The third and final part of my three part guest blog series on Optogenetics has been published on the Addgene blog. Addgene is a nonprofit organization dedicated to making it easier for scientists to share plasmids and I’m thrilled to be able to contribute to their blog! This post covers the running behavioral experiments utilizing optogenetics.
The second part of my three part guest blog series on Optogenetics has been published on the Addgene blog. Addgene is a nonprofit organization dedicated to making it easier for scientists to share plasmids and I’m thrilled to be able to contribute to their blog! This post covers the material science aspects of running optogenetic experiments.
The first part of my three part guest blog series on Optogenetics has been published on the Addgene blog. Addgene is a nonprofit organization dedicated to making it easier for scientists to share plasmids and I’m thrilled to be able to contribute to their blog!
Why is it that one person becomes an addict and another does not?
This is a central question in addiction field and one that I’ve touched on in some of my posts (and will continue to explore in the future). Two recent papers may help to shed more light on this difficult and complicated question. Both studies have revealed changes that occur in the brain as a result of childhood trauma that may cause an individual to be more susceptible to risky behavior such as drug abuse.
Both papers are neuroimaging studies meaning they use living human subjects and look at brain activity in response to different scenarios. There are many ways to image a living brain but these studies both use functional magnetic resonance imaging (fMRI). Basically, fMRI measures blood flow into the brain. As neurons turn “on” (that is, when they conduct an electrical signal), they require energy. Neurons use glucose as their primary energy source, which is delivered to them through blood flow. Therefore, the more blood flowing to a region of the brain = the more energy required by neurons = more neurons “firing”.
The analysis of fMRI data is very complicated and beyond the scope of my knowledge or this discussion. But in essence, when you think or read about something, certain areas of your brain process that information. Using fMRI, you can actually visualize regions of the brain that are turning “on” or “off” when a patient thinks about a particular situation! Watch these YouTube videos for additional explanations on fMRI.
In both of the studies featured in today’s post, subjects would read different scripts while in the fMRI scanner and the scientists would image the entire brain and identify the regions that were active during the test. Then data from multiple subjects can be compiled and a composite image that represents the averages all the subjects can be produced. The picture to the right is an example of this type of composite image. Finally, you can see which regions of the brain are active for most of the patients during the different experiments. Keep this information in mind as I go over the papers.
The first paper performed fMRI scans on adolescents that had or had not experienced maltreatment or trauma during childhood (less than 18 years old). 67 subjects were recruited from a larger study looking at disadvantaged youth and 64 were eventually used in the study. The adolescents filled out a standard survey that allowed the scientists to learn which of the subjects had experienced maltreatment/trauma during childhood.
The experiment involved having the different subjects read a script about either a stressful moment, their favorite food, or something neutral or relaxing while their brains were being imaged in the fMRI scanner.
Amazingly, for the stressful scenario, a difference in brain activity was detected in multiple regions of the prefrontal cortex only in subjects that had experienced childhood maltreatment! What this means is those youths that were abused as kids responded to stress differently than youths that were not abused. Their brain function has literally been changed later in life as a result of the abuse they suffered as children.
The prefrontal cortex is a part of the mesocorticolimbic system, a group of brain areas especially involved in addiction. The prefrontal cortex is also involved in decision making, impulsivity, and other functions. It’s not clear what this change in prefrontal cortex activity actually means but it is possible that the altered activity could make the youth more vulnerable to stress or more likely to engage in risky activities, such as drug abuse.
The second study was also interested in subjects that had experienced maltreatment or trauma during childhood but it instead of adolescents, this study used subjects that are adult men dependent on cocaine. Similarly, the subjects were grouped into those that had been mistreated as kids and those that had not.
In a parallel design to the other study, the subjects read a script describing a situation while being scanned in the fMRI machine. The scripts in this study included stress, cocaine-associated, and neutral. Interestingly, an increase in activity in a specific region of the prefrontal cortex and an area of the brain involved in motor activity were detected in the subjects that had been abused during childhood. And even more important, these changes were correlated to enhanced drug craving. These results suggest that childhood trauma can affect drug craving for addicts, which may be relevant factor in triggering relapse. That is to say, addicts that have been abused as children may be more vulnerable to not only acquiring addiction but also relapse.
