TLR Biosensor
137 views
Skip to first unread message
Simon Rose
Jan 31, 2015, 2:52:54 PM1/31/15
to diy...@googlegroups.com
Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6 The idea is to ligate the pathogen-sensing, extracellular domains of each TLR (or possibly just the recognition module, consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise. Using a prokaryotic chassis to carry the TLRs might not be possible owing to the absence of endoplasmic reticula in prokaryotes, and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs in the cell wall. Any TLRs expressed might not fold properly. I think that TLR signal transduction is not well understood hence the use of GFPs as reporters, as bacteria would lack the requisite intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used. Anyway, it's a learning curve!
Simon Rose
Jan 31, 2015, 3:07:29 PM1/31/15
to diy...@googlegroups.com
ps, I won't be doing this in a biohacklab but in a university lab under supervision. I've just seen the post about recombinant factor 8 by the way. Sterile techniques, if required should be no problem
Simon Rose
Jan 31, 2015, 3:10:11 PM1/31/15
to diy...@googlegroups.com
Feedback and suggestions would be welcome
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Sunil Phani
Jan 31, 2015, 11:49:20 PM1/31/15
to diy...@googlegroups.com
hey Simon Rose ,
"synthetic pathogen sensor using human toll " such a cool idea ...
have you seen this Multiplex Automated Genome Engineering (MAGE) project http://wyss.harvard.edu/viewpage/330/
every thing is possible once you have decided to do it ... so keep doing the good work.. tell me more about what you know
about factor VIII ...
and always remember one thing no matter what they say ...
"Positive anything is better than negative nothing."
Filip Hasecke
Feb 1, 2015, 7:20:15 AM2/1/15
to diy...@googlegroups.com
Hi Simon,
as far as I understand you want to use two TLR monomers, that upon pathogen contact dimerize and result in a GFP signal, right?To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/33140523-9954-406f-b0a8-96b022361b15%40googlegroups.com.--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diy...@googlegroups.com. To unsubscribe from this group, send email to diybio+un...@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+un...@googlegroups.com.
To post to this group, send email to diy...@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.
GavinScott
Feb 2, 2015, 9:13:46 AM2/2/15
to diy...@googlegroups.com
Simon writes:
> Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6.
Wikipedia and [2][3] indicate that these operate by "recogniz[ing] the gross, primarily structural features of molecules not innate to the host organism[1]". So really the resulting "OMG! IT'S A PATHOGEN!" signal is probably the product of two things: the fact that the receptor recognized the molecule AND the fact that the molecule was actually there where the receptor was looking for it. So inside a human it works somewhat like a guard patrolling a locked warehouse at night, who doesn't have to be terribly discriminating as pretty much anyone they run into at 2am has a high likelihood of being an intruder. But if you put that guard on a crowded city street, the fact that they're suddenly seeing lots of unknown people is not terribly interesting. So I wonder if the proposed system would be specific enough to be useful outside of the biological environment it evolved to serve.
You might have a hard time finding a procaryote to express this system in (if I correctly understand that to be your goal) that isn't itself something that the system will recognize as a "pathogen". [2] indicates that TLR2 for example is sensitive to components of gram-positive bacteria and yeasts. And if you express it in a gram-negative organism like E. coli, does the more complicated cell-envelope mean that it's hard to get the recognition end of the protein sufficiently "outside" for it to encounter the things it's looking for? But looking at [2] again another TLR2 substrate is triacylated lipoprotiens which I think you'll find in E. coli too.
> Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6.
Wikipedia and [2][3] indicate that these operate by "recogniz[ing] the gross, primarily structural features of molecules not innate to the host organism[1]". So really the resulting "OMG! IT'S A PATHOGEN!" signal is probably the product of two things: the fact that the receptor recognized the molecule AND the fact that the molecule was actually there where the receptor was looking for it. So inside a human it works somewhat like a guard patrolling a locked warehouse at night, who doesn't have to be terribly discriminating as pretty much anyone they run into at 2am has a high likelihood of being an intruder. But if you put that guard on a crowded city street, the fact that they're suddenly seeing lots of unknown people is not terribly interesting. So I wonder if the proposed system would be specific enough to be useful outside of the biological environment it evolved to serve.
You might have a hard time finding a procaryote to express this system in (if I correctly understand that to be your goal) that isn't itself something that the system will recognize as a "pathogen". [2] indicates that TLR2 for example is sensitive to components of gram-positive bacteria and yeasts. And if you express it in a gram-negative organism like E. coli, does the more complicated cell-envelope mean that it's hard to get the recognition end of the protein sufficiently "outside" for it to encounter the things it's looking for? But looking at [2] again another TLR2 substrate is triacylated lipoprotiens which I think you'll find in E. coli too.
