L03 Research Methods

To link neurochemicals and behavior, we will need to…

• observe neurotransmission

  • e.g. Measure DA levels, DA neurons and DA receptors

• manipulate neurotransmission

  • e.g. Activate/inhibit DA projections and receptors

• measure behavior

  • e.g. Animal behavioral tests and human behavioral tests

---

• we examine brain-behaviour correlations

Observational vs. Experimental Studies

• Observed alterations in neurotransmitter system

  • Observational studies

  • Correlational, not causal association

  • More common in humans

• Manipulation of neurotransmitter system

  • Experimental studies

  • Allow for causal inferences

  • More common in animals

• Both correlate to Behavior or Behavioral Disorder

---

• observational studies - exploring associations - observe NT + behaviour/disorder

Observing NT systems

1. Localization and quantification of system elements (e.g. receptors, enzymes, etcetera) - see where pieces are + how much (what receptor on what cell)

2. Measuring transmitter levels (via metabolites)

3. Measuring gene-trait associations

4. Cell activity

Quantification + Localization

• various strategies for quant + localiz.

• can do for enzyme, transporter, etc.

• high mRNA levels correlated w protein → quantification

add more notes from TB*

Quantification (How much)

• Baseline gene + protein expression

• Expression can change

  • e.g. social isolation increases mRNA and protein levels (α4/δ GABAA subunits)

• Implications of these changes can be explored

  • Do GABAA subunits contribute to the psychological effects of isolation?

---

• % change in receptor sub units on y-axis - expression changes depending on env.

• expression only changes in area where already expressed?

Localization (Where)

• take AB for receptor, it attaches to AB, if you produce signal from AB —> you have dark signal that reflects the receptor

• binding reflects presence of receptor

• treatments do not change localization - they change quantitation (done in post-mortem studies due to invasiveness)

Positron Emission Tomography

• We can measure receptor occupancy/availability in the live human brain under different conditions

• Reduced GABA_A receptor occupancy in Fragile X Syndrome - could impact GABA_A receptor sensitivity, or GABA levels, etc.

• Reduced DA2 receptor availability in addiction

---

• imaging of NT systems in live organisms

• DA theory of SUD - less DA2 receptors for binding, less DA levels overall?

PET for exploring drug mechanisms

• PET studies can show a relationship between drugs, receptor binding and psychological function

• Subjective intensity of psilocybin experience correlates with 5-HT2A receptor occupancy (see L06)

---

• looking at amount of receptor binding via PET for serotonin receptor w psychedelics (which are similar in structure to serotonin) - occupancy of receptor by drug is related to intensity of experience

Measuring neurotransmitter levels

• microdialysis

• fiber photometry

• magnetic resonance spectroscopy

Microdialysis in animals

Invasive + labor-intensive, rarely done even in animals

---

• sample CSF fluid to determine amount of NTs

Fiber photometry in animals

• genetically engineered system where fluorescing compounds are used with fibers - can switch system on or off + measure signal

• DA signal? look at DA fibers, and create neurons that fluoresce when high levels of dopaminergic activity occurs

add info from TB

Alternative options in humans

• Need for less invasive techniques

• An example is magnetic resonance spectroscopy - works for specific NTs as shown in figure - good for in vivo (don’t need to know NAA)

• Useful for in vivo imaging, but limited spatial/temporal resolution and only works for certain neurotransmitters

• Metabolites can be used as a surrogate measure

• Metabolites are easier to measure (e.g. blood, urine)

  • 5-Hydroxy indoleacetic acid (5-HIAA) = metabolite for serotonin

  • Homovanillic acid (HVA) = metabolite for dopamine —> high levels of HVA may be present for SZ

  • 3-Methoxy-4-hydroxyphenylglycol (MHPG) = metabolite for noradrenaline (gambling, bipolar..?)

