Open Access
Add to BibSonomyAbstract
Light sheet fluorescence microscopy has recently emerged as the technique of choice for obtaining high quality 3D images of whole organisms/embryos with low photodamage and fast acquisition rates. Here we present an open source unified implementation based on Arduino and Micromanager, which is capable of operating Light Sheet Microscopes for automatized 3D high-throughput imaging on three-dimensional cell cultures and model organisms like zebrafish, oriented to massive drug screening.
© 2015 Optical Society of America
1. Introduction
Whole organisms have recently emerged as promising models for drug screening. Between them, zebrafish [1,2] has been proposed as an alternative to mammalian models for cancer cell studies, due to its easy handle and reproduction, its transparency, mainly on early stages of development, and its relatively simple genetic modification, to assess the efficacy and toxicity of antitumor drugs. The possibility of individual testing of patients’ particular cancer type on zebrafish could yield, in the near future, to tailor-made therapeutics [3]. Alternatively, three dimensional (3D) cell cultures, derived from human cells, are becoming a powerful tool to address biological relevant problems, from cell proliferation, cell death, differentiation, or metabolism using both patient-derived cells and cell lines in an attempt to achieve a realistic representation of the physiological environment, closer to the observed in tissues [4–6]. The analysis of these complex, 3D, highly differentiated and dense samples needs advanced imaging techniques in order to extract relevant information for the evaluation of drug response mechanisms. Although confocal microscopy [7] can provide highly detailed 3D cell structures, it is not suited to be used as a screening tool due to its low throughput. On the other hand, modern flow cytometry can deliver measurement speeds exceeding 70,000 cells per second [8]. However, its high throughput (HT) has drawbacks such as sacrificing the spatial information which is essential for analyzing three dimensional structures. In this context, the development of new imaging techniques, such as Light-Sheet Fluorescence Microscopy [9, 10] (LSFM), can contribute to increase the potential and widespread adoption of these models in biological research and preclinical stages of drug development. LSFM seems to be an ideal tool, providing reduced light dosage (since only those parts of a cell or tissue in the focal plane are illuminated) and high resolution volumetric information at fast acquisition rates. It has already been used to image cell aggregates and zebrafish showing its great potential, not only on fixed samples but also on-line/live monitoring of cell events such as apoptosis and calcium signaling [6, 11, 12].
In this paper we have explored new designs for 3D imaging based on LSFM combined with flow cytometry and fluidics approaches providing not only high throughput screening (HTS) but high spatial resolution, which can be comparable with confocal microscopy. Using the open software and hardware framework of OpenSpinMicroscopy platform [13] we have extended the functionalities of the system by adding a sample aspiration and positioning system, including motorized syringes and multi-well plate loading.
In order to facilitate the automated sample loading, essential for HTS, we have designed special chambers for the different types of objectives. Across those chambers, at 45 degrees, a Fluorinated Ethylene Propylene (FEP) tube positioned at the intersection of the illumination and detection focal plane transports the samples, which can be aspired and pushed back and forward with an Arduino controlled stepper motor attached to a syringe. Our home-made controller allows performing micro-steps up to 0.225°, creating a 2 microns steps movement of the sample (for a FEP tube of 1mm inner diameter), and to detect sample passage using photodiodes, in order to automatize the acquisition process. The system also permits to control a secondary camera, which can be placed at different locations, thus increasing the flexibility of the system. We have also created a dedicated java plugin for Micromanager [14] software, which enables to easily control sample positioning and data acquisition from a single window, creating a modular open source platform for HTS, here tested on both, zebrafish and 3D cancer cell cultures.
2. Materials and methods
2.1 Description of the set up
A scheme of the SPIM-Fluid (SPIM, Selective Plane Illumination Microscopy) high throughput light-sheet microscope set-up is shown in Fig. 1(a).
