PREreview of Abdominal symptoms of invasive meningococcal disease are associated with the induction of plasminogen activator inhibitor in omental adipocytes

Abdominal symptoms of invasive meningococcal disease are associated with the induction of plasminogen activator inhibitor in omental adipocytes

1. Summary and Overall Impression

Recommendation: Major revisions required. 

This preprint investigates the pathophysiological mechanisms underlying abdominal symptoms as a result of invasive meningococcal disease (IMD) using a transgenic mouse model to examine thrombotic lesions and inflammatory responses in the omentum. The authors provide compelling evidence that Neisseria meningitidis (Nm) isolates of serogroup W cluster into subgroups on the basis of association with abdominal presentation. This is further supported by the finding that abdominal isolates lack the beta chain of LOS, while non-abdominal isolates retain it, illuminating the possibility that Nm has a genetic basis for differential symptoms. Histological analysis shows evidence of thrombosis in Wabdo isolate-infected mice, suggesting coagulation lesions in the omentum. However, the methodological preparations of histological specimen is not clearly presented in the manuscript. RNA sequencing was conducted of omentum fractions from infected mice, showing that the PAI-1-encoding gene associated with clot stability was differentially upregulated in mice infected with abdominal isolates, but the magnitude of upregulation is not quantitatively addressed. RT-rtPCR was performed to validate the RNAseq findings and suggested that PAI-1 was primarily expressed in omental adipocytes. Induction of the PAI-1-encoding gene was also observed in mice injected with LOS from abdominal isolates, but the significance of this finding is not quantitatively presented. The authors claim that these findings suggest that local induction of PAI-1 contributes to thrombosis and hypoperfusion, causing abdominal symptoms associated with IMD. While these findings contribute to our pathophysiological understanding of extra-meningeal clinical manifestations of IMD, it is not clear that this conclusion drawn by the authors is reasonably justified by the findings of the study due to a lack of data transparency and statistical transparency. The conclusions are theoretically sound, but not experimentally validated. The methodology appears robust with the inclusion of whole-genome sequencing, transcriptomic analyses (RNA-seq), in situ RT-rtPCR, and histological imaging, but it is unlikely that this manuscript will be prepared for publication until major revisions are made to quantitatively elucidate the results. 

1. Section-by-Section Review

Abstract

Major issues:

1. Sentence 1 of paragraph 2 states, “...isolates associated with abdominal presentations (mainly serogroup W isolates of clonal complex 11) caused thrombotic lesions in the blood vessels of the omentum, and they also induced a higher inflammatory response in the omentum with elevated levels of IL-6, TNF-alpha, and KC.” According to the reported results, these increased cytokine levels were insignificant. It is misleading to assert that insignificant results are biologically meaningful in this case. The authors should include p-values or confidence intervals to support the level of association between cytokine levels in the omentum and infection with abdominal isolates. 

2. The end of the abstract reads “These lesions can lead to thrombosis and hypoperfusion in the omentum, resulting in clinical abdominal presentations”. The findings demonstrated by this study contribute significantly to the current understanding of clinical abdominal presentations, but it is a gross overstatement of conclusions to insinuate exclusive causality of abdominal symptoms without reasonable exclusion of other biologically plausible mechanisms for abdominal pain. The study provides strong correlative evidence but does not establish direct causality, so the authors should remove any language suggesting causality.  

3. The Author Summary claims “...this is the first study to explore a mechanism underlying an extra-meningeal clinical form of N. meningitidis infection” (paragraph 2, sentence 3), however, many other extra-meningeal pathophysiological mechanisms have been elucidated by the current literature, so this conclusion is not fully supported. Instead, the authors should state that this is the first study to explore the pathophysiology of omental lesions associated with IMD. 

4. Paragraph 2, sentence 1 claims that this study provides evidence that abdominal symptoms are caused by the coagulation (clotting) of blood in the microvessels that surround the abdominal organs. The study potentially addresses the presence of coagulative lesions, but lacks data to suggest that this causes abdominal symptoms. The authors must reference existing literature in support of the conclusion that omental lesions have been previously shown to cause abdominal symptoms associated with IMD. 

Introduction

Major issues:

1. State objectives and central hypothesis in the abstract or introduction. The last sentence of Discussion states, “Our study supports the hypothesis that a local induction of an inflammatory response leads to thrombotic lesions, which seem to result from the specific action of PAI-1 on neighboring blood vessels”, but this hypothesis was not stated prior. 

2. Paragraph 4, sentence 1 states, “This work aimed to mimic the pathophysiological process of these abdominal presentations in mice and test the role of meningococcal LOS in causing lesions that may lead to these abdominal symptoms.” It is reasonable to state that these lesions may lead to abdominal symptoms but this testament is inconsistent throughout the paper which suggests that causation was established by this study. Please remove language suggesting a causal link between abdominal symptoms of IMD and upregulation of PAI-1. 

3. Paragraph 2, sentence 10 is missing a critical reference. “These abdominal presentations were hypothesized to be associated with the formation of microinfarcts in the blood vessels of the omentum, a highly vascularized fold of the peritoneum connecting the various abdominal viscera”. Reference 10 cited in the previous sentence (Paragraph 2, sentence 9) does not draw this association. This hypothesized statement seems to be the basis of this study, so a reference is critical. 

Results

Subheading: Characterization of selected isolates

Minor Issues: 

1. Table 1 extends beyond the right margin, making the last column difficult to interpret. 

Subheading: Meningococcal isolates with abdominal syndromes induce micro-infarcts in the omentum

Major issues:

1. Quantify how many infected mice exhibited clinical decline compared to non-infected controls and if that difference was significant. 

2. Figure 3 should explicitly mention how representative the histological images are of each group, or how many mice showed these lesions. 

Subheading: Transcriptomic analysis of the omentum

Major issues:

1. The bar labels of Figure 4 are unreadable, preventing accurate interpretation of results. 

2. Paragraph 2 sentence 3 states that the PAI-I gene was upregulated, but please specify by how much it is upregulated or whether this difference is statistically significant. RT-rtPCR results should quantify gene expression fold-changes.

3. Paragraph 3, sentence 4 suggests that genes encoding IL-6, TNF-alpha, serpin e1, serpin b2, and factor 3 were hyperexpressed, but the following sentence suggests that only the serpin e1 gene was significantly hyperexpressed. If the other genes were not overexpressed to a level of statistical significance, clearly make that distinction in this paragraph for clarity and accurate interpretation of RT-rtPCR results. 

Discussion and Conclusion

Major issues:

1. Paragraph 1, sentence 6 states that the L3,7,9 immunotype corresponding with the LOS alpha chain is associated with virulence and significant induction of TNF-alpha and IL-6, but previous reporting of findings in the Results section suggests nonsignificance of this finding. The authors should remove this sentence, as it is misleading to assign biological relevance and establish an association if this finding lacks statistical support.  

Minor issues:

1. Paragraph 2, sentence 6 “Our study also supports the findings that high production of PAI-1 is reported in IMD and high production of PAI-1 were reported in IMD” is grammatically incorrect and unclear to the reader.

Methods

Major issues:

1. The study does not provide a clear rationale for the number of isolates used (n=20) or the number of mice per group. Was a power calculation performed to justify this choice? If so, stating the rationale explicitly would allow for reproducibility. 

Competing interests

The authors declare that they have no competing interests.