ComplexEye: a multi-lens array microscope for high-throughput embedded immune cell migration analysis

Abstract

Autonomous migration is essential for the function of immune cells such as neutrophils and plays an important role in numerous diseases. The ability to routinely measure or target it would offer a wealth of clinical applications. Video microscopy of live cells is ideal for migration analysis, but cannot be performed at sufficiently high-throughput (HT). Here we introduce ComplexEye, an array microscope with 16 independent aberration-corrected glass lenses spaced at the pitch of a 96-well plate to produce high-resolution movies of migrating cells. With the system, we enable HT migration analysis of immune cells in 96- and 384-well plates with very energy-efficient performance. We demonstrate that the system can measure multiple clinical samples simultaneously. Furthermore, we screen 1000 compounds and identify 17 modifiers of migration in human neutrophils in just 4 days, a task that requires 60-times longer with a conventional video microscope. ComplexEye thus opens the field of phenotypic HT migration screens and enables routine migration analysis for the clinical setting.

Fig. 1. Optical performance of the ComplexEye lens.

a Left: Photograph of a standard 20x microscope lens. Right: The ComplexEye lens with its schematic optical path. The dimensions above show the difference in diameter of the lenses. b Images acquired with a standard 20x lens (left) and just minutes later with the ComplexEye lens (right) of the identical cell sample consisting of Cutaneous Melanoma (CM cells) and neutrophil granulocytes. This experiment was repeated three times with similar results. c Images acquired with a standard 20x lens (left) and the ComplexEye lens (right) of a 250 µm grid in a Neubauer chamber. This experiment was repeated twice with similar results. The size of the FOV in the respective captured images are displayed. d Above: Schematic structure of the 16 individual microscope units without the imagers. Below: Photograph of the top view of the ComplexEye in its working setup. LED Light Emitting Diode.

Fig. 2. The ComplexEye: a multi-lens microscope for high-throughput, embedded cell migration analysis.

a Above: Top view of the lenses with one lens removed to allow a zoom onto the underlying CMOS-imager. Below: photographs of the circuit board with imagers and FPGA from top and bottom. b 3D model of the ComplexEye excluding the dividing chamber (see also supplementary movies 1 and 2). c Left: Schematic overview of the ComplexEye components and structure inside the tempered cabin with the division into dry and humid climate zones. Right: Photograph of the real-world ComplexEye corresponding to the view shown in the schematic figure (left). FPGA Field Programmable Gate Array.

Fig. 3. ComplexEye performance and comparison with a conventional microscope.

a 16-well measurement of freshly prepared human peripheral blood neutrophils from a single donor (n = 1) migrating in response to the indicated stimuli. Each condition was prepared in quadruplicates in independent wells of a 96-well plate and imaged for one hour (8 s between frames) with ComplexEye using a 96-well plate. The same cells were measured simultaneously on a commercial video microscope from Leica. Shown are the tracking results of the four independent runs on ComplexEye as dots (light gray bars) and the parallel run on the Leica as single bar with dot (dark gray bars). Orange lines indicate published reference data of n = 25 healthy individuals. b 64-well recording of neutrophils (one hour, 8 s between frames) from a single donor (n = 1) on ComplexEye using a 384-well plate with the indicated stimuli in 16-fold repetitions. c 64-well measurement as in (b) but with cells simultaneously prepared from 16 donors (n = 16). Each cell preparation was measured with four different conditions as indicated. d Comparison of energy consumption between ComplexEye and a commercial system from Leica for the generation of a one-hour movie with 8 s between frames. The striped bar demonstrates the 64-well measurement at the ComplexEye, the light gray bar the 16-well measurement at the ComplexEye and the dark gray bar the 4-well measurement at the Leica. Statistical significances were calculated via Kruskal–Wallis test with multiple comparisons. Data are presented as median values ± interquartile range. ***p < 0.001. Source data are provided as a Source Data file.

Fig. 4. ComplexEye high-throughput screening of migration modifying compounds.

a Experimental setup of the screening assay. Briefly, neutrophils from human blood were isolated and plated on a 384-well plate, treated with one of the 1000 compounds from a library of known bioactives and stimulated with fMLP. Neutrophil motility was then recorded simultaneously in 64 wells of a 384-well plate for one hour (8 s between frames) using ComplexEye. Afterwards the motility was analyzed via single cell tracking. b Data represent 1000 movies, ~800 tracks/movie and show the impact of 1000 compounds screened in 17 rounds, each round with three controls (PBS, DMSO and fMLP or fMLP/DMSO). The heatmap shows each round with 64-wells with the relative speed of imaged neutrophils indicated as color code compared to the fMLP-control in that run (artificially set to 1.0). Compounds that reduced the speed are shown in green-blue (low speed). Indicated gray wells were non-evaluable due to production residues of the 384-well plates inhibiting clear sight of the cells. Source data are provided as a Source Data file.

Fig. 5. ComplexEye high-throughput screening identifies neutrophil migration modifiers.

a Relative speed data with 17 compounds reducing motility by more than 40% compared to fMLP as detected by the screen illustrated in Fig. 4. Compound R14D2 had less effects on speed, but strongly affected the cell shape. b Relative activity data with 27 compounds reducing the number of migrating neutrophils by more than 20% compared to fMLP. c Sorting of inhibitory compounds into classes according to their effect on speed and activity of migrating neutrophils. In every square the left vertical line is for relative speed and the right vertical line is for relative activity. Class 1: compounds strongly decreasing the speed whereas the number of migrating cells was not affected (lines with blue dots). Class 2: compounds strongly decreasing the number of moving cells without affecting their speed (lines with red dots). Class 3: compounds decreasing both, speed and activity (lines with gray dots). d Comparison of neutrophil morphology between fMLP-treated cells and cells treated with fMLP and the indicated compounds. This experiment was not repeated, Polar plots show migration tracks of all cells in the experiment normalized to one common center. Rings in polar plots define 100 µm distances. The scale bar is given as 50 µm. All datasets shown in this figure are based on cells from a total of eight single donors analyzed within 4 consecutive days. Each condition was measured once (n = 1). Source data are provided as a Source Data file.