Ilse Van Aelst

The Hemopexin and O-Glycosylated Domains Tune Gelatinase B/MMP-9 Bioavailability via Inhibition and Binding to Cargo Receptors

Gelatinase B/matrix metalloproteinase-9 (MMP-9), a key regulator and effector of immunity, contains a C-terminal hemopexin domain preceded by a unique linker sequence of ∼64 amino acid residues. This linker sequence is demonstrated to be an extensively O-glycosylated (OG) domain with a compact three-dimensional structure. The OG and hemopexin domains have no influence on the cleavage efficiency of MMP-9 substrates. In contrast, the hemopexin domain contains a binding site for the cargo receptor low density lipoprotein receptor-related protein-1 (LRP-1). Furthermore, megalin/LRP-2 is identified as a new functional receptor for the hemopexin domain of MMP-9, able to mediate the endocytosis and catabolism of the enzyme. The OG domain is required to correctly orient the hemopexin domain for inhibition by TIMP-1 and internalization by LRP-1 and megalin. Therefore, the OG and hemopexin domains down-regulate the bioavailability of active MMP-9 and the interactions with the cargo receptors are proposed to be the original function of hemopexin domains in MMPs.

CXCR3 determines strain susceptibility to murine cerebral malaria by mediating T lymphocyte migration toward IFN‐γ‐induced chemokines

Cerebral malaria (CM) results from the binding of infected erythrocytes and leukocytes to brain endothelia. The precise mechanisms underlying lymphocyte recruitment and activation in CM remain unclear. Therefore, the expression of various chemokines was quantified in brains of mice infected with Plasmodium berghei ANKA (PbA). Several chemokines attracting monocytes and activated T‐lymphocytes were expressed at high levels. Their expression was almost completely abrogated in IFN‐γ ligand and receptor KO mice, indicating that IFN‐γ is an essential chemokine inducer in vivo. Surprisingly, the expression levels of chemokines, IFN‐γ and also adhesion molecules in the brain were not lower in CM‐resistant Balb/c and DBA/2 mice compared to CM‐sensitive C57BL/6 and DBA/1 mice, although T lymphocyte sequestration in the brain was significantly less in CM‐resistant than in CM‐sensitive mice. This difference correlated with a higher up‐regulation of the CXC chemokine receptor (CXCR)‐3 on splenic T cells and a higher chemotactic response to IFN‐γ‐inducible protein‐10 (IP‐10) in C57BL/6 compared to Balb/c mice. In conclusion, parasite‐induced IFN‐γ in the brain results in high local expression levels of specific chemokines for monocytes and lymphocytes. The strain‐dependent susceptibility to develop CM is more related to the expression of CXCR3 in circulating leukocytes than to the chemokine expression levels in the brain.

Matrix metalloproteinases, tissue inhibitors of MMPs and TACE in experimental cerebral malaria

Cerebral malaria (CM) is a life-threatening disorder and a major medical problem in developing countries. It is caused by the sequestration of malaria-infected erythrocytes onto brain endothelia, followed by blood–brain barrier (BBB) damage and neurological deficit. In the present study, matrix metalloproteinases (MMPs) were analysed in a mouse model of CM with Plasmodium berghei ANKA. Increased numbers of gelatinase B (MMP-9)-positive cells, which were also CD11b+, were detected in the brain. In addition, activation of gelatinase B occurred in CM brains, and not in brains of mice with non-CM. However, selective genetic knockout of gelatinase B did not alter the clinical evolution of experimental CM. To study other protease balances, the mRNA expression levels of nine matrix metalloproteinases (MMPs), five membrane-type MMPs, TNF-α converting enzyme (TACE) and the four tissue inhibitors of metalloproteinases (TIMPs) were analysed during CM in different organs. Significant alterations in expression were observed, including increases of the mRNAs of MMP-3, -8, -13 and -14 in the spleen, MMP-8, -12, -13 and -14 in the liver and MMP-8 and -13 in the brain. Net gelatinolytic activity, independent of gelatinase B and inhibitable with EDTA, was detected in situ in the endothelia of blood vessels in CM brains, but not in brains of mice with non-CM, suggesting that metalloproteases, different from gelatinase B, are active in the BBB environment in CM. The increase in MMP expression in the brain was significantly less pronounced after infection of C57Bl/6 mice with the noncerebral strain P. berghei NK65, but it was similar in CM-susceptible C57Bl/6 and CM-resistant Balb/C mice upon infection with P. berghei ANKA. Furthermore, in comparison with C57Bl/6 mice, a larger increase in TIMP-1 and a marked, >30-fold induction in MMP-3 were found in the brains of Balb/C mice, suggesting possible protective roles for TIMP-1 and MMP-3.

