Establishment of a Human Breast Cancer Model by Fusion PCR for In Vivo and In Vitro Fluorescence Imaging of Human Breast Cancer

Abstract

This study aimed to construct a breast cancer model that could continuously express the genes of luciferase and green fluorescent protein. The genes luciferase, EGFP, and Neo were obtained by fusion polymerase chain reaction (PCR) and inserted into pAAV-MCS. The pAAV-Luciferase-EGFP-Neo vector was transfected into MDA-MB-231 cells. After antibiotic resistance gene screening and limiting dilution assay, we constructed a monoclonal stable cell line that expresses the fusion protein Luciferase-EGFP. In comparison with the polyclonal stable cell line, the monoclonal cell line had good genetic stability and was not different from the parental cell line MDA-MB-231. The monoclonal stable cell line would be ideal for a breast cancer model. Indices of fluorescence imaging can be applied to fluorescence imaging in vitro and in vivo, providing a straightforward and reliable system for breast cancer and drug discovery research.

Introduction

Breast cancer is one of the most common malignant tumors among women. Incidences of breast cancer increase every year, and breast cancer tends to progress rapidly in younger women (Ademuyiwa et al., 2015). The optimal treatment time is usually missed because breast cancer does not have typical symptoms at the initial stage. The key factors in the death of breast cancer patients are recrudescence and distant metastasis of tumor cells. The number of early breast cancer cells is too small to be detected with routine inspection methods. Therefore, it is necessary to construct cells that can be fluorescently labeled (Kijanka et al., 2016). In vivo bioluminescence imaging is being increasingly utilized as a method to continuously monitor the behavior of living cancer cells, which can be used to continuously and dynamically monitor the same individual cells.

The luciferase gene reporter system is a system that involves biological fluorescence. Light generated by the system has high sensitivity and penetration at an emission wavelength of 570 nm. With this system, the emission of fewer than 100 cells can be detected (Bahmani and Hosseinkhani, 2016). Green fluorescent protein, a highly efficient, stable, nontoxic, and sensitive marker for in vitro gene expression, can be excited at 488 nm and emitted at around 597 nm. The fluorescence intensity at an excitation wavelength of 488 nm and emission wavelength of 597 nm is highly efficient, stable, nontoxic, and easy to monitor. The luminous intensity of breast cancer cells dual labeled with luciferase and EGFP is positively correlated with cell number and tumor growth status.

Animal models of breast cancer have been widely used in various fields of scientific research. The different growth stages of breast cancer cells are closely correlated with tumor formation, growth rate, invasion, metastasis, and other biological behaviors (Li et al., 2016). Insertion of exogenous genes and long-term in vitro culture will lead to abnormalities in tumor cell growth; thus, it may not reflect the actual biological characteristics of tumors. Ultimately, this would result in preclinical efficacy results that are not consistent with the therapeutic effect of clinical patients, leading to inability in predicting drug efficacy. Therefore, it is important to establish new human breast cancer models that are efficient.

We constructed a plasmid vector pAAV-Luciferase-EGFP-Neo, which expresses the recombination fusion protein Luciferase-EGFP using fusion polymerase chain reaction (PCR) technology. Then, we obtained a breast cancer cell line with stable expression of two fluorescent proteins after transfection of MDA-MB-231 cells and resistance screening. However, the screened stable cell line had many clones with different biological characteristics such as light intensity, growth characteristics, migration ability, and cell morphology. Therefore, we screened out a monoclonal stable cell line with the same characteristics as nontransfected cells, which can form tumors in nude mice. The tumor model uses two optical imaging methods, biological luminescence and fluorescence labeling, which are suitable for progressively marking breast cancer cells and in vivo tracking.

