h a l f b a k e r y
Still more entertaining than cricket.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
This Idea involves something we don't yet know, but are
going to know in the not-distant future.
We start with the topic of stem-cell research. A stem cell has
ability to produce multiple types of other cells. When a sperm
fertilizes an egg, the resulting zygote is a "totipotent"
meaning it can generate any other type of cell in the body. It is
an important fact that any cell descended from a stem cell has
ALL the DNA of the stem cell.
An ordinary specialized cell, like a muscle cell, normally only
pays attention to some of the overall DNA code in the cell.
is of course the code that tells it how to be a muscle cell. What
stem-cell researchers are trying to do is locate the DNA code
can tell an ordinary specialized cell to behave like a stem cell,
and stimulate the cell to actually start processing the stem-cell
We know that various aspects of the goal can be done; an
virus, for example, can tell a cell to stop paying attention to its
normal DNA code, and start running the code in the virus (which
causes many copies of the virus to get made, until the cell dies
from exhaustion). And when the DNA of a normal specialized
is extracted, and used to replace the DNA in an ovum, somehow
that event causes the stem-cell/zygote code in that DNA to get
processed, such that a clone begins to exist.
Imagine a specially crafted virus. It infects an ordinary cell,
tells the cell to stop processing its normal DNA code, and to
processing stem-cell code. The goal of stem-cell researchers
could in theory be that simple.
But that is only the beginning of this Idea. Imagine identifying
the blocks of DNA code normally associated with all the
types of cell in the body. Want to be more muscular? Get a
specially crafted virus to tell fat cells to start processing
cell DNA code! And yes, a virus can be that specialized, such
it can infect only one type of cell. (This implies someone must
specially craft one virus for each cell, but perhaps we could
the virus first tell the cell to make just a couple copies of the
before switching DNA gears from fat-storage DNA to muscle
Now consider the testicles. They contain specialized cells that
manufacture sperm, and possibly other specialized cells that
produce hormones like testosterone. I'm not completely sure
about all the specializations; maybe one type of cell does both
(and more). Anyway, what we want here is a custom virus that
can tell the appropriate testicular cells to stop running --
perhaps by skipping-- the DNA
them to make sperm. Think of it as doing the equivalent of
inserting a JUMP
into a block of computer code. If only sperm-production is
affected, then the man will be perfectly healthy otherwise, and
yet is now basically sterile, with respect to having offspring (or
will be after the backlog of previously-created sperm has left
It doesn't have to be permanent, though. Another custom virus
could be made, which removes that JUMP instruction.
images of mechanical on off valves [popbottle, Feb 01 2016]
||Customizable sterility sounds like a convenience. And a
weapon. Perhaps even a weapon of convenience?
I'll bun that [+]
||Not sure I want a virus running around my balls that tells
them to become sterile, and another one that is trying to
shut it off. Who knows what damage they'll do during the
O.K. Corral shootout.
||" This Idea involves something we don't yet know "
||I think that takes it into the category of "magic".
||Vernon has a unique lack of regard for the perils of
producing a virus that could cause human extinction.
||[WcW], nonsense! Some humans are always immune to a
given virus. But, yes, the population might go down a lot if
this Idea was mis-used.
||A few strains of Vernon's magical elixir "accidentally" lost in
ISIS-controlled countries might slow the production of future
terrorists and help develop a following for fried food, sugar
and a lethargic lifestyle. Better than a bomb.
||I think it would be pretty trivial to use a lenti virus to
CRISPR out fertility, targets and specific promoters are
pretty easy with such a specialized tissue. You should be
able to transiently put it back, with a scrotal injection of
adenoviral rescue DNA. I can't be bothered with the details
right now, but viral delivery of a drug-inducable system to
switch fertility on and off should be pretty doable.
||Did anyone else notice the phrase "scrotal injection"
||//Some humans are always immune to a given virus.//
||So far, no readily transmissible and sterilising (or highly
lethal) virus has been human-to-human infectious enough to
spread to any highly connected human population and
become pandemic. That's not actually the same thing, and
there have been some diseases which have been close.
||[Loris], even the AIDS virus has run up against a few
humans who were immune to it. I do need to be clear that
it is different humans who are immune to different viruses;
not one group immune to all. We could be wiped out by
enough different-and-fatal viruses.
||I'm a bit sceptical about the computer code analogy. Computer code is generally executed in a linear way. DNA is more ... 3D ... isn't it?
||The immune system is not some sort of infinitely diverse magical system which (within the species) has enough diversity to fight off any possible attack. While I don't know of evidence of any species driven to extinction by a virus, I'd be surprised if none had.
||//I'm a bit sceptical about the computer code analogy. Computer code is generally executed in a linear way. DNA is more ... 3D ... isn't it?//
||DNA is a 3D molecule, but then, so are the computer chips where computer code functions.
