(2-methoxyethyl acrylate) (PMEA) and its analogous polymers exhibited blood-compatibility.¹⁴ Generally, conventional blood-compatible polymers, such as polyethylene glycol (PEG) and poly(2-methacryloyloxyethyl phosphoryl choline), suppressed the adhesion of both blood cells and adherent cancer cells.^15,16 In contrast with these blood-compatible polymers, we have recently reported that blood-compatible PMEA, and its analogous polymers, allow for non-blood cells, but not blood cells, to adhere to these polymer-coated substrates.^17–19 Therefore, it is expected that CTCs can be isolated using PMEA, or its analogous polymer-coated substrates, via an adhesion-based mechanism. More recently, we have demonstrated that some of the PMEA analogous polymers possess a lower critical solution temperature (LCST).²⁰ Thermo-responsive polymers such as poly(N-isopropyl acrylamide) (PNIPAAm) have been frequently used for the detachment of cells in an intact condition from their substrates by their incubation at temperatures under the LCST.^21,22 Therefore, it is expected that the substrates coated with PMEA analogous polymers will exhibit both blood-compatibility and thermo-responsiveness, which could be used to collect intact CTCs from peripheral blood.

In this study, we examined the possibility of capturing CTCs via an adhesion-based method using a new PMEA analogous polymer, poly(2-(2-ethoxyethoxy) ethyl acrylate-co-2-(2-methoxyethoxy) ethyl methacrylate) (P(Et2A-Me2MA)) by an EpCAM-independent mechanism. In addition, we attempted to release the adherent cells from the polymer-coated substrate by incubating it under the LCST of P(Et2A-Me2MA) to facilitate the collection of intact cells. We also examined whether the detached CTCs could be further cultured in vitro.

Experimental

Preparation of polymer-coated substrate

PMEA was synthesized according to a previous report.¹⁴ Poly(2-methacryloyloxyethyl phosphorylcholine-co-butyl methacrylate) (30 : 70 mol%, PMPC) was kindly gifted by the NOF Corporation (Tokyo, Japan). P(Et2A-Me2MA) was synthesized as follows; the monomers of Et2A (Tokyo Chemical Industry, Tokyo, Japan) and Me2MA (Tokyo Chemical Industry) were mixed at a ratio of 50 : 50 mol% in 1,4-dioxane and were then polymerized via free radical polymerization using azobis(isobutyronitrile) (AIBN) as an initiator at 75 °C for 6 h. Synthesized P(Et2A-Me2MA) was purified by precipitation using tetrahydrofuran (THF)/hexanes as the good/bad solvent pair system, collected and dried under reduced pressure at 60 °C for 18 h. The chemical structure of P(Et2A-Me2MA) was confirmed by ¹H-NMR spectroscopy (JEOL, ECX 500 MHz). The LCST determined by turbidity measurement was 18.0 °C in phosphate buffered saline (PBS). The chemical structures of the polymers used in this study are shown in ESI Fig. S1.†

Polymer-coated substrates were prepared according to previous reports.^17–19 The polymers were dissolved in methanol at concentrations of 0.2, 1.0, and 5.0 wt%. The polymers were spin-coated on polyethylene terephthalate (PET) discs (ϕ = 14 mm, thickness = 125 μm, Mitsubishi Plastics, Tokyo, Japan) with prepared polymer solutions. The thickness of polymer layers spin-coated at concentrations of 0.2, 1.0, and 5.0% were approximately 70–80 nm, 240–280 nm, and 730–800 nm, respectively (ESI Fig. S2†). Additionally, the polymers were cast-coated on tissue culture polystyrene (TCPS, Iwaki, Chiba, Japan) with 37.5 μl cm⁻² of each polymer solution (0.2 wt%). UpCell was purchased from CellSeed (Tokyo, Japan). The prepared substrates were exposed to UV light on a clean bench for 2 h to sterilize them and were stored at room temperature until further use.

Measurement of the water contact angle

The substrates spin-coated with 1.0% of polymer solution were immersed in PBS and were incubated at 10 or 15 °C for the indicated times. Incubated substrates were then washed once with water at 70 °C and dried overnight. The water contact angle on the dried substrates was measured using the sessile drop method (2 μl of water droplet) to confirm whether the coated polymer was dissolved by incubation under the LCST.

In addition to the confirmation of polymer dissolution, the temperature of the media was also monitored. One milliliter of Dulbecco's modified Eagle Medium/F-12 Mixture (1 : 1) (DMEM/F-12, Gibco, Carlsbad, CA) was incubated in 24-well plate at 10 or 15 °C. The temperature of the media was monitored for 40 min with a digital thermometer (Asone, Osaka, Japan).

Platelet adhesion test

The polymer-coated substrates were cut into squares (8 × 8 mm). The square substrates were fixed on a SEM specimen stage using double-faced adhesive tape. Platelet-rich plasma (PRP) was obtained from human whole blood (Tennessee Blood Services, Memphis, TN) as the supernatant after centrifugation at 1500 rpm for 5 min. The platelet poor plasma (PPP) was obtained from the same blood sample as the supernatant after centrifugation at 4000 rpm for 10 min. The PRP and PPP were mixed to prepare the platelet suspension at a concentration of 2 × 10⁸ cells per ml and then, 200 μl of the suspension was dropped onto the polymer-coated substrates. Platelets were allowed to adhere to the substrate for 1 h at 37 °C. After the incubation time, non-adherent platelets were washed twice with PBS. The adherent platelets were then fixed with 1% glutaraldehyde (Polysciences, Inc., Warrington, PA) for 2 h at 37 °C. Finally, the samples were washed four times in the following order; PBS, diluted PBS (1 : 1, PBS/water) and twice with pure water. Air-dried samples were observed with a scanning electron microscope (SEM).

