A platform for high throughput two-photon-targeted in vivo cellular physiology
Lead Research Organisation:
Imperial College London
Department Name: Dept of Bioengineering
Abstract
The whole-cell patch clamp recording technique has had much impact in the Life Sciences by allowing the currents in ion channels to recorded: this has been crucial to advances in our understanding of cellular function. Use of the patch-clamp technique in vivo, however, has been difficult, because it essentially has to be carried out "blind", with the only feedback obtained by monitoring the impedance of the micropipette used to make the recording.
Recent developments in multiphoton microscopy now allow us to make targeted recordings from cells that have been fluorescently labelled. It has also recently been shown that in vivo patch-clamp recordings can be automated. In this project, we will combine these principles to develop a new approach for whole cell patch clamp electrophysiology. Our platform will allow genetically targeted classes of cells to be visually selected ("point and click") by a human operator, and then automatically patched by a group of robotic micromanipulators capable of obtaining recordings from up to six cells simultaneously. As well as allowing high throughput characterization of cells in vivo for basic scientific or drug discovery purposes, our platform will allow new scientific questions to be asked involving interactions between cells that have not hitherto been addressable. Finally, the precision afforded by the automated robotic control system will allow the patch-clamping of subcellular structures in vivo, which has not previously been systematically achievable. We will demonstrate the utility of our two-photon targeted robotic patch clamp platform by using it to target two particular classes of pyramidal cell in the mouse cerebral cortex, in order to ascertain their respective roles in processing sensory information.
Recent developments in multiphoton microscopy now allow us to make targeted recordings from cells that have been fluorescently labelled. It has also recently been shown that in vivo patch-clamp recordings can be automated. In this project, we will combine these principles to develop a new approach for whole cell patch clamp electrophysiology. Our platform will allow genetically targeted classes of cells to be visually selected ("point and click") by a human operator, and then automatically patched by a group of robotic micromanipulators capable of obtaining recordings from up to six cells simultaneously. As well as allowing high throughput characterization of cells in vivo for basic scientific or drug discovery purposes, our platform will allow new scientific questions to be asked involving interactions between cells that have not hitherto been addressable. Finally, the precision afforded by the automated robotic control system will allow the patch-clamping of subcellular structures in vivo, which has not previously been systematically achievable. We will demonstrate the utility of our two-photon targeted robotic patch clamp platform by using it to target two particular classes of pyramidal cell in the mouse cerebral cortex, in order to ascertain their respective roles in processing sensory information.
Technical Summary
Our objectives are
1. To develop a platform for making high yield robotically assisted whole cell patch clamp recordings from multiple genetically targeted and visually selected cells in vivo.
2. To demonstrate the utility of this technology by using it to resolve an unanswered question about mammalian cortical circuit function: what are the respective roles of slender and thick-tufted layer V pyramidal neurons in processing sensory information?
We will achieve our first objective by building a customised two-photon microscope, utilising ScanImage, an open source toolbox for microscope control. An additional module will be developed for ScanImage which allows point-and-click acquisition of cellular target locations, resulting in automated patch-clamp recording from the targeted cell. The robotic patch-clamp control algorithm will monitor impedance and an electropneumatic converter to control micropipette pressure. Pressure will be maintained at a high level until the pipette is in the target location. Multiple pipettes will be operated simultaneously.
Once the platform has been developed, it will be used to address the second objective, by making use of two available Cre mouse lines, Etv1 and Glt25d2, together with an AAV vector made in house, to obtain targeted patch-clamp recordings from layer 5 pyramidal cells of defined class, while somatosensory stimulation is performed.
1. To develop a platform for making high yield robotically assisted whole cell patch clamp recordings from multiple genetically targeted and visually selected cells in vivo.
2. To demonstrate the utility of this technology by using it to resolve an unanswered question about mammalian cortical circuit function: what are the respective roles of slender and thick-tufted layer V pyramidal neurons in processing sensory information?
We will achieve our first objective by building a customised two-photon microscope, utilising ScanImage, an open source toolbox for microscope control. An additional module will be developed for ScanImage which allows point-and-click acquisition of cellular target locations, resulting in automated patch-clamp recording from the targeted cell. The robotic patch-clamp control algorithm will monitor impedance and an electropneumatic converter to control micropipette pressure. Pressure will be maintained at a high level until the pipette is in the target location. Multiple pipettes will be operated simultaneously.
Once the platform has been developed, it will be used to address the second objective, by making use of two available Cre mouse lines, Etv1 and Glt25d2, together with an AAV vector made in house, to obtain targeted patch-clamp recordings from layer 5 pyramidal cells of defined class, while somatosensory stimulation is performed.
Planned Impact
The project will impact upon:
Life Scientists (see Academic Beneficiaries)
The UK Life Science Instrumentation Industry
This will be of immediate benefit to our industrial partner, Scientifica Ltd, a rapidly growing UK Life Sciences instrumentation company, who will be involved in commercializing developments from the project. However, we expect the benefits to extent to the wider UK Life Sciences industry, who will be galvanized by our developments.
The advances gained over competing Life Sciences instrumentation companies in other countries will improve the economic competitiveness of the UK
These benefits are likely to follow reasonably quickly after the end of the project (within 1-3 years), as the potential product takes advantage of a number of separate things we already know how to do, and could be brought to market quickly
Drug discovery companies
This includes both "Big Pharma" and the new breed of smaller, more dynamic drug discovery companies.
Products focused on drug discovery may require scaling up our system ("higher throughput"), and thus are likely to follow 3-5 years from the end of the project.
Some parts of the Charity Sector will in the longer run reap many benefits from the technology we propose, as many problems of interest to Charities such as the Alzheimer's Trust involve specific cell types that are not well physiologically characterized in vivo, and can be genetically targeted for basic physiology studies using our platform
In the longer run, Society, through economic benefits ensuing from the use of our technology to advance the understanding of basic processes underlying disease mechanisms, aging, and the discovery of new drugs.
Life Scientists (see Academic Beneficiaries)
The UK Life Science Instrumentation Industry
This will be of immediate benefit to our industrial partner, Scientifica Ltd, a rapidly growing UK Life Sciences instrumentation company, who will be involved in commercializing developments from the project. However, we expect the benefits to extent to the wider UK Life Sciences industry, who will be galvanized by our developments.
The advances gained over competing Life Sciences instrumentation companies in other countries will improve the economic competitiveness of the UK
These benefits are likely to follow reasonably quickly after the end of the project (within 1-3 years), as the potential product takes advantage of a number of separate things we already know how to do, and could be brought to market quickly
Drug discovery companies
This includes both "Big Pharma" and the new breed of smaller, more dynamic drug discovery companies.
Products focused on drug discovery may require scaling up our system ("higher throughput"), and thus are likely to follow 3-5 years from the end of the project.
Some parts of the Charity Sector will in the longer run reap many benefits from the technology we propose, as many problems of interest to Charities such as the Alzheimer's Trust involve specific cell types that are not well physiologically characterized in vivo, and can be genetically targeted for basic physiology studies using our platform
In the longer run, Society, through economic benefits ensuing from the use of our technology to advance the understanding of basic processes underlying disease mechanisms, aging, and the discovery of new drugs.
Publications

