Evaluation of Inoculation Methods for Determination of Winter Wheat Resistance to Fusarium Head Blight
Round 1
Reviewer 1 Report
Why authors used each year different sets of genotypes? Only 5 cultivars were repeated all years. For objectives of the study: "(1) to compare three inoculum sources in terms of their efficacy in inducing FHB symptoms, (2) to determine the relationship between FHB scores within methods, and (3) to estimate heritability of investigated traits" authors should use the same set of cultivars/lines in four experimental years to remove effect of genotype.
You should apply mist irrigation also on non-inoculated plots to promote natural infection.
Did you transform % data prior to the analysis?
How variance components were calculated?
In my opinion authors cannot compare of heritabilities between years. Each year was different, with unique and random weather conditions, so each year should be treated as a separate experiment in this case (different sets of genotypes in each year!). In such a situation, we are not authorized to compare heritabilities with each other (they are obtained under different incomparable environmental conditions), and even more there is no basis for calculating the average value for heritability coefficients over several years.
Author Response
Dear Reviewer,
Thank you for your valuable comments and suggestions. Below are the responses to your comments. Following your suggestions, we have added the description of the calculation of the variance component to the Materials and Methods chapter, and removed the average heritability values from the table 2 and made appropriate corrections in the Results and Discussion chapters.
Why authors used each year different sets of genotypes? Only 5 cultivars were repeated all years. For objectives of the study: "(1) to compare three inoculum sources in terms of their efficacy in inducing FHB symptoms, (2) to determine the relationship between FHB scores within methods, and (3) to estimate heritability of investigated traits" authors should use the same set of cultivars/lines in four experimental years to remove effect of genotype.
We compared three inoculation methods within each year, not between years. We agree with you that it would be beneficial to include the same set of genotypes to draw conclusions for all three objectives of the present study. On the other hand, our study is consistent with real breeding programs (especially those of small breeding companies), where new line sets are introduced each year in the field tests for FHB resistance. Despite the different genotype sets tested each year, infection levels measured as VRI and FDK were satisfactory for the two methods of artificial inoculation, as were estimates of heritability and correlations between methods. In addition, there is only one study by Mesterhazy et al. (2015) reporting the results of comparing maize stalk inoculation method with standard spray methods. Therefore, we believe that our results add to the existing knowledge on the potential of the maize stalk inoculation method in FHB resistance testing.
You should apply mist irrigation also on non-inoculated plots to promote natural infection.
In the present study, mist irrigation was not used in either artificial inoculation or natural infection. Therefore, the genotypes were grown under the same conditions in all three inoculation methods used. This was not clearly stated in the description of inoculation methods in the Materials and Methods chapter, which we have corrected.
We recognize that mist irrigation may have favoured Fusarium infection and created more uniform conditions for infection over the years, but breeders aim to develop varieties for real environmental conditions in a given region, which we believe were well represented during the four years of our study. Furthermore, the breeding program for FHB resistance at the Bc Institute, on which this research was based, has no equipment and does not use mist irrigation after artificial inoculations. In addition, mist irrigation would be difficult to implement under maize stalk inoculation at distant locations, which was noted as a potential advantage of this method in the Conclusions chapter of this study.
Did you transform % data prior to the analysis?
We did not transform the data prior the analysis, which is in line with other studies cited in the present manuscript.
The situation is similar when we looked at the mapping and GWAS studies on FHB resistance.
We also have been studying pre-harvest sprouting resistance in wheat for years, in which % data are common. We noticed that in some earlier studies transformation of data were sometimes performed prior to the statistical analyses. However, we did no find reports on % data transformation in newer papers (last two decades).
How variance components were calculated?
Genotypic variance (s2G) was calculated as (genotype mean square – error mean square) / r), where r is the number of replicates, and error variance (σ2ε) is equal to the error mean square. This sentence is added in the description of the statistical analysis (Materials and Methods chapter).
In my opinion authors cannot compare of heritabilities between years. Each year was different, with unique and random weather conditions, so each year should be treated as a separate experiment in this case (different sets of genotypes in each year!). In such a situation, we are not authorized to compare heritabilities with each other (they are obtained under different incomparable environmental conditions), and even more there is no basis for calculating the average value for heritability coefficients over several years.