Enzyme-linked immunosorbent assay (ELISA)

The polymer-cast-coated substrates were immersed in PPP or 5 μg ml⁻¹ of human fibronectin (Sigma, St Louis, MO) solution and then were incubated for 1 h at 37 °C. After the adsorption of proteins, the samples were incubated with Blocking-One (Nacalai Tesque, Kyoto, Japan) for 30 min at room temperature to prevent non-specific reactions of the antibodies (Abs). After blocking, the samples were then incubated with conformation-specific anti-human fibrinogen γ chain Ab (clone 2.G2.H9, Millipore, Temecula, CA) and anti-human fibronectin Ab (HFN7.1, Abcam, Cambridge, UK) for 2 h at 37 °C and then with peroxidase-conjugated anti-mouse IgG Ab for 1 h at 37 °C. After the incubation with the Abs, the samples were incubated with 2, 2′-azinobis (3-ethylbenzothiazoline-6-sulfonic acid ammonium salt) (ABTS) substrate (Roche Diagnostics) at 37 °C for a colorimetric assay. The absorbance was measured at a wavelength of 405 nm.

Cell culture

Colorectal cancer cell lines, HT-29 and SW480, and a breast cancer cell line, MDA-MB-231 were obtained from the American Type Culture Collection (ATCC, Manassas, VA). A breast cancer cell line, MCF-7, hepatocarcinoma cell line, HepG2, fibrosarcoma cell line, HT-1080, and a cervical carcinoma cell line, HeLa, were obtained from the Health Science Research Resources Bank (Osaka, Japan). All cells were maintained in TCPS dishes in DMEM/F-12 containing 10% fetal bovine serum (FBS; Equitech-Bio, Kerrville, TX) (serum DMEM/F-12). Prior to further experiments, the cells were detached from the dishes with a 0.25% trypsin/EDTA solution (Gibco).

Semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was extracted from the semi-confluent cultured cells using Sepasol-RNA I Super reagent according to the manufacturer's instructions (Nacalai Tesque). One microgram of total RNA was used as a first-strand reaction that included random hexamer primers and ReverTra Ace-α reverse transcriptase (TOYOBO, Osaka, Japan). Semi-quantitative RT-PCR was performed using HybriPol DNA polymerase (Nippon Genetics, Tokyo, Japan) with specific human primer sets. All primers were obtained from Nihon Gene Research Laboratories (Sendai, Japan). The sequence of primer sets were as follows, GAPDH: (forward): 5′-GGGCTGCTTTTAACTCTGGT-3′, (reverse): 5′-TGGCAGGTTTTTCTAGACGG-3′, EPCAM: (forward) 5′-GCTCTGAGCGAGTGAGAACC-3′, (reverse) 5′-GATGTCTTCGTCCCACGCAC-3′. For each experiment, GAPDH was amplified to normalize the expression of EPCAM in the sample. The PCR products were analyzed by 1% agarose gel electrophoresis.

Cell adhesion assay

Prior to cell culture, the polymer-spin-coated substrates were immersed in serum DMEM/F-12 for 1 h at 37 °C. The cells were seeded on the substrates at a density of 1 × 10⁴ cells per cm² and were allowed to adhere to the substrates in serum DMEM/F-12 for 30 and 60 min. Non-adherent cells were removed from the culture by washing twice with PBS. Adherent cells were fixed with 0.1% glutaraldehyde and were stained with 0.2% crystal violet (Wako, Osaka, Japan) containing PBS for visualization. The adherent cells in three randomly selected fields were counted using an optical microscope (Olympus, Tokyo, Japan).

The adhesion of HT-1080 cells spiked in human whole blood was also examined. The cells were seeded on P(Et2A-Me2MA)-coated substrates at a density of 1 × 10⁴ cells per cm² in the blood and were allowed to adhere to the substrates for 1 h at 37 °C. The adherent HT-1080 cells were immunostained with anti-cytokeratin 19 antibody (Progen Biotechnik GmnH, Heidelberg, Germany) and corresponding Alexa-568-conjugated secondary antibody (Invitrogen, Carlsbad, CA). All cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, Invitrogen). Finally, the adherent cells were observed with CQ1 image analyser (Yokogawa Electric Corporation, Tokyo, Japan).

