Limiting nitrous oxide production in Pseudomonas putida
Limiting nitrous oxide production in Pseudomonas putida
As part of the ammonia pillar, we aimed to reduce the nitrous oxide (N2O) production associated with the HNAD process. N2O is harmful for the environment. It is known that external factors such as: substrate type, organism, pH and carbon /nitrogen ratio influence the accumulation of N2O . Moreover, in some conditions, electrons are allocated unevenly to the four denitrification enzymes, leading to accumulation of nitrous oxide . To study whether N2O accumulates in P. putida we first expressed the denitrification machinery from P. stutzeri.
P. putida EM42 is strictly aerobic, which means that it uses terminal oxidases to dissipate electrons through oxygen respiration. To ensure N2O is not accumulated due to an insufficient electron supply, we attempted to redirect the electron flux towards the denitrification machinery. This was done by CRISPR interference (CRISPRi) downregulation of P. putida’s terminal oxidases. Because of this downregulation, less electrons are consumed in oxygen reduction leaving more electrons available to fuel the denitrification machinery.
We successfully measured nitrite accumulation and detected traces of produced nitrous oxide, suggesting that at least three enzymes of the integrated machinery worked. However, due to time constraints, we were unable to test the effect of CRISPRi downregulation on nitrous oxide production.
Both nitrification and denitrification processes (i.e. HNAD) can contribute to the formation of nitrous oxide. Nitrous oxide (N2O) is the penultimate compound in the to-be-engineered pathway (Figure 1) and is considered detrimental to the environment. N2O exhibits a global warming potential that is 300-fold that of CO2. Moreover, it can lead to depletion of the ozone layer . Thus, accumulation and consequential escape of nitrous oxide should be prevented at all costs.
Multiple factors influence the efficiency of the HNAD system. For instance, more N2O is being produced if the dissolved oxygen (DO) is low, or when ammonia (NH3) and nitrite (NO2-) concentrations are high . Moreover, the substrate type, biomass type, pH and C/N ratio (among others) impact the production of the intermediate N2O . Another explanation for N2O accumulation, is that the distinct denitrification steps (Figure 2) influence each other through electron competition. Pan et al.  have shown that the competition arose when the supply of electrons did not meet the demand for electrons by the four reduction steps. During electron-limiting conditions, electrons were allocated differently to the denitrification enzymes. Nitrite reductase (Nir) was found to receive more electrons. The difference in allocation creates a discrepancy in reduction rates of NO2- and other downstream N-species (i.e. nitric oxide (NO) and (N2O), leading to N2O accumulation .
Electron balance in our denitrifying P. putida
Upon establishing the denitrification machinery into P. putida EM42 ΔnasT, electron transfer pathways will resemble to those of the closely related species P. aeruginosa. Electrons are released through oxidation of organic compounds by respiratory dehydrogenases . In P. aeruginosa, electrons can be consumed in both aerobic and anaerobic respiration processes. Aerobic respiration in P. aeruginosa is identical to P. putida, with five terminal oxidases, catalyzing the four-step reduction of oxygen to water. Moreover, the respiratory chain further branches into the denitrification enzymes (Figure 2), which enable P. aeruginosa to grow anaerobically .
In the context of Cattlelyst, operated aerobically, both oxygen and nitrate will be used as oxidizing agents. This means that P. putida will dissipate electrons to oxygen and nitrate reduction at the same time. This process, referred to as co-respiration or aerobic denitrification is seen for both P. aeruginosa and P. stutzeri . However, the exact reasons why these closely related species co-respire remains elusive. A big known downside of aerobic denitrification is the higher risk for accumulating N2O .
A conceivable explanation could be the electron competition between the denitrification enzymes. If a microbe is co-respiring, part of the electrons is ‘lost’ in oxygen respiration and the remaining electrons are to be allocated between the four denitrification enzymes (Figure 2) The fact that another electron sink (oxygen respiration) is active could increase the possibility that the electron supply and consumption by denitrification are unbalanced, resulting in N2O production. Thus by generating a P. putida EM42 strain with the ability to co-respire, there is a potential to accumulate N2O , which, due to its harmful effects on the environment would defeat the purpose of Cattlelyst. Therefore, we dedicated this wetlab project to prevent N2O accumulation.
