Découverte de médicament

Drug discovery is an interdisciplinary field focused on the identification and pre-clinical testing of potential medications in vitro, ex vivo, and in vivo. This includes primary assays of the drug’s pharmacological properties, as well as secondary assays of absorption, distribution, metabolism, elimination, and toxicology (ADMET) and related safety factors that are evaluated in the discovery process. High throughput screening/imaging has occupied a central position in drug discovery testing over the past couple decades – providing automated and highly quantitative options for screening large numbers of candidate compounds for hits. Furthermore, imaging complements genomics, proteomics, and other -omics type methods by providing a spatial context. Combined with advanced techniques such as cell phenotype profiling using artificial intelligence, light microscope-based imaging/screening seems poised to become an even more important (and accessible) component of the drug discovery process.

Produits pour le Découverte de médicament

For high throughput and high content imaging applications in drug discovery, Nikon offers the BioPipeline LIVE and BioPipeline PLATE high content imaging systems. Both systems are based on our ECLIPSE Ti2-E motorized inverted microscope and feature a robotic arm for automatic exchange and imaging of up to 44 vessels, including well plates. The BioPipeline LIVE additionally includes full incubation coverage for long-term live cell imaging-based assays. Both systems are suited to the scale of high throughput screening in academic research labs.

With the use of larger model systems, such as 3D cell cultures and model organisms, an optical sectioning technique such as confocal microscopy may be necessary. The Nikon AX / AX R confocal microscopes are point-scanning instruments that support imaging over a large 25 mm field of view with up to 8192-pixel resolution. The AX R model additionally features a resonant scanner for high-speed imaging (30 frames per second at 2048 x 512-pixel resolution (2K), 1024 x 512-pixel resolution (1K)), supporting high throughput.

The Yokogawa CSU series spinning disk confocal systems can also be configured on BioPipeline instruments. Later stage in vivo and intravital imaging is supported by the AX R MP multiphoton microscope system, which allows for imaging up to 1.4 mm deep when combined with 1300 nm illumination.

An important component of a Nikon microscope system for drug discovery is our NIS-Elements software, an integrated solution for acquisition, analysis, and device control. The NIS-Elements HC version is specifically designed for high throughput and high content imaging applications. Heat maps, sample images, binary masks, assay results, and more are centralized, allowing for quick filtering and further analyses. Furthermore, our NIS.ai deep learning-based software modules can be integrated in imaging/analysis pipelines to bring the power of artificial intelligence (AI) to bear for various analysis tasks, such as image segmentation.

●: Inclus, ⚬: Optionnel

BioPipeline LIVE
High Content Imaging System
BioPipeline PLATE
High Content Imaging System
Maximum Sample Capacity 44 44
Automated Sample Exchange yes yes
Intended Vessel/Slide Type(s) 96-well plates
384-well plates
96-well plates
384-well plates
Z Focus Correction Perfect Focus System 4 (PFS4)
Auto-Focus
Perfect Focus System 4 (PFS4)
Auto-Focus
Suitable for Long-Term Live imaging yes no
Compatible Contrasting Techniques BioPipeline LIVE BioPipeline PLATE
Brightfield yes yes
Phase Contrast* yes yes
Nikon Advanced Modulation Contrast (NAMC)* yes yes
Volume Contrast yes yes
Widefield Fluorescence yes yes
Confocal yes yes

*These contrasting techniques may not be well-suited for all applications and magnification ranges due to the meniscus effect within individual wells.

Discussion sur le Découverte de médicament

Intestinal organoid stained for DNA (blue; DAPI), mucin (green; produced by goblet cells), and somatostatin (red; produced by enteroendocrine cells), imaged using point-scanning confocal microscopy.

Selecting a Microscope System for your Drug Discovery Model

Model systems for drug discovery run just about the entire range of possibility, from in vitro adherent cell cultures to whole model organisms and almost everything in between. Complex 3D cell culture models such as spheroids, organoids, and organs-on-chips can be composed of multiple cell types to better recapitulate various physiological characteristics lost in traditional adherent cell cultures, which are the traditional standard for high throughput screening. Additionally, organoids can even be cultured from autologous (patient-derived) cells for developing precision medicines.

Widefield fluorescence imaging is an appropriate choice for relatively flat samples, such as adherent cells cultured in multi-well plates, the traditional standard for high throughput screening. It is fast, sensitive, and cost effective, but does not provide intrinsic optical sectioning – the ability to effectively image a single plane in a thick 3D specimen.

Physically larger 3D models, such as organoids and tissue, may require an imaging technique with optical sectioning capabilities to capture the features of interest without being overly compromised by out-of-focus blur. Confocal microscopy is the standard for such applications, Nikon offers the AX / AX R confocal microscopes, point-scanning systems which can be configured on the BioPipeline LIVE and BioPipeline PLATE systems for imaging discrete sections up to several hundred micrometers deep in various samples.

While confocal imaging is a suitable choice for optical sectioning at depth, sometimes it is insufficient. In vivo imaging in the presence of thick and scattering tissues often necessitates the use of multiphoton imaging, such as the Nikon AX R MP multiphoton microscope system, which uses multiphoton excitation with near-infrared to infrared light to minimize out-of-focus excitation and scattering.

