Phenotypic and Molecular Characterization of ESBL-Producing Enterobacterales from Intestinal Colonization in Loja, Ecuador

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Only 89 of the 195 ESBL-producing isolates were available for molecular characterization. The authors should clarify whether these isolates are representative of the entire ESBL-positive collection and provide a comparison between characterized and non-characterized isolates to assess potential selection bias.

 

The molecular analysis was limited to a selected panel of resistance genes. Additional discussion should be included regarding other possible resistance mechanisms that may explain the observed phenotypes, particularly the high frequency of fluoroquinolone resistance.

 

The statistical analysis requires additional detail: confidence intervals for odds ratios, which p-values were subjected to multiple-testing adjustment, and whether all reported pairwise comparisons were included in the adjustment.

 

Several resistance estimates are based on extremely small denominators, including estimates of 100% based on 1/1 isolates. These estimates should be explicitly identified as descriptive observations rather than robust prevalence estimates, and the very small denominators should be emphasized when interpreting differences between species or healthcare settings.

 

This study is explicitly about intestinal colonization but elsewhere the manuscript moves into the direction of infections, e.g., “E. coli was the predominant species particularly among outpatients, whereas K. pneumoniae was more frequently recovered from ICU patients, supporting the distinct epidemiological patterns of community and healthcare-associated infections.” The manuscript should consistently distinguish intestinal colonization from clinical infection.

 

The methodology for phenotypic ESBL confirmation requires further clarification. The manuscript states that double-disc synergy testing and combination discs were used, but it does not specify the interpretive criterion used to define an ESBL-positive isolate. In particular, please report the required difference in inhibition-zone diameter between the antimicrobial tested alone and in combination with the β-lactamase inhibitor.

 

Rather than just a reference to CLSI, please provide the breakpoint values used to improve methodological transparency and reproducibility.

Author Response

Dear Editor and Reviewers of Antibiotics,

 

Many thanks for all your comments and suggestions, which have been used to improve our manuscript entitled “Phenotypic and Molecular Characterization of ESBL-Producing Enterobacterales from Intestinal Colonization in Loja, Ecuador”. This new version has been updated following the recommendations of the reviewers.  Please, find below the point by point answers to the reviewer comments.

All modifications from previous version are highlighted in yellow in the manuscript.

 

Reviewer comments:

Reviewer #1 (Comments for the Author):

- Only 89 of the 195 ESBL-producing isolates were available for molecular characterization. The authors should clarify whether these isolates are representative of the entire ESBL-positive collection and provide a comparison between characterized and non-characterized isolates to assess potential selection bias.

Authors: We thank the reviewer for this important observation. Of the 195 ESBL-producing isolates, 89 (45.6%) could be recovered after storage and were therefore available for molecular characterization, whereas 106 (54.4%) could not be recovered. To assess potential selection bias, characterized and non-characterized isolates were compared according to sex, age group, and healthcare setting. No significant differences were observed in sex distribution (p = 0.637; Cramér’s V = 0.035) or age group (p = 0.100; Cramér’s V = 0.181). Healthcare-setting distribution differed between the two groups (p = 0.043; Cramér’s V = 0.170), with ICU isolates proportionally more frequent among the characterized isolates (11.2% vs. 2.8%). These findings indicate that the characterized isolates were broadly comparable to the non-characterized isolates with respect to sex and age group, although a modest overrepresentation of ICU isolates was observed. Species distribution was summarized descriptively as different identification methods were used in the two groups, precluding a reliable statistical comparison.

These findings have been incorporated into Section 4.3 (Inclusion and Exclusion Criteria), the opening paragraph of Section 2 (Results), and the Limitations paragraph in Section 3 (Discussion). The detailed comparison is provided in Supplementary Table S2.

- The molecular analysis was limited to a selected panel of resistance genes. Additional discussion should be included regarding other possible resistance mechanisms that may explain the observed phenotypes, particularly the high frequency of fluoroquinolone resistance.

Authors: We thank the reviewer for this important comment. We agree that the molecular panel used in this study was limited and does not allow other resistance mechanisms to be excluded. Accordingly, we have expanded the Discussion section to explicitly address this limitation, and to discuss possible mechanisms underlying the high frequency of fluoroquinolone resistance observed in our collection.

