Urease

From 2006.igem.org

Finding a suitable urease

To generate an increase in pH, we decided to take advantage of the urease reaction, in which urea is cleaved by urease to yield ammonium and carbon dioxide. This reaction is used in clinical microbiology to identify urease-positive bacteria such as Proteus vulgaris, which in the presence of urea can raise the pH of the growth medium to 9 or higher. The most obvious source of urease would be the urease operon of a urease-positive E. coli strain such as EDL933. The sequence of this operon is available in Genbank. However, the gene cluster consists of almost 5 kb of DNA encoding 7 genes, ureDABCEFG, where ureABC encode the three subunits of urease and the other genes encode proteins required for insertion of the nickel cofactor; also, sequence analysis showed the presence of 5 PstI sites and 1 EcoRI site, all of which would have to be individually mutated out to make this DNA into a biobrick.

Analysis of other sequenced urease operons, and a search of the associated literature, showed that the Bacillus subtilis 168 urease operon consisted of only three genes, ureABC, which nevertheless can be assembled into a functional urease in E. coli without the requirement for accessory proteins, although urease activity was relatively low (Kim, J.K., Mulrooney, S.B., and Hausinger, R.P. 2005. Biosynthesis of active Bacillus subtilis urease in the absence of known urease accessory proteins. Journal of Bacteriology 187, 7150-7154). Unfortunately, the sequence still contained one EcoRI and one SpeI site, which would need to be removed before a biobrick could be made. We therefore decided to test this urease to see whether it would be suitable for our purposes.

The genes ureABC were amplified from B. subtilis 168 genomic DNA using primers with a SacI site in the forward primer and a BamHI site in the reverse primer. The PCR product was initially ligated into pGemT-easy (Promega), a vector designed for cloning PCR products with AT overhangs. This vector possesses a lac promoter at one end of the insertion site and a T7 promoter at the other. We had hoped to recover a clone with the insert in the correct orientation for expression from the lac promoter, but all clones tested using SacI digests had the insert in the reverse orientation, which could only be expressed from the T7 promoter. One of these plasmids was therefore introduced into E. coli BL21(DE3), which expresses T7 RNA polymerase in the presence of IPTG. Also, in case activity might be different in our normal host strain, JM109, the insert was excised as a SacI-BamHI fragment and inserted into pBluescript SK+ (Stratagene), so that it could be expressed from the lac promoter.

Expression was tested in the two clones: BL21(DE3)/pGemTe-ureABC and JM109/pBluescript-ureABC. In both cases, when expression was induced with IPTG in LB medium with 2% w/v urea, a slow pH increase was seen, with the pH reaching 9 after overnight incubation. In control cultures lacking urea, or using control plasmids pT7-7 or pBluescript SK+, no such pH increase was seen. This indicated that the B. subtilis urease was potentially suitable for our purposes.

Making a urease biobrick

We then sought to remove the offending EcoRI and SpeI sites using a PCR-based strategy. The operon was amplified as three separate PCR products. The SpeI site was removed silently by converting the SpeI site in one fragment to an XbaI site. Ligation of these compatible sticky ends would not regenerate either site. The EcoRI site was more problematic. It was removed by adding an XhoI site adjacent to it. This could not be achieved silently, but required a conservative Asp to Glu mutation (D526E) in UreC, which we hoped would not affect activity. The three PCR products were digested respectively with XbaI, SpeI+XhoI, and XhoI, and ligated together. The ligation was used as template for a PCR reaction with the forward primer from fragment 1 and the reverse primer from fragment 3. This PCR was successful, and the product was cloned in pGemT-easy as before. In this case, both orientations of the construct were recovered, allowing testing of activity in both JM109 and BL21(DE3). Restriction analysis showed that the two undesired restriction sites had been successfully removed, and the XhoI site introduced.

Unfortunately, in all activity tests, clones bearing the mutated operon showed no pH increase, though controls bearing the wildtype operon continued to work well. Sequencing of the construct showed that ureA and ureB had the expected sequence, but that several apparent point mutations were present in ureC, two of which (D326G and S338T) were not silent, though one resulted in a conservative substitution. More seriously, there appeared to be a frameshift error due to a missing base in ureC in the region of amino acid residue 528. Oddly, this lay within the sequence of one of the mutagenic primers. If this is not a sequencing artefact, then it suggests either an error in the primer synthesis, or an unusual PCR-induced mutation. In any case, all we can say for certain is that the mutated clones we have obtained to date are inactive and cannot be used to make a biobrick. A future approach would be to repeat the mutagenesis using a kit such as Quickchange (Stratagene), which involves less PCR and thus offers less scope for introduction of undesired mutations.

Table 1: a sample result showing pH from constructs in JM109

Making the hybrid promoter

In order to ensure that urease was not produced during growth of the inoculum prior to assays, our design called for a hybrid promoter that could be repressed by lambda cI repressor (if enough arsenate was present to induce this) and also by LacI, so that the system could be switched on by the presence of lactose or IPTG in the assay medium. On close examination of the relevant promoter regions, we found that cI binding sites OR1, OR2 and OR3 in bacteriophage lambda overlapped the -35 and -10 regions of the PR and PRM promoters, whereas in the lac promoter, the LacI binding site was downstream of the -10 site. This suggested that we could achieve our aim by fusing the LacI binding site from the lac promoter to the OR1-OR3 region from bacteriophage lambda. This was accomplished by amplifying the two DNA fragments in separate PCR reactions. The reverse primer for the lambda fragment, and the forward primer for the lac fragment, each included an XhoI site. The reverse primer for the lac fragment was the same as that used in generating our lacZ' biobrick, BBa_J33202, so that the lac fragment also included the 5' region of lacZ encoding the first 76 amino acids of LacZ followed by an introduced stop codon. This would allow a simple means of testing whether the construct was repressed by lambda cI.

The two PCR products were digested with XhoI and ligated together to generate the final construct, which was designated BBa_J33205. The sequence of the DNA, with full annotation showing all of the operator sites and the fusion site, may be found in the Registry entry for this part.

Proper testing of the regulation of this hybrid promoter will require its introduction into a system expressing lambda cI. Unfortunately, we did not have time to accomplish this, so we are unsure whether or not this part will function as desired. However, strong expression of LacZ activity on Xgal/IPTG plates suggests that the basic PR promoter function is intact.

If it works properly, this hybrid promoter may also have other uses. Lambda promoter PRM is also present, so should drive expression in the reverse direction in the presence of cI. Thus the presence of cI should cause switch off transcription in the forward direction (as defined by the Biobrick ends) from promoter PR, and switch on transcription in the reverse direction from promoter PRM. This functionality is not used in our system, but may be useful for other projects in the future.

[http://2006.igem.org/University_of_Edinburgh_2006 Main page]