DNA is however an essentially linear store of information (some DNA molecules are circular; this doesn't affect the basic point). A linear series of the four standard bases encodes the information required to produce RNA, most of which encodes for various proteins (some is functional directly as RNA). The DNA holds both the expression information (whether to produce a protein) and the amino acid sequence of that protein. DNA molecules can be very long, and encode many proteins. The structure of a protein molecule depends on its sequence - different amino acids flex and attract each other in different ways. So the one-dimensional DNA sequence encodes the 3D shape of proteins.
||However, one big difference between computer code and DNA is that biological processes are very probabalistic (from the perspective of the 'DNA program'). If a particular event has to happen reliably then quite a lot of work has to be done to ensure that.
||[Loris], one reason some species are considered
"endangered" is because their gene pools have shrunk so
low that they are indeed susceptible to extinction by
disease. Humanity, however, is not currently suffering
from that danger.
||What I am saying is that this:
||//Some humans are always immune to a given virus.//
||For what it's worth, humans are high in numbers, sure, but are not particularly genetically diverse.
||Actually, while I'm here, flaunting my ignorance, there's something else that I may have misunderstood about DNA-editing.
||I'm picturing a virus spreading, from cell to cell, through some tissue. Let's say it's the "off" Vernovirus, in someone's testicle. Presumably its spread would be limited, in a somewhat unpredictable way, by the person's immune system. If cell-division is still happening in that part of the body, then infected cells will make new infected cells, while uninfected cells will make new uninfected cells (with the old "factory-settings" DNA).
||What mechanism causes the infected cells to win out over the uninfected ones? And how do you know when they have? And, once they have, how do you clean up the infection? (That is, once the virus has delivered its payload to all relevant cells, I assume you wouldn't want a permanent viral infection using up immune-system resources).
||In this case, you'd probably want a very high level of confidence about whether you were fertile or not - but my curiosity is more about the general practicalities of CRISPR-ing stuff in multi-cellular organisms such as myself.
||So, the editing is a separate technology from the
delivery. The delivery is what you'd have to get right
here. The most commonly used virus technologies, are
adenovirus, adeno-associated virus and lentivirus. The're
all engineered to be replication deficient, that is a viral
protein necessary for building more viruses is missing.
That means that the virus now enters the cell, delivers
the DNA then stops short of making millions of new
viruses and breaking open the cell like a wild-type
adenovirus would do. Instead the replication has to take
place in special lab grown cells which have the missing
protein. So instead of a disease, you now have a
breathtakingly efficient DNA delivery system. Adenovirus
is pretty non-specific and will infect almost any cell it
runs into, infecting a whole body would be tricky, but
individual tissues with a well understood blood supply
(like balls) should be pretty easy. Then there's the
question of specificity. Because adenovirus will infect
practically anything, how do you target your effect? The
answer to that is in the DNA you put int he virus, you
need something called a "tissue-specific promoter".
Choosing these is sometimes easy, sometimes hard, but
you essentially want to find a gene that is only expressed
in your target tissue. If you're lucky, on the front end of
this will be the promoter, a region of DNA recognised and
read by the cell. For example, insulin is only made by the
pancreatic beta cells. The DNA for insulin is in every cell,
but on the front of that DNA is a promoter only
recognized by beta cells. So if you wanted to selectively
build proteins in beta cells, you'd copy the insulin
promoter and the beta cells will make whatever you put
behind it. So your virus can infect many types of cells, but
largely nothing happens because non-beta cells will just
ignore the insulin promoter.
||So, if you find yourself a good, sperm specific promoter,
you can put the CRISPR system in behind it, targeted to
break a gene necessary for viable sperm. Then, package
it all up in an adenovirus. Adenovirus efficiency is >95%,
and CRISPR is around the same, once its in the cell, so
you might miss 10% of cells. But you can just re-dose a
few times, remember the virus doesn't replicate, and the
DNA doesn't even get read in most cell types.
||I mentioned adeno-associated viruses, these are a group
of viruses which have some specificity, they are physically
able to infect some cells, but not others. For example,
AAV8 preferentially infects the liver, this is great,
because you can inject the virus into a vein, and the
viruses won't just jump into the first cell they meet. They
hang around until the encounter a liver cell. The
specificity is useful if you can't get privileged access to
the tissue you are interested in. Not really an issue with
||The last virus type is lentivirus. This is a retro virus. It
infects, integrates itself into the host genome and from
there becomes just like a regular gene, only you get the
choice of what that gene is. After infection, the host cell
will just crank out the proteins you told it to make
forever. Theoretically, this is great for putting a working
copy of a gene into cells to offset the effect of a broken
one. The downside is that we don't get to choose where
the DNA gets integrated. It' a roll of the dice whether the
new DNA inserts itself in a gene critical for preventing
cancer. It's pretty unlikely, but with billions of cells and
billions of viruses, you're rolling the dice a lot, and you
only have to get unlucky once.
||Excellent explanation. Thank you very much, [bs0u0155].