Materials and Methods

Primers, plasmids, bacteria, cells, and animals

Based on the DNA sequence (luciferase, EGFP, IRES, and Neo), forward primers and reverse primers were designed by Primer Premier 5.0 software. PCR primers were synthesized by Sangon Biotech (Shanghai, China) and purified by HPLC. All primers used in this study are listed in Table 1. The plasmids pmirGLO, pTR-UF11, pTR-EGFP-BFGF, pAAV-iRFP713, and bacterial SURE2 were preserved in our laboratory. The AAV-293 and MDA-MB-231 cell lines were purchased from the CAS Typical Culture Collections Committee cell library. Female BALB/c nude mice were purchased from Slaccas Experimental Animal Co. Ltd. (Shanghai, China). Animal welfare and experimental procedures were carried out strictly according to national laws and covered by license from the China government.

Table 1.

The Primer Used in This Study

Primer name	Primer sequence (5′–3′)	Restriction site
Luciferase-F	GGAACGAATTCGCCACCATGGAAGATGCCAAAAACAT	EcoRI
Luciferase-R	CCTGAAGCAGCAGCAGCAGAGCCGCCGCCGCCCACGGCGATCTTGCCGCC
EGFP-F	GCTGCTGCTGCTTCAGGAGGAGGAGGATCCGTGAGCAAGGGCGAGGAG
EGFP-R	AGGGGCGGATCTGTCGACTTACTTGTACAGCTCGTC
IRES-F	AGTAAGTCGACAGATCCGCCCCTCTCCCT
IRES-R	ATGGTGGCACTAGTTGTGGCCATATTATCATCGT
Neo-F	CCACAACTAGTGCCACCATGATTGAACAAGATGGATT
Neo-R	GGAACCTCGAGTCAGAAGAACTCGTCAAGAA	XhoI

F, forward primer; R, reverse primer.

Construction of recombinant vectors

Luciferase was amplified with pmirGLO as the template, Neo was amplified with pTR-UF11 as the template, and IRES and EGFP were amplified with pTR-EGFP-BFGF as the template. The PCR was performed as follows: 3 min at 96°C (15 s at 95°C and 2 min at 72°C) × 25 cycles, and 5 min at 72°C. After extraction and purification, PCR products were amplified as templates by fusion PCR. The PCR was performed as follows: 3 min at 96°C (15 s at 95°C and 5 min at 72°C) × 25 cycles, and 10 min at 72°C. After extraction and purification, the recombinant PCR product was digested with EcoRI and XhoI. The pAAV vector was generated by digesting the plasmid pAAV-iRFP713 with restriction enzymes EcoRI and XhoI, followed by the insertion of the PCR product into the pAAV vector. The mixture was then transformed into SURE2 competent cells, and the new plasmid was confirmed by DNA sequencing (Fig. 1).

FIG. 1.

Schematic diagram of recombination fusion PCR. PCR, polymerase chain reaction.

Functional identification of recombinant plasmids

AAV-293 cells were cultured in 24-well plates with the DMEM containing 10% fetal bovine serum (FBS) at 37°C and 5% CO₂. When the cells reached 70% confluence, the recombinant plasmid pAAV-Luciferase-EGFP-Neo was used for cell transfection. After transfection for 24 h, a 510/20-nm BP filter was used for the detection of EGFP. After transfection for 48 h, cells were lysed and proteins were collected. We added 100 μL of luciferin solution to the protein samples to detect the catalytic reaction of luciferase in a multifunctional microplate reader.

Screening and detection of MDA-MB-231 monoclonal stable cells

We obtained a linear plasmid containing a distinct structure (inverted terminal repeat [ITR]) after pAAV-Luciferase-EGFP-Neo was digested with the restriction enzyme ApaLI. MDA-MB-231 cells were cultured in 24-well plates with the L15 medium containing 10% FBS at 37°C. After transfection of MDA-MB-231 cells with the linear plasmid, the L15 medium was replaced with a fresh medium containing 600 μg/mL of G418 every 72 h. We obtained a polyclonal stable cell line 7 weeks later after the death of all nonfluorescent cells. Then, we selected single cells that were in good condition by limiting the dilution assay to expand and establish a monoclonal stable cell line.