To reduce nitrous oxide production and redirect the electron flux towards the denitrification machinery we developed a wetlab project comprising two subprojects. In project 1 we downregulated the five terminal oxidases with CRISPR interference (CRISPRi). In project 2 we introduced the complete denitrification machinery from P. stutzeri into P. putida. After the initial cloning phase, we tested the effect of downregulation of the terminal oxidases on growth as well as whether the inserted denitrification machinery was functional.
Establishing the CRISPRi system for downregulation of terminal oxidasesarrow_downward
Why downregulate the terminal oxidases?
Pseudomonas putida is an obligate aerobic bacterium, which is not always practical in biotechnological applications . Thus far, considerable research has been made to make P. putida grow in anoxic conditions . Unfortunately, these efforts have not been successful, which illustrates the importance of oxygen respiration for P. putida. Therefore, we decided not to knock-out P. putida’s ability to respire with oxygen, but to downregulate the responsible enzymes. Batianis et al.  have shown that CRISPR interference can be employed to efficiently control transcription levels in P. putida. Additionally, they showed that multiple genes can be targeted at the same time, which a useful feature given that P. putida has five terminal oxidases.
We designed 15 different spacers, three for each terminal oxidase complex, to be inserted in pSEVA231-CRISPR  plasmids by means of Golden Gate cloning. The spacers were designed to target the non-template strand, right at the 3’ end of a Protospacer Adjacent Motive (PAM) sequence. For every terminal oxidase a spacer was designed for: (1) the promoter sequence, (2) sequence between the promote r and start codon and (3) start codon (Figure 3).
Figure 3: Spacer numbers corresponding to target
The pSEVA231-CRISPR plasmids were transformed into E.coli dh5α, and confirmed with sequencing. Subsequently, the plasmids were transformed into P. putida EM42 by electroporation.
Figure 4: example pSEVA231-CRISPR plasmid with spacer targeting startcodon of Cio.
Testing the effect of the CRISPRi plasmids on growth
P. putida EM42 strains with the pSEVA231-CRISPR plasmid were grown at 30°C overnight in 10 mL LB supplemented with 50 μg/ml kanamycin. 15 different targeting spacers were used to downregulate the different terminal oxidases and a non-targeting spacer was used as control.
The next day, the cultures were diluted to an OD600 of 0.2 in M9 supplemented with kanamycin and 30 mM acetate as carbon source to a total volume of 200 μl in 96-well plates. The plate was directly incubated at 30°C in the Synergy MX plate reader for 48 hours, with OD600 measurements every 5 minutes. For these experiments both biological and technical triplicates were included.
Plug and play approach: how did we integrate the complete denitrification machinery?arrow_downward
To be able to transport the complete machinery, we designed and developed a 42.5 kb bacterial artificial chromosome (BAC) that carried the four denitrification operons. Then, we prepared a ‘landing pad’ in P. putida EM42 ΔnasT such that it could accept a 32kb genomic insertion.
Given that the genomic insert would be around 32kb, the BAC was designed in a way that it could be assembled in yeast. Therefore, a centromere (CEN), an autonomous replicating (ARS), and Leucine biosynthesis (LEU) were included in the design . Leu was added because the yeast strain that would have been used is auxotrophic for leucine. Moreover, we used the PCC1fos backbone which is a low copy number plasmid. The rationale for this is that a large, high copy number plasmid >30kb can excessively burden bacteria and are prone to mutate . The denitrification machinery is flanked by two lox sites: lox66 BBa_K3747604 and lox2m/71 BBa_K3747605, and comprises from 5’ to 3’:
- A gentamycin resistance cassette
- The Periplasmic nitrate reductase (Nap) operon BBa_K3747600 from P. stutzeri with at the 5’ end a ribosome binding site (RBS):BBa_J34801
- The nitrite reductase (Nir) operon BBa_K3747601 from P. stutzeri with at the 5’ end an RBS: BBa_J34801
- The nitric oxide reductase (Nor) BBa_K3747602 operon from P. stutzeri with at the 5’ end a T7 promoter: BBa_K3633015, RBS: BBa_J34801
- The nitrous oxide reductase (Nos) operon BBa_K3747603 from P. stutzeri with at the 5’ end a RBS: BBa_J34801
Figure 5: The designed BAC carrying the complete denitrification machinery.