Annecchino LA
(2018)
Progress in automating patch clamp cellular physiology.
in Brain and neuroscience advances

Annecchino LA
(2017)
Robotic Automation of In Vivo Two-Photon Targeted Whole-Cell Patch-Clamp Electrophysiology.
in Neuron

Annecchino, LA
(2013)
Two-photon targeted robotic patch-clamp electrophysiological recording in vivo.
in Abstract Viewer/Itinerary Planner. Washington, DC: Society for Neuroscience. Online

Cheung K
(2015)
NeuroFlow: A General Purpose Spiking Neural Network Simulation Platform using Customizable Processors.
in Frontiers in neuroscience

Delgado Ruz I
(2014)
Localising and classifying neurons from high density MEA recordings.
in Journal of neuroscience methods

Jager P
(2016)
Tectal-derived interneurons contribute to phasic and tonic inhibition in the visual thalamus.
in Nature communications

Jarvis S
(2015)
Prospects for Optogenetic Augmentation of Brain Function.
in Frontiers in systems neuroscience

Jarvis S
(2018)
Neuronal gain modulability is determined by dendritic morphology: A computational optogenetic study.
in PLoS computational biology

Jarvis S
(2013)
Controlling the neuronal balancing act: optical coactivation of excitation and inhibition in neuronal subdomains
in BMC Neuroscience

