Norovirus Detection in Ready-To-Eat Salads by Propidium Monoazide Real Time RT-PCR Assay

Round 1

Reviewer 1 Report

Manuscript Number: applsci-861348

The manuscript by Valentina Terio et al., describes the “Norovirus detection in ready-to-eat salads by 2 Propidium monoazide Real Time RT-PCR assay”. The authors evaluated norovirus PMA-qPCR methods and applied on 135 bags ready to eat vegetables specimens from 2 plants to differentiate infectious/ non-infectious noroviruses.

Comments:

Norovirus is a leading cause of acute diarrhea disease of all population in the world. Several factors are currently increasing the challenge posed by norovirus infections to global healthy, such as the rapid rate of the genetic and antigenic evolution of circulating noroviruses complicates the development of vaccine and treatment, several transmission routes and cause disease between contaminated environment to human with only low dose et al . Foodborne or waterborne are important transmission routes from investigations of outbreaks. The issue of control and prevent virus contaminated food providing to the market is important policy.  

1. In this study, the authors setup a norovirus PMA-qPCR methods to differentiate infectious/ non-infectious noroviruses. The results should include the limitation of this method, compare the reproducibility, bias, and thermal inactivated..
2. Though, the samples were collected from only 2 brand plants. But these 2 are best-known brands on the market. The positive rates are high in pre-treatment and post-treatment, 74.8% (101/135) and 28.1% (38/135). Can this norovirus PMA-qPCR method determine the viral titer in each positive sample?
3. What’s the detection limitation in this norovirus PMA-qPCR method?
4. Page 4 line 150, check “supernatant cointeined virus RNA”
5. Check inactivation temperature in Method 2.3.2 and Table 1
6. Page 5 line 175, check “Table explanantion” ?
7. In Table 2, may add “Total” in the button of the table.

Author Response

We gratefully acknowledge the interest in our work and the valuable comments, suggestions, and corrections. Our responses immediately follow, with indication of the changes made in the revised version of the manuscript. Moreover, in the text, the corrections are highlighted in yellow. Please note that our responses to your comments are reported below in italics.

General comment

The manuscript by Valentina Terio et al., describes the “Norovirus detection in ready-to-eat salads by 2 Propidium monoazide Real Time RT-PCR assay”. The authors evaluated norovirus PMA-qPCR methods and applied on 135 bags ready to eat vegetable specimens from 2 plants to differentiate infectious/ non-infectious noroviruses.

Norovirus is a leading cause of acute diarrhea disease of all populations in the world. Several factors are currently increasing the challenge posed by norovirus infections to global healthy, such as the rapid rate of the genetic and antigenic evolution of circulating noroviruses complicates the development of vaccine and treatment, several transmission routes and cause disease between contaminated environment to human with only low dose et al. Foodborne or waterborne are important transmission routes from investigations of outbreaks. The issue of control and prevent virus contaminated food providing to the market is important policy.

 Comment #1: In this study, the authors setup a norovirus PMA-qPCR methods to differentiate infectious/ non-infectious noroviruses. The results should include the limitation of this method, compare the reproducibility, bias, and thermal inactivated.

Response #1: The statistical analysis was made for the reproducibility of the method, as described in paragraph 2.4. Since we are evaluating the quantitative aspect for future developments, the parameters useful for validation are being set up. Moreover, after your suggestion, a section has been added where the limits of this work and future perspectives are analyzed (paragraph 5.1 L. 279-280).

Comment #2 and #3: Though, the samples were collected from only 2 brand plants. But these 2 are best-known brands on the market. The positive rates are high in pre-treatment and post-treatment, 74.8% (101/135) and 28.1% (38/135). Can this norovirus PMA-qPCR method determine the viral titer in each positive sample?

 Response #2 and #3: Thanks for your question, having used the ISO/TS 15216-2:2017 qualitative method for the detection of viral RNA, it is not possible to know the viral titer.

Comment #4: Page 4 line 150, check “supernatant cointeined virus RNA”

Response #4: Thanks, I modified cointeined with contained as required.

Comment #5: Check inactivation temperature in Method 2.3.2 and Table 1

Response #5: Thank you, done.

Comment #7: In Table 2, may add “Total” in the button of the table.

