Never underestimate the power of a thorough literature review

As someone who is currently writing their thesis, I have a major piece of advice for PhD students embarking on their studies…

Never underestimate the power of a thorough literature review!

Always, always, always read as much as you can about the field of research you are working in. Especially if there are numerous inconsistencies and unresolved controversies surrounding the expression/function of proteins/receptors and the pharmacology of their ligands. I’m specifically talking about the field of cannabinoid research and its relatives.

I carried out one of my early experiments* under strict instructions from my project manager and supervisor. Their rationale at the time sounded ok to me: look at the effects of drug A in the presence of drug B, which happens to be a selective antagonist for receptor X. My knowledge of the literature at the time (which was quite substantial, with over 100 references for the literature review I wrote 9 months after starting the PhD) concurred that this seemed like a legitimate thing to do: If drug B, which has no effect on its own, blocks the effects of drug A, then drug A must bind to receptor X.

Seems legit

Recently however, while doing a bit of extra research on the drugs I used**, I came across a few research articles (written by one of the most famous scientists in the field) that described both drug A and drug B to be inverse agonists of receptor X. Basically, that means both drugs should produce an effect by themselves. The same effect. At the same receptor.

*face palm*

Great. How on earth am I supposed to explain that in my thesis, let alone in my viva?

Luckily, my trump card was the fact that neither drug did the same thing when applied individually, which kinda goes against the ‘inverse agonists at the same receptor’ theory. Also, drug A appeared to be antagonised by drug B afterall. To top it all off, I found more research articles with evidence to suggest that not only do drug A and B bind to receptor X, but they also bind to receptor Y!

Hurrah for the inexplicable and frustrating weirdness of cannabinoid research!

Seriously though, do lots of reading. Read the reviews written by the most famous and respected scientists. Read the original articles that they refer to. Read articles about similar research conducted at other laboratories. Even read the letters and commentaries on these articles written by other scientists in the same field – they might dispute the findings, or highlight problem areas. Even attending conferences can open your eyes to what other scientists really think.

However, try to stay open minded as science is subject to fashion and popularity. Established scientists can denounce the results of perfectly good work just because it challenges the dogma. But sometimes it can take a fresh pair of eyes and a different way of thinking to make real progress.

Never make the excuse that you don’t have enough time to read research papers. There is always time to read papers. Your supervisors/advisors may be an expert in a particular field of research, but you are the expert in your PhD. Try to make sure that every experiment is properly thought through because you don’t want to be struggling to find reasons and explanations when it comes to your thesis/viva. Saying ‘My supervisor told me to do it’ is simply not good enough.

*This was supposed to be a simple control experiment to show what the majority of the literature has already described. My ‘proper’ experiments involved trying to figure out how drug C works.

**Drug A and drug C are similar, but there’s so much more written about drug A that I use this to help me interpret what drug C does.

A typical day – In the beginning

A typical day at the beginning of my PhD was very different to a typical day towards the end of my third year.

When starting a PhD (at least in Life Sciences), before anyone is let loose in a lab, there’s a lot of form-filling, health and safety inductions and even occupational health assessments that need to be done first. Sometimes this can take a while and usually depends on how organised and attentive your supervisor is.

In the meantime, there’s a huge amount of background literature that needs to be thoroughly read, understood and reviewed. This part can be quite tedious, as sitting at a desk reading doesn’t feel like you’re doing much at all, especially when you’re surrounded by busy PhD students and post-doctoral researchers.

In the beginning I used to turn up at around 9.30am, to find most PhD students and post-docs already at their desks or in the lab doing experiments. It used to make me feel a bit guilty, especially as almost all of them would still be there when I left the office at 5pm. But in actual fact, it’s entirely normal to feel like you’re at a loose end when everyone around you seems to be working incredibly hard.

After all the necessary paperwork and admin stuff was out the way, I was able to get into the lab and start learning how to perform my first experiment. Again, this was a little tedious because it mainly involved shadowing other people while they tried to explain what they’re doing and why they’re doing it. For me, this was learning how to ‘patch’ (patch clamp) cells in fresh brain slices and it took about 3 months to go from absolute beginner to first successful attempt. I think it was another month or so before I was confident enough to do a ‘proper’ experiment.

Once I’d found my feet and got into a bit more of a routine, I started to feel less guilty about not working as hard as some of the more seasoned researchers in the lab. I suppose it’s difficult to make yourself work hard when you’re still trying to fathom exactly what you’re PhD is about. I know numerous people, myself included, who didn’t include any data from the first year in their thesis. For some, it was simply due to the direction that the science took them in and for others it was because their experiments didn’t work or weren’t optimised enough.

My advice to people who have just started a PhD; Be patient. The stress will come later.

Guilty feelings

The feelings of a PhD student just starting out


When the stress gets too much

It’s not easy being an electrophysiologist. Most of the time we’re under a lot of pressure and not just from our supervisors and managers, but from ourselves too. Sometimes we go for weeks, even months, without a single successful experiment. We know that it has worked for us before, but now that that all important deadline is coming up, electrophysiology refuses to play ball.

