The Lock-and-Key mechanism
Envision yourself in an endless hallway with countless locked doors. Behind each door is a room filled with a scent - vanilla, jasmine, sea spray, you name it - even some that you've never smelled before in your life!
Each lock is unique, but it doesn't require the key to be a perfect fit to be opened; as long as the key is of roughly the right shape and size, it'll be able to open the door, sending the hallway awash with its contained aroma - and sometimes, even an unfurled paperclip would suffice.
In some cases, the locks are similar, too - a key that opens one door can open some others, and as the scents mix in the hallway, you pick up a mixture of smells.
That is the lock-and-key mechanism of olfaction; the keys are the molecules that make their way into your nose, which are known as ligands, and the locks are the olfactory receptors in your nasal cavity, which have a complex 3D structure that the ligand will need to be complementary to in order to bind.
Just like regular keys, the teeth are what controls whether the key will be able to turn the lock; that is analogous to the osmophore and profile, key parts (see what I did there?) of the molecule which enables it to be inserted into and bind to the receptor, triggering nervous signals that your brain interprets as a scent. Without it, the ligand would be completely unable to engage the receptor.
On the other hand, while necessary for the functioning of the key, you could get away with a locksmith making you a duplicate key with a slightly misshapen stem - the key would probably still be able to fit inside the lock, and as long as the tip can be inserted and the teeth line up with the mechanism, you'll be able to open the door - given that the stem isn't completely out of shape or of the wrong size. This is akin to the rest of the ligand besides the osmophore and profile.
(Note: While the lock-and-key analogy suggests an all-or-none response, the "stem" in odourants or the positioning of the osmophore can help the molecule achieve better binding, which in turn triggers a stronger response - so imagine a key with a poor fit allowing you to open the door by a gap and pick up wisps of the aroma, while a perfect fit would enable you to throw the door wide open, completely flooding the hallway with its scent.)
As you can see, the molecular structures of limonene, which smells like citruses, and cedrol, which is found in cedarwood, which look nothing alike, also smell nothing alike - cedrol is too large to fit into the limonene receptor, while limonene can't achieve good binding to the cedrol receptors as the teeth of the key don't match.
On the other hand, considering beta-santalol, which is one of the principal odourants of sandalwood, and sandalore, which is a synthetic molecule but also smells like sandalwood, you can see that they share a similar general shape; a bulky cyclic structure on one end of the molecule and an OH group on the other, separated by a chain of carbon atoms (they look just like keys, don't they?). This allows them to both fit into the same receptor and elicit the same response from our brain ("oh, this smells like sandalwood!").
With the knowledge of the receptor's shape, new odourant molecules can also be designed, and may even achieve better binding than the natural ligand - javanol, for example, is an extremely powerful sandalwood molecule that was discovered through intentional molecular design.
Back to the titular question: What makes a rose smell like a rose?
If you analysed the components of rose oil, you'd find that it has myriad compounds that contribute to its scent - citronellol, geraniol, phenylethyl alcohol, beta-damascone... You get the idea. None of them are singularly responsible for the complex, rich scent of roses, but instead contribute to the various olfactory facets that rose possesses - lemony, floral, balsamic, et cetera. It is important to note that smells are usually highly complex mixtures of hundreds or thousands of molecules, and oftentimes, no single molecule can realistically replicate a natural scent.
If you attempted to recreate the scent of a rose by taking the most abundant molecules like geraniol and phenethyl alcohol, you'd end up with a rather flat caricature; somewhat counterintuitively, relative abundance isn't correlated to the impact that they have on the overall scent! That is why even though citronellol makes up 38% of rose oil, it's only responsible for 4.3% of the olfactory impact, whereas beta-damascenone is a trace component (0.14%) but is responsible for 70% of the olfactory impact (Source: Scent and Chemistry, by Ohloff, Pickenhagen, and Kraft)!
This caused a lot of headaches for chemists a century ago, as they only had the techniques to identify the more abundant compounds back then, and were thus unable to convincingly recreate the scent of roses as a cheaper alternative to rose oil!
While analytical techniques have advanced immensely since then, many questions remain - even a commonly used perfumery material like vetiver oil only recently had its odour principle, which is present in trace amounts, discovered by Kraft et al. Nevertheless, new discoveries are being made every day, and I'm excited to see what the future brings!
I hope you enjoyed this first part of a series on olfaction. Please consider subscribing below to stay tuned for upcoming posts!
Brilliant article! The "lock and key" comparison was really helpful in understanding this concept. I also didn't realize that certain molecules can make up a large percentage of an oil while only contributing slightly to the olfactory impact.