Cell detachment from the polymer substrates

HT-1080 cells were seeded on the polymer-spin-coated substrates at a density of 1 × 10⁴ cells per cm² prior to the incubation of the substrates in serum DMEM/F-12 and were cultured in serum DMEM/F-12 for 1 day. After culturing, the samples were treated with 3 types of processes as follows; (a: non-treat) non-adherent cells were immediately washed with PBS once after culturing, (b: shake) the media was changed to fresh DMEM/F-12 with serum and the samples were shaken at 500 rpm for 30 s on a plate shaker (IKA, Wilmington, DE). Non-adherent cells were removed by washing once with PBS, (c: treat) the media was changed to fresh DMEM/F-12 with serum and the samples were incubated at 10 or 15 °C for the indicated time. After the incubation, the samples were shaken at 500 rpm for 30 s on a plate shaker. Non-adherent cells were removed by washing once with PBS. The remaining cells were fixed with 0.1% glutaraldehyde containing PBS and were then visualized by crystal violet staining. The numbers of the remaining cells were counted in three randomly selected fields using an optical microscope. All detached HT-1080 cells obtained by the above treatment were also re-seeded on TCPS and were cultured in DMEM/F-12 with serum. In addition, cell shape was observed by phase contrast microscopy after 1 and 3 days of culture.

Cytotoxicity assay with soluble P(Et2A-Me2MA)

P(Et2A-Me2MA) was dissolved from the spin-coated substrate in 1 ml of DMEM/F-12 with serum by incubation at 10 °C for 90 min. Freshly harvested HT-1080 cells were seeded on TCPS at a density of 1 × 10⁴ cells per cm² and were cultured in P(Et2A-Me2MA) containing DMEM/F-12 with serum. The number of growing cells was assessed by a WST-8 assay after the culture for the indicated time.²³

Statistical analysis

All of the data are expressed as the means ± SD. The significance of the differences between the two samples was determined by an unpaired Student's t test using Microsoft Excel 2010. P values less than 0.05 were considered statistically significant.

Results and discussion

Thermo-responsiveness of P(Et2A-Me2MA)-coated substrate

We confirmed whether the substrates coated with the new PMEA analogous polymer, P(Et2A-Me2MA), exhibited the desired thermo-responsive properties. The water contact angle of the substrates was measured after incubation in water at 10 °C to check the dissolution of coated P(Et2A-Me2MA) (Table 1). The contact angles of the substrates coated with P(Et2A-Me2MA) at 0.2, 1, and 5 wt% did not change after their incubation at 37 °C. In contrast, they exhibited contact angles similar to bare PET and to the substrates-coated with PMEA even after their incubation at 10 °C.

Table 1 Water contact angle of polymer-coated substrates (θ)^a

Polymer	Coating conc. (wt%)	Before treatment	37 °C for 90 min	10 °C for 90 min
a N.D.: not determined.
PET	—	72.8 ± 1.1	72.1 ± 1.2	73.6 ± 1.6
PMEA	0.2	40.6 ± 1.1	N.D.	N.D.
	1.0	46.0 ± 0.4	38.9 ± 2.1	39.8 ± 2.0
P(Et2A-Me2MA)	0.2	24.1 ± 1.6	24.0 ± 1.4	71.4 ± 1.8
	1.0	30.2 ± 2.6	29.0 ± 1.6	74.3 ± 1.9
	5.0	41.3 ± 1.4	36.9 ± 1.6	75.6 ± 0.9

---

In addition, we checked the kinetics of contact angle alteration at 10 and 15 °C (Fig. 1A). After 10 min of incubation at 10 °C, the contact angle of P(Et2A-Me2MA)-coated substrates started to increase and the contact angle reached the angle of bare PET after 30 min. Furthermore, the water contact angle of P(Et2A-Me2MA) started to gradually increase after approximately 15 min of incubation at 15 °C and the contact angle reached the angle of bare PET after 60 min. In contrast, the contact angle of bare PET did not change during incubation at 10 and 15 °C. These results suggested that almost all the P(Et2A-Me2MA) dissolved and the bare PET surface was exposed by the incubation at 10 and 15 °C. Therefore, it is expected that cells cultured on P(Et2A-Me2MA)-coated substrate will lose their anchoring substrate and detach during this incubation.

Fig. 1 The rate of polymer dissolution from the polymer-coated substrates. (A) The change in the water contact angle of the 0.2 wt% polymer-coated substrates. Circle: P(Et2A-Me2MA) at 15 °C, Diamond: PET at 10 °C. Data represent means ± SD (n = 5). (B) Time dependency of the temperature change of the media. Circle: 15 °C, Square: 10 °C.

To compare the kinetics of contact angle alteration, we also monitored the temperature of the media during the incubations at 10 and 15 °C (Fig. 1B). The temperature of the media reached 18 °C, the LCST of P(Et2A-Me2MA), after 10 and 15 min of incubation at 10 and 15 °C, respectively. This is consistent with the time required to alter the contact angle of P(Et2A-Me2MA). These results indicated that the substrate coated with P(Et2A-Me2MA) was thermo-responsive.