• Metabolites also measured in cerebrospinal fluid (but this is invasive)

---

• NTs break down + make metabolites

• measure metabolites to infer NT levels - good in theory, but not in reality

• metabolites collected in compartments + fluids

Gene variant-trait correlations

• measuring these associations.. we can..

• Assess links between gene variants and traits in humans (see PSY390)

• Early research showed links between DA system genes and substance use

  • e.g. if you have a certain form of gene, higher risk

• However, most links are weak + nonreproducible, but this doesn’t mean studying genes is useless

---

• most gene variations very weakly linked to traits

• finding gene for extroversion very unlikely

• finding gene for depression also unlikely, but less so is the idea

Different approaches to genes…

Different approaches that genes (i.e. transgenic knockouts) are in fact a common and useful approach

• missing a KO receptor or enzyme..

Where do we record cell acitivity?

• It depends – if we wanted to understand the DA system (L04), we could record from cells in DA pathways

• Aside: these same areas are good targets for manipulations (see next section)

---

• add more notes

“Micro level”: Electrophysiology

• Study of electrical + physiological properties of neurons

• We can study individual cells or large groups of cells

• In vivo (live brain) or in vitro brain slices

---

• lots of insight on small group of cells

Example: DA neuron recording

DA neuron signaling changes w/most rewards (e.g. food, social interaction), not just drugs of abuse

---

• dopaminergic firing changes during feeding

Related: Calcium Imaging

• create calcium imaging + generate fluorescing calcium-sensitive ..

• voltage-gated calcium channels - Ca rushes out

add more notes

“Macro” level

• As we are often focused on specific cells, we prefer “micro techniques” for measuring activity

• “Macro techniques” that observe many cells at once are also available

  • fMRI, PET, EEG and more

• Though less common in our area, these techniques are very useful in certain cases

  • For example, characterizing brain activation with psychedelics via fMRI (as in the work of Carhart-Harris and colleagues)

---

• looking at whole brain regions - but not usually done in this field

• fMRI not very good to observe NT systemsy

Manipulating the NT systems

1. Diet

2. Drug treatment (e.g. w/cannulation)

3. Electrode targeting

4. Lesion (many methods)

5. Cell targeting w/optogenetics and chemogenetics

6. Transgenic manipulation

---

• 
  

1) Diet

• Some neurotransmitters are made with nutrients that from the diet (e.g. tryptophan > serotonin)

• If we change the diet, we can change tryptophan availability and perhaps serotonin levels

  • Tryptophan loading and tryptophan depletion (L06)

---

• if NT production ends on precursor → levels of diet corresponds to NT?

• add notes

2) Drug treatment

• Commonly: give varying doses of a drug (agonist or antagonist; X axis) and observe behavior (Y axis)

• For example – if you are studying receptor Q, you’d give varying doses of an agonist for Q and then measure behavior

---

• influence receptor activity with drug (turn on/off)

• effects of removing = antagonist

• effects of adding = like agonist

“Selective” drug delivery

• Many ways to deliver a drug.. but Most methods are not specific (e.g. affect the whole body), which is NOT ideal when we’re trying to understand a restricted transmitter system.

• If we wanted to understand the DA system (L04), we could look at the effects of DA drugs in certain areas (e.g. where there are DA receptors)!

“Selective” method: Via Cannula

• Cannula implanted into amygdala

• GABA agonist (muscimol) prior to learning inhibited the amygdala and impaired fear conditioning

---

• surgical manipulations

• coordinate system to rodent brain to locate specific area

• implant in brain area of interest

• can reversibly inactivate - implant canula into amygdala to deliver drugs there

  • what happens when you turn on GABA receptor = inhibitory

  • neurons would turn off below - briefly neurons are off - and can examine effects

  • learning less effective when muscimol used to turn off..