Cell migration assay

The MDA-MB-231 negative control cell line, the polyclonal stable cell line, and the monoclonal stable cell line were cultured in six-well plates with a medium containing 10% FBS at 37°C. The medium was replaced with the L15 basic medium until the cells reached 90% confluence. A sterile 200-μL pipet tip was used to scratch four separate wounds through the cells 12 h later. The cells were gently rinsed twice with PBS to remove floating cells and incubated in the L15 medium containing 1% FBS at 37°C. Images of the scratches were taken using an inverted microscope at 10× magnification at 0, 24, and 48 h of incubation.

Cell proliferation analysis by MTT assay

The MDA-MB-231 negative control cell line, the polyclonal stable cell line, and the monoclonal stable cell line were cultured in 96-well plates with 200 μL medium containing 10% FBS at 37°C and incubated for 6 days. Every 24 h, the medium was replaced with 180 μL of L15 basic medium containing 20 μL MTT (5 mg/mL), which was used for the MTT assay. Finally, growth curves were generated based on the results.

Establishment of a tumor model in nude mice

Initially, 1 × 10⁷ MDA-MB-231 polyclonal stable cells and monoclonal stable cells in 200 μL PBS were injected into the fourth pair of the mammary fat pad of mouse at 10 weeks of age. Mice were maintained in a pathogen-free environment. When the tumor was formed, the growth of tumor cells was observed by in vivo imaging. We injected 200 μL of luciferin (150 mg/kg) into the abdominal cavity of each nude mouse before observation and then performed in vivo imaging within 30 min. Imaging was performed by using a Kodak In-Vivo MultiSpectral Imaging System FX (2D; Kodak, Rochester, NY) in epifluorescence mode equipped with 570 nm filters for emission. Images were taken with 60 s exposure. The signal intensity is represented by radiance and encoded by pseudocolors on the iRFP image. All animal experiments were performed in a facility approved by the Association for Assessment and Accreditation of Laboratory Animal Care.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 5.0 software (GraphPad Software, Inc., San Diego, CA). One-way repeated measure ANOVA and Student's t-test were used. Differences were considered significant at p values of <0.05.

Results

Construction of the recombinant plasmid pAAV-Luciferase-EGFP-Neo by fusion PCR

The fragments luciferase, EGFP, IRES, and Neo were amplified using the corresponding primers (Fig. 2A). The PCR product Luciferase-EGFP-IRES-Neo was obtained by the fusion PCR method (Fig. 2B). The PCR product Luciferase-EGFP-IRES-neo and the vector pAAV-MCS were digested by restriction enzymes EcoRI and XhoI. Finally, we obtained the recombinant plasmid pAAV-Luciferase-EGFP-Neo by linking the enzyme-digested products. The correct plasmid was verified by digestion with SmaI (Fig. 2C). We observed for green fluorescence under a fluorescence microscope after transfection of AAV-293 cells with the recombinant vector (Fig. 2D). Then, we used a multifunctional microplate reader to verify the expression of the luciferase gene. The fluorescence intensity from transfected cells are up to 1.5 × 10⁶, while the negative control cells are just 1.5 × 10³ (Fig. 2E).

FIG. 2.

Construction and verification of pAAV-Luciferase-EGFP-Neo (bars: 100 μm). (A) PCR amplification of fragments of luciferase, EGFP, IRES, and Neo. Lane M: DNA Marker; Lane 1: Luciferase; Lane 2: EGFP; Lane 3: IRES; Lane 4: Neo. (B) Fusion PCR. Lane M: DNA marker; Lane 1: Luciferase-EGFP-IRES-Neo obtained by fusion PCR. (C) Digestion verification of the recombinant plasmid pAAV-Luciferase-EGFP-IRES-Neo. Lane M: DNA marker; Lane 1: undigested; Lane 2: digested with SmaI. (D) Expression of EGFP in AAV-293 cells after pAAV-Luciferase-EGFP-IRES-Neo transfection. (E) Expression of luciferase in AAV-293 cells after pAAV-Luciferase-EGFP-IRES-Neo transfection. Data represented mean ± SD of triplicate (*P < 0.05).

