#structural biology

As a thank you in advance, I offer a picture of the cells I’ve been working with

Lies of P weaponry is fun because the weapons system is a good analogy for antibody class switching due to the modular nature of the weapons.

Antibodies have a portion called the "variable region" that varies based on the target. So an antibody against SARS-CoV2 looks different from an antibody against influenza virus due to the variable portion of the antibodies having different shapes tailor-made for each pathogen. Thus, the variable region determines how the antibody interacts with enemies.

Antibodies also have a portion called the "constant region" that only comes in a few shapes. This region serves as the interface with different cell types or other molecules, i.e. it determines how the antibody interacts with its user.

For example, IgM ("immunoglobin M") antibodies' constant regions allow them to be chained together to form a pentamer. (Yes, it's effectively a 5-part antibody Megazord.) This allows them to bind a ton of their target antigen, effectively minesweeping it out of the organism's system. It also allows them to collect lots of the invader to show to cells that can then act on that information.

Meanwhile, the IgG ("immunoglobulin G") antibody's constant region allows it to bind directly to a bunch of hungry immune cell surfaces and trigger the cells to eat ("phagocytose") the antibody-antigen pair, removing it from the organism's system.

You can probably see the metaphor forming already: the variable region is much like the blade and the constant region is much like the handle.

The blade of the weapon allows you to apply specific status effects (including none) and determines the base power of the weapon based on the blade's upgrade level. Thus, the blade determines how that weapon interfaces with enemies, just like the variable region of an antibody does.

In contrast, handles in Lies of P determine the weapon's main power scaling stat (i.e. how its total power gets boosted based on the user's stats) as well as the weapon's movement type (sweeping, stabbing, etc.). Thus, the handle of the weapon determines how the user interacts with the weapon, just like an antibody's constant region does.

But wait, there's more!

Something neat that antibody-producing B cells do is called "class switching". Maybe a B cells start off by producing antibodies of the IgM variety. But what if I want to keep my ability to bind a specific target, but I want an IgG instead? That's where class switching comes in! The B cell can do this by changing only the constant region, leaving the variable region intact. This puts a different "handle" on the antibody without changing the "blade". This changes which users (i.e. cells) the antibody can interact with but without changing which invaders it can bind to.

Glad to see there are others like me out there.

What is your favourite science-y thing?

Hey,

That's such a nice message. Thank you.

My favourite sciency thing?

I am a structural biologist, so proteins I guess. Or any kind of molecular machines.

Building a Sperm

Combining protein structure predictions with cryoelectron tomography reveals proteins – novel and known – their structures, interactions and locations in intact mammalian sperm

Read the published research paper here

Image from work by Zhen Chen and colleagues

University of California San Francisco, San Francisco, CA, USA

Image originally published with a Creative Commons Attribution 4.0 International (CC BY 4.0)

Published in Cell, October 2023

Our sense of smell is dependent on recognition of a vast array of odorants despite having a finite number of receptors. Odorants are mostly detected by G protein-coupled receptors called odorant receptors in olfactory sensory neurons. While there are only approximately 400 odorant receptors in humans, combinatorial activation of these odorant receptors enables sensing of odorants with diverse chemical structures. However, the mechanistic basis of odorant binding to odorant receptors in humans has remained unclear.

SBGrid member Aashish Manglik and other researchers have been working to develop a structural understanding of how odorants are recognized by odorant receptors. Using cryo-EM, they report the structure of active human odorant receptor OR51E2 bound to fatty acid propionate.

Above: Structure of human OR51E2 bound to propionate in complex with miniGs399. CC BY SBGrid.

Based on this structure, they determined that propionate is bound within a pocket in the odorant receptor and makes specific contacts to achieve activation. After mutating the odorant-binding pocket, they observed altered recognition of fatty acids with varied chain length. This suggests that activation of odorant receptors by odorants is influenced by tight packing interactions. Through molecular dynamics simulations, they show that propionate induces conformational changes in a specific region of odorant receptor OR51E2, extracellular loop 3. This work provides a foundational understanding of odorant recognition by human odorant receptors at a high-resolution, structural level.

So I’ve been away from Tumblr for a while, but I’m back! 

While I’ve been away I’ve started a new job as a Science Technician on a Crystallography platform. I thought i’d give a brief overview of what this technique is, and why it’s useful for scientific research. There will be more detailed posts on crystallography in the future!

To begin with we need a protein crystal. So we mix pure protein solution with different solutions which we hope will encourage the protein to come out of solution in an organised crystal form. To get a protein crystal, you may need to try 100′s of conditions to find the right one, and some proteins will refuse to crystallise for a number of reasons. 

We then pick these tiny crystals up (the droplet in the image above is less than 1mm accross) into loops and send them to a synchrotron, where powerful x-rays are fired at the crystal. Some of these x-rays bounce off of the electrons in the atoms of the protein, and are recorded as a pattern of spots on a detector. 

Each of the diffracted spots contains information about the position of every atom in the protein. The signal from one protein molecule would be almost impossible to get data from, but with millions of molecules of protein in each crystal, and capturing thousands of images like this with lots of spots, we can get a strong signal, allowing us to decode and visualise the space taken up by each atom. This space is shown by the blue mesh below, and is called a map:

We can then use specialised programs, and manual model building, to model the amino acid sequence of the protein into the blue map, giving a representation of the protein structure in the crystal. 

Knowing the shape of a protein can provide huge insights into it’s biological function, and potential for targeting with drugs.