Single-cell optical stretching and sorting into an integrated microfluidic device

Optical stretching emerged in the recent years as a very promising technique for testing mechanical properties of single cells in an efficient and non-destructive way. Several papers already demonstrated that many different pathologies result in changes in cellular deformability, and optical stretching can thus be used as a technique to evaluate the health of unlabelled cells [1]. Up to now, what is still unreported in the current scientific literature is the ability to use a single integrated chip to analyze the stretching response of cells, and to use the measured deformation as a parameter for selective sorting. In this summary we briefly describe the design, fabrication and validation of a similar device. A double-Y microfluidic chip integrating different couples of optical waveguides, symmetrically placed with respect to the channel, is realized in a glass substrate by the FLICE technique [2]. In order to avoid surface roughness problems, which would reduce the chip imaging quality [3], a specific fabrication strategy was devised. The strong improvement observed in the quality of realized surfaces is probably due to the highly ellipsoidal shape of the writing voxel. After chip fabrication and pigtailing, the device is connected to a 10-W fiber laser (emitting @ 1070 nm), and to tubings (see setup in Fig.1, left panel). System validation is obtained by injecting at one of the input a buffer solution, while at the other a suspension containing two different cellular lines of human melanoma (metastatic A375-P, and highly-metastatic A375-MC2). In the starting suspension the ratio of the cells belonging to the two lines is 1:1. Thanks to the fact that the two lines show the same cellular size, but a slightly different optical deformability (see center panel of Fig.1) we aimed at selecting a population of cells with an enriched content of A375-MC2 cells. To achieve this result we defined a "deformation threshold" (e.g. 9%, as shown in Fig.1 center) and we sorted to a "selected-cells" vial only those cells showing a deformation higher than the threshold. By repeatedly performing "stretching & sorting" tests, also considering different "deformation thresholds", we verified that both the experimentally obtained enrichment-factor and the cells "acceptance rate" (i.e. the percentage of cells in the input sample exhibiting a deformability higher than the threshold) closely match the theoretically expected values, thus demonstrating the efficiency and reliability of the proposed chip.