ABSTRACT

At the moment, the most popular system in use for protein expression is the pET expression system. The pET expression system is a lactose induced T7 polymerase dependent system. This system, however, has few downsides: i) the system tends to leak; ii) it is very hard to control the protein expression level once it is induced; iii) being a viral polymerase,T7 polymerase has a tendency to insert the mismatches (mutations) in the transcripted sequence; iv) if the produced protein is toxic for the host, the bacterial culture can decline before the desired concentration of the protein of interest is manufactured; v) the organized and stable transfer of the produced protein to i.e. a medium can prove difficult to carry out.

Therefore, we decided to seek an alternate solution to use in protein expression. We prepared a series of plasmids expressing his-tagged superfolder GFP under the control of four nontoxic inducers: arabinose, melibiose, rhamnose and xylose. The four promoters are more controllable than the most popular lactose induced T7RNA polymerase dependent systems. The coding sequences are transcribed by the cellular RNA polymerase which is slower and more accurate than T7 RNA polymerase. SfGFP can be used to compare activities of standard, modified, and minimal promoters during growth on different carbon sources and as well as a probe of a fused protein solubility. We present a comprehensive comparison of the sfGFP expression driven by inducible promoters with the insulated, constitutive promoters proC and proD as standards. Two pairs of primers should be sufficient to clone an ORF of interest as a protein fused to sfGFP or to replace it in all vectors of the mulltipromoter system.

Our next step is to compare the efficiency between these promoters and test their function when grown on the media consisting of different components.

All promoter fragments derive from Escherichia coli K12, strain DH5 alpha, some elements are synthetic. Chromosomal DNA of E. coli was isolated as described in the methods section.

Arabinose induced promoters

sfGFP under arabinose promoter s1 without AraC (Arashort1) - BBa_K1741000

In commercially available expression vectors arabinose promoter araBAD is usually used together with a gene for AraC transcription activator / repressor to make the system independent of chromosomal copy of AraC gene. We fused the araBAD promoter without AraC reading frame, as other sugar induced promoters to sfGFP. The shorter version of the popular promoter is still functional i.e. induced by arabinose. Arabinose induced (0,4%) in a a short time (1-8h) expression is a bit lower than driven by AraC-araBAD but after a longer time like 18h the GFP level is higher than from AraC-araBAD promoter. New shorter versions of arabinose induced promoters, all originating from E. coli genome were compared to the biobrick BBa_K1481002, provided last year by Poznan_Bioinf team (labeled as AraWT in the comparison chart).
The sequence of this promoter starts from O2 region the left part of the binding site of AraC in absence of arabinose.

You can find this part's sequence with more information in registry: BBa_K1741000.

sfGFP under a short arabinose promoter s2 without O1 and O2 - BBa_K1741002

This shortened version of araBAD promoter does not contain O1 and O2 regions, but still contains the CRP binding site. We can expect than this version of araBAD will be still repressed by CRP but not by AraC without arabinose. From our preliminary experiments it appears that the promoter is more strongly induced by arabinose than AraC-AraBAD but is weaker than arashort1. All three arabinose pomoters are repressed by glucose.

You can find this part's sequence with more information in registry: BBa_K1741002.

Arabinose induced promoters comparison

The fluorescence was measured after 18h-growth. Arashort1 (without AraC element) seems to be the strongest. Arabinose promoters are blocked by glucose.

Melibiose induced promoters

sfGFP under melibiose promoter - BBa_K1741003

MelR chromosomal copy of MelR is sufficient to activate the melibiose melAB promoter copied from E. coli K12 chromosome when melibiose is added to M9 minimal medium or to a rich medium 2xLB (0,4%). In both media induction is much weaker than of arabinose or rhamnose induced promoters.
The hairpin sequence below.

You can find this part's sequence with more information in registry: BBa_K1741003.

sfGFP under melibiose promoter, with improved 5'UTR - BBa_K1741004

Twice higher expression has been obtained by 5’UTR editing – removal of potential secondary structure which involved RBS and a better positioning of RBS - 7 nt upstream AUG start codon. The modified 5’UTR results in about twice higher expression upon melibiose addition to the M9 or 2xLB medium. The promer is still weaker than araBAD or rhaBAD, so can be a promoter of choice if lower expression level is necessary for efficient folding or secretion of a recombinant protein.
The modified sequence below.

You can find this part's sequence with more information in registry: BBa_K1741004.

Melibiose induced promoters comparison

The fluorescence was measured after 18h-growth. Mel2 (modified version) seems to be the strongest. Both versions of melibiose promoters are blocked by glucose, arabinose and xylose.