After the design phase, seven fragments were PCR amplified with 60-bp overhangs and assembled by Gibson Assembly to the final BAC (Figure 5) .
Then the BAC was transformed into E.coli EPI300, screened, isolated and checked with sequencing. After verification, the BAC was conjugated into P. putida EM42 ΔnasT. Subsequently, we verified that the pathway was successfully integrated with colony PCR and sequencing.
Preparing the landing pad in P. putida EM42 ΔnasT
To transport the complete denitrification machinery, we prepared a ‘landing pad’ in P. putida EM42 ΔnasT (Figure 6). The landing pad comprises a T7 promoter that controls the expression of Cre recombinase flanked by two lox sites: lox71 and lox66 2m. The landing was integrated at the PP_5388 site (downstream of the cusF gene) . Genomic integration in this locus specifically has been shown to be innocuous, moreover expression at this locus is low . As mentioned, the denitrification machinery and gentamycin resistance cassette on the BAC were flanked by compatible lox sites. The landing pad was integrated through homologous recombination with the assistance of the integrative vector pGNW . After successful conjugation of the BAC into P. putida EM42 ΔnasT, Cre recombinase recognized the lox sites and consequently integrated the denitrification machinery.
Figure 6: landing pad cargo for the integrative pGNW vector. With this construct, from 5’ to 3’, the T7 terminator, lox66, Cre recombinase, lox71 and the T7 promoter are integrated. This is because H1-cusF and H2-cusF correspond to regions at the 3’ and 5’ end of the PP_5388 site respectively.
How did we ensure transcription of 31.7 kb?
Given that we integrate 32 genes at once, in total 31,7kb, we wanted to make sure that the overall pathway was expressed. Therefore, we placed the denitrification pathway under the control of two T7 promoters located in the genome in front of lox71 and in front of the nor operon on the BAC. The T7 polymerase (T7pol) transcribes DNA only downstream of a T7 promoter. Given that P. putida does not encode for such RNA polymerase, we integrated it at the attn7 site (Figure 7). We chose the T7pol specifically because it synthesizes RNA at a high rate and it ignores terminators frequently . We prevented constitutive expression of the integrated denitrification machinery, and with that the associated metabolic burden by making T7pol transcription IPTG inducible.
Figure 7: IPTG inducible T7pol cargo for the integrative pGNW vector. With this construct, from 5’ to 3’, the lacI promoter, lacI, the lac operator, T7 polymerase and &lambda0 is integrated. This is because H1-glmS and H2-glmS correspond to regions at the 5’ and 3’ end of the attn7 site respectively.
The final strain with T7pol, the gentamycin resistance cassette and the denitrification machinery was called P. putida :SD, SD standing for Synthetic Denitrification.
Testing of P. putida :SD
To test whether the integrated Nap is active in P. putida EM42 :SD we used P. putida EM42 ΔnasT as negative control and P. putida EM42 ΔnasT + Nap operon from the mosaic approach as positive control. The specific strains were cultured on M9 medium with 30 mM acetate as C-source and 2 g/L (NH4)2SO4 as N-source supplemented with 5 mM NaNO3 as substrate for Nap. After 24 hours, the supernatant was taken and the NO2- concentration was quantified by performing a Griess Assay .
Quantifying N2O accumulation
In P. putida :SD, the complete denitrification machinery is present. The machinery is a concatenation of enzymatic reactions where the product of one enzymatic reaction is consumed by the next enzymatic step. For this wetlab project specifically, we aim to limit N20 accumulation. In this experiment, we validated the concatenated activity of Nap, Nir, and Nor by quantifying N2O production, with gas chromatography - mass spectrometry (GC-MS). For this experiment we compared P. putida :SD with and without IPTG to P. putida ΔnasT. The strains were cultured on M9 medium with 30 mM acetate as C-source and 2 g/L (NH4)2SO4 as N-source supplemented with 5 mM NaNO3 as substrate in sealed bottles air.
P. putida :SD with the Nos accessory plasmid was cultured in M9 with 30 mM acetate as C-source and 2 g/L (NH4)2SO4 as N-source in sealed bottles containing 21% (v/v) oxygen and 0.5% N2O. To test whether the Nos construct was active in vivo, the depletion of N2O was measured with GC-MS. According to our GC technician, produced N2 was too difficult to measure with GC-MS, as N2 is abundant in air and minor differences cannot be measured.
