Re: [DIYbio] Curious about zink finger nuclesases / gene therapy

The size issue you raise is pretty significant, I'll grant you. When
there was any chance of me working with interesting things like
ZFNs/TALEs, it was with the possibility of using Adeno-Associated Virus,
and you can easily fit ZFNs into an AAV vector, but probably not a TALE.

So, when delivery size is a constraint, ZFNs still have a significant
advantage. A zinc finger binding 3/4 nucleotides can be as little as
20-30 amino acids long, barely longer than the loose consensus of a
"peptide". TALEs seem, on cursory inspection, to be significantly larger.

But, when size isn't an issue (cost of synthesis falling, vector insert
size not a constraint), I think the effort and wo/man-hours saved by not
"optimising" makes TALEs far more attractive.

But then, I'm lazy. ;)

On 27/09/12 10:23, jhwangc wrote:
> Although a lot of the field is moving to working on TALEN's zinc fingers
> haven't yet been completely replaced.
>
> Although zinc finger proteins do have context-dependent effects like you
> mentioned (you can't string together any two zinc fingers into an array)
> there are currently two different assembly methods for zinc finger arrays
> which work around this (OPEN and CoDA) by building arrays using zinc
> fingers that are known to work well in combination together or optimizing
> the zinc finger array at every step of construction. These processes are
> pretty long and cumbersome to just stringing together TALE's (TALENs are
> the the TALE protein + a nuclease domain) though which don't have any known
> context dependent effects.
>
> BUT the size of the TALE protein arrays is a concern for a lot of
> researchers because for a zinc finger array, each individual zinc finger
> binds 3 base pairs of DNA but for a TALE array, each TALE only binds one
> base pair. As a result, a zinc finger array only needs 6 zinc finger
> proteins to bind to an 18 base pair sequence while a TALE array would
> need18 proteins to bind to an 18 bp sequence.
>
> The field is definitely moving toward TALE's but zinc fingers will still be
> around for at least a little longer until more research on TALE's is
> done/the kinks get ironed out.
>
> On Wednesday, September 26, 2012 11:37:05 AM UTC-4, Cathal wrote:
>>
>> I studied Zinc Finger Nucleases long enough to know that they are a dead
>> end. Seriously, six-feet-under dead end. They're already being replaced
>> by TALENs because TALENs aren't as Patent-encumbered.
>>
>> Essentially, about half of the known zinc finger families are patented
>> to the hilt, and the patent troll that owns them charges in the 10,000s
>> for licenses. The remaining zinc fingers aren't sufficient for reliable
>> assembly; it turns out that, despite their glowing reputation for
>> "modular DNA targeting proteins", they are actually very hard to get
>> working correctly. You can add a zinc finger for "TAG" to a zinc finger
>> for "GAC" and have it totally fail to bind to TAGGAC, and instead bind
>> to TTAGWGAC, or something like that.
>>
>> I haven't studied TALENs enough to know for sure yet, but I gather they
>> are somewhat more predictable. Being less patented is ALWAYS a boon, and
>> it seems that it's paying off.
>>
>> As to how they work; when using them in-vitro, you can use them as
>> proteins are normally used; protein + buffer + DNA etc.
>> When used for gene therapy, it's a totally different situation; you
>> don't deliver the protein, you deliver a DNA agent that encodes the
>> protein. So, you design a plasmid containing your ZFN/TALEN, plus the
>> DNA you want to replace the chromosomal target with (it must have
>> significant homology to the chromosomal target to encourage crossover
>> after enzyme-cleavage), and you deliver that through electroporation, or
>> chemical treatment, or viruses.
>>
>> The TALEN/ZFN gets transcribed/translated and, if you've got the
>> appropriate nuclear targeting peptides attached, gets sent back into the
>> nucleus, where it cuts the target site. Then, homology-directed DNA
>> repair (homologous recombination) leads to the target site getting
>> replaced with your alternative sequence in the plasmid at some efficiency.
>>
>> If your replacement is designed not to contain the target site, this is
>> one-way and reasonably high efficiency, provided that the DNA-cleaving
>> TALEN/ZFN works as intended and gets into the nucleus. You have to
>> provide lots of plasmid to ensure there's plenty of template lying
>> around when the break occurs in the target strand.
>>
>> Efficiency per targeted cell is good, but don't expect to transform
>> whole organisms, it will not happen. Efficiencies of transduction in
>> animals is very low for naked DNA, often less than 10%. When using
>> viruses, you can get higher efficiencies, but at the cost of high
>> specificity (only certain cell types get transformed) and potential
>> immune overreactions. Methods for non-naked-DNA, non-viral DNA delivery
>> include electroporation, high pressure delivery, and liposomal delivery.
>>
>> Options for Viral delivery which aren't insanely risky to attempt on
>> humans include: Adeno-Associated-Virus (NOT Adenovirus). And that's it.
>> The others are highly risky in terms of immune response, and if you use
>> them as they are most often used, they also carry significant cancer
>> risks due to random-ish, gene-preference integration. AAV is remarkable
>> as it has such low immune stimulation effects, and it has a target site
>> in the human genome that seems to have very low risks of any cancerous
>> side-effects (provided it's unoccupied by wild AAVs).. but it has really
>> low capacity for DNA, so it's hard to deliver anything useful.
>>
>> Naked DNA has low efficiency of transformation, but at least it's A)
>> Easy and B) More controllable; integration is only likely where you
>> direct it to occur, and immune reaction to naked DNA is generally mild,
>> particularly if you produce it via PCR or DAM/DCM negative strains of
>> E.coli, so it's not methylated in ways that the immune system treats as
>> suspicious.
>>
>> On 26/09/12 08:54, Mega wrote:
>>> Hi,
>>>
>>>
>>> I read about zink finger nucleases, and that they may be used in living
>>> organisms to alter their genome (after birth, as an adult) .
>>>
>>> I was wondering how that was posssible, how do you get the enzyme in
>> every
>>> cell of the body? Is it just injected in the blood stream and then gets
>>> distributed automatically?
>>> Or do you inject it into the organ needed? But how can it penetrate the
>>> cell wall? Is it so small that it fits through the celll membranes?
>>>
>>>
>>>
>>
>> --
>> www.indiebiotech.com
>> twitter.com/onetruecathal
>> joindiaspora.com/u/cathalgarvey
>> PGP Public Key: http://bit.ly/CathalGKey
>>
>


--
www.indiebiotech.com
twitter.com/onetruecathal
joindiaspora.com/u/cathalgarvey
PGP Public Key: http://bit.ly/CathalGKey

--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To post to this group, send email to diybio@googlegroups.com.
To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.