Author Archives: downhousesoftware

Dick Hallorann lends his grandmother’s words to describe people like himself and Danny Torrance. She called it “Shining.” The Overlook Hotel is one of those places that shines too.

Here are a few random thoughts, sounds, and images from the many lives of The Shining.

Room 217 of the Stanley Hotel

Before there was an Overlook, there was the Stanley Hotel. Stephen King once stayed in the Presidential Suite of the Stanley, where he was visited by bad dreams, hallucinations, and was inspired to sketch out an outline for his best novel.

But The Shining is not just King’s best novel. It also served as the basis for the film of the same name, as written and directed by Stanley Kubrick. Like the hotel, this Stanley would also haunt King as he took his novel and stretched and twisted it into one of the most influential films ever made. So different from the book, King has said that it is Kubrick’s film, not his.

A reviewer calls the film, “a brilliant, ambitious attempt to shoot a horror film without the Gothic trappings of shadows and cobwebs so often associated with the genre.”

One of the most striking things, aside from masterful performances by Jack Nicholson and Shelley Duvall (heresy!? Yes, I think her performance was nearly as perfect as Nicholson’s – just drawn more from Kubrick’s haranguing rather than her innate talent.) is the spellbinding score and incidental music. From the first intonations of Dies Irae to the frenetic energy of the final chase scene, the music shapes your emotions and pulls you into the haunted world of the Overlook.

Rolling Stone, on the music accompaniment to Stanley Kubrick’s The Shining…

“Kubrick and music editor Gordon Stainforth cherry-picked maximally queasy passages from works by a clutch of Eastern European mavericks: “Lontano” by György Ligeti, the Hungarian genius whose music gave 2001: A Space Odyssey its unearthly atmosphere, and even more importantly Krzysztof Penderecki, the Polish radical whose strangulated strings, barbaric brass yawp, clatter-bone rhythms and hissing choruses in “Utrejna,” “De Natura Sonoris,” “The Awakening of Jacob” and more provided the Overlook Hotel and its denizens with an appropriately unhinged environment.”

To address the differences between the novel and the film, King sought to make a new version of the film that tracked more with the original storyline. In 1997, this version was released as a miniseries staring Steven Weber and Rebecca De Morney as the Torrances and using the Stanley Hotel as the Overlook. This version eschewed the hedge maze that Kubrick had created in place of the technically difficult hedge lions and, most notably stayed faithful to the novel’s end with Jack dying in the fiery explosion sparked by the Overlook’s overpressurized boiler.

Three decades after the original novel, King released Doctor Sleep, following the troubled life of Danny in the years after his ordeal at the Overlook. Fans of the Shining particularly enjoyed the last act of Doctor Sleep which brought us back to the remains of the Overlook, a place that exists only in the world of the original Shining film.

Radiolab did a good piece on Challenge trials for the COVID-19 vaccine which you can find here.

Challenge trials are an expedient, but fraught way to generate data on a vaccine’s efficacy quickly. Briefly, they involve taking a (relatively) small group of people, say 100, and dividing them into two groups of 50 each. One group gets the vaccine and the other group gets the placebo. Then, after waiting some period of time for the immune system to generate a protective response, you expose all 100 people to live virus.

This is essentially the same thing as what is done in the more traditional trial, with the exception of directly exposing subjects.

In the end, it’s a simple matter to calculate the efficacy (or not) of the vaccine by comparing the number of people who got sick in each group. If these numbers are equal, the vaccine is ineffective, etc. Even if all 100 people got sick from this challenge, it would still be less than the number of people who got sick (170) during the Pfizer trial that involved 43,000+ total participants (reported here).

Further, a challenge trial can be completed in as little as two months, while more traditional trials are ongoing over two years.

In the Radiolab piece, trial participants were asked why they participated. Answers ranged from appeals to mathematics (one participant said exposure to COVID did not significantly increase his risk of dying in any given year) to a need to make their experience meaningful (if they caught covid in the trial, at least it would provide important evidence toward approving a drug, whereas if they caught it any other time, it would just be a waste.)

However, the one reason I did not hear was one of financial inducement. Although I could not find actual data on this, challenge trials may pay up to $4000 for the 3-6 weeks of legal quarantine. There are a number of issues associated with these payments – the most obvious being that it may induce those without means to participate because they feel the money if too good to pass up. Jennifer Blumenthal-Barby and Peter Ubel write that, in fact, the payment for these trials may not be enough as it fails to meet the proposed minimum living wage of $15/hr.

I recommend that you listen to the Radiolab piece (it’s less than 30 minutes long) and take a look at the Blumenthal-Barby and Ubel, which is an interesting read.