High and stable expression of the monoclonal stable cell line

After transfection of MDA-MB-231 cells with the recombinant plasmid, 1% of the cells were successfully labeled (Fig. 3A). After resistance screening, all live cells expressed green fluorescence, and MDA-MB-231 polyclonal stable cells were generated. Each cell clone was different in cell morphology and light intensity because of random plasmid integration (Fig. 3B). The green fluorescence intensity of the monoclonal cell line was moderate, stable, and robust based on flow cytometry detection (Fig. 3C, D). The number of detected fluorescent cells was far below the actual number of cells since most of the cell clones expressed proteins with low fluorescence (Fig. 3E). The monoclonal stable cells had high protein expression; and fluorescence intensity increased with an increase in the number of cells, which can distinguish the emission intensity from <10 cells (Fig. 3F).

FIG. 3.

Screening and identification of MDA-MB-231 stable cells (bars: 100 μm). (A) Transduction of MDA-MB-31 cells with the pAAV-Luciferase-EGFP-IRES-Neo vector. (B) Fluorescence imaging of the MDA-MB-231 polyclonal stable cell line. (C) Fluorescence imaging of the MDA-MB-231 monoclonal stable cell line. (D) Detection of fluorescent cells in the MDA-MB-231 polyclonal stable cell line by flow cytometric analysis. (E) Detection of fluorescent cells in the MDA-MB-231 monoclonal stable cell line by flow cytometric analysis. (F) Correlation between bioluminescence intensity and cell number in the MDA-MB-231 monoclonal stable cell line. Data represented mean ± SD of triplicate (*P < 0.05).

Difference in migration ability between the negative control cell line, polyclonal stable cell line, and monoclonal cell line

Cell wounds of the negative control cell line and monoclonal stable cell were healed at 72 h when the cell wounds were no longer visible. The migration speed of the polyclonal stable cell line was significantly lower than the negative control cell line. The results revealed that the migration speed of the monoclonal stable cell line was not different from the negative control cell line, and it was more easily observed in the fluorescence background (Fig. 4). The cell migration rate was determined as 1−(migration distance of cells/initial distance of cells).

FIG. 4.

Migration ability of MDA-MB-231 cells (bars: 100 μm). (A) Migration ability of the MDA-MB-231 negative control cell line. (B) Migration ability of the MDA-MB-231 polyclonal stable cell line. (C) Migration ability of the MDA-MB-231 monoclonal stable cell line. (D) Migration rate of the MDA-MB-231 negative control cell, polyclonal stable cell line, and monoclonal stable cell line. Data represented mean ± SD of triplicate (*P < 0.05).

Measurement and comparison of cell proliferation capacity

The proliferation of MDA-MB-231 polyclonal stable cells was lower than the nontransfected cells; however, the monoclonal stable cell line was significantly different from the nontransfected MDA-MB-231 cell line (Fig. 5). It demonstrated that random integration of foreign genes into chromosomes may restrict cell proliferation.

FIG. 5.

Proliferation curves of the MDA-MB-231 negative control cell line, monoclonal stable cell line, and polyclonal stable cell line. Data represented mean ± SD of triplicate (*P < 0.05).

Superiority of the monoclonal stable cell line over the polyclonal stable cell line in a tumor model

The nude mice had raised vesicles after subcutaneous injection. A week later, nude mice injected with MDA-MB-231 monoclonal stable cells had distinct hard lumps after the vesicles were no longer visible. This was a solid tumor where we can detect luciferase bioluminescence by in vivo imaging (Fig. 6A). The mice injected with polyclonal stable cells had no obvious lumps; nevertheless, we were able to detect luciferase bioluminescence (Fig. 6B). However, we found that a tumor formed 3 weeks later, indicating that the system could detect potential tumors. Comparison of tumor measurement by vernier caliper and in vivo imaging, the latter is more sensitive. At these settings, even though very few cancer cells existed, it detected a strong signal of over 200-fold above background in potential tumor. And small tumor (mouse 1) gave a robust signal that was 25-fold higher than potential tumor (mouse 2) of comparable size (Fig. 6C).

