Production of Glyoxylate from Glucose in Engineered Escherichia coli

Round 1

Reviewer 1 Report

This manuscript describes construction of 10 E. coli strains for improved production of glyoxylate, a chemical of industrial interest, and their evaluation in shake flask cultures. Although the authors clearly describe how they stepwise improve the glyoxylate titers to 2.4 g/L, the study has some fundamental flaws that need attention: (i) The authors do not mention the theoretical maximum yield of glyoxylate from glucose or which would be their ideal overall reaction, given that conversion of glucose to glyoxylate will result in a redox imbalance. (ii) I would be alarmed to find out that addition of 1.5% glyoxylate completely abolishes growth as shown in Figure 2. How do the authors think they can ever come to commercial relevant titers if their strain cannot withstand 15 g/L? (iii) The analytical data are limited to glucose, acetate, and glyoxylate (and intracellular oxaloacetate for EG_9 and EG_10). Are there no other products like organic acids such as succinate, malate, formate, or lactate produced? Is the carbon balance more or less closed? The authors do not mention that. They do refer to other studies that show accumulation of lactate and malate when boosting the glyoxylate shunt. (iv) Overexpression of pyc may be beneficial at this stage, but in essence it is competing with production of acetyl-CoA from pyruvate, which is the main carbon source for glyoxylate. (v) The strains consume glyoxylate. (vi) All fermentation data are from shake flasks, which will acidify. pH-controlled batch fermentations will probably give better results. 

 

 

The English generally is fine. The authors just need to check for some minor errors in the text like "The expression vector" ("expression vector") in Table 1, "related genes" ("competing genes"?) in line 206, "This model" ("approach"?) in line 370.

Author Response

Dear Editor of the MDPI Fermentation Journal

 

Dear Reviewers

 

Manuscript ID: fermentation-2404326

 

We would like to express special thanks to the reviewers and the editor for all comments. We have edited our manuscript according to the comments. All changes have been highlighted in yellow. We hope that we will make them suitable for acceptance.

 

Reviewer 1: 

1. The authors do not mention the theoretical maximum yield of glyoxylate from glucose or the ideal overall reaction, given that the conversion of glucose to glyoxylate results in a redox imbalance.

 

Thank you for the valuable comment. Calculation of the theoretical yield of glyoxylate from glucose has been included in lines 331–345. Theoretically, 1 mole of glucose generates 2 moles of glyoxylate. Additionally, we have compared the production of glyoxylate in E. coli EG_7 with the theoretical yield.

 

2. I would be alarmed to find out that addition of 1.5% glyoxylate completely abolishes growth as shown in Figure 2. How do the authors think they can ever come to commercial relevant titers if their strain cannot withstand 15 g/L

 

Thank you for the valuable comment. There is room to develop glyoxylate production of up to 15 g/L. In this study, the highest titer was 2.4 g/L which is much lower than the limit. Further studies are highly recommended to investigate methods for overcoming the toxicity of glyoxylate on E. coli growth. Possible solutions to overcome the toxicity may be the utilization of extractive fermentation or continuous fermentation to avoid the accumulation of glyoxylate in the culture. In addition, appropriate channel proteins woulod be improved to efflux extracellular glyoxylate. The discussion has been added to lines 193–199.

 

3. Analytical data were limited to glucose, acetate, and glyoxylate (and intracellular oxaloacetate for EG_9 and EG_10). Are there no other products, such as organic acids such as succinate, malate, formate, or lactate produced? Is the carbon balance more or less closed? The authors do not mention that. They do refer to other studies that show accumulation of lactate and malate when boosting the glyoxylate shunt. 

 

Thank you for the valuable comment. We have included the quantification of the side products, including acetate, lactate, and formate, in Table 2. We also did not detect malate in the culture (data not shown) since aceB and glcB gene disruption stopped malate production from glyoxylate. The text has been added to lines 344–345.

The text describing the carbon balance has been added to lines 331–345 which showed that carbon flow was 60.6% satisfied. Reference [44] was added to state the accumulation of malate in a previous study by glyoxylate cycle. We found no reference indicating the increase in lactate concentration by overexpression of the glyoxylate shunt. There is a reported paper stating that glyoxylate shunt consumed better pyruvate and acetyl CoA, therefore reducing lactate.

 

4. Overexpression of pycmay be beneficial at this stage, but in essence it is competing with production of acetyl-CoA from pyruvate, which is the main carbon source for glyoxylate. 

 

Thank you for the valuable comment. A possible solution is to overexpress acetyl-CoA synthetase to recover acetyl-CoA from acetate since acetate concentrations were remarkable during the fermentation (Figures 4A and B). The discussion has been added to lines 371–374.

 

5. The strains consume glyoxylate?

 

Yes, we have understood the challenges created by glyoxylate assimilation. Since the eda gene which functions in glyoxylate assimilation was not disrupted in this study, E. coli consumed glyoxylate to produce 4-hydroxy-2-oxoglutarate. The authors have discussed the assimilation of glyoxylate in lines 222–232 and 294–302.

