Introduction to Neuroimaging 06-10-2023 Animal Research and Research Ethics PDF

Introduction to Neuroimaging Animal research 06.10.2023a Ruth M Krebs, Dept. of Experimental Psychology, Ghent University 1 Overview 1. Why discuss animal research? 2. Selected techniques 2.1 Recording and inducing neuronal activity 2.2 Pharmacological manipulations and lesions 2.3 Local field p...

Introduction to Neuroimaging Animal research 06.10.2023a Ruth M Krebs, Dept. of Experimental Psychology, Ghent University 1 Overview 1. Why discuss animal research? 2. Selected techniques 2.1 Recording and inducing neuronal activity 2.2 Pharmacological manipulations and lesions 2.3 Local field potentials, EEG, and fMRI 3. Notes on comparability 2 1. Why discuss animal research? This course is on human neuroimaging, but we need the animal-model perspective  Human cognitive neuroscience is heavily influenced by animal research  Human neuroimaging methods are less precise (lower spatio-temporal resolution)  Hence we need to relate our results to animal models for a comprehensive view  As experimental psychologists you might want to do animal research yourself 3 2. The basis Neuronal activity - Sodium-potassium pump in the membrane pumps sodium (Na+) out and potassium (K+) in. Negative resting potential (more Na+ exported) - Neurotransmitters (released from pre-synaptic cell) open NA+ or K+ channels: excitatory or inhibitory stimulation - Based on the summed inputs, the electrical membrane of the neuron rapidly rises and falls: Action potential - The Action potential travels along the axon and contributes to activating the next neuron… 4 2. Selected techniques Measuring and manipulating neuronal activity at different levels small to large scale  Recording and inducing neuronal activity: recording and inducing electrophysiological activity just outside the cell(s)  related to action potentials  Optogenetic imaging: Manipulating cell function with light via genetically modified neurons  Pharmacological manipulations: inducing receptor agonists and antagonists; changing re-uptake, synthesis, break-down  Local field potentials (LFPs): measuring the summed dendritic synaptic current in the tissue (not action potentialsas such)  (intracranial) EEG and fMRI (similar to human neuroimaging) 5 2. Selected techniques 2.1 Recording and inducing neuronal activity (incl. optogenetic imaging)  activity patterns of neurons and neuron clusters provide insights into a region’s function 2.2 Pharmacological manipulations (and lesions)  allow causal conclusions regarding the function of a region or neurotransmitter system 2.3 Local Field Potentials (LFPs), (intracaranial) EEG, and fMRI  more indirect than single-unit recordings, but comparable with human data This is not a comprehensive overview of animal research methods in neuroscience (restricted to in-vivo, not including histological post-mortem analyses or patch-clamp etc.) Some of the following studies have inspired important research lines in human neuroscience 6 2.1 Recording and inducing neuronal activity RECORDING neuronal activity via electrodes - Microelectrodes (tip 1-10 µm) can isolate activity of a single neuron: voltages generated in the extracellular matrix when an action potential is generated in the cell (“spikes”) - Carrier device mounted on skull; electrodes are moved down through the target area until well-isolated neuronal activity is observed (supported by stereotactic reference frame) 7 2.1 Recording and inducing neuronal activity RECORDING neuronal activity via electrodes Example: Neurons coding for specific head direction in the hippocampus (Preston-Ferrer et al. 2016) (https://elifesciences.org/articles/14592#content) 8 2.1 Recording and inducing neuronal activity RECORDING neuronal activity via electrodes Example: Single-cell recordings from dopaminergic neurons in the substantia nigra during reinforcement learning (Schultz et al. 1997, 2001) - Increase in firing rate when reward (R) is delivered - Increase in firing rate when a conditioned stimulus (CS) reliably predicts this reward (replaces the response to R itself) - Decrease in firing rate when R is predicted but does not occur  These changes in neuronal activity lead to variying dopamine release in the target regions (e.g., nucleus accumbens)  This model of complex neuronal coding of reward information became the basis of human motivation studies using fMRI and PET 9 2.1 Recording and inducing neuronal activity INDUCING neuronal activity via electrodes  The same set-up can be used to stimulate the neurons in the vicinity of the electrode which can provide insights into the function of an area and its projection regions Example: Eletrical stimulation of the septal area (Olds & Milder 1954) - Electrodes were placed in the septal area of the rat forebrain, part of the dopaminergic system -The electrodes induced electrical pulses