Quantifying growth defects by the CRISPRi downregulation of the terminal oxidases
Multiple growth experiments were performed to measure the effect of the spacers on growth. We noticed that there was a lot of variability between biological replicates. This could suggest that growth was affected by the spacer, but we could not know for sure. In the final experiment, we were able to test all 15 spacers, 2 or 3 biological replicates in technical duplicates.
After 48 hours, the curves showed that generally that the spacers (Sp) did not affect the growth of P. putida EM42 drastically (Figure 8). However, significantly longer doubling times were computed for spacers targeting the promoter of Cyo oxidase (Sp4), the promoter of the Aa3 oxidase (Sp7) and the startcodon of the Aa3 oxidase (Sp9). This does suggest that targeting the Cyo oxidase and Aa3 oxidase can impair growth, whereas the largest effect is found for targeting the promoter of the Cyo oxidase, which can be seen more clearly in Figure 9.
We also computed the time at which the exponential phase started. We found that for Sp4, targeting the promoter of the Cyo oxidase, the exponential phase was delayed most. Overall, the values for exponential delay were quite variable for the different spacers and therefore difficult to be conclusive alone. From these analyses we can conclude that targeting the promoter for the Cyo oxidase is most impactful in impairing growth.
This spacer most likely affects growth as it targets the Cyo terminal oxidase which has a significant effect on the cell transcriptome . Moreover, this oxidase is the only oxidase active during the exponential phase, which explains why we see the growth defect only during this growth stage (Figure 9). The other oxidases, i.e. the CIO oxidase, the Aa3 oxidase, the Cbb3-1 oxidase and Cbb3-2 oxidase are active during the stationary phase (Table 1). When other oxidases take over, which has been shown upon knocking out the Cyo oxidase, the effect of downregulation could have been lost.
|Terminal oxidases||When active||Effect knockout|
|Cyo oxidase||Exponential phase||34% decrease growth rate|
|CIO oxidase||Stationary phase||Negligible|
|Aa3 oxidase||Stationary phase||Negligible|
|Ccb3-1 oxidase||Stationary phase||Negligible|
|Ccb3-2 oxidase||Stationary phase||Negligible|
To further pinpoint which spacers are most effective in downregulating the terminal oxidases, we performed several quantitative PCR runs. However, due to insufficient time and resources we could not pursue this for all spacers. The next step is to develop a multiplex CRISPRi plasmid, targeting several terminal oxidases, such as the Cyo oxidase and Aa3 oxidase, at the same time. To test whether nitrous oxide production is can be reduced with our approach, the multiplex plasmid will be transformed into P. putida :SD. Unfortunately, due to time constraints, we were not able to combine the CRISPRi with the plug-and-play strain.
Can P. putida :SD denitrify?
We integrated the complete denitrification machinery in P. putida and tested whether the pathway worked by measuring nitrate reduction, nitrite accumulation, nitrous oxide accumulation. Nitrate reduction was measured to test the overall efficiency of the pathway. However, this assay was not sensitive enough to conclude whether the pathway was active.
Is Nap active in P. putida :SD?
To test Nap activity for P. putida :SD we performed a Griess assay to quantify nitrite accumulation (Figure 11). Additionally, we tested the effect of IPTG induction on the accumulation of nitrite. For this condition, :SD+, the denitrification pathway is expressed to a higher extend. With IPTG, overall expression of the system was higher, which was also reflected in the OD600. The higher metabolic burden resulted in a OD600 66% lower than that for :SD without IPTG.
Compared to the control P. putida ΔnasT, more nitrite accumulates for both :SD and :SD+, demonstrating that the integrated Nap works. However, when compared with the Nap plasmid (Nap from Paracoccus denitrificans), less nitrite accumulates. This could imply that (1) the nitrate reduction was not as optimal as for the Nap plasmid or (2) that Nir was active. Nir reduces nitrite to nitric oxide, this concatenated reduction step could explain the lower amounts of nitrite accumulated for :SD. The fact that :SD+, accumulated less nitrite compared to :SD could be explained by the higher expression of all denitrification genes increasing pathway efficiency.
Does nitrous oxide accumulate?
