The impact of the PCSK-9/VLDL-Receptor axis on inflammatory cell polarization.

BACKGROUND
Studies have shown that lipoproteins, such as LDL and VLDL, as well as its major protein component ApoE2 impact on macrophage polarization important in atherosclerosis. Proprotein convertase subtilisin/kexin 9 (PCSK9) is a key regulator of lipoprotein receptor expression. The present study investigated the effect of the VLDL/VLDL-receptor (VLDL-R) axis on mononuclear cell polarization, as well as the role of PCSK9 and PCSK9 inhibitors (PCSK9i) within this network.


METHODS
Human monocytic THP-1 cells and human monocyte-derived macrophages isolated from peripheral blood mononuclear cells (PBMC) were treated with either LPS/IFN-γ to induce a pro-inflammatory phenotype, or with IL-4/IL-13 to induce an anti-inflammatory phenotype. Cells were then subjected to further treatments by lipoproteins, PCSK9, PCSK9i and lipoprotein receptor blockers.


RESULTS
LPS/IFN-γ treatment promoted a pro-inflammatory state with an increased expression of pro-inflammatory mediators such as TNF-α, CD80 and IL-1β. VLDL co-treatment induced a switch of this pro-inflammatory phenotype to an anti-inflammatory phenotype. In pro-inflammatory cells, VLDL significantly decreased the expression of pro-inflammatory markers e.g., TNF-α, CD80, and IL-1β. These effects were eliminated by PCSK9 and restored by co-incubation with a specific anti-PCSK9 monoclonal antibody (PCSK9i). Migration assays demonstrated that pro-inflammatory cells displayed a significantly higher invasive capacity when compared to untreated cells or anti-inflammatory cells. Moreover, pro-inflammatory cell chemotaxis was significantly decreased by VLDL-mediated acquisition of the anti-inflammatory phenotype. PCSK9 significantly lessened this VLDL-mediated migration inhibition, which was reversed by the PCSK9i.


CONCLUSION
VLDL promotes mononuclear cell differentiation towards an anti-inflammatory phenotype. PCSK9, via its capacity to inhibit VLDL-R expression, reverses the VLDL-mediated anti-inflammatory action, thereby promoting a pro-inflammatory phenotype. Thus, PCSK9 targeting therapies may exert anti-inflammatory properties within the vessel wall.


the gene of interest was normalized with 18S (forw
rd primer: CGGCTACATCCAAGGAA, reverse primer: GCTGGAATTACCGCGGT).Duplicate tests on each sample were done.


Western blot analysis

The protein lysates were blotted as previously described [40] and

ELISA
analysis for TNF-α (BioLegend, Germany) was performed.ELISA was done as recommended by the manufacture.


Flow cytometry

The purity of the CD14 isolated human monocytes was analyzed using followi

DCFDA (Life Tech, Germany
for total ROS measurement for 30 min at 37°C as described in [41].Brie y, cells were washed twice with PBS containing calcium-chloride (1 mmol/L) and lysed with Triton X-100 buffer.The uorescence intensity was analyzed by excitation at 485 ± 10 nm and emission at 530 ± 10 nm using a ViktorX Multilable plate reader and subsequently normalized to protein level.


Migration assay

Transwell cell culture chamber with gelatin-coated (0.2%) polycarbonate membr

es (8mm pores) (
ecton Dickinson) was used.The number of cells per high power eld (HPFs; magni cation x 320) migrated to the lower surface of the lters after 4 hr.incubation was determined as described [42].


Phagocytosis assay

Phagocytotic effect of untreated, LPS/IFN-γ, and IL-4/IL-13 treated THP-1 ce

s was measured with
the Vybrant™ Phagocytosis Assay Kit (V-6694, Molecular Probes, USA) according to the instructions of the manufacturer.


Statistical analysis

Data were evaluated using the Mann-Whitney test (nonparametric test), Krus

l-Wallis test (nonpar
metric test) or with two-way ANOVA.Statistical signi cance was de ned as p<0.05.Statistical analysis was performed by using Prism 5 for Windows (GraphPad Software, 2003, San Diego, USA).