Blood-compatibility of P(Et2A-Me2MA)-coated substrates

We evaluated the blood-compatibility of P(Et2A-Me2MA)-coated substrates by performing a platelet adhesion test after exposure of the coated substrate to a platelet suspension for 1 h (Fig. 2A). On a bare PET substrate which is not blood-compatible, massive platelet adhesion was observed. In contrast with the bare PET substrate, platelets hardly adhered to blood-compatible PMEA and PMPC substrates. Similar with blood-compatible PMEA and PMPC substrates, adherent platelets were hardly observed on the P(Et2A-Me2MA)-coated substrate, indicating that the P(Et2A-Me2MA) substrate is blood compatible. In contrast with thermo-responsive P(Et2A-Me2MA), it has been reported that platelets adhere well on a PNIPAAm-immobilized substrate, which is another thermo-responsive polymer.²⁴

Fig. 2 Platelet adhesion on 0.2 wt% polymer-coated substrates. (A) Photos of the polymer-coated substrate surface after their immersion in platelet-containing plasma. Bar indicates 10 μm. (B) Detection of exposed platelet adhesion sites on the fibrinogen γ chain adsorbed on polymer-coated substrate by ELISA. Data represent means ± SD (n = 5). *: P < 0.05 vs. TCPS.

It has been reported that the conformational alteration of adsorbed fibrinogen is required to expose cell adhesion sites for platelet adhesion on the polymer substrate.²⁵ The cell adhesion sites that were exposed were compared among polymer-coated substrates by ELISA with a cell adhesion site-specific Ab after 1 h of the incubation in PPP (Fig. 2B). Many fibrinogen cell adhesion sites were detected on the TCPS, which was assessed as a non-blood-compatible substrate instead of bare PET. Significantly, fewer fibrinogen cell adhesion sites were detected on blood-compatible PMEA- and PMPC-coated substrates. Similar with these blood-compatible substrates, the cell adhesion sites were barely detected on the P(Et2A-Me2MA)-coated substrate. Therefore, the P(Et2A-Me2MA)-coated substrate exhibited blood-compatibility due to the suppression of fibrinogen's conformational change.

We have reported that a unique water structure (termed as intermediate water) is formed on many blood-compatible polymers under hydrating conditions.^14,26 The intermediate water can act as a barrier that inhibits the interaction between the polymer and proteins. It has also been reported that the intermediate water weakens the force that is necessary to induce conformational changes in the adsorbed proteins (e.g., hydrophobic interaction).^14,18,20 Additionally, we have reported that the homopolymers of both PEt2A and PMe2MA exhibit an intermediate water structure under hydrated conditions.²⁰ Therefore, we speculated that the intermediate water structure would be formed in hydrated P(Et2A-Me2MA), which could potentially inhibit fibrinogen's conformational change.

Cancer cell adhesion on a thermo-responsive and blood-compatible P(Et2A-Me2MA)-coated substrate

As a next step, we examined whether various cancer cells could adhere to the P(Et2A-Me2MA)-coated substrate. Seven types of cancer cell lines were used for this purpose. It has been noted that EpCAM-negative cancer cells cannot be detected by the CellSearch System; which is the only device approved by the FDA for CTC-based cancer diagnosis.^5,6

Therefore, seven different cancer cell lines were examined in this study; 4 EpCAM-positive cancer cell lines (HT-29, MCF-7, HepG2, and SW480) and 3 EpCAM-negative cell lines (MDA-MB-231, HT-1080, and HeLa) (Fig. 3A).

Fig. 3 Cancer cell adhesion on polymer-coated substrates. (A) EPCAM expression in 7 types of cancer cell lines. The adhesion of (B) HT-29, (C) MCF-7, (D) HepG2, (E) SW480, (F) MDA-MB-231, (G) HT-1080, (H) HeLa on 0.2 wt% polymer-coated substrates. White and black bars indicate the numbers of adherent cells after 30 and 60 min incubations, respectively. PMPC is a negative control for cell adhesion. Data represent means ± SD (n = 3). P < 0.05 (PET, PMEA, and P(Et2A-Me2MA) vs. PMPC). (I) Detection of exposed cell adhesion sites in human fibronectin adsorbed on polymer-coated substrates by ELISA. Data represent means ± SD (n = 5). ***: P < 0.005 vs. PMPC.

A cell adhesion assay was performed to assess the adhesion of cancer cells after 30 min and 1 h (Fig. 3B–H). All examined cancer cells started to adhere to bare PET within 1 h. On the other hand, all the examined cancer cells barely adhered to the PMPC-coated substrate. The number of adherent cancer cells on the PMEA-coated substrate was higher than that on bare PET, which is consistent with our previous report.^17–19 All examined cancer cells could also adhere to the P(Et2A-Me2MA)-coated substrate and the numbers of adherent cells were also higher than those on bare PET. These results indicated that both EpCAM-positive and negative cancer cells can adhere to a thermo-responsive and blood-compatible P(Et2A-Me2MA)-coated substrate. These results suggesting that our adhesion-based method of CTC detection has the potential to detect CTCs that cannot be detected by the CellSearch System.

To understand why cancer cells, but not platelets, can adhere to the P(Et2A-Me2MA)-coated substrate, cell adhesion sites on adsorbed fibronectin, a major cancer cell adhesion protein, were detected by ELISA with cell adhesion site-specific Abs (Fig. 3I). The cell adhesion sites were barely detected on the PMPC-coated substrate which is a non-cell-adhesive substrate. In contrast with the PMPC-coated substrate, the cell adhesion sites were well-detected on TCPS, which acted as a control for a cancer cell adhesive substrate, and the PMEA-coated substrate, which is consistent with our previous report.¹⁷ Similar with TCPS and PMEA-coated substrates, the cell adhesion sites were well-detected on the P(Et2A-Me2MA)-coated substrate. Therefore, we suggest that the exposed cell adhesion site on fibronectin allows the cancer cells to adhere to the P(Et2A-Me2MA)-coated substrate.

