A Microimage-Processing-Based Technique for Detecting Qualitative and Quantitative Characteristics of Plant Cells
Round 1
Reviewer 1 Report
Automated comparison of plant cell samples in terms of cell abundance, protein tag content, and morphometric characteristics is an important task for evaluating the genetic transformation efficiency and stress tolerance of cell lines. The authors describe their advances in computer-operated analysis of fluorescence images of Dunaliella microalga and Arabidopsis protoplasts, which might be related to plant breeding purposes. The manuscript can be published, although after an extensive revision.
The manuscript title is too general which makes it misleading. Plant cells contain a number of fluorescent substances (Chl, NAD(P)H, phycobilins, etc.) and, in addition, they can carry GFP or other fluorescent labels. Without a note on fluorescence origin in the title, the reader may expect a different content in the main text. The work deals with morphometric characteristics, such as the cell size, diameter, and cell counts, but this is not clear from the title.
A work part related to chlorophyll fluorescence is rather weak. The authors measured Chl fluorescence with a hope to identify healthy functionally active cells (lines 378-380; 423-424, 460-461; 467-468), but their belief is incorrect. The photosynthetic activity is assessed with a special method, called PAM, i.e., pulse-amplitude-modulated (micro)fluorometry. The method is based on measuring Chl fluorescence under weak (actinic) light and saturating light pulses; these data provide information on the condition of photosynthetic apparatus and electron transport rates. When Chl fluorescence is measured under high-intensity light, as probably is the case in the present study, the photosynthetically active and inactive (even poisoned) cells can show equally bright fluorescence. Thus, it is not surprising that Chl fluorescence of salt-treated and untreated cells showed no significant difference in these experiments.
The ACFVA method proposed in this study is based on measuring green (GFP) fluorescence and red (chlorophyll) fluorescence (see lines 157-159). However, the formula given at lines 175-176 includes blue (B) fluorescence in addition to green (G) and red (R) emissions. Why blue fluorescence is taken into account here, given that blue light is entirely cut off by band-pass filters and is not measured? It is not clear why the formula is designated XX and why it is given without a reference to its source.
Table S1 is not adequately described. There are no comments on how the values in columns were obtained. The term “chloroplast intensity” in this table needs editing: “chloroplast fluorescence intensity” was probably meant. The terms “information entropy” and “information utility” are not even mentioned in the main text. In which way these quantities are helpful; are they really needed?
Experiments were mostly performed with Arabidopsis protoplasts and Dunaliella cells. However, the Introduction does not explain why these materials were chosen. The description of plant growth conditions and vector construction in Materials and Methods is presented without any comments on the significance of these species within the framework of this study. The suitability of these materials is commented in Results, but these comments should appear earlier, already in the Introduction.
What is the independent variable (plotted along x axis) for straight lines in Fig. 2C and 3 (the right column)? I agree that the data points are defined, but the line segments between the points have no physical meaning in my view.
Lines 447-449. The authors claim that ACFVA in combination with Arabidopsis protoplast is a reliable system for the "intuitive and accurate" detection of NaCl in solutions. In order to support this claim, a calibration curve should be presented. Surely, the presence of NaCl in solutions can be reliably detected by other means, e.g., ion-selective electrodes or flame photometry.
The quality of English writing is not uniform and needs editing. For example, we read at lines 79-80: “researchers …need to identify characteristics of plant cell by humans.” It sounds like researchers use humans in their work. Please check grammar at lines 139-140; 225; 287-288; 298-299 (“…diameters are more stable and accurate than the Cellprofiler”; 307-312 (mannual -> manual); 324-326; 331-333 (coefficient was lower than ACFVA); 338-339; 354-355 (“Moreover” is a wrong word choice); 357 (“make ACFVA can”?); 374 (“positive fluorescent light”?); 447-448 (“ACFVA can… intuitively …find liquids with NaCl”; 477, spelling).
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper “An image-processing-based technique for detecting quantitative fluorescence in plant cells”, authors Jun Feng et al. is an encouraging trial to propose a new method for automatic cell analysis.
1. Why only plant cells? Fluorescence parameters may be used for the other cell types as well.
The authors used only plant cells without cell wall. More generally, plant cells contain the cell wall which should be taken into consideration.
2. “The significant errors of Dunaliella cell and cowpea pods cell numbers counted by CellProfiler may be due to their irregular cell shape and contaminants, which are similar to the target cells in some ways.”
The protoplasts also vary in shape and size.
3. “The chloroplast fluorescent value was also quantified by 460 ACFVA to evaluate the cells' activity of two systems.”
How was calculated chloroplast fluorescent value? The method quantifying red and blue fluorescence in plant cells and using their ratio for different purposes is not new (see Z.G. Cerovic et al., Ultraviolet induced fuorescence for plant monitoring: present state and prospects. Agronomie, EDP Sciences, 1999, 19 (7), pp.543-578). Also, fluorescence of the plant cells obtained on the microscope was used in the other studies for quantitative purposes (see A. Savić et al., Fluorescence-Detected Linear Dichroism of Wood Cell Walls in Juvenile Serbian Spruce: Estimation of Compression Wood Severity, Microsc. Microanal. 22, 361–367, 2016).
In this manuscript, there is no enough information about the method applied, especially after cell binarization, i.e. in the part about using internal references.
Also, using cell binarization and further analysis of such images, in the case of plant cells, has been used before as a new proposed automatic method for (plant) cell analysis (A. Nedzved et al., Automatic image processing morphometric method for the analysis of tracheid double wall thickness tested on juvenile Picea omorika trees exposed to static bending, Trees volume 32, pages 1347–1356, 2018).
The main drawback of the manuscript is its novelty. The provided information about the new method is poorly explained. Besides, the authors did not make a satisfactory inspection of the literary data existing in this field. Therefore, they are not convincing that the method that they propose is really new.
My opinion is to reject the paper. If the authors complete their study, they are encouraged to submit the new version of the manuscript.
Minor editing of English language required
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors added some text, but it is not clear yet how they use fluorescence to detect cells’ characteristics, so that it is a new method. There is a lot of text, but the needed information is not abundant. They show many examples – the images and final results. Fluorescence of plant cells is complex, including emission from chloroplasts and cell walls. There is also the possibility of overlapping bands of various compounds. The authors should include a simple scheme on one cell example, to show how the method works. They claim that they have patent applications and cannot write on these subjects in this paper. However, if the patent is under application or protected intellectual property, the subject or a part of it can be included in a published manuscript.
Based on the above, I recommend that the authors simplify the explanations and make a scheme in which they will explain all steps included in the method.
Moderate editing of English language.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
