Fully automated cellular-resolution vertebrate screening platform with parallel animal processing

Abstract

The zebrafish larva is an optically-transparent vertebrate model with complex organs that is widely used to study genetics, developmental biology, and to model various human diseases. In this article, we present a set of novel technologies that significantly increase the throughput and capabilities of our previously described vertebrate automated screening technology (VAST). We developed a robust multi-thread system that can simultaneously process multiple animals. System throughput is limited only by the image acquisition speed rather than by the fluidic or mechanical processes. We developed image recognition algorithms that fully automate manipulation of animals, including orienting and positioning regions of interest within the microscope's field of view. We also identified the optimal capillary materials for high-resolution, distortion-free, low-background imaging of zebrafish larvae.

Fig. 1

Multi-thread Vertebrate Automated Screening Technology (VAST). The platform consists of three subsystems that operate simultaneously: loading, imaging, and unloading. Larvae are automatically loaded to the platform from individual wells of a mesh-filter multiwell plate positioned by a motorized x y stage. The mesh-filter insert allows easy transfer of larvae into the system. In order to avoid a drop in the fluid level within multiwell plate (while larva is being aspirated by the loading nozzle), a circulator is set nearby the loading nozzle (inset A) that consists of a fluid source and an aspirator where the tip of the aspirator is slightly elevated with respect to the tip of the fluid source. A zebrafish discriminator with a brightfield and a fluorescence photodetection system (inset B) discriminates the passage of fluorescent larvae from non-fluorescent ones, air bubbles and debris. Two step motors hold a capillary immersed in a water bath along its axis of rotation; this assembly is mounted on a three-axis position stage (not shown) and held between an upright microscope and an inverted microscope. A multifocal confocal head with a cooled electron-multiplying charge-coupled device (EM-CCD) camera and a second large-area charge-coupled device (CCD) are used for high-speed confocal and wide-field fluorescence imaging, respectively. A high-speed CCD camera connected to the inverted microscope allows rapid bright-field imaging for positioning and orienting the larvae.

Fig. 2

Timings of multi-thread processing of zebrafish. Blue, red, and green bars indicate the processes comprising loading, imaging, and unloading, respectively. (a) The time required for each handling step (n = 192). Multiwell plate loading process includes positioning the multiwell plate and the loading nozzle, and aspirating a larva into the nozzle from the multiwell plate. Transportation time is the duration it takes to move a larva within the fluidic tube from one point to another. (b) Multi-thread operation: The system simultaneously performs loading, imaging, and unloading operations with three different larvae. As a result, the overall processing time is dictated solely by the duration of the slowest process (i.e. imaging; 9.6 sec), not by the total duration of all processes.

Fig. 3

Zebrafish discriminator. (a) Schematic representation of the zebrafish discriminator. The system is composed of a bright-field discrimination system and a fluorescence-activated zebrafish sorter. (b) Schematic representation of the mechanism of action of the bright-field discrimination system. By combining both the scattered and transmitted signals the system distinguishes a zebrafish larva from air bubbles and debris. (c) Detection and discrimination reliabilities at increasing flow rates. The reliability is near 100 % at normal operating speeds of 330 μL/s. (n=150).

Fig. 4

Identification of the angle and position of larvae. (a) Dorsally and ventrally oriented larvae. Red arrows indicate the dominant morphological features for distinguishing dorsally vs. ventrally views. It is too difficult to reliably determine the larva’s orientation based solely on comparisons with dorsal and ventral reference images. (b) The larva is rotated along its longitudinal axis through a full 360^° and snapshots are acquired at two degree increments at 180 frames per second. (c) The system then correlates the images with a library of prerecorded dorsal and lateral images from stage-matched control larvae (inset). The blue and red curves in the radar chart are results of the image correlations with the dorsal and lateral templates. The blue, green, and red arrows indicate the orientations of maximum correlations with the dorsally, ventrally and laterally orientated templates, respectively. The reliability of the algorithm is 99 % (n=100 larvae)

Fig. 5

Analysis of capillary materials for distortion-free low-background high-resolution imaging. (a) From left to right, the point-spread functions (PSF) with different materials/conditions; no capillary, PTFE capillary, FEP capillary, borosilicate glass capillary, and ultra-thin borosilicate glass capillary. The wall thicknesses of capillaries are 250 μm, 150 μm, 170 μm and 10 μm, respectively (shown in brackets). The ultra-thin glass capillary produces the least image distortion. Scale bar is 1 μm. (b) Radial (blue) and axial (red) resolutions of candidate materials. PTFE was not included due to the significant non-Gaussian PSF it produces. (c) Autofluorescence analysis of candidate materials. PTFE and FEP capillaries cause significant autofluorescence.