Results


VLDL has anti-in ammatory effects on proin ammatory polarized monocytic cells

First, th

LDL-R gene and protein expression in undifferentiated and differentiated cells an
investigated the impact of VLDL and LDL on these lipid receptors in pro-in ammatory cells.All three cell phenotypes expressed the VLDL-R and LDL-R, with pro-in ammatory cells displaying higher mRNA and protein levels of the LDL-R (p < 0.05 vs M0; Fig. 3A-C).When pro-in ammatory cells were treated with VLDL, VLDL-R mRNA and protein levels were unaffected (p = non-signi cant vs untreated proin ammatory cells; Fig. 3A, D).In contrast, LDL-R mRNA and protein signi cantly decreased upon VLDL treatment (p < 0.01 vs untreated pro-in ammatory cells; Fig. 3B, D).When these cells were treated with LDL, VLDL-R mRNA signi cantly increased and LDL-R mRNA levels decreased (both p < 0.01 vs untreated LPS/IFN-γ treated cells; Fig. 3E, F).

3.3.PCSK9 inhibits VLDL-mediated anti-in ammatory effects in pro-in ammatory polarized monocytic cells via VLDL-R egulation VLDL-R and LDL-R regulation by PCSK9 was investigated in VLDL-treated pro-in ammatory polarized cells.Receptor gene and protein expression was compared in control (untreated) cells, cells co-incubated with PCSK9 and cells co-incubated with PCSK9 and concurrently treated with a speci c monoclonal anti-PCSK9 antibody.Immunoblotting demonstrated, that PCSK9 signi cantly lessened LDL-R and VLDL-R protein levels (Fig. 4A).However, no effect was found on VLDL-R mRNA (Fig. 4B).Co-incubation and thus rescue of PCSK9 inhibitory effects with a speci c monoclonal anti-PCSK9-antibody restored LDL-R and VLDL-R protein expression (Fig. 4A).

To investigate the impact of PCSK9 on VLDL/VLDL-R signaling cascade and thus VLDL-mediated antiin ammatory effects in p o-in ammatory cells, mRNA expression of TNF-α and CD80 was investigated.VLDL-treatment signi cantly decreased both in ammatory markers (p < 0.05 vs pro-in ammatory cells; Fig. 5A, B).Upon co-incubation of VLDL-treated pro-in ammatory cells with PCSK9, mRNA levels of TNFα and CD80 signi cantly increased (p < 0.01 vs. VLDL-treated pro-in ammatory cells; Fig. 5A, B).

Inhibition of PCSK9 with a speci c monoclonal anti-PCSK9-antibody reversed the effects of PCSK9 on CD80 mRNA levels (p < .05 vs. VLDL/PCSK9-treated pro-in ammatory cells; Fig. 5B).In contrast, a speci c monoclonal LDL-R antibody had no impact on TNF-α or CD80 mRNA levels, neither in controls nor in VLDL-treated pro-in ammatory cells (Fig. 5C, D).To further elaborate VLDL-R dependent antiin ammatory actions, we co-incubated pro-in ammatory differentiated cells with VLDL and either TSP-1

or the PKC-activator PMA as mediators of a sustained in ammatory response.In both cases, TSP-1 or PMA-mediated in ammation, evident by increases in TNF-α and CD80, were prevented by VLDL (p < 0.05 vs PMA or TSP-1 treated cells; Fig. 5E-H).


VLDL impacts on pro-in ammatory polarized monocytic cell phagocytosis and invasion

To assess the impact of VLDL on pro-in

mmatory cell functionality, we investigated cell phagocytosis and invasion in untre
ted, LPS/IFN-γ and IL-4/IL-13 treated cells post treated with VLDL and cells post treated with VLDL and PCSK-9, co-incubated with or without a speci c monoclonal anti-PCSK9 antibody.

Pro-in ammatory differentiated cells displayed a signi cantly higher phagocytotic capacity when compared to untreated or an i-in ammatory cells (p < 0.05 and p < 0.01, respectively; Fig. 6A).VLDL signi cantly reduced phagocytosis in pro-in ammatory cells when compared to untreated proin ammatory cells, down to the level found in anti-in ammatory cells (p < 0.05 vs pro-in ammatory cells; Fig. 6A).In Boyden-chamber invasion assays with gelatin-coated membranes, VLDL signi cantly inhibited the migratory capacity of pro-in ammatory cells towards MCP-1, when compared to untreated pro-in ammatory cells (p < 0.001; Fig. 6B).The number of migrated cells was comparable in proin ammatory cells post treated with VLDL to anti-in ammatory cells.PCSK9 eliminated this antimigratory effect of VLDL on pro-in ammatory cells (p < 0.001 vs VLDL-treated pro-in ammatory cells; Fig. 6B).Accordingly, this was signi cantly lessened by co-treatment with a speci c monoclonal anti-PCSK9 antibody (p < 0.001 vs VLDL/PCSK9-treated pro-in ammatory cells; Fig. 6B).