In particular, fluoroquinolone resistance was not molecularly characterized in this study. In Enterobacterales, this resistance is commonly associated with chromosomal mutations in the quinolone resistance determining regions of gyrA and parC, which were not investigated here. In addition, plasmid-mediated quinolone resistance determinants, including qnr, aac(6′)-Ib-cr, oqxAB, may contribute to reduced fluoroquinolone susceptibility and can be carried on plasmids also harbouring ESBL determinants including bla_CTX-M. The potential co-selection of these resistance determinants is therefore discussed as a possible contributing factor to the high frequency of ciprofloxacin resistance observed in our study, although this hypothesis could not be assessed with the molecular data available.

This discussion has been added to the revised manuscript (highlighted in the attached file).

- The statistical analysis requires additional detail: confidence intervals for odds ratios, which p-values were subjected to multiple-testing adjustment, and whether all reported pairwise comparisons were included in the adjustment.

Authors: We thank the reviewer for this important comment. The Statistical Analysis section has been revised to provide greater methodological detail and to clarify the approach to multiple testing.

We now specify the use of Fisher’s exact test for categorical associations, Monte Carlo simulation for sparse R × C contingency tables, and Cramér’s V as the measure of association strength. For clinically relevant 2 × 2 comparisons, odds ratios are now reported together with their corresponding 95% confidence intervals. Regarding multiple testing, no post hoc pairwise comparisons were performed to support the primary inferential conclusions. Therefore, no multiplicity adjustment was applied to the p-values from the global Fisher’s exact tests. The corresponding Results and table notes were revised accordingly.

- Several resistance estimates are based on extremely small denominators, including estimates of 100% based on 1/1 isolates. These estimates should be explicitly identified as descriptive observations rather than robust prevalence estimates, and the very small denominators should be emphasized when interpreting differences between species or healthcare settings.

Authors: We thank the reviewer for highlighting this important limitation. We agree that several resistance estimates are based on very small denominators and should be interpreted as descriptive observations rather than as robust estimates of resistance prevalence.

To address this point, a footnote was added to Table 2 explicitly stating that resistance percentages calculated from strata containing fewer than five isolates are descriptive only and should not be interpreted as reliable prevalence estimates. This clarification is particularly relevant for estimates of 100% based on a single isolate (e.g., E. coli in hospitalization, K. oxytoca in the outpatient clinic, and E. hormaechei in the ICU).

We have also revised the Results and Discussion to emphasize that comparison between species or healthcare settings involving very small subgroups should be interpreted with caution and that such percentages may be highly unstable. No conclusions regarding differences between species or healthcare settings are drawn from these small denominators.

In addition, Figure 1 was updated following recalculation of the statistical analyses using the final dataset. The revised figure includes the corresponding global and stratum-specific p-values, together with the final denominators and percentage distributions. To facilitate visual interpretation, strata containing  fewer than five isolates are now identified with a dashed border (Hospitalization–Female, n = 1). Leader lines were also added to label the two minority species profiles in the female outpatient-clinic group (K. oxytoca and E. hormaechei), which were previously difficult to identify due to the limited width of their segments.

These changes have been incorporated into the revised manuscript and in Table 2. 

- This study is explicitly about intestinal colonization but elsewhere the manuscript moves into the direction of infections, e.g., “E. coli was the predominant species particularly among outpatients, whereas K. pneumoniae was more frequently recovered from ICU patients, supporting the distinct epidemiological patterns of community and healthcare-associated infections.” The manuscript should consistently distinguish intestinal colonization from clinical infection.

Authors: We thank the reviewer for this observation. The entire manuscript was reviewed to verify consistent use of the terms ‘colonization’ and ‘infection.’ We identified one sentence in the Conclusions section that incorrectly referred to ‘community and healthcare-associated infections’ when describing the findings of the present study. As the study specifically investigated intestinal colonization, this sentence has been corrected to accurately reflect the colonizing, rather than infectious nature of the analysed isolates. The remaining references to ‘infection’ were reviewed and retained only where they refer to the broader epidemiology of infections described in the referenced literature (Introduction) or to established terminology in the context of infection prevention and control programs (Discussion).

The corresponding sentence in the Conclusions has therefore been revised to ensure that the findings of this study are consistently presented in the context of intestinal colonization.

