Why Do Side Effects Happen?

Side effects happen because a molecule cannot tell one part of your body from another. It fits a receptor — a structure on a cell that a particular molecule slots into, causing a change inside that cell. Wherever that receptor exists, the molecule does its one thing. Your body is full of places you were not thinking about. The molecule does not know which of them interested you, and it has no way of finding out.

That is the short answer, and it is worth sitting with, because most people picture something quite different. The phrase sounds like a malfunction. Something slipped, or the body reacted badly, or the compound was impure. Usually none of that happened. The compound did exactly the one thing it does, in every place equipped to receive it.

Is a side effect the molecule going wrong?

The phrase itself misleads. "Side effect" sounds like a side road — a second process that branched off the main one. Two separate things happening: the effect you wanted, and then something else that crept in.

In most cases there is only one process. A molecule fits a receptor and switches it on. That happens identically everywhere the receptor exists. What varies is not the molecule's behavior. What varies is the job of the cell it lands on.

So the effect you wanted and the effect you did not want are frequently the same event. You gave them different names based on which one you were hoping for.

This is not word-play. It changes what you can reasonably expect. If side effects were accidents, you might hope for a cleaner batch, or a purer version with none. If they are the intended process happening in an extra location, the only way to remove them is to change where the molecule can act at all.

The same effect, somewhere you were not thinking about

Picture a master key cut for the front door of a building. It works. It also opens the basement, the roof hatch, and a storage room on the third floor, because those locks happen to be the same model. The key is not defective. It is doing precisely what a key does, at every lock it fits.

Receptors are like that far more often than people assume. Cells across the body build the same receptor whenever they need the same signal. A receptor involved in one organ's work will often appear in two or three others as well. Evolution reused parts. It did not arrange the body for the convenience of anyone designing drugs.

A compound circulating in your blood reaches all of those places within about a minute. Every cell carrying the matching receptor responds. The cells you cared about respond, and so do the rest, at the same moment, in the same way.

Let that be concrete, because the general version slides past too easily. Suppose one kind of receptor sits in fat cells and also in heart muscle. A compound built to act on the fat cells will act on the heart too. Not because something went wrong. Because heart muscle put out the same receptor, and the molecule cannot read the label on the room it walked into.

This is the single most clarifying idea in the subject, and it is the reason side effects are not a separate topic from effects. They are the same topic, viewed from a room you were not watching.

What does selectivity mean?

Selectivity is the word for how narrowly a molecule fits. A highly selective molecule slots into one kind of receptor and largely ignores everything else. A poorly selective one fits several related receptors, and does something at each of them.

Why both exist comes down to shape. Receptors come in families — groups of related structures built on the same underlying plan, differing in the details 12. Members of a family can resemble one another closely. A molecule shaped to fit one will sometimes fit a cousin, loosely, but well enough to switch it on.

Think of the family as a row of doors with similar locks. A key cut precisely for one may still rattle a neighboring lock open. How precisely it was cut is exactly what selectivity measures.

• Highly selective — fits one receptor and mostly ignores the rest. Fewer effects, and easier to predict.
• Poorly selective — fits several related receptors. More effects, wanted and unwanted, arriving together.
• Neither is automatically better. Sometimes reaching several receptors at once is the whole point.
• Selectivity is a matter of degree, never a yes or no. Nothing is perfectly selective.

Improving selectivity is one of the slowest and most expensive parts of developing a medicine. Teams spend years making thousands of small variations on a molecule, testing each one against a panel of related receptors, hunting for a shape that keeps its grip on the target and loses its grip on the neighbors. Most attempts fail.

This is one of the main differences between a compound that became a medicine and one that did not. Plenty of molecules do the wanted thing perfectly well. They never made it, because they also did four other things, and nobody could separate the four from the one.

One receptor family, several jobs

Melanocortin receptors make the point cleanly. They are a family: several related receptors, built on the same plan, sitting in different tissues. Some are in the skin, where they influence pigmentation — the amount of dark pigment your skin produces. Others sit in the brain and take part in appetite regulation. Others again are elsewhere, doing other work.

Now imagine a molecule that fits several members of that family at once. It will affect pigmentation. It will affect appetite. It will do whatever the remaining members do, simultaneously, because all of those receptors are equally available to it 4.

Notice what is not happening. Nothing malfunctioned. Skin darkening is not the molecule going astray. It is the molecule doing its one trick at a receptor that happens to live in skin. The word "side" is carrying a lot of hidden work here. It means "the part I was not asking about," and nothing more than that.

Once you see this shape, a great deal stops being surprising. Two results in two tissues are not two events. They are one event, reported twice, from different rooms of the same building.

Why does nausea come up so often?

There is a second reason effects turn up where you did not want them, and it is simpler still. Sometimes the target sits right next to something else.

The brain is not one uniform organ. It is a dense arrangement of small regions, each doing particular work, packed tightly against each other. A signal aimed at one region does not stop politely at its border. Molecules spread through the surrounding tissue, and nearby cells carrying the same receptor respond too.

Nausea from compounds that affect appetite is the standard example. Appetite and fullness signaling in the brain happens very close to the region that triggers nausea and vomiting 3. They are neighbors. Acting on one tends to touch the other.

So nausea, in that setting, is not a sign that something went wrong. It is close to unavoidable given the floor plan. If you want to reach the appetite regions with a signal, the nausea region is next door, and it is listening.

This one is geography rather than chemistry, and it is genuinely useful to hold onto. Some unwanted effects come from receptors in distant organs. Others come from receptors a fraction of a millimeter away from the ones you were aiming at.

An empty list is not a short list

This is the part that matters most on a site about compounds that are mostly unstudied, and it is easy to get backwards.

When a medicine reaches approval, its effects have been tracked deliberately. People took it under observation, for a set period, and were asked afterward what happened to them. Everything reported got written down, including things nobody predicted and things that later turned out to be unrelated. The result is a long and fairly unflattering list.

A compound that never went through that has no such list. Not a short one. None at all. Nobody was watching, so nobody wrote anything down, so there is nothing to read.

From the outside, the two situations can look similar. One page says "known effects include" and then runs on for a while. The other page says nothing. Reading the silence as reassuring is a very natural mistake. It is only silence.

There is a further wrinkle worth knowing. Even where people describe their experiences informally, that is not systematic collection. Nobody knows how many had a given experience and never mentioned it, or mentioned it somewhere no one was reading. Counting only works when somebody set out to count.

None of this means an unstudied compound is dangerous. It means nobody knows, including the people selling it and the people writing about it, and quiet should not be mistaken for a result.

What side effects actually tell you

A long list of effects is not proof that something is dangerous, and an empty one is not proof that something is safe. Both are mostly evidence about how carefully anyone looked.

Nor are side effects a sign that a compound is fake or badly made. They point the other way. Effects appear because something is fitting something and switching it on. A molecule that does nothing anywhere is a molecule that is not fitting anything, and that is the one case where nothing at all happens — including the thing you were hoping for.

So the useful way to read this is not as a list of risks bolted onto the end of an article. It belongs in the description of what the compound is. Which receptor does it fit? How narrowly does it fit it? Where else does that receptor live, and what sits next door? What do the cells in those places do for a living?

Answer those four questions and you have described the wanted effect and the unwanted ones in the same breath, because they were never two separate things. Knowing what something does elsewhere is part of knowing what it does at all.