Review Reports - Seasonal Dynamics of ‘<i>Candidatus</i> Phytoplasma prunorum’ in Selected <i>Prunus</i> Species Revealed by Quantitative PCR

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The work is devoted to the seasonal dynamics of Candidatus Phytoplasma prunorum in various tissues of four Prunus species using quantitative PCR and multilocus genotyping. The study addresses the practically important problem of optimizing the diagnosis of European stone fruit yellows (ESFY) and is of interest both for fundamental phytopathology and for applied diagnostics of planting material. Of particular note is the use of absolute quantitative PCR, monthly monitoring, and an attempt to compare the seasonal dynamics of phytoplasma titer with the genotypes of the pathogen and the symptoms of the disease.

However, despite the relevance of the topic, the manuscript has a number of significant limitations, primarily related to experimental design, statistical interpretation of results, and excessive extrapolation of conclusions. In its present form, in my opinion, the article requires significant revision (Major Revision).

Main remarks

1. The main limitation of the research is the experimental design. The work is based on a total of 21 infected trees of four Prunus species, with each species represented by two varieties and 2-3 trees. This sample size significantly limits the statistical power of the study, especially when analyzing seasonal differences and interspecific comparisons.

2. The factors "type", "grade" and "rootstock" are completely mixed together. The authors themselves note that the rootstock effect was not interpreted as an independent factor, since it is related to the species and variety. However, a similar problem applies to the comparison of species: the differences may reflect the characteristics of specific varieties or rootstocks, rather than biological differences between Prunus species. Consequently, the formulations about species-specific patterns should be significantly softened.

3. The study covers only one growing season (2025). Therefore, the revealed seasonal patterns cannot be considered stable without confirmation in subsequent years. The interannual variability of weather conditions can significantly affect the distribution of phytoplasma.

4. The authors repeatedly use the term "seasonal dynamics", but actually describe the changes only in one year. For such formulations, it is advisable to have long-term observations or to explicitly emphasize that the results relate exclusively to the conditions of one season.

5. The analysis of genetic diversity is limited to one genotyping of each tree. The detection of ten haplotypes is interesting in itself, but the sample is too small to draw any conclusions about the relationship between the phytoplasma genotype, titer, and symptoms. The authors correctly note this limitation in the abstract, however, the Discussion in some places still goes beyond the limits of the available evidence base.

6. It is necessary to justify in more detail the use of only four loci for genotyping. It should be discussed to what extent the chosen scheme really reflects the intraspecific diversity of 'Ca. P. prunorum' and whether there are more informative typing schemes.

7. When analyzing the titer of phytoplasma, negative samples were excluded, and points without detection were displayed separately. This approach is convenient for visualization, but it can shift the estimate of the average concentration towards higher values. It is advisable to discuss the possible impact of such a decision on the interpretation of seasonal dynamics.

8. The statistical model should be described in more detail. It is not entirely clear from the manuscript how repeated measurements on the same trees during the year were taken into account. If each tree was examined monthly, the observations cannot be considered independent, and this should be reflected in the statistical analysis.

9. The work pays great attention to the differences between plant organs, but the physiological mechanisms of phytoplasma redistribution are discussed rather briefly. It would be useful to link the results obtained more deeply with the seasonal physiology of phloem transport and changes in cambium activity.

10. The conclusion that roots are the most reliable material for year-round diagnostics requires more careful formulation. The practical applicability of this approach is limited by the fact that the selection of root material is much more difficult and traumatic compared to shoots. The diagnostic value and practical feasibility of such a selection should be discussed separately.

11. There is practically no discussion of the possible influence of the age of trees. The study used apricot trees about 20 years old, while the rest of the species were about 10 years old. The age of plants can potentially affect the distribution of phytoplasma and the intensity of symptoms.

12. Interpretation of symptoms requires more caution. The authors attribute the more pronounced symptoms in P. persica to the accumulation of phytoplasma, but the sample is small and the causal relationship has not been proven. More careful formulations should be used ("associated with", "may reflect").

13. The possible influence of the climatic conditions of the year under study should be discussed in more detail. Since the work was carried out in the same garden in South Moravia, the weather patterns of the season could significantly affect the identified patterns.

14. Discussion does not compare the results obtained sufficiently with similar studies of other phytoplasmas of the Apple proliferation group ('Ca. P. mali' and 'Ca. P. pyri'). Such a comparison would make it possible to better assess the commonality of the identified patterns.