Muzzu T
(2018)
Encoding of locomotion kinematics in the mouse cerebellum.
in PloS one
Description | We have developed a robotically automated system for performing electrophysiological recordings from cells in live tissue, targeted by fluorescent labels. Our platform is based on a multiphoton fluorescence microscope, augmented with custom developed electronics and software. A controller system controls internal pipette pressure, while monitoring impedance, as a micropipette is lowered into the brain towards a target cell. A computer vision system monitors the elastic deformation of the brain due to the pipette moving through brain tissue, and adjusts the pipette trajectory in order to reach the desired target. Upon reaching the target, the system automatically obtains a whole cell patch clamp recording with the cell. The system was validated by using it to perform recordings from genetically targeted cells in the mouse neocortex and cerebellum. |
Exploitation Route | Our platform is widely applicable to problems in basic neuroscience (understanding neural circuit properties) and to characterising the effects of pharmaceuticals on single neurons in the brain. We aim to disseminate both the hardware and software, through open source release, and/or commercialisation. The latter will depend on access to further funding for early stage commercialisation, for which we are now applying. We have obtained some funding from the Royal Society (however this does not include salary costs, so we are applying for additional schemes). |
Sectors | Pharmaceuticals and Medical Biotechnology |
Description | The development of the two-photon targeted robotic patch clamp made in this project has not been commercialised, due to the conflicting underlying patent landscape (ie. there are overlapping patents for an underpinning technology that currently prevent any operator from commercialising this space). For this reason, our industry partner opted not to license the technology. We would like to further disseminate the technology via open source means (including the hardware, via a third party that will manufacture kits), and are exploring this possibility (however even to do this would require further funding for salaries, which are not available on the Impact funding route). Nevertheless, the work has had impact upon industry, resulting in a number of invitations to speak about it to the pharma industry (invitations to speak in house at Eli Lilly and Janssen, and at the Neuroscience R&D pharma industry oriented conference. |
First Year Of Impact | 2019 |
Sector | Pharmaceuticals and Medical Biotechnology |
Description | The Robot Neuroscientist: Automating Two-Photon Targeted Electrophysiology |
Amount | £48,000 (GBP) |
Funding ID | TA\R1\170047 |
Organisation | The Royal Society |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 03/2018 |
End | 02/2019 |
Title | Platform for two photon targeted robotic patch clamp electrophysiology |
Description | As part of this project, we have developed a platform for two photon targeted robotic (automated) whole cell patch clamp electrophysiology. This comprises a two photon microscope coupled to a patch clamp amplifier, and custom-developed electronics for closed loop, analogue pneumatic control of internal pipette pressure. Custom software was developed to control the platform; this incorporates a computer vision system for guidance of pipette motion through the brain towards a target, compensating for deformation of the brain. A paper on this platform was published in the journal Neuron. Software and hardware details were disseminated upon publication of our paper. |
Type Of Material | Technology assay or reagent |
Year Produced | 2017 |
Provided To Others? | Yes |
Impact | We are developing further applications of this tool which will impact the pharmaceutical/ life sciences industry, as well as university research. |
URL | https://www.sciencedirect.com/science/article/pii/S089662731730733X |
Description | Data analysis tools for two photon calcium imaging |
Organisation | Eli Lilly & Company Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | The Schultz group, in collaboration with the Dragotti group (Dept of EEE) have developed a number of tools for online and offline analysis of two photon calcium imaging data. This stems in part from the BBSRC grant "A platform for high throughput two-photon targeted in vivo cellular physiology", in which we developed a new method for performing robotically automated in vivo patch clamp recordings, guided by a two-photon microscope. Many of the applications of this tool involve it being used in conjunction with calcium imaging. |
Collaborator Contribution | McGill University - the lab of Jesper Sjostrom provided "ground-truth" quadruple patching in vitro electrophysiology data, which enabled us to test many aspects of the algorithms we developed. As a result, the McGill team (who performed these difficult experiments on our request) and co-authors of two papers - one in eNeuro in 2017, and a second which is currently under review. Eli Lilly: supported a PhD studentship to take this work further, with application to studying mouse models of Alzheimer's Disease (the student, supported by the EPSRC CDT in Neurotechnology, and the government of the Philippines, began in Oct 2017). Eli Lilly are providing transgenic mice, and shipping costs, as required. |