Fig. 1 (a) Schematic of the SPIM-Fluid set up. The different elements of the system are: lasers (L), dichroic mirrors (DM), filter wheels (FW), shutter (S), galvo mirror (GM), telescope (T), cylindrical lens (CL), cameras (CAM) and photo-detection system (PDS). In the basic system CAM 1 is used for the Flow-Mode and CAM 2 for the Vertical-Mode. Alternatively, CAM2 can be used for double side Flow-Mode or fast double color Flow-Mode are plotted in gray. The HT-sample management system consist on bottom rounded multiwall plates mounted on a XY stage, and a third motor controlling the FEP tip height. (b) Detailed scheme of the automatic sample mounting system based on an Arduino controlled injector with a photodiode system and Fluorinated Ethylene Propylene (FEP) tubes. (c) Front panel of the Micromanager SPIM-Fluid plugin. (d) Arduino controlled photodiode circuit for automatized sample positioning control. (e) Detailed scheme of the sample chamber, where the FEP tube crosses at 45°. Laser illumination is shown green and FEP tube in blue. For detailed info check our website.
The illumination block is a home-made laser combiner including three lasers lines (L): 473 nm (DPSSL MBL-III-473-50), 561 nm (Coherent OBIS 561-50 LS) and 642 nm (Vortran Stradus 165 mW) and two dichroic beam splitters (DM): 505DCLP and FF593-Di03-25x36 (Semrock). Excitation laser lines are selected using an Arduino controlled filter wheel (FW1) with three different filters (Semrock 473/10, 561/10 and 640/8). The laser illumination is controlled using a shutter (S) (Uniblitz electronics LS3T2) and a varying neutral density filter. The laser scanning is carried in the vertical axis using an Arduino controlled galvanometric mirror (GM) (6210H Cambridge Technologies) which optical plane is conjugated with the back focal aperture of an objective lens (Plan Fluor 4 × 0.13 WD17.4 mm) using a 3.5 × telescope system consisting in a 50 mm and a 180 mm achromatic lenses (Thorlabs). Alternatively, light sheet can be created using a 50 mm cylindrical lens placed 5 cm before the back focal aperture of the illumination objective. For detection, an air objective (Nikon 10x 0.3NA WD 16.7 mm), placed perpendicularly to the excitation plane, is used to collect fluorescence emission. Excitation light is rejected using emission filters placed in infinity space before the camera, with filters mounted in a second automatic filter wheel (FW2), consisting of the following: HQ 535/70m-2p, HQ 580/25m-2p, HQ 620/90m-2p, and ET 700/75. Finally a 200 mm tube lens creates the image on the chip of the sCMOS cameras (Hamamatsu Orca-Flash4).
Sample loading is performed using a home-made pump system consisting in a manual injector (Eppendorf CellTram) connected to a stepper motor (Mercury) (see Fig. 1(b)). The syringe is connected to FEP tubes of different diameter (0.5, 0.8, 1 µm) depending on the sample size. As FEP has a refractive index close to water (1.341), if surrounded by water, it is possible to image sample inside the tube without affecting optical quality. We have created specialized chambers in acrylic plastic filled with water for different types of detection objectives (Fig. 1(e)). A FEP tube crosses the chamber with an angle of nearly 45° to the illumination axis at a focal position of the detection objective. This configuration permits fully access to the sample from five sides of the chamber with objective lenses (two side illumination, two side detection and vertical detection). In that way, sample mounting procedure for some specific samples, such as zebrafish embryos and larvae or 3D-cell cultures, gets less restrictive and extremely facilitated compared with agarose embedding methods, since samples can travel through the FEP and flow in a controlled fashion across the light-sheet plane and detection focal plane. For the sake of comparison, conventional sample mounting using agarose needs for each sample over 15 minutes preparation before agar solidification, insertion of the holder onto the system and positioning on the camera field of view. With the system presented here few seconds are enough to load hundreds of samples onto the tube.
We have also designed a motorized loading platform in order to fully automatize the sample management for large scale experiment. The system controls the stepper motor of the injector in order to load sample through FEP tubes from a multi-well plate in complex experiments, like for example for drug screening studies. Multi-well plates are mounted in a XY stage consisting in two stepper motor stages (Thorlabs LTS150/M). A third DC motor (Thorlabs MTS25/M-Z8 with TDC001 controller) moves the FEP tube tip vertically at every position, in order to load the sample. Multi-well positions are defined using the plugin user interface (XY List button) (Fig. 1(c)). To speed up the automated mounting system, i.e. bring the sample to the detection zone faster, we use a photo-detection system (PDS) to control sample positioning, which circuit is shown in Fig. 1(d). Sample loading is a complex task which requires different programs at different acquisition stages of the loading/unloading process. Initial aspiration to the tube has to be fast, while sample positioning during acquisition has to be soft. To minimize delays, the approach to the photodetector unit and the sample chamber should be performed at higher speeds. Moreover, the aspirated volume needs to be controlled in order to deposit back the sample at its original well. In order to control externally the functionalities of this mode, we stablished a sequence of STATE events on the Arduino firmware, which can be easily reconfigured. The sequence on multi-well plate acquisition, as well as full descriptions of the apparatus, Arduino controllers, acquisition software source code and instructions to build it are available through our webpage (https://sites.google.com/site/openspinmicroscopy/).