Often simply called a gel shift assay, these are performed to identify the binding of proteins (or other molecules, conceivably) to DNA. The basic idea is that you have a section of DNA in excess supply that contains a sequence that you suspect may be bound by known or unknown proteins. You first label your DNA by incorporation of labelled bases via PCR or end-labeling. This used to be done using radioactive bases, but there are other alternatives available now. For our purposes, I’m going to assume a radioactive label as it is simple to imagine.

Once you have your radiolabelled DNA, you now need to add some proteins that you suspect might bind. This could be isolated protein from a recombinant source or it could be a cell lysate from cells that have been treated in some way. Again, I’ll imagine that we have stimulated cells with something for a progressively longer time (say, 0 minutes stimulation – 30 minutes of stimulation at 5 minute intervals). Finally, after the get runs, you can blot it onto a membrane and expose it to film where the radioactive DNA will light up. Assuming this, I guess we also have a radioactive DNA ladder too

Here’s an example… (remember, this is entirely fictitious)

We stimulate cultured B cells given the survival cytokine, BLyS. At the timepoints indicated above (0-30 minutes), we harvest cells and lyse them in the cold to obtain nuclear fractions. Our DNA is derived from the promoter region of a protein we are interested in and we are wondering if we will see transcription initiation factors assemble.

What we are seeing is a 100bp DNA ladder run with our samples. Keep in mind that the ONLY thing we can see here is labeled DNA. In the first several lanes, none of our DNA is being bound and it is running according to its size (~50bp). At 15 minutes, we start to see the shadow of a band that has shifted our DNA up to an apparent ~700bp. IMPORTANT: DNA ladders are not protein ladders! Also, this is a native protein with unknown charge that is bound to our DNA. All we know is that we are getting binding, we can’t assume knowledge of the exact size of the protein (however, in general, we do see shifts going up as more proteins accumulate)

By 30 minutes, we see an appreciable amount of protein binding our DNA. But how can we know what protein(s) are binding?

One way, if we know what we are expecting, is to use an antibody against that suspected protein. Assuming we have extra lysate + DNA mixture to run a second gel, we could spike our antibody into one tube with DNA and lysate and not into the other (or spike an irrelevant antibody).

If we see that the antibody results in a ‘supershift’ where the DNA/protein band has moved up to a larger apparent size, this means that our antibody is binding its target and that target is binding the labeled DNA.

Let’s assume we suspect that Nf-kB is what is binding our DNA, so we use and Nf-kB antibody to do a supershift assay.

These data support our model and we can now go on to ask new questions.

Francis Crick is well known, not only for his work in elucidating the structure of DNA along with James Watson, but also for his hypothesizing about the nature of biology that he thought would be discovered in the wake of the publication of the structure.

Among these hypotheses was the notion of a ‘central dogma’ that described the flow of information from DNA, through RNA, to protein. It was clear how information could flow from DNA to RNA as these were written in the same language of nucleic acids. What was less clear was how this information could then be translated into a protein.

 In 1955 Crick proposed the idea of adaptor molecules that would be required to read the information on the nucleic acid and translate this into the language of proteins, comprised of Amino Acids (AA). “In its simplest form there would be 20 different kinds of adaptor molecules, one for each amino acid”

We now understand these adaptor molecules to be transfer RNAs (tRNAs), which bind both messenger RNA (mRNA) and AAs in a way that the AAs can engage in protein synthesis. These tRNAs bind the AAs on one end of the molecule and have three bases, called anticodons, that interact with the codons of the mRNA. This alone would be enough to prove Crick as prescient.

However, that only covers the first part of his statement. He goes on to say, “…and twenty different enzymes to join the amino acid to their adaptor.”

Twenty enzymes to build the adaptor molecules (tRNA with AA).

Yet again, Crick’s mental model was correct, we have a number of aminoacyl tRNA synthetase (aaRS) enzymes that specifically bind the tRNAs and AAs individually and then promote their conjugation. One might expect these synthetase enzymes to be, themselves, at least partially comprised of nucleic acids that can identify the tRNAs by their anticodons and then charge them with their appropriate, cognate, AAs. However, this is not always the case. Each tRNA is, instead, recognized by an identity element that is often discrete from the anticodon, but nevertheless serves to distinguish the correct tRNA that should be bound. These identity elements (aka identity determinants) are often located along the anticodon stem or the acceptor arm of the tRNAs, however they are not restricted to such (see Fig 2 from Giegé et al.)

Figure 2 from Giege et al: Cloverleaf folding of tRNA with location of known identity determinants and its three-dimensional L-shaped organisation.