15. The authors draw conclusions about the biological characteristics of individual Prunus species, but the sample includes a limited number of varieties. It should be emphasized that the results relate primarily to the studied varieties and do not necessarily reflect the characteristics of the species as a whole.

16. It would be useful to present the data not only in the form of averages, but also to show the individual dynamics of individual trees. This would allow us to assess the degree of intraspecific variability.

17. The reasons for the lack of phytoplasma detection in individual organs in some months should be discussed in more detail. It is unclear whether this reflects the actual absence of the pathogen, its concentration below the detection threshold, or the distribution of infection.

18. The conclusion is worded too broadly in some places. The results obtained do indeed demonstrate pronounced seasonal and tissue-specific variability of the phytoplasma titer, however, their use for the development of universal diagnostic recommendations requires confirmation on a larger number of varieties, rootstocks, regions and seasons.

Author Response

Dear Reviewer, please see the attached file with responses to your comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Comments to the Authors,

In this study, the authors clarified the seasonal dynamics of ‘Candidatus Phytoplasma prunorum’ in selected Prunus species by real-time PCR, Roots, one-year-old shoots and annual shoots were sampled monthly from 21 infected trees of Prunus armeniaca, Prunus domestica, Prunus persica and Prunus salicina. Phytoplasma titer was signiffcantly affected by host species, plant part and sampling month, with species- and tissue speciffc seasonal patterns. Roots provided the most consistent year-round detection, whereas above-ground tissues showed stronger seasonal variation. P. domestica displayed the most distinct pattern: roots remained frequently positive, while one-year-old shoots and particularly annual shoots had lower detection rates and lower phytoplasma titers. However, there some major revisions should be clarified or made in the manuscript. Comments as following:

The current plan only generalizes the construction of a multi-level monitoring and early warning system for diseases and pests, but lacks refined sampling standards and seasonal monitoring optimization strategies based on pathogen biological characteristics. Relevant studies have confirmed that phytoplasma pathogens infecting the crops show significant seasonal tissue colonization differences. Pathogens persist stably in root tissues throughout the year, while the pathogen titer and detection rate in above-ground shoots (annual shoots and one-year-old shoots) fluctuate greatly with seasons and host species. To improve the accuracy and stability of field monitoring and early warning, the project should supplement differentiated sampling specifications: take root tissues as the core perennial monitoring sample to avoid false-negative detection in dormant seasons, and adjust the sampling frequency and tissue types of above-ground shoots according to seasonal growth rhythms of the crops.
Pictures about the disease symptoms of the samples should be provided in the text.

Author Response

Dear reviewer, please see the attached file with responses to your comments.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This work is interesting and in general, experimental designing and implementation are well done. In complex the manuscript is good; the methodology is appropriate with correct analytical procedures. The figures and the tables are appropriate.

I would suggest discussing the genetic variants and the lineages of ‘Ca. P. prunorum’ strains identified according to what is already published. This part can be implemented and improved and should be also mentioned in the introduction, since genetic diversity in phytoplasma strains is important for their epidemiology.

The different behaviors in development can vary according to the species and maybe this can affect the differences in titer in the plant.

The role of the insect vectors is not mentioned in the manuscript. I would discuss the possible implications in the spreading, in the phytoplasma presence and titer. The detection in annual shoots can be influenced by the presence of insect vectors during the vegetative period actively feeding on the new shoots.

Following are my comments that might help to improve the manuscript.

Line 28: I prefer to use “is associated with” instead of “causal agent” (no Koch's postulates)

Line 97: Add the storage conditions of the sample collected

Line 105: How is the ‘Ca. P. prunorum’ identified? The sample can be infected also with other phytoplasmas. The primers don’t seem to be species specific. Explain this aspect.

Line 106: specify the names of the primers employed in the PCR and the target gene of the amplification and the level of specificity for the ‘Ca. P. prunorum’.

Line 122: I would use the term “lineage” instead of haplotype.

Line 132: which software has been used for the assembly and alignment?

Line 184: add space… 503tested

Line 200: indicate the meaning of the letters

Line 204: Specify what F stands for

Figure 2: I would add the months under every picture for a better visualization

Author Response

Dear reviewer, please see the attached file with responses to your comments.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I thank the authors for the substantial revision of the manuscript. The second version is significantly more rigorous from a methodological and interpretive perspective. In particular, the authors now clearly indicate that the differences between the studied groups cannot be unequivocally attributed to the Prunus species, since the species, variety, and rootstock are interrelated; they correctly account for repeated measurements on the same trees using mixed models; they describe the processing of negative qPCR results more transparently; they discuss the possible seasonal redistribution of phytoplasma with greater caution; and they also explicitly acknowledge the limitations of the small number of trees, the single plot, and the single year of observation. These changes address most of the fundamental issues with the original version.

