Quantitative phase imaging (QPI) measures the growth rate of individual cells by quantifying changes in mass versus time. Here, we use the breast cancer cell lines MCF-7, BT-474, and MDA-MB-231 to validate QPI as a multiparametric approach for determining response to single-agent therapies. Our method allows for rapid determination of drug sensitivity, cytotoxicity, heterogeneity, and time of response for up to 100,000 individual cells or small clusters in a single experiment. We find that QPI EC50 values are concordant with CellTiter-Glo (CTG), a gold standard metabolic endpoint assay. In addition, we apply multiparametric QPI to characterize cytostatic/cytotoxic and rapid/slow responses and track the emergence of resistant subpopulations. Thus, QPI reveals dynamic changes in response heterogeneity in addition to average population responses, a key advantage over endpoint viability or metabolic assays. Overall, multiparametric QPI reveals a rich picture of cell growth by capturing the dynamics of single-cell responses to candidate therapies.
The optical properties of polymer materials used for microfluidic device fabrication can impact device performance when used for optical measurements. In particular, conventional polymer materials used for microfluidic devices have a large difference in refractive index relative to aqueous media generally used for biomedical applications. This can create artifacts when used for microscopy-based assays. Fluorination can reduce polymer refractive index, but at the cost of reduced adhesion, creating issues with device bonding. Here, we present a novel fabrication technique for bonding microfluidic devices made of NOA1348, which is a fluorinated, UV-curable polymer with a refractive index similar to that of water, to a glass substrate. This technique is compatible with soft lithography techniques, making this approach readily integrated into existing microfabrication workflows. We also demonstrate that this material is compatible with quantitative phase imaging, which we used to validate the refractive index of the material post-fabrication. Finally, we demonstrate the use of this material with a novel image processing approach to precisely quantify the mass of cells in the microchannel without the use of cell segmentation or tracking. The novel image processing approach combined with this low refractive index material eliminates an important source of error, allowing for high-precision measurements of cell mass with a coefficient of variance of 1%.
SignificanceImaging changes in subcellular structure is critical to understanding cell behavior but labeling can be impractical for some specimens and may induce artifacts. Although darkfield microscopy can reveal internal cell structures, it often produces strong signals at cell edges that obscure intracellular details. By optically eliminating the edge signal from darkfield images, we can resolve and quantify changes to cell structure without labeling.AimWe introduce a computational darkfield imaging approach named quadrant darkfield (QDF) to separate smaller cellular features from large structures, enabling label-free imaging of cell organelles and structures in living cells.ApproachUsing a programmable LED array as the illumination source, we vary the direction of illumination to encode additional information about the feature size within cells. This is possible due to the varying levels of directional scattering produced by features based on their sizes relative to the wavelength of light used.ResultsQDF successfully resolved small cellular features without interference from larger structures. QDF signal is more consistent during cell shape changes than traditional darkfield. QDF signals correlate with flow cytometry side scatter measurements, effectively differentiating cells by organelle content.ConclusionsQDF imaging enhances the study of subcellular structures in living cells, offering improved quantification of organelle content compared with darkfield without labels. This method can be simultaneously performed with other techniques such as quantitative phase imaging to generate a multidimensional picture of living cells in real-time.
10 September 2018 in Journal of Visualized Experiments
The use of microfluidic devices has emerged as a defining tool for biomedical applications. When combined with modern microscopy techniques, these devices can be implemented as part of a robust platform capable of making simultaneous complementary measurements. The primary challenge created by the combination of these two techniques is the mismatch in refractive index between the materials traditionally used to make microfluidic devices and the aqueous solutions typically used in biomedicine. This mismatch can create optical artifacts near the channel or device edges. One solution is to reduce the refractive index of the material used to fabricate the device by using a fluorinated polymer such as MY133-V2000 whose refractive index is similar to that of water (n = 1.33). Here, the construction of a microfluidic device made out of MY133-V2000 using soft lithography techniques is demonstrated, using O2 plasma in conjunction with an acrylic holder to increase the adhesion between the MY133-V2000 fabricated device and the polydimethylsiloxane (PDMS) substrate. The device is then tested by incubating it filled with cell culture media for 24 h to demonstrate the ability of the device to maintain cell culture conditions during the course of a typical imaging experiment. Finally, quantitative phase microscopy (QPM) is used to measure the distribution of mass within the live adherent cells in the microchannel. This way, the increased precision, enabled by fabricating the device from a low index of refraction polymer such as MY133-V2000 in lieu of traditional soft lithography materials such as PDMS, is demonstrated. Overall, this approach for fabricating microfluidic devices can be readily integrated into existing soft lithography workflows in order to reduce optical artifacts and increase measurement precision.
The field of microfluidics provides a robust toolkit for biomedical applications such as disease diagnosis and drug discovery, especially when combined with advanced microscopy techniques. An important challenge facing the combination of microfluidic devices with quantitative microscopy techniques, such as quantitative phase imaging (QPI), is the mismatch in refractive index between channel structures and aqueous media. This mismatch can introduce artifacts at material interfaces due to scattering and, in the case of QPI, phase unwrapping. We will show that these issues can be addressed through the use of MY133-V2000, a UV-curable, fluorinated polymer with a low refractive index similar to water (n = 1.33). MY133-V2000 can be fabricated into microfluidic devices using standard soft lithography techniques based on an SU-8 or polydimethylsiloxane (PDMS) mold. The addition of fluorine reduces the overall polarizability of the material, lowering refractive index. However, this introduces a new challenge due to the typically low adhesion of fluorinated polymers. We will discuss device integration and packaging strategies to overcome this limitation. Using QPI, we will demonstrate measurement of the distribution of cell biomass in live, adherent cells, both in the center of the channel and at the interface with microchannel structures, to demonstrate the dramatic reduction in artifacts due to the matching indices of refraction. We will also discuss applications to other microscopy techniques, including fluorescence. MY133-V2000 therefore provides QPI researchers with the opportunity to leverage the advantages of microfluidics for a diverse range of biomedical applications.
Functional precision oncology seeks to match patients with effective therapies by empirically testing patient-derived samples for drug sensitivity in the laboratory. However, existing approaches require significant sample expansion time and expense prior to analysis, rendering them impractical for routine clinical testing. Quantitative phase imaging (QPI) provides a potential path forward by directly measuring responses at single cell resolution without the need for extensive sample expansion. In previous work, we demonstrated that multiple, independent parameters of cellular response to therapeutic agents can be derived from QPI data, an approach we call multiparametric QPI (mQPI). Here, we demonstrate application of mQPI using cells from patient derived xenograft organoid (PDxO) models, as well as cells viably frozen direct from patients. Using mQPI with breast cancer PDxO models, we uncover distinct drug responses for cells originating from different anatomic sites in the same patient and resolve cellular heterogeneity of response in a model of acquired therapeutic resistance. We also show that mQPI can detect drug responses in viably frozen primary patient samples, either direct from thaw or after a short term expansion of only 2 weeks. Overall, these data provide proof-of-principle for application of mQPI to a range of sample types, including cryopreserved material direct from patients. This underscores the clinical potential of mQPI as a time- and materials-efficient alternative to current methods in functional precision oncology.