The configuration here proposed allows us to perform large scale experiments in two complementary ways. The flow detection arm, siting horizontally on the optical bench, permits to continuously image the sample plane, recoding images sequentially as the sample travels through this plane. Using the so called F-Mode (Flow Mode) the Arduino controller provides small rotation angles to the syringe motor (up to 0.225°) synchronized with the camera for precise control of the sample flow imaging. Alternatively, using the second detection arm, placed vertically to the sample plane (Vertical or V-Mode), samples are just positioned on the camera field of view inside the FEP tube, with the pump system. Once there, scanning of the sample is performed without any further movement of the sample, by the synchronized movement of the light sheet by the galvo and the repositioning of the upper water dipping detection objective (Olympus x10 NA0.3 WD 3.5 mm), which is mounted on a DC motor (Thorlabs MTS25/M-Z8 with TDC001 controller). Using a 50 mm cylindrical lens (CL) (Thorlabs) (or alternatively a secondary galvo), we create a plane of illumination that can be scanned using the primary galvo mirror in coordination with the motorized focusing system.
In order to increase the system’s functionality in F-Mode, CAM2 can be mounted on different positions depending on the experimental needs, as shown in Fig. 1(a) (in gray). In the basic design, multicolor imaging is performed by filter wheels selection, introducing undesired delays. In order to obtain fast two color imaging of relatively transparent samples, such as the 3D cell cultures presented here, the fluorescent signal can be split using a dichroic mirror (DM3) and imaged simultaneously with two cameras. Moreover for thick, high scattering samples, such as zebrafish embryos and larvae, double side imaging is able to properly reconstruct the full sample volume. Due to space limitations, the secondary arm is mounted vertically perpendicular to the optical table in our set up. An additional objective (Nikon 10x 0.3NA WD 16.7 mm) is used to collect fluorescence signal and a mirror is used to deflect the light onto the tube lens.
2.2 Sample positioning controller
A micro-stepper shield (Sparkfun EasyDriver) is connected to an Arduino UNO board which controls the rotation of a stepper motor connected to the injector through a home-made holder. With this configuration we can perform rotations steps from 0.225° to 360°.
To properly place the sample on focus we designed a flow control system using photodiodes (OPT301; Texas Instruments). The system uses a blue LED as a light source and a photodiode (LD), located around the FEP tube, to detect the difference in transmitted light. To reduce the noise, LED and photodiode are assembled into a small unit. Using the circuit shown in Fig. 1(d) the measured amplified voltage feeds an Arduino analog input. On the automatic system operation, the aspiration stepper motor will rotate until the photodiode signal is below a user defined threshold. When the sample is detected, acquisition mode is triggered, moving the sample to the focus position in order to start the data acquisition. If needed, different LED intensities and the photodiode sensitivities can be achieved replacing resistors R1 and R4.
2.3 Bioreactor cell cultures
2D cell culture
MCF7 cells expressing RFP were kindly provided by Professor Cathrin Brisken (EPFL, Switzerland) within the scope of the PREDECT consortium. Cell expansion was performed in DMEM High Glucose supplemented with 1% Penicillin-Streptomycin, 4 mM Glutamax, 1 mM Sodium Pyruvate and 10% FBS. Human Dermal Fibroblasts (HDF), from Innoprot, were passaged once weekly at a seeding density of 0.5 x 104 cell/cm2, in Iscove's Modified Dulbecco's Medium (IMDM) supplemented with 1% Penicillin-Streptomycin and 10% Fetal Bovine Serum (FBS) (all from Life Technologies). Cells were cultured in static conditions in an incubator at 37°C with humidified atmosphere containing 5% CO2.