FIG. 6.

In vivo bioluminescence of tumorigenesis after subcutaneous inoculation. (A) In vivo bioluminescence after injecting monoclonal stable cells. (B) In vivo bioluminescence after injecting polyclonal stable cells. (C) Quantification of (A) and (B). Data represented mean ± SD of triplicate (*P < 0.05).

Discussion

A human breast cancer tumor model is an important tool for studying the mechanisms of breast cancer, evaluating drug efficacy, and developing new therapies. An ideal model is not only similar to the pathological and physiological processes of the human body but also convenient for producing, observing, and monitoring the characteristics of tumor cells in different conditions (Gu et al., 2016). The MDA-MB-231 cell line, which is the most commonly used cell line in human breast cancer research, has characteristics including fast growth, high metastasis, and high invasion. Because of these characteristics, patients often miss the optimal time for diagnosis and treatment (Abdel-Zaher and Eldeib, 2016). An optical molecular imaging technology would make it possible to observe the growth, invasion, and metastasis of tumor cells in vitro and in vivo.

With regard to the establishment of breast cancer tumor models, previous studies have generally used lentiviruses to infect breast cancer cells. However, the random chromosomal integration of lentiviruses may lead to changes in cell growth characteristics (Wang et al., 2012; Li et al., 2015). In this study, we used a vector with an ITR structure that increases the probability of insertion into the chromosome 19 locus, which is widely acknowledged as a safe harbor locus (Liu et al., 2015; Torres et al., 2015). ITR structures are easily lost in the transformation process; thus, we used SURE2 cells that lack the ability to remove complex structures, which can greatly reduce the probability of losing the ITR structure (D'Costa et al., 2016). The ITR structure cannot be detected by PCR and sequencing, but can be verified by the restriction enzyme SmaI. Since potential issues of viral vectors are still unclear, we did not use a viral packaging system (Huang and Chen, 2014). This study used nonviral vectors to deliver nucleic acids and obtained a polyclonal stable cell line by resistance screening.

In terms of vector construction, we designed the terminal sequence of one PCR product, which had the same sequence homology as the starting sequence of another PCR product according to the fusion PCR principle. All linker sequences of the PCR products were complementary pairs, and they resulted in a double-stranded fragment in the PCR annealing process and formed a complete double-stranded fragment in the extension of both primers. We cloned three genes and a structure from a plurality of different plasmids and then spliced them into a large fragment in vitro. Finally, we obtained the gene expression vector pAAV-Luciferase-EGFP-Neo, which contains the ITR structure, by molecular cloning. In comparison with traditional molecular cloning methods, our method solved complex problems involving multiple transformations, screening, and long experimental period encountered when using these genes with different vectors.

We derived a stable monoclonal cell line from MDA-MB-231 polyclonal cells and compared various aspects of the polyclonal stable cell line and monoclonal stable cell line. We found great differences in the protein expression of the polyclonal stable cell line and monoclonal stable cell line. The fluorescence intensities of some clones were very low and less than the self-luminous threshold, resulting in a fluorescent cell ratio of only around 50%, which can lead to a large experimental error in flow cytometry (Nedbal et al., 2015). However, the fluorescence intensity of monoclonal cells was bright and stable, and the fluorescent cell ratio detected by flow cytometry was near 100%. In addition, our study demonstrated that the proliferation and migration abilities of the polyclonal cell line were lower than the monoclonal cell line. A few months later, part of the polyclonal stable cells lost their fluorescence, while the monoclonal stable cells were continuously cultured for over a year with no changes in vivo and in vitro.

