Lysyl oxidase (LOX) limits VSMC proliferation and neointimal thickening through its extracellular enzymatic activity

Abstract

Lysyl oxidase (LOX) plays a critical role in extracellular matrix maturation and limits VSMC proliferation and vascular remodeling. We have investigated whether this anti-proliferative effect relies on the extracellular catalytically active LOX or on its biologically active propeptide (LOX-PP). High expression levels of both LOX and LOX-PP were detected in the vascular wall from transgenic mice over-expressing the full-length human LOX cDNA under the control of SM22α promoter (TgLOX), which targets the transgene to VSMC without affecting the expression of mouse LOX isoenzymes. TgLOX VSMC also secrete high amounts of both mature LOX and LOX-PP. Wild-type (WT) mouse VSMC exposed to VSMC supernatants from transgenic animals showed reduced proliferative rates (low [³H]-thymidine uptake and expression of PCNA) than those incubated with conditioned media from WT cells, effect that was abrogated by β-aminopropionitrile (BAPN), an inhibitor of LOX activity. Lentiviral over-expression of LOX, but not LOX-PP, decreased human VSMC proliferation, effect that was also prevented by BAPN. LOX transgenesis neither impacted local nor systemic inflammatory response induced by carotid artery ligation. Interestingly, in this model, BAPN normalized the reduced neointimal thickening observed in TgLOX mice. Therefore, extracellular enzymatically active LOX is required to limit both VSMC proliferation and vascular remodeling.

Figure 1

Mature LOX and LOX-PP were over-expressed in the vascular wall and in VSMC from TgLOX mice. (a) Representative immunohistochemical analysis showing LOX and LOX-PP staining (brown color) in aorta from both wild-type and TgLOX mice. The indicated areas are shown at high magnification (bars = 20 µm). (b,c) Mature LOX and LOX-PP protein levels were determined by Western-blot in VSMC supernatants from these animals. The position of the pro-enzyme (ProLOX), mature LOX (mLOX) and LOX-PP forms are indicated. GAPDH was analyzed as loading control. Representative immunoblots from 3 independent experiments were shown. WT: wild-type; Tg: TgLOX. Displayed blots are not cropped from different gels or different parts of the same gel and images conform the digital image and integrity policies of the journal.

Figure 2

Extracellular forms are involved in the attenuated cell proliferation induced by LOX transgenesis. (a) Aortic VSMC isolated from wild-type (WT) mice were serum-starved for 24 h and then exposed to conditioned media from transgenic (TgLOX-CM; black bars) or WT VSMC (WT-CM; white bars). Cell proliferation was analyzed by the [³H]-thymidine incorporation method. Results are represented as mean ± SD. *P < 0.002 vs. WT-CM stimulated cells (Mann–Whitney test; n = 6). (b) Endogenous LOX and LOXLs mRNA levels were determined in these cells by real-time PCR. Data are expressed as mean ± SD (Mann–Whitney test; n = 4).

Figure 3

BAPN prevents the reduction in VSMC proliferation induced by LOX transgenesis. Aortic VSMC isolated from wild-type (WT) mice were serum-starved for 24 h and then exposed to conditioned media from transgenic VSMC (Tg; black bars) stimulated or not with BAPN (Tg + BAPN; grey bars) or from WT VSMC (white bars). (a) Cell proliferation was analyzed by the [³H]-thymidine incorporation method. Results are represented as mean ± SD. P < 0.0001: * vs. cells stimulated with conditioned media from WT VSMC; ^# vs. cells stimulated with conditioned media from Tg VSMC (One-way ANOVA; at least n = 6). (b) PCNA protein levels were assessed by Western-blot (upper panel). The bars graph (lower panel) shows the values, obtained by densitometric analysis of Western-blots. Results normalized by β-actin levels are shown as mean ± SD (n = 4; P < 0.05: * vs. cells stimulated with conditioned media from WT VSMC; ^# vs. cells stimulated with conditioned media from Tg VSMC; Kruskal-Wallis). Displayed blots are not cropped from different gels or different parts of the same gel and images conform the digital image and integrity policies of the journal.

Figure 4

Over-expression of LOX-PP did not affect VSMC proliferation. Human VSMC were transduced with a lentiviral vector encoding for full-length LOX (pLOX; black bars), LOX-PP (pLOX-PP; grey bars) or with the corresponding empty vector (pLVX; V; white bars). (a) Immunoblots corresponding to mature LOX (mLOX; left panel) and LOX-PP (right panel) are shown. The position of the pro-enzyme (ProLOX, right panel), detected with the antibody against the propeptide, was also indicated. GAPDH was analyzed as a loading control. Displayed blots are not cropped from different gels or different parts of the same gel and images conform the digital image and integrity policies of the journal. Representative immunoblots from 3 independent experiments were shown. (b) Transduced VSMC were serum-starved for 24 h and then stimulated with 20% FCS. Cell proliferation was evaluated by the [³H]-thymidine incorporation method (left panel) or by cell count (right panel) in these cells. Results represented as mean ± SD. *P < 0.003 vs. pLVX (Kruskal-Wallis; at least n = 6). (c) [³H]-thymidine incorporation into DNA assessed in human VSMC transduced with lentiviral particles for full-length LOX (pLOX; black bars), or the corresponding empty vector (pLVX; white bars) treated or not with BAPN (500 µM). Results are represented as mean ± SD. P < 0.0001: * vs. pLVX; ^# vs. pLOX transduced cells (Two-way ANOVA; at least n = 9).

Figure 5

LOX transgenesis did not affect the systemic and local inflammatory response after carotid artery ligation. (a) Number of circulating white cells in wild-type (WT) and TgLOX mice after carotid artery ligation. (b) Count of the number of circulating monocytes (MID: Mid-sized cells; a subgroup of white blood cells that consists primarily of monocytes), lymphocytes (LYMF) and granulocytes (GRAN) in these animals. Data are expressed as mean ± SEM (*p < 0.001 vs. WT or TgLOX mice at time 0; Two-way ANOVA; n = 6). (c) Mac3 immunostaining in hematoxylin counterstained cross-sections of ligated carotid arteries from wild-type and TgLOX mice 3 weeks after ligation (n = 3). Bar: 20 µm.

Figure 6

BAPN ameliorates the reduced vascular remodeling after carotid artery ligation. Left common carotid arteries were harvested from TgLOX mice and wild-type (WT) littermates treated or not with BAPN 21 days after permanent ligation. Morphometric analysis was performed as described in the Material and Methods section. (a) Representative microphotographs of injured carotid arteries stained with hematoxylin and eosin in these animals. Bar: 100 µm. (b) Analysis of the percentage of stenosis and intima and media areas from sections covering the region 1.4 to 1.6 mm from the ligation site. Three cross-sections for each carotid artery were measured, and the data were averaged. Data are expressed as mean ± SEM. P < 0.05: * vs. wild-type animals; ^# vs. TgLOX mice (Two-way ANOVA; at least n = 6).