Given that we showed that Nap, being first enzyme in the pathway worked in P. putida :SD, we further tested nitrous oxide accumulation. Nitrous oxide accumulation would be the concatenated effect of Nap, Nir and Nor, and measuring its production verifies the activity of these enzymes.
Within the timeframe of iGEM, we were able to perform one GC-MS experiment for which we measured the N2O concentration at t0, t24, t48 and t139 in the headspace. Over the timeframe, no N2O accumulated for P. putida :SD. However, for two out of three biological replicates of P. putida :SD+ we noticed a slight increase in N2O (Figure 12). These preliminary results suggest that the combined Nap, Nir, Nor enzymes work as nitrous oxide is accumulating slightly. This accumulation could happen due to a lack of electrons, as illustrated by Pan et al. . This could also be explained by the fact that Nos is active in this strain and that it further reduces nitrous oxide, explaining only the small accumulation. However, to know for sure, this experiment needs to be repeated.
Can P. putida :SD reduce nitrous oxide?
Thus far, heterologous expression of Nos in vivo has not been successful , . Therefore, to test Nos activity separately, we expressed the Nos accessory plasmid, Nos accessory plasmid containing multiple putative electron donors for this enzyme.
Nos activity was tested by culturing P. putida :SD with 0.5% (5000 ppm) N2O in the headspace and subsequently measuring the reduction in N2O with GC-MS at t=0, t=24, t=48. Unfortunately, no significant reduction of N2O consumption was observed. We used P. stutzeri JM300 as a positive control to check if this native denitrifier could convert N2O into N2, but also no reduction of N2 was observed. This could indicate that the experimental conditions were not optimal for Nos activity, or that the reduction was so little that it could not be measured with the current GC-MS settings. Similar to the mosaic approach, within the time frame of iGEM it was not possible to repeat this experiment.
To reduce nitrous oxide production, we developed the strain P. putida :SD encoding for the complete denitrification machinery from P. stutzeri. To test the pathway, we measured if intermediates were accumulated. These tests showed that nitrite accumulated, and traces of N2O were detected, suggesting that the first three enzymes of the pathway are active. However, more research needs to be done to derive conclusions.
In-parallel we developed 15 CRISPRi plasmids to downregulate the terminal oxidases in P. putida. We learned that quantifying the effect of the spacers on growth was difficult. Biological replicates showed a lot of variability and the results obtained in different growth experiments were not consistent. We learned that for the spacer targeting the promoter of the Cyo oxidase (Sp4) growth was impaired. Unfortunately, this effect was lost soon, which could be explained by the upregulation of the other oxidases. To be able to pinpoint the most efficient spacers in downregulating terminal oxidases, we performed quantitative PCR experiments. However, due to insufficient time and resources, no conclusions could be drawn from these experiments. Our next step is to combine both projects and test the effect of the CRISPRi downregulation on nitrous oxide production.
- B.-J. Ni, M. El Ruscalleda, C. Pellicer-N Acher, and B. F. Smets, “Modeling Nitrous Oxide Production during Biological Nitrogen Removal via Nitrification and Denitrification: Extensions to the General ASM Models,”Environ. Sci. Technol, vol. 45, pp. 7768–7776, 2011, doi: 10.1021/es201489n.
- Y. Pan, B.-J. Ni, P. L. Bond, L. Ye, and Z. Yuan, “Electron competition among nitrogen oxides reduction during methanol-utilizing denitrification in wastewater treatment,” 2013, doi: 10.1016/j.watres.2013.02.054.
- B. MD, W. YY, C. E, and S. T, “Nitrous oxide emissions for early warning of biological nitrification failure in activated sludge,” Water Res., vol. 43, no. 5, pp. 1265–1272, 2009, doi: 10.1016/J.WATRES.2008.12.027.
- H. Arai, “Regulation and function of versatile aerobic and anaerobic respiratory metabolism in Pseudomonas aeruginosa,” doi: 10.3389/fmicb.2011.00103.
- D. Lloyd, “Aerobic denitrification in soils and sediments: From fallacies to factx,” Trends Ecol. Evol., vol. 8, no. 10, pp. 352–356, Oct. 1993, doi: 10.1016/0169-5347(93)90218-E.
