A Plant-Based Artificial Haemagglutinin (A/H5N1) Strongly Induced Neutralizing Immune Responses in Mice
, Udo Conrad 3, Trang Huyen Vu 1,2 and Ha Hoang Chu 1,2,*Round 1
Reviewer 1 Report
In this study Pham et al designed and produced artificial haemagglutinin constructs representing A/H5N1 and H5.c2 strains. One of them, H5.c1 protein, was selected for purification and further characterization. Authors showed by cross-linking reaction and size exclusion chromatography trimeric structure of produced protein. Next they demonstrated that artificial H5.c1protein elicit HA-specific IgG and neutralizing antiboies.
The manuscript is difficult to read and sometimes incomprehensible. Authors explain trivial things e.g. basis for protein purification on IMAC. Many misspelled sentences e.g. line 391 "The difference in protein expression level between artificial H5.c2 and native H5Dk protein might be explained by the different in amino acid sequence because amino acids act as substrates for protein synthesis as well as protein expression". There is a repetition of text from line 186 to line 195. In my opinion the manuscript requires intensive work on content and style.
Table 1 should contain a description of what is color-coded.
I think that the authors should show the purity of the obtained proteins H5.c1 and H5TG on the gel stained with Commassie not only on Western blot.
The figure 5 is poor quality.
The specificity of the obtained antibodies should be tested on HA containing extracts not only on purified proteins (Fig 6b).
Author Response
Thank you for shepherding the review of our manuscript. We appreciate the thoughtful and constructive comments from all reviewers to improve the quality of the manuscript. We have now revised the manuscript to accommodate the points raised by the reviewers.
|
Reviewer 1:
Open Review (x) I would not like to sign my review report English language and style (x) Extensive editing of English language and style required
Comments and Suggestions for Authors - In this study Pham et al designed and produced artificial haemagglutinin constructs representing A/H5N1 and H5.c2 strains. One of them, H5.c1 protein, was selected for purification and further characterization. Authors showed by cross-linking reaction and size exclusion chromatography trimeric structure of produced protein. Next they demonstrated that artificial H5.c1protein elicit HA-specific IgG and neutralizing antiboies. The manuscript is difficult to read and sometimes incomprehensible. Authors explain trivial things e.g. basis for protein purification on IMAC. Many misspelled sentences e.g. line 391 "The difference in protein expression level between artificial H5.c2 and native H5Dk protein might be explained by the different in amino acid sequence because amino acids act as substrates for protein synthesis as well as protein expression". There is a repetition of text from line 186 to line 195. In my opinion the manuscript requires intensive work on content and style. Response: We corrected the sentence line 391 in the manuscript and deleted the text from the line 186 to line 195. We clarified some other parts in the manuscript.
- Table 1 should contain a description of what is color-coded. Response: We added the description for table 1 in the text from the line 277 to line 280 in the manuscript.
- I think that the authors should show the purity of the obtained proteins H5.c1 and H5TG on the gel stained with Commassie not only on Western blot. Response: We also showed the purity of the obtained proteins H5.c1 and H5TG on the gel stained with Commassie in supplementary data.
- The figure 5 is poor quality. Response: We enhanced the quality of the figure 5 in the manuscript.
- The specificity of the obtained antibodies should be tested on HA containing extracts not only on purified proteins (Fig 6b). Response: We thank reviewer for the comment. The main purpose of this experiment (Figure 6b) is to detect HA-specific IgG antibodies in mouse sera after immunization with IMAC-purified HA protein by Western blot. We have tried to use crude extract containing HA to detect specificity of HA-specific IgG antibodies, however the noise background is still high. Because we could not remove 100% of plant-derived protein from purifed H5 protein in IMAC purification, and antibody against plant-derived protein produced in mouse sera that can interact with plant antigens resulting to noise background in Western blot. Therefore, to reduce the noise background, we used SEC-purified H5 proteins that had higher purity than IMAC-purified H5 proteins as antigens to detect HA-specific IgG antibodies in mouse sera.
|
Reviewer 2 Report
In this manuscript the authors designed two artificial haemagglutinins using sequences from HA proteins from viruses detected in poultry in Vietnam. One of these proteins (H5.c1) showed good expression in Nicotiana benthamiana and induced strongly neutralizing antibodies in mice against homologous H5.c1 and 3 heterologous native strains in haemagglutination inhibition assays. Therefore, H5.c1 could be a promising vaccine candidate against A/H5N1 virusses.
However, I have some critical comments.