3D cell culture
3D cell culture was performed as described in [16]. Briefly, MCF-7 cells were aggregated in a stirred-tank for 24h and further encapsulated in 3 mL of 1.1% (w/v) of Ultrapure Ca2+ MVG alginate (UP MVG NovaMatrix, Pronova Biomedical, Oslo, Norway) dissolved in NaCl 0.9% (w/v) solution together with HDFs in a 1:1 proportion (co-cultures). Microencapsulation was performed using an electrostatic bead generator (Nisco VarV1) as previously described [15]. Encapsulated co-cultures were kept in a stirred-tank, in a humidified incubator with 5% CO2, with 50% medium exchange every 3 days.
Sample preparation (Whole mount Immunofluorescence using SPIM)
Culture samples were collected at day 5 of culture. Encapsulated cells were fixed with 4% paraformaldehyde (PFA) + 4% Sucrose for 20 minutes at RT and processed for immunostaining as described in [15]; anti-vimentin antibody (Abcam) was used for fibroblast detection.
2.4 Zebrafish
Fli:GFP (expressing GFP on the vascular system) and Tg(β-actin:HRAS-EGFP) (expressing membrane-bound GFP) transgenic zebrafish embryos were used to test the performance of the SPIM-Fluid microscope on its different acquisition modes. Due to the lack of fluorescence in Fli:GFP strain during the first 24 hours we used HRAS:GFP zebrafish to visualize embryonic stages. Zebrafish larvae were anesthetized with tricaine methane sulfonate (MS222) while no further protocol was follow on embryos.
For Fig. 3(i) we used a Fli-EGFP transgenic fish showing fluorescence in the vascular system (green) and human tumor cells labeled with the lipophilic cell tracer carbo-cyanine red dye, Dil (magenta). Sample was injected with human tumor cell, which after 24 hours propagate through the body. The sample was then fixed using 2 ml of 4% PFA and stained with Dil. We acknowledge Dr. A. C. Borges, Dr. M. Ferreira and Dr. R. Fior for fish sample.
All the procedures here reported were approved by the IGC ethical committee and the Direção Geral de Alimentação e Veterinaria (DGAV) of Portugal.
3. Results
The system was tested with two models developed for pre-clinical drug screening: 3D cell aggregates and zebrafish larvae and embryos. For the 3D cell culture, we used a novel cancercell model consisting on tumor cell aggregates co-cultured with fibroblasts entrapped in alginate microcapsules [16]. These types of co-cultures could be extremely useful to understand the role of fibroblasts on cancer cell proliferation, phenotypic alterations over time, and mechanisms of drug resistance [16]. The developed F-Mode allows us to perform, in a fast way, automatic imaging and analysis of a statistically relevant number of samples, providing 3D information about the number, size and distribution of the cell sub-populations in co-cultures. Visualization 1 illustrates in both, bright-field and fluorescence modes, the flow of the microcapsules in FEP tubes as they pass through the light sheet plane. The heterogeneity of these microcapsules in terms of number of aggregates, morphology and distribution inside the microcapsules can be observed in Fig. 2(a)-2(c), where the maximum projection of three different co-cultures is shown. Compared with other techniques, SPIM-Fluid provides not only fluorescence based quantitative information but also 3D imaging, allowing the acquisition of more accurate data of the different samples. The 3D segmentation capability also allows correct determination of features such as the number of aggregates per microcapsule, which could have been miscalculated (due to occultation behind other aggregates) using methods such as bright field or wide-field microscopy. In order to demonstrate the usefulness and efficiency of the methods here presented, we have quantified, using semi-automatic FIJI macros, the number of cancer cell aggregates per capsule, the number of fibroblasts per capsule, the average aggregate volume size and also phenotypic features. Namely, 120 aggregates were classified according to circularity in regular and irregular shaped and the ratio between both determined (Fig. 2(d)-2(g)) over a population of 40 alginate microcapsules.
Fig. 2 (a-c) Maximum projection of three different 3D co-culture of cancer cell aggregates (green) and fibroblasts (magenta) entrapped in alginate microcapsules. Visualization 1 shows the microcapsules flow inside FEP tubes, in bright-field and fluorescence modes. The system allows the semi-automatic quantitative analysis of different parameters of the 3D co-cultures, such as (d) number of aggregates (n = 40 microcapsules; mean = 3.00; standard deviation (SD) = 2.15) (e) number of fibroblast per capsule (n = 40 microcapsules; mean = 489.90; SD = 126.70), (f) estimation of cell aggregate volumes (n = 120 cell aggregates; mean = 3.15 × 10−03; SD = 3.02 × 10−03) and (g) phenotypic classification of aggregates according to their circularity. The analysis was performed on a population of 40 microcapsules, produced and collected on the same conditions without any randomization or blinding process; only microcapsules without cancer cell aggregates were removed from the analysis. Scale bar: 200 µm.