Volume Contrast
Phase Contrast

Quantitative Phase Imaging for Cell Analysis

Nikon’s ECLIPSE Ti2-E inverted microscope, which forms the base of the BioPipeline LIVE and BioPipeline PLATE systems, is capable of quantitative phase imaging via the Nikon Volume Contrast technique, which only requires a small Z stack of brightfield images (as few as three images per stack) to create the phase distribution image. The resulting phase distribution image is brightest where the optical path difference is greatest – in the middle of the cell when imaging a culture. This is beneficial for cell segmentation via thresholding and other methods.

Unlike other transmitted light (diascopic) imaging techniques, such as phase contrast, Volume Contrast is unaffected by the meniscus effect, which can be significant in well plates due to the small diameter of individual wells. Performance of phase contrast and Volume Contrast in the presence of a meniscus is compared in the image on the right.

For more information, see our recent application note detailing the use of Volume Contrast as part of a label-free cell proliferation assay, a commonly performed assay in drug discovery research.

Segment.ai
Original

Neurites in phase-contrast were not possible to define accurately by traditional thresholding. Segment.ai was trained on hand-traced neurites (human recognized) and learned how to trace neurites in subsequent images.

Enhancing Drug Discovery Research using Artificial Intelligence

One of the great advantages of image-based assays is their rich information content. Until recently, however, relatively little of this information could be leveraged in a practical manner. However, new approaches based on artificial intelligence (AI), and particularly deep learning (DL) methods using artificial neural networks (ANNs), allow for deeper correlations between image features to be inferred and applied towards characterizing morphology and phenotype. Such analytical approaches have been referred to as “cell profiling” or “image-based profiling,” and represent a very active area of development.

In addition to profiling, DL can be used to help accelerate and strengthen image analyses in other ways. Nikon is committed to the design of reliable DL-based image analysis tools under the umbrella of NIS.ai – a series of software modules available for the NIS-Elements software. For example, the Segment.ai module can be trained to automatically segment difficult image features that are difficult to isolate using classical approaches.

Other NIS.ai modules of potential application for drug discovery work includes Convert.ai, which can be trained to predict image features from a fluorescence channel using only a brightfield or another transmitted light channel as a reference, which can help cut down on both cytotoxicity (from the fluorescent label) and phototoxicity (from the high intensity light used for fluorescence imaging). Relatedly, the Enhance.ai module can be trained to predict a higher signal-to-noise version of noisy image data. This allows for a reduction in illumination intensity, and thus phototoxicity.

High throughput widefield fluorescence imaging of larger model systems, such as organoids, can benefit from our Clarify.ai module, which is pre-trained to provide automatic blur removal from widefield fluorescence microscopy images. This allows for users to benefit from the speed of widefield fluorescence imaging, but with enhanced optical sectioning and without the need for a confocal system.

Nikon contract research services at a glance.

Nikon Contract Imaging Services for Pre-Clinical Drug Discovery Research

The Nikon BioImaging Labs provide contract research services to their local biotech and research communities, as well as remote services* for clients located elsewhere. With locations in Cambridge, MA, USA, Leiden, The Netherlands, and Shonan, Japan, the Nikon BioImaging Labs have significant experience working with customers in the biopharma research space and performing confocal microscopy of organoids and other 3D cell culture systems now used for drug discovery studies, including various commercially produced organ-on-chip models.

The Nikon BioImaging Labs are not only capable of data acquisition, but their full-service capabilities also extend to assay design, assay development, assay validation, cell culture, sample preparation, and data analysis. If you would like to learn more and see if services at one of the Nikon BioImaging Labs is right for you, please don’t hesitate to contact us to set up a free consultation.

* The service may vary depending on the facility. Please contact the Nikon BioImaging Lab near your location for details.

Glossaire

Automated Sample Exchange
This refers to automated exchange of sample vessels or microscope slides between the microscope stage and incubator or loader/hotel.
Compatible Contrasting Techniques
Typically, high throughput imaging microscopy systems are capable of a transmitted light technique, such as phase contrast, as well as a fluorescence imaging modality, such as widefield or confocal.
Intended Vessel/Slide Type(s)
The BioPipeline LIVE and PLATE systems are based on inverted microscopes, making them better suited to imaging chambered culture vessels, such as multi-well plates, than upright microscopes. In this case, the bottom of the vessel should be made of an optical quality material. The BioPipeline SLIDE is based on an upright microscope – the usual choice for imaging traditional glass microscope slides.
Maximum Sample Capacity
This is the maximum number of vessels, well plates, or microscope slides (as appropriate) that can be accommodated in the loader/hotel for use in automated experiments. See “Intended Vessel/Slide Type(s)” for reference.
Suitable for Long-Term Live imaging
Suitable systems integrate environmental control to the imaging and vessel storage areas. Standard incubation parameters include temperature, humidity, and percent CO2.
Z Focus Correction
Being able to rely upon having the correct Z plane in focus is especially important for high throughput, as mistakes may go unnoticed during automated acquisitions. Nikon’s Perfect Focus System is a leading hardware solution for maintaining a user or software-selectable plane in focus. Auto-focus is a software-based routine that identifies the correct focal plane by scanning in Z to find the plane with the highest contrast/sharpness.