MANUSCRIPT NUMBER: MS: 1469101899129162
TITLE: Influence of IFN-gamma and its receptors in human breast cancer
AUTHORS: Garcia-Tuñón, Ignacio; Ricote, Monica; Ruiz, Antonio; Fraile, Benito; Paniagua,
Ricardo; Royuela, Mar.
CHANGES MADE FOLLOWING THE REVIEWER-1 COMMENTS:
Major Compulsory Revisions (that the author must respond to before a decision on
publication can be reached).
1. While not a scientific issue, the article needs to be re-written to give better format,
clarity, and to minimize typographical errors to increase readability to readers. Here
are some examples: the "MATERIALS AND METHODS" section need to have headings
for each experiment, e.g., "Western Blot Analysis" "Immunohistochemistry staining",
etc. In the "RESULTS" section, the headings should a summary of the finding, rather
than the method used. There are also too many typographical errors: e.g. "Four"
instead of "For" on page 7, 1st paragraph; "star" rather than "start" on page 8, the
10th line from bottom; "have been reported" rather than "have reported" on page 9,
line 4; "together the nuclear ..." rather than "together WITH the nuclear ..." in abstract
and on page 10, last paragraph. Some statements are not clear, e.g., "... could be
critical in the tumor response" - response to what? PCNA needs to be defined. This kind
of editorial changes could make the article more understandable.
The manuscript has been revised by an English-speaking colleague, and we will
please to receive editorial assistance. In this way, several typographical errors have
been corrected. Although, “material and methods” and results sections have been rewritten and divided into small paragraphs with subheading. In order to clarify the
text, a new section (abbreviations section) has been included in this version (pages
12), following Instructions for BMC Cancer authors published in BioMed Central
web. Also we changed the abstract and conclusion section the sentence “could be
critical in the tumor response” by “could be a tumoral cell response” (page 2, line
24; page 11, line 25).
2. In Figure 1: I cannot appreciate much difference between the bands for benign
lesions vs in situ carcinoma for IFNr, yet, in table 1, the density showed a two-fold
difference. Please recheck the numbers.
In our study, all the analysis (western blot, for examples) was made five times. The
optical density of the five western blot analyses was measured by image analysis
software and was included the statistical analyses: means and SD for each breast
group and antibody used. In Table 1, we showed an increase to IFN, from 9.75 to
20.25. In Figure 1, it is true that cannot appreciate much difference between bands
for benign lesions vs in situ carcinoma. In order to clarify the results, Figure 1 has
been changed.
3. In table 1, IFNr-Ra for in situ-carcinoma is 21.1+/-3.6 (17.5-24.7), for IC it is 13.51
+/- 4.2 (9.31-17.7). The confidence interval overlap. Please re-check statistical
calculation to see if P value is indeed < 0.05.
We have re-check the statistical analysis and we have correct a typography error in
table 1: IFN-R for in situ carcinoma is 21.1+/-3.6 (17.5-24.7) and for IC it is 13.51
+/- 2.2 (11.31-15.7).
4. In situ carcinoma has "higher intensity" but "lower percentage of positive cases" for
IFNr. What is the implication of this discrepency? How was the conclusion "These data
suggest a loss of IFN-r antitumoral activity in these tumors" drawn?
The implication of to find higher intensity immunoreactions to IFN but in a lower
number of patients, could be interpret as that the most of the samples has lost the
IFN expression, but in the few cases that there is IFN expression, it is acting against
the exceed of proliferation of tumor cells, as occurs in normal epitheliums.
We found IFN-R in 94% in situ samples located in nucleus, and it was reported
that it is a membrane receptor, and is necessary the binding of two receptors and
IFN in membrane to start the signalling pathway of IFN and to produce its cell
effects. So the nuclear location of IFNr-beta suggests that it is not available to bind
the IFN, and so it can not exerts its effects.
5. The study showed IFNr has higher intensity in in situ tumors than in benign lesions,
and has same intensity in infiltrative tumors have similar to benign breast lesions. How
do these observation suggest important roles of IFNr? It was also speculated that "IFNr could be non-functional". With these two points, how was the conclusion "present
study suggest that IFN-r could be a potential therapeutic tool in breast cancer"
obtained?
We also showed a great decrease of the number of patients that expressed IFN
from benign lesions (70%) to tumor samples (just 35-40%), and it could be an
evidence of the protecting role of IFN against the abnormal proliferation that occurs
in normal tissues. The fact of IFN could be non-functional in this tumor is due to
the abnormal location in nuclei observed for IFN-R: It is known that is necessary
the presence of both receptors in the membrane to exerts its effects in the epithelial
cells. So, with this data, it is true that present study suggests that IFN could be a
potential therapeutic tool in breast cancer. In this way, several authors described and
evaluated the clinical role of IFN in overcoming tamoxifen as a therapy in the
treatment of metastatic breast cancer. At the same time we think that it would be
interesting more studies about the location of the IFN-receptors in tumor tissues, to
elucidate the status of the IFN receptors in order to know the possible cellular
response to IFN treatment.
Minor Essential Revisions (such as missing labels on figures, or the wrong use of a
term, which the author can be trusted to correct)
1. In the tables 1 and 2. the denote of "a" and "b" should be defined.
In this new version, in Tables legends sections (new section added in this version,
following reviever-3 comments), we described that “a”, “b” or “c” (superscript
letter) indicate differences significant between the different patients groups. If
appeared the same superscript, are not statistically different for each group.
Statistical analysis refers to each antibody separately.
CHANGES MADE FOLLOWING THE REVIEWER-2 COMMENTS:
Major Compulsory Revisions (that the author must respond to before a decision on
publication can be reached).
