[Hplusroadmap] Cheap bioreactor mostly for the construction of further bioreactors
Bryan Bishop
You might know about
http://biohack.sf.net/
And one of the biggest problems with a simple collection of notes and
documents is mainly that you can't really do anything with it except
read, which is fine, if you have the time, but it's just not "up to
date" and it never can be due to the intense amount of effort that it
would take to make sure you have everything covered, that a rewritten
and new version is scientifically accurate, etc.
So, simultaneously, there are other 'problems' in diy tech. A few days
ago I was going through some neurosci papers on GABAergic and ACh
circuits in the prefrontal and visual cortices:
http://heybryan.org/mediawiki/index.php/Sustained_attention
To do experimentation on biopsied organotypical neuron slices would
require the neurochemicals. And if you go look up on Sigma-Aldrich, the
neurotransmitters, agonists, antagonists and related substances are
prohibitively expensive. $1/mg. That's ridiculous. We *know* that we
can have bacteria/algal bioreactor tanks that will produce it for free.
So let's do it.
http://heybryan.org/mediawiki/index.php/Neurochem_kit
http://heybryan.org/mediawiki/index.php/Bioreactors
^ this last page is where I'm centralizing most of the notes. But please
excuse the current state of the page, it's mostly just a link to an IRC
log, and at the moment I fail to recall the other documents on the wiki
that are of relevant note.
The idea is to build a biroeactor tank out of scrap parts. There are
lots of problems to be solved like contamination, piping, purification
of substances, cloning and making new versions for friends (or
yourself), reprogramming the ecoli farms, etc. But I'm certain it can
work, and once it's able to produce the reactants for PCR and the
chemicals necessary to do oligonucleotide synthesis in the DNA
synthesizer, it's basically self-supporting except for the metals that
would be required to make a new bioreactor over all. That's pretty good
results, if it ever gets that far.
And on top of that, if you're incorporating an entire DNA synthesizer
(which is really just a giant, over-sized inkjet printer that uses
phosphodiesters if I recall correctly), you can connect it to a cheap
PIC or microcontroller and download genes from the internet, perhaps
from GenBank or Entrez or something, get new biobricks, incorporate
synthesis reaction pathways from other species (i.e., integrate
dopmaine synthesis).
There are other possible uses of a biotan. Tissue engineering, biofuels,
organ farms, the PETA in vitro meat prize, which I happen to have notes
on here:
http://heybryan.org/mediawiki/index.php/Meat_on_a_stick
It could also be used in neural tissue experimentation, longevity
experiments, but as a first step ecoli or algal farming is excellent.
I'm still outlining the overall protocols that could be used, but I'm
optimal so far. The main problem seems to be the tens of thousands of
RPMs in common protocols for using plasmids. Purification of
synthesized neurotransmitters would /normally/ be painful because I
frankly have no ideas on synthesizing filtration membranes, however an
interesting alternative is to use aptamers:
http://en.wikipedia.org/wiki/Aptamers
.. which have the added bonus of having a synthesis largely dependent on
a DNA synthesizer in the first place. So that's good news, killing (or
giving life to?) a few birds with one stone.
What's needed:
* cheap, but effective protocols (as 'biologically-dependent' as
possible) for cloning, cultures, subcloning, PCR, expression
experiments, and selection experiments (ampicillin, etc.)
* ideas on the physical setup of the bioreactor, i.e. how to do piping
with glassware, PVC pipes, or catching produced materials via simple
things like balloons tied to faucets, etc.
* the DNA synthesizer - http://bioinformatics.org/pogo/
* eventually some server space where we can store all of this
information as a git repository. I'd offer my server space, but I'm
quickly approaching zero available space, although the cheap 12 cents
per GB announced on some websites today has me scouting for the $64 500
GB steals. So that might change soon. :)
I'd like to hear some ideas.
- Bryan
________________________________________
http://heybryan.org/
cowbert
chromatography for a bunch of the purification. (Although you can sort
of build your own if you could obtain a spectrophotometer.)