Very true

Very true

When asked “what’s wrong?”, most of the time the answer will be “I just don’t know”. This is not because electrophysiologists are lazy and have no idea what they’re doing. On the contrary, we’re very acutely aware that many, many different things have to be just right in order for an experiment to work – and sometimes even that is not enough.

Disaster can strike at any time

Disaster can strike at any time

Going through the list of things that have to be absolutely perfect can be time consuming, but any electrophysiologist worth their salt will have a vague idea of what the weak links are likely to be.

Some days are a bit 'meh'


Some days the electrodes  are not absolutely perfect

Sometimes the problem can be really obvious.

Another experiment ruined and a day wasted

My experiment may not have worked, but I do enjoy a bit of DIY

I hate it when the perfusion system doesn't co-operate

Occasionally stress can be released in the form of exclamation marks and profanities.

Sometimes the tools you work with can let you down and spoil your day

And right after I put on some drugs too

Sometimes it's the computer program that can let you down

But when the stress gets too much, sometimes the best thing to do is walk away, have a cup of tea and calm down.

Next time you see an electrophysiologist alone in the corner with a cup of tea, approach with caution

Next time you see an electrophysiologist alone in the corner with a cup of tea, approach with caution


Be careful! Sometimes pent up stress can erupt when you least expect it:

A group of PhD students were in the breakout space one afternoon, taking a break from work. Everyone was happily chatting away and conversation turned to favourite Family Guy sketches. Someone quoted a particularly funny phrase and everyone laughed. One person in particular laughed so hard, he could barely breathe. Everyone starts laughing again and this time two more people laughed so much that they struggled to speak. The uncontrollable laughing continued, but not everyone was doing it anymore. Only three people were laughing, but their sounds of laughter started to sound more and more like sobs. There was a bit of an awkward silence and one by one people started to leave and return to the lab. The PhD students wiping their eyes and taking deep breaths looked round at each other and realised that they were electrophysiologists.

Like I said before: if we didn’t laugh, we’d cry and believe me, we’ve cried a lot.

You have been warned


“So what do you do?”…

We’ve all been there. Whether it’s at a family gathering, a friend-of-a-friend’s party or in the queue for the beer festival. At some point we’ve been asked by someone we vaguely know (or a complete stranger) “so what do you do?”.

Sometimes we’re waiting for it, as sooner or later the small talk will turn to what we do on a day-to-day basis. Sometimes the question is completely unexpected and results in a slightly blank expression as we’re trying to gather our thoughts.

I’m always ready to answer this question.

“So what do you do?” Well, the general area of my research is using cannabis to treat epilepsy.

Photo courtesy of Marcus Haag

Presenting my research in the form of a Science Slam. Photo courtesy of Marcus Haag

As you can probably imagine, people tend to latch on to the ‘cannabis’ part of that statement and I’ve found that I usually get two types of responses. I either get something like “Wow! That’s really cool” or “Hmm, ok”. To be honest, I’m not surprised that some people are sceptical about using an illicit drug to treat a disease – I certainly am! But the thing is, there’s so much more to it than simply giving cannabis to a person with epilepsy.

The cannabis plant contains around 500 chemicals. Of those ~500 chemicals, approximately 100 are unique to the plant and are known as cannabinoids. Of those ~100 cannabinoids, only 1 produces the euphoric effects associated with taking cannabis. This chemical is known as ‘delta-9-tetrahydrocannabinol’, or simply ‘THC’ for short. It’s described as being ‘psycho-active’ because it binds to, and activates, cannabinoid receptors that are expressed within the brain (CB1 receptors).

What about the other 100 or so cannabinoids found in the plant? Well, some of these have been studied in a variety of disease states including cancer, diabetes, multiple sclerosis, ulcerative colitis, rheumatoid arthritis, schizophrenia and epilepsy. The main one studied is known as ‘cannabidiol’, or simply ‘CBD’ for short. But why is CBD so special? Well, it’s NOT psycho-active. It doesn’t produce the euphoric effects that THC is famous for because it simply doesn’t bind to, and activate, CB1 receptors. It’s also the second most prevalent cannabinoid found in the cannabis plant, which makes growing and extracting it relatively easy.

“If CBD doesn’t bind to CB1 receptors then how does it work?” The thing is we just don’t know. There are many other receptors that have been reported to be implicated in the mechanism of action of CBD, such as the ‘capsaicin receptor’ (TRPV1), the most common serotonin receptor subtype (5-HT1A) and the ‘orphan cannabinoid receptor’ (GPR55). However, none of the effects of CBD at these receptors are convincing enough for the world’s leading cannabinoid scientists to stand up and say “CBD is a potent agonist/antagonist at this receptor and that is how CBD mediates all of its pharmacological effects”.

“What has all this got to do with your PhD?” Over the last couple of years there’s been a surge in interest, particularly the USA, in people with epilepsy using whole cannabis extracts to reduce the frequency of their seizures. This isn’t anything particularly new as cannabis has been used as a medicine for thousands of years. However, the thing that has made people sit up and notice is the fact that some children with severe epilepsy, in which their seizures cannot be controlled and is therefore life-threatening, have been given extracts of cannabis, in which CBD is the main component, and their seizures have dramatically reduced in both severity and frequency. You can read a couple of articles from CNN here: Jayden David and Charlotte Figi.