The authors used two methods to determine the oligomeric state of H5.c1 and concluded from both methods that the protein is trimeric. However, these two methods are not very suitable to determine accurate molecular weights of proteins in solution since they are not based on the mass but on the shape of the protein of interest. Therefore, no accurate conclusion can be drawn from the position of the largest band in the SDS-PAGE/Western blot or the position of the largest peak in the SEC elution profile. Alternative methods such as analytical ultracentrifugation and static light scattering in combination with SEC (SEC/MALLS) are better suited for this purpose.
On top of that, it can’t be stated that the top HA band runs at 201 kDa since the top marker band in the SDS-PAGE/Western blot analysis is only 170 kDa. A significantly higher molecular weight is more likely.
In the SEC experiment the largest peak of H5.c1 seems to overlap exactly with the ferritin marker (440 kDa), which would indicate a higher oligomeric state than the trimer mentioned in the text.
The authors should use another, more reliable method for the determination of the oligomeric state of H5.c1 (see above).
In Figure 5 the authors show the SEC elution profiles for two different HA proteins (H5TG and H5.c1). Based on the SDS-PAGE/Western analysis they pooled for both proteins the elution fractions B1-B6 (indicated with a red box in the elution profile). Here we can clearly see that although all fractions contain HA the pooled fractions are not homogenous. Especially in the case of H5.c1 a significant second peak is also pooled that most likely represents HA of a higher oligomeric state or even aggregated protein. This fraction is most likely less or even not at all functional and if present in the preparations used for functional studies will affect the results negatively. The shape of the H5TG peak also shows that the protein is not homogenous and also here aggregated protein is added to the preparation.
Other comments:
In par 2.3 more details should be given for the described method.
In par 2.4 more details should be given such as the volume of Ni-NTA resin used, with how many column volumes was the column washed/eluted, how were the proteins stored?
In par 2.4 did the authors test the different concentration of imidazole in the wash buffer of H5.c1 or was a sequencial series of wash steps used? If so, why going down in the imidazole concentration?
Did the authors also purify the second artificial HA protein, H5.c2?
In par 2.6 what was the running buffer of the SEC experiment?
The first paragraph 2.9 is exactly the same as paragraph 2.8 and can be deleted.
In par 3.2 the imidazole concentration in the elution buffer of H5 TG is 125 mM. In par 2.4 it is 50 mM. Please clarify.
In par 3.2 the authors mentioned that the different HA proteins contain a C-terminal His-tag. From Figure 2 it can be concluded that this should be an N-terminal His-tag.
The accuracy of the molecular weights obtained from the SDS-PAGE/Western blot gel is unrealistic (e.g. 201.45 kDa). Especially when these are supposed to be only approximations (~201.45 kDa).
In par 3.3 Figure 5c should be 6c.
In par 3.3 (see also Figure 6c) I don’t agree with the conclusion that the mean OD450 values of group 1 (H5.c1) are not significantly different from that of group 2 (H5TG).
Figure 6a has is well described in the text and could be deleted.
In the first paragraph of the Discussion the authors mix up a number of times the proteins H5TG and H5Dk.
Author Response
Thank you for shepherding the review of our manuscript. We appreciate the thoughtful and constructive comments from the reviewer to improve the quality of the manuscript. We have now revised the manuscript to accommodate the points raised by the reviewer.
Reviewer 2
Open Review
(x) I would not like to sign my review report
( ) I would like to sign my review report
English language and style
( ) Extensive editing of English language and style required
( ) Moderate English changes required
(x) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style
|
Yes |
Can be improved |
Must be improved |
Not applicable |
|
|
Does the introduction provide sufficient background and include all relevant references? |
(x) |
( ) |
( ) |
( ) |
|
Is the research design appropriate? |
(x) |
( ) |
( ) |
( ) |
|
Are the methods adequately described? |
( ) |
(x) |
( ) |
( ) |
|
Are the results clearly presented? |
(x) |
( ) |
( ) |
( ) |
|
Are the conclusions supported by the results? |
(x) |
( ) |
( ) |
( ) |
Comments and Suggestions for Authors
In this manuscript the authors designed two artificial haemagglutinins using sequences from HA proteins from viruses detected in poultry in Vietnam. One of these proteins (H5.c1) showed good expression in Nicotiana benthamiana and induced strongly neutralizing antibodies in mice against homologous H5.c1 and 3 heterologous native strains in haemagglutination inhibition assays. Therefore, H5.c1 could be a promising vaccine candidate against A/H5N1 virusses.
However, I have some critical comments.
- The authors used two methods to determine the oligomeric state of H5.c1 and concluded from both methods that the protein is trimeric. However, these two methods are not very suitable to determine accurate molecular weights of proteins in solution since they are not based on the mass but on the shape of the protein of interest. Therefore, no accurate conclusion can be drawn from the position of the largest band in the SDS-PAGE/Western blot or the position of the largest peak in the SEC elution profile. Alternative methods such as analytical ultracentrifugation and static light scattering in combination with SEC (SEC/MALLS) are better suited for this purpose.