every time the rats pushed a certain lever in their box - The rats “enjoyed” the stimulation instead of avoiding it; some rats even neglected eating and drinking in favour of the stimulation  Demonstration that electrical stimulation of neurons can be used as an operant reinforcer (and evidence for the existence of some kind of “pleasure center”)  Functionally, the stimulation of the septal area leads to dopamine release in the nucleus accumbens, similar to the effects of primary rewards (previous slide) 10 2.1 Recording and inducing neuronal activity INDUCING neuronal activity via optogenetic imaging  Method to control and monitor the activities of individual neurons in living tissue in freely-moving animals  Genes for light-activated ion channels (opsins) are introduced to a population of cells by an engineered virus Blue light activates ON opsin (channelrhodopsine) Yellow light activates OFF opsin (halorhodopsin) Source: http://optogenetics.weebly.com/why--how.html https://www.youtube.com/watch?v=I64X7vHSHOE 11 2. Selected techniques 2.1 Recording and inducing neuronal activity (incl. optogenetic imaging)  activity patterns of neurons and neuron clusters provide insights into a region’s function 2.2 Pharmacological manipulations (and lesions)  allow causal conclusions regarding the function of a region or neurotransmitter system 2.3 Local Field Potentials (LFPs), (intracaranial) EEG, and fMRI  more indirect than single-unit recordings, but comparable with human data 12 2.2 Pharmacological manipulations and lesions Pharmacological manipulations Example: Manipulating dopamine transmission during effort-based choice (Salamone et al. 1994) - T-maze with a high (HD) and low (LD) reward density arm; one group had to cross a barrier to reach the food in the HD arm (effortful choice) - The rats mostly chose the HD arm, even those in the barrier group - Dopamine depletion (6-hydroxydopamine) and dopamine receptor blocking (haloperidol) in the nucleus accumbens abolished these high effort choices  Disrupting dopaminergic signalling did not alter reward evaluation, but biased the decision process towards investing less effort 13 2.2 Pharmacological manipulations and lesions Lesions Example: The role of the anterior cingulate during effort-based choice (Walton et al. 2003) - In this study the same T-maze task was used as on the previous slide Anatomical lesions via toxine injection (anterior cinglulate: ACC, limbic cortex: PL-IL, sham lesion: surgery without lesion) Only the ACC lesion rats stopped climbing the barrier to get to the high reward. Same result as dopamine depletion 14 2. Selected techniques 2.1 Recording and inducing neuronal activity (incl. optogenetic imaging)  activity patterns of neurons and neuron clusters provide insights into a region’s function 2.2 Pharmacological manipulations (and lesions)  allow causal conclusions regarding the function of a region or neurotransmitter system 2.3 Local Field Potentials (LFPs), (intracaranial) EEG, and fMRI  more indirect than single-unit recordings, but comparable with human data 15 2.3 Local Field Potentials (LFPs), EEG, and fMRI Local Field Potential (LFP’s): These are second-level signals which represent the local activity of a few thousand neurons together (after low-pass filtering), rather than spikes reflecting action potentials directly. EEG and fMRI: (EEG often intracranial = under the skull) EEG (electroencephalography) fMRI (funtional magnetic resonance tomography) Combining LFP recordings with EEG and fMRI in animals reveals what these more indirect measures reflect in the human brain! 16 2.3 Local Field Potentials (LFPs), EEG, and fMRI LFPs and EEG Relate EEG to neuronal activity in the same monkey Whittingstall et al. 2009 Compare EEG data of monkeys and humans Reinhart et al. 2009 17 2.3 Local Field Potentials (LFPs), EEG, and fMRI LFPs and fMRI Relate fMRI data to neuronal activity in the same monkey Compare fMRI data of monkeys and humans  Presenting visual stimuli while recording fMRI, multiple-unit activity, and local field potentials  fMRI signal (red) corresponds most closely to local field potentials (blue) (summed dendritic activity) Logothetis &Wandell 2004 Tootell et al. 2003 18 3. Notes on comparability Comparibility in terms of function  different research questions require different animal models Rodents in the lab (mostly mice and rats)  Rodent models are highly valuable for cognitive neuroscience especially for processes related to “old” brain structures (brainstem, basal ganglia, hippocampus)  Not comparable to humans on the neocortical level and more limited regarding complex cognitive tasks  Cheap, easy in terms of breeding, handling, training Primates in the lab (mostly macaques and rhesus)  Primate models are more comparable on the neocortical level and thus more valuable for investigating higher cognitive functions (e.g., monkeys can