Induction of neutrophil gelatinase-associated lipocalin in vascular injury via activation of nuclear factor-kappaB

Abstract

Neutrophil gelatinase-associated lipocalin (NGAL) has recently emerged as an important modulator of cell homeostasis. Elevated plasma NGAL levels, possibly because of activation of blood leukocytes, are associated with atherosclerosis. However, little is known about induction of NGAL expression in blood vessels. Using a rat carotid artery injury model, we found that NGAL was highly induced in the intima after angioplasty but was attenuated by adenovirus-mediated expression of a dominant-negative mutant of inhibitor of nuclear factor (NF)-kappaB kinase beta (dnIKKbeta). Expression of NGAL mRNA and protein was also up-regulated in an NF-kappaB-dependent manner in rat and human vascular smooth muscle cells (SMCs) in response to interleukin-1beta stimulation. Rat SMC-produced NGAL was present as mono- and homomeric forms in the cytosol and in a complex containing matrix metalloproteinase-9 (MMP-9) after secretion. In agreement with levels of NGAL, proteolytic activity of MMP-9 was markedly high in the intima of injured vessels and in the culture supernatant of activated intimal SMCs but was reduced in the vessels transduced with dnIKKbeta. The present study reveals a previously unrecognized vascular response to an-gioplastic injury, characterized by NF-kappaB-dependent expression of NGAL in vascular SMCs. Further-more, SMC-produced NGAL interacts with MMP-9, a mechanism by which NGAL may modulate MMP-9 proteolytic activity in the vascular repair process.

Figure 1

Expression of NGAL and MMP-9 in the injured vessels. A: Transcripts of NGAL and MMP-9 were determined by real-time quantitative RT-PCR at the indicated times in the injured carotid arteries transduced with recombinant adenoviral vector expressing β-galactosidase (β-gal) or dominant-negative IKKβ (dnIKKβ). Levels of transcripts are expressed in relative units normalized to hypoxanthine guanidine ribonucleosyltransferase (HPRT) mRNA. Data are presented as mean ± SEM from six to eight rats for each time point. *P < 0.05, versus β-gal group at day 14; ^#P < 0.05, β-gal group at day 14 versus the uninjured group; ^&P < 0.05, β-gal group at day 3 versus the uninjured group. B: Immunostaining for NGAL in the carotid arteries transduced with β-gal or dnIKKβ at day 3 and day 14 after injury and in an uninjured vessel. Control for the NGAL immunostaining was conducted by applying excess blocking peptide in a β-gal-transduced vessel at day 14 (top right). The arrows indicate internal elastic lamina, and the arrowheads indicate cells showing representative positive signals for NGAL (brown color). All of the sections are counterstained with hematoxylin. Original magnifications, ×400.

Figure 2

Effect of dnIKKβ on NF-κB activation and the expression of TNF-α, NGAL, and MMP-9 in the medial and intimal SMCs. A: NF-κB activation was assessed by electrophoretic mobility shift assay in both medial and intimal SMCs. Cells are stimulated with IL-1β for 30 minutes with or without preinfection with β-gal or dnIKKβ for 1 hour (h). The specificity of binding was verified by using excess amounts of unlabeled NF-κB or AP-1 probes. The arrow indicates NF-κB-specific DNA-protein complex. B: Cells are stimulated with IL-1β for 6 hours with or without preinfection with β-gal or dnIKKβ. Transcripts of TNF-α, NGAL, and MMP-9 in SMCs were determined by real-time quantitative RT-PCR and normalized to hypoxanthine guanidine ribonucleosyltransferase (HPRT) mRNA. Data are presented as mean ± SEM from three independent experiments. *P < 0.05, versus β-gal group treated with IL-1β; ^&P < 0.05 versus the untouched intimal SMCs; ^#P < 0.05, versus the untouched medial SMCs.

Figure 3

Western blot analysis of NGAL and MMP-9. Proteins extracted from medial and intimal SMCs stimulated with IL-1β for 24 and 48 hours with or without preinfection of β-gal or dnIKKβ were immunoblotted with NGAL antibody, reprobed with MMP-9 antibody after stripping and β-actin antibody for loading control under reducing conditions. Human recombinant NGAL (rNGAL) with a molecular mass of 20.5 kd was loaded as positive control. The identities of all proteins are indicated on the right. Molecular mass markers are indicated on the left. A representative from three independent experiments is shown.

Figure 4

NGAL-MMP-9 complex formation and interaction. A: Western blot analysis of NGAL and MMP-9 under nonreducing condition in the medial and intimal SMCs stimulated with IL-1β for 24 hours in the presence of β-gal or dnIKKβ (see Materials and Methods). Proteins were immunoblotted with NGAL antibody and reprobed with MMP-9 antibody after stripping; rNGAL was loaded as a positive control. β-Actin antibody was used for loading control. B: IP of intimal cell lysates and supernatants with NGAL antibody and Western blot analysis with MMP-9 antibody (top) and reverse IP of supernatants from medial and intimal SMCs with MMP-9 antibody and Western blotting with NGAL antibody (bottom). SMCs were stimulated with or without IL-1β for indicated times. The first lane both in top and bottom panels is the control without extract but immunoprecipitated as other lanes. C: Gelatinase zymographies were analyzed for MMP activities. Three major gelatinase activities were detected with apparent molecular masses of ∼125/115, ∼86, and ∼68 kd from SMC supernatants. SMCs were stimulated with or without IL-1β for 24 and 72 hours. The identities of all proteins are indicated by arrows. Molecular mass markers are shown on the left. A representative result from three independent experiments is shown.

Figure 5

Kinetics of the IL-1β-induced NGAL in human coronary artery SMCs (hCASMCs). A: Transcripts of NGAL in hCASMCs were determined by real-time quantitative RT-PCR. Cells were stimulated with IL-1β for the indicated times. Levels of transcripts are expressed in relative units normalized to cyclophilin A mRNA. Data are presented as mean ± SEM from three independent experiments. *P < 0.05, versus control hCASMCs. B: Western blot analysis with polyclonal goat NGAL antibody on protein extracts from hCASMCs stimulated with IL-1β for indicated times (top); the same membranes were reprobed with β-actin antibody for loading control (bottom). A representative from three independent experiments is shown. C: NF-κB activation in hCASMCs was assessed by electrophoretic mobility shift assay using the NF-κB sequence of the human NGAL promoter as probe. Nuclear extracts are from cells treated with or without IL-1β for 2 hours. The specificity of binding was verified by using excess amounts of unlabeled NF-κB or AP-1 probes. The arrow indicates NF-κB-specific DNA-protein complex.

Figure 6

Gelatinolytic activity in the angioplastic injured carotid arteries by in situ zymography. Frozen sections were prepared at day 14 from the uninjured artery (left) and the injured artery transduced with β-gal (middle) or dnIKKβ (right). A representative result is shown from four of the uninjured, seven of the injured transduced with β-gal, and five of the dnIKKβ transduced. The arrows denote the internal elastic lamina, and the arrowheads indicate external elastic lamina, defining the borders of intima, media, and adventitia.