Meth is a weak base, yo

September 20, 2012


[The above image was lifted from Mike Mitchell, and I added the chemistry bits]


The idea for our $25k meth crowdfunding project came to me earlier this summer after a meeting with David Sulzer and members of his Columbia Med School lab. Sulzer’s lab has studied how amphetamines work for over 20 years. I had emailed Sulzer a few weeks prior to our meeting with an invitation to discuss the broader implications of my lab’s April PLOS ONE paper, which revealed new insights into the effects of the SSRI antidepressant sertraline, aka Zoloft, on the structure and function of cell membranes.


The foundation of our mutual interest is the 1990 Neuron paper by Sulzer and his mentor Stephen Rayport, in which they debuted the “weak base” hypothesis of amphetamine action, to which I’ll return after a brief historical interlude.


It’s been known for decades that amphetamine and amphetamine-like compounds stimulate the release of catecholamine neurotransmitters, e.g., dopamine, from the synapses of neurons, yet we (pharmacologists) still haven’t enumerated all of the cellular targets of amphetamines. By cellular targets I don’t just mean protein targets, but also the vital chemical equilibria that are perturbed by amphetamine accumulation in cells, e.g., ion gradients.


Conventional wisdom holds that amphetamine directly competes with dopamine for access to the dopamine transporter protein (DAT), which is expressed in dopaminergic neurons, and selectively slurps up dopamine after each round of neurotransmission so that it can be repackaged into fresh synaptic vesicles for future rounds of excitation. This model is based on structural similarity between amphetamine and dopamine, as shown here:



However, the affinity of amphetamine for DAT is reportedly weak in some studies but strong(er) in others. The trouble with measurements of drug binding affinity is that different pharmacologists do the experiment differently. In other words, people compare the ability of a panel of drugs A-Z to displace a primary compound that binds to DAT with high affinity and selectivity, but the choice of primary compounds, which can bind to different locations on the DAT protein, often varies between studies.


A comparison is often made between cocaine and amphetamine, since their addictive properties are both thought to involve increasing dopamine levels. Remember those self-administration drug studies with lab rats? This excerpt from a 1987 Science paper of that ilk by Ritz et al sums up the situation well:


CNS stimulants can be divided into two classes on the basis of their biochemical effects on catecholamine-containing neurons in brain: the amphetamines and the nonamphetamines (cocaine and methylphenidate). Drugs in the former class inhibit reuptake and are potent releasers, while drugs in the latter class inhibit reuptake but are more restricted in their releasing properties.”


Alas, the rabbit hole goes deeper still. Amphetamine and methamphetamine have been shown by some labs to cause DAT to be removed from the synaptic membrane, resulting in DAT sequestration in intracellular compartments, e.g., endosomes, lurking below the membrane surface. But as is often the case in experimental pharmacology, the reproducibility of these results depends on the duration of drug treatment and whether cells were exposed to drug in vitro vs in vivo. Annette Fleckenstein’s lab does solid work here, and her lab’s publications are required reading for budding psychopharmacologists.


So, getting back to the weak base hypothesis. It’s really a simple idea. Even a fleeting glance by the trained eye reveals the chemical propinquity between amphetamine and dopamine: both molecules possess an amine that is ionized, in this case protonated, at physiological pH. Nature exploits this amine in an ingenious way. Synaptic vesicles are loaded with amine-containing neurotransmitters like dopamine on the basis of a pH gradient between the inside of the synaptic vesicle, which the cell spends energy to make acidic, and the surrounding cytoplasm, which is buffered at a comfortable near-neutral pH. The rest basically flows from acid-base equilibrium theory, e.g., Henderson-Hasselbach equation.


So the reason why amphetamine causes dopamine release from synapses is that amphetamine is a stronger base than dopamine, and competes with dopamine for the finite storage capacity of tiny synaptic vesicles. Now there’s more dopamine in the cytoplasm, which seeps out of synapses through DAT.


A hypothesis is only as good as its testable predictions. To wit, 22 years after the original weak base paper was published by Sulzer & Rayport, a group led by Teja Groemer and collaborators showed – coincidentally also in the pages of Neuron – that antipsychotic drugs, which are also have a pesky amine, are subject to weak base forces; are accumulated in synaptic vesicles; are released into the synaptic space in response to neuronal activity. (If anyone’s interested, I started a running discussion of Groemer’s paper on my website, which includes comments by Groemer himself, here).