The interest in using zebrafish as a model organism has been expanding rapidly and, because of its transparency on early stages, imaging is a preferred screening tool. However the practical size of 3D screens has been mainly limited by the speed of manual manipulation. For these reasons we extended our system in order to fulfill the requirements of such delicate and highly scattering samples. Using the SPIM-Fluid system it is possible to replace tedious manual manipulation of zebrafish larvae by automatically, precisely and reproducibly position larvae on the camera field of view for HTS while maintaining viability of the fish for further testing. Samples are loaded from multi-well plates using an Arduino custom made microcontroller, positioned on the camera field of view, imaged using one of the two available modes and placed back onto the original well.
Examples of different developmental stages and acquisition modes are shown in Fig. 3. We used Fli:GFP and Hras:GFP mutant embryos passing through a 1 mm diameter FEP tube and x10 detection objective lens, providing a resolution of 0.65 µm and a field of view of 1.33 mm. Figure 2(a)-2(f) were acquired in F-Mode, i.e., pushing the sample with the aspiration system and sequentially acquiring images at a frame rate of 3 images per second.
Fig. 3 Maximum projections of zebrafish embryos and larvae volume reconstructions: (a-c) three of the twelve Hras:GFP embryos at early developmental stages automatically loaded from a multi-well plate. 90° projection of an anesthetized Fli:GFP larvae acquired with F-Mode (d) single and (e) double side mode. (f) 6 day old Hras:GFP zebrafish imaged in F-Mode. (g) Same larvae imaged with V-Mode after stitching of seven 3D stacks. (h) 8 day Fli:GFP larvae acquired with V-Mode. The image is build up from four different 3D stacks. (i) Vascular system (green) of a 6-day fixed Fli:GFP larvae injected with human cancer cells, stained with Dil (magenta). The image was acquired using single side F-Mode. Sample courtesy of Dr. M. Ferreira and Dr. R. Fior. Volume reconstructions of Fli:GFP zebrafish are shown in Visualization 2 while Visualization 3 presents the development of the vascular system on Fli:GFP zebrafish using V-Mode. Scale bar: 200 microns.
As described above, we have created a sample loading management system in order to perform HTS analysis on large zebrafish populations, automatizing the sample mounting from multi-well plates. We have tested with a population of 12 Hras:GFP embryos, expressing membrane bound GFP, imaged in double side F-Mode. Figure 3(a)-3(c) shows three of those embryos. Following a user defined protocol, loaded on the Arduino microcontroller memory, samples are automatically aspired from each well while approach to the imaging plane is controlled with a photodiode system. After imaging, ten of the samples continue developing normally while two of them collapsed.
To properly reconstruct the entire fish body double side illumination and detection must be used in most of the cases. As can be observed in Fig. 3(d)-3(e) double side detection allows to combine the information of both cameras on a single image, revealing features of the Fli:GFP zebrafish that otherwise would be hidden, providing full volumetric information of the sample with reasonable resolution. The sample flow control performed a 0.9 degree rotation for each plane, corresponding to approximately 3 µm movements of the sample. Alternatively, the V-Mode provides higher resolution images, avoiding artifacts due to sample translation. In that mode, sample is positioned on the field of view of the camera, and images are acquired with a water dipping x10 objective that refocuses the image following the scanned light sheet. For adult zebrafish the syringe pump moves the sample in a controlled way to acquire sequentially 3D stacks that can be afterwards stitched.
