After we created the constructs that express all genes seperately and in a polycistronic frame, we characterized the associated enzyme performances.


  • prooving enzyme expression on SDS-PAGEs
  • showing the enzyme activity of Mdh in varying strains under different conditions
  • demonstrating the growth performance of a strain with the polycistronic methanol condensation plasmid to be better on high methanol concentrations compared to others
  • performing a complex 13C labelling experiment

Single Expression

To check the heterologous expression of mdh, hps, phi and xpk in E.coli, we inserted them into a vector downstream a strong lacI repressible T7 promoter. Therefore, we used the pSB1A30 plasmid that was built by Team Heidelberg 2014[1]. In this vector, you find the BioBrick scar downstream the promoter.

All four genes were cut with EcoRI and PstI and ligated into a pSB1A30 plasmid that was digested with the same enzymes. As these construct require a T7 polymerase, they were transformed into BL21 Gold (DE3). Thus, we created microorganisms that, after IPTG induction, expressed the four MCC genes separately. For testing, we ran SDS-PAGEs with cell pellets that should contain the enzymes.

BioBrick enzyme molecular weight [kDA]
BBa_K1585210 Mdh 40.70
BBa_K1585211 Hps 22.60
BBa_K1585212 Phi 19.94
BBa_K1585213 Xpk 92.63

Results of Single Expression

After successfully building the plasmids, we tested the expression of the four genes separately. The SDS-PAGEs show that each of the enzymes is present in the cell after IPTG induction. As a negative control, we used a BL21 Gold (DE3) strain that expresses RFP.

Aachen 15-08-10 Mdh, Hps, Phi f. Wiki 2.png
Mdh, Hps and Phi expressed in a pSB1A30 backbone on SDS page after IPTG induction with a RFP expression strain as negative control
all genes expressed in BL21 Gold (DE3), samples taken from 50 ml main culture in shake flasks after 6 h and 21 h after IPTG induction
Aachen 15-08-10 Mdh, Xpk f. Wiki 2.png
Mdh and Xpk in a pSB1A30 backbone on SDS page after IPTG induction with a RFP expression strain as negative control
all genes expressed in BL21 Gold (DE3), 2 ml samples taken from 50 ml main culture in shake flasks after 6h and 21 h after IPTG induction

Laboratory Notebook of Single Expression

Mdh Characterization

The expression of the Mdh was confirmed via a SDS-PAGE showing clearly visible bands at the expected weight of 40.7 kDa. In parallel to whole cells, supernatant of lysed cells and the respective cell fragments were checked via SDS-PAGE. The cell fragments showed the strongest bands indicating the formation of inclusion bodies containing Mdh when expressed under the control of a T7 promoter. To avoid this, we transformed different strains of E. coli (SHuffle T7 Express and C43) as alternatives to BL21 Gold (DE3) which are known for more efficient protein expression and also tested lower cultivation temperatures. Additionally, it was confirmed that all strains are able to grow on M9 medium because we made the experience that not every strain is able to grow on M9.

However, the variation of strains and cultivation temperatures showed no significant effect on the Mdh becoming incorporated into inclusion bodies.

To test the functionality of the expressed Mdh we used a modified version[2] of the colorimetric and fluorometric formaldehyde assay described by Nash[3]. The whole-cell samples of BL21 Gold (DE3) showed the highest formation of formaldehyde proving the strongest activity of the Mdh compared to the other strains. Neither in the respective cell fragments nor in the supernatant of the lysed cells a similarly high activity could be observed. This indicates that the Mdh functions best in its natural environment within the cell and the cell membrane possibly representing a barrier protecting the Mdh from the harsh assay environment.


  • Proving the catalytical activity of the Bacillus methanolicus methanol dehydrogenase 2 in E. coli via Nash Assay despite the formation of inclusion bodies

Expression verification & localization of the Mdh

The expression of the Mdh was verified by doing an SDS-PAGE. To localize the Mdh not only whole cells were used but also the fragments of lysed cells and the respective supernatant. The expected bands were clearly visible in all samples proving the expression of the Mdh. However, the Mdh specific band had the highest intensity in the sample of the cell fragments whereas the same band was only of slight intensity in the supernatant. This indicates that the Mdh is incorporated into insoluble inclusion bodies.

Aachen 15-09-03 MDH Assay expression check + new constitutive MDH + Poly.png
Expression test of #IGEM#
An SDS-PAGE of the whole cell, the cell fragments and the lysate supernatant of the Mdh expressing BL21 Gold (#IGEM#) and the GlgC expressing BL21 Gold (#OHES#) as a negative control was run. The expected band at the weigth of the Mdh can be clearly seen in all #IGEM# samples. The according band with the strongest intensity occurs in the lane with #IGEM# fragments indicating the formation of inclusion bodies containg the Mdh.

