Motion microscopy for label-free detection of circulating breast tumor cells

https://doi.org/10.1016/j.bios.2020.112131Get rights and content

Highlights

  • Motion microscopy amplifies micro motions of breast tumor cells, which we referred to herein as cellular trail.

  • Cellular trails were associated with cell status, surface proteins, and flow rates rather than mitochondrial activity.

  • Cellular trails were present only around tumor cells in blood samples obtained from breast cancer patients.

Abstract

Circulating tumor cells (CTCs) are cancer cells that have been shed from a primary tumor and circulate in the bloodstream during progression of cancer. They may thus serve as circulating biomarkers that can predict, diagnose and guide therapy. Moreover, phenotypic and genotypic analysis of CTCs can facilitate prospective assessment of mutations and enable personalized treatment. A number of methodologies based on biological and physical differences between circulating tumor and non-tumor cells have been developed over the past few years. However, these methods did not have sufficient sensitivity or specificity. In this work, a remote analysis protocol was designed using motion microscopy that amplifies cellular micro motions in a captured video by re-rendering small motions to generate extreme magnified visuals to detect dynamic motions that are not otherwise visible by naked eye. Intriguingly, motion microscopy demonstrated fluctuations around breast tumor cells, which we referred to herein as cellular trail. Phenomena of cellular trail mostly emerged between 0.5 and 1.5 Hz on amplified video images. Interestingly, cellular trails were associated with cell surface proteins and flow rates rather than mitochondrial activity. Moreover, cellular trails were present only around circulating tumor cells from individuals with breast cancer under conditions of 20–30 μm/s and 0.5–1.5 Hz. Thus, motion microscopy based CTC detection method can offer a valuable supplementary diagnostic tool for assessment of drug efficacy and identifying physical characteristics of tumor cells for further research.

Introduction

The metastatic process is complex with CTCs ultimately seeding in a distant site where they adapt to a new environment by neoangiogenesis (Lambert et al., 2017). CTCs in most individuals with cancer are lower than 10 to 100 cells per milliliter of blood (He et al., 2008). CTCs have been identified in many cancers, including breast, lung, prostate, pancreas, stomach, and colon (Bidard et al., 2018; Dong et al., 2002; He et al., 2008; Tsai et al., 2016). Several techniques have been used and applied for CTC detection (Supplementary Table 1). These techniques are composed of density-based separation, microfilters, microfluidic sorting, immunoaffinity, or combination of these methods. Traditionally, density gradient centrifugation is employed to enrich the mononucleocyte fraction, which includes CTCs due to their similar buoyant density. However this method can be time consuming (Rosenberg et al., 2002; Campton et al., 2015). On the other hand, CTC isolation using microfilters has been demonstrated to be efficient, by exploiting the size of CTCs which are significantly larger than circulating blood cells. However, the track etching of microfilters often results in fusion of two or more pores, resulting in lower CTC capture efficiency to 50–60% (Gallant et al., 2013; Harouaka et al., 2014). An alternative technology based on microfluidic system uses deterministic lateral displacement with continuous high-throughput to separate the whole blood from isolate nucleated cells, including CTCs and white blood cells, efficiently with minimal damage in an antigen-independent manner (Hvichia et al., 2016; Hosokawa et al., 2010). Additionally, CTC surface epithelial marker enrichment such as EpCAM, HER2, EGFR can be used for immunomagnetic separation (Miltenyi et al., 1990). Moreover, negative enrichment, through absence of CD45 on CTCs but presence on normal cells, can also be used (Sajay et al., 2014). This method is expensive owing to cost of antibodies conjugated to magnetic beads with wide ranging yield (9–90%) due to variable expression of surface markers. Recently, combinations of technologies have been developed to overcome various shortcomings of these detection methods (Yoon et al., 2013; Park et al., 2017). More recently, vibrational profiling of cancer cells have been applied using atomic force microscopy (Nelson et al., 2017). However, there are technological difficulties due to the requirement of proximity of cantilevers to tumor cells. We describe here the application of motion microscopy whereby vibrational profiling of CTCs can be measured remotely.

The motion microscopy is a computational tool that quantifies micro motions from videos by generating a new image whereby the motions are magnified enough to be visualized by human eye (Adiv, 1985; Hurlburt and Jaffey, 2015; Sellon et al., 2015; Wadhwa et al., 2017). Through this methodology, otherwise invisible micro-movements are recorded which can subsequently be visualized by pixel camera. Therefore, more pixels covering the object of interest would yield better signals for extraction. For every pixel at location (x, y), time t, scale r, and orientation θ, spatial local phase information was combined in different sub-band of frames using the least squares objective function (Wadhwa et al., 2017), argminu,viAri,θi2[(φri,θix,φri,θiy)(u,v)Δφri,θi]2.