To be totally fair though, the notion of weak base accumulation may have been new to the field of psychopharmacology in 1990, but Christian de Duve, the discoverer of the organelle called the lysosome, appreciated back in the 1960s that anti-malarial compounds, e.g., chloroquine, which accumulate in acidic organelles, are also weak bases.


Like a lot of powerfully simple ideas in biology, they seem to be forgotten and re-remembered over and over again.


So why did I email Sulzer in the first place? On top of all the weak base considerations, there’s another facet of amphetamines that actually distinguish them from their natural counterparts. They are more hydrophobic. Go back to the above comparison figure. You’ll notice that the catechol group in dopamine is plain vanilla benzene ring in amphetamine. By way of a culinary analogy, that means that given a choice, amphetamine is much more likely to choose oil over water if dissolved in vinaigrette.


My lab has shown that the antidepressant sertraline/Zoloft accumulates in yeast cell membrane with both rapid and slow-acting effects of yeast cell physiology. In 1994, Sulzer’s lab showed that methamphetamine, which has an extra methyl group vs amphetamine, causes gross changes in the endocytic pathway of cultured neurons, including the induction of autophagy, which my lab showed is induced in yeast cells treated with Zoloft, and so may be a general property of what hydrophobic weak base drugs. I’ll review the literature on the direct effects of amphetamines on membranes in my next post.


So where exactly are amphetamines accumulating inside brain cells, and what is the physiological significance of this accumulation? Seems like a question ripe for Crowdsourcing Discovery


Here is the previous post on perks and price points.

  • Giuseppe Gangarossa

    Cocaine and amphetamines share (more or less) a common mechanism of action: blockade of DAT. This raises many questions! Do cocaine and amphetamine induce same behavioral effects? For the majority of neuroscientists the answer is: if you inject these psychostimulants to mice (or rats) they will run and they will be conditioned. Right but…! Other answer: both drugs activate the rewarding system. Right, but…! (Scientifically speaking I love saying but…!). Unfortunately (fortunately) the reality is much more complex and our mania-like behavior of oversimplification usually drives a lot of mistakes. For some unknown reasons cocaine and amphetamines do not induce the same pattern of neuronal activity, do not induce the same transcriptional and translational changes and even more interesting they act on completely different brain territories even within the same brain structure (i.e. striatum, prefrontal cortex, nucleus accumbens, thalamus). Why, when we talk about DAergic psychostimulant drugs, do we focus preferentially on the ventral tegmental area (VTA) and not on the substantia nigra pars compacta (SNpc)? Both of them are the sources of DAergic neurons which express both DAT and TH, transporter and synthesis enzyme of dopamine, respectively…Again the answer is: matter of oversimplification. We like saying VTA= rewarding system and SNpc= motor control. Is it true? Not really.
    Other thing…amphetamine and methamphetamine are really close each other (chemically speaking). So, why their behavioral and biochemical effects are not similar?
    All these questions just to say that we don’t know so much about psychostimulants’ mechanisms and that all the efforts to understand how addictive drugs work are more than welcome and needed!!!!!

  • bill

    OK, time for a dumb question — the bete noir of Open Science, being flooded with morons like me. But you can always ignore.

    What makes a weak base tend to accumulate in an acidic environment? Is the protonated state more energetically favourable? E.g. amines carry a charge when protonated –> more stabilizing electrostatic interactions with e.g. water?

    Actually, make that two questions: why do you consider the storage space of synaptic vesicles to be finite? Can it not be expanded by vesicle synthesis, if there’s more to package? Presumably the kinetics of vesicle synthesis vs vesicle cycling comes into play here…?

    • Ethan Perlstein

      Weak base accumulation in acidic compartments has to do with acid-base equilibria. The ionizable group of weak base drugs is the amine, which is protonated in acidic environments. (It is also protonated in neutral environs, but the tendency to remain protonated increases with acidity). The protonated form of weak base drugs cannot cross cell membrane, so when they diffuse into an acidic compartment from a neutral medium, they get stuck. And, as you say, the protonated form of a weak base drug is more water-soluble.

      In terms of synaptic vesicles, I was referring to the volume of a single synaptic vesicle. Of course the cell can modulate the overall rate of vesicle formation to create more vesicles en toto, though their size distribution is thought to be relatively constant.

  • Ethan O. Perlstein

    Give credit where it’s due, man. That illustration is by Mike Mitchell. Other sites are claiming it as yours.