- The methodology for phenotypic ESBL confirmation requires further clarification. The manuscript states that double-disc synergy testing and combination discs were used, but it does not specify the interpretive criterion used to define an ESBL-positive isolate. In particular, please report the required difference in inhibition-zone diameter between the antimicrobial tested alone and in combination with the β-lactamase inhibitor.

Authors: We thank the reviewer for this observation. The description of the methodology used for phenotypic ESBL detection has been expanded and clarified. Specifically, for the double-disc synergy test, we now specify that a 20/10 µg amoxicillin/clavulanic acid (AMC) disc was used, and that distortion or expansion of the inhibition zone toward the AMC disc (keyhole effect) was interpreted as indicative of ESBL production. For the combination-disc test, we have explicitly stated that an increase of ≥5 mm in the inhibition-zone diameter between the antimicrobial agent tested alone and in combination with clavulanic acid was considered indicative of ESBL production.

These methodological details have been incorporated into Section 4.4, Antimicrobial Susceptibility Profile, Phenotypic Detection of ESBL, and Carbapenemases, of the manuscript.

- Rather than just a reference to CLSI, please provide the breakpoint values used to improve methodological transparency and reproducibility.

Authors: We thank the reviewer for this suggestion. Supplementary Table S1 has been added, presenting the inhibition-zone diameter breakpoints used to classify the 10 antimicrobials evaluated, according to CLSI M100, 34th edition (2024), already cited as reference [37] in the manuscript. An explicit reference to the new supplementary table has also been added to the Methods (Section 4.4), to facilitate access to these criteria.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript presents potentially valuable data on intestinal carriage of ESBL-producing Enterobacterales in Loja, Ecuador, particularly because molecular characterization of ESBL determinants and the exploratory use of FT-IR provide complementary information. I support its possible further processing after appropriate modifications/ clarifications as outlined below:

Comment 1: The authors must better justify the selection of isolates for molecular and FT-IR analyses needs better justification. Of 195 ESBL-producing isolates, only 89 were subjected to molecular characterization because of viability/recovery after storage. This represents a substantial reduction and may introduce important selection bias. The authors acknowledge this limitation, but they should provide more information comparing the 89 characterized isolates with the remaining 106 isolates, if possible, particularly regarding species, healthcare setting and patient characteristics. It should also be clearly stated why FT-IR analysis was performed on 87 rather than 89 isolates.

Comment 2: In the reviewer opinion the authors must revise the terminology concerning “clonal relationship” and and FT-IR throughout the manuscript. The Abstract states that “clonal relationship characterization was determined by FT-IR,” which is too strong. The manuscript later appropriately states that FT-IR-defined groups are phenotypic spectral clusters and not confirmed clonal lineages. This distinction should be maintained consistently, including in the Abstract and Conclusions. FT-IR may suggest phenotypic relatedness, but without WGS, MLST or another validated high-resolution typing method, clonality cannot be established.

Comment 3: Please provide some additional methodological detail about FT-IR methodology. The authors should provide sufficient information to allow reproducibility, including spectral preprocessing/normalization, distance metric, clustering parameters and the criteria used to define a cluster and to classify a replicate/isolate as “robust” or “non-robust.” The current statement that cluster assignment was based on the Bruker software and distance matrix is insufficient.

Comment 4: There is an inconsistency in the antimicrobial susceptibility testing methodology. The Methods state that 11 antimicrobials were evaluated, including ampicillin, whereas Table 2 presents results for only 10 antimicrobials and does not include ampicillin. This should be corrected or explained. In addition, the fosfomycin methodology requires clarification, particularly the disk concentration used and the applicability of the selected breakpoint to the tested organisms. Also, the molecular characterization should distinguish detection of resistance genes from identification of ESBL variants. Detection of blaTEM or blaSHV by PCR does not by itself establish that the detected gene encodes an ESBL, since non-ESBL variants such as TEM-1 may also be detected. The terminology in the Results and Table 1 should therefore be revised to distinguish “β-lactamase genes” from confirmed ESBL determinants. Similarly, “blaCTX-M-1 group enzymes” should preferably be described as blaCTX-M-1-group genes unless sequencing was performed to identify specific variants.