The authors discovered that the expression of the pyc gene could decrease the assimilation of glyoxylate. This may be because low pyruvate accumulation by pyc expression decreases eda function. Since the eda gene plays an important role in the bacterial Entner–Doudoroff pathway, the eda gene disruption was not conducted to avoid further effects on cell growth. The discussion has been added to lines 243-245.

 

6. All fermentation data are from shake flasks, which will acidify. pH-controlled batch fermentations will probably give better results.

 

Thank you for the valuable comment. We have already tested to conduct the fermentation by pH-stat jar fermentation, but it was difficult to control the assimilation of glyoxylate at a high metabolic rate. Upscale optimization requires further in-depth investigation. We have added the text on the Jar fermentation to lines 381–382.

 

7. The English generally is fine. The authors just need to check for some minor errors in the text like "The expression vector" ("expression vector") in Table 1, "related genes" ("competing genes"?) in line 206, "This model" ("approach"?) in line 370.

 

 Thank you for the valuable comment. “The expression vector”, “related genes”, and “This model” have been changed to “Expression vector” in Table 1, “competing genes” in line 203, and “approach” in line 384.

Reviewer 2 Report

Glyoxylate is an important chemistry midbody and have a broad range of industrial applications. In the manuscript entitled “Production of glyoxylate from glucose in engineered Escherichia coli” the authors developed several engineered E. coli strains to enhance glyoxylate production from glucose. The study will be of interesting to readers. However, all the following comments should be addressed:

1.The correlation equation of OD600 and cell weight and how the equation was obtained should be presented in Materials and Methods.

2. L181, “All data were analyzed in duplicate”. In most cases, the experiment was repeated three times.

3. In Figures 2-5 the symbols were described in the text of the figures legend. Please, could you put them in figure, it would be easier to understand then.

4. “3.1. Toxicity of Glyoxylate on E. coli growth”, it seems that this part is not necessary, it would be better to delete.

5. As shown in Fig. 4E, just like E.G-10, E.G-9 also showed a slower growth and glucose consumption, can adding yeast exact improve E.G-9 growth and its glyoxylate production?

6. From Fig.3, it could be seen that E.G-4 has the highest glyoxylate production and productivities, why not just use E.G-4 as the original strain to overexpress genes? E.G-6, E.G-7, E.G-8 may be not necessary.

7. Why 10 times diluted was used for oxaloacetate analysis? According to Fig. S2, using original solution may be better.

8. A formal form should be used instead of Table.s1

Author Response

Dear Editor of the MDPI Fermentation Journal

 

Dear Reviewers

 

Manuscript ID: fermentation-2404326

 

We would like to express special thanks to the reviewers and the editor for all comments. We have edited our manuscript according to the comments. All changes have been highlighted in yellow. We hope that we will make them suitable for acceptance.

 

Reviewer 2:

1. The correlation equation of OD600 and cell weight and how the equation was obtained should be presented in Materials and Methods.

 

Thank you for the valuable comment. The text has been included from lines 166–173 to describe the method of quantification.

 

2. L181, “All data were analyzed in duplicate”. In most cases, the experiment was repeated three times.

 

Thank you for the valuable comment. In this study, most of the data provided in this study were within the margin of error. We believe that our data reflect reliable tendency.

 

3. In Figures 2-5 the symbols were described in the text of the figures legend. Please, could you put them in figure, it would be easier to understand then.

 

Thank you for the valuable comment. All symbols have been displayed in the corresponding figures. Accordingly, the authors have edited the figure legends.

 

4. “3.1. Toxicity of Glyoxylate on  coligrowth”, it seems that this part is not necessary, it would be better to delete.

 

Thank you for the valuable comment. Figure 2 has been changed to Figure S1. A short description of the toxicity of glyoxylate has been added to lines 193–199.

 

5. As shown in Fig. 4E, just like E.G-10, E.G-9 also showed slower growth and glucose consumption, can adding yeast extract improve E.G-9 growth and its glyoxylate production?

 

Yes, yeast extract can support the growth of most disruptants and glyoxylate production since the yeast extract can supply amino acids for E. coli. A description that yeast extract supports for cell growth has been included in line 330.

 

6. From Fig.3, it could be seen that E.G-4 has the highest glyoxylate production and productivities, why not just use E.G-4 as the original strain to overexpress genes? E.G-6, E.G-7, E.G-8 may be not necessary.

 

Thank you for the valuable suggestion. The former EG_6, EG_7, and EG_8 have been deleted from the manuscript. Figure 3 has been modified to eliminate EG_6, EG_7, and EG_8. Accordingly, EG_9 and EG_10 have been changed to EG_6 and EG_7, respectively.

 

7. Why 10 times diluted was used for oxaloacetate analysis? According to Fig. S2, using original solution may be better.

 

Thank you for the valuable comment. The dilution samples were used to ensure that the results were within the range of the quantification (0–2.5 g/L). We have added the explanation to lines 186–188.

 

8. A formal form should be used instead of Table.s1

 

Thank you for the valuable suggestion. We have edited Table S1 to be a formal form.

 

Round 2

Reviewer 2 Report

The manuscript has been sufficiently improved, I have no more questions.