Other: Blankets Huge Chemistry Nerds Knit

I haven't done any knitting in many years, but at one time, it was kind of my thing, instead of messing with plants. And for several years I was also very very bored, so I had a lot of time to do it, too. At the same time, I also had a chemistry degree, and a particular fondness for biochemistry, so it was only natural, in an odd, not-inevitable-at-all kind of way, that I would find a way to combine the two.

So I did a series of what I call "protein blankets." First, to explain protein: proteins are what do most of the actual work in your body: antibodies are protein, hair is protein, your metabolism is by and large executed by proteins. Your DNA contains, among other things, the instructions for making these proteins, and it works like this:

DNA is more or less a line of text. Different bases (adenine, cytosine, guanine, thymine) are attached to a backbone of sugars and phosphates, and these bases are read three at a time, and these base triplets determine how the protein gets put together. Proteins are made of twenty different amino acids, which can link together, also in a line, and their specific properties are determined by which parts of that chain are attracted to one another, or to water, or are repelled by water, and so forth. Each set of three bases specifies a particular amino acid for that slot in the protein. It gets complicated really fast, but the upshot is that just as DNA is a line of bases which read like CCT GCA ACG TCT CCC, the protein that is formed when these bases are read is also a line, of amino acids. (The particular example above would translate as proline-alanine-threonine-serine-proline.¹)

So what I did for the blankets was, I assigned a yarn color to each of the twenty amino acids,² and then knitted one row for each amino acid in the sequence until I was done, and then I went on to the next blanket. I don't know exactly how long it took to do this one, though the smaller ones, once I got a rhythm down, were taking me about a week each. (This is a week spent doing nothing but knitting and watching TV, granted.)

The blankets that result when this procedure is followed generally look like I was just grabbing random balls of yarn with no particular rhyme or reason, and are kind of ugly unless one knows why the colors follow the sequence they do. And even then, they're still ugly. It's just that the ugly means something.

They're also hard to photograph, but here is one:


That is the major prion precursor UJHU blanket. Major prion precursor UJHU³ is associated with, and found in high concentrations in, certain neurodegenerative diseases like Creutzfeldt-Jakob disease, fatal familial insomnia, kuru, and so forth. The animal equivalent would be the agents responsible for mad cow disease (bovine spongiform encephalopathy), though I don't believe it's precisely the same protein in both cases. I should probably look into that.

We don't know exactly what purpose this protein normally serves in the body, though the fact that everybody makes it, and yet we don't all have degenerative neurological diseases, suggests that it must do something. (Wikipedia reports that it might have something to do with forming long-term memories: we know Wikipedia isn't the most reliable source on these things, but these kinds of diseases do all involve the brain, so memory is at least plausible, and the article does contain a reference.) It's also not clear how the disease is actually triggered. Prions are thought to work when a protein normally found in the body is induced to adopt a different, unusual shape, and then this misfolded protein is able to induce other copies of the same protein to also adopt this altered shape. Over time, this results in small clumps, called plaques,⁴ of these misfolded proteins, which as they get larger interfere with normal function of the brain.

Most (all?) cases of prion-related disease appear to be the result of mutations in the DNA which encodes this protein, which in turn causes the amino acid sequence to change. In some cases, the eight amino-acid repeat of PHGGGWGQ on the left side of the blanket (It's mostly white, with a little pink and blue. It starts just after the left side and takes up most of the left third of the blanket.) may be missing a repeat, or may get extra repeats, either of which could change its shape and function. (Repetition of a chunk of DNA is known to happen, though ordinarily I think they're smaller chunks than that.) More commonly, certain kinds of damage to DNA might change a single base in the DNA, causing the body to use a different amino acid in the finished protein. In blanket terms, this would barely be noticeable: we're talking about a white line changing to a light green one, or a dark green turning into a different dark green. Although these changes are not huge, they can affect how the protein folds itself, making the bad, plaque-forming shape more likely and leading to disease over time.

The whole blanket thing is mostly just interesting because most proteins interpreted this way actually look like what they are: random assemblages of amino acids that happen to be good at doing something. If you don't know the background, the blankets just look like I was grabbing random balls of yarn out of a bin and doing a line or two of them and then reaching in for a different ball, and the blankets are ugly. I mean, the blankets are still ugly if you do know the background, but it's very much not the case that I was grabbing random balls of yarn. It's all very, very specific and precise: it only looks sloppy.

The picture above of the actual blanket is kind of crap, since I haven't figured out a good way to photograph them yet. I think the camera needs to be like ten feet away to capture the whole thing, but I can only get the camera eight feet off the floor before running into ceiling. So here's a kind of idealized version: same colors, same sequence, but in a neat, tidy computer file instead of as a blanket. Just in case anybody's really dying to try to read the sequence of colors for themselves.