It is important to keep in mind that, like the previous study, the real functional importance of these different changes in unknown. However, clearly there are real changes that occur in the brain as a result of abuse/maltreatment during childhood. Imaging data must be taken with a grain of salt because it is difficult to show real causality. Yet, both studies (and many others) suggest long-lasting changes in brain activity, especially in response to stress, as a result of childhood trauma/maltreatment.
The conclusions we can draw from these studies is that childhood mistreatment, or trauma can have lasting changes on the brain. How these changes affect behavior is a much more difficult question to answer. Nevertheless, the changes that occur may be one of the factors that can contribute to susceptibility to addiction. These studies are supported by a previous post in which animal studies have shown that stress during early age leads to greater drug use as an adult.
And a broader point, these two neuroimaging studies help to put a different perspective on disadvantaged youth and importance of a stable home life, the lack of which can significantly affect you as an adult and may even contribute to susceptibility of become a drug addict.
Excellent new article on optogenetics in The New Yorker. Optogenetics is a powerful, cutting-edge tool developed by Karl Deisseroth’s lab (profiled in the article) and is one of most significant advances in neuroscience research in decades. I recently spent two months learning the technique and we will be implementing it in the lab I work in at Rockefeller University. Optogenetics allows researchers to turn specific neurons “on” and “off” and see how those neurons are directly involved in a particular behavior. The article does a great job of profiling Deisseroth himself and explaining a little bit of the history of optogenetics and other developments in the Deisseroth lab. Enjoy!
When a news article starts with the headline “A new study finds…” do you know what that means? The article is (allegedly) referring to a peer-reviewed scientific research paper. Research papers are the heart of the scientific research field and are a report of a series of experiments conducted by a scientist or team of scientists. In a future post, I’ll do a break down what a paper looks like but for now all you need to know is that the heart of the paper is the data. The data are the pieces of information that scientists have acquired from their experiments and are reporting in the scientific paper.
But how do scientists generate data?
This is one of the crucial questions in the scientific field because it refers to experimental design: 1) what is the question the scientists wishes to answer, 2) which experiments does the scientist need to design in order to answer those questions and 3) what are the different techniques and tools needed in those experiments?
This is my second post in series of posts I’m doing to show how scientists actually collect data and the various experimental techniques and tools we have at our disposal. Right now I’m only talking about neuroscience and techniques specific to the addiction field but may discuss more general biological tools and experimental techniques in the future.
In my last post in this series, I discussed the locomotor activity test (also known as the open field test), intravenous self-administration, and microdialysis. Today, I’ll discuss a behavioral technique that’s an alternative to self-administration: conditioned place preference.
Conditioned Place Preference
Recall our discussion on self-administration. It’s a powerful technique that allows animals to administer drugs to themselves. The technique also has the potential to model initiation of drug taking, maintenance/escalation in drug taking, and even relapse-like behaviors. However, there is one major flaw with this technique. It is extremely difficult and very time consuming! After all, a mouse jugular vein is really small, which makes doing the surgeries not a trivial exercise…
Is there an easier way to study addiction that doesn’t require surgery? Thankfully, there is! Conditioned place preference (CPP) is another model to test whether animals find a drug of abuse pleasurable/rewarding or not pleasurable/aversive.
The technique is based on a Pavlovian or classical conditioning mechanism. Perhaps you’ve heard of the famous Russian scientist Ivan Pavlov? In a series of very famous experiments, he was able to cause dogs to salivate anytime he rang a bell (or any neutral stimulus for that matter). Like most famous discoveries, he wasn’t trying to do this but through careful observations he uncovered one of the basic mechanisms that underlies learning.
Pavlov’s conditioning experiment was done by presenting the dogs with an unconditioned stimulus, that is to say something that will cause a response in the animal no matter what, which is called an unconditioned response. In Pavlov’s case, he would present the dog with the unconditioned stimulus of food, which would cause the unconditioned response of salivating (Figure 1). Through careful observation, he was able to identify that dogs would salivate even before he put the food in front of them, sometimes just the site of the food dish was enough to cause the dogs to salivate. He followed up on this intriguing observation.