- L. F. C. Kampers, J. J. Koehorst, R. J. A. van Heck, M. Suarez-Diez, A. J. M. Stams, and P. J. Schaap, “A metabolic and physiological design study of Pseudomonas putida KT2440 capable of anaerobic respiration,” BMC Microbiol. 2021 211, vol. 21, no. 1, pp. 1–15, Jan. 2021, doi: 10.1186/S12866-020-02058-1.
- A. Steen et al., “Construction and characterization of nitrate and nitrite respiring Pseudomonas putida KT2440 strains for anoxic biotechnical applications,” J. Biotechnol., vol. 163, no. 2, pp. 155–165, Jan. 2013, doi: 10.1016/J.JBIOTEC.2012.09.015.
- C. Batianis et al., “An expanded CRISPRi toolbox for tunable control of gene expression in Pseudomonas putida,” doi: 10.1111/1751-7915.13533.
- T. Aparicio, V. de Lorenzo, and E. Martínez-García, “CRISPR/Cas9-Based Counterselection Boosts Recombineering Efficiency in Pseudomonas putida,” Biotechnol. J., vol. 13, no. 5, p. 1700161, May 2018, doi: 10.1002/BIOT.201700161.
- N. Kouprina and V. Larionov, “Exploiting the yeast Saccharomyces cerevisiae for the study of the organization and evolution of complex genomes,” FEMS Microbiol. Rev., vol. 27, no. 5, pp. 629–649, Dec. 2003, doi: 10.1016/S0168-6445(03)00070-6.
- M. Miyahara et al., “Potential of aerobic denitrification by pseudomonas stutzeri TR2 to reduce nitrous oxide emissions from wastewater treatment plants,” Appl. Environ. Microbiol., vol. 76, no. 14, pp. 4619–4625, Jul. 2010, doi: 10.1128/AEM.01983-09.
- G. DG, Y. L, C. RY, V. JC, H. CA, and S. HO, “Enzymatic assembly of DNA molecules up to several hundred kilobases,” Nat. Methods, vol. 6, no. 5, pp. 343–345, 2009, doi: 10.1038/NMETH.1318.
- “Evaluation of chromosomal insertion loci in the Pseudomonas putida KT2440 genome for predictable biosystems design | Elsevier Enhanced Reader.” https://reader.elsevier.com/reader/sd/pii/S2214030120300195?token=E0DC4D1101BA00821983D4B4451E3CB13F22D7E61D0930AB0BE0D7F3BEEBF1A339F6993FB745C14BD0A1F0612C883EA3&originRegion=eu-west-1&originCreation=20211017181619 (accessed Oct. 17, 2021).
- N. T. Wirth, E. Kozaeva, and P. I. Nikel, “Accelerated genome engineering of Pseudomonas putida by I-SceI―mediated recombination and CRISPR-Cas9 counterselection,” Microb. Biotechnol., vol. 13, no. 1, pp. 233–249, Jan. 2020, doi: 10.1111/1751-7915.13396.
- T. S, “Expression using the T7 RNA polymerase/promoter system,” Curr. Protoc. Mol. Biol., vol. Chapter 16, no. 1, Jul. 2001, doi: 10.1002/0471142727.MB1602S11.
- A. Ugidos, F. Rojo, G. Morales, A. Ugidos, and F. Rojo, “Inactivation of Pseudomonas putida cyo terminal oxidaseG Inactivation of the Pseudomonas putida cytochrome o ubiquinol oxidase leads to a significant change in the transcriptome and to increased expression of the CIO and cbb3-1 terminal oxidases,” Environ. Microbiol., vol. 8, no. 10, pp. 1764–1774, 2006, doi: 10.1111/j.1462-2920.2006.01061.x.<.li>
- P. Wunsch, M. Herb, H. Wieland, U. M. Schiek, and W. G. Zumft, “Requirements for CuA and Cu-S center assembly of nitrous oxide reductase deduced from complete periplasmic enzyme maturation in the nondenitrifier Pseudomonas putida,” J. Bacteriol., vol. 185, no. 3, pp. 887–896, 2003, doi: 10.1128/JB.185.3.887-896.2003.
- L. Zhang, A. Wüst, B. Prasser, C. Müller, and O. Einsle, “Functional assembly of nitrous oxide reductase provides insights into copper site maturation,” Proc. Natl. Acad. Sci. U. S. A., vol. 116, no. 26, pp. 12822–12827, 2019, doi: 10.1073/pnas.1903819116.