ALA to PPIX
One of the aims of this project was to re-engineer E.coli to produce chlorophyll a which E.coli does not naturally produce. It produces heme which shares the same precursor as a chlorophyll a, Protoporphyrin IX (PPIX) (figure 1) (Willows, 2004). The PPIX pathway begins from aminolevulinic acid (ALA), which is produced naturally in E.coli but can be also be added to increase the production of PPIX (Willows, 2004).
Fig 1: Biosynthetic pathway from ALA to the first coloured intermediate protoporphyrin IX. a. ALA dehydratase; b. porphobilinogen deaminase; c. uroporphyrinogen III synthase; d. uroporphyrinogen decarboxylase; e. coproporphyrinogen oxidase; f. protoporphyrinogen oxidase (Willows, 2004) .
Our initial aim was to model the pathway from ALA to PPIX in native E. coli cells to determine what concentration of ALA would give us the optimal yield of PPIX. Ultimately, we plan to model the entire Chlorophyll a biosynthesis pathway, and the production of H2 gas.
Mathematical Model Overview-
Michaelis-Menten type enzyme kinetics were studied first prior to establishing appropriate parameters from reported literature to propose our models. The adopted kinetic equation was then used to determine enzyme concentration through first-hand experiments and then fedback on our model which contains an arbitrary amount of enzyme concentrations.
The initial velocity of the reaction (Vi) directly depends on the rate of the conversion of ES to P. It is also dependent on the total enzyme concentration (Keleti & Kramer, 1986). By using the following formula, the total enzyme concentration can be determined from the experimental values of Vi and S, which also corresponds to initial ALA concentration. (Keleti & Kramer, 1986)
- ALA -δ-aminolevulinic acid
- PPBS -porphobilinogen synthase
- PPB -porphobilinogen
- PPBD -porphobilinogen deaminase
- HMB -hydrymethylbilane
- UROS -uroporphyrinogen III synthase
- URO -uroporphyrinogen III
- UROD -uroporphyrinogen decarboxylase
- CPO -coproporphyrinogen
- CPOO -coproporphyrinogen oxidase
- POIX -protoporphyrinogen IX
- POO -protoporphyrinogen oxidase
- PPIX -protoporphyrin IX
In our modelling experiment, the initial concentration of ALA, [ALA]0, was used as a known independent variable while initial enzyme concentration, [E]0, was used as a dependent variable . We assumed that all the substrates and products had a concentration of 0 at time 0 sec (t=0). [ALA]0 was varied from 1mM to 5mM to 10 mM to 20mM. [E]0 was arbitrarily chosen in a realistic interval of 0.1 µM to 1 µM. This is the natural concentration of enzyme seen in E.coli . For this first model, we assume that there would be no inhibition of any enzymatic activity and that ALA would only be converted to PPIX only and it only follows this biosynthesis pathway shown below.
In this model we considered the following reaction for ALA to PPIX transformation:
Table 1- Table displaying enzyme properties involved.
The above equations were used to obtain the following graphs via Matlab, and was adopted for all the intermediates. In this equation, S1 denotes the substrate and S2 denotes the intermediate produced from that substrate. For example, for the first step of reaction, ALA to porphobilinogen, S1 denotes ALA and S2 denotes porphobilinogen;for the following step of the reaction, porphobilinogen to hydroxymethylbilane, S1 denotes porphobilinogen and S2 denotes hydroxymethylbilane.
Graph 1: The above graph was plotted using initial ALA concentration of 1mM and 0mM of the remaining intermediates, enzyme and products involved. From the graph, it was observed that [ALA] became 0 at an accelerated rate and [PPIX] reached its peak value in approximately 0.7h . Combined values of Kcat for all the intermediates contribute to the time for formation of PPIX. This peak [PPIX] value was 1/8th of [ALA]0, which was expected as the conjugation of eight ALA molecules yields one protoporphyrin IX (PPIX) molecule(Wachowska et al., 2011).
Graph 2: This consists of graph 1 zoomed in to view the change in concentration of the intermediates involved at the initial stage of the reaction. PPB was the first intermediate produced, thus it appeared first and began to decrease as HMB was produced. As Uroporphyrinogen III synthase has a very high Kcat (500/s) , URO production from HMB was almost instant, hence, HMB was rapidly consumed.
Based on this, we performed a preliminary experiment with measurements at time intervals of 5, 10, 20, 30, 60 and 120 minutes. However, no brown coloured solution was observed to indicate PPIX formation, which was further verified by nanodrop absorbance readings.