While the food is the unconditioned stimulus, the food dish or scientist bringing the food served as a neutral stimulus that normally would have no effect on the dogs ability to salivate. Pavlov tested if he could induce this salivating effect with other neutral stimuli. A neutral stimulus that normally has no effect on the animal, called a conditioned stimulus, would become associated with the unconditioned stimulus to produce a response (the conditioned response). In Pavlov’s experiments, the conditioned stimulus (food), when paired with the unconditioned stimulus (bell), would then produce a conditioned response (salivating).
Now let’s see how Pavlov’s conditioning experiment was actually done. If he rang the bell before the conditioning (the conditioned stimulus), it would have no effect. The dogs don’t really care about the noise from the bell because it is not associated with anything in the dog’s brains. But every time the food (unconditioned stimulus) is presented to the dogs, Pavlov would ring the bell (conditioned stimulus). Now the ringing of the bell became associated in the dog’s brain with the presence of the food.
Finally, after the conditioning sessions, Pavlov would ring the bell and would remarkably cause the dogs to salivate (conditioned stimulus)! They had learned to associate the sound of the bell with the presence of the food. Just to clarify, the dogs are not “choosing” to associate the bell with food. This type of conditioning is hard wired into the brain itself—forming these type of associations is one of the things that brain does best. In fact, classical conditioning is a basic mechanism in many types of learning. To this day, Pavlov’s work remains some of the foundational experiments in the biological basis of learning.
Here’s a video I found on YouTube that summarizes everything that you just read:
The taking of a psychoactive drug can actually have a similar type of classical conditioning effect. Think about it this way, a drug is never taken in a vacuum,it is always taken in a particular context. A drug may be frequently taken in a particular location, or under particular circumstances, or even with certain people.
I’m a former cigarette smoker and this is an example of conditioning that I personally experienced. Every time I got in the car I would light up a cigarette. After months and years of smoking, I caused a classical condition effect in myself. The cigarette (unconditioned stimulus) produces that “nicotine high” and relaxing feeling that smokers crave (unconditioned response). However, driving in a car (conditioned stimulu) normally does not cause that feeling. But every time I would need to drive someplace I would smoke. Eventually, simply being in the car would cause a craving for a cigarette! The conditioned stimulus of driving became associated with the unconditioned stimulus of smoking to produce the conditioned response of nicotine craving every time I go into the car.
Classical conditioning is exactly how conditioned place preference works. In the laboratory, we can use this basic mechanism to force mice to experience a conditioned response when placed in a distinctive chamber. The mouse will even seek out that chamber and spend time in it because they know that they received a “good feeling” anytime they were in the chamber before.
This is what the chamber looks like.
It consists of three connected boxes: a central grey one with normal flooring, a white-walled one on the left with a mesh grating as the floor, and a black-walled one on the right with steel bars on the floor. There are special trap doors (white knobs in the picture below) that can be opened or closed so that a mouse is allowed to either explore the whole apparatus or be confined to one of the chambers. When the mouse is being conditioned, the trap doors are closed and the mouse stays in only one chamber the whole time.
A CPP experiment consists of four main steps: 1) the pre-test day, 2) the conditioning sessions (multiple of these), 3) the post-test day, and 4) data analysis. (See Figure 2 below).
Step 1: A mouse in placed in the central grey chamber and it is allowed to explore the entire apparatus all it wants want. Both the white and black chambers represent a conditioned stimulus because right now, they have no association with anything in the mouse’s brain. The time spent in each chamber is recorded.
Step 2: Now the mouse receives an injection of drug (or saline as a control substance) and then is placed in either the white or black chamber. The mouse is forced to stay in the chamber for the entire session (usually 15-30min). That way the features of the chamber (wall color and floor texture) become associated the unconditioned stimulus of the drug. One conditioning session occurs a day for several days.