> The idea is to ligate the pathogen-sensing, extracellular domains of each TLR (or possibly just the recognition module,
> consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise.
Today I Learned that "In split GFP, two fusion proteins are produced, each one is fused to “half” of a GFP protein (its not exactly half but let not go into that now). If the two fused proteins are in close proximity, the two halves associate to produce an GFP that fluoresce irreversibly[4]." Neat.
However I think this may not help in your application because while the dimerization of the TLRs in the presence of the things they're looking for may bring the split GFPs together, there's probably nothing stopping them from from finding each other without the TLR dimerization, so you may end up with a bunch of TLRs flopping around while connected by their linked GFPs, and lots of florescence under all circumstances.
However I think this may not help in your application because while the dimerization of the TLRs in the presence of the things they're looking for may bring the split GFPs together, there's probably nothing stopping them from from finding each other without the TLR dimerization, so you may end up with a bunch of TLRs flopping around while connected by their linked GFPs, and lots of florescence under all circumstances.
> Using a prokaryotic chassis to carry the TLRs might not be possible owing to the absence of endoplasmic reticula in prokaryotes,
> and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs
> in the cell wall. Any TLRs expressed might not fold properly.
Also potentially the procaryote's collection of lipid species in the membrane might not be what it's expecting (compare the simplicity of the E. coli cell membrane lipid composition to that of a human cell (no cholesterol in E. coli (IIRC) for example)) so even if it folds reasonably well it still might not function perfectly (but for your application I guess you don't care). And somehow all of the recognition end of the thing is going to have to get through (at least) the plasma membrane along with this GFP that's stuck to it.
> I think that TLR signal transduction is not well understood hence the use of GFPs as reporters, as bacteria would lack the requisite
> intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used.
You could always do Drosophila, which already include TLRs and have the advantage of providing a self-propelled mobile aerial sensor platform. There you just need to connect some downstream part of the
signaling cascade with a GFP reporter and watch out for glowing flies (something, something, automated florescence detector, something, something, banana).
signaling cascade with a GFP reporter and watch out for glowing flies (something, something, automated florescence detector, something, something, banana).
> Anyway, it's a learning curve!
Gavin Scott
Feb 2, 2015, 1:54:58 PM2/2/15
to diy...@googlegroups.com
Simon writes:
> Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6.
Wikipedia and [2][3] suggest that these operate by "recogniz[ing] the
> Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6.
gross, primarily structural features of molecules not innate to the
host organism[1]". So really the resulting "OMG! IT'S A PATHOGEN!"
signal is probably the product of two things: the fact that the
receptor recognized the molecule AND the fact that the molecule was
actually there where the receptor was looking for it. So inside a
human it works somewhat like a guard patrolling a locked warehouse at
night, who doesn't have to be terribly discriminating as pretty much
anyone they run into at 2am has a high likelihood of being an
intruder. But if you put that guard on a crowded city street, the fact
that they're suddenly seeing lots of unknown people is not terribly
interesting. So I wonder if the proposed system would be specific
enough to be useful outside of the biological environment it evolved
to serve.
You might have a hard time finding a procaryote to express this system
in (if I correctly understand that to be your goal) that isn't itself
something that the system will recognize as a "pathogen". [2]
indicates that TLR2 for example is sensitive to components of
gram-positive bacteria and yeasts. And if you express it in a
gram-negative organism like E. coli, does the more complicated
cell-envelope mean that it's hard to get the recognition end of the
protein sufficiently "outside" for it to encounter the things it's
looking for? But looking at [2] again another TLR2 substrate is
triacylated lipoprotiens which I think you'll find in E. coli too.
host organism[1]". So really the resulting "OMG! IT'S A PATHOGEN!"
signal is probably the product of two things: the fact that the
receptor recognized the molecule AND the fact that the molecule was
actually there where the receptor was looking for it. So inside a
human it works somewhat like a guard patrolling a locked warehouse at
night, who doesn't have to be terribly discriminating as pretty much
anyone they run into at 2am has a high likelihood of being an
intruder. But if you put that guard on a crowded city street, the fact
that they're suddenly seeing lots of unknown people is not terribly
interesting. So I wonder if the proposed system would be specific
enough to be useful outside of the biological environment it evolved
to serve.