To avoid the incorporation of the Mdh into inclusion bodies, the alternative E. coli strains SHuffle T7 Express and C43 were used. SHuffle T7 Express allows more efficient protein folding in the cytoplasm and lacks proteases whereas C43 allows the expression of toxic proteins. Additionally, cultivation at a lower temperature of 30°C was tested. It was shown that all strains were able to grow on M9 medium, however, another test for the expression and localization of the Mdh did not reveal major differences between the strains. In every case the Mdh was still incorporated into inclusion bodies.

Functionality of the expressed Mdh

To test if the expressed Mdh is functional, we modified a colorimetric and fluorescent formaldehyde assay described by T. Nash. In the presence of formaldehyde the yellow and fluorescing diacetyl-dihydro lutidine is formed. In the reaction catalyzed by the Mdh methanol is converted to formaldehyde which can be detected by the mentioned assay.

First, the different cells as well as their respective fragments and lysate supernatants were screened for formaldehyde production. The samples were taken 6 h after induction from shake flask cultures. Several cells, fragments and supernatants showed a production of formaldehyde but the Mdh activity in the intact cells was higher for each strain.

The formaldehyde assay was repeated only with whole cells and additional samples taken 20 h after induction. Moreover, the assay was conducted not only at 37°C but also at 30°C. The GlgC expressing BL21 Gold was used as a negative control. Again, in several strains functional Mdh could be detected and some general conclusions could be drawn:

  • More formaldehyde was produced in the samples taken 6 h after induction
  • Strains cultivated at 37°C show a stronger response than the same ones cultivated at 30°C
  • The assay works better and faster at an incubation temperature of 37°C
Aachen Comparison strains @ different conditions.png
Comparison of different strains at varying cultivation conditions.
The highest activity could be shown in BL21 Gold (DE3) at a cultivation temperature of 37°C in M9 in a sample taken 6 h after induction. Fluorescence measured at a gain of 75.

By far the highest formaldehyde production was observed in the BL21 Gold (DE3) cells 6 h after induction cultivated at 37°C despite its formation of inclusion bodies.

To show the significance of the production of formaldehyde the assay was conducted with the Mdh expressing BL21 Gold (DE3) cultivated in the labled methanol experiment along with the BL21 Gold (DE3) expressing GlgC as a negative control, both in multiple replicates.

Aachen Comparison IGEM vs OHES new.png
Formaldehyde formation during in vivo assay
Expression of Mdh and glgC as a negative control (4.5 h after induction). Both constructs were incorporated in vector pSB1A30 and the BL21 strains were cultivated at 37°C. Each value is made from at least 35 replicates of each construct. Fluorescence measured at a gain of 100.

The Mdh expression strain showed significantly more formaldehyde production indicating a functional expression of the Mdh.

Proof of Mdh Activity in Methanol Conversion Strain

As the Mdh represents the bottleneck of the whole MCC pathway we tested its activity in the strains with the polycistronic plasmid. For this the assay to detect formaldehyde first described by Nash was done with these strains. Whole cells, cell fragments and lysate supernatants were tested for the polycistronic strains. The highest formation of formaldehyde could be observed in the assay using whole cells.

Aachen Nash Poly.png
Activity of the Mdh in strain harboring the polycistronic construct
The strain with the polycistronic plasmid (#POLY#) shows significantly more Mdh activity than the negative control (#OHES#) expressing GlgC and the blank consisting of Nash reagent and buffer. Fluorescence measured at a gain of 75.

It was proven that the Mdh is expressed functionally in the strains containing the polycistronic construct. However, if compared to the activity of the strain only expressing the Mdh it is rather low (for comparison see data above).

Conclusion & Discussion

The functional Mdh was successfully expressed in multiple strains. The strongest response in the formaldehyde assay was reached in BL21 Gold (DE3) cultivated at 37°C in M9 medium. Although inclusion bodies containing the Mdh are formed in BL21 Gold, a high activity in the intact cells could be observed. This suggests that either a sufficient amount of Mdh still remains functional and in solution or the enzymes on the surface of the inclusion bodies still have a catalytic activity. The fact that a higher activity is reached in whole cell samples compared to the cell fragments and the lysate supernatant might be due to the protective property of the intact cell membrane shielding the enzyme from the harsh assay conditions and keeping it in its native environment.

Laboratory Notebook of Mdh Characterization


After successfully building strains that genetically meet all conditions to metabolize methanol, we wanted to examine the growth performance of a strain that carries the engineered constructs. The strain with the polycistronic plasmid behind J23119 promoter and the one expressing the Mdh in pSB1A30 were cultivated on different media with varying concentrations of methanol. Our growth experiments were performed in shake flasks and BioLectors [4]. To measure the OD we used, among other things, the automated cell density monitoring devices of Aquila Biolabs.