In my view, the remaining issues no longer require any further substantial revision; however, it would be advisable to make a few clarifications before final acceptance.

The main remaining methodological issue is the analysis of the detection rate. The authors correctly use a random intercept for tree identity when analyzing the quantitative titer, but differences in detection rates are assessed using pairwise tests for equality of proportions. At the same time, the same trees were examined multiple times across months and tissues, so individual detection results are also not independent observations. A simple comparison of proportions does not take into account the repeated structure of the data. For a formal analysis of detection probability, a binomial generalized linear mixed model with tree identity as a random effect would be more appropriate. If the authors do not want to carry out additional analysis, the results of pairwise proportion tests should be clearly labeled as descriptive/auxiliary and should not be used for strong statistical conclusions.
The approach to quantitative analysis of only qPCR-positive samples is now well explained, and the authors rightly point out that the mean values are conditional on detection. Nevertheless, this has an important consequence for biological interpretation, especially for P. domestica, where only 5 out of 96 samples were positive in annual shoots. The average titer among these five positive samples cannot be interpreted as a characteristic of a typical annual shoot of this host group. In the Results, Discussion, and graphs, it is necessary to consistently distinguish between two characteristics: the probability of detection and the titer provided that detection has occurred.
The related issue concerns LMM for P. domestica. Given the extremely low frequency of positive annual-shoot samples, the model for the quantitative titer may rely on a very small number of observations and yield unstable estimates for individual month × plant-part combinations. It is necessary to check and, preferably, report the number of positive observations that actually entered the relevant models. The interpretation of quantitative differences for groups with very low detection frequency should be carried out with the utmost caution.
The authors indicate the analytical detection limit as 2 log10 cells g-1 phloem, but mark it as “data not shown”. Since the decision to exclude negative results from quantitative analysis is directly based on this limit, it is desirable to present the method for determining LOD/LOQ or the corresponding validation data in the Supplementary Materials. The range of the standard curve up to 10^1 copies µL-1 is not in itself a sufficient description of the analytical LOD.
The absolute quantitative assessment is expressed as “phytoplasma cells g-1 phloem”, although qPCR actually measures the number of copies of the 16S rDNA target relative to the plasmid standard. It is necessary either to justify the transition from copy number to cell number, including the number of copies of the corresponding rRNA operon in the genome of ‘Ca. P. prunorum’, or to use a more rigorous designation such as “target copies g-1 phloem”. This is especially important for an article whose central result is the absolute quantitative assessment.
Multilocus genotyping is now correctly limited by the authors as an additional characteristic of the material, and they rightly refrain from making statistical conclusions about the genotype–titer–symptom relationship. Nevertheless, selecting a single sample with the maximum titer from each tree means that the study does not allow for an assessment of possible intra‑tree or temporal variability of genotypes. It should be briefly noted that the assigned multilocus profile characterizes the detected dominant/amplified profile of the selected sample and does not rule out mixed infection or temporal genotype variation.
The authors have significantly improved the interpretation of the предполагаемого seasonal movement of phytoplasma and now explicitly state that the direction of movement was not measured. This is correct. However, the expression “progressive colonization of newly formed shoots” should still be used as an interpretation compatible with the data, rather than as an established process. The design itself represents successive slices of the titer distribution and does not track the movement of a specific phytoplasma population. The discussion of the possible influence of Cacopsylla pruni is interesting, but somewhat lengthy relative to the data presented in this work. Since the vector population, feeding sites, and new transmission events have not been studied, it is advisable to shorten this fragment to a brief alternative explanation so as not to dilute the main result of the study.
The practical conclusion about roots as the most stable material for year‑round diagnostics within the scope of this experiment is supported by the data. However, it is necessary to maintain the distinction between analytical reliability and practical convenience. Root sampling is more labor‑intensive and potentially traumatic, so it is advisable to formulate the recommendation as using root tissue when there is a high risk of false‑negative results in above‑ground tissues, rather than as a universally preferred material for routine diagnostics.
The conclusion in the new version is generally correctly limited to one site and one year. I recommend saving