A comparison between flow and vertical modes is displayed in Fig. 3(f)-3(g), where the same Hras:GFP zebrafish is imaged, showing good correlation. In order to obtain Fig. 3(g) seven consecutive 3D-stacks were recorded as the sample flow performed 18 degrees rotations between stacks. After stitching we created the maximum projection of the data set here shown. In order to better visualize the high quality volumetric information obtained with the different modes Visualization 2 shows the volume reconstruction of Fli:GFP zebrafish in Fig. 3(e) and 3(h). It is worthy to mention that the total amount of data gets reduced over 2 fold on F-Mode (1.1 Gb), compared with V-Mode (2.5 Gb), something to take into account when large scale experiments are considered. It is also important to notice that the two modes can be complementary, allowing to perform a quick analysis on a large population using the flow imaging mode, helping to select the candidate for a higher resolution long term experiment using the vertical mode. Moreover, using V-Mode, the system also allows to obtain high quality 3D time-lapse recordings. As an example, the vascular development on Fli:GFP over 7 and 14 hours, in two different embryos, is shown in Visualization 3.
Finally, as a proof of concept of the great potentialities of this microscope on cancer therapies, a 6-day Fli:GFP larvae (green) injected with human cancer cells (magenta) was imaged using flow fast dual color mode (Fig. 3(i)). Fluorescence signal is collected with the same objective lens and then spitted using a dichroic mirror. Two cameras record an image of the same plane before the sample is pushed to the next position by the syringe pump. This configuration allows eliminating lead times associated with filter wheels rotation. The system is able to precisely localize, in 3D, cancer cells propagated through the zebrafish body 24 hours after injection.
4. Discussion
LSFM pivots around the idea of building a microscope around the sample [17], forcing to rethink the way of classical sample mounting. Current LFSM microscopes, due to their revolutionary architecture, revealed to be complex instruments requiring special sample preparation and manipulation with limited flexibility. In the majority of the systems, relatively large samples need to be mounted one by one, embedded in agar gel, limiting the number of samples analyzed and making it unpractical for large scale experimental application. Furthermore, it has also been proved to compromise correct development of zebrafish embryos [18]. New protocols of sample mounting using FEP polymers (with a refractive index close to water) have been recently presented [18] as a promising alternative to maintain sample liability, while keeping image quality. Recently, new advances have been presented for combining light sheet imaging approaches to microfluidics on 3D spheroids [19], where the sample is immobilized and different agents circulate around the sample; and on flow cytometry [20] applied to algae imaging for water quality control [21]. In this work we have designed a new configuration, using FEP tubes at 45°, for sample loading and positioning that overcome the described limitations, facilitating mounting of large set of samples, and opening the door to large volume screenings on zebrafish and 3D cell cultures. In that way, samples can be maintained in a controlled environment, such as microfluidics bioreactors [22] or multi-well plates with different drug dosages, being only temporally loaded for imaging.
Following our open-source philosophy, through our webpage we provide the source code of a new plugin able not only to control the different components of the system, but also to perform new complex imaging modalities. To the previous imaging configurations (Selective Plane Illumination Microscopy (SPIM), Digital Scanned Light-sheet Microscopy (DSLM) and Optical Projection Tomography (OPT)) we added two more modes: F-Mode and V-Mode. F-Mode provides low resolution high speed imaging of the samples as they flow through the FEP tube, crossing a static light-sheet plane, pushed by a controlled syringe pump. V-Mode uses a camera placed vertically above the sample chamber, while the detection objective is mounted on a translational stage that moves in coordination with the galvo mirror, in order to refocus the selected plane. This allows obtaining high-resolution 3D images keeping the sample static at the FEP tube on the field of view, while the aspiration system allows large sample compositions by stitching. We have also shown that using V-Mode it is possible to track developmental processes on zebrafish embryos for extended periods of time. It is worthy to mention that sample drift may represent a problem for long term imaging. This can be minimized by avoiding air bubbles inside the FEP tube and keeping the loading tip immersed in water during acquisition in order to prevent water evaporation. Moreover, the plugin also provides, in parallel with acquisition, simple quantifications using custom made macros, such as, overall cell counting on 3D cell cultures.
As the design of the set-up is created on a modular way, the secondary camera can be placed on different positions allowing, depending on the sample under analysis, to obtain double side or fast dual color imaging on F-Mode. As described previously, the system here presented enables fast and automated loading and imaging of a big amount of samples. However acquisition speed is still limited by the motors delays to 4 frames per second. The imaging procedure can be speed up using advanced digital piezo-electric pump systems. Moreover, it is worthy to mention that this FEP configuration can be easily adapted to other microscope configurations. Nowadays different LSFM configuration, such as single objective SPIM [23], double inverted SPIM [24] or decoupled illumination detection LSFM [25], provide imaging speeds of tens of volumes per second, optically scanning the sample. The combination of FEP mounting, V-Mode detection and any of those fast acquisition approaches may allow to increase the throughput of the system.