Step 3: The test day. Now the trap doors are raised and the animal is allowed to explore all three chambers again. If the experiment worked, the mouse will spend most of its time in the chamber that it received the drug injections! In other words, the mouse was conditioned to expect the drug in either the white or black chamber and, given the choice, prefers to spend time in that chamber in anticipation of the drug.
Step 4: Analysis. The time spent in the drug or saline-paired chamber on the test day is subtracted from the time spent in that chamber on the pre-test day. This difference in time is considered the quantitative measure of a successful conditioning session.
This figure summarizes a CPP experiment:
If an animal likes a drug and finds it pleasurable and rewarding, it will spend a lot of time in the conditioning chamber (see the graph). If the mouse hates the drug, it will not spend time in the conditioning chamber. By using this setup, we can test how different drugs and doses of drugs, and other types of experimental manipulations can effect how a the mouse perceives the drug.
If we were to compare self-administration to CPP, a conditioned response in the CPP experiment would be a similar measure as self-administration of a drug. Both experiments reveal that the animal likes the drug and wants to take it.
And as with self-administration, many variations on the basic setup exist but I’ll spare you those details for now…
One of the most important questions that every scientist learns to ask is “How do you know that…?” As scientists, we are trained to be skeptical. When we consider a bit of research done by a colleague, before we are inclined to believe the data, we need to be sure that they conducted the right experiments and that those experiments were done correctly. This doesn’t mean that scientists are stubborn or closed-minded. The reality is quite the opposite. Scientists are ready to incorporate new ideas and new results but first we need to know that the data are real. That’s what being a skeptic is all about: reserving judgment until you know all the facts.
The question “How do you know that..?” is one of the intellectual tools we use when considering whether or not data are real or not. This question has two parts: 1) how do you measure the thing that you interested in and 2) how do you know the effect you are seeing is actually based on what you think it is? What type of comparisons do you need to make in order to test the effect you’re interested in?
The first point of the question relies on special tools, equipment/technology, and experimental setups that are used to take measurements. For example, if you want to know how much a mouse likes taking a drug, then you need a way to measure how much drug it takes and how often it takes the drug (more on this in a bit). Today, I’ll go over a few of the tools that we use in addiction research.
The second part is more important (and more difficult to explain) but is really at the heart of the scientific method. It is all about experimental design and making sure you make the proper comparisons and analyses. I won’t discuss these details any more right now but will save this discussion for a future post.
Instead, let’s take a look at a few of the tools a scientist studying drug addiction has in his/her toolbox.
Locomotor Activity Test
The psychostimulants amphetamine and cocaine act in very similar ways and have very similar effects on the brain. We know that stimulants sort of “amp you up” or make you feel like you have more energy. Think of how you feel after drinking too much coffee. And what do you do when you have more energy? You tend to move around more (maybe you feel a little twitchy/antsy after too much of that coffee…). The same thing happens to mice and rats.
We can measure the amount of movement using a locomotor activity test. This test uses a special piece of equipment that uses light beams and a light-sensitive detector. Whenever the animal moves around the test box, the light beams are broken and the detector records that information. One way to analyze the data is by simply plotting beam-breaks (photo-cell counts are the same thing) that occur over the time of the test period. This way you have a measure of how much the animal moves around in a certain amount of time (more beam-breaks/time unit equals greater movement). A more sophisticated analysis of this same data can actually give you information on where in the box the animal spends its time. Does is just pace back and forth in a small area of the box or does it explore the entire chamber? This type of exploratory behavior data is valuable information and can be useful to other fields that may or may not study drug addiction. The general test for this exploratory behavioral analysis, regardless of speed of the movement caused by drugs, is the open field test.
An interesting phenomenon has been identified with psychostimulants. If you give an animal an injection of cocaine it will move around more compared to regular animals. But if you give it another dose of cocaine the next day it will move around even more than it did on the first day. This is called locomotor sensitization and is an important property of psychostimulants like amphetamine and cocaine.
The graphs below are real data that I took from a figure from one of our lab’s papers so you can see what locomotor sensitization looks like.