You might have a hard time finding a procaryote to express this system
in (if I correctly understand that to be your goal) that isn't itself
something that the system will recognize as a "pathogen". [2]
indicates that TLR2 for example is sensitive to components of
gram-positive bacteria and yeasts. And if you express it in a
gram-negative organism like E. coli, does the more complicated
cell-envelope mean that it's hard to get the recognition end of the
protein sufficiently "outside" for it to encounter the things it's
looking for? But looking at [2] again another TLR2 substrate is
triacylated lipoprotiens which I think you'll find in E. coli too.
> The idea is to ligate the pathogen-sensing, extracellular domains of each TLR (or possibly just the recognition module,
> consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise.
> consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise.
Today I Learned that "In split GFP, two fusion proteins are produced,
each one is fused to “half” of a GFP protein (its not exactly half but
let not go into that now). If the two fused proteins are in close
proximity, the two halves associate to produce an GFP that fluoresce
irreversibly[4]." Neat.
However I think this may not help in your application because while
the dimerization of the TLRs in the presence of the things they're
looking for may bring the split GFPs together, there's probably
nothing stopping them from from finding each other without the TLR
dimerization, so you may end up with a bunch of TLRs flopping around
while connected by their linked GFPs, and lots of florescence under
all circumstances.
each one is fused to “half” of a GFP protein (its not exactly half but
let not go into that now). If the two fused proteins are in close
proximity, the two halves associate to produce an GFP that fluoresce
irreversibly[4]." Neat.
However I think this may not help in your application because while
the dimerization of the TLRs in the presence of the things they're
looking for may bring the split GFPs together, there's probably
nothing stopping them from from finding each other without the TLR
dimerization, so you may end up with a bunch of TLRs flopping around
while connected by their linked GFPs, and lots of florescence under
all circumstances.
> Using a prokaryotic chassis to carry the TLRs might not be possible owing to the absence of endoplasmic reticula in prokaryotes,
> and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs
> in the cell wall. Any TLRs expressed might not fold properly.
> and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs
> in the cell wall. Any TLRs expressed might not fold properly.
Also potentially the procaryote's collection of lipid species in the
membrane might not be what it's expecting (compare the simplicity of
the E. coli cell membrane lipid composition to that of a human cell
(no cholesterol in E. coli (IIRC) for example)) so even if it folds
reasonably well it still might not function perfectly (but for your
application I guess you don't care). And somehow all of the
recognition end of the thing is going to have to get through (at
least) the plasma membrane along with this GFP that's stuck to it.
membrane might not be what it's expecting (compare the simplicity of
the E. coli cell membrane lipid composition to that of a human cell
(no cholesterol in E. coli (IIRC) for example)) so even if it folds
reasonably well it still might not function perfectly (but for your
application I guess you don't care). And somehow all of the
recognition end of the thing is going to have to get through (at
least) the plasma membrane along with this GFP that's stuck to it.
> I think that TLR signal transduction is not well understood hence the use of GFPs as reporters, as bacteria would lack the requisite
> intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used.
> intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used.
You could always do Drosophila, which already include TLRs and have
the advantage of providing a self-propelled mobile aerial sensor
platform. There you just need to connect some downstream part of the
signaling cascade with a GFP reporter and watch out for glowing flies
(something, something, automated florescence detector, something,
something, banana).
the advantage of providing a self-propelled mobile aerial sensor
platform. There you just need to connect some downstream part of the
signaling cascade with a GFP reporter and watch out for glowing flies
(something, something, automated florescence detector, something,
something, banana).
> Anyway, it's a learning curve!
Simon Rose
Feb 3, 2015, 1:32:17 PM2/3/15
to diy...@googlegroups.com
Hi Blenderkid and Gavin, sorry about the delay in responding, I've been offline for a few days !
In response to Blenderkid's question, yes I intend to try to couple the TLR's to a cell. Initially I thought of using e.coli as a chassis, and as TLR2/6 respond to gram- positive bacteria and e.coli is gram negative, self- detection should not be a problem. It would be something on the lines of a little "machine" for sensing gram-positive pathogens. However, I am very unsure if the transmembrane domains of both TLRs are even capable of being anchored in the cell -wall of a gram negative bacterium, so I might have to use yeast as a chassis.
The TMs of TLRs 2 and 6 are both 21 AAs in length by the way. There is some information about them on PLOS1. Apparently isolated TMDs are capable of oligermerisation independently of any PAMPS detected by the extracellular portions (PLOS 1 November 2012 | Volume 7 | Issue 11 | e48875 )
As for detection one idea would be to extract the TLR from say, HeLa ( assuming they're not too badly mutated) having tagged the primers with histidine codons, ligate into an expression vector and then detect any expressed protein with anti-6His antibodies, just to see if anything happens.