Although it is conceivable that every individual tRNA would have a specific aaRS to charge it with its AA, “[w]ith some exceptions there is only one aaRS per isoacceptor tRNAfamily (the cognate set of tRNAs).”¹ However, I will not go into the specifics of this flexibility here.

Once a tRNA is recognized and bound by its associated aaRS, the appropriate adenylated AA is brought in to the synthetic site where it will become charged to the tRNA. Once charged, the AA will be shifted to a second binding pocket, the editing site, where it is again checked. Together, this is known as the double sieve model, where the first, coarse, sieve (occurring in the synthesis site) allows placement of correctly sized (or smaller) AAs, while the second sieve (occurring in the editing site), allows only AAs of the correct hydrophilicity, mischarged AAs are cleaved by deacetylase enzymes.²

References

1. Giegé, Richard & Eriani, Gilbert. (2014). Transfer RNA Recognition and Aminoacylation by Synthetases. 10.1002/9780470015902.a0000531.pub3.
2. Dino Moras, Proofreading in translation: Dynamics of the double-sieve model PNAS December 21, 2010 107 (51) 21949-21950; https://doi.org/10.1073/pnas.1016083107
3. Shuya Fukai et al. Structural Basis for Double-Sieve Discrimination of L-Valine from L-Isoleucine and L-Threonine by the Complex of tRNA^Val and Valyl-tRNA Synthetase. Open ArchiveDOI:https://doi.org/10.1016/S0092-8674(00)00182-3

My Molecular Biology class can print, complete, and bring in this crossword puzzle for 2pts of extra credit on the chapter 10 (Eukaryotic Transcription) quiz.

Link to puzzle

For Credit, complete, paper copies of this puzzle will be accepted at the start of the 28Oct2020 class period – ABSOLUTELY NO EXCEPTIONS

Once you finish the puzzle, there is an icon in the bottom right side that you may use to print out your work.

Quite a while ago, I wrote a blog entry here discussing the regulation of sigma factors in E. coli.

Briefly, sigma factors are the subunit of prokaryotic DNA-Dependent RNA Polymerases (RNAP) that mediates the attachment of the polymerase to the promoter region.  See the illustration below (taken from The Cell, ASM Press), where the various subunits of the RNAP are illustrated along with a DNA strand. The s subunit is required for the precise positioning of the RNAP on the promoter of the amplified gene. Without this subunit, the polymerase does not know where to bind or initiate transcription.

In that earlier post, I referenced an article by De Las Peñas et al., which predicted that a regulator of s24, RseA, was a membrane protein. This week, in another class I teach, we are looking at ways to determine if a specific protein contains membrane-spanning regions, so I thought I would use this as an example.

The first thing we need to do is to get the amino acid sequence for RseA. The easiest way to do this is to query the NCBI’s protein database.

It looks like a fair number of RseA proteins are known, but I’m selecting the fourth one down, from E. coli strain K-12.

This gives us the AA sequence:

MQKEQLSALMDGETLDSELLNELAHNPEMQKTWESYHLIRDSMRGDTPEVLHFDISSRVMAAIEEEPVRQPATLIPEAQPAPHQWQKMPFWQKVRPWAAQLTQMGVAACVSLAVIVGVQHYNGQSETSQQPETPVFNTLPMMGKASPVSLGVPSEATANNGQQQQVQEQRRRINAMLQDYELQRRLHSEQLQFEQAQTQQAAVQVPGIQTLGTQSQ

With this in hand, we can go to a hydrophobicity plotter that will scan the AA sequence and provide a moving average score (similar to those sometimes used in the stock market) of the hydrophobicity of each amino acid. The moving average includes scores from the neighboring AAs in a way that we can get a sense for the hydrophobicity of a region of a protein taken together. I prefer the Kyte-Doolittle plot which I use through the Swiss protein group’s ProtScale tool found here.

Pasting the AA sequence into the box and checking “Hphob. / Kyte & Doolittle” will return the following plot:

Note that a score of ‘0’ is midway along the y axis. ‘0’ corresponds to neither hydrophobic nor hydrophilic. As this is a hydrophobicity plot, positive scores identify hydrophobic regions. The largest hydrophobic region found on this plot occurs just after AA 100 and goes on for about 20 amino acids after that. This tells us that this bit of the protein may be hydrophobic enough to be a transmembrane region.

Although this is suggestive, it would take some experimental work to demonstrate that it is true. However, because this protein is well characterized, we can check this using another web tool known as UniProt.

On UniProt’s website, we can find an entry for this exact protein which includes a subcellular location and topology section with references to back up our conclusion.