5. Conclusions
With the system here presented, sample mounting becomes much easier than other configurations, and allow performing a high variety of experiments, from large population analysis, 3D imaging of big organisms by stitching and time lapse recordings. Compared with agarose mounting, as the FEP tube remains always in focus, it enables to easily choose the right samples amongst many, based on 3D fluorescence information prior to image acquisition. Samples can be loaded either from single micro-tube or multi-well plates and, in the case of 3D cell cultures, it will also allow to automatize loading directly from the culture system, where samples are produced, providing more realistic evaluation checkpoints. High throughput light-sheet microscopy for fast sample screening can be extended to different types of neural, cancer and hepatic models as well as other organisms to address target validation of newly developed therapeutics and their toxicological effects.
Acknowledgments
The authors acknowledge support from Fundação para a Ciência e Tecnologia, Portugal –SFRH/BD/52208/2013, SFRH/BPD/80717/2011 and EXPL/BBB-IMG/0363/2013. We gratefully acknowledge Dr. Cathrin Brisken, for the supply of the MCF-7-rfp cell line within the scope of the IMI-funded project PREDECT (grant agreement n° 115188). We also thank R. Fior and M. Ferreira (Instituto Gulbenkian de Ciência) and A.C. Borges (Fish Facility) for providing zebrafish embryos and larvae. This research has also received support from the Innovative Medicines Initiative Joint Undertaking (IMI grant agreement n° 115188); resources composed of financial contribution from EU - FP7 and EFPIA companies in kind contribution and co-founded MINECO/FEDER project BIO2014-59614-JIN.
References and links
1. C. A. Lessman, “The developing zebrafish (danio rerio): a vertebrate model for high-throughput screening of chemical libraries,” Birth Defects Res. C Embryo Today 93(3), 268–280 (2011). [CrossRef] [PubMed]
2. J. R. Mathias, M. T. Saxena, and J. S. Mumm, “Advances in zebrafish chemical screening technologies,” Future Med. Chem. 4(14), 1811–1822 (2012). [CrossRef] [PubMed]
3. B. Pruvot, A. Jacquel, N. Droin, P. Auberger, D. Bouscary, J. Tamburini, M. Muller, M. Fontenay, J. Chluba, and E. Solary, “Leukemic cell xenograft in zebrafish embryo for investigating drug efficacy,” Haematologica 96(4), 612–616 (2011). [CrossRef] [PubMed]
4. J. Friedrich, C. Seidel, R. Ebner, and L. A. Kunz-Schughart, “Spheroid-based drug screen: considerations and practical approach,” Nat. Protoc. 4(3), 309–324 (2009). [CrossRef] [PubMed]
5. F. Pampaloni, E. G. Reynaud, and E. H. K. Stelzer, “The third dimension bridges the gap between cell culture and live tissue,” Nat. Rev. Mol. Cell Biol. 8(10), 839–845 (2007). [CrossRef] [PubMed]
6. F. Pampaloni, B. J. Chang, and E. H. K. Stelzer, “Light sheet-based fluorescence microscopy (LSFM) for the quantitative imaging of cells and tissues,” Cell Tissue Res. 360(1), 129–141 (2015). [CrossRef] [PubMed]
7. C. Pardo-Martin, T. Y. Chang, B. K. Koo, C. L. Gilleland, S. C. Wasserman, and M. F. Yanik, “High-throughput in vivo vertebrate screening,” Nat. Methods 7(8), 634–636 (2010). [CrossRef] [PubMed]
8. S. H. Cho, J. M. Godin, C. Chen, W. Qiao, H. Lee, and Y. Lo, “Recent advancements in optofluidicflow cytometer,” Biomicrofluidics 4(4), 043001 (2010). [CrossRef]