You are right. Heritability estimates depend on both the set of genotypes and the set of environments included. In the present study, we did not compare heritabilities between years but between different inoculation methods within each year, including the same set of genotypes tested in the same environment. The average heritability values presented in table 2 were intended to represent only a descriptive statistics of a set of heritabilities (data set) estimated in four years of the study. Just as we reported minimum and maximum heritability values for each trait, we also reported the corresponding average heritabilities. A similar approach was presented in the study by authors from seven U.S. universities published in Plant Diseases in 2022 (Gaire R, Sneller C, Brown-Guedira G, Van Sanford D, Mohammadi M, Kolb FL, Olson E, Sorrells M, Rutkoski J. Genetic Trends in Fusarium Head Blight Resistance from 20 Years of Winter Wheat Breeding and Cooperative Testing in the Northern U.S.A. Plant Dis. 2022 Feb;106(2):364-372. doi: 10.1094/PDIS-04-21-0891-SR. Epub 2022 Feb 6. PMID: 34282926). Using 20 years of data (1998 to 2018) from the Northern Uniform and Preliminarily Northern Uniform winter wheat scab nurseries (different sets of genotypes were included in the trials conducted at different sets of locations each year), in which the authors presented trends in means for two FHB scores, FDK and DON content. They also estimated genetic and error variances and heritability on a single location basis using the same equation as we used: h2=σ2a/(σ2a+σ2ε). At the beginning of the Results chapter (page 368) the authors reported the ranges of heritability from single locations over 20 years and the average heritabilities over the same period for four traits studied. They also performed a statistical test (Tukey HSD test) to compare the mean heritabilities of four FHB-related traits studied. The results showed that DON had significantly higher heritability (0.33) compared with incidence and severity (0.22 and 0.26, respectively).
However, your comment is correct, since average values in the present study may confuse the reader by suggesting that heritability averages presented in the table 2 resulted from combined ANOVA across years, which we did not conduct. It is also possible that there is no theoretical (statistical) basis for calculating the average values of heritability over several years. Therefore, considering your comment, we have removed the average values of heritability from table 2 and adjusted the corresponding text in the Results chapter and in the Discussion chapter accordingly. In addition, in table 2 we corrected the value for heritability for FDK in 2014 in natural infection (typing error). Instead of 0.12 it was written 0.2. This correction did not affect the range of heritabilities for FDK presented in the text of the Results chapter.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors, I think that the research results presented in the paper are interesting and the publication of the manuscript is worth considering. However, I have a few comments whose resolution is very important for the publication process of the paper.
(1) The field experiment should be designed in such a way that an equal quantity of inoculum is used to infect a wheat sample representative of each variety. Each method of artificial inoculation should be described very precisely in the methodology. This is essential for correct conclusion.
(2). Please give the location where the isolates were stored, the name of collection and their codes (F. graminearum).
(3) How were the spores prepared for inoculation?
(4) How were the plants sprayed to ensure homogeneity of infection?
(5) How was uniformity of infection ensured for all wheat genotypes tested using infected maize stalks? There may have been different levels of inoculum and strains of F. graminearum with different genetic characteristics and aggressiveness towards wheat. In addition, there may have been various other Fusarium causing FHB present.
(6) What sample was used to make a visual assessment of the level of infection? What scale was used? Please include a detailed description of this procedure in your paper.
(7) Please describe in detail how FDK was assessed.
(8) Was fungal reisolation performed to confirm the effectiveness of infection by F. graminearum?
(9) The Spearman's Rank method should be used for correlation. A description should be included in the methodology.
Author Response
Dear Authors, I think that the research results presented in the paper are interesting and the publication of the manuscript is worth considering. However, I have a few comments whose resolution is very important for the publication process of the paper.
Dear Reviewer,
Thank you for your valuable comments and suggestions. Below are the responses to your comments. Following your suggestions, we have added a detailed description of the inoculum production methodology, the inoculation procedure for the spray method and the maize stalk method, the description of the visual assessment of the level of infection, and the recalculation of the correlations between traits using Spearman's Rank method.
(1) The field experiment should be designed in such a way that an equal quantity of inoculum is used to infect a wheat sample representative of each variety. Each method of artificial inoculation should be described very precisely in the methodology. This is essential for correct conclusion.
We added a precise description of applied inoculation methods in the manuscript.
(2). Please give the location where the isolates were stored, the name of collection and their codes (F. graminearum).
Each year new isolates were prepared and used for artificial inoculation by the spray method. This has been a common practice for more than 30 years in the breeding program for FHB resistance in wheat conducted at the Bc Institute. To reduce the influence of isolates from different years in this research four different isolates (most aggressive for the respective year) were used, probably resulting in better data quality. On the other hand, the focus was on comparing inoculation methods within years. Consistently, for the maize stalk inoculation method, new stalks infected with the dominant Fusarium strains in that year were collected each year because infected maize stalks cannot be stored for several years.
(3) How were the spores prepared for inoculation?
In the Materials and Methods chapter we added a detailed description of the method of preparing the inoculum.
(4) How were the plants sprayed to ensure homogeneity of infection?
A detailed description of the inoculation procedure was added in the materials and Methods chapter.
Each genotype was sprayed separately on two occasions and 40 ml of inoculum was applied to each plot in both sprays.