Response #7: Thanks for your suggestion, I added “Total” in table 2, (please see table 2).

Reviewer 2 Report

Ready-to-eat salads is increasingly involved into foodborne norovirus outbreaks. Sensitive and accurate detection of norovirus in RTE salads play a critical role in RTE salads safety. RT-qPCR is the good standard of detecting of norovirus in food. However, PCR cannot differentiate infectious and inactivated virus. Presence of the later in food does not pose a risk to consumers. Therefore, detection of norovirus with PCR could cause unnecessary health concern and economic burden. PMA is a nucleic acid interaction dye. Nucleic acid bound with PMA will become not available for PCR reaction. Inactivated viruses do not have intact capsids which allows PMA entering into damaged capsids and then interact with nucleic acid to inhibit PCR amplification. In contrast, infectious viruses with intact capsids which prevent PMA from entering into viruses to interact with nucleic acid. Therefore, the nucleic acids can be for PCR amplification. Application of PMA-PCR to differentiate infectious viruses and inactivated viruses have been widely studied.  The manuscript introduced the first study of detection of infectivity of norovirus in naturally contaminated RTE salads. This an innovation. However, the experimental design, conduction, or results presentation is confusing. I wish the authors can explain the following:

Table 1. There are four treatments to both NoV GI and NoV GII. They are, for example to NoV GI:

NT：sample -> photoactivation -> RNA extraction -> RT-PCR, Ct: 28.66

PMAxx: sample -> PMAxx treatment -> photoactivation -> RNA extraction -> RT-PCR, Ct: 33.79

99^oC: sample -> 99^oC inactivation -> photoactivation ->RNA extraction ->RT-PCR, Ct: 32.79

99^oC+PMAxx: sample -> 99^oC inactivation -> PMAxx treatment -> photoactivation -> RNA extraction -> RT-PCR, Ct: 36.35

              PMAxx treatment has a higher Ct than NT. It is understandable since in the NT sample, there are some inactivated viruses which cannot amplified in PMAxx.

              99^oC treatment has a higher Ct than NT. It is also understandable. During the heating, virus capsid was damaged or destroyed. The question is if all capsids were damaged or destroyed. A part or all RNA in the damaged/ destroyed capsids had degradation in the amplification region. RNA in intact capsids and RNA in the damaged/ destroyed capsids but having the intact amplification region still amplified in the PCR. So, the PCR still had CT but CT value is higher.

              99^oC+PMAxx treatment still can amplify. Does it mean 99^oC did not damage all capsids? So, there were still some RNA which did not react with PMA and then amplified while RNA in the damaged/ destroyed capsids was blocked by PMA for amplification. So, the treatment still has Ct but the Ct value is higher than the 99^oC treatment.  The explanation sounds logical. But the experiment design cannot prove PMA only enters into damaged/destroyed capsids and then intact with RNA. It cannot exclude the possibility that PMA cannot enter into intact capsids. In other words, the higher Ct value in the treatment compared with 99^oC treatment may not be just caused by PMA interacting with RNA in damaged/destroyed capsids but also caused by PMA entering into intact capsids.

              An inactivation approach before PMA treatment which can damage/destroy all capsids will be a better choice. In the situation, I expect the inactivation + PMA would not have Ct values. The complete inactivation of the approach should be supported by cell infection or published references.

              The section title of Table 1 is “PMA Real Time RT-PCR assay optimization”. Two PMA treatments were studied in the test, PMAxx and 99^oC+PMAxx. What is the conclusion of the optimization? Which one is better or they are the same? In the RTE salads test, PMAxx treatment instead of 99^oC+PMAxx was used. I think it is based on the optimization test. Why do you think PMAxx treatment instead of 99^oC+PMAxx treatment is optimal?        

              BTW, were the samples heated at 95^oC or 99^oC? In the method section, it is 99^oC. But, in Table 1 it is 95^oC.

Line 194: where is the Fig. 1.

Author Response

We gratefully acknowledge the interest in our work and the valuable comments, suggestions and corrections. Our responses immediately follow, with indication of the changes made in the revised version of the manuscript. Moreover, in the text the corrections are highlighted in the in yellow. Please note that our responses to your comments are reported below in italics.