As for the problem with split GFPs flopping around and dimerising independently, well this should not be a problem if the TLR GFP constructs are anchored in a cell membrane/wall. There has actually been some fairly recent work on fluorescent biosensors here;
Eiji Nakata, FongFong Liew, Shun Nakano and Takashi Morii (2011). Recent progress in the construction
methodology of fluorescent biosensors based on biomolecules, Biosensors - Emerging Materials and
Applications, Prof. Pier Andrea Serra (Ed.), ISBN: 978-953-307-328-6,rey ss b (
These ideas are still in progress by the way.
Its nice to have an international community to help out. The immune system fascinates me, by the way which is why I'm doing this! Thanks for the responses so far!
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6 The idea is to ligate the pathogen-sensing, extracellular domains of each TLR (or possibly just the recognition module, consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise. Using a prokaryotic chassis to carry the TLRs might not be possible owing to the absence of endoplasmic reticula in prokaryotes, and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs in the cell wall. Any TLRs expressed might not fold properly. I think that TLR signal transduction is not well understood hence the use of GFPs as reporters, as bacteria would lack the requisite intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used. Anyway, it's a learning curve!
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Hi everyone I'm interested in making a synthetic pathogen sensor using human toll- like receptors 2 and 6 The idea is to ligate the pathogen-sensing, extracellular domains of each TLR (or possibly just the recognition module, consisting of just a single residue) to one half of a GFP, producing a visible signal when the TLRs dimerise. Using a prokaryotic chassis to carry the TLRs might not be possible owing to the absence of endoplasmic reticula in prokaryotes, and the consequent problems with protein folding, as well as with being able to anchor the transmembrane domains of the TLRs in the cell wall. Any TLRs expressed might not fold properly. I think that TLR signal transduction is not well understood hence the use of GFPs as reporters, as bacteria would lack the requisite intracellular signalling pathways. If these problems prove insuperable, then a eukaryotic chassis might have to be used. Anyway, it's a learning curve!
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Filip Hasecke
Feb 3, 2015, 3:31:11 PM2/3/15
to diy...@googlegroups.com
Hi Simon,
First of all, this might be interesting for you: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1300016&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D1300016
I'm currently at home so I could not download the full text, but the abstract sounds promising.
2015-02-03 19:32 GMT+01:00 Simon Rose <excalib...@googlemail.com>:
Hi Blenderkid and Gavin, sorry about the delay in responding, I've been offline for a few days !In response to Blenderkid's question, yes I intend to try to couple the TLR's to a cell. Initially I thought of using e.coli as a chassis, and as TLR2/6 respond to gram- positive bacteria and e.coli is gram negative, self- detection should not be a problem. It would be something on the lines of a little "machine" for sensing gram-positive pathogens. However, I am very unsure if the transmembrane domains of both TLRs are even capable of being anchored in the cell -wall of a gram negative bacterium, so I might have to use yeast as a chassis.
As for the display of the TLRs on the surface of E.coli, I'd recommend a fusion to ompA or any other kind of surface display:
http://www.sciencedirect.com/science/article/pii/S016777991000199X
http://www.sciencedirect.com/science/article/pii/S016777991000199X
Simply search for "E.coli surface display" for more information.
Also removing the transmembrane domain of a TLR does not obliterate its recognition ability per se, so using the entire extracellular region of the TLRs fused to ompA might be promising.
An example of TLR recognition sites (entire extracellular domain) used in fusion proteins: http://www.google.com/patents/EP1657257A1?cl=en
An example of TLR recognition sites (entire extracellular domain) used in fusion proteins: http://www.google.com/patents/EP1657257A1?cl=en
As for the problem with split GFPs flopping around and dimerising independently, well this should not be a problem if the TLR GFP constructs are anchored in a cell membrane/wall.
I would doubt that, as a membrane is somehow a fluid itself. Your protein will not be tightly bound on the membrane surface unless you attach it to the cytoskeleton. There will always be some background signal as a result of random dimerization, but hopefully it will be less than the positive signal when pathogens are present.
Anyway, one other concern might be the sensitivity of this approach, as the underlying signaling cascade, which amplifys the reaction to pathogen contact, is missing. You might end up with only a low fluorescence signal reducing the applicability of your method to wastely infested water puddles, instead of a blood samples for example.
.
.
One other note for the time when you start testing your sensor: The samples that contain pathogens should be lysated prior to the detection experiments, as it virtually increases the amount of pathogen in your sample by dispersing the bacteria into tiny bits, instead of having intact bacteria in your solution. This might improve your signals.