Based on this, we constructed a second model taking into account the intermediates consisting of enzyme complexes.
In order to estimate the quantities of the enzymes involved in the ALA to PPIX pathway, we measured the concentrations for each of the pathway’s intermediates.
The E. coli cultures were grown overnight with shaking at 300 rpm at 37°C. The cells were harvested by centrifugation, supernatant discarded. The cells were resuspended in 5 mL of 0.1 mM glucose in 1 mM ALA, then incubated in the dark for up to 14 days. Approximately 250 uL of cells were extracted each time over 24 hour intervals, and lysed with lysis buffer and neutralized with neutralization buffer (Sigma plasmid prep kit). The precipitates were separated by centrifugation for 10 mins, and the absorbance of the supernatant measured with the nanodrop (Thermoscientific Nanodrop 2000 spectrophotometer). The absorbance maxima of PPIX is 404 nm.
When the intermediates (enzyme complexes) were added, the following reactions were obtained:
The first reaction was translated into a mathematical equation to give:-
The second mathematical model was then based on these equations.
Fig 2 : 2 ALA molecules combine to produce 1 PPB molecule. 4 PGB molecules form 1 HMB molecule (Ryter & Tyrrell, 2000).
The stoichiometric relationships displayed in figure 2 were introduced into matlab and used to obtain the following graph:
Graph 3: This graph was obtained using 1mM [ALA]0 and 0.1 µM as initial enzyme concentrations. [PPIX] reached its peak after approximately 7 days. From our first experiment, a significant concentration of PPIX was observed after 1 week, and peaked after another 7 days (shown in experimental part).
The second experiment
The second experiment aimed to examine the effect of cell concentration on PPIX production. It was initially speculated that intracellular and extracellular concentration of ALA was homogeneous, and our model assumed that the cells occupied the entire space. That is, we assumed that the added ALA solution enters directly into the cells, which is not the case. Thus, we took this into consideration and remodeled our code to take into account the fact that cells only occupy a certain percentage of the total volume. Here are the explanations:
- 1 =solution with which we resuspended cells (ALA)
- Vi=volume of the media i
- V=volume total = V1 + V2
- xvol=proportion of cells in the media
- C1(t)= concentration at t
- ni(t)=molar quantity of the media i at t
- de=molar quantity which enter in cells between t and t+dt
- dc=molar quantity which is consumed between t and t+dt
- We know dc and we want to find de knowing that C1(t) = C2(t) at any time:
- LB Media (250 mL)
- E.coli cells
- lysis solution (200 µL doses)
- Binding buffer (350 µL doses)
- 0.1 mM glucose in 1 mM ALA (5 mL)
- Cell concentration - oxygen is a key requirement for this pathway. If the current oxygen levels were maintained while reducing the quantity of the cells that are present, more oxygen is available for each cell and could increase the rate of the reaction.
- ALA concentration - the substrate concentration greatly influences the final concentration of the product. By varying the concentration of ALA, an optimum value can be determined for the production of PPIX.
- Addition of Zinc, Magnesium and PBS - ALA dehydratase catalyzes the first step of the PPIX biosynthesis from ALA. It exists mainly as a high activity octamer or a low activity hexamer (Lawrence, Ramirez, Selwood, Stith, & Jaffe, 2009). Magnesium acts as an allosteric regulator of ALA dehydratase and contains a binding site only in the octamer. As well as this, it has been demonstrated that the absence of magnesium promotes the formation of these low activity hexamers (Breinig et al., 2003) However, ALA dehydratase purifies with eight Zn2+ ions per octamer(Jaffe, Martins, Li, Kervinen, & Dunbrack, 2001).
- This suggests that the addition of both zinc and magnesium would promote the formation of the more active octamers and hence increase the rate of the first step of the PPIX biosynthesis
- Performed an in vitro experiment: this allows us to check the assumption [ALA]int = [ALA]ext
- LB Media (250 mL)
- E.coli cells
- Lysis solution (20 µL)
- ALA (20 mM)
- glucose (20 mM)
- PBS solution
- ZnCl2 (10 µM)
- MgCl or MgSO4 (2 mM)
- Alwan, A. F., Mgbeje, B., & Jordan, P. M. (1989). Purification and properties of uroporphyrinogen III synthase (co-synthase) from an overproducing recombinant strain of Escherichia coli K-12. Biochem. J, 264, 397-402.