General Comment

Ready-to-eat salads is increasingly involved into foodborne norovirus outbreaks. Sensitive and accurate detection of norovirus in RTE salads play a critical role in RTE salads safety. RT-qPCR is the good standard of detecting of norovirus in food. However, PCR cannot differentiate infectious and inactivated virus. Presence of the later in food does not pose a risk to consumers. Therefore, detection of norovirus with PCR could cause unnecessary health concern and economic burden. PMA is a nucleic acid interaction dye. Nucleic acid bound with PMA will become not available for PCR reaction. Inactivated viruses do not have intact capsids which allows PMA entering into damaged capsids and then interact with nucleic acid to inhibit PCR amplification. In contrast, infectious viruses with intact capsids which prevent PMA from entering into viruses to interact with nucleic acid. Therefore, the nucleic acids can be for PCR amplification. Application of PMA-PCR to differentiate infectious viruses and inactivated viruses have been widely studied. The manuscript introduced the first study of detection of infectivity of norovirus in naturally contaminated RTE salads. This an innovation. However, the experimental design, conduction, or results presentation is confusing. I wish the authors can explain the following:

Comment #1: Table 1. There are four treatments to both NoV GI and NoV GII. They are, for example to NoV GI: NT：sample -> photoactivation -> RNA extraction -> RT-PCR, Ct: 28.66

PMAxx: sample -> PMAxx treatment -> photoactivation -> RNA extraction -> RT-PCR, Ct: 33.79

99^oC: sample -> 99^oC inactivation -> photoactivation ->RNA extraction ->RT-PCR, Ct: 32.79

99^oC+PMAxx: sample -> 99^oC inactivation -> PMAxx treatment -> photoactivation -> RNA extraction -> RT-PCR, Ct: 36.35

PMAxx treatment has a higher Ct than NT. It is understandable since in the NT sample, there are some inactivated viruses which cannot amplified in PMAxx.

99^oC treatment has a higher Ct than NT. It is also understandable. During the heating, virus capsid was damaged or destroyed. The question is if all capsids were damaged or destroyed?

A part or all RNA in the damaged/ destroyed capsids had degradation in the amplification region. RNA in intact capsids and RNA in the damaged/ destroyed capsids but having the intact amplification region still amplified in the PCR. So, the PCR still had CT but CT value is higher.

Response #1: Thank you for your comment, the viral inactivation normally is conducted by heating at 95-99°C. In this way, you can destroy most of the viral particles, but this doesn’t assure that all capsids are destroyed or damaged, most of all when the time is short like 5’. The 95 ° C treatment is used to destroy viral particles by damaging the capsid to free RNA from the capsid and make it available for amplification or treatment with PMAxx. it is to be considered that the treatment at 95 ° C for 5 minutes does not degrade the RNA, as also described by Randazzo et.al 2016, but it is a necessary treatment to be able to validate the method. So, after 95°C treatment and addition of PMAxx, the Ct value will be higher than the sample treated exclusively at 95 ° C for 5 minutes. In Table 1, the four different treatments on NoV suspensions are described in detail in paragraph 2.2.1. 

 Comment #2: 99^oC+PMAxx treatment still can amplify. Does it mean 99^oC did not damage all capsids? So, there were still some RNA which did not react with PMA and then amplified while RNA in the damaged/ destroyed capsids was blocked by PMA for amplification. So, the treatment still has Ct but the Ct value is higher than the 99^oC treatment. The explanation sounds logical. But the experiment design cannot prove PMA only enters into damaged/destroyed capsids and then intact with RNA. It cannot exclude the possibility that PMA cannot enter into intact capsids. In other words, the higher Ct value in the treatment compared with 99^oC treatment may not be just caused by PMA interacting with RNA in damaged/destroyed capsids but also caused by PMA entering into intact capsids.

 Response #2: Thanks, there is multiple scientific evidence to support the impossibility of the PMA intercalating dye to penetrate an infectious virus. (Randazzo et al., 2016; Fuster et al., 2016; Parshionikar et al., 2010) The PMAxx RT-PCR assay consists of 2 steps: the first concerns the binding of the PMAxx to free RNA outside the viral capsid. The second concerns the impossibility of amplification, during the PCR, of RNA bound to PMAxx. For this reason, the Ct value will be higher than the sample not treated with PMAxx. It should be remembered that the Ct value is inversely proportional to the initial RNA concentration.