9. J. Huisken, J. Swoger, F. Del Bene, J. Wittbrodt, and E. H. K. Stelzer, “Optical sectioning deep inside live embryos by selective plane illumination microscopy,” Science 305(5686), 1007–1009 (2004). [CrossRef] [PubMed]
10. P. J. Verveer, J. Swoger, F. Pampaloni, K. Greger, M. Marcello, and E. H. K. Stelzer, “High-resolution three-dimensional imaging of large specimens with light sheet-based microscopy,” Nat. Methods 4(4), 311–313 (2007). [PubMed]
11. E. J. Gualda, D. Simão, C. Pinto, P. M. Alves, and C. Brito, “Imaging of human differentiated 3D neural aggregates using light sheet fluorescence microscopy,” Front. Cell. Neurosci. 8, 221 (2014). [CrossRef] [PubMed]
12. M. B. Ahrens, M. B. Orger, D. N. Robson, J. M. Li, and P. J. Keller, “Whole-brain functional imaging at cellular resolution using light-sheet microscopy,” Nat. Methods 10(5), 413–420 (2013). [CrossRef] [PubMed]
13. E. J. Gualda, T. Vale, P. Almada, J. A. Feijó, G. G. Martins, and N. Moreno, “OpenSpinMicroscopy: an open-source integrated microscopy platform,” Nat. Methods 10(7), 599–600 (2013). [CrossRef] [PubMed]
14. A. D. Edelstein, M. A. Tsuchida, N. Amodaj, H. Pinkard, R. D. Vale, and N. Stuurman, “Advanced methods of microscope control using μManager software,” J. Biol. Methods 1(2), 10 (2014). [CrossRef] [PubMed]
15. S. P. Rebelo, R. Costa, M. Estrada, V. Shevchenko, C. Brito, and P. M. Alves, “HepaRG microencapsulated spheroids in DMSO-free culture: novel culturing approaches for enhanced xenobiotic and biosynthetic metabolism,” Arch. Toxicol. 89, 1347–1358 (2014). [PubMed]
16. M. F. Estrada, Biomaterials, Under submission (2015).
17. E. G. Reynaud, J. Peychl, J. Huisken, and P. Tomancak, “Guide to light-sheet microscopy for adventurous biologists,” Nat. Methods 12(1), 30–34 (2014). [CrossRef] [PubMed]
18. A. Kaufmann, M. Mickoleit, M. Weber, and J. Huisken, “Multilayer mounting enables long-term imaging of zebrafish development in a light sheet microscope,” Development 139(17), 3242–3247 (2012). [CrossRef] [PubMed]
19. T. Bruns, S. Schickinger, R. Wittig, and H. Schneckenburger, “Preparation strategy and illumination of three-dimensional cell cultures in light sheet-based fluorescence microscopy,” J. Biomed. Opt. 17(10), 1015181 (2012). [CrossRef] [PubMed]
20. J. Wu, J. Li, and R. K. Y. Chan, “A light sheet based high throughput 3D-imaging flow cytometer for phytoplankton analysis,” Opt. Express 21(12), 14474–14480 (2013). [CrossRef] [PubMed]
21. R. Regmi, K. Mohan, and P. P. Mondal, “High resolution light-sheet based high-throughput imaging cytometry system enables visualization of intra-cellular organelles,” AIP Adv. 4(9), 097125 (2014). [CrossRef]
22. P. Neuži, S. Giselbrecht, K. Länge, T. J. Huang, and A. Manz, “Revisiting lab-on-a-chip technology for drug discovery,” Nat. Rev. Drug Discov. 11(8), 620–632 (2012). [CrossRef] [PubMed]
23. R. Galland, G. Grenci, A. Aravind, V. Viasnoff, V. Studer, and J. B. Sibarita, “3D high- and super-resolution imaging using single-objective SPIM,” Nat. Methods 12(7), 641–644 (2015). [CrossRef] [PubMed]
24. A. Kumar, Y. Wu, R. Christensen, P. Chandris, W. Gandler, E. McCreedy, A. Bokinsky, D. A. Colón-Ramos, Z. Bao, M. McAuliffe, G. Rondeau, and H. Shroff, “Dual-view plane illumination microscopy for rapid and spatially isotropic imaging,” Nat. Protoc. 9(11), 2555–2573 (2014). [CrossRef] [PubMed]
25. O. E. Olarte, J. Andilla, D. Artigas, and P. Loza-Alvarez, “Decoupled illumination detection in light sheet microscopy for fast volumetric imaging,” Optica 2(8), 702–705 (2015). [CrossRef]