- Boynton, T. O., Daugherty, L. E., Dailey, T. A., & Dailey, H. A. (2009). Identification of Escherichia coli HemG as a novel, menadione-dependent flavodoxin with protoporphyrinogen oxidase activity. Biochemistry, 48(29), 6705-6711.
- Breckau, D., Mahlitz, E., Sauerwald, A., Layer, G., & Jahn, D. (2003). Oxygen-dependent coproporphyrinogen III oxidase (HemF) from Escherichia coli is stimulated by manganese. Journal of Biological Chemistry , 278(47), 46625-46631.
- Breinig, S., Kervinen, J., Stith, L., Wasson, A. S., Fairman, R., Wlodawer, A., . . . Jaffe, E. K. (2003). Control of tetrapyrrole biosynthesis by alternate quaternary forms of porphobilinogen synthase. Nature Structural & Molecular Biology, 10(9), 757-763.
- Camadro, J.-M., Matringe, M., Scalla, R., & Labbe, P. (1991). Kinetic studies on protoporphyrinogen oxidase inhibition by diphenyl ether herbicides. Biochem. J, 277, 17-21.
- Dixon, J. M., Taniguchi, M., & Lindsey, J. S. (2005). PhotochemCAD 2: A Refined Program with Accompanying Spectral Databases for Photochemical Calculations¶. Photochemistry and photobiology, 81(1), 212-213.
- Jaffe, E. K., Martins, J., Li, J., Kervinen, J., & Dunbrack, R. L. (2001). The molecular mechanism of lead inhibition of human porphobilinogen synthase. Journal of Biological Chemistry, 276(2), 1531-1537.
- Keleti, T., & Kramer, M. (1986). Basic enzyme kinetics: Akademiai Kiado Budapest.
- Lawrence, S. H., Ramirez, U. D., Selwood, T., Stith, L., & Jaffe, E. K. (2009). Allosteric inhibition of human porphobilinogen synthase. Journal of Biological Chemistry, 284(51), 35807-35817.
- Maneli, M. H., Corrigall, A. V., Klump, H. H., Davids, L. M., Kirsch, R. E., & Meissner, P. N. (2003). Kinetic and physical characterisation of recombinant wild-type and mutant human protoporphyrinogen oxidases. Biochimica et Biophysica Acta (BBA)-Proteins and Proteomics, 1650(1), 10-21.
- Ryter, S. W., & Tyrrell, R. M. (2000). The heme synthesis and degradation pathways: role in oxidant sensitivity: heme oxygenase has both pro-and antioxidant properties. Free Radical Biology and Medicine , 28(2), 289-309.
- Stamford, N. P. J., Capretta, A., & Battersby, A. R. (1995). Expression, purification and characterisation of the product from the Bacillus subtilis hemD gene, uroporphyrinogen III synthase. European Journal of Biochemistry , 231(1), 236-241.
- Willows, R. (2004). Chlorophylls.
- Wachowska, M., Muchowicz, A., Firczuk, M., Gabrysiak, M., Winiarska, M., Wańczyk, M., . . . Golab, J. (2011). Aminolevulinic acid (ALA) as a prodrug in photodynamic therapy of cancer. Molecules, 16(5), 4140-4164.
Thereby, we obtained the following graph with an enzyme concentration of 0.3 µM by taking into account of the above cell percentage:
Graph 4- Concentrations of product and intermediates produced over time using enzyme concentration of 0.3 µM, 1mM initial ALA concentration and considering effects of cell concentration.
It was observed from preliminary experiments that certain key factors such as ALA concentration, co-factor metal ion concentrations, cell number could significantly increase the production of PPIX. With this in mind, several experiments were performed with the aim of optimising the production of PPIX. The preliminary experiment involved the resuspension of the E.coli cells in a solution of 1mM ALA. The formation of PPIX is indicated by a colour change of the media from colourless to brown, taking approximately one week to be detectable by eye. Several assumptions were made and then new experiments were carried out with slight changes to the original experiments including a decrease in cell concentration, varying the ALA concentration and the addition of zinc and magnesium to the cell culture.
In order to quantify the activity of the enzymes involved in the ALA to PPIX pathway, we measured the concentrations of the intermediates within in the pathway.