 Comment #3: An inactivation approach before PMA treatment which can damage/destroy all capsids will be a better choice. In the situation, I expect the inactivation + PMA would not have Ct values. The complete inactivation of the approach should be supported by cell infection or published references.

Response #3: as previously reported, the viral inactivation normally is conducted by heating at 95-99°C. In this way, you can destroy most of the viral particles, but this doesn’t assure that all capsids are destroyed or damaged.

 Comment #4: The section title of Table 1 is “PMA Real Time RT-PCR assay optimization”. Two PMA treatments were studied in the test, PMAxx and 99^oC+PMAxx. What is the conclusion of the optimization? Which one is better or they are the same?

 Response #4: Thank you for your comment, but as described below we have modified the title of paragraph 2.2

 Comment #5: In the RTE salads test, PMAxx treatment instead of 99^oC+PMAxx was used. I think it is based on the optimization test. Why do you think PMAxx treatment instead of 99^oC+PMAxx treatment is optimal?

 Response #5: Thanks, as first step of my research, I have validated the PMAxx RT PCR assay described by Randazzo et al., 2016 in my lab on viral suspension. So I think that the term optimization maybe is incorrect and the term validation is more appropriate. (I have modified the section title of paragraph 2.2)

 Comment #6: BTW, were the samples heated at 95^oC or 99^oC? In the method section, it is 99^oC. But, in Table 1 it is 95^oC.

 Response #6: Thanks for this evidence, has been an error, the value in the paragraph 2.2.1 lines 80-81 has been corrected.

Reviewer 3 Report

Terio et al. describe that Real Time RT-PCR combined with PMAxx employed to investigate a contamination of NoV GI and GII in bagged RTE vegetables showed the presence of GII in 74.8% of samples, of which 37.6% were infectious. In this study, the authors describe an importance of the PMAxx Real Time RT-PCR assay to distinguish between infectious and non-infectious viruses existing in the bagged RTE vegetables.

This reviewer agree the importance to discriminate between infectious and non-infectious viruses, both of which are equally detected by a Real Time RT-PCR though only infectious virus could affect human health. Therefore, a trial of the NoV detection methodology performed practically to estimate contamination of bagged RTE vegetables should be worthwhile.

However, the result showing in this manuscript seems not to be persuasive thus more consideration should be required; e.g., Table 1 must have p-values or at least the significance that are explained in the line 169. Ct values showing in Table 1 could be evaluated by comparative Ct or 2(-delta delta Ct) methods if the authors don't have a standard curve.

Moreover, this reviewer think the term "infectious virus" is not appropriate to explain "untreated virus" (e.g., in the line 170) unless data or references to consider a link between the PMA Real Time RT-PCR and NoV infectivity. Although this reviewer agree that there is no standardized cell culture as described in the line 260, some cultivation systems such as B cells and enteroids have been established and might have been applied to verify the link. Further discussion about the link could lead this manuscript to more worthwhile paper.

Finally, this is for my personal curiosity. Has there been a news to broadcast a foodborne incidence? The high detection rate of NoV showed in the current study could be affect local communities...

Author Response

We gratefully acknowledge the interest in our work and the valuable comments, suggestions and corrections. Our responses immediately follow, with indication of the changes made in the revised version of the manuscript. Moreover, in the text the corrections are highlighted in the revision mode and in yellow. Please note that our responses to your comments are reported below in italics.

 General comment

Terio et al. describe that Real Time RT-PCR combined with PMAxx employed to investigate a contamination of NoV GI and GII in bagged RTE vegetables showed the presence of GII in 74.8% of samples, of which 37.6% were infectious. In this study, the authors describe an importance of the PMAxx Real Time RT-PCR assay to distinguish between infectious and non-infectious viruses existing in the bagged RTE vegetables.

This reviewer agree the importance to discriminate between infectious and non-infectious viruses, both of which are equally detected by a Real Time RT-PCR though only infectious virus could affect human health. Therefore, a trial of the NoV detection methodology performed practically to estimate contamination of bagged RTE vegetables should be worthwhile.