First, we wanted to know how methanol in the medium affects the growth of E. coli in general. Therefore, we cultivated a BL21 Gold (DE3) strain with an empty pSB1KRDP backbone without and with different methanol concentrations on M9 medium. As expected, high concentrations of methanol in the medium clearly decelerate the growth of the organisms.

Aachen MeOH control growthrate.png
Influence of MeOH on maximal growth rate of pSBKRDP in BL21 Gold (DE3)
Error bars indicate standart error of linear regression trough averages of log(backscatter); n=4; 750 µl cultivation in BioLector for 23 h at 37°C with 900 rpm

Next, we wanted to test the overall growth performance of our polycistronic plasmid in BL21 Gold (DE3). In this experiment, a strain with the polycistronic construct, the Mdh expression plasmid and a control were grown on M9 medium. The best growth could be detected for the strain with the four genes in a polycistronic frame. This findings surprised us because the burden of expressing the four genes should inhibit growth in comparison to strains with less additional overexpressed genes. In later attempts, this results could not be reproduced and the strain with the polycistronic version of the methanol conversion genes grew worst compared to the others.

Aachen 0Mcomparison.png
Comparison of polycistronic plasmid in pSB1KRDP, mdh in pSB1A30 and control on M9 medium without methanol
All strains are BL21 Gold (DE3). The control carries an empty pSB1KRDP backbone; cultivation of 750 µl cell suspension in BioLector for 23 h at 37°C with 900 rpm; n=4

Subsequently, we were able to show that the lag phase of a strain with our polycistronic construct can be much shorter compared to a strain with the Mdh expression plasmid and a control with an empty pSB1KRDP backbone when it is grown on M9 medium with 1.6 M methanol. This amount of methanol equals 6.48 percent per volume of the fermentation broth. The polycistronic strain is the only one that grows at all on on this methanol concentration that actually is the highest concentration where we ever observed any growth.

Aachen 1.6Mcomparison.png
Comparison of polycistronic plasmid in pSB1KRDP, mdh in pSB1A30 and control on M9 medium with 1.6 M methanol
all strains are BL21 Gold (DE3). The control carries empty pSB1KRDP backbone; cultivation of 750 µl cell suspension in BioLector for 23 h at 37°C with 900 rpm, n=4

In a following shake flask experiment, we saw that the strain with the methanol conversion plasmid reaches in general lower OD values in its stationary phase. But in contrast to the control, the strain grows to the highest densities at the highest methanol concentrations that were used in this experiment.

Aachen MethanolControlVSPolyOnline.png
growth comparisions on different methanol concentrations in 250 ml shake flasks of strain with polycistronic plasmid and conctrol with empty pSB1KRDP vector
methanol concentrations: 0 M, 0.75 M, 1.25 M; both strains BL21 Gold (DE3), experiment with 25 ml LB medium, methanol added after 2h, cultivation at 37°C with 250 rpm, OD600 measured with Aquila devices; n=3

Even though strains with our polycistronic plasmid grow best on high methanol concentrations compared to the others, methanol in the medium still affects the performance of it a lot. The strain grows to the highest densities in the shortest time, when there is no methanol at all in the medium. The lowest growth rate was detected when it was grown on 1.6 M methanol. Nevertheless, it is still remarkable that this strain has an exponential phase at all of these methanol concentrations.

Aachen FlowerPlateSeries.png
Growth of polycistronic construct in BL21 Gold (DE3) at different methanol concentrations
cultivation of 750 µl cell suspension in BioLector for 23 h at 37°C with 900 rpm on M9 medium with methanol concentrations of 0.0 M, 0.75 M, 1.25 M and 1.6 M (n=4 [c: 0.0 / 1.25 M MeOH]; n=3 [c: 0.75 M MeOH/ c: 1.6 M MeOH])


  • measuring the toxicity of methanol on the growth rate of E. coli BL21 Gold (DE3) without any additional genes by calculating the respective growth rate
  • showing that the strain with the polycistronic methanol conversion plasmid was able to start growing on methanol concentrations when other strains were not
  • demonstrating that the strain with the polycistronic construct does't grow to the same densities as control strains (probably due to the high burden of four addional expressed genes)
  • showing that the strain with the polycistronic plasmid shows an exponential growth phase at each tested methanol concentration

Laboratory Notebook of Physiology

Labeling Experiment

This really cool 13C labeling experiment was done at the research center 'Forschungszentrum Jülich', where Prof. Dr. Wiechert offered all necessary equipment and components to us. Jannick Kappelmann instructed two iGEM Team Aachen members (Jonas R. & Tobias S.) during the whole experiment.


A batch fermentation with 13C labeled methanol as an additional carbon source was done to demonstrate the functionality of our methanol assimilation pathway. Afterwards 13C labeled carbon atoms were measured via mass spectrometry.

