Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Mack and co-authors conducted as study on the association among inflammatory monocyte-derived macrophages and blood lipid levels, with a primary focus on oxLDL. In brief, the authors used flow cytometry to quantitate macrophage phenotype markers (e.g., CD86, CD163) and cytokine production (e.g., TNF, IL-6) among the blood of 20 human subjects. Cytokine responses were also measured in a subset of 6 blood samples with a normal lipid profile that were cultured with oxLDL and then stimulated with LPS/interferon gamma. As a result of their findings, authors conclude that the inflammatory phenotype of macrophages is dependent on the level of lipids that contribute to atherogenesis (e.g., low LDL particle size). The manuscript is concise and clear throughout premise, methods, results, and discussion (+ experimental limitations). Thus, I do not have recommendations for substantial improvement of the study/manuscript.

Author Response

Reply: We thank reviewer 1 for their review. We note there are no recommended changes or comments we need to take action on. 

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript by Mack et al analyses whether the inflammatory state of monocyte-derived macrophages is associated with blood lipid levels. The authors have included a very low number of subjects which limits the results observed. However, the idea raised is interesting. However, there are some aspects that may be addressed before the manuscript could be published.

1. The authors say that included healthy volunteers. Furthermore, the also say that monocytes were isolated from 6 normolipidemic individuals. They must define what is normolipidemic. When you see lipid profile in Table 1, most of the individuals are normolipidemic. However, they say that “All participants entering our study were generally healthy, yet many were unaware that they had dyslipidaemia”. Dyslipidemia criteria must be added in the methods section.

2. The authors conclude that “Monocyte-derived macrophages adopt an inflammatory phenotype relative to the levels of circulating lipid factors that are characteristic of atherogenic dyslipidaemia (such as high TG, TG/HDL-C and low LDL particle size), but not LDL-C.” and “Our findings that (monocyte-derived) macrophages are inflammatory in generally healthy individuals with atherogenic dyslipidaemia (independent of age) calls for equally early attention to be paid to atherogenic dyslipidaemia”. However, most of the subjects do not present atherogenic dyslipidemia. Show normal-high HDL-C, normal TG. Thus, this conclusion is not correct.

3. I consider, according to both previous commentaries, that the authors have 2 options: divide the “healthy volunteers” in two groups according to the presence of dyslipidemia (dyslipidmeia criteria must be added in the methods section) and to reanalyse the data comparing true healthy with dyslipidemic subjects. However, if the number of individuals in any of the groups is low (probable because of the number of subjects included) they shoud add more subjects to complete both groups and perform the experiments again.

4. Why there is no analysis including apoB? ApoB is the most important atherogenic particle (includes all the atherogenic lipidic particles). ApoB is the best marker for cardiovascular risk (taking into account lipid profile) far better than TG, TG/HDL-C, and cholesterol/HDL-C. in fact, the recent guidelines for cardiovascular risk that not consider them as primary goals. This analysis must be performed considering apoB and non-HDL cholesterol. The authors consider that the results

5. At the discussion section the authors say “Notably, the individuals with high TG/HDL-C ratio (equivalent to >3.5 when using mg/dL units) also had high insulin levels, consistent with a high ratio identifying insulin resistance [22] and identifies them as having atherogenic dyslipidaemia, independent of their LDL-C levels [23]”. It is true that insulin resistance is cahracterized by atherogenic dyslipoidemia, but many/most of the subjects included in the present study do not show insulin resistance. They have an average insulin of 7, normal glucose and an average HOMA of 1.3. Is there specific cut-off for insulin resistance in Australia? They must be added and to discuss about it.

It is true that TG/HDL ratio is a good surrogate for insulin resistance. However, it is used when insulin is not available. The authors have measured insulin. Thus, they must consider insulin or HOMA instead of TG/HDL ratio.

6. LDL abbreviation must be added in line 43 instead of line 45.

Author Response

Please see the attachement

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This exploratory study (n=20) links an atherogenic dyslipidaemia profile (high TG, elevated TG/HDL-C, smaller LDL size) to a more inflammatory phenotype in monocyte-derived macrophages and shows that oxLDL exposure increases IL-1β and lowers IL-10 upon LPS/IFNγ stimulation. The work extends prior monocyte findings to macrophages and may inform early-risk biology.