Selective effects on DA projections

• targeted drug release not the only way to affect an area

• If we wanted to understand the DA system (L04), we could manipulate the activity of key DA projections

3) Targeted electrode

• Electrodes can be implanted into the rodent brain at specific sites in neurotransmitter systems

• Includes the intracranial self-stimulation paradigm

• If the electrode is implanted in the ventral tegmental area* (with DA neurons), stimulation is reinforcing

• Lesion of DA projections prevents this behavior

---

• gives animal power to turn on - will always keep it on

• why do this? give insight

• activate once makes you very likely to activate it again - dopamine hypothesis for SUD

• can stimulate brain but no DA fibers, no reinforcement..?

• add notes

4. Lesion

• Damage a brain area critical for neurotransmission (w/heat, chemicals or other methods)

• The drug 6-hydroxydopamine (OHDA), for example, destroys NA + DA fibers while also impairing memory

• Sometimes used to model Parkinson’s Disease

---

• artificially destroy fibers - to stimulate the loss of DA occurring in PD

5. Chemogenetics vs. optogenetics

• Selectively target a specific cell in a specific area (e.g. 5-HT cells in DRN, NA cells in LC)

• Reversibly activate or inactivate that cell type

• Optogenetics more popular; much better temporal resolution and fewer confounds

---

• optogenetics (way more common than chemogenetics) - activation of neurons w light that have been genetically sensitized to light - almost every study is like that (can turn neuron off / on)

• plant fiber to deliver light - observe (behavioural?) consequences when light delivered - extremely selective + fast

• great spatial + temporal resolution - due to reliance on light

Activating LC neurons (Opto)

• Stimulating the LC (which are NA cells) affects arousal and movement

---

• can also design optogenetic system that switches things off

Activating DRN neurons (Chemo)

Chronic but not acute activation of dorsal raphe neurons (which are 5-HT cells) is antidepressant

---

• artificial system generating cells specific to a drug

• drug-sensitive switch in certain cells

• what if receptor is active under other circumstances?

• acute activation associated with higher levels of anxiety

• acute activation associated w lower levels of depression ?

6) Transgenic modification

Variations in genes can lead to variations in behavior.

If we want to know how a transmitter affects behavior…

…we can see what happens when we manipulate genes in that transmitter system.

Transgenic mice

• Remove a gene (knock-out or KO mouse)

  • Most common, a lot of the data we discuss is from KO mice

• Add a gene (knock-in mouse)

  • Informative but far less common (not physiologically-relevant, doesn’t fit the reductionist approach)

• Many ways to modify genes

  • Global: Gene modification in all cells

  • Selective: Gene modification in certain cells (e.g. only 5-HT cells)

  • Inducible: Gene modification can be turned on at a certain time (e.g. 6 months)

---

• knock-out 5HT1B → aggressiveness - infer that it is involved

• knock-in more complex, so most of the time KO

• selective modification best

• inducible activation in adulthood for instance, to give narrow understanding of what it does in adulthood

δ subunit KO mice

Deletion of the δGABAA receptor subunit is associated with depression-like behavior and abnormal maternal behavior post-partum (L08)

---

• missing gene for subunit

• can’t make recptors for it

• behavioural differences occur

• receptors known to be hormone-sensitive

Our test subjects

• Humans are not frequently used in early research

  • Cannot always use the experimental method (invasive, unethical and sometimes impossible)

  • Potential for causal inferences is limited

• Animal models (in particular mouse models) are often used instead

• Mice are preferred for many reasons

  • Genome well-understood; similar structure to humans

  • Most human genes have mouse counterparts

  • Mouse populations are easy to produce and maintain

---

• mice - genetic similarity to humans - hence transgenics useful

Limitations of animal models

• Cost is high

  • Human research can be cheaper in some cases

• Modeling human behaviors is challenging

  • Even good models (e.g. Depression) are imperfect

  • Many sophisticated behaviors (e.g. language) cannot be investigated at all

• Not all findings are generalizable

  • Many drugs that work in animals do not work in humans (most fail, ~80%)

---

• 
  

Concerns

• 
  

Properties of “good” tests

• Reliability

  • Similar results are obtained every time you repeat the test

• Predictive validity (i.e. construct validity)

  • Animal test predicts the appropriate human behavior

  • Example - anxiety tests in rats identify anxiolytics in humans

• Selectivity

  • Test performance mostly reflects the appropriate behavior

  • Example – a test for pain is not affected by cognitive ability

• (In drug tests) Behavior affected by drugs in a dose-dependent manner.

