A (Brief) Love Letter to the Chemistry of Molecules
The microscopic world is fascinatingly geometric. Look at viruses under a microscope (and you would need a particularly powerful one to do so, the tiny buggers that viruses are), and you’d see the various polyhedral shapes they take on. The smaller the virus, the more geometric they seem to be, giving them an unnatural shape to the untrained eye.
Most famously, bacteriophages (a class of viruses that infect and replicate inside bacteria and archaea) look shockingly like an alien robot. It’s a fun bit of science communication to build models or show illustrations of bacteriophages, in a kind of “Hey look! Aren’t viruses cool?” manner that was a lot more effective prior to the SARS-CoV-2 pandemic.
Regardless, bacteriophages seem to elicit a jarring amount of disbelief and frankly even distress. After all, how could nature create something with such perfectly shaped polygons? They look like they could be built out of Legos! Go further down into the world of proteins and molecular motors, and you’ll find yourself faced with a whole lot of disbelievers. “That kinesin thing can’t be a protein, that’s gotta be a microorganism! There’s life there,” they exclaim, while pointing at a 3D animation of kinesin carrying its payload of cellular cargo along a microtubule. “See, it walks!”
The answer is, maybe, frustratingly nuanced: they are in fact all parts of life, but no, these motor proteins or viruses don’t have the metabolic or reproductive capabilities of cells. They’re far simpler, actually. From their shape to how they move, what they attach to and how they perform their functions—everything is chemistry. And chemistry, in turn, is at the mercy of electric forces at the atomic level. So, let’s start with bonding.
If you’re familiar at all with the periodic table of elements, you’re likely already aware that each element has its own electrochemical properties. These properties include the electron orbitals surrounding an atom of your chosen electron, that is to say, the position surrounding the nucleus of an atom where its electrons are most likely to be found. Orbital theory typically visualizes this as “clouds,” and each element has a different type/shape and number of orbitals surrounding their nuclei that determines the stability they can donate to bonds.
This bonding typically only occurs in the valence (or outer) shell of the atom. For the most part, chemistry doesn’t care about the particle physics behind why electrons like to be paired, but that too many electrons closely together are energetically unfavorable, leading to VSEPR theory (valence shell electron-pair repulsion), which determines the bond angles based on electron-pair electrostatic repulsion. Meaning that due to the negative charge of electrons, they repel each other, and as a result, bonds form at specific but common angles.
Linear (180º), trigonal/triangular (120º), and tetrahedral (109.5º) are the most commonly encountered angles in molecules relevant to organic chemistry due to the valence shell of carbon. Other elements can contribute more valence shell electrons, leading to bond angles like trigonal bipyramidal (120º, 90º), octahedral (90º), pentagonal octahedral (72º, 90º), and square antiprismatic (70.5º, 99.6º, 109.5º), which result in more geometric shapes.
As we go up in size, some larger molecules are made up of smaller ones, sometimes uniformly like polymers or using an array of a class of molecules like proteins and amino acids. When it comes to proteins, which are generally very big molecules, we have to consider the way the building blocks of proteins join together.
Amino acids connect to each other via peptide bonds, forming the peptide chain, and have three important features: an amino group, a carboxylic acid group, and their side chains. It’s in these side chains that we see varying levels of polarity, which makes some amino acids more hydrophobic than others. Much like oil, hydrophobic particles repel water rather than mixing with them, a property linked to the polarity of a molecule. This hydrophobicity along with the actual, physical shape and size of these side chains determines where the amino acid will go in the structure of a protein.
Hydrophobic side chains will be more likely found within the protein structure such that it’s shielded from water while bulky side chains might be more likely found at turns, a secondary structure element that does exactly what it sounds like: reversing the direction of the polypeptide chain. The size of these side chains can also affect the flexibility of the peptide chain and looking along the peptide bond between two amino acids, we can view the structural planes between them and assign them ϕ (phi), ψ (psi), and ω (omega) angles, which have important roles in protein folding. For those who are curious, the [ϕ, ψ] plot of an amino acid is called a Ramachandran plot, which visualizes the common regions the angles of an amino acid takes in a protein.
Ultimately, we can think of a peptide chain like a charm bracelet, where each link has a charm attached to it. The bracelet itself can be folded, twisting as far as the links and charms will allow. It isn’t a perfect analogy, but it gives us a decent image of the primary structure of a protein. From there, well, we’d need a very long charm bracelet to continue the analogy.