Good luck with your project and please keep us up to date ;)
-Filip
Simon Rose
Feb 4, 2015, 11:31:01 AM2/4/15
to diy...@googlegroups.com
Thanks Filip, just seen the articles
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Mike Horwath
Feb 4, 2015, 11:53:47 AM2/4/15
to diy...@googlegroups.com
Hi Simon,
Immunobiology grad student here. Sounds like a very interesting although potentially challenging project! It's great to explore the possibilities.
Regarding "background" dimerization of the split-GFP: This is a concern, but careful design and testing could avoid it. Some TLR's exist as monomers before they see the ligand, some exist as dimers. I think TLR2 and TLR6 are in the category that exist primarily as monomers, but there's still likely to be some random interaction. I would try to take advantage of the change in protein conformation that occurs when TLRs bind ligand, which brings the cytoplasmic TIR domains closer together. The length and "floppiness" of the linker to the GFP-fragment could be optimized so that interaction between split-GFPs is unlikely unless ligand binding occurs.
A few links:
http://www.ncbi.nlm.nih.gov/pubmed/24419035 --Overview of molecular mechanisms of TLR ligand binding
http://www.ncbi.nlm.nih.gov/pubmed/19378018 ---Book chapter exploring ideas very similar to yours, although they don't propose bacterial expression. They discuss both split-GFP and FRET methods. From personal experience, FRET gives you beautiful data when it works but can be a major PITA!
Some speculation...I'm not too familiar with function of animal membrane proteins in bacteria, so I can't comment on how feasible that would be. However, if you can't get bacterial expression to work, maybe consider plant cells? Plants have their own set of pattern-recognition receptors (PRRs) for bacteria, virus, etc...not TLRs, but similar LLR structure. (Toll-like-receptor-like-receptors?)
Final comment...Blenderkid's IEEE link shows a good point, immune cells are already exquisite little contamination sensors. I routinely treat mouse dentritic cells with 1ng/mL lipopolysacharide, and 6 hours later they have changes in gene expression, surface markers, and even cell shape. This is just a few thousand molecules of LPS per cell!
Cheers,
Mike
Immunobiology grad student here. Sounds like a very interesting although potentially challenging project! It's great to explore the possibilities.
Regarding "background" dimerization of the split-GFP: This is a concern, but careful design and testing could avoid it. Some TLR's exist as monomers before they see the ligand, some exist as dimers. I think TLR2 and TLR6 are in the category that exist primarily as monomers, but there's still likely to be some random interaction. I would try to take advantage of the change in protein conformation that occurs when TLRs bind ligand, which brings the cytoplasmic TIR domains closer together. The length and "floppiness" of the linker to the GFP-fragment could be optimized so that interaction between split-GFPs is unlikely unless ligand binding occurs.
A few links:
http://www.ncbi.nlm.nih.gov/pubmed/24419035 --Overview of molecular mechanisms of TLR ligand binding
http://www.ncbi.nlm.nih.gov/pubmed/19378018 ---Book chapter exploring ideas very similar to yours, although they don't propose bacterial expression. They discuss both split-GFP and FRET methods. From personal experience, FRET gives you beautiful data when it works but can be a major PITA!
Some speculation...I'm not too familiar with function of animal membrane proteins in bacteria, so I can't comment on how feasible that would be. However, if you can't get bacterial expression to work, maybe consider plant cells? Plants have their own set of pattern-recognition receptors (PRRs) for bacteria, virus, etc...not TLRs, but similar LLR structure. (Toll-like-receptor-like-receptors?)
Final comment...Blenderkid's IEEE link shows a good point, immune cells are already exquisite little contamination sensors. I routinely treat mouse dentritic cells with 1ng/mL lipopolysacharide, and 6 hours later they have changes in gene expression, surface markers, and even cell shape. This is just a few thousand molecules of LPS per cell!
Cheers,
Mike
Mike Horwath
Feb 4, 2015, 12:13:00 PM2/4/15
to diy...@googlegroups.com
One more link of potential interest, figure 2 especially:
http://www.jbc.org/content/281/41/31002.full
They labeled TLR2 with fluorescent antibody and used FRET to measure heterodimerization with other surface proteins after stimulation.
http://www.jbc.org/content/281/41/31002.full
They labeled TLR2 with fluorescent antibody and used FRET to measure heterodimerization with other surface proteins after stimulation.
Simon Rose
Feb 5, 2015, 10:24:07 AM2/5/15
to diy...@googlegroups.com
Thanks Mike
On Saturday, January 31, 2015 at 7:52:54 PM UTC, Simon Rose wrote:
Reply all
Reply to author
Forward
0 new messages