---

• reliability - same results under same conditions

• not enough for test to be repeatable - must help us reliably predict

• above two features most important

Types of Behavioral Testing

• Anxiety-like Behavior

• Depression-like Behavior

• Learning, Memory and Cognition

• Social Behavior

Anxiety

• ‘Feeling of worry, nervousness or unease’ about something

• A trait we all have, to a degree - that’s useful (absence of anxiety would backfire)

• Adaptive to an extent

  • Avoid danger, survive to reproduce

• Though anxiety is viewed as useful, problems arise when it is not properly regulated

Elevated Plus Maze (EPM)

• Plus-shaped maze with four arms

  • Closed arms: unexposed (safe space)

  • Open arms: exposed (to light/height)

• Time spent in each arm measured

• Animals strongly prefer closed arms (by ~x4-10 times)

• Avoidance of open arms (low open arm time) is therefore an indicator of anxiety

• EPM- sample results

  • Closed arm time is high in both groups

  • Closed arm time is significantly higher in Transgenics

---

• look at approach + avoidance

• risk-taking behaviour still important (e.g., for finding food) so time spent in open arms is low but not 0

Other tests

• Open field test

• Light-dark box

• Successive alleys test

• Zero maze (modified version of EPM)

EPM - In summary…

• EPM regarded as industry standard for anxiety

• No pure tests of anxiety; other traits can interact

  • Locomotor activity + pain are factors in all tests

  • Novelty interest is a factor in the open field

• To interpret any one test (e.g. EPM), you should do several other tests • Notably, anxiety is a factor in other behavioral tests

Depression

• Loss of interest or pleasure even when performing activities normally enjoyed (e.g. Anhedonia)

• Decreased energy (e.g. Fatigue)

• Feelings of guilt or low self-worth

• Disturbed sleep, appetite and activity

• Inability to concentrate

• Thoughts of suicide

---

• when modelling, we try to look for these behaviours

Depression in Animal Models

• Behavioral despair or hopelessness

  • Modeled with forced swim test + tail suspension test

• Anhedonia (more commonly modeled/..?)

  • Modeled with sucrose preference test (decline of sweets consumed by mice is indicator of anhedonia)

Forced Swim Test (FST)

• When placed in water, animals demonstrate an escape response – swimming + climbing

• Less swimming/climbing, early cessation of swimming/climbing was said to be “depression-like”

  • More recent views is that it is “stress coping” responses

• Sample data:

  • Deletion of the delta GABAA receptor subunit is associated with depression-like behavior post-partum (GABA receptor KO mouse - they’re immobile faster)

---

• animals don’t like swimming - when do they stop so how much swimming/climbing?

Memory

• Process whereby information is stored, consolidated + retrieved

• Several types of memory proposed (sensory, working/short-term + long-term)

• These stores work over different time periods

Animal models of memory

• Fear conditioning test (LTM)

• Novel object recognition (LTM + STM)

• T-maze (STM + WM; not covered)

• Morris water maze (LTM; not covered

Why use fear to test memory?