Rhamnose induced promoters

sfGFP under rhamnose promoter with removed EcoRI site - BBa_K1741005

rhaBAD promoter has been copied from E. coli genome: ggtgagcatcacatcaccacaattcagcaaattgtgaacatcatcacgttcatctttccctggttgccaatggcccattttcctgtcagtaacgagaaggtcgcgaattcaggcgctttttagactggtcgtaatgaaattcagcaggatcacatt,
then Eco RI site was removed by PCR to obtain Rha1 in a biobrick standard: ggtgagcatcacatcaccacaattcagcaaattgtgaacatcatcacgttcatctttccctggttgccaatggcccattttcctgtcagtaacgagaaggtcgcgaatttaggcgctttttagactggtcgtaatgaaattcagcaggatcacatt.
The only one point mutation makes the promoter slightly stronger and more sensitive to rhamnose.

You can find this part's sequence with more information in registry: BBa_K1741005.

sfGFP under short rhamnose promoter (very weak) - BBa_K1741006

From rhaBAD1 promoter we removed both CRP binding sites. It seems likely thatt RhaS binding site has been also affected, so however the resulting promoter is still induced by rhamnose, the induction is very weak.

You can find this part's sequence with more information in registry: BBa_K1741006.

Rhamnose induced promoters comparison

The fluorescence was measured after 18h-growth. Rha1 (without EcoRI site) seems to be slightly stronger. Rhamnose promoters are blocked by glucose.

Xylose induced promoters

sfGFP under xylose promoter xylA (xylWT) - BBa_K1741007

SfGFP under xylose promoter xylA (wild type).

You can find this part's sequence with more information in registry: BBa_K1741007.

sfGFP under xylose promoter xylA1 with improved 5'UTR - BBa_K1741008

Xylose induced promoter xylA is weaker than arabinose and rhamnose promoters. To enhance the efficiency of xylA driven transcription/translation we have transplanted 5'UTR from the strong constitutive promoter proD. In some experiments it seems to be slightly stronger than original xylA promoter from part BBa_K1741007.

You can find this part's sequence with more information in registry: BBa_K1741008.

sfGFP under shortened xylose promoter xylA1 with improved 5'UTR BBa_K1741009

Short version of xylA promoter with known functional elements is stronger than two other versions.

You can find this part's sequence with more information in registry: BBa_K1741009.

sfGFP under xylose promoter xylF (very weak) - BBa_K1741010

We also tested the xylF promoter from the same double sided regulatory element, and xylF appears to be a very weak promoter slightly induced by xylose.

You can find this part's sequence with more information in registry: BBa_K1741010.

Xylose induced promoters comparison

The fluorescence was measured after 18h-growth. Xylose promoters are blocked by glucose and weakened by arabinose.

XylS is the strongest. Strength was calculated using sfGFP fluorescence.

Other promoters

sfGFP under T7 promoter - BBa_K1741011

This part imitates the most popular IPTG/lactose dependent T7 driven expression systems. To express sfGFP or another sequence with which one can substitute sfGFP ORF using Gibson or CPEC assembly, host strain of choice expressing T7 RNA polymerase is necessary, not only DE3.

You can find this part's sequence with more information in registry: BBa_K1741011.

T7 promoter leaks on all sugars that were used in the experiment, it’s activity in Tuner strain seems to depend only on E. coli growth rate.

sfGFP under proC promoter - BBa_K1741012

proC - moderately strong constitutive promoter seems to reflect growth rate of E. coli bacterial strains on different media, and potentially can be a measure of protein biosynthesis rate, to be determined in future. Expression is higher than of inducible promoters.

You can find this part's sequence with more information in registry: BBa_K1741012.

sfGFP under proC promoter with a hairpin in 5'UTR BBa_K1741013

A weak hairpin introduced to 5’UTR slightly lowers transcription/translation. Will be extended to a thermometer, actually we see no temp. dependence.

You can find this part's sequence with more information in registry: BBa_K1741013.

OD600 was measured after 18h, 22h and 26h. M9 mimimal medium was supplemented with different sugars (0,4%) , glycerol (1%) and glycine (1%) as a control.

sfGFP under proD promoter - BBa_K1741014

A strong insulated constitutive promoter proD is stronger than proC. Expression profiles of proC and proD are similar on different media = both seem to be really constitutive and the expression level does depend on E. coli growth rate.

You can find this part's sequence with more information in registry: BBa_K1741014.

The fluorescence of E. coli cells was measured after 24h of growth on M9 mimimal medium suplemented with different sugars (0,5%)

Other comparisons

The fluorescence of E. coli cells was measured after 24h of growth on M9 mimimal medium suplemented with different sugars (0,5%).

The fluorescence was measured after 48h-growth on M9 medium with 0,5% of glucose. T7 promoter leaks, whilst reporter constructs with EcoFactory promoters show no fluorescence of sfGFP. The photograph shows sfGFP fluorescence under blue light.

Team:UAM Poznan/Parts

ABSTRACT