The conformation of fibronectin is linked to “beads on a string”²⁷ and fibronectin sees more flexible than fibrinogen. Thus, the formation of the intermediate water structure might be insufficient to inhibit the conformational change of fibronectin, but no fibrinogen, due to the difference in their flexibilities. Indeed, we have previously reported that cell adhesion sites on adsorbed fibronectin decreased as the levels of the intermediate water structure increased in hydrated PMEA analogous polymers.¹⁸ Moreover, we have shown that cells can adhere to several PMEA analogous substrates via both integrin-dependent and independent mechanisms.^17–19 We cannot exclude the possibility that cancer cells adhere to the P(Et2A-Me2MA) substrates via both integrin-dependent and independent mechanisms.

Cell detachment from P(Et2A-Me2MA)-coated substrate mediated by incubation at temperatures under the LCST

Intact CTC isolation is also important for further analyses such as anti-cancer drug screening and additional biological studies that use collected CTCs. We next examined whether adherent cancer cells can be detached from a P(Et2A-Me2MA)-coated substrate by incubation at temperatures under the LCST of the polymer. PNIPAAm-grafted substrate (UpCell) was used as a positive control of cell detachment and HT-1080 cells were used for the following experiments.

HT-1080 cells were cultured on polymer-coated substrates for 1 day (Fig. 4). The cells spread on PET, PMEA-coated substrate, and UpCell. On the P(Et2A-Me2MA)-coated substrate, the cell shape changed depending on the concentrations of the coating material. Cells spread on substrates coated with P(Et2A-Me2MA) at 0.2 wt%. In contrast, cells maintained their round shape on substrates coated with P(Et2A-Me2MA) at concentrations of 1.0 and 5.0 wt%.

Fig. 4 Cell shapes on the polymer substrates after 1 day of culture. PMEA was spin-coated with 1.0 wt% of PMEA solution. Bar indicates 100 μm.

Next, we attempted to detach these cells from their substrates with incubation at 10 and 15 °C (Fig. 5). HT-1080 cells were barely detached from bare PET and PMEA-coated substrates even after incubation at 10 °C for 30 min and a shaking process. In contrast, cells completely detached from UpCell (Fig. 5A and Table 2). Similar to UpCell, HT-1080 cells detached from a P(Et2A-Me2MA)-coated substrate after 30 min of incubation at 10 °C and a shaking process. HT-1080 cells barely detached from a P(Et2A-Me2MA)-coated substrate and UpCell even after shaking without incubation at 10 °C. The detachment ratio of the cells from P(Et2A-Me2MA) increased with the increase in concentration of the coating material (Fig. 5A and Table 2). The cell shape remained round on the substrate coated with P(Et2A-Me2MA) at a concentration of 5.0 wt%, suggesting that the strength of the cell adhesion was weak (Fig. 4). Therefore, the cells could be detached from the substrate coated at 5.0 wt% more easily than those coated at 0.2 and 1.0 wt%.

Fig. 5 Detachment of HT-1080 from polymer-coated substrates. (A) The effect of polymer-coated concentration on cell detachment at 10 °C. Time dependency of HT-1080 detachment from 5.0 wt% P(Et2A-Me2MA)-coated substrate at 10 °C (B) and 15 °C (C). Non-treat indicates the cells without any treatment. Shake indicates the cells with shaking process at 37 °C. Treat indicates the cells after the incubation at 10 or 15 °C and shaking process. Data represent means ± SD (n = 3). *: P < 0.05, **: P < 0.01, ***: P < 0.005 vs. non-treat.

Table 2 The effect of coating concentration on detachment ratio of HT-1080 cells^a

Polymer	Coating conc. (wt%)	Detachment ratio (%)
a The cells were incubated at 10 °C for 30 min.
PET	—	13.5 ± 10.9
PMEA	1.0	9.2 ± 6.4
P(Et2A-Me2MA)	0.2	50.8 ± 11.6
	1.0	67.4 ± 14.1
	5.0	96.2 ± 0.7
UpCell	—	99.0 ± 0.2

---

We also tried to detach the cells from the substrate coated with the homopolymer of Et2A (PEt2A, LCST: 15 °C) by the incubation at 10 °C for 90 min. However, the cells could not detach from the PEt2A-coated substrate even after the incubation and a shaking process (ESI Table S1†). This result suggested that not only thermo-responsivity but also other factors can influence cell detachment from the substrate although it is unclear why the cells cannot detach from PEt2A-coated substrate by the incubation at 10 °C at this time.

The time dependency of cell detachment was also assessed after incubation at 10 °C (Fig. 5B and Table 3). Cells minimally detached from bare PET and PMEA-coated substrates even after 90 min. On UpCell, the cells began to detach after 15 min and 89% of the cells were detached after 30 min. Similar with the cells adhered to UpCell, the cells began to detach from the P(Et2A-Me2MA)-coated substrate after 15 min and 88% of the cells were detached after 30 min. As compared with the incubation at 10 °C, the cells did not begin to detach from the P(Et2A-Me2MA)-coated substrate after incubation at 15 °C whereas 67% of the cells were detached from UpCell (Fig. 5C and Table 4). After 30 min of incubation at 15 °C, the cells began to detach from the P(Et2A-Me2MA)-coated substrate and 94% of the cells detached within 90 min. These results indicated that cancer cells can detach from blood-compatible and thermo-responsive P(Et2A-Me2MA)-coated substrate after their incubation at temperatures under the LCST of the polymer.