• Fear is a feeling, induced by perceived danger, that is associated with physiological + behavioral changes (e.g. fleeing, hiding or freezing)

• Some fears are innate (like shadow over mouse, it thinks it’s a predator) but most are learned

• If exposed to an aversive stimulus, animals will remember that experience (memory is key)

• If re-exposed to an aversive stimulus that they remember, animals will show fear (expression of learned fear indicates memory)

---

• fear (1) easily (and reliably) induced + (2) easily observed

• expose to fear-induced aversive event

• expression of fear to aversive event an indicator of memory (shock + sound)

• more response = stronge rlearning

Fear Conditioning

• Shock paired w/an environment (blue box) + tone (bell)

• Box + tone become signals for the shock

• When re-exposed to box or tone days later, the animal ‘expects’ another shock and displays freezing

  • As test is conducted days later, assesses LTM, Freezing ~ memory

• sample data - Long-lasting memory resistant to decay (expressed for months later)*

---

• measure purple box but sound to infer learning

• learning + memory indicate conditioning - what else? maybe not better learners? maybe more pain sensitive, less movement (lazy) - must look into these and exclude

• more freezing doesn’t guarantee learning, just suggests it - must do more tests to strengthen correlation

Object recognition (OR) test

• Non-aversive memory test in 3 stages:

• Training: Animal exposed to a set of identical objects

• Break where animal is removed from chamber (variable time)

• Testing: Animal returned to the chamber; one object replaced with a novel object

• Animals recalling the familiar objects will show increased exploration of a novel object

• Sample data: Deletion of the δGABAA receptor subunit is associated with poor performance in this task

---

• level of interaction indicative of memory when exposed to increased exploration of new object

• 
  

Other tests of cognition

• 5 choice serial response time task - Attention and impulse control (could be used to study ADHD, ? transmission etc.)

• Puzzle box - Cognitive flexibility (switching b/w strategies)

Social behavior

• Many disorders are characterized by abnormal social behavior (e.g. Schizophrenia, Autism, Anxiety…)

• Social behavior in humans is complex; studying it in animals is challenging

  • Three-chambered social interaction test (SIT) - for general interactions b/w animals

  • Intruder test (for aggression)

  • Maternal behavior

Social interaction test (SIT)

• Occurs in three phases

• Part 1: Habituation to the apparatus

• Part 2: Mouse (M1) in one chamber

• Part 3: Novel mouse (M2) in new chamber + old mouse (M1) in new chamber

• Sample data: How does the transgenic mouse differ in traits?

---

• mice are singly-housed

• like interacting, but also with newer mice

• there’s an expected result - anything different is interesting (but must consider other factors as well)

• interacting w new mouse = indicative of memory + preference for new

• compared to some new, some old = not so much memory + preference

The road to a new drug - drug development

• 10+ years and billions in USD

• Assumes the drug works and is safe in humans (not guaranteed)

• Drugs can be pulled from the market at any time (due to adverse effects not shown in earlier tests)

Can we tell in advance if a drug has a high potential for abuse?

• drug reinforcement

• Studying reinforcing properties

• Drugs as reinforcers

Drug reinforcement

• Addiction has been defined as “compulsive engagement in a behavior (drug use) despite harmful consequences”

  • Term is being phased out; the term substance use disorder now preferred

• Some drugs might carry a higher risk for abuse

  • Consideration for both recreational and therapeutic drugs

• To gauge the addictive potential of a substance, we can test its value as a reinforcer (S+)

Studying reinforcing properties

• Drugs given with a certain number of bar presses (fixed ratio schedule)

• The number of presses is key; if you require too many they will stop performing for drug (break point/BP)

  • BP indicates “effort for drug”

• Higher BPs suggest reinforcing properties

---

• reinforcement = amount of bar presses they’ll do for drug

• BP = amount of bar presses to get response

• not high BP for psychedelics

Drugs as reinforcers

• 
  

Drug development + drug designs

• Few new drugs developed based on rational drug design

  • Many major drugs in use today were discovered by accident or via screening

  • Even when we develop new drugs, they tend to use “old” mechanisms

• Possible reasons for lack of progress

  • Lack of valid measures for disorders (face validity is insufficient)

  • Focus on traits rather than disorders (EPM)

  • Research practices (bad stats)

  • Species differences (generalization)

---

• rational drug design = selection of drugs based on models

• screening is the other way drugs developed? add onto a bit