Biobricking of “IP-Free” Fluorescent Protein Genes

We started the competition period with a lofty goal- to create a small library of biobrick parts that would be open to the DIYbio community. The realities of an adult team that all have full time day jobs made it necessary to scale back that ambition, so we decided to focus on some fluorescent protein genes since turning E. coli K-12 green has become the "Hello World" experiment of citizen scientist groups everywhere. Little did we realize how twisted the path to open source biology is!

The Registry of Standard Biological Parts does not appear to contain many fluorescent proteins that are free of intellectual property limitations. Four are listed (BBa_J97003, BBa_J97002, BBa_J97000, BBa_J97001) but none of these are physically deposited in the Registry. So we decided to focus our efforts on submitting an IP-free GFP, RFP, CFP, YFP and OFP that we could give out to other community labs just starting up.

We were delighted to see that the company DNA2.0 listed a collection of proteins on its product site that it called "IP-Free© synthetic non-Aequorea fluorescent proteins". We reasoned that by using the right PCR primers we could easily copy these genes out of the commercial plasmid and biobrick them.

The registered trademark sign after the words "IP-Free" should have alerted us to dig deeper, and by the time we did we had already committed too much time and resources to the project to take an alternate path. DNA 2.0's intellectual property policy is very specific- you can use the gene for a further invention only if you buy it in their plasmid at a cost of $245 per gene. You are not allowed to have it synthesized. You are not allowed to distribute it unless it is modified such that the function is changed or enhanced. We did not realize this, and so our biobricked parts can only be used for purely academic pursuits, which are allowed by DNA2.0. However, if you then want to commercialize your invention you must purchase the gene directly from them.

A lesson learned to dig deeply whenever IP issues arise.

This left us with a dilemma. Do we continue and potentially submit genes that are of no use? We did not know whether or not the restriction on use by DNA2.0 would result in no team being able to get them from the registry. Since non-commercial use of the genes appears to be OK, we reasoned that perhaps the registry would make them available under those conditions. But giving them out to other DIYbio labs might not be possible.

We considered three alternative courses of action. The first would be to abandon the idea of giving out these genes based on the IP restrictions. That would be a shame, since we worked all summer to clone them. The second would be to contact DNA 2.0 and see if we could get some sort of exception or special licence to distribute to community labs with the understanding that if any commercial product came out of the work they would have to buy the plasmid from DNA 2.0. The third most radical course would be to challenge the "IP-Free" designation as deceptive advertising and give out the plasmid to precipitate legal action. This would be the most interesting, but also risky and somewhat combatitive in spirit- we were not sure that this was a fight we wanted to get into if we were choosing our battles wisely.

In the end, we decided to go with the middle alternative and contact DNA 2.0 as a first action. We are awaiting an introduction to key individuals at DNA 2.0 and will see where that leads...

Our cloning strategy was simple.

1. Design primers that bracket the gene and upstream RBS but add biobrick prefix and suffix
2. Clone the fragment into pSB1C3
3. Test the color generation by cutting out the fragment and cloning into pUC19 which puts it under the control of the constituitive T7 promoter

All the genes were in the DNA2.0 plasmid backbone FPB-xx-444 that includes AmpR. The upstream and downstream regions surrounding the open reading frame were the same for all plasmids. We designed one forward primer that could be used for all five genes, but we had to design specific reverse primers for them all since there was an EcoR1 site immediately after the termination codon.

Forward primer for all: ATGAATTCGCGGCCGCTTCTAGAGCGAGACCTTAGGAGGTAAAC

We realized later that the YPF had an internal EcoR1 site that we had missed, and for some reason we failed to get good clones of the GFP, so we submitted and tested the RFP, OFP and CFP protein genes as documented on our registry pages.

They look great!

But our next step is to pursue the cloning of truly IP-free fluorescent protein genes.