It’s a little hard to read but there are two groups of animals: one that receives cocaine injections (the top line) and the other that receives saline injections (the bottom line). Saline is a saltwater solution that is a standard control solution that has no biological effects. Each data point represents an average of several animals from each group. The baseline graph shows the locomotor activity before injections (no differences). As you can see, at day 1 the cocaine animals are already moving more than the saline group. This increase in movement continues over the 14 days of the experiment, evidence of locomotor sensitization.
This video shows an analysis of locomotor activity using video tracking software instead of light-beam breaks.
Locomotor activity is all good and well but not all drugs of abuse cause locomotor sensitization. More directly related to addiction in humans, how do we even know if the animal likes the drug or wants to take the drug? Humans addicts crave the drug and compulsively use it, meaning the desire to do the of the drug overpowers the addict’s self-control. Is there a way we can study this type of drug-taking behavior in animals? The answer is yes!
Self-administration is a very versatile and powerful technique used throughout the addiction field. This technique allows the animal to control whenever it takes the drug and however much it wants. We can study many different aspect of drug taking using self-administration.
The basic idea is is simple: The rodent (mouse or rat) is placed in a chamber and presented with two levers. If the mouse the presses one lever (the active lever) it receives a dose of drug but if it presses the other lever (inactive lever) it does not. The self-administration sessions are run for a set period of time and the number of presses is recorded for each lever. Over the course of several days the animal steadily increases the amount of lever presses, thus the amount of drug it takes. Meaning the animal learns how to take drug and then takes more and more of it. Just like a human addict would do!
Alternatively, the mouse can poke its nose at a special hole that acts just like the active lever. I’ll use “lever press” and “nose poke” interchangeably because they essentially mean the same thing.
Here’s a little cartoon I found on YouTube of a rat that is self-administering nicotine.
Here’s another video that shows a real mouse self-administering a natural reward (meaning not a drug of abuse but food in this case).
There are several important variations to this basic idea that help scientists to not only make the experiments easier to control and data better/easier to analyze, but allow different aspects of drug taking to be studied.
For example if you are studying alcohol addiction, then when the mouse presses the lever a spout may appear that allows the animal to drink the alcohol (the inactive lever produces a bottle of water only). This is perfect for testing alcohol self-administration because both humans and mice drink alcohol. But what if you want to study heroin or cocaine self-administration? Humans (nor mice) drink or eat these drugs. So how does the drug get delivered to the mouse when it presses the lever?
The answer is intravenous self-administration. In this version, a small surgery is performed where a small tube (a cathether) is threaded into the jugular vein of the animal. This tube is fixed to the mouse back and attached to another tube that is part of the self-administration apparatus. This time when the mouse hits the lever, a dose of drug is pumped directly into its vein! See the diagram and videos above for more details.
Intravenous self-administration has several advantages.
As explained above, it allows us to deliver drugs to animals that won’t take them orally.
It allows the drug to act immediately on the animal because the drug is being delivered directly into the bloodstream.
It allows us to control the dose of the drug. When the mouse hits the lever (or nose pokes) it receives a fixed amount of drug that the scientist decides on ahead of time. That way we know how much total drug the mouse takes during a single self-administration session.
There is no variability in whether the animal is receiving the full dose or not. For example, if the lever press results in a food pellet, there is no guarantee the animal will eat the whole thing. But if you set the self-administration apparatus to deliver 0.5mg of heroin every time the lever is pressed, then there is no doubt if the full 0.5mg dose is delivered to mouse ever time.
Warning: not for the squeamish! This video shows you how to do the catheter implantation surgery on a mouse that will be used for intravenous self-administration!
Finally, best of all, self-ad can be used to address many different types of questions related to different stages in the addiction cycle. Here I briefly describe some of the more common experimental questions and applications that self-ad can help to address.
Initial use and escalation of use. How much will the animal take when it is first exposed to the drug? Will the animal reach a ceiling in the amount of drug it will take in a single session?
Maintenance of drug taking. One cool variation is you can make it more difficult for the animal to get the same dose of drug. This is called a progressive ratio self-administration. For example, the animal may need to press the lever 5 times before it receives a dose. You can keep increasing the number of presses during each session to see how hard the animal will work for a dose. One way this experiment can be interpreted is how badly does the animal want the drug? Some animals will press the lever many, many times just to get a small dose. This type of behavior is similar to the intense cravings that human addicts can experience.
