Variability of Fruit Quality among 103 Acerola (Malpighia emarginata D. C.) Phenotypes from the Subtropical Region of Brazil

Round 1

Reviewer 1 Report

In the submitted manuscript by Daniela Farinelli, Silvia Portarena, and colleagues entitled “Variability of fruit quality among 103 acerola (Malpighia emarginata D. C.) phenotypes from the subtropical region of Brazil’’, the authors evaluated 103 phenotypes of acerola concerning physiological traits, vitamin C and polyphenols content. The authors found that the vitamin C varied from 425 mg to 2625 mg/100 g pulp, with a mean value of 1240 mg/100 g, and polyphenols content ranged from 84 mg/kg to 3196 mg/kg with a mean value of 1397 mg/kg. Overall, the manuscript is well written, and the tables have a good format, but the quality of the figures should be improved. To my opinion, this manuscript is in line with the aims and scope of the Agronomy journal. However, there are some issues in the text, tables and figures that should be carefully addressed.

Major and minor comments,
Abstract section

Line 22-23: The authors should avoid mentioning phenotype numbers, especially in the abstract section. Please, replace with physiological traits of the selected phenotypes. For instance, ‘Seven phenotypes were selected as candidates in the next breeding program,’ that have high SSC and fruit weight and belongs …..

Line 24-27: Similarly, please avoid mentioning phenotype numbers, replace them with vitamin C content.

To my opinion, the same approach should you follow in the conclusion section. (avoid phenotype numbers and replace them with desirable traits).

Table 1

In the first location (Anápolis, Goiás Sate) at TA, the authors mention 7.5 – 17.4 % which is extraordinarily high TA concentration. Taking into account the pH values (2.3 - 3.5) from the same location probably the authors make a mistake. Please, replace it with the correct one or else remove this source.

Furthermore, Flor Branca and Junko cultivars seem to have very low vitamin C content (2.16 - 2.77). Do they are the actual values or making a mistake?

Figure 3 legend

The authors should mention the number of replicates and each dot what does the symbol means? For instance, 4 replicates of 10 fruits, each dot represents or indicates the replicate.

Figure 5

The authors should improve the quality.

Figure 7.

The authors should improve the quality and provide a more suitable PCA because vitamin C and polyphenols have low representation in the current PCA model. Perhaps in PC1 - PC3 have a better one or in PC2 with PC3. Please incorporate and PC1 - PC3 and PC2 -PC3.

Author Response

In the submitted manuscript by Daniela Farinelli, Silvia Portarena, and colleagues entitled “Variability of fruit quality among 103 acerola (Malpighia emarginata D. C.) phenotypes from the subtropical region of Brazil’’, the authors evaluated 103 phenotypes of acerola concerning physiological traits, vitamin C and polyphenols content. The authors found that the vitamin C varied from 425 mg to 2625 mg/100 g pulp, with a mean value of 1240 mg/100 g, and polyphenols content ranged from 84 mg/kg to 3196 mg/kg with a mean value of 1397 mg/kg. Overall, the manuscript is well written, and the tables have a good format, but the quality of the figures should be improved.

Answer：The authors improved the quality of figures as much as possible.

To my opinion, this manuscript is in line with the aims and scope of the Agronomy journal. However, there are some issues in the text, tables and figures that should be carefully addressed.

Major and minor comments,
Abstract section

Line 22-23: The authors should avoid mentioning phenotype numbers, especially in the abstract section. Please, replace with physiological traits of the selected phenotypes. For instance, ‘Seven phenotypes were selected as candidates in the next breeding program,’ that have high SSC and fruit weight and belongs Changed

Line 24-27: Similarly, please avoid mentioning phenotype numbers, replace them with vitamin C content. Changed

To my opinion, the same approach should you follow in the conclusion section. (avoid phenotype numbers and replace them with desirable traits).

To answer to the three reviewers, we added only desirable traits, keeping the phenotype number. see revised manuscript.

Table 1

In the first location (Anápolis, Goiás Sate) at TA, the authors mention 7.5 – 17.4 % which is extraordinarily high TA concentration. Taking into account the pH values (2.3 - 3.5) from the same location probably the authors make a mistake. Please, replace it with the correct one or else remove this source.

You are right, the values are corrected but they are expressed as ml NaOH N, we added this to the cell.

Furthermore, Flor Branca and Junko cultivars seem to have very low vitamin C content (2.16 - 2.77). Do they are the actual values or making a mistake?

Sorry, this range is expressed as %. We add it to the cell.

Figure 3 legend

The authors should mention the number of replicates and each dot what does the symbol means? For instance, 4 replicates of 10 fruits, each dot represents or indicates the replicate.

We added that each dot represents the mean value of 4 replicates of ten fruits each.

Figure 5

The authors should improve the quality. done

Figure 7.

The authors should improve the quality and provide a more suitable PCA because vitamin C and polyphenols have low representation in the current PCA model. Perhaps in PC1 - PC3 have a better one or in PC2 with PC3. Please incorporate and PC1 - PC3 and PC2 -PC3.

