Week 1 HW: Principles and Practices

Homework #1 - Idea: Gene silencing pesticides

First, describe a biological engineering application or tool you want to develop and why.

My idea is RNAi-based pesticides, also known as gene silencing pesticides. These pesticides are short strands of dsRNA sequences that are converted to siRNA (small interfering RNAs) by an enzyme called Dicer, then integrated into a RNA-induced silencing complex (RISC) as a template strand, which binds to equivalent mRNA strands and cleaves it, blocking the expression of important genes inside pests, therefore killing them. Because RNAi-based pesticides are made out of RNA and not toxic chemicals, they are environmentally non-toxic and naturally degrade over time. RNAi-based pesticides can also be targeted towards certain organisms, only killing harmful pests while leaving other beneficial organisms unaffected.

Next, describe one or more governance/policy goals related to ensuring that this application or tool contributes to an “ethical” future, like ensuring non-malfeasance (preventing harm). Break big goals down into two or more specific sub-goals.

Policy goal: Ensure that the introduction of RNAi-based pesticides protects the ecosystem and avoids causing drastic changes to it. Specific subgoals:

• Ensure that the reduction in population of the intended target of the RNAi-based pesticide will not cause unexpected or drastic changes in the ecosystem. For example, if the pest is an important food source for a specific species of frogs, we would want to ensure that its removal will not harm the survival viability of these frogs.
• Ensure that RNAi-based pesticides are sufficiently targeted to only affect pest species and not accidentally kill other types of insects, such as pollinators.

Next, describe at least three different potential governance “actions” by considering the four aspects below (Purpose, Design, Assumptions, Risks of Failure & “Success”).

1. A regulatory framework stipulating testing requirements for RNAi pesticides before widespread usage

The purpose would be to codify regulations on what would need to be proven, what kinds of standards an RNAi pesticide would need to meet in order to be allowed to be sold commercially. Currently, there are not many specific regulations in this area and it is unclear whether RNAi pesticides would be covered under other frameworks such as GMO organisms. A committee including scientific experts and government regulators could establish a series of standards that an RNAi pesticide would need to meet and the required tests needed to prove it. Regulatory authorities, such as maybe the FDA, would need to review and approve RNAi-based pesticides against this framework. Some of the risks around new technologies are unknown unknowns, or risks that we only realize after their creation. Regulations enacted before widespread usage of RNAi pesticides might not anticipate effects that only show up rarely and are noticed after they are deployed widely.

2. A monitoring framework to detect any negative effects

The purpose would be to help quickly catch unanticipated negative environmental impacts related to RNAi pesticides and stop the continued usage of such pesticides if they arise. Since RNAi pesticides are not yet in widespread usage, this might be needed to catch side effects that didn’t show up in initial field studies. A committee including scientific experts and government regulators could establish reporting standards for farms that use RNAi pesticides which mandate certain tests to be done regularly, with the results reported to a government agency. Regulatory authorities, such as the FDA, would need a system for receiving such records and need to periodically review RNAi pesticides on the market. Any harmful effects caused by such pesticides might already have taken place by the time that this is detected. This does not prevent harm from happening in the first place, it merely prevents it from spreading further. The costs of monitoring could also be prohibitive in time or money and prevent small farmers from using RNAi pesticides even if they are later proven to be more effective or safer.

3. Providing funding and resources to test RNAi pesticides

The purpose would be to better predict the impact of RNAi-based pesticides in actual usage by testing them on a larger scale, in a more realistic environment, while preventing any potential harm from spreading widely. More field testing might catch side effects that show up rarely, before these pesticides are sold commercially. The government could fund the construction of a specific greenhouse or farm at a public land-grant university whose purpose is to test RNAi pesticides against conventional methods of repelling pests, and fund more research into this area. The government could partner with companies to conduct studies investigating e.g. potential off-target effects of RNAi pesticides. The government funding RNAi pesticide safety experiments may cause moral hazards by allowing companies to spend less money on ensuring the safety of their products and push the costs onto taxpayers instead. Government scientific funding is also scarce, and money used to test RNAi-based pesticides could have been used to fund other research instead.

Next, score (from 1-3 with, 1 as the best, or n/a) each of your governance actions against your rubric of policy goals.

Last, drawing upon this scoring, describe which governance option, or combination of options, you would prioritize, and why. Outline any trade-offs you considered as well as assumptions and uncertainties. 

I would probably prioritize option 1 because it poses relatively little cost compared to its upside. For example, option 1 would help clarify the legal difficulties around RNAi-based pesticides, allowing startups and companies greater certainty in creating successful products and therefore making it easier for them to attract future investors, helping to bring these products into commercialization while increasing safety. Option 1 is especially urgent as businesses right now are close to developing viable commercial prototypes for RNAi-based pesticides and clarity in their testing stage will help inform their future research. Meanwhile, monitoring regulations can be delayed and drafted with better information after initial data is collected from experiments. Funding RNAi-based research in universities would help the field but poses less urgency as startups are already commercializing this invention.

Reflecting on what you learned and did in class this week, outline any ethical concerns that arose, especially any that were new to you. Then propose any governance actions you think might be appropriate to address those issues.