Extinction and Relapse. You can run a special type of experiment where you run a self-administration experiment like normal and then change it so that the active lever no longer gives the animal a dose of drug. Eventually, the animal presses the lever less and less as it learns that it will no longer get the drug. This is called extinction of self-administration. This is like being in a rehab clinic where you are prevented from taking the drug. However, after the extinction sessions, if the scientist gives the animal another does of drug this will causes animal to start pressing the lever at high rates again. This a called reinstatement of self-administration and is model of relapse. What other types of conditions or factors can cause reinstatement (relapse behavior)? This situation is just like an abstinent cocaine addict who may not be craving cocaine but if he/she takes even a single hit, this can be sufficient for that person to sink back into full-blown addiction.
Let’s take a look at some real data. The graph below is from a paper from our group that looks at oxycodone self-administration in mice.
This study is interested in comparing oxycodone self-administration between adult mice and adolescent mice. As you can see, the number of nose pokes at the active hole (remember, same thing as a lever presses) increases during the course of the experiment (don’t worry about FR1 vs FR3) while the inactive hole is ignored, because it does not result in drug administration. Note that the nose pokes are plotted over the time of the administration sessions (2 hours) and that 9 sessions are run (one every day).
The types of experiments I’ve described so far are great ways of studies addictive behaviors but they don’t really tell you about what’s going on in the brain. These behavior experiments are useful in themselves but they are much more powerful if they can be combined with another type of experiment that gives you a window into what’s changing in the brain at the same time as the behaviors.
In my post Synapse to it, I described how neurotransmitters are released by the pre-synaptic neurons into the synaptic cleft so that they can act on receptors located on the post-synaptic neuron. Using microdialysis, you can sample the fluid that exists in the synaptic cleft and actually measure the amount of neurotransmitters being released!
This is an extremely difficult and very technically complicated technique and I will only go into the basics about it. First, the microdialysis probe is surgically placed into a region of the brain that you are interested in studying.
The microdialysis probe itself is like a very thin piece of tubing that allows the experimenter to flow fluid into it one side(inlet) and collect the fluid that flows out of the other side (outlet). At the tip of the probe (the part that’s actually inside the brain) is a special type of material that allows fluid from inside the brain to flow into the tubing (a semi-permeable membrane).
After the surgery, you run your behavioral experiment, and while you are doing that you start flowing fluid into the brain. The fluid that the microdialysis probe flows in is of a similar consistency to the fluid that exists naturally in the brain. As the fluid inside the probe moves through the tubing, it causes fluids in the brain to enter into the probe and through the tubing where it can be collected when it flows out of the tubing.
Let’s say you give an animal a drug that causes a neurotransmitter to be released in the brain region you are interested in. Then some of those released neurotransmitters will enter the microdialysis probe because some of the fluid that enters the probe is from the synaptic cleft.
You keep collecting fluid at different time points during your experiment. When the experiment is over, then you can use chemistry to determine what neurotransmitters are in the fluid you collected. Best of all, you can determine how much of those neurotransmitters you have! How you do actually use chemistry to do this is a very technical part of the procedure and is not important to this discussion.
And all that work gives you a nice graph of the neurotransmitters that are released at different times during your experiment.
Now for some real data. Below are figures from a paper that our lab produced that uses microdialysis to study release of the neurotransmitter dopamine.
In this study, the effect of cocaine on dopamine release in a region of the brain called the caudate putamen is being studied. The first image shows you that the microdialysis probe was placed in the right area of the brain (the white line that pierces through the dark area is the tract in the caudate putamen). The graph shows that injection of cocaine (the arrows) causes an increase in dopamine release in this brain region. Interestingly, the dopamine levels have returned to normal by the end of the experiment. Note: C57Bl/6J is the strain of mouse used in this study.
These are just three of the techniques that are used in addiction research. But we scientists have very big toolboxes! I’ll to explain some more in a later post.
Feel free to contact me or comment if you have questions!