The authors improved the quality of Figure 7 and added a new Figure, named Figure 8, in which Scatter plot of the scores of acerola samples on the two-dimensional plane defined by PC1 and PC3 as calculated from the complete dataset by PCA. Authors did not incorporate PC2-PC3 because did not explain al lot of total variance.

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript has some serios flaws:

• The accesions under study are not properly described (origin, exact location, age of the trees, common name, etc)
• The evaluation of the trees has been carried out only during one year, which is not enough to have robust data in fruit trees.
• A genetic study and a population structure analysis is needed to identify homonimes and synonimes among the trees studied. 
• The authors do not specify how many trees are evaluated from each accesion. 
• Material and methods are not described with enough detail. 

Author Response

(1)The accesions under study are not properly described (origin, exact location, age of the trees, common name, etc).

The accessions are 103 trees located in different gardens of the Marechal Cândido Rondon city, in Western Paraná State, Brazil, located at a latitude of 24°33'23.26"S and a longitude of 54°3'28.33"W, 420 meters above sea level. The authors tried to draw a map of the city showing the exact location of the 103 phenotypes, but it was not legible in detail. All the trees are adult, in average 7 years old, and have a high and costant yield. The phenotypes do not have a common name and the authors had labeled each one by a single number code.
(2)The evaluation of the trees has been carried out only during one year, which is not enough to have robust data in fruit trees.

One year pre-screening of tree crops germplasm is a methodology already used for other fruit species as well as olive and passion fruit (Bianco et al. 2013; Galeano Mendoza et al. 2018). In specific, many studies carried out to evaluate acerola fruit quality have been performed during only one year as reported in cited lecterature i.e. [16, 17, 23, 37, 38, 41]. This is due to climatic condition recorded in the brazilian acerola growing area. In addition, the subtropical mesothermal climate of the area of study is characterized by almost stable wether contidions with no significant inter-annual temperature and precipitation amount of variations. Annual rainfall varies from 1600 to 1800 mm, with rainfall well distributed throughout the year and hot summers. In specific, the average value of precipitation in October, November and December of the last ten years is, respectively 134±22.3 mm; 167±26.6mm and 183±41 mm, showing very constant climatic conditions. This situation allows to limit the fruit quality evaluation during one year.
(3)A genetic study and a population structure analysis is needed to identify homonimes and synonimes among the trees studied.

The authors did not carry out genetic characterization of acerola's phenotypes because nowadays SSR markers are not available yet, as well as, genomic assembly from which SSR markers may be extracted. To do this it is necessary to sequence the DNA and identify the set of suitable SSR markers.
(4)The authors do not specify how many trees are evaluated from each accesion.

In this prescreening phase each accession is composed of a single tree. The 7 most interesting phenotypes will be clonally multiplied and then compared.
(5)Material and methods are not described with enough detail.

The request is very generic and therefore difficult to replicate. The authors have however tried to integrate this part with further details.

Author Response File: Author Response.pdf

Reviewer 3 Report

Farinelli et al compared 103 acerola phenotypes based on critical biochemical and physiological markers. The article is appropriate for Agriculture, however, I cannot propose the acceptance of the article because as it stands, it is mainly descriptive as it lacks genetic data. I propose the inclusion of data (such as SSR markers or similar) at least from the phenotypes that the authors found to have the best properties. Moreover, because of its importance, the ascorbic acid levels should have been assayed using a more accurate method.

Minor comments

Line 14 remove “ever known”

Line 69 revise to “specific cultivars or to “a limited number of cultivars”

Line 150 revise to “described in”

Line 318 the genetic differences should be documented

Line 483 describe in-depth and provide reference

Author Response

Farinelli et al compared 103 acerola phenotypes based on critical biochemical and physiological markers. The article is appropriate for Agriculture, however, I cannot propose the acceptance of the article because as it stands, it is mainly descriptive as it lacks genetic data. I propose the inclusion of data (such as SSR markers or similar) at least from the phenotypes that the authors found to have the best properties. Moreover, because of its importance, the ascorbic acid levels should have been assayed using a more accurate method.

Answer: This study, as pointed out by the reviewer, is a descriptive characterization of 103 phenotypes using statistical methods such as Principal Component and Hierarchical Cluster analysis performed on all dataset to explore the variability among samples and to identify the main clusters.

The authors did not carry out the genetic characterization of acerola's phenotypes because SSR dominant markers are not available for the acerola species, as well as, genomic assembly from which SSR markers may be extracted. To do this it is necessary to sequence the DNA and identify the set of suitable SSR markers.

Previous genetic studies (Salla et al., 2002 and Chowdhury et al., 2005) used RAPD markers and reported the associations observed with molecular markers were, for many accesses, the same of those determined on the basis of morphological characters.

Anyway next step, as reported on line 508-510 of the submitted manuscript, will be to try genotyping phenotypes using microsatellite markers.

However, developing these SSR dominant markers requires a lot of effort and a distinct study.