Comment 5: The statistical analysis need some clarification. Given the very small numbers in several categories (particularly hospitalization and ICU groups), the authors should be cautious when interpreting statistically significant associations. The manuscript should report confidence intervals for important effect estimates, particularly the OR of 13.2 for blaSHV carriage in K. pneumoniae versus E. coli. The rationale for the selected pairwise comparisons and the application of Holm correction should also be described more clearly.

Comment 6: Table 1 is difficult to interpret because the gene-profile categories and percentages are presented without sufficient explanation of whether percentages represent the total 89 isolates or the corresponding subgroup. The table should be reformatted for clarity.

Last, but not least, please standardize terminology throughout: “healthcare setting,” “health service,” “outpatient clinic,” and “hospitalization” are used somewhat inconsistently.

Author Response

Dear Editor and Reviewers of Antibiotics,

 

Many thanks for all your comments and suggestions, which have been used to improve our manuscript entitled “Phenotypic and Molecular Characterization of ESBL-Producing Enterobacterales from Intestinal Colonization in Loja, Ecuador”. This new version has been updated following the recommendations of the reviewers.  Please, find below the point by point answers to the reviewer comments.

All modifications from previous version are highlighted in yellow in the manuscript.

 

Reviewer #2 (Comments for the Author):

The manuscript presents potentially valuable data on intestinal carriage of ESBL-producing Enterobacterales in Loja, Ecuador, particularly because molecular characterization of ESBL determinants and the exploratory use of FT-IR provide complementary information. I support its possible further processing after appropriate modifications/clarifications as outlined below:

- Comment 1: The authors must better justify the selection of isolates for molecular and FT-IR analyses. Of 195 ESBL-producing isolates, only 89 were subjected to molecular characterization because of viability/recovery after storage. This represents a substantial reduction and may introduce important selection bias. The authors acknowledge this limitation, but they should provide more information comparing the 89 characterized isolates with the remaining 106 isolates, if possible, particularly regarding species, healthcare setting and patient characteristics. It should also be clearly stated why FT-IR analysis was performed on 87 rather than 89 isolates.

Authors: We thank the reviewer for this observation, which coincides with a similar point raised by Reviewer 1. Of the 195 ESBL-producing isolates identified in the study, 89 (45.6%) could be recovered after storage and were therefore available for molecular characterization, whereas the remaining 106 (54.4%) could not be recovered for confirmatory testing. In a previous version of the manuscript, this number was inadvertently reported as 87; this has been corrected to 89 throughout the manuscript to reflect the isolates actually included in the molecular analysis.

To assess potential selection bias, the characterized and non-characterized isolates were compared according to sex, age group, and healthcare setting. No significant differences were observed in sex distribution (p = 0.637; Cramér’s V = 0.035) or age group (p = 0.100; Cramér’s V = 0.181); however, the distribution by healthcare setting differed significantly between the two groups (p = 0.043; Cramér’s V = 0.170), with a higher proportion of ICU isolates among the characterized group (11.2%) than among the non-characterized group (2.8%). Species distribution was summarized descriptively rather than formally compared, because different identification methods were used for the two groups (MALDI-TOF MS for characterized isolates and Enterosystem 18R for non-characterized isolates).

Accordingly, although the two groups were comparable with respect to sex and age, the molecularly characterized subset should not be considered fully representative of the complete collection of ESBL-positive isolates, given the observed overrepresentation of ICU isolates. This potential selection bias has now been explicitly acknowledged as a limitation of the study.

The FT-IR analysis was performed on all 89 molecularly characterized isolates; the number 87 reported in the previous version was a typographical error, which has been corrected throughout the manuscript.

These changes were incorporated into Section 4.3 (Inclusion and Exclusion Criteria), the opening paragraph of Section 2 (Results), and the Limitations paragraph of Section 3 (Discussion). The detailed comparison between characterized and non-characterized isolates is presented in Supplementary Table S2.

- Comment 2: In the reviewer’s opinion the authors must revise the terminology concerning “clonal relationship” and FT-IR throughout the manuscript. The Abstract states that “clonal relationship characterization was determined by FT-IR,” which is too strong. The manuscript later appropriately states that FT-IR-defined groups are phenotypic spectral clusters and not confirmed clonal lineages. This distinction should be maintained consistently, including in the Abstract and Conclusions. FT-IR may suggest phenotypic relatedness, but without WGS, MLST or another validated high-resolution typing method, clonality cannot be established.