Response: We thank reviewer for the suggestion of the alternative methods such as analytical ultracentrifugation and static light scattering in combination with SEC (SEC/MALLS) to determine accurate molecular weights of proteins in solution. However, based on our laboratory facility, we just only use two methods as described in the manuscript: i) cross-linking reaction with BS3, separating protein in the SDS-PAGE/Western blot; This method was based on the study of Weldon and his colleagues (2010) and Phan and his colleagues (2013). ii) Size exclusion chromatography that is a chromatographic method in which molecules in solution are separated by their size. SEC is a widely used polymer characterization method because of its ability to provide good molar mass distribution results for polymers. The oligomeric state of HA using SEC was previously described in the study of Phan and his colleagues (2017).
- On top of that, it can’t be stated that the top HA band runs at 201 kDa since the top marker band in the SDS-PAGE/Western blot analysis is only 170 kDa. A significantly higher molecular weight is more likely.
Response: We thank reviewer for this comment. In fact, we only have marker with top band in the SDS-PAGE/Western blot analysis is 170 kDa. We corrected the state in the manuscript. The apparent molecular weights shown in Figure 4 are higher than the expected sizes predicted from unglycosylated polypeptide. This could be explained by the fact that glycosylation influences the running behavior during the electrophoretic separation.
- In the SEC experiment the largest peak of H5.c1 seems to overlap exactly with the ferritin marker (440 kDa), which would indicate a higher oligomeric state than the trimer mentioned in the text.
The authors should use another, more reliable method for the determination of the oligomeric state of H5.c1 (see above).
Response: We corrected the stated in the manuscript. The apparent molecular weight shown in Figure 5 is higher than the expected sizes of trimeric protein. This might be explained by the fact that glycosylation influences the running behavior during SEC separation. The SEC result indicated that most of the purified H5 proteins were oligomeric.
- In Figure 5 the authors show the SEC elution profiles for two different HA proteins (H5TG and H5.c1). Based on the SDS-PAGE/Western analysis they pooled for both proteins the elution fractions B1-B6 (indicated with a red box in the elution profile). Here we can clearly see that although all fractions contain HA the pooled fractions are not homogenous. Especially in the case of H5.c1 a significant second peak is also pooled that most likely represents HA of a higher oligomeric state or even aggregated protein. This fraction is most likely less or even not at all functional and if present in the preparations used for functional studies will affect the results negatively. The shape of the H5TG peak also shows that the protein is not homogenous and also here aggregated protein is added to the preparation.
Response: We agree with the reviewer that all fractions contain HA the pooled fractions are not homogenous, and a second peak that most likely represents HA of a higher oligomeric state or even aggregated protein. However, this could be explained by the fact that glycosylation influences the running behavior during the SEC separation. In addition, in this study, the main purpose of SEC experiment is to characterize the protein structure, and collect the purified protein fractions (B1-B6) for ELISA and Western blot to detect the HA-specific IgG antibodies, but not for immunization in mouse. We used purified protein by IMAC for mouse experiment and demonstrated that the purified HA proteins elicited strong HA-specific IgG antibodies and neutralizing antibody responses, shown in the manuscript.
Other comments:
- In par 2.3 more details should be given for the described method.
Response: We gave more details according to the reviewer’s request in the par 2.3
- In par 2.4 more details should be given such as the volume of Ni-NTA resin used, with how many column volumes was the column washed/eluted, how were the proteins stored?
Response: More details were added in par 2.4 at the request of reviewer.
- In par 2.4 did the authors test the different concentration of imidazole in the wash buffer of H5.c1 or was a sequencial series of wash steps used? If so, why going down in the imidazole concentration?
Response: The 30mM imidazole in washing buffer was previously used for purification of H5TG protein (Phan et al., 2017) but not for H5.c1 protein. Since a large amount of target H5.c1 protein was removed in wash step using 30 mM imidazole, we reduced the concentration of imidazole in the wash step. In our study, the different concentration of imidazole in the wash buffer of H5.c1 (30 mM, 25 mM, 20 mM or 5 mM imidazole) in washing buffer was tested. The supplementary data 1 results showed that the 5 mM Imidazole in the wash step is the optimal concentration for purification of H5.c1 protein.
- Did the authors also purify the second artificial HA protein, H5.c2?