The authors agree that the ascorbic acid levels could have been assayed using a more accurate method, but, being this research a qualitative characterization of the fruits and not a study of characterization of the metabolic profile of the 103 phenotypes, the authors have decided to follow the method following Mapa (Ministério da Agricultura Pecuária e Abastecimento) guidelines (BRASIL, 2005), used in Brazil by other colleagues ([16], [17], [23]).

Minor comments

Line 14 remove “ever known” done

Line 69 revise to “specific cultivars or to “a limited number of cultivars” done

Line 150 revise to “described in” done

Line 318 the genetic differences should be documented

Answer：We agree; in fact, the authors, since have eliminated all other factors of variability, such as stage of maturation, conservation and place of cultivation, postulated that the physical and biochemical variation are due to intrinsic genetic differences. The next step will be to try to genotype acerola’s phenotypes.

Line 483 describe in-depth and provide the reference

Done, see revised manuscript.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

All my concerns have been successfully addressed.

Author Response

The authors thank the reviewer because his comments were very useful. We checked english language and style

Reviewer 2 Report

Several considerations to authors: 

• along the text: when citing literature in the text, if refering to the study in a sentence, you should name at least the author to make the sentence readable and complete (see lines 164, 151, 474, etc).
• along the text: do not start a sentece with a figure (see line 23). 
• Table 1 and Table S3. If you define the units in the first row of a table, then you need to convert the units shown in the rest of the data in the units defined in that first row (for example, if using mm, do not use cm in the body of the table). See Vit C content and TA in Table 1. 
• Table S2. Why all the TA data is multiple of 5? How can you explain that SE of TA is in most cases 0.10?
• Methodology is still no clear. In section 2.2, 4 repetitions of 10 fruits were used. Then, in section 2.3, 3 repetitions of 5 fruits. Why? Also, in section 2.3, are those parameters measured from fresh fruit or from the stored pulp in 2.2? If measured from the stored pulp, why 3 reps instead of 4?
• Did you measure SSC, TA, etc from each fruit or from the homogenate of several fruits? how many?
• Was Vitamin C measured from stored tissue? If so, it is well known that vit C is easily degraded when freezing. HPO3 must be added to the tissue before freezing to stablize vitamin C. Therefore you must have underestimate the vit C content in you samples. 
• Along the text (and in the tables and figures), when refering to soluble solids content, use the common SSC abbreviation. 
• Fig. 5. Are all the shown correlations statistically significant? A symbol should be added to identify significant ones and their level. 

Author Response

First of all, the authors thank the reviewer because his/her comments were very useful, below, the answers to the questions.

• along the text: when citing literature in the text, if refering to the study in a sentence, you should name at least the author to make the sentence readable and complete (see lines 164, 151, 474, etc). We agree and add the author’s name.
• along the text: do not start a sentence with a figure (see line 23). Thanks, we checked along the text and corrected these errors
• Table 1 and Table S3. If you define the units in the first row of a table, then you need to convert the units shown in the rest of the data in the units defined in that first row (for example, if using mm, do not use cm in the body of the table). See Vit C content and TA in Table 1. We converted all the units according to the units defined in the first row.

• Table S2. Why all the TA data is multiple of 5? How can you explain that SE of TA is in most cases 0.10? Thanks. The number of decimal places reported for the mean and SE exceeded that of the precision of the measurements. We now report these results to one digit after the decimal point.
• Methodology is still no clear. In section 2.2, 4 repetitions of 10 fruits were used. Then, in section 2.3, 3 repetitions of 5 fruits. Why? Also, in section 2.3, are those parameters measured from fresh fruit or from the stored pulp in 2.2? If measured from the stored pulp, why 3 reps instead of 4?

We have written better these parts. In specific, we reported, from line 105 to line 107, that for each phenotype, a fruit sample, composed by 4 repetitions of 10 fruit each, were used to determine the physical and biochemical parameters of the fruits. Since, from each repetition, a sub-sample of 5 fruits was de-pulped (lines 136-138), and the other sub-sample of 5 fruits was use for the determination of biochemical parameters. The pulps were manually separated from the seeds and ground (lines 140-143).

• Did you measure SSC, TA, etc from each fruit or from the homogenate of several fruits? how many?  We homogenate the pulps of 5 fruits, for each repetition.
• Was Vitamin C measured from stored tissue? If so, it is well known that vit C is easily degraded when freezing. HPO3 must be added to the tissue before freezing to stablize vitamin C. Therefore you must have underestimate the vit C content in you samples. Yes, it was. To avoid any degradation, the pulps were stored in small plastic bags and kept in a freezer at temperature of -80 °C until the time of analysis, performed one / two days later (lines 142-143).
• Along the text (and in the tables and figures), when refering to soluble solids content, use the common SSC abbreviation. Thanks, done
• Fig. 5. Are all the shown correlations statistically significant? A symbol should be added to identify significant ones and their level. As now reported in the text (lines 175- 177 of the new manuscript) and in the legend of Figure 5 the values of correlation reported in the heat map matrix resulted significantly different per p < 0.05.

Reviewer 3 Report

After the revision, the article can be accepted for publication

Author Response

The authors thank the reviewer because his comments were very useful. We checked english language and style.