On May 29, 6:25 pm, Bryan Bishop <kanz...@gmail.com> wrote:
> Hi all,
>
> You might know about
> http://biohack.sf.net/
> And one of the biggest problems with a simple collection of notes and
> documents is mainly that you can't really do anything with it except
> read, which is fine, if you have the time, but it's just not "up to
> date" and it never can be due to the intense amount of effort that it
> would take to make sure you have everything covered, that a rewritten
> and new version is scientifically accurate, etc.
>
> So, simultaneously, there are other 'problems' in diy tech. A few days
> ago I was going through some neurosci papers on GABAergic and ACh
> circuits in the prefrontal and visual cortices:
> http://heybryan.org/mediawiki/index.php/Sustained_attention
> To do experimentation on biopsied organotypical neuron slices would
> require the neurochemicals. And if you go look up on Sigma-Aldrich, the
> neurotransmitters, agonists, antagonists and related substances are
> prohibitively expensive. $1/mg. That's ridiculous. We *know* that we
> can have bacteria/algal bioreactor tanks that will produce it for free.
>
> So let's do it.
>
Bryan Bishop
> If you had access to chromatography equipment you could use affinity
> chromatography for a bunch of the purification. (Although you can
> sort of build your own if you could obtain a spectrophotometer.)
How would the spectrophotometer help?
http://heybryan.org/instrumentation/instru.html
^ though there's no explicit spectrophotometer listed there, there's a
spectrometer and a few other tools that we can splice together. The
cereal box + flash light + box setups for spectro stuff showing up on
the internet might be useful, though only for recording relative
differences in absortion spectrums (IIRC?).
http://en.wikipedia.org/wiki/Affinity_chromatography
> Usually the starting point is an undefined heterogeneous group of
> molecules in solution, such as a cell lysate, growth medium or blood
> serum. The molecule of interest will have a well known and defined
> property which can be exploited during the affinity purification
> process. The process itself can be thought of as an entrapment, with
> the target molecule becoming trapped on a solid or stationary phase
> or medium. The other molecules in solution will not become trapped as
> they do not possess this property. The solid medium can then be
> removed from the mixture, washed and the target molecule released
> from the entrapment in a process known as elution.
That's certainly an interesting method. How hard is it to chug through
the chemical catalogs to come up with the reaction that you need to
make the phase changes for the separation process?
cowbert
go from rough to refined
The typical first step is using dialysis tubing and taking advantage
of osmosis and ion exchange across the membrane. If you say maintain a
hypertonic solution outside the tubing you can extract the solvent out
of the tubing, concentrating your product. Be careful of precipitation
though! (may require partial denaturation with urea or what not).
Next, there are other prelim liquid chromatogrpahy methods as well- a
common one being sephadex bead size exclusion column chromatography,
which separates mixtures based on molecular mass. Sepharose beads
compose the solid phase of a column. Small particles take longer to
traverse through the beads of a given diameter because they have more
pathfinding opportunites than larger particles. So larger particles
elute faster from the column than smaller ones.
We use the spectrophotomer in column chromatography to detect the
product being eluted from the column. It will also serve to tell us
the concentration in each fractoin we are collecting from the column.
As an example, tyrosine-rich proteins show marked abosrbance at 280nm.
As to affinity chromatography, a common approach is to use antibody
affinity column LC. What you do is again, you take sephadex but this
time you cross link the FC portion of mABs which show affinity your
product of interest to the beads. On the first pass, your product will
bind to the mABs attached to the beads. You then change the conditions
within the column to change the binding constant (usually pH related),
which causes your product to unbind.
Next we have ion exchange chromatography, which uses special columns
that have alterable binding constants based on things like solvent
polarity or salt concetration. While these are very effective in
purification, they are probably unfeasible for DIY due to materials
problems (these columsn are constructed from exotic alloys or
zeolites and so on).
On May 29, 8:54 pm, Bryan Bishop <kanz...@gmail.com> wrote:
> On Thursday 29 May 2008, cowbert wrote:
>
> > If you had access to chromatography equipment you could use affinity
> > chromatography for a bunch of the purification. (Although you can
> > sort of build your own if you could obtain a spectrophotometer.)
>
cowbert
overview of the currently used technologies for protein purification.
Oligonucleotides can benefit from antiDNA ABs, or if you manage to
cross link some ssDNA or somehthing to the solid phase you might be
able to capture a pattern as well. In the worst case, treat
nucleotides as an organic compound you need to capture.