(5) How was uniformity of infection ensured for all wheat genotypes tested using infected maize stalks? There may have been different levels of inoculum and strains of F. graminearum with different genetic characteristics and aggressiveness towards wheat. In addition, there may have been various other Fusarium causing FHB present.
Uniformity of infection for all wheat genotypes tested using infected maize stalks was ensured by scattering an approximately equal number of pieces of maize stalks to each experimental plot? The same methodology was used in the study by Mesterhazy et al. (2015), cited in this manuscript.
Different (unknown) levels of inoculum and strains of F. graminearum with different genetic characteristics and aggressiveness towards wheat are to be expected and cannot be avoided. The method itself assumes that the inoculum source is collected de novo each year, since the pieces of infected stalks cannot be stored for several years.
The presence of various other Fusarium spp. causing FHB in collected maize stalks also cannot be avoided. However, monitoring of Fusarium spp. in naturally infected wheat under the agroclimatic conditions of Croatia, revealed that the predominant species causing FHB was F. graminearum (represented with 80% in infected grains), while F. culmorum, F. avenaceum and F. poae were represented with a total of 20% (Španić et al., 2010- this reference is cited in the manuscript in the Introduction chapter). Španić V., Lemmens M., Drezner G. (2010). Morphological and molecular identification of Fusarium species associated with head blight on wheat in East Croatia. European journal of plant pathology 128(4): 511-516.
(6) What sample was used to make a visual assessment of the level of infection? What scale was used? Please include a detailed description of this procedure in your paper.
A visual assessment of the level of infection was conducted on sample of 100 spikes per plot. Two assessments were made, 21 and 25 days after the second infection in the spray method. The genotypes of all three methods were evaluated on the same day. The scale used is shown in Table S1, which is added to the manuscript.
(7) Please describe in detail how FDK was assessed.
The percentage of Fusarium damaged kernels (FDK) was determined on ten randomly selected spikes taken from each experimental plot after harvest. The spikes were threshed by hand and the Fusarium damaged and normal kernels were counted. Only the pinkish white colored grains along with the slightly infected whitish powdered kernels were considered as Fusarium damaged, while the normally coloured but shrivelled kernels were not considered. This description has been added to the manuscript.
(8) Was fungal reisolation performed to confirm the effectiveness of infection by F. graminearum?
Fungal reisolation was not performed.
(9) The Spearman's Rank method should be used for correlation. A description should be included in the methodology.
The correlations between traits were recalculated using Spearman's Rank method and appropriate changes were made in the tables and text of the manuscript.
Author Response File: Author Response.docx
Reviewer 3 Report
Congratulations for fine research!
Please clarify the isolates which were used for inoculation. In one sentence there is a remark "For this method, four most aggressive strains of F. graminearum grown on PDA medium were used." That is al right. I do not understand the inoculum source for next year. Whether same four isolates were used. There is a sentence "The source of inoculum production were Fusarium infected wheat grains from the previous year"., but in that case you isolated new isolates which are not the same as original ones.
Other question is maintanence of isolates. It is well known that they are loosing their pathogenicity after several subculturing to PDA. How you stored isolates for next vegetation?
Author Response
Dear Reviewer,
Thank you for your comments. Below are the responses to your comments. In the Materials and Methods chapter, we have added a detailed description of the inoculum production methodology, the inoculation procedure for the spray method and the maize stalk method as well as an extended description of the visual assessment of the level of infection.
Please clarify the isolates which were used for inoculation. In one sentence there is a remark "For this method, four most aggressive strains of F. graminearum grown on PDA medium were used." That is al right. I do not understand the inoculum source for next year. Whether same four isolates were used. There is a sentence "The source of inoculum production were Fusarium infected wheat grains from the previous year"., but in that case you isolated new isolates which are not the same as original ones.
Each year new isolates were prepared and used for artificial inoculation by the spray method. This has been a common practice for more than 30 years in the breeding program for FHB resistance in wheat conducted at the Bc Institute. To reduce the influence of isolates from different years in this research four different isolates (most aggressive for the respective year) were used, probably resulting in better data quality. On the other hand, the focus was on comparing inoculation methods within years, not between years. Consistently, for the maize stalk inoculation method, new stalks infected with the dominant Fusarium strains in that year were collected each year because infected maize stalks cannot be stored for several years. However, despite the different genotype sets tested each year, infection levels measured as VRI and FDK were satisfactory for the two methods of artificial inoculation, as were estimates of heritability and correlations between methods
Other question is maintanence of isolates. It is well known that they are loosing their pathogenicity after several subculturing to PDA. How you stored isolates for next vegetation?
As described in the first answer, four most aggressive isolates were selected each year and used for spray inoculation.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Authors responded fully to my comments and made significant changes in the manuscript.
Reviewer 2 Report
The manuscript has been revised. All questions requiring clarification have been positively resolved. I recommend acceptance of the work for publication.