Arguments have been suppressed for readability: Ari,θi (x, y, t) and φri,θi (x, y, t) are the spatial local amplitude and phase of a steerable pyramid representation of the image, and u (x, y, t) and v (x, y, t) are the horizontal and vertical motions at every pixel (Wadhwa et al., 2017). The original purpose of motion microscopy was to measure vibrations of building structures or earthquakes, and the detectable movement was 0.3 nm–100 nm in macro-spheres (Adiv, 1985; Wadhwa et al., 2017). However, this has not been described in micro-spheres, such as in areas of cellular properties. Applying the image from a microscope, we expected to be able to detect vibrations between 1 nm and 0.003 nm using a motion microscope and examined whether it could be applied to detect fluctuations of tumor cells. We therefore hypothesized that motion microscopy can be used to detect tumor cells. As CTCs of breast cancer have been studied extensively and can be a promising potential for liquid biopsy (Bidard et al., 2018), we analyzed wavelength profiles in breast cancer cells using motion microscopy.

Section snippets

Cell lines and culture

MCF-7 (HTB-22, ATCC, USA), MDA-MB-231 (HTB-26, ATCC, USA), and SK-BR-3 (HTB-30, ATCC, USA) were maintained in Dulbecco's MEM (11885, Gibco, USA) with 10% fetal bovine serum (16000044, Gibco, USA). All cultures were maintained at 37 °C under an atmosphere of 95 % O2 and 5 % CO2. Before each experiment, cells were detached from the surface of culture flasks by 0.05 % trypsin (15400054, GIbco, USA). The resulting cell suspensions were then centrifuged at 118×g and resuspended with

Results and discussion

Detection of CTCs is an important technique for the initial diagnosis of cancer metastasis and for monitoring the response of various cancer therapies (Bidard et al., 2018; He et al., 2008; Tsai et al., 2016; ). A number of different techniques have been developed for CTC detection to date. However, many methods have pitfalls with suboptimal efficiencies and specificities. For example, detecting CTCs based on size differences is insufficient due to lack of ideal biomaterial for selection based

Conclusion

Micro motions of cells can reveal important dynamic changes under various biological conditions. Motion microscopy facilitates visualization of tumor cell micro movements and formation of cellular trails using a digital camera. We show in this study that cellular trails were observed specifically in tumor cells and CTCs were detected from human blood samples with a consistent detection rate.

CRediT authorship contribution statement

Hyueyun Kim: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Visualization. Young-Ho Ahn: Conceptualization. Bom Sahn Kim: Conceptualization. Sanghui Park: Conceptualization. Joo Chun Yoon: Conceptualization. Junbeom Park: Conceptualization. Chang Mo Moon: Conceptualization. Dong-Ryeol Ryu: Conceptualization. Jihee Lee Kang: Conceptualization. Ji Ha Choi: Conceptualization. Eun-Mi Park: Conceptualization. Kyung Eun Lee: Conceptualization. Minna Woo: Writing -

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: I have numerous grants from Korea government and Ewha Womans University.

Acknowledgements

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government (MIST) (2019R1C1C1003384 and 2010-0027945) and Ewha Womans University Research Grant of 2018. We thank medical students (Sunbeen Kim, Seoyeong Jung, and Eun young Baek) of Ewha University for helping to culture tumor cells.

References (27)

  • Q. Dong et al.

    Hematogenous dissemination of lung cancer cells during surgery: quantitative detection by flow cytometry and prognostic significance

    Lung Canc.

    (2002)
  • A.W. Lambert et al.

    Emerging biological principles of metastasis

    Cell

    (2017)
  • G. Adiv

    Determining three-dimensional motion and structure from optical flow generated by several moving objects

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1985)
  • F.C. Bidard et al.

    Circulating tumor cells in breast cancer patients treated by neoadjuvant chemotherapy: a meta-analysis

    J. Natl. Canc. Inst.

    (2018)
  • S. Byun et al.

    Characterizing deformability and surface friction of cancer cells

    Proc. Natl. Acad. Sci. U. S. A.

    (2013)
  • D.E. Campton et al.

    High-recovery visual identification and single-cell retrieval of circulating tumor cells for genomic analysis using a dual-technology platform integrated with automated immunofluorescence staining

    BMC Canc.

    (2015)
  • F.A. Coumans et al.

    Filtration parameters influencing circulating tumor cell enrichment from whole blood

    PloS One

    (2013)
  • J.N. Gallant et al.

    Predicting therapy response in live tumor cells isolated with the flexible micro spring array device

    Cell Cycle

    (2013)
  • D. Harari et al.

    Molecular mechanisms underlying ErbB2/HER2 action in breast cancer

    Oncogene

    (2000)
  • R.A. Harouaka et al.

    Flexible micro spring array device for high-throughput enrichment of viable circulating tumor cells

    Clin. Chem.

    (2014)
  • W. He et al.

    Quantitation of circulating tumor cells in blood samples from ovarian and prostate cancer patients using tumor-specific fluorescent ligands

    Int. J. Canc.

    (2008)
  • M. Hosokawa et al.

    Size-selective microcavity array for rapid and efficient detection of circulating tumor cells

    Anal. Chem.

    (2010)
  • N. Hurlburt et al.

    A spectral optical flow method for determining velocities from digital imagery

    Earth Sci. India

    (2015)
  • Cited by (0)

    View full text