Comment #1: However, the result showing in this manuscript seems not to be persuasive thus more consideration should be required; e.g., Table 1 must have p-values or at least the significance that are explained in the line 169. Ct values showing in Table 1 could be evaluated by comparative Ct or 2(-delta delta Ct) methods if the authors don't have a standard curve.

Response #1: Thank you for your comments, Table 1 shows the comparison between Ct of different NoV suspension treated and non-treated and after your suggestion, the table was modified as required for better understanding (please see the table 1)

Comment #2: Moreover, this reviewer think the term "infectious virus" is not appropriate to explain "untreated virus" (e.g., in the line 170) unless data or references to consider a link between the PMA Real Time RT-PCR and NoV infectivity

Response #2: Thank you for your comment, we have kept the same term because in several references the authors reported intact viral particles not linked with intercalating dye (PMA and EMA) as infectious virus.

Comment #3: Although this reviewer agree that there is no standardized cell culture as described in the line 260, some cultivation systems such as B cells and enteroids have been established and might have been applied to verify the link. Further discussion about the link could lead this manuscript to more worthwhile paper.

Response #3: Thank you for your suggestion, the “link” has been better explored and discussed within the manuscript, please see the paragraph (5.1)  In general, PMAxx RT-PCR assay is a very valid method to be able to discriminate between infectious viruses and inactivated viruses as widely documented (Randazzo et al., 2016; Randazzo et al., 2018; Fuster et al., 2016; Parshionikar et al., 2010; Elizaquivel et al., 2013).  On one hand, the advantage of this method is that in a few hours it is possible to have a result that can apply good risk analysis and safeguard public health. On the other hand, cell cultures, although they represent an excellent and even more specific method, cannot give short answers but need a few days and are not currently standardized. In this way, it is impossible to apply a risk assessment and safeguard public health.

Comment #4: Finally, this is for my personal curiosity. Has there been a news to broadcast a foodborne incidence? The high detection rate of NoV showed in the current study could be affect local communities.

Response #4: Thank you, about your personal curiosity, it is important to highlight that we are involved in several epidemiological manuscripts on NoV circulation in Italy and in the Apulia region too. We are members of the Department of Veterinary Medicine and we collaborate with Competent Authorities responsible for animal origin food. In Italy, unfortunately, vegetables are controlled by other Competent Authorities and we hope that our studies are taken into consideration in the future to improve opportune measures to control NoV risk in this food chain.

Round 2

Reviewer 2 Report

The revised manuscript solved the problems pointed out the Round 1 review. 

Author Response

I again appreciate the kindness of the Reviewer in helping
improve the manuscript.

Reviewer 3 Report

Terio et al politely answers the questions and have revised the manuscript where my comments are adequately reflected. This reviewer thinks this manuscript would satisfy to accept in Applied Sciences after minor revision as follow.

1) Line 16, NoV should be spelled out here or before.

2) The addition of a space in front a unit should be checked (especially in the MATERIALS AND METHODS section), though MDPI printer would tell during proof checking.

3) In Table 1, REDUCTION RATIO could be showed as REDUCTION because the numbers in the column are not ratio. And the REDUCTION column might contain REDUCTION numbers of 95oC sample and add hyphens to the NT sample rows.

Author Response

I thank the reviewer’s carefully reading and comments.
Considering your comments, i made the following responses:

COMMENT 1): Line 16, NoV should be spelled out here or before.

Reply: NoV has been spelled before, in Line 14

COMMENT 2): The addition of a space in front a unit should be checked (especially in the MATERIALS AND METHODS section), though MDPI printer would tell during proof checking.

Reply: I checked the spacing of the text as you requested.

COMMENT 3): In Table 1, REDUCTION RATIO could be showed as REDUCTION because the numbers in the column are not ratio. And the REDUCTION column might contain REDUCTION numbers of 95oC sample and add hyphens to the NT sample rows.

Reply: In Table 1 I deleted "RATIO" in the column "Reduction Ratio", I  added  the reduction numbers of 95 °C samples (NoV GI and NoV GII) and I added the hyphens to the NT samples, as you suggested.

Moreover, I have better specified the table 1 explanation.