Two different constructs were tested in E. coli BL21 Gold (DE3).

Aachen MCC.png
Methanol Condensation Cycle scheme
Sugars are the first metabolites that will contain 13C if the pathway works successfully. The initial step is the assembly of formaldehyde.

One of them was the polycistronic expression cassette and the other one was the mdh behind a IPTG inducable T7 promoter in pSB1A30.

Different amounts of methanol were added to each reactor to make the polycistronic construct comparable with itself to generate more independent data.

  • Offgas analysis of one cultivation with the mdh construct was done constantly during the fermentation. Ideally, it detects labled carbondioxide which is produced via the detoxification pathway of E. coli as a response to the formaldehyde produced by the Mdh.
  • The samples for analysis of cytosolic metabolites were taken after one doubling time (ca. 2:25 h).
  • The analysis was done by mass spectrometry of fructose-bis-phosphate, glucose-6-phosphate and ribose-5-phosphate.

The MS-detection of cytosolic metabolites covered additionally DHAP, Xylulose-5-phosphate, Fructose-6-phosphate, E4P, 2Phosphoglycerat/3Phosphoglycerate, S7P, AMP & ADP. Nevertheless, these metabolites were not analysed because of natural isotope distributions, coelutions or too low concentrations.

Aachen Reactors overview 13C.jpg
Experimental Setup
Four 200 ml reactors for cultivation in the 13C experiment.


  • We proved at methanol is not assimilated into the central metabolism of E. coli BL21 Gold (DE3) when the bacteria carries the polycistronic plasmid of the four missing enzymes to complete the MCC.
  • Despite no detectable production of 13C carbondioxide in the offgas we showed significant Mdh activity with samples from the respective reactor (see Mdh Characterization)

Expected Results

3 different sugars were analyzed: fructose-bis-phosphate, glucose-6-phosphate and ribose-5-phosphate.

  • Expected mass shift if 13C,d4 is assimilated: methanol - m+5, formaldehyde - m+3
    • all other sugars in the subsequent pathway should have a shift of m+3 in the first cycle of the functional MCC.
  • Expected mass shift if 13C is assimilated: methanol - m+1, formaldehyde - m+1
    • all other sugars in the subsequent pathway should have a shift of m+1 in the first cycle of the functional MCC.


In no sample of the polycistronic clones was a mass shift of m+3 observable. The detected peak areas are similar to the analytic standard samples of 12C carbohydrates. The mass shifts m+1 & m+2 are as frequent as it is usual in the natural composition.

Aachen 13C analytstanards plot.png
Plot of natural FBP, G6P & R5P as control
This figure shows the natural mass shift composition of the measured metabolites.
  • The same is valid for reactor 6 with (#AW9K#) our polycistronic construct with 0.3 M methanol.

Conclusion & Discussion

It was not possible to show a completely working "Methanol Condensation Cycle" [5] (MCC) in E. coli.

Although no labeled cytosolic metabolites could be detected we proved qualitatively the functionality of the Mdh. This leads to the assumption that the activity of the Mdh is too low to incorporate enough formaldehyde. Another possible explanation is that subsequent enzymes of the pathway do not function properly. The measured metabolites give no hint which of the implemented steps of the MCC does not work because the measured carbohydrates are only labeled if the assimilated molecule has passed the whole MCC.

The Xpk should have no influence on the assimilation of formaldehyde because without this enzyme the pathway equals the Ribulose-Monophosphate-Pathway (RuMP) which was already shown as a functional pathway if it is implemented in E. coli[6].

Laboratory Notebook


  1. Backbone Collection of Team Heidelberg 2014
  2. Kleeberg & Klinger. 1982. Sensitive formaldehyde determination with Nash's reagent and a tryptophan reaction. Journal of Pharmacological Methods, 8(1), 19-31.
  3. Nash. 1953. The colorimetric estimation of formaldehyde by means of the Hantzsch reaction. Biochemical Journal, 55(3), 416-421.
  5. Bogorad IW, Chen CT, Theisen MK, Wu TY, Schlenz AR, Lam AT, Liao JC. Building carbon-carbon bonds using a biocatalytic methanol condensation cycle. Proc Natl Acad Sci U S A. 2014 Nov 11;111(45):15928-33. doi: 10.1073/pnas.1413470111. Epub 2014 Oct 29. PubMed PMID: 25355907; PubMed Central PMCID: PMC4234558.
  6. 1. Metab Eng. 2015 Mar;28:190-201. doi: 10.1016/j.ymben.2014.12.008. Epub 2015 Jan 14.Engineering Escherichia coli for methanol conversion. Müller JE(1), Meyer F(1), Litsanov B(1), Kiefer P(1), Potthoff E(1), Heux S(2), Quax WJ(3), Wendisch VF(4), Brautaset T(5), Portais JC(2), Vorholt JA(6).