Secondary structures of proteins consist of things like α-helices, β-sheets, turns, and less common helices that vary in size via the number of amino acids (called “residues” here) per turn. The α-helices have approximately 3.6 residues per turn; 310 helices have 3 residues per turn; and π helices have 4.4 residues per turn. Other helices include Polyproline I, Polyproline II, Polyproline III, and the collagen helix; though, these are specialized helices. Secondary structures are found in parts of an overall protein and proteins often comprise of multiple of these structures. Most proteins are quite large and thus multiple domains of different secondary structures are quite common. Structural motifs describe the connections between these structures, such as a beta hairpin, Greek key, helix-loop-helix, and helix-turn-helix, among others.
The tertiary and quaternary structures are then larger, with the final, folded form of a protein frequently being its tertiary structure. This is when the helices and sheets of the secondary structure have folded into its most energetically favorable position. Which is to say, the parts that don’t want to be exposed to water are all happily nestled inside the protein itself. We may see channels forming here, with amino acid residues sticking out to guide stray ions or small molecules.
They may also perform important catalytic tasks, making these sites “active sites” where the protein can bind other proteins, bind to ligands, or catalyze reactions. With proteins that have a quaternary structure, we can take these tertiary structures and think of them as modular blocks of a larger structure. They’ll come together like the parts of a machine, executing more complex functions than proteins that only have a tertiary structure.
Ultimately, the shape of these proteins is determined by their amino acid sequence, the sequence we get from our DNA (or sometimes simply RNA in the case of proteins). From there, it’s a matter of energetics: is it thermodynamically favorable for this molecule to be here? Molecules aren’t thinking about the process, but chemistry guides their folding from the thermodynamics of hydrophobicity to reducing the energy requirements in protein folding via other proteins as catalysts. And the shape is important, it’s what allows proteins to interact with other proteins and nonprotein molecules.
The modular nature of larger proteins also means that we see a type of repetition, from one or two axes of symmetry (Cyclic [one axis of rotational symmetry] and dihedral [one axis of rotational symmetry, a perpendicular axis with two-fold symmetry]) to cubic symmetry groups—and here is where the icosahedral structures of protein shells emerge, as the three-fold symmetry of cubic symmetry groups form hollow shells suitable for storage and transportation, including the genetic material and simple proteins found in viruses.
Biochemists have spent a lot of time solving the overall structure of a protein, predicting their final shapes based off their sequences alone. It isn’t as easy of a task as one might expect, given that so many variables are known, and that these secondary and tertiary structures are already well-understood. Still, it’s an ongoing area of research, and scientists are utilizing machine learning to identify which amino acid sequence matches to their function in the protein.
There’s a degree of redundancy in the genetic code. A quick glance at a codon chart shows just how many amino acids can be coded by multiple combinations of nucleic acids while important, “instructional” amino acids, such as Methionine (a codon that “starts” a gene sequence), is coded by a single codon. This allows for mutations to occur without losing important functions of proteins, yet we also know that we can make changes to the amino acid sequence in order to induce a change in function.
This is the basis of protein engineering, where we typically design proteins by manipulating the regions of reactivity, from ligand-binding to enzymatic active sites, and even in antibody engineering to improve their effectiveness. These tweaks might lower the activation energy of a reaction or strengthen the binding of a ligand, all by swapping out the amino acids at these active sites because it’s in these fundamental bonds and the shapes they arrange themselves in that determine what they do and how well they do it. It’s a fascinating area of study, used in things like cancer drug delivery and improving yield in industrial chemical synthesis. It’s also a fairly new field, spawned out of research into solving protein structures and the thermodynamics of bonding. Structure informs function, and structure is informed by entropy: order is simply more stable than chaos, those neatly geometric viruses are simply because proteins just fit better that way. We didn’t need to have a hand in their creation, they were probably the simplest forms of early “life” that thermodynamics slotted together and traveled alongside us as we, in turn, evolved.
Ashley Deng is a Canadian-born Chinese-Jamaican author of dark fantasy and horror. She holds a BSc in biochemistry, specializing her studies toward making accessible the often-cryptic world of science and medicine. When not writing, she is a hobbyist medical/scientific illustrator and spends her spare time overthinking society and culture. Her work has appeared at Nightmare Magazine, Fireside Magazine, Augur Magazine, and others. Her climate horror novella, DEHISCENT, is available July 2023 from Tenebrous Press.