Major prion precursor UHJU isn't my best protein-related work, I should note. Everybody's favorite is salivary glue protein sgs-3 from Drosophila yakuba, which I only have as a blanket, not as a computerized picture like the one above. Perhaps someday I'll figure out how to take better pictures. Zein, a nitrogen-storage protein from corn seeds, is also nice. If there is any interest, I may occasionally post more blanket pictures, assuming I can get better photos than this. And also I'll probably have to look up which proteins they go with: at one time I could tell them apart pretty easily, but it's been a while. I've also got an RNA blanket (potato tuber spindle viroid), a neon emission spectrum blanket, and other science-related stuff I'm failing to remember right now. 'Cause I'm a big nerd.

ADDITIONAL TEXT FOR BIOCHEMISTS:

The N-terminal end of the protein is on the left. This is a precursor protein: about 20 amino acids on the C-terminal side are cleaved off after transcription. I do not know whether there are any disulfide bonds between the cysteine residues, though I'd be surprised if there weren't.

For the right amount of money, I could be convinced to take requests. The best proteins to do are between 200 and 400 amino acids long. Although I said above that I used to be able to knock out a smallish one in about a week, I would probably require considerably longer than that to do one now, and I'm not sure how we could set it up so that neither of us could back out of the agreement once it was made. But if you have a particular pet protein that you'd really like a blanket of, well, I do still have a lot of yarn, and the husband would like to see it go away, so . . . my e-mail address is in the left sidebar.

ADDITIONAL TEXT FOR KNITTERS:

I maybe should have done stockinette stitch for this, as it would have left me a smoother, more finished-looking surface in the end. Unfortunately, when I started these, I could knit but not purl, and even if I could have purled, the blanket would have curled up on me, which is not a particularly desirable quality for a blanket.

Of course, the stretchiness of the all-knit blankets leaves something to be desired, too. I'm really a very unsophisticated knitter (or, you know, was, since I haven't done any knitting in about three years). The blankets work more on a, you know, conceptual level.

If I had a whole lifetime for knitting and didn't have to take care of plants and stuff, I'd do some of the spider silk proteins: a lot of them would wind up very pastel (white, light green, aqua, lavender), and they'd maybe be nice for a baby blanket or something. Though it'd be weird to notice that you were wrapping your baby in spider silk, even if only on a really meta level.

-

¹ For the sake of convenience, faster typing, etc., each of the amino acids has been assigned a single-letter abbreviation. Our example, in the single-letter code, would be "PATSP". (What a crazy random happenstance!)
² Alanine (A) -- light green
Cysteine (C) -- yellow
Aspartic Acid (D) -- red
Glutamic Acid (E) -- dark red
Phenylalanine (F) -- black
Glycine (G) -- white
Histidine (H) -- medium blue
Isoleucine (I) -- dark gray-green
Lysine (K) -- dark blue-green
Leucine (L) -- dark green
Methionine (M) -- orange
Asparagine (N) -- light pink
Proline (P) -- aqua
Glutamine (Q) -- medium pink
Arginine (R) -- royal blue
Serine (S) -- lavender
Threonine (T) -- royal purple
Valine (V) -- kelly green
Tryptophan (W) -- brown
Tyrosine (Y) -- dark purple
Readers with some biochemistry knowledge will recognize that for the most part, I've coded small amino acids with light colors and heavy amino acids with darker ones. Also hydrophobic amino acids are varying shades of green, basic amino acids are shades of blue, and acidic amino acids or their close relatives are shades of red and pink. Purples indicate neutral hydrophilic groups, and yellow and orange involve sulfur.
There are some things I'd change about this setup now. Like, methionine really should have been more of an acid green / chartreuse color, cysteine should have been a pastel yellow, lysine should have been more of a straight blue, and asparagine and glutamine maybe should have been purplish (maybe fuchsia/magenta). But I was also having to make do with what was available at the Wal-Mart in town, too, and they had a limited range of colors.
³ I think the "UJHU" part is an internal identifier code for this particular protein at the specific site where I got the sequence, and not, properly speaking, part of the protein's actual name. I include it anyway because, in theory, this will make it easier to look up. Incidentally: uniprot.org is incredibly slow and balky looking anything up for me, so don't take it personally if you have to try multiple times to load the page. It took me, no lie, like three hours to find and load the pages. I don't know what their problem is.
⁴ Totally different from the plaque one gets on one's teeth, though.