The luciferase and EGFP genes were inserted into the pAAV vector, and these two proteins were used to construct a fusion protein by a peptide linker. After transfection and screening, we generated the monoclonal stable cell line, and its proliferation and migration abilities were not different from the unlabeled cell line. In addition, we found that the characteristics of the monoclonal stable cell line were better than the polyclonal stable cell line in all aspects. Finally, we established a breast cancer tumor model in nude mice in which we were able to detect a tumor in the early stages before solid tumor formation. We not only can easily study the dynamic growth of tumor cells through real-time observation of the EGFP protein in vitro but also can easily calculate the number of cells by analyzing the fluorescence emission intensity. The study demonstrated that the monoclonal stable cell line could be an ideal reporter gene system with good stability and accuracy in terms of simultaneous imaging in vivo and in vitro, and it provided a solid foundation for further research in breast cancer.

Footnotes

Acknowledgments

This work was partially supported by the National Natural Science Foundation of China [Grant No. 311101101] and the Natural Science Foundation of the Jiangsu Province [grant number: BK20130342&BK20140381] and Suzhou Municipal Science and Technology Program [grant number: ZXY201432].

Disclosure Statement

No competing financial interests exist.

References

Abdel-Zaher

A.M.

, and Eldeib

A.M.

(2016). Breast cancer classification using deep belief networks. Expert Syst Appl, 46, 139–144.

Ademuyiwa

F.O.

, Gao

, Hao

, Morgensztern

, Aft

R.L.

, Ma

C.X.

, and Ellis

M.J.

(2015). US breast cancer mortality trends in young women according to race. Cancer, 121, 1469–1476.

Bahmani

, and Hosseinkhani

(2016). Increase of segmental mobility through insertion of a flexible liker in split point of firefly luciferase. Int J Biol Macromol, 055, 1–9.

D'Costa

, Blouin

, Broucque

, Penaud-Budloo

, François

, Perez

I.C.

, Le Bec

, Moullier

, Snyder

R.O.

, and Ayuso

E.F.

(2016). Practical utilization of recombinant AAV vector reference standards: focus on vector genomes titration by free ITR qPCR. Mol Ther, 16019, 1–9.

, Li

, Peng

, Zhang

, Liang

, Zhao

, Qi

, Wang

, and Guo

(2016). Autophagy-related prognostic signature for breast cancer. Mol Carcinog, 55, 292–299.

Huang

, and Chen

(2014). Research progress on adeno-associated virus. Biol Disaster Sci, 37, 333–336.

Kijanka

M.M.

, Brussel

A.S.

, Van der Wall

, Mali

W.P.

, Van Diest

P.J.

, van Bergen En Henegouwen

P.M.

, and Oliveira

(2016). Optical imaging of pre-invasive breast cancer with a combination of VHHs targeting CAIX and HER2 increases contrast and facilitates tumour characterization. EJNMMI Res, 6, 1–13.

, Wei

, Li

, Zhang

, and Xiao

(2015). Down-regulating HIF-1α by lentivirus-mediated shRNA for therapy of triple negative breast cancer. Cancer Biol Ther, 16, 866–875.

T.M.

, Lan

W.J.

, Huang

, Zhang

, and Liu

X.M.

(2016). Establishment and identification of the near-infrared fluorescence labeled exosomes in breast cancer cell lines. Hereditas, 38, 427–435.

10.

Liu

, Siriwon

, and Rohrs

J.A.

(2015). Generation of targeted adeno-associated virus (AAV) vectors for human gene therapy. Curr Pharm Design, 21, 3248–3256.

11.

Nedbal

, Visitkul

, Ortiz-Zapater

, Weitsman

, Chana

, Matthews

D.R.

, Ng

, and Ameer-Beg

S.M.

(2015). Time-domain microfluidic fluorescence lifetime flow cytometry for high-throughput Förster resonance energy transfer screening. Cytom Part A, 87, 104–118.

12.

Torres

, Garcia

, Jimenez

, Rodriguez

, and Ramirez

(2014). An integration-defective lentivirus-based resource for site-specific targeting of an edited safe-harbour locus in the human genome. Gene Ther, 21, 343–352.

13.

Wang

, Xie

, Ren

, Xia

, Zhang

, and He

(2012). Establishment of a bioluminescent MDA-MB-231 cell line for human triple-negative breast cancer research. Oncol Rep, 27, 1981–1989.