The E. coli cultures were grown overnight with shaking at 300 rpm at 37°C. The cells were harvested by centrifugation, supernatant discarded. The cells were resuspended in 5 mL of 0.1 mM glucose in 1 mM ALA, then incubated in the dark for up to 14 days. 250 uL of cells were extracted each time over 24 hour intervals and lysed with lysis buffer (Sigma plasmid prep kit) and neutralized (buffer). The precipitates were separated by centrifugation for 10 mins, and the absorbance of the supernatant measured with the nanodrop (Thermoscientific Nanodrop 2000 spectrophotometer). The absorbance maxima of PPIX is 404 nm.
Graph 5- PPIX was observable after approximately one week of incubation and continued to increase for a further 6 days. The theoretical absorbance value for a complete conversion of ALA to PPIX was calculated to be 2.0625. At 14th day, 2.0625 mM concentration of PPIX was produced.
As a result of conducting these experiments, it was established that 14 days was required to produce a significant quantity of PPIX. Optimal PPIX concentration was produced in 14 days. Considering the first traces of PPIX were detected 1 week after the beginning of the experiment, several possible optimisation strategies were examined:
SECOND EXPERIMENT (from 03/08)
As described previously, we measured the concentrations for each of the pathway’s intermediates to quantify the activity of the enzymes involved in the ALA to PPIX pathway.
The E. coli cultures were grown overnight with shaking at 300 rpm at 37°C. The cells were harvested by centrifugation, supernatant discarded. The cells were resuspended in a solution of phosphate buffered saline, Zinc(10µM) and Mg(2mM), then incubated in the dark for up to 19 days. The cultures were adjusted to an OD of 0.2 and 0.02 to observe the effect of cell concentration on PPIX output. Four concentrations of ALA were tested against these conditions as follows:
Table 2- Showing different initial concentrations of ALA added to different concentrations of E.coli cell culture.
Approximately 25 uL of cells were extracted each time over 24 hour intervals, and lysed with lysis buffer (20 uL) (Sigma plasmid prep kit) and neutralized (neutralisation buffer). The precipitates were separated by centrifugation for 10 mins and the absorbance of the supernatant measured with the nanodrop (Thermoscientific Nanodrop 2000 spectrophotometer).
Graph 6 - Shows concentration of PPIX produced from various initial concentrations of ALA for E.coli cell culture concentration of 0.2mM. The most PPIX was produced from an initial ALA concentration of 5 mM as greater concentrations resulted in substrate inhibition (shown in figure 2 below).
Graph 7 - Shows concentration of PPIX produced from various initial concentrations of ALA for E.coli cell culture concentration of 0.02mM. The most PPIX was produced from an initial ALA concentration of 5 mM as greater concentrations resulted in substrate inhibition (shown in figure 2 below).
An initial ALA concentration of 5mM produced the optimum concentration of PPIX. This is likely to be due to substrate inhibition at greater concentrations where Protoporphyrinogen IX inhibits the conversion of ALA to porphobilinogen via porphobilinogen synthase (Stamford, Capretta, & Battersby, 1995).
Moreover, the change in cell concentration did not change the production of PPIX, indicating that oxygen availability did not greatly influence the production of PPIX. The addition of glucose resulted in a subtle increase in PPIX production. As such, it cannot be deduced that glucose significantly impacts the production of PPIX. A slight difference was observed between the experimental PPIX formation and the predicted PPIX formation. We predicted the formation of 0.12 mM PPIX (Graph 1) from 1 mM ALA but only 0.065 mM of PPIX was produced (Graph 7 ).
Fig 3- Protoporphyrinogen IX inhibiting the conversion of ALA to porphobilinogen via porphobilinogen synthase (Stamford, Capretta, & Battersby, 1995). This is a process of substrate inhibition which allows a certain concentration of ALA to produce an optimum amount of PPIX which in our case was 5mM (Stamford, Capretta, & Battersby, 1995).
Our experiments enabled us to improve our understanding of how the ALA to PPIX pathway operates. A substantial amount of PPIX was produced from ALA and the ratio of ALA:PPIX concentration was 1:8 as expected (Wachowska et al., 2011). It was speculated that ALA enters cells via passive transport, as no excess energy was needed to introduce ALA to E.coli. Oxygen availability did not have a significant impact on PPIX production as the change in concentration of cells little effect. Furthermore, due to substrate inhibition, an initial ALA concentration of 5mM resulted in the greatest yield of PPIX compared to the other concentrations used. Enzyme concentrations were not determined due to the excessive length of the experiments. This was also the case with other concentrations such as those of the intermediates.