Part of me was a bit confused about how to deal with unknown concerns that we can’t fully anticipate. On the one hand, they pose legitimate concerns. On the other hand, it feels a bit unfair to force companies to rule out every single concern before being able to launch a product. Delaying the development of RNAi pesticides could also be considered unethical if doing so increases the usage of conventional pesticides, which have very negative environmental impacts of their own. RNAi-based pesticides, for all their potential flaws, sidestep around most of the environmental issues caused by typical pesticides and are likely to be more environmentally friendly compared to them.

I think some of this could be resolved by adding a middle stage between a ban and a widespread rollout, where a biological product is tested on a wider area, but is still under close scrutiny for any negative impacts, so that such issues can be found and fixed promptly, while allowing companies to proceed quickly. Committees could also weigh the harms of delaying a new technology alongside the harms of the new technology itself when crafting regulation

Professor Jacobson

Nature’s machinery for copying DNA is called polymerase. What is the error rate of polymerase? How does this compare to the length of the human genome.

The error rate is one per 10^6 bases. One set of human chromosomes have 3.1 billion base pairs, two sets have 6.2 billion base pairs, so you’d expect ~6,200 errors in the replicate of a diploid genome, naïvely.

How does biology deal with that discrepancy?

For mismatched bases, e.g. those which are not AT or CG pairs, a protein called MutS detects them, binds to the site, then recruits another protein called MutL to identify which strand is correct, which is usually selected based on which one is methylated. The unmethylated strand’s base is cut by exonuclease and is refilled with the correct base by DNA polymerase and DNA ligase. This usually brings down errors by around 100-fold or more.

How many different ways are there to code (DNA nucleotide code) for an average human protein?

It depends on the length of the human protein, but is extremely large for most normal proteins. Amino acids can usually be coded by multiple different codons. We have an average of ~3 codons for every amino acid after excluding the stop codons, so for example, if we have a 100 amino acid long protein, there are 3^100 ways to represent it, which has 48 digits in it. So, quite a lot.

In practice what are some of the reasons that all of these different codes don’t work to code for the protein of interest?

Bonds can form between different nucleotides. The strongest is the G-C bond which has 3 hydrogen bonds. The A-T/A-U bond has 2 hydrogen bonds, and in RNA, a G-U bond could form. When DNA is transcribed into mRNA to be translated later, this is important because if a protein sequence uses many G-C bonds, the DNA will bind very tightly to itself and slow down or even stop transcription. Also, when transcribed, G-C rich mRNA can fold to form very strong bonds between the nucleotides like a G-C bond, which would bind the mRNA into a specific shape, such as a hairpin loop, which is difficult to read for ribosomes and is hard to unfold.

Dr. LeProust

What’s the most commonly used method for oligo synthesis currently?

It’s the phosphoramidite method by Caruthers, where sequences are produced by adding a nucleotide one by one. Each addition usually has an around 99% success rate.

Why is it difficult to make oligos longer than 200nt via direct synthesis?

Although each nucleotide has a ~99% success rate, when multiplied over a 200 nucleotide sequence where every nucleotide needs to be correct, you get 0.99^200 = 13.4% success rate for the whole thing to be successful. This reduces yields dramatically as you start making longer sequences, as you have more attempts to mess up and getting every single nucleotide right becomes statistically implausible.

Why can’t you make a 2000bp gene via direct oligo synthesis?

To get a >10% chance of it to work, you’d need an error rate of ~0.1% compared to the existing 1%, which is a 10-fold reduction. Other than that, oligo synthesis involves a step where acid is used to remove the DMT-protecting group from the last nucleotide so you can add a new one. A 2000bp gene would also require you to add acid 2,000 times to a DNA sequence which is very likely to change the structure of at least one nucleotide and therefore break the sequence.

George Church

What are the 10 essential amino acids in all animals and how does this affect your view of the “Lysine Contingency”?

Wikipedia says 9. This made me sufficiently confused so I asked Gemini what the 10th one could be and it suggested arginine which is listed as conditionally essential. I also went into the Wikipedia talk page to see whether there was any debate regarding the 9 or 10 essential amino acids but didn’t find any but I did notice a discussion regarding a failed redirect from “PVTTIMHALL” to the essential amino acid page, where the author calls it a “stupid mistake” because A is apparently not an essential amino acid. Maybe HTGAA has something to say to that editor.

Anyway, the other 9 essential amino acids (as in, those which cannot be produced themselves by animals and usually require it to be obtained through diet) are, according to Wikipedia:

• Histidine (H)
• Isoleucine (I)
• Leucine (L)
• Lysine (K)
• Methionine (M)
• Phenylalanine (F)
• Threonine (T)
• Selenocysteine (U)
• Tryptophan (W)
• Valine (V)

The Lysine Contingency, in the context of Jurassic Park, apparently means that dinosaurs need lysine to survive and can’t make it themselves. First of all, this is already true of humans and most other animals, according to Wikipedia. If the Lysine Contingency was a thing, then we would all need to eat lysine supplements or die. Fortunately, there are other sources, such as our diet. If the dinosaurs are really lacking lysine, they could just eat other animals and tourists to get their needed fix. Or eat plants, but I doubt they are as appetizing.