perform comparable computer tasks)  Still, there are neuroanatomical differences on the cortical level, which is why comparative studies often use the expression: “monkey homologue” of a region  Much more time consuming and challenging to breed and train primates 19 3. Notes on comparability Comparability in terms of methods Animal research procedures (mostly invasive): - closer to the actual neuronal substrate (firing rate, LFPs, receptor binding, etc.) - provide insights into causal relationships (pharmacological intervention, lesions, post-mortem histology, ...) Human research procedures (mostly non-invasive): - Recording / inducing neuronal activity: only as part of therapeutic approach (e.g., parkinson’s patients) - Pharmacological manipulations: mild pharmacological manipulations (healthy) and treatment (patients) - Lesions: “virtual lesions” TMS (healthy); real lesions (e.g., stroke); therapeutic lesion (e.g., epilepsy) - EEG and fMRI: in healthy population and patients [see lecture on Clinical groups] In human neuroscience, we mostly work with ‘indirect’ measures of neuronal activity and need to integrate results from animal research 20 21 Introduction to Neuroimaging Research ethics 06.10.2023b Ruth M Krebs, Dept. of Experimental Psychology, Ghent University 22 Research ethics 23 1. Some history on human research ethics Landmarks for today’s human research ethics:  1947 Nuremberg Code: subject consent and the right to quit, no unnecessary harm, scientific foundation, performed by experts. This was in response to “medical” experiments by the Nazi regime without any ethical standards.  1950-1970 Clinical and behavioral studies without consent, on sick or dependent participants, and/or clearly deceptive: Willowbrook hepatitis study; Ohio prison/Brooklyn Hospital cancer studies; Obedience study (Milgram); Tearoom trade study (Humphreys)  1964 Declaration of Helsinki: ethical principles for human experimentation (World Medical Association)  1966 Protection of Human Subjects policy: independent review boards (US National Institute of Health)  1930-1970 Tuskegee Syphillis study (revealed in 1972): study on poor african-americans, painful procedures, omission of treatment, unethical financial benefits  1974 National Research Act: formation of the US National Commission “Protection of Human Subjects in Biomedical and Behavioral Research”  1975-2000 revisions to the Declaration of Helsinki: still the basis for today’s ethical procedures  1979 Belmont Report: basic principles for human subject research are Respect, Beneficence, Justice  1991 Common Rule: establishing the Institutional Review Boards to ensure ethical procedures 24 2. Guidelines human research ethics Basic rules for human (neuroscience) research today:  Ethical Board approval: Each study has to be approved by the local ethical review board, which checks for accordance with overarching guidelines  Informed consent: Human subjects have to give their voluntary consent after being informed properly (procedure, risks, benefits), and they have to be able to give consent  Unbiased selection of subjects: Same procedure, benefits, rights (no “minority” studies)  Special protection for special groups: Children, patients, prisoners, pregnant women  Subjects have the right to quit: At any time without negative consequences  Minimal risk: No health risks for the subjects, both physically and mentally. This is less relevant for experimental and clinical psychology, but there are cases were it is (e.g., physical pain manipulations; fear conditioning; food/water deprivation; addiction; depression; obedience, etc.) These basic rules apply to all sudies conducted in our faculty! 25 3. Guidelines animal research ethics Guidelines for the care and use of mammels in neuroscience research: (Committee on Guidelines for the Use of Animals in Neuroscience and Behavioral Research 2003)  Use the minimum number of animals required to obtain valid results.  Conduct limited pilot studies if little is known about the involved processes to assess the effects of the procedure on the animals.  Manage animal pain/distress. If pain processing is not part of the research question, pain has to be minimized (anesthesia, analgesia, pharmacological treatment, handling)  Define humane endpoints a-priori (‘when to stop’). e.g., therapeutic failure (tumor growth), disease exacerbation (increased seizure frequency), or general clinical deterioration (decreased weight, alertness, respiration, increased temperature).  Define method of euthanasia a-priori and train staff to perform and confirm death. Common methods include decapitation, cervical dislocation, carbon dioxide inhalation, barbiturate overdose. Animals cannot give consent or file complaints, so they rely on the researchers’ integrity. Human research often relies on animal studies, so we cannot ignore or condemn these altogether. 26 27