Authors: We thank the reviewer for this observation. The terminology related to ‘clonal relationship’ and FT-IR was carefully reviewed throughout the entire manuscript to ensure that phenotypic spectral clustering is clearly distinguished from confirmed clonal relatedness.

In the Abstract, the statement ‘Clonal relationship characterization was determined by FT-IR’ was considered too strong and was replaced with ‘Phenotypic clustering was assessed by FT-IR.’. The Conclusions were also reviewed to ensure that FT-IR findings are not interpreted as evidence of clonality.

 Throughout the Results, Discussion, Methods, and Conclusions, FT-IR defined groups are consistently described as phenotypic spectral clusters rather than confirmed clonal lineages. We explicitly state that, in the absence of a validated high-resolution typing method such as WGS, MLST, or PFGE, FT-IR cannot establish clonality. The revised terminology therefore reflects that FT-IR may identify phenotypic relatedness but does not demonstrate clonal relationships.

- Comment 3: Please provide some additional methodological detail about FT-IR methodology. The authors should provide sufficient information to allow reproducibility, including spectral preprocessing/normalization, distance metric, clustering parameters and the criteria used to define a cluster and to classify a replicate/isolate as “robust” or “non-robust.” The current statement that cluster assignment was based on the Bruker software and distance matrix is insufficient.

Authors: We thank the reviewer for this important comment. We agree that the methodological description of the FT-IR analysis required additional detail to ensure reproducibility. Accordingly, we have expanded the Methods section to provide further information on spectral pre-processing and normalization, the spectral region analysed (1300–800 cm⁻¹), the distance metric and hierarchical clustering parameters, the criteria used for cluster definition, and the quality-control criteria used to classify replicates/isolates as robust or non-robust. We have also clarified the respective roles of principal component analysis (PCA), hierarchical clustering analysis (HCA) and distance matrix analysis in the workflow. These details have been added to Material and Methods section.

- Comment 4: There is an inconsistency in the antimicrobial susceptibility testing methodology. The Methods state that 11 antimicrobials were evaluated, including ampicillin, whereas Table 2 presents results for only 10 antimicrobials and does not include ampicillin. This should be corrected or explained. In addition, the fosfomycin methodology requires clarification, particularly the disk concentration used and the applicability of the selected breakpoint to the tested organisms. Also, the molecular characterization should distinguish detection of resistance genes from identification of ESBL variants. Detection of blaTEM or blaSHV by PCR does not by itself establish that the detected gene encodes an ESBL, since non-ESBL variants such as TEM-1 may also be detected. The terminology in the Results and Table 1 should therefore be revised to distinguish “β-lactamase genes” from confirmed ESBL determinants. Similarly, “blaCTX-M-1 group enzymes” should preferably be described as blaCTX-M-1-group genes unless sequencing was performed to identify specific variants.

Authors: We thank the reviewer for this detailed observation. The manuscript has been revised to address each of the methodological and terminological issues raised.

• Antimicrobial susceptibility testing. The apparent inconsistency regarding the number of antimicrobials was corrected. Ampicillin results have now been included in Table 2 (AM column). Fosfomycin was excluded from the panel of interpretable antimicrobials because the applicability of the selected disk diffusion methodology and breakpoint to the organisms investigated could not be reliably established. Thus, the final panel presented in Table 2 comprises 10 interpretable antimicrobials, including ampicillin. 
• β-Lactamase genes and ESBL determinants. We agree that detection of bla_TEM-1 and bla_SHV-12 by PCR alone does not establish that the detected gene encodes an ESBL. In our study, the bla_TEM-1 and bla_SHV-12 variants identified by PCR were confirmed by Sanger sequencing. The purified PCR product was sequenced at Macrogen Europe (Macrogen Inc., Amsterdam, the Netherlands) using Applied Biosystems 3730XL high-throughput sequencers. This methodological information has been incorporated into the corresponding section of the manuscript. To avoid implying that all detected β-lactamase genes are ESBL determinants, the Table 1 heading has been changed from ‘ESBL Genes’ to ‘β-Lactamase Gene Profile’. The manuscript has also been revised to distinguish β-lactamase gene detection from confirmation of specific ESBL variants. Therefore, in isolates in which bla_TEM-1 was the only gene detected (7, 7.9%), the molecular determinant responsible for the observed ESBL phenotype remains unidentified.
• bla_CTX-M For bla_CTX-M, sequencing confirmed the corresponding CTX-M group (CTX-M-1/CTX-M-9) but did not resolve the specific allelic variant. Accordingly, the manuscript now consistently uses the terminology ‘bla_CTX-M-1/bla_CTX-M-9 group genes’ rather than referring to specific CTX-M variants or enzymes.