Unfortunately there are some issues with using homegrown cannabis preparations such as medical marijuana.

Firstly, the preparations still contain a significant amount of THC, despite growers’ best efforts to produce plants that predominantly contain CBD. Also, we still have no idea what the impact of regular THC exposure has on an adult’s brain, let alone a child’s brain and their neurological development.

Secondly, the preparations are unregulated and inconsistent. The relative proportions of cannabinoids present will vary depending on the growing conditions, the methods of extraction and the preparation of the final product. This is a major problem for people with epilepsy. I spoke to someone from California, who sometimes takes medical marijuana for their epilepsy, and he told me that whenever he went to the dispensary to get medical marijuana he had no idea what he was getting in terms of THC levels. He wants to be able to lead a normal life and have a job, but he can’t do that when he’s high on medical marijuana with too much THC in it. Also, in terms of being able to control seizures, it’s extremely important for medication to be consistent. Even a slight change in regime, dose or drug can potentially result in rebound seizures and do more harm than good.

Finally, there are legal restrictions on the use of cannabis-derived medicine in research. Currently in the USA, CBD is classified as a Schedule 1 controlled substance and federal law prohibits its use, even for research, despite the fact that it is NOT psycho-active. Yet, paradoxically, medical marijuana containing THC is available for ‘medical’ uses in about a third of the states in the USA. When I attended a conference in San Diego, California last year, I was astounded by the sheer number of American researchers telling me how incredibly lucky I was to be able to do research on plant-derived cannabinoids (particularly CBD).

I personally think that we need to move away from medical marijuana and pay more attention to regulated botanical drug substances (BDSs) derived from cannabis, in addition to isolated and purified non-psycho-active phytocannabinoids. Thankfully this has already been going on for many years by the pharmaceutical company who sponsor my PhD (GW Pharmaceuticals).

Presenting my research at Neuroscience 2013. Photo courtesy of Immy Smith

Presenting my research at Neuroscience 2013. Photo courtesy of Immy Smith

“OK, but I still don’t get what your PhD is all about…” Right, so I’ve already explained that (anecdotally) cannabis and components of the cannabis plant (eg. CBD) can treat people with epilepsy. For over a decade now, scientists have been conducting pre-clinical studies in order to put together enough evidence that justifies putting non-psychoactive plant-derived cannabinoids through clinical trials for epilepsy. This basically means that several PhD students, like myself, have been doing experiments on live animals or tissue taken from dead animals, in order to justify giving these drugs to people who have a life-threatening disease. I must also point out here that the animals used in these experiments are rats and mice, and that every person doing experiments with live animals MUST hold a personal Home Office license – I will go into more detail about this another time.

Preparations for electrophysiological experiments. Photos courtesy of Tom Hill

Preparations for my electrophysiological experiments on fresh rat brain. Photos courtesy of Tom Hill

So, clinical trials are currently underway at GW, which means that we know at least 2 non-psycho-active phytocannabinoids (CBD and its propyl analogue CBDV) definitely work in animal models of seizure and epilepsy.

But the burning question is “How do these compounds work as anti-convulsant and anti-epileptic drugs?” Well, that’s where my PhD project steps in. I’ve been doing electrophysiological experiments for the last 3 years in an attempt to figure out how these compounds, CBDV in particular, work as anti-convulsant and anti-epileptic drugs. I do this by taking recordings from individual brain cells, as well as networks of brain cells, from fresh rat brain slices. I record what brain cells do under ‘normal’ conditions and then record them again under ‘drug’ conditions and see if there’s a difference. I also record the ‘normal’ and ‘drugged’ activity of brain cells from rats that were epileptic. To then add another level of complexity, I compare the differences between non-epileptic brain cells and epileptic brain cells in terms of their response to the drug.

“Yikes! Now my brain hurts” If you want something that’s a little easier to understand (and A LOT more fun) then you can watch my Science Slam.

“So what have you discovered?” That there are no easy answers in science. Every time I think I’m getting close to answering a particular question, more questions spring up in its place. I suppose in that regard, conducting scientific research is a bit like fighting the Lernaean Hydra. I’ve also discovered that electrophysiology is a cruel and fickle mistress.

In terms of actually discovering something that makes a substantial contribution to science, I’m afraid I can’t say at this point. This is not necessarily because I haven’t found something, it’s more about keeping everything under wraps until I am permitted to reveal all. But don’t worry, there are a few papers in the pipeline.

You can read more about the case for medical marijuana in epilepsy, the case for assessing cannabidiol in epilepsy and a critical review about CBD and its therapeutic role in epilepsy and other neurological disorders for free in the journal Epilepsia.


05/08/2014 edit: It’s been brought to my attention that the Epilepsia articles are no longer free to view. If you really want to read them, but can’t get access – let me know and I’ll see what I can do 🙂