On May 29, 8:54 pm, Bryan Bishop <kanz...@gmail.com> wrote:
> On Thursday 29 May 2008, cowbert wrote:
>
> > If you had access to chromatography equipment you could use affinity
> > chromatography for a bunch of the purification. (Although you can
> > sort of build your own if you could obtain a spectrophotometer.)
>
Bryan Bishop
> On Thursday 29 May 2008, cowbert wrote:
> > If you had access to chromatography equipment you could use
> > affinity chromatography for a bunch of the purification. (Although
> > you can
Cowbert, I'll have to get back to your recent responses soon. But I
wanted to elaborate on the aptamer method of purification.
http://en.wikipedia.org/wiki/Aptamers
The idea is to use these oligonucleotide sequences in a way that allows
binding, maybe some hybridization, and this is binding between RNA and
the target molecule. This method could be used kind of like the
antibody separation processes, but has the advantage of being cheaply
manufactured. Any thoughts?
- Bryan (who really, really will get to the other emails)
________________________________________
http://heybryan.org/
Julie E Norville
proteins, one simple way to check and see if the protein is there is to
tap the
tube and see if bubbles appear. If bubbles readily appear that indicates the
presence of a protein at high concentration.
This is what they used to do to find potentially good fractions in labs that
didn't want to buy a spectrophotometer. You will not know from this whether
the protein is pure.
Polyprep columns from Biorad are pretty cheap.
Affinity resins are not very reasonably priced this point of time.
Frog
Been watching from the sidlines for a while, but now something I can
contribute. I was at LTI (now know as the mud sucking dogs
Invitrogen) for 15 years, making Taq and DNA polymerases and RT's and
restriction enzymes, proteases, kinases, yada, yada, yada. Also did
quite a bit of Oligo synthesis
In order for your "aptamers" concept to work, our protein of interest
would need to have some affinity for the oligonucleotide/peptide
"backbone". The main technical hurdle here would be the length of the
aptamer molecule. It would need to be pretty damn specific to the
protein of interest. Typically, this kind of chromatography would be
used as a "polishing" step and really would be characterized as
"affinity chromatography". I believe it would likely be simpler, less
technically challenging and less expensive to make the antibodies.
Ion exchange is much simpler, first step. This would separate protein
on the basis of their net charge (either positive or negative). It
can be done in aggregate batch mode or in a simple fritted glass or
plastic column.
I am not sure why you think you'll need a "bioreactor" per se. Most
of the bugs you'd need to grow could be done in a mini-prep format:
1-2 Liters in any clean glass jar. The proverbial "mayonnaise jar"
approach.
It will be important to have the purest protein prep achievable, as
impurities will compound and amplify mistakes in any synthesis
reactions.
Then there is the necessity of a thermostable DNA polymerase for PCR
and sources of pure nucleotides for the synthesis.
So, I have taken a slightly different approach.We have been successful
scooping up gear (like fermentors, spectrophotometers, stir plates,
heat blocks, microscopes, hoods etc.) and chemicals (including 25L of
Sephadex resin and 25L of Protein G agarose, open market value @ ~
$1,000 per liter) for nearly free. The catch is you need to find
Biotechs that are going under (which mean you'll need a target rich
environment) and have a means to haul the stuff out of their
facility. It's simple economics. Most of these places have already
expensed or capitalized their losses and are not looking to recuperate
any revenue from their "junk".
Bryan Bishop
> Bryan,
Sorry for my delay in response. I was munching over your suggestions.
> Been watching from the sidlines for a while, but now something I can
> contribute. I was at LTI (now know as the mud sucking dogs
> Invitrogen) for 15 years, making Taq and DNA polymerases and RT's and
> restriction enzymes, proteases, kinases, yada, yada, yada. Also did
> quite a bit of Oligo synthesis
Neat. I think we'd all love to hear from you more frequently. I don't
quite have an as-impressive resume built up quite yet.
> In order for your "aptamers" concept to work, our protein of interest
> would need to have some affinity for the oligonucleotide/peptide
> "backbone". The main technical hurdle here would be the length of
> the aptamer molecule. It would need to be pretty damn specific to
> the protein of interest. Typically, this kind of chromatography
> would be used as a "polishing" step and really would be characterized
> as "affinity chromatography". I believe it would likely be simpler,
> less technically challenging and less expensive to make the
> antibodies.