Table 3 The effect of incubation time on the detachment ratio of HT-1080 cells at 10 °C^a

Polymer	Time (min)	Detachment ratio (%)
a PMEA and P(Et2A-Me2MA) were coated at concentrations of 1.0 and 5.0 wt%, respectively.
PET	15	11.7 ± 9.9
	30	20.1 ± 11.3
	90	31.5 ± 9.0
PMEA	15	5.8 ± 24.6
	30	9.2 ± 6.4
	90	10.2 ± 2.6
P(Et2A-Me2MA)	15	30.9 ± 16.6
	30	87.9 ± 8.8
	90	92.8 ± 1.7
UpCell	15	59.9 ± 3.9
	30	89.0 ± 7.0
	90	96.5 ± 1.5

---

Table 4 The effect of incubation time on the detachment ratio of HT-1080 cells at 15 °C^a

Polymer	Time	Detachment ratio (%)
a PMEA and P(Et2A-Me2MA) were coated at concentrations of 1.0 and 5.0 wt%, respectively.
PET	15	Undetached
	30	Undetached
	90	Undetached
PMEA	15	14.0 ± 24.6
	30	15.9 ± 26.1
	90	11.6 ± 14.5
P(Et2A-Me2MA)	15	13.9 ± 21.7
	30	51.4 ± 25.6
	90	93.9 ± 0.3
UpCell	15	66.9 ± 12.6
	30	84.3 ± 6.3
	90	98.3 ± 0.3

---

During 10 °C incubation, there was a marked change in the contact angle on the P(Et2A-Me2MA)-coated substrate surface that began at 10 min and P(Et2A-Me2MA) was almost completely dissolved by 30 min (Fig. 1A). In contrast, during incubation at 15 °C, the contact angle of P(Et2A-Me2MA)-coated substrate started to increase gradually during the first 15 min, and markedly increased after 40 min, suggesting that P(Et2A-Me2MA) dissolved more slowly at 15 °C than 10 °C (Fig. 1A). This differences in the rate of P(Et2A-Me2MA) dissolution led to the differences in detachment rate.

Cytotoxicity of P(Et2A-Me2MA) on detached HT-1080 cells

We examined the cytotoxic effects of P(Et2A-Me2MA) on detached HT-1080 cells (Fig. 6). First, we checked the cytotoxicity of P(Et2A-Me2MA) itself. The WST-8 assay was performed to compare cell growth between cells incubated with or without polymer (Fig. 6A). Regardless of the presence of P(Et2A-Me2MA), the cells grew similarly on TCPS. These results suggested that P(Et2A-Me2MA) exerted few cytotoxic effects on the cells.

Fig. 6 The cytotoxicity of P(Et2A-Me2MA). (A) Proliferation of freshly harvested HT-1080 cells in the media with dissolved P(Et2A-Me2MA). Data represent means ± SD (n = 3). N.S. indicates no significant differences. (B) The shapes of re-seeded HT-1080 cells after the detachment process. The cells were detached from P(ET2A-Me2MA) substrates with incubation at 10 °C. After the detachment, the cell suspension was re-seeded on TCPS. The photos were taken before and after the one wash with PBS. Bar indicates 100 μm.

As a next step, the cytotoxic effect of P(Et2A-Me2MA) dissolving from the substrate on detached cells was observed. Cells that detached from the P(Et2A-Me2MA)-coated substrate were re-seeded on TCPS and were cultured in the presence of dissolved P(Et2A-Me2MA) at the concentration after cell detachment. After 1 day of culture, the cells were normally spread on TCPS even in the presence of the polymer (Fig. 6B). After 3 days, the cells were growing on TCPS even in the presence of the polymers. Furthermore, the dissolved polymer can be easily washed out from the culture. Thus, it is suggested that dissolved P(Et2A-Me2MA) had few cytotoxic effects on detached cells from the substrate.

In our previous report, we showed that the chemosensitivity of A549, a lung carcinoma cell line, was similar when compared with TCPS.²⁸ Similar to these polymers, the PMEA analogous polymer, P(Et2A-Me2MA), is expected to exhibit few effects on chemosensitivity, which is one of the most important parameters used during in vitro screening of anti-cancer drugs.

HT-1080 cell adhesion on P(Et2A-Me2MA)-coated substrate in whole blood

Lastly, we examined the possibility to capture CTCs in the blood of metastatic cancer patients by our adhesion-based method. Adhesion assay of HT-1080 cells spiked in human whole blood was performed (Fig. 7). Cytokeratin 19-positive nucleated cells can be observed on P(Et2A-Me2MA)-coated substrate after 1 h of incubation. In contrast, cytokeratin 19-negative nucleated cells hardly observed. This result suggested the possibility to capture CTCs from the blood of metastatic cancer patients by our adhesion-based methods.

