Summary of information from Eric P on Nanno Engineering

84 views
Skip to first unread message

Cameron Clarke

unread,
May 14, 2015, 4:05:41 PM5/14/15
to wet-lab-san-...@googlegroups.com
Here are a summary of notes from Eric on approaches we can take on engineering the Nanno. We'll be making a plan so anyone interested, please advise.
Thanks
Cameron
-------------------
ON ELECTROPORATION: 
There is no setting for field strength but should be ~500 ohms resistance and 50 uF capacitance may work. 
Before each Electroporation (Gene Pulser II) after setting all the parameters: Measure actual sample resistance (manual p. 12). a. Sample resistance> press buttons for ‘actual volts’ and ‘time constant’ at the same time b. Electroporation in 2 mm cuvettes: 2.2 kV, capacitance at 50 µF, resistance at 600 Ohms For all electroporations record time constant (expected to be around 25ms) Note: actual volts, capacitance and time constant can be read after each electroporation, actual resistance of the sample can be read before the electroporation (usually 500-600 OHMS).
--------------------
I can send over the plasmids next week, no problem. I have vectors for overexpression and epitope tags luciferase (firefly, renilla, and nanoluc), fluorescent proteins (venus, cerulean) and HA tags. I have transformed Nanno with zeocin connected to luciferase by F2A and P2A peptides and will be testing those for cleavage over the next month. Also I am waiting for CRISPR transformants to appear to test those for knockouts. 

The Nature paper discusses using bacteria to transform diatoms which may be a method that may be useful for DIYBIO (and has many advantages), though it seems it would take a far amount of cloning. Otherwise electroporation and sonication are two methods which seem to work.  

If the 2A peptides work in Nanno those could have many used and combined with bidirectional promoters in Nanno for gene stacking. In particular I would propose to express the bacterial luciferase biosynthetic pathway using this approach and express the luciferase under a night specific promoter (cellulose synthase).  

----------------
For electroporation of algae the field strength, capacitance, and resistance have to be optimized, meaning the electroporator needs to have those as settings - most times a Biorad gene pulser II is used but there are probably others out there. The conjugation system sounds quite attractive due to its simplicity and ability to transfer large pieces of DNA. It seems feasible to use any plasmid after inserting an oriT and using an e coli with the conjugation genes (genome integrated strain S-17-1 or on plasmid pTA-MOB). SGI is doing some pretty amazing things, though our vector system is gateway based I would like to move to yeast recombination for complex projects (I think the gateway site is still useful for the positive + negative selection).  

 I will send my selection, overexpression vectors, and luciferase vectors, which have been tested and work- you can transform the overexpression yfp or cfp vectors to express a fluorescent protein. The vectors will dried on paper and don't need refrigeration - will try to send this week. I also am testing a series of chlamy luciferase vectors (firefly, renilla, nanoluc) and we can discuss if you have an interest in chlamy. I am using the EFa promoter for Cas9 expression, and the endogenous U6 region for sgRNA expression. We'll see if we get expression and mutations, after that I need to talk to my PIs about sharing it. I have also attached our light:dark RNA-Seq which we are working on publishing now.

Gene stacking approaches using bidirectional promoters and 2A peptides would be a project that I am really interested in and could assist the Wetlab with. The lux operon would be an elegant way to test the approach. I have identified a number of bidirectional promoters in our strain (VCPs, histones, and ribosomal proteins) which I can share. I have attached a map of the vector (pNOC stacked) I would like to make for gene stacking with yeast based cloning. Hypothetically this construct could use one resistance gene to express three other proteins, three genes make up the lux luciferin pathway. A construct like this could be cut at the unique cloning sites and combined with overlapping PCR products to clone in multiple proteins at once through yeast TAR. Also only hyg or zeo should be necessary so two vectors could be produced from the backbone. I have determined the codon bias of Nanno which prefers GC codons, like chlamy, and it would probably be best to synthesize the genes of the lux operon which are 900-1200 bp (if the genes are available from a bacteria we could try those first). For expressing large numbers of transgenes bidirectional promoter cassettes could be combined, and this could be a backup plan if the 2A peptides doesn't work well. Anyway somes ideas to work with.
-----------------------
HERE is a link to all of the related documentation
---------------------------------
I've been talking with a colleague who uses and has designed conjugation systems and he can provide me with an e coli strain for conjugations. I am not sure if there are weaknesses to using this strain vs the helper plasmid approach but we are going to discuss tomorrow about the different options. The vectors developed in the episome paper are available from addgene, though they come by MTA. So it seems conjugation based transformation is an option to consider.

With the electroporator you got I think the field strength maxed out at 1.0 kV so you would want to use .1 cm cuvettes (I think).
CAMERON NOTE: I've ordered 1mm cuvette samples to test
-----------------------------------
http://www.nature.com/ncomms/2015/150421/ncomms7925/full/ncomms7925.html#affil-auth
Pretty interesting. We have observed Nannochloropsis seems to be able to maintain plasmids as well.
-------------------------------------------------
There is a new website http://www.nannochloropsis.org/index.php which has some high quality data on it, they also have a nice new paper in Molecular Plant. They have a keyword genome search and have started a biocyc database- Nagacyc based on N. gaditana. The nagacyc part of the website comes and goes but you can download the database (and biocyc) and run it locally. It's a very good way to visualize the metabolism of nanno and allows some in silico metabolic analysis. A database like this is necessary to organize info of highly characterized species and metabolic engineering and we will probably attempt to do the same with N. oceanica.
-----------------------------------------------


eric poliner

unread,
May 14, 2015, 5:53:09 PM5/14/15
to wet-lab-san-...@googlegroups.com, selur...@gmail.com
In regards to the pathway tools database, I have a rough draft version which I have shared with Cameron. The database incorporates the go terms, pfams, and functional names from vieler et al, and reciprocal best hit blast results of the two genomes that were available when I made this (gaditana ccmp526, gaditana B31). The gene models are not correct, just includes the start and stop of the transcript based on our gff file. I used the pathologic wizard without taxonomic pruning - so their are a lot of pathways which aren't real but it caught a lot of stuff it missed with pruning. 

Also check out http://www.bioenergychina.org:8989/index.html. This the website of the N oceanica imet1 genome, they have blastn, tblastn, and a genome browser up and running.
Reply all
Reply to author
Forward
0 new messages