So I talked this over with a few buddies re: the aptamer specifity. It
turns out that there's another conceptual problem to work off of too.
If you're going to be making aptamers, you're probably going to be
doing it from a pool of random RNA sequences. And then from there you
can run your selection experiments once they are in aptamer form, and
then clean everything else off, denature, transcription run, some gels,
PCR, all of that fun stuff. Then you replicate the same sequences and
get to make many many more aptamers. But if there's not a match in a
random DNA pool of strands of length-N bases or whatever, it's going to
take a DNA synthesizer in the process, which means length issues (both
in the ability to build the molecule, but also its simulation and
analytical constraints on the possibility space of the N units). Could
be fun, if we already had all of those setups. How much easier are the
antibody protocols?
> Ion exchange is much simpler, first step. This would separate
> protein on the basis of their net charge (either positive or
> negative). It can be done in aggregate batch mode or in a simple
> fritted glass or plastic column.
>
> I am not sure why you think you'll need a "bioreactor" per se. Most
> of the bugs you'd need to grow could be done in a mini-prep format:
> 1-2 Liters in any clean glass jar. The proverbial "mayonnaise jar"
> approach.
Yeah, I'm all for the mayo jar approach.
> It will be important to have the purest protein prep achievable, as
> impurities will compound and amplify mistakes in any synthesis
> reactions.
On that note, how hard would it be to come up with sets of reactants and
either the antibodies, aptamers, or chromatography setups to repeat the
process on demand? The idea is to eventually have these same cells, or
at least a subset of them in some strain/batch, produce the chemicals
needed to do the transcription kits, etc.
> Then there is the necessity of a thermostable DNA polymerase for PCR
> and sources of pure nucleotides for the synthesis.
The pure nucloetide problem sounds like a hard one, so what about the
oligo synthesis reaction routes for synthesizing individual
nucleotides, or the possibilities of centrifuging cells and collecting
nucleotides? Maybe even denaturing lots and lots of protein from
acquirable sources, like grass or meat?
As for thermostability in DNA polymerase. A good number of protocols
that I have been looking at indicate that there's a few steps at 90 deg
Celsius and then back down to stuff well below room temperature. I
don't know if this is practical for do-it-yourself setups. There's
certainly the possibility of using heat lamps coupled to a digital
thermometer plus sensors in a box, but I don't know if this can be used
accurately enough. Ideally we could just find something that works in
something 'close to' room temperature. Is this what you mean by a
thermostable DNA polymerase?
> So, I have taken a slightly different approach.We have been
> successful scooping up gear (like fermentors, spectrophotometers,
> stir plates, heat blocks, microscopes, hoods etc.) and chemicals
> (including 25L of Sephadex resin and 25L of Protein G agarose, open
> market value @ ~ $1,000 per liter) for nearly free. The catch is you
How? Networking? Dumpster diving? Foul play?
> need to find Biotechs that are going under (which mean you'll need a
> target rich environment) and have a means to haul the stuff out of
> their facility. It's simple economics. Most of these places have
> already expensed or capitalized their losses and are not looking to
> recuperate any revenue from their "junk".
Hm, so targeting the rich environment might really involve relocation
from a 'non' biotech area. Kinda of an important step in the whole
process. For those on the list up in Boston, I envy you. ;-)
Frog
Bryan,
Let me see if I can answer your questions succinctly:
"So I talked this over with a few buddies re: the aptamer specifity.
It
turns out that there's another conceptual problem to work off of too.
If you're going to be making aptamers, you're probably going to be
doing it from a pool of random RNA sequences. And then from there you
can run your selection experiments once they are in aptamer form, and
then clean everything else off, denature, transcription run, some
gels,
PCR, all of that fun stuff. Then you replicate the same sequences and
get to make many many more aptamers. But if there's not a match in a
random DNA pool of strands of length-N bases or whatever, it's going
to
take a DNA synthesizer in the process, which means length issues (both
in the ability to build the molecule, but also its simulation and
analytical constraints on the possibility space of the N units). Could
be fun, if we already had all of those setups. How much easier are the
antibody protocols? "
"On that note, how hard would it be to come up with sets of reactants
and
either the antibodies, aptamers, or chromatography setups to repeat
the
process on demand? The idea is to eventually have these same cells, or
at least a subset of them in some strain/batch, produce the chemicals
needed to do the transcription kits, etc. "
OK, so now I am taking a deep breath.........