Fig. 7 Adherent cells from HT-1080 cell-spiked whole blood. The cell in HT-1080 cell-spiked whole blood were allowed to adhere on 5.0 wt% P(Et2A-Me2MA)-coated substrate for 1 h. Cell nuclei and cytokeratin 19 were labelled in the adherent cells. Blue and red indicate cell nuclei and cytokeratin 19, respectively. Adherent HT-1080 cells were shown as cytokeratin 19-positive nucleated cells. * Indicates a cytokeratin 19-negative nucleated cell (i.e., a blood cell). Bar indicates 60 μm.

At present, several techniques have been proposed for the detection and isolation of CTCs. PNIPAAm conjugated with an antibody has been reported to isolate CTCs when incubated at a cold temperature.²⁹ Aptamer-immobilized substrates have also been reported to isolate CTCs from peripheral blood, which can then be released after incubation with a DNase.³⁰ However, these substrates may potentially fail to detect several types of CTCs because their detection principle is based on specific interactions between the cells and the substrates. In contrast, our method is based on integrin-dependent (and possibly integrin-independent) adhesion which is a system universal used by cancer cells. Therefore, it is expected that our system can be applied for broader capture of CTCs.

Compared with microchip technology-based method, our method is expected to be more easily applied to large-scale processing while avoiding contamination with leukocytes. Moreover, our method can be combined with column technology³¹ and/or nanotopographic patterning techniques³² to increase the specificity of detection. In addition to the combination of these techniques, the blood samples can be pre-treated to remove erythrocytes with the dilution of blood samples with a erythrocyte lysis buffer or density-gradient centrifugation with ficoll for the increase of CTC capture efficiency.^33,34 Therefore, it is expected to increase the specificity and efficiency of CTC capture with our adhesion-based method by the combination with above techniques. Further optimization is undergoing for future application.

Conclusions

We have demonstrated that a blood-compatible and thermo-responsive P(Et2A-Me2MA) substrate captures both EpCAM-positive and negative cancer cells for the isolation of CTCs from peripheral blood. Additionally, the adherent cancer cells can be readily detached from the substrate in an intact state by simple incubation of the sample at temperatures under the LCST of P(Et2A-Me2MA), allowing the collection of the CTCs for further analysis, such as in a biological assay. Therefore, we concluded that our blood-compatible and thermo-responsive P(Et2A-Me2MA) substrate shows great potential as a material for the development of innovative devices for CTC-based cancer diagnosis, in vitro anti-cancer drug screening for personalized cancer medicine, and research into cancer biology.

Acknowledgements

We thank Ms Wakui of Yamagata University for technical support for the platelet adhesion test. We also thank Prof. Kobayashi of Kyushu University for fruitful discussions relating to this work. This work was supported by the Funding Program for Next-Generation World-Leading Researchers (NEXT Program, LS017) of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan and the Center of Innovation Program from the Japan Science and Technology Agency (JST). Also, this study was partially funded by Sumitomo Rubber Industries Ltd. The Sumitomo Rubber Industries Ltd had no control over the interpretation, writing, or publication of this work. T. Hoshiba was also supported in part by a Grant-in-Aid for Young Scientists (A) (26702016), funded by MEXT, Japan.

