We applied a stepped approach in our study to provide evidence that endotoxin exposure in vivo results in a proatherogenic phenotype in circulating monocytes. First, we proved that the LPS infusion model we applied results in both clinical symptoms and increased IL-6, a marker for inflammation associated with TLR4 signaling (17), with IL-6 levels peaking at three hours post LPS infusion. Second, circulating monocytes were profiled on cDNA arrays pre and post LPS infusion. We generated a list of 39 genes differentially expressed at one or four hours post LPS infusion. A lack of overlap between genes at both time points was observed, with only AKIP being differentially expressed at both time points. This observation is likely due to the sequential activation of pathways after the LPS stimulus. In addition, this might reflect different populations of monocytes contributing to the transcriptome, since only 45% of monocytes remain in circulation one hour after LPS stimulation. Of these genes, 7 of 8 randomly selected candidates were confirmed by RT-PCR. Finally, comparisons with our in vitro data identified 6 overlapping genes (BATF, BID, C3aR1, IL1RN, SEC61B and SLC43A3). Since C3aR1 has been previously associated with inflammation and atherosclerosis this might be one of the interesting candidates in our experiments to link both disorders. We will focus our discussion on the genes which were differentially expressed in both our in vivo and in vitro experiments.
The C3aR1 gene is located on chromosome 12. The gene product is a highly ligand-specific membrane receptor belonging to the family of the seven transmembrane domain G-protein-coupled receptors. C3aR1 is expressed in monocytes, macrophages and endothelial cells. Binding of C3 to C3aR1 induces a wide rang of inflammatory and immune effects . There are multiple lines of evidence for the role of C3aR1 in atherosclerosis.
At gene expression level, in samples of patients with advanced peripheral artery disease (PAD) the gene was expressed at a five times higher level in advanced compared to intermediate atherosclerotic lesions . At protein level, advanced human coronary atherosclerosis plaques express C3aR1 in contrast to normal coronary intima . In addition, signaling via C3aR1 promotes plaque instability and can thus result in clinical sequelae of acute coronary syndromes .
The direct causal role of C3aR1 is provided by the fact that pertubaration of this gene in knock out mice models result in significant decrease in atherosclerotic lesion size .
Putting the evidence together, we speculate the following sequence of events. We realize that several pathways will co-exist and this is merely one of them. Low grade inflammation due to endotoxinemia results in the higher expression of C3aR1 in vivo. Increased C3aR1 expression in turn results in a proatherogenic monocyte phenotype. Firstly, this results in increased atherosclerotic lesion size. Secondly, it results in increased plaque instability. Ultimately all these effect might result in more clinical CVD events.
The up regulation of IL1RN and BATF, in contrast, does not support our hypothesis of the circulating monocyte with a proatherogenic phenotype.
IL1RN is a cytokine gene located on chromosome 2. It is a negative regulator of IL-1 signalling and plays a role as an anti-inflammatory cytokine in acute and chronic- inflammation of the vascular wall. It is expressed by macrophages, endothelial cells and smooth muscle cells. Endogenous IL-Ra suppresses atherosclerosis in humans. Decreased expression and not increased expression has been associated with atherosclerotic plaques in mice . The increased expression of IL1RN might be an internal protective mechanism to downlplay the effects of systemic inflammation due to the LPS stimulus.
BATF is a nuclear basic leucine zipper protein and a member of the AP-1 family of transcription factors located on chromosome 14. It directly regulates key components of the formation and function of follicular helper T cells and antibody class switching in B cells . In B-cells the expression of BATF is also induced by LPS and IL-6 . Expression of BATF in rat fibroblast suppressed the production of Metalloproteinase (MMP)-2 and MMP-9. MMPs are key players in atherothrombosis due to their extracellular matrix remodelling properties and their functional effects on cells involved in atherogenesis and atherosclerotic complications .
The role of BID, SEC61B, SLC43A3 in atherosclerosis is not yet established.
BID is located on chromosome 22q and is a critical mediator of inflammation and innate immunity . Mechanistically, BID interacts with NOD1, NOD2 and the IκB kinase (IKK) complex, impacting NF-κB and extracellular signal-regulated kinase (ERK) signalling . Targeting BID by small molecules has been proposed as a way to treat immune-mediated inflammatory disease including inflammatory bowel disease.
SEC61B is located on chromosome 9q22. In the Endoplasmatic Recticulum (ER) membrane, the heterotrimeric Sec61 complex comprises three transmembrane subunits (Sec61α, Sec61β, and Sec61γ in mammals) and forms protein-conducting channels, collectively termed a translocon. Sec61α is known to be stabilized by Sec61γ and mainly responsible for the translocation activity in the ER. In contrast to the other two subunits, Sec61β can be stable on its own, and its function is not as well defined. It is known that Sec61β in the inner nucleus membrane (INM) is required for the release of epidermal growth factor from the INM to the nucleus .
SLC43A3 is located on chromosome 11 and is known to be highly expressed in macrophages. It is thought to function as a transporter of metabolites and nutrients that are necessary during developmental events, such as organogenesis . It is part of the specific expression pattern of the micro vascular endothelium and has been proposed as putative drug target to pathological angiogenesis .
Our study has several limitations. The experiments were conducted in a small number of individuals. However the results are supported by the fact that the selected candidate genes were validated by RT-PCR. In addition, several genes have been previously described in the literature as being differentially expressed in response to endotoxin exposure [22, 23]. In our experiment we determined the effect of LPS on gene expression in circulating monocytes. However, a substantial part of the monocytes migrated into the vessel wall, reflected by a fall in monocyte count in the peripheral blood after LPS infusion. Therefore it begs questioning how representative these circulating monocytes are for the entire population of these cells . In this respect, the overlap in genes differentially expressed in our in vivo study compared to previous in vitro LPS challenges, lends further support to the validity of the identified candidate genes in circulating monocytes. However, the low level of homology between in vivo and in vitro transcriptomes makes us question the value of studying expression profiles of circulating cells as (prognostic) markers for disease states. The poor overlap between in vivo and in vitro experiments could be due to either the intrinsic differences between the two systems or due to the small sample sizes in both studies.
Another limitation is that the procedure of monocyte isolation is time consuming and might thus effect transcription and stability. To minimize these potential effects, for all in vivo and in vitro experiments the same optimized protocol was used by the same analyst. Furthermore genes differentially expressed due to handling would have showed up in the placebo experiments. No genes were differentially expressed in these experiments.
Finally, in comparison to in vitro experiments only a small number of genes were differentially expressed. This is in part due to the stringent quality control criteria we applied during the analysis, the limited coverage of the microarrays and the fact that TLR4 signaling is modulated in vivo by a number of specific pathways.