Major comments

1. Confounding & model choice
• Please add multivariable models (sparse is fine for n=20) adjusting at least for age, sex, fasting insulin (proxy for IR), and hs-CRP, given the tight relation between TG/HDL-C and insulin resistance discussed. Include sensitivity analyses to show that associations of TG/HDL-C and LDL size with macrophage phenotype persist after adjustment.
2. Risk calculation handling
• The use of imputed values (BP 179 mmHg for three females; Chol/HDL = 8 for one participant) to compute CVD risk is problematic and may misclassify risk. Either (i) exclude these individuals from risk analyses or (ii) present with/without analyses showing unchanged conclusions. Please also clarify that these imputations were not used in the core macrophage–lipid correlations.
3. oxLDL experiment—controls and dosing
• Specify the oxLDL source, oxidation method, and endotoxin testing, and include a vehicle/native LDL or albumin control if feasible. Please justify that 30 µg/mL is physiologically relevant and, if possible, add a minimal dose–response (2–3 points).
• You switch to GM-CSF on day 6 before LPS/IFNγ; GM-CSF itself biases toward an M1-like state. Please justify this design, include a parallel M-CSF–only arm, or temper causal language accordingly.
4. Culture conditions & variability
• Autologous serum (20%) can introduce between-subject batch effects. Indicate donor fasting status, how serum was standardized, and whether differentiation conditions were normalized for serum TG/oxLDL content, which could partially recapitulate in-vivo exposure.
• Methods §2.7: clarify the rationale for switching to 10% FBS in that experiment.
• For the LEGENDplex readout, detail the normalization factor (per-well cell counts, recovery, and/or viability) and the exact formula used.

Minor comments

• LDL particle size: Specify the analytic method (e.g., gradient gel vs NMR), justify the units (Å), and provide laboratory QC details.
• Cytokine timing: You report LPS 1 µg/mL for 6 h (flow) and LPS/IFNγ 50 ng/mL each for 24 h (LEGENDplex). Please state the rationale for dose and kinetics and, if appropriate, harmonize or explain why distinct conditions were required.
• Data sharing: Deposit de-identified source data (CSVs) for Figure 1 (and key supplementary plots) to enhance transparency and reproducibility.

Comments for author File: Comments.pdf

Author Response

please see the attachement

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have supplied adequate answers to all the questions and have modified the manuscript according to the suggestions.

 

Reviewer 3 Report

Comments and Suggestions for Authors

I thank the authors for their careful revision and clarifications. Several of my initial concerns have been addressed, including specification of fasting status, clarification of the use of imputed values for CVD risk calculation, and a detailed description of the normalization procedure for cytokine analyses. These additions improve transparency. Nevertheless, important issues remain unresolved.

Major Points

1. Statistical Adjustment
   The revised manuscript continues to present only univariate Pearson correlations. My original request was for at least minimal multivariable adjustment (age, sex, insulin, hs-CRP), given the well-established relationship between TG/HDL-C and insulin resistance. Without such adjustment, the reported associations remain susceptible to confounding. Even sparse models would substantially strengthen the robustness of the findings.
2. oxLDL Experimental Design
   Key methodological details remain missing: the source of oxLDL, method of oxidation, and endotoxin testing are not reported. The physiological justification for using 30 µg/mL is not provided, and no dose–response data are included. In addition, while the switch from M-CSF to GM-CSF is acknowledged as pro-inflammatory, no parallel M-CSF–only arm was included. This limits causal inference and should be either addressed experimentally or reflected in more cautious interpretation.

Minor Points

1. CVD Risk Calculation
   The clarification that imputed values were used only for risk estimation (and not for macrophage–lipid correlations) is helpful. However, a sensitivity analysis excluding these individuals would provide reassurance that overall conclusions are unaffected.
2. Culture Conditions
   The specification of fasting blood draws and normalization methodology is appreciated. However, potential variability introduced by the use of 20% autologous serum remains unaddressed, particularly with regard to differing TG/oxLDL content between donors.

While certain limitations remain—notably the lack of multivariable statistical adjustment and the incomplete oxLDL control design—the authors have made a good-faith effort to clarify methodology and improve interpretive caution. In my view, the manuscript now presents a sufficiently transparent and informative account of the work, and the residual limitations do not compromise its overall contribution to the field.