1. The data as presented are descriptive in nature. While the authors offer a potential
explanation for the differential expression of IFNg and its receptor in various types of
breast cancer, without additional experimentation the claims are speculative. For
example, the authors could evaluate co-localization of IFNg and its receptor units and
perhaps perform functional experiments testing the antiproliferative/apoptotic effect of
IFNg on receptor positive versus negative breast cancer cells, and/or evaluate signal
transduction elements downstream of the receptor. Several of these aspects are
mentioned in the discussion, but they could be experimentally tested.
Different authors studied the clinical role of IFN in overcoming tamoxifen as a
therapy in the treatment of metastatic breast cancer or studied the effect product
when different breast cancer cell lines were treated with IFN. This cytokine is
secreted by several blood cells but in the last decade several authors described IFNg
and its two receptors in different tissues as endothelial cells, fibroblast, neuronal
cells, prostate cancer cells… At the present, and in our Knowledge, there are no
studies of IFN and its receptors expression reported by inmunohistochemistry and
Western blot, in different non malignant and carcinomatous human breast tissue. We
think that these data are descriptive but necessaries in order to know the possible
cellular response to IFN treatment. The authors also evaluate co-localization of IFN
and its receptors (IFN-R and IFN-R) as we showed in this additional Figure.
Although, these photos are not included in the manuscript because we think that are
not new relevant data in the results section.
A-B Inmunostainig to IFN- (black) and IFN-R (red) in benign (A) and in infiltrating cancer
(B). C-D Immunostainig to IFN- (black) and IFN-R (red) in in situ (C) and in infiltrating breast
cancer (D).
Minor Essential Revisions (such as missing labels on figures, or the wrong use of a
term, which the author can be trusted to correct)
The M&M section could be divided into small paragraphs with subheadings to improve
clarity.
In order to clarify the manuscript, the “material and methods section has been rewritten and divided into small paragraphs with subheadings
CHANGES MADE FOLLOWING THE REVIEWER-3 COMMENTS:
Major comments:
1. This manuscript is poorly written and contains numerous spelling and grammatical
errors throughout the text. In many instances, statements within the manuscript are
ambiguous, incomprehensible, and inaccurate. For example, in the Introduction
Section (last paragraph). The authors state that:
“At present and in our knowledge, no studies of IFN- in human breast cancer have
been reported.”.
However, in the previous paragraph, the authors state:
“In breast cancer patients with skin metastasis, local injection of IFN- results in the
total or partial regression of the skin lesions [15]. Other authors have been showed
that IFN- increases the growth inhibitory effects of tomaxifen in breast metastatic
carcinomas [16-17]. In this way, it has been show that IFN- produce this
antitumoral effect up-regulating the expression of p21 and resulting cell cycle arrest
in breast cancer cell lines [18].”
Such statements are contradictory, disjointed and appear too frequently throughout the
manuscript. This, in part, conceals and obstructs many of the important points that the
authors need to convey about their data.
We agree with your observation and we have check and correct these contradictory
sentences (page 3; lines 22-24). There are several authors that evaluated the clinical
role of IFN in overcoming tamoxifen as a therapy in the treatment of metastatic
breast cancer (Macheledt et al., 1991; Seymour and Bezwoda, 1993). Gooch et al.
(2000) studied the different effects product to IFN in two breast cancer cell lines
and its relation with another factor as p21 or STAT1. At the present, and in our
Knowledge, there are no studies of IFN and its receptors expression reported by
inmunohistochemistry and Western blot, in different non malignant and
carcinomatous human breast tissue.
2. Another major problem with this manuscript is that their data is merely descriptive
and does not support their conclusions. Although potentially interesting, the study
design and data obtained make it difficult to accurately attribute with confidence their
conclusions that IFN- and its receptors influence breast tumor development. Their
conclusions are highly speculative and the data does not adequately and/or reasonably
address their hypothesis.
Following your suggestion we have changed the conclusion and delete the
speculating sentences
3. With respect to the data in Table 1, have the authors attempted to analyze the data by
normalizing their results to the Actin “house keeping” protein concentrations (i.e. used
as internal controls) to obtain “relative units” for comparative analysis? This may
actually strengthen their results and provide a more accurate and reliable
interpretations of the data.
In order to normalize western blot data we measured the total protein in each sample
through the Bradford method, and then we used the same protein amount for each
sample (50 g/ml), so we obtained a semi-quantitative analysis. Actin is just another
internal control to ensure that the protein measure was right, and all samples has the
same total protein quantity. This is mentioned in material and methods section,
western blot analysis (page 4, lines 31-32; page 5, lines 15, 18 and 21-22)
Minor comments:
1. The figure legends and table footnotes are somewhat anemic. The reader should be
able to interpret how the experiments were accomplished and analyzed without having
to review the main-body text. Each table and figure should be able to “stand alone”.
The table footnotes and legends need to be more detailed about experimental procedures
and methods of data presentation.
The figure and table legends (new section added in this version) have been rewritten
and more detailed about experimental procedures and methods data has been
included.
2. The author needs to initially spell out the abbreviations used in the text (i.e. PCNA,
TUNEL, IFN-, IFN-R, etc…). Once they are established, the authors can use the
abbreviation throughout the remainder of the paper.
We have spell out all abbreviations in the manuscript. In this way, a new section has
been added in this version (page 12), following Instructions for BMC Cancer
authors published in BioMed Central web page
3. Their nomenclature and abbreviations are inconsistent throughout the text.
A new section (abbreviations section) has been added in this version (page 12),
following Instructions for BMC Cancer authors published in BioMed Central web
page
4. On the title page, the corresponding authors should attach their highest degree of
training (for example, Dr. Mar Royuela).
We have included the highest degree of all authors.