(11-26-2018 09:05 PM)Antarius Wrote: https://www.statnews.com/2018/11/26/rice...y-project/
Curious on what people think, especially those who understand the science.
I figured I'd weigh in because I do understand the science and have dabbled in the technique myself, although not to create genetically-modified babies!!!
So an overview is that CRISPR is generally thought of as a major component of the bacterial immune system. Since bacteria are susceptible to being attacked by viruses (which generally work by inserting their own DNA into the bacteria), the basic idea is that CRISPR is a way to recognize viral (foreign) DNA and to render it ineffective, typically by introducing a mutation into the viral DNA.
So how does it introduce the mutation? Generally, the genetic code is made up of a sequence of four letters - A,T,G,C. Each cell 'reads' that genetic code in order to produce proteins, which may be anything that is necessary for the cell's function - metabolism, growth, signaling to other cells, etc. The key is that each cell 'reads' the genetic code three letters at a time, in non-overlapping function. For example, a sequence such as: ATGCTGACCAAG will be read precisely in that order, without overlapping (ie, ATG CTG ACC AAG), and that generally produces a functional protein. CRISPR works generally by either deleting a random number of letters in that code, or inserting a random number of letters into that code. CRISPR therefore requires that we are able to identify this target sequence within a soup consisting of your whole genetic code. CRISPR is able to do this based on a template called a 'guide RNA (gRNA)'. Suffice to say gRNA is produced by bacteria, but not by human cells. I won't go into further details on this unless you're really interested, because it's not useful for this discusion.
So given this background, what does this mean for this to work in humans? There are two key components here:
1. We need to deliver a 'Cas9' protein. This is the basis for the CRISPR system, and is present in bacteria but not naturally in humans.
2. We need to deliver the gRNA template that as I said is produced by bacteria but not by human cells.
Keep in mind again, we have to know which genes we want to modify. Because this defines component #2 above.
There is a major problem with gene editing in humans, and that is that most traits - height, metabolism rates, susceptibility to cardiovascular diseases or neuropsychiatric disorders - all of these come from an interplay between your biology (your DNA) and the environment. This is where most literature you may read on the biology behind our physical traits is 'multifactorial' and 'polygenic' in biologic jargon. For example, genes Height1, Height2, Weight1, and Weight3 may all contribute to your height, so if you'd like your baby to be tall one day, you may have to modify all four genes. However, modifying Weight1 and Weight3 may contribute to unforeseen effects, such as increasing the possibility of your baby to be overweight one day. Furthermore, editing Height1 and Height2 may not be guaranteed to increase height if the environment is not correct (such as not having an appropriate diet).
Final note on CRISPR: keep in mind this is all predicated upon the idea that the Cas9/gRNA system is highly specific, which it is definitely NOT. Cas9, even with the correct gRNA, has the tendency to be error-prone, leading to introduction of mutations where you did not intend for mutations to be introduced. This is particularly problematic in biology because there is a general tendency for patterns within our genetic code - basically proteins that structurally are very similar to one another, thereby being part of a 'family' of related proteins. However, a difference in maye 10-30% of their sequence was enough to produce a different function.
Hopefully based on this information we can directly tackle the study in question here:
https://www.bbc.com/news/health-46342195
The Chinese group that modified the baby claims to have made a baby more resistant to HIV by modifying a protein called 'CCR5', which is commonly thought to be the human receptor recognized by HIV, leading to viral insertion into CCR5-expressing cells. On my note above about CRISPR being error-prone, the group claimed that they did not see off-target mutations so we may (for the purposes of this discussion, although I am skeptical!) assume that they are right that they only modified CCR5 and nothing else. The problem of this approach is that CCR5 is also necessary for a very important function of the immune system - which is the recruitment of some cells of your immune system into sites of infection or inflammation. In essence, the group may have made babies more resistant to HIV, but impaired the babies' ability to fight off a common cold or flu. Therefore the plan to interfere with CCR5 was shortsighted and has an extremely high chance of impacting the livelihood of two babies.
Anyway - we are still far from the brave new world of 'designer babies'. Chinese labs generally have lower ethical barriers to performing experiments than labs in most other developed countries, so unfortunately, they will likely keep on doing things like this to add to their already amazing human rights track record.