Impact | Multi-Disciplinary: neuroscience, engineering (signal processing, data analysis, photonics). Outputs: (1) S Reynolds, T Abramsson, R Schuck, PJ Sjöström, SR Schultz and PL Dragotti (2017). ABLE: an Activity-Based Level Set Segmentation Algorithm for Two-Photon Calcium Imaging Data. eNeuro 4(5): ENEURO-0012. DOI: https://doi.org/10.1523/ENEURO.0012-17.2017. (2) R Schuck, MA Go, S Garasto, S Reynolds, PL Dragotti and SR Schultz (2017). Multiphoton minimal inertia scanning for fast acquisition of neural activity signals. Journal of Neural Engineering, 15(2), 025003. (3) S. Reynolds, T. Abramsson, P. J. Sjöström, S.R. Schultz and P.L. Dragotti (2018). CosMIC: a consistent metric for spike inference from calcium imaging. Submitted to Neural Computation. |
Start Year | 2016 |
Description | Data analysis tools for two photon calcium imaging |
Organisation | McGill University |
Country | Canada |
Sector | Academic/University |
PI Contribution | The Schultz group, in collaboration with the Dragotti group (Dept of EEE) have developed a number of tools for online and offline analysis of two photon calcium imaging data. This stems in part from the BBSRC grant "A platform for high throughput two-photon targeted in vivo cellular physiology", in which we developed a new method for performing robotically automated in vivo patch clamp recordings, guided by a two-photon microscope. Many of the applications of this tool involve it being used in conjunction with calcium imaging. |
Collaborator Contribution | McGill University - the lab of Jesper Sjostrom provided "ground-truth" quadruple patching in vitro electrophysiology data, which enabled us to test many aspects of the algorithms we developed. As a result, the McGill team (who performed these difficult experiments on our request) and co-authors of two papers - one in eNeuro in 2017, and a second which is currently under review. Eli Lilly: supported a PhD studentship to take this work further, with application to studying mouse models of Alzheimer's Disease (the student, supported by the EPSRC CDT in Neurotechnology, and the government of the Philippines, began in Oct 2017). Eli Lilly are providing transgenic mice, and shipping costs, as required. |
Impact | Multi-Disciplinary: neuroscience, engineering (signal processing, data analysis, photonics). Outputs: (1) S Reynolds, T Abramsson, R Schuck, PJ Sjöström, SR Schultz and PL Dragotti (2017). ABLE: an Activity-Based Level Set Segmentation Algorithm for Two-Photon Calcium Imaging Data. eNeuro 4(5): ENEURO-0012. DOI: https://doi.org/10.1523/ENEURO.0012-17.2017. (2) R Schuck, MA Go, S Garasto, S Reynolds, PL Dragotti and SR Schultz (2017). Multiphoton minimal inertia scanning for fast acquisition of neural activity signals. Journal of Neural Engineering, 15(2), 025003. (3) S. Reynolds, T. Abramsson, P. J. Sjöström, S.R. Schultz and P.L. Dragotti (2018). CosMIC: a consistent metric for spike inference from calcium imaging. Submitted to Neural Computation. |
Start Year | 2016 |
Title | Patch Clamp Technology (software) |
Description | A system for performing two-photon targeted robotic patch clamp electrophysiology was developed. This included LabView software for closed loop control of a robotic patch clamp manipulator, under the guidance of a two photon microscope, which was disclosed to Imperial Innovations in 2017. |
IP Reference | |
Protection | Copyrighted (e.g. software) |
Year Protection Granted | 2017 |
Licensed | No |
Impact | None as yet. |
Title | Platform for two-photon targeted robotic patch clamp electrophysiology |
Description | This platform, which includes both hardware and software developments, was disclosed to Imperial Innovations in 2016. They made the decision that a patent would not be sought. |
IP Reference | |
Protection | Protection not required |
Year Protection Granted | 2016 |
Licensed | No |
Impact | None as yet. |
Title | Software for two-photon targeted robotic patch clamp electrophysiology |
Description | We developed LabView software for controlling a two-photon targeted robotically automated patch clamp electrophysiology system. A version of this software, frozen as of the publication of the Annecchino et al (2017) paper in Neuron, was made available via GitHub.com |
Type Of Technology | Software |
Year Produced | 2017 |
Open Source License? | Yes |
Impact | None as yet. |
URL | https://github.com/schultzlab/rita |
Description | Interviews for news articles about our robotic patch clamp paper |
Form Of Engagement Activity | A press release, press conference or response to a media enquiry/interview |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Media (as a channel to the public) |
Results and Impact | Imperial College issued a press release on our paper "Robotic, two photon targeted in vivo whole cell electrophysiology" in Neuron. This led to interviews (eg with The Scientist, Technology Networks, and the Italian edition of Wired), and coverage in numerous media, including Wired (Italy), The Engineer, The Scientist, Medical News Today, Eurekalert.org and Technology Networks. On Twitter the post had 6450 impressions, and on Facebook 9,274, within two weeks. Technology Networks covered it on their "Neuronewsresearch" webcast which is followed by 1.2 million people. The news item was featured on the front page of the BBSRC website. |
Year(s) Of Engagement Activity | 2017 |
Description | Top Technical Advances of 2017 |
Form Of Engagement Activity | Engagement focused website, blog or social media channel |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Public/other audiences |
Results and Impact | The Scientist magazine selected our technology, as published in the Annecchino et al (2017) paper in Neuron, as one of the Top Technical Advances of 2017, alongside CRISPR-Cas developments, and DNA Origami. |
Year(s) Of Engagement Activity | 2017 |
URL | https://www.the-scientist.com/?articles.view/articleNo/51220/title/Top-Technical-Advances-in-2017/ |