References

1. D. Hanahan and R. A. Weinberg, Cell, 2011, 144, 646.
2. C. Alix-panabières, H. Schwarzenbach and K. Pantel, Annu. Rev. Med., 2012, 63, 199.
3. M. Yu, S. Scott, M. Toner, S. Maheswaran and D. A. Haber, J. Cell Biol., 2011, 192, 373.
4. M. Yu, M. A. Bardia, N. Aceto, F. Bersani, M. W. Madden, M. C. Donaldson, R. Desai, H. Zhu, V. Comaills, Z. Zheng, B. S. Wittner, P. Stojanov, E. Brachtel, D. Sgroi, R. Kapur, T. Shioda, D. T. Ting, S. Ramaswamy, G. Getz, A. J. Iafrate, C. Benes, M. Toner, S. Maheswaran and D. A. Haber, Science, 2014, 345, 216.
5. M. C. Miller, G. V. Doyle and L. W. Terstappen, J. Oncol., 2010, 2010, 617421,  DOI:10.1155/2010/617421.
6. H. Wang, G. Yue, C. Dong, F. Wu, J. Wei, Y. Yang, Z. Zou, L. Wang, X. Qian, T. Zhang and B. Liu, ACS Appl. Mater. Interfaces, 2014, 6, 4550.
7. M. Hosokawa, T. Hayata, Y. Fukuda, A. Arakaki, T. Yoshino, T. Tanaka and T. Matsunaga, Anal. Chem., 2010, 82, 6629.
8. S. J. Tan, L. Yobas, G. Y. Lee, C. N. Ong and C. T. Lim, Biomed. Microdevices, 2009, 11, 883.
9. G. Vona, A. Sabile, M. Louha, V. Sitruk, S. Romana, K. Schütze, F. Capron, D. Franco, M. Pazzagli, M. Vekemans, B. Lacour, C. Bréchot and P. Paterlini-Bréchot, Am. J. Pathol., 2000, 156, 57.
10. E. Ozkumur, A. M. Shah, J. C. Ciciliano, B. L. Emmink, D. T. Miyamoto, E. Brachtel, M. Yu, P. I. Chen, B. Morgan, J. Trautwein, A. Kimura, S. Sengupta, S. L. Stott, N. M. Karabacak, T. A. Barber, J. R. Walsh, K. Smith, P. S. Spuhler, J. P. Sullivan, R. J. Lee, D. T. Ting, X. Luo, A. T. Shaw, A. Bardia, L. V. Sequist, D. N. Louis, S. Maheswaran, R. Kapur, D. A. Haber and M. Toner, Sci. Transl. Med., 2013, 5, 179ra47,  DOI:10.1126/scitranslmed.3005616.
11. S. Maheswaran, L. V. Sequist, S. Nagrath, L. Ulkis, B. Brannigan, C. V. Collura, E. Inserra, S. Diederichs, A. J. Iafrate, D. W. Bell, S. Digumarthy, A. Muzikansky, D. Irimia, J. Settleman, R. G. Tompkins, T. G. Lynch, M. Toner and D. A. Haber, N. Engl. J. Med., 2008, 359, 366.
12. S. Riethdorf and K. Pantel, Ann. N. Y. Acad. Sci., 2010, 1210, 66.
13. S.-F. Chang, C. A. Chang, D.-Y. Lee, P.-L. Lee, Y.-M. Yeh, C.-R. Yeh, C.-K. Cheng, S. Chien and J.-J. Chiu, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 3927.
14. M. Tanaka, T. Motomura, M. Kawada, T. Anzai, Y. Kasori, T. Shiroya, K. Shimura, M. Onishi and A. Mochizuki, Biomaterials, 2000, 21, 1471.
15. K. Jang, K. Sato, Y. Tanaka, Y. Xu, M. Sato, T. Nakajima, K. Mawatari, T. Konno, K. Ishihara and T. Kitamori, Lab Chip, 2010, 10, 1937.
16. K. Yoshimoto, M. Ichino and Y. Nagasaki, Lab Chip, 2009, 9, 1286.
17. T. Hoshiba, M. Nikaido and M. Tanaka, Adv. Healthcare Mater., 2014, 3, 775.
18. T. Hoshiba, E. Nemoto, K. Sato, T. Orui, T. Otaki, A. Yoshihiro and M. Tanaka, PLoS One, 2015, 10, e0136066,  DOI:10.1371/journal.pone.0136066.
19. T. Hoshiba, T. Otaki, E. Nemoto, H. Maruyama and M. Tanaka, ACS Appl. Mater. Interfaces, 2015, 7, 18096.
20. K. Sato, S. Kobayashi, M. Kusakari, S. Watahiki, M. Oikawa, T. Hoshiba and M. Tanaka, Macromol. Biosci., 2015, 15, 1296.
21. H. Takahashi and T. Okano, Adv. Healthcare Mater., 2015, 4, 2388–2407.
22. K. Nagase, Y. Hatakeyama, T. Shimizu, K. Matsuura, M. Yamato, N. Takeda and T. Okano, Biomacromolecules, 2015, 16, 532.
23. T. Hoshiba and M. Tanaka, Biochem. Biophys. Res. Commun., 2013, 439, 291.
24. B. Sivaraman and R. A. Latour, Biomaterials, 2010, 31, 832.
25. K. Uchida, K. Sakai, E. Ito, O. H. Kwon, A. Kikuchi, M. Yamato and T. Okano, Biomaterials, 2000, 21, 923.
26. M. Tanaka, T. Hayashi and S. Morita, Polym. J., 2013, 45, 701.
27. D. J. Leahy, I. Aukhil and H. P. Erickson, Cell, 1996, 84, 155.
28. T. Hoshiba, M. Nikaido, S. Yagi, I. Konno, A. Yoshihiro and M. Tanaka, J. Bioact. Compat. Polym., 2016, 31, 361–372.
29. S. Hou, H. Zhao, L. Zhao, Q. Shen, K. S. Wei, D. Y. Suh, A. Nakao, M. A. Garcia, M. Song, T. Lee, B. Xiong, S.-C. Luo, H.-R. Tseng and H.-h. Yu, Adv. Mater., 2013, 25, 1547.
30. Y. Wan, Y. Liu, P. B. Allen, W. Asghar, M. A. Mahmood, J. Tan, H. Duhon, Y. T. Kim, A. D. Ellington and S. M. Iqbal, Lab Chip, 2012, 12, 4693.
31. A. Mahara and T. Yamaoka, Biomaterials, 2010, 31, 4231.
32. W. Chen, S. Weng, F. Zhang, S. Allen, X. Li, L. Bao, R. H. Lam, J. A. Macoska, S. D. Marajver and J. Fu, ACS Nano, 2013, 7, 566–575.
33. H. J. Kahn, A. Presta, L.-Y. Yang, J. Blondal, M. Trudeau, L. Lickley, C. Holloway, D. R. McCready, D. Maclean and A. Marks, Breast Cancer Res. Treat., 2004, 86, 237–247.
34. H. W. Hou, M. E. Warkiani, B. L. Khoo, Z. R. Li, R. A. Soo, D. S.-W. Tan, W.-T. Lim, J. Han, A. A. S. Bhagat and C. T. Lim, Sci. Rep., 2013, 3, 1259,  DOI:10.1038/srep01259.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra15229e

‡ These authors equally contributed to this work.

---