Incredibly difficult, which explains why people invest billions of
dollars and years of training & basic research to create elaborate
systems for doing all this stuff. If you just take a single element
of this proposed system, say the "cells". Assuming we're talking
about fermentation of recombinant bacteria (since this would likely be
the route of least resistance), you need to control every element of
the fermentation process in order to get consistent results: the
"feed stocks" (water, sugars, vitamins, & other medium components),
dissolved gasses (through the degree of agitation, using compressed
gasses etc), temperature, time and storage of the bugs. Then there is
the pesky problem of mutation rates, so you'll have to continuously
monitor those little bugger to make sure they haven't mutated/evolved
and start producing some undesirable nasty that will foul everything
up. So even with the fanciest fermenter known to man, it is not
"easy" to consistently produce a cell paste that will yield the same
amount and quality of a DNA polymerase and the contaminants you'll
need to remove from the prep.
So this is why "synthetic biology" is so appealing, because it takes a
lot of the crap out of the equation. Unfortunately, we still don't
know enough about the complex systems that create these things to be
able to create them de novo, so we need to rely on the complex
mechanisms of in vivo systems to produce the desired molecule and then
cull what we want out of the resultant mix.
"The pure nucloetide problem sounds like a hard one, so what about the
oligo synthesis reaction routes for synthesizing individual
nucleotides, or the possibilities of centrifuging cells and collecting
nucleotides? Maybe even denaturing lots and lots of protein from
acquirable sources, like grass or meat?"
You can isolate nucleotides from nucleic acids or you can synthesize
nucleic acids from scratch using a series of organic reactions. But I
am definitely not an organic chemist and have no idea where to start
with the organic synthesis route. From an in vivo source first you
would need to isolate a bunch of DNA (from any plant or animal
source), degrade the DNA into nucleotides using a nuclease. Mung bean
nuclease <http://www.epibio.com/pdftechlit/006pl092.pdf> would work,
but you'd have to make single stranded DNA, which you can do by
boiling it or heating it. You can buy pounds of mung beans at the
grocery store from not a lot of money.
"As for thermostability in DNA polymerase. A good number of protocols
that I have been looking at indicate that there's a few steps at 90
deg
Celsius and then back down to stuff well below room temperature. I
don't know if this is practical for do-it-yourself setups. There's
certainly the possibility of using heat lamps coupled to a digital
thermometer plus sensors in a box, but I don't know if this can be
used
accurately enough. Ideally we could just find something that works in
something 'close to' room temperature. Is this what you mean by a
thermostable DNA polymerase? "
So the advantage of Taq polymerase is that it can survive the
"denaturation" heating step. "Thermostable" means it remains stable
at elevated temperatures. You could use a different DNA polymerase,
like DNA Pol-1 or the Klenow fragmnent of DNA Pol-1 <http://
en.wikipedia.org/wiki/DNA_polymerase>. Taq was originally isolated
from deep ocean vent bug Thermus aquaticus, but is usually produced
from a cloned e coli now, since it's an easier bug to grow.
The reaction temperature doesn't drop below RT ever (maybe 60 C, not
60 F), because at lower temperature the DNA will reanneal and just gum
everything up. But the "thermal cycling" element can be reproduced
using water baths pretty easily, one would just have to move it
manually back and forth to the desired temperature for the desired
amount of time.
"Hm, so targeting the rich environment might really involve relocation
from a 'non' biotech area. Kinda of an important step in the whole
process. For those on the list up in Boston, I envy you. ;-)"
And, by the way, they don't have much government funded research in
Boston. As a matter of fact, Maryland has the lions share of
government research institutions and much more vibrant biotech than
Boston. I am not a Boston fan, and not just that they kicked my Cavs
out of the playoffs and knocked out the Tribe last year.
I will have my sweet revenge when the Cleveland teams emerge to
dominate the NFL , MLB & NBA when Boston's brief respite at the top of
the heap comes to a close..