Discussion

The present study investigated the impact of VLDL on mononuclear in ammatory cell polarization.Here, we demonstrate

hat VLDL pr
motes an anti-in ammatory phenotypic change in in ammatory polarized cells, thereby signi cantly lessening cell invasion and phagocytosis.Furthermore, we show that PCSK9, via its capacity to regulate VLDL-R cell surface expression, is a master switch in VLDL/VLDL-R-mediated pro-in ammatory suppression.

Studies have shown, that apolipoprotein B-containing lipoproteins such as LDL and VLDL are crucial to atherosclerosis, although th y vary in their apolipoprotein and triglyceride content [43].Among the triglyceride-rich lipoproteins (TRLs), not all are atherogenic, as it was shown that large VLDL (in contrast to small VLDL remnants) do not cross the endothelial barrier [43].It has been proposed that VLDL remnants are involved in macrophage foam cell formation via the TRL-lipoprotein lipase (LPL)-VLDL-R pathway; however, the effect of large VLDL on macrophages is still a matter of debate [48].

In our study, treatment of pro-in ammatory primed monocytic THP-1 cells or human monocyte-derived macrophages with VLDL, but not wit LDL, signi cantly decreased the mRNA expression of M1 marker genes, including TNF-α, CD80, and IL-1β.This was accompanied by increases in CD206 used as an M2 marker.VLDL-treatment but not LDL-treatment increased the IL-10 mRNA expression in relation to IL-12 mRNA in pro-in ammatory primed cells, when IL-12 is the pro-in ammatory M1-cytokine, and IL-10 denotes the anti-in ammatory M2-cytokine in macrophage polarization [7].Thus, a shift in the ratio of IL-12/IL-10 towards IL-10 (IL-12 low /IL-10 high ) indicates an anti-in ammatory phenotype [49].Accordantly, treatment of pro-in ammatory primed monocytic cells with the in ammatory stimulators PMA and TSP-1 resulted in a signi cant increase of TNF-α and CD80, which was reversed by VLDL co-treatment.TSP-1 is upregulated during in ammatory conditions and regulates the secretion of pro-in ammatory Th1 cytokines such as IL-6, IL-1β, and TNF-α via the NF-κB pathway [50].PMA induces THP-1 macrophage differentiation/polarization and stimulates a high phagocytic capacity and expression of cytokines in response to TLR-ligands [51].

Others have demonstrated that apoE induces macrophage polarization from a pro-in ammatory to an anti-in ammatory phenotype involving the a oER2 or the VLDL-R [22].VLDL-R-or apoER2-expressing cells not only downregulated the expression of M1-markers, but also reduce M1-macrophage responses after treatment with ApoE in that study [53].Likewise, ApoE-producing macrophages obtained from VLDL-R-de cient mice displayed a reduced secretion of M2 marker-cytokines, promoting a pro-in ammatory phenotype [22].Interestingly, the VLDL-R and LDL-R display about 50% sequence homology, with the Olinked sugar (OLS)-and the intracellular domains as the regions with the most considerable divergence [52].

Several studies demonstrated that the cytokine repertoire of macrophages strongly depends on their local environment [54] [55] [56].In our st dy, we show VLDL anti-in ammatory effects, since the proin ammatory differentiated THP-1 cells as well as human monocyte-derived macrophages produced less pro-in ammatory mediators and ROS formation following VLDL treatment.However, we could not show the generation of a "clear-cut" pro-in ammatory suppression following VLDL treatment, since M2 markers such as CD206 were not strongly regulated.However, others also found that a pro-in ammatory-or antiin ammatory-speci c stimulus e.g., with HDL or free fatty acids caused a different spectrum of macrophage activation than the initial M1 or M2 phenotype [57].

To further investigate VLDL-signal transduction, we assessed the expression of the VLDL-R and LDL-R, as well as their regulation by LDL and VLDL We found that both receptors are expressed at mRNA and protein level in all cell phenotypes investigated.The VLDL-R was moderately lessened in proin ammatory derived cells, while the LDL-R was signi cantly increased, suggesting that a proin ammatory environment affects the expression of both receptors.