Figure 2 was also updated to reflect the confirmed bla_TEM-1 and bla_SHV-12 variants. In addition, leader lines and explicit labels were added for the two minority genetic profiles (1.1% each), which were previously indistinguishable due to the minimal width of their bar segments.

 

- Comment 5: The statistical analysis needs some clarification. Given the very small numbers in several categories (particularly hospitalization and ICU groups), the authors should be cautious when interpreting statistically significant associations. The manuscript should report confidence intervals for important effect estimates, particularly the OR of 13.2 for blaSHV carriage in K. pneumoniae versus E. coli. The rationale for the selected pairwise comparisons and the application of Holm correction should also be described more clearly.

Authors: We thank the reviewer for highlighting the need for greater caution in the interpretation of the statistical findings. The Statistical Analysis and Results sections were revised to explicitly acknowledge the small sample sizes in several analytical strata, particularly hospitalization and ICU groups, and to ensure that associations based on small numbers are interpreted cautiously as descriptive or exploratory findings where appropriate. The association between bla_SHV carriage and bacterial species was recalculated using the final dataset. The comparison between K. pneumoniae and E. coli is now reported with Fisher’s exact test, with an OR = 16.54, and 95% CI: 3.56–89.46. This confidence interval has been incorporated into the revised Results to provide a more transparent representation of the uncertainty associated with this estimate.

Regarding the rationale for pairwise comparisons and multiplicity adjustment, we clarified that no post hoc pairwise comparison was used to support the primary inferential conclusions and, consequently, Holm correction was not applied to the global Fisher tests. Any pairwise comparisons presented in the revised manuscript are descriptive and were not used as independent inferential tests. Table notes were also revised to clarify denominators and the interpretation of estimates based on very small strata. Figure 1 was updated using the final dataset, including the corresponding statistical results and denominators (see our response to Reviewer 1 regarding small denominators for further details).

- Comment 6: Table 1 is difficult to interpret because the gene-profile categories and percentages are presented without sufficient explanation of whether percentages represent the total 89 isolates or the corresponding subgroup. The table should be reformatted for clarity.

Authors: We thank the reviewer for this observation. To improve the clarity and interpretation of Table 1, an explanatory footnote has been added specifying that all values are presented as n (%), with percentages calculated using the total number of molecularly characterized isolates (n = 89) as the denominator. In addition, the layout of Table 1 was simplified by consolidating two extremely infrequent genetic profiles, bla_CTX-M + bla_SHV-12 and bla_CTX-M + bla_TEM-1 + bla_SHV-12, into a single ‘Other combinations’ category. Each of these profiles was represented by a single isolate (1.1%) in the complete molecularly characterized cohort. This consolidation reduces the number of sparsely populated columns and improves the readability of the table.

For consistency with this approach, Table 2 was also revised: species × healthcare-setting strata containing fewer than five isolates are now shaded in gray to facilitate the visual identification of cells whose percentages should be interpreted as descriptive observations only, as specified in the corresponding footnote.

- Last, but not least, please standardize terminology throughout: “healthcare setting,” “health service,” “outpatient clinic,” and “hospitalization” are used somewhat inconsistently.

Authors: We thank the reviewer for this observation. The terminology describing healthcare context was reviewed throughout the manuscript and standardized to ensure consistency. The table headings ‘Health Service’ in Table 1 and ‘Healthcare Department’ in Table 2 were both replaced with ‘Healthcare Setting’, consistent with the terminology used in the main text. The terms ‘outpatient clinic,’ ‘hospitalization,’ and ‘intensive care unit (ICU)’ were also reviewed and used consistently throughout the entire manuscript.

 

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors carefully and completely addressed all of the raised concerns by the reviewer in the first review round. The manuscript can be accepted for publication in its present form. Congratulation!