I hope this exercise is useful in describing how these tools can be used to learn more about your favorite protein before you even step into the lab.

In bioethics, we discussed the film Inferno. See my notes from this discussion on the companion site:

https://100filmsin100days.wordpress.com/2020/10/19/inferno/

“Most of my nightmares involve me forgetting my lines in a stage play.”

We live in a time when cloning a gene is a trivial matter done by technicians in the lab using tools that have been around for decades. However, this does not mean that it can be accomplished without knowing precisely how the biology works.

It is often said, even in molecular biology texts, that the DNA of the target gene and a plasmid vector simply need to be cut with the same restriction enzyme so that their ‘sticky ends’ can come together and be ligated into a new construct. This is often diagrammed as a cut and paste procedure (below), suggesting that no further attention to detail is required.

There are two issues (the bulleted items) I have with such a simplified cartoon:

• This cartoon does not indicate any preference for the direction that the insert takes once it is incorporated into the plasmid.

Inserts have directionality. A gene has a start codon on one end and a stop codon on the other. These must be in line with the promoter sequence and polyA signals that are typically not a part of the insert, but are contributed by the plasmid.

The Promoter sequence is required for binding of the DNA-Dependent RNA-Polymerase that will make the mRNA. This must be upstream of the start codon, ATG. The Poly Signal is responsible for attaching untemplated ‘A’ residues to the 3′ end of the mRNA. It must be downstream of the stop codon, [TAA/TAG/TGA].

The easiest way to prevent this problem is to ensure directional cloning by cutting each end of the vector and insert with different enzymes that don’t possess compatible ends.

The second easiest way to control for cloning your insert in the wrong way is to have a digest with another enzyme to screen for the proper orientation after miniprepping multiple clones. In the figure below, we can see that there is one XhoI site within the insert and one within the plasmid. Because the one in the insert is offset to one side, the size of the fragments generated by this digestion will depend on the orientation of the insert. Knowing which fragment size is associated with proper orientation of the insert will help us to quickly screen our minipreps by XhoI digestion.

• This cartoon suggests that the recombinant plasmid is the only product of this ligation reaction.

The other problem with this cartoon is that it assumes that the desired ligation of one insert with one plasmid is the most abundant product, when in fact, it probably occurs in a tiny minority of cases. More likely, the plasmid curls up on itself and religates without the insert being incorporated.

The reason that the plasmid religates is because the two ends of the plasmid can religate and they’re close enough together that they probably run into one another more often than an insert does. The way to prevent this is to recall the chemistry involved in this ligation reaction.

This reaction requires a 3′ hydroxyl group and a 5′ phosphate group to come together to form the phosphodiester bond. This happens on each strand of the DNA and is mediated by a ligase enzyme.

However, only one strand being ligated is sufficient to hold the recombinant plasmid together until it can be fixed inside the cell by DNA repair mechanisms. If you remove the phosphate from the plasmid molecule, it cannot come together and be ligated to itself. Instead, only when an insert is incorporated are there the phosphates required to ligate one strand at either end.

The phosphates can be removed from the plasmid DNA using the enzyme, Calf Intestinal Phosphatase (CIP). The only thing you need to remember is that the CIP must be destroyed at the end of its reaction with the plasmid. Otherwise, it can go on to also dephosphorylate the insert DNA when that is added for a ligation reaction. Luckily, CIP is easily inactivated by heat (NEB suggests 80 C for 2 minutes).

An alternative to CIP is Shrimp Alkaline Phosphatase (SIP), also available from NEB.

Using either of these two techniques individually will result in an abundance of your preferred recombinant product.

Happy cloning!

The ultimate goal of the new and emerging SARS-CoV-2 antibody tests is to determine when and how we can safely get back to work and restart our stalled economies. While knowing how many people are sick is valuable in tracking the rate of disease spread and identifying people who need to isolate or seek treatment, it misses out on telling us who has had the virus and has recovered (or may have not been symptomatic at all). The idea is that people with antibodies may be protected from (re-)infection and are therefore safe to return to “normal” life. (caveats are that these antibodies are actually protective and will last, conferring immunity).

A good article discussing what antibodies are and how these tests operate was recently published in USA Today. The form of many of these tests is very similar to home pregnancy tests (lateral flow ELISAs), enabling quick, rapid results with no special training to perform.

Eventually, once enough people have been exposed to the virus and recovered or been vaccinated against the virus (probably early next year), then we will have herd immunity sufficient to protect the vulnerable people around us from being exposed to a life-threatening illness.

In the meantime, we’ll probably have to be content with a ‘new normal’ of living in a way that doesn’t prevent the spread of infection but limits it to a level that our hospitals can bear.