Other authors demonstrated that VLDL and LDL (among other lipoproteins) negatively regulated the LDL-R, while the VLDL-R is not regulated by lipop oteins in monocytic THP-1 cells [58,59].In accordance, LDL negatively regulated the expression of its receptor in this study.Treating pro-in ammatory polarized cells with VLDL resulted in a signi cant decreased of the LDL-R at mRNA and protein expression, while the levels of the VLDL-R did not change.Likewise, it has been shown that β-VLDL does not regulate LDL-R expression [58].

To further examine the importance of the VLDL-R, we investigated its regulation by PCSK9.Here, we demonstrate that PCSK9 is a key player in VLDL/VLDL R mediated pro-in ammatory suppression in monocytic cells.Previous studies have shown that PCSK9 is related to a heightened expression of proin ammatory cytokines, chemokines, and adhesion molecules [60].In endothelial cells, PCSK9 is induced by in ammatory mediators [61].Furthermore, PCSK9 stimulates the secretion of pro-in ammatory markers in different cell-types including macrophages and is associated with the induction of proin ammatory pathways, including TLR4 and NF-κB, along with apoptosis and autophagy regulation [62].Both, VLDL-R and apoER2 are degraded by PCSK9 [38].For this reason, we treated pro-in ammatory differentiated monocytic cells with PCSK9, as well as with a speci c monoclonal PCSK9-blocking antibody.

PCSK9 co-treatment resulted in a signi cant resurge of the pro-in ammatory markers TNF-α and CD80 in VLDL-treated in ammatory polarized monocytic cells.Con urrent inhibition of PCSK9 with its speci c monoclonal antibody restored the VLDL-mediated anti-in ammatory phenotypic transition.Using immunoblotting and RT-PCR, we show that PCSK9 regulates VLDL-R expression post-translationally, comparable to the LDL-R [63].To further clarify the importance of the VLDL-R, we compared VLDL-R regulation by PCSK9 to LDL-R regulation by PCSK9.Our study shows that both are downregulated at protein levels by PCSK9 in pro-in ammatory cells and that the inhibition of PCSK9 restores the protein expression of both receptors.This is in accordance with previous studies, showing that PCSK9 governs the degradation, not only of the LDL-R but also of the VLDL-R, as the latter is regulated in a LDL-R independent manner [64].The VLDL-R comprises an EGF-A repeat, sharing about 60% sequence identity with the LDL-R to which PCSK9 can directly bind [64] [65].Among the lipoprotein receptor-family, LDL-R and VLDL-R, but not apoER2, contend for PCSK9-mediated degradation, with VLDL-R being more prone to the effects of PCSK9 [38].Taken together, PCSK9 is a key player in VLDL/VLDL-R anti-in ammatory signaling.

To further investigate PCSK9/VLDL/VLDL-R mediated inhibition of in ammation, we investigated its impact on cell functionality using pro-in ammatory polarized cells Alternative activated macrophages undergo cell-functional changes, as they are less motile and cytotoxic [22].Our results showed that treating these pro-in ammatory cells with VLDL signi cantly decreased the amount of phagocytosis, comparable to an anti-in ammatory phenotype.Treatment with VLDL also signi cantly decreased invasion of pro-in ammatory polarized cells.This was signi cantly inhibited by PCSK9, thus promoting invasion and abolished by PCSK9 inhibition.


Conclusions

In summary, our study demonstrates that VLDL via VLDL-R promotes the anti-in ammatory phenotype in mononuclear cells.Within this cascade, PCSK9 is a key

Figure 6 VLDL
6
Figure
Table 1
1
Validation of the M1 and M2 phenotype of THP-1 cells.Relative mRNA expression analysis of TNF-α,

oteinDec
arationsEthics approval and consent to participate Not applic ble.Consent for publicationAll authors approved the manuscript, and given consent of submission and subsequent publication of the manuscript.Competing interest PS has received consultancy and lecture honoraria from Amgen, Novartis, Sano -Aventis, Bristol-Myers Squibb/P zer, Daiichi-Sankyo, Bayer, Boehringer Ingelheim, BerlinChemie, B. Braun, AstraZeneca, editor honoraria from Springer Nature and was an investigator in PCSK9-inhibitor outcome trails.The remaining authors declare that they have no con ict of interest.We con rm that the manuscript has been read and approved by all