Response: We have tried to purify the artificial H5.c2 protein, however the low H5.c2 protein expression level in planta made a difficulty in purification. We did not have the purified artificial H5.c2 protein for further experiments. Therefore, we did not show the purification of H5.c2 protein in the manuscript.
- In par 2.6 what was the running buffer of the SEC experiment?
Response: The running buffer of the SEC experiment was phosphate buffer
- The first paragraph 2.9 is exactly the same as paragraph 2.8 and can be deleted.
Response: We deleted the par.2.9 in the manuscript.
- In par 3.2 the imidazole concentration in the elution buffer of H5 TG is 125 mM. In par 2.4 it is 50 mM. Please clarify.
Response: we corrected that the imidazole concentration in the elution buffer of H5 TG is 125 mM in par 2.4.
- In par 3.2 the authors mentioned that the different HA proteins contain a C-terminal His-tag. From Figure 2 it can be concluded that this should be an N-terminal His-tag.
Response: we corrected that the different HA proteins contain a N-terminal His-tag in the par 3.2.
- The accuracy of the molecular weights obtained from the SDS-PAGE/Western blot gel is unrealistic (e.g. 201.45 kDa). Especially when these are supposed to be only approximations (~201.45 kDa).
Response: We agree with the reviewer this comment. We corrected this state in the result section in the manuscript.
- In par 3.3 Figure 5c should be 6c.
Response: we corrected the Figure 6c in par 3.3.
- In par 3.3 (see also Figure 6c) I don’t agree with the conclusion that the mean OD450 values of group 1 (H5.c1) are not significantly different from that of group 2 (H5TG).
Response: We based on the statistical result of Welch's t-test in Sigma plot software.
After the second immunization, the difference in the mean values of the two groups (H5.c1 and H5TG) is not great enough to reject the possibility that the difference is due to random sampling variability. There is not a statistically significant difference between the input groups (P = 0.130).
After the third immunization, the difference in the mean values of the two groups is not great enough to reject the possibility that the difference is due to random sampling variability. There is not a statistically significant difference between the input groups (P = 0.070).
We would like to show you the results of Welch's t-test analysis as the following:
Comparing H5.c1 and H5TG after 2nd immunizationt-test Wednesday, September 04, 2019, 10:21:55 AM
Data source: Data 1 in Notebook1
Normality Test (Shapiro-Wilk): Passed (P = 0.334)
Equal Variance Test (Brown-Forsythe): Passed (P = 0.488)
The result of the equal variance test indicates the likelihood that the two groups are sampled from populations with equal variances, but does not guarantee the equality or inequality of the two variances.
Group Name N Missing Mean Std Dev SEM
H5.c1 5 0 0.367 0.222 0.0994
H5TG 5 0 0.172 0.104 0.0463
Difference of means 0.194
Use the results of Welch's test, where equal variances are not assumed, if the equality of the population variances of the two groups is in doubt.
Equal Variances Not Assumed (Welch's t-test):
t = 1.773 with 5.658 degrees of freedom.
95 percent two-tailed confidence interval for difference of means: -0.0779 to 0.467
Two-tailed P-value = 0.130
The difference in the mean values of the two groups is not great enough to reject the possibility that the difference is due to random sampling variability. There is not a statistically significant difference between the input groups (P = 0.130).
Comparing H5.c1 and H5TG after 3rd immunization
t-test Wednesday, September 04, 2019, 10:46:27 AM
Data source: Data 1 in Notebook1
Normality Test (Shapiro-Wilk): Passed (P = 0.357)
Equal Variance Test (Brown-Forsythe): Passed (P = 0.247)
The result of the equal variance test indicates the likelihood that the two groups are sampled from populations with equal variances, but does not guarantee the equality or inequality of the two variances.
Group Name N Missing Mean Std Dev SEM
H5.c1 5 0 0.748 0.347 0.155
H5TG 5 0 0.369 0.128 0.0573
Difference of means 0.379
Use the results of Welch's test, where equal variances are not assumed, if the equality of the population variances of the two groups is in doubt.
Equal Variances Not Assumed (Welch's t-test):
t = 2.286 with 5.069 degrees of freedom.
95 percent two-tailed confidence interval for difference of means: -0.0454 to 0.803
Two-tailed P-value = 0.0703
The difference in the mean values of the two groups is not great enough to reject the possibility that the difference is due to random sampling variability. There is not a statistically significant difference between the input groups (P = 0.070).
- Figure 6a has is well described in the text and could be deleted.
Response: We deleted Figure 6a in the manuscript.
- In the first paragraph of the Discussion the authors mix up a number of times the proteins H5TG and H5Dk.
Response: We corrected the name of the proteins (H5TG and H5Dk) in the first paragraph of the Discussion.