Tumor Angiogenesis Induced by Granulocyte Chemotactic Protein-2 as a Countercurrent Principle

Chemokine production by tumors is a well-known phenomenon, but its role in tumor biology remains debatable. Although intratumoral injection of granulocyte chemotactic protein-2 (GCP-2) had no effect on tumor parameters, needle-free stable expression of the chemokine resulted in enhanced tumor growth. It is shown here that tumors that express a potent form of GCP-2 induce a strong influx and activation of tumor-associated neutrophils. The production of GCP-2 leads to intratumoral expression of gelatinase B and advantage for tumor growth by increased angiogenesis. These results are in line with the countercurrent principle of chemokine action and support the notion that paraneoplastic expression of ELR-positive CXC chemokines has to be blocked rather than stimulated in cancer therapy.

Immunopathology and Dexamethasone Therapy in a New Model for Malaria-associated Acute Respiratory Distress Syndrome

RATIONALE Malaria infection is often complicated by malaria-associated acute respiratory distress syndrome (MA-ARDS), characterized by pulmonary edema and hemorrhages. No efficient treatments are available for MA-ARDS and its pathogenesis remains poorly understood. OBJECTIVES Development of a new animal model for MA-ARDS to explore the pathogenesis and possible treatments. METHODS C57BL/6 mice were infected with Plasmodium berghei NK65, and the development of MA-ARDS was evaluated by the analysis of lung weight, histopathology, and bronchoalveolar lavages. Cytokine and chemokine expression in the lungs was analyzed by reverse transcription-polymerase chain reaction, and the accumulation of leukocyte subclasses was determined by flow cytometric analysis. MEASUREMENTS AND MAIN RESULTS In this model, the pulmonary expression of several cytokines and chemokines was increased to a higher level than in mice infected with Plasmodium chabaudi AS, which does not cause MA-ARDS. By depletion experiments, CD8(+) T lymphocytes were shown to be pathogenic. High doses of dexamethasone blocked MA-ARDS, even when administered after appearance of the complication, and reduced pulmonary leukocyte accumulation and the expression of a monocyte/macrophage-attracting chemokine. CONCLUSIONS We developed a novel model of MA-ARDS with many similarities to human MA-ARDS and without cerebral complications. This contrasts with the more classical model with P. berghei ANKA, characterized by fulminant cerebral malaria. Hence, infection with P. berghei NK65 generates a broader time window to study the pathogenesis and to evaluate candidate treatments. The finding that high doses of dexamethasone cured MA-ARDS suggests that it might be more effective against MA-ARDS than it was in the clinical trials for cerebral malaria.

Beta-hematin interaction with the hemopexin domain of gelatinase B/MMP-9 provokes autocatalytic processing of the propeptide, thereby priming activation by MMP-3.

Gelatinase B or matrix metalloproteinase-9 is involved in inflammation and in autoimmune and vascular diseases. In contrast to the constitutive and homeostatic matrix metalloproteinase-2, matrix metalloproteinase-9 is an inducible enzyme. Furthermore, it needs tight regulation, and a major control mechanism of its enzymatic activity is the activation of the latent enzyme by proteolysis of the 87 residue propeptide. Activated matrix metalloproteinase-9 is detected in many vascular or hematological disease states, including in an experimental model for cerebral malaria with Plasmodium berghei ANKA. However, insight into its activation mechanism is incomplete. In view of the association with hemorrhagic and hemolytic diseases, it was studied whether and how hemoglobin and its derivatives might activate pro-matrix metalloproteinase-9. Incubation of matrix metalloproteinase-9 with hemin or beta-hematin, the core constituent of hemozoin or malaria pigment, leads to differential autocatalysis of the propeptide, mediated by allosteric interaction with the hemopexin domain. The cleavage catalyzed by beta-hematin coincides with the first cleavage by stromelysin-1/matrix metalloproteinase-3, and preincubation of matrix metalloproteinase-9 with beta-hematin enhances the activation rate by matrix metalloproteinase-3 at least 6-fold. These findings suggest that reduction of hemorrhage and hemolysis might prevent matrix metalloproteinase-9-mediated inflammatory and vascular damages.

Human monocyte-derived dendritic cells produce bioactive gelatinase B: inhibition by IFN-beta.

We studied the secretion of gelatinase B by dendritic cells (DC) generated by culturing human peripheral blood monocytes in granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-4 (IL-4). First, we found the intracellular expression of gelatinase B on sections of fixed DC pellets. Zymography analysis of the supernatants of DC cultured for 72 h demonstrated the presence of gelatinase B. To determine if DC produce net enzymatic activity, bioactive gelatinase, a novel sensitive fluorescent-activated substrate conversion (FASC) assay was used to complement the zymography data. Culture media of unstimulated DC demonstrated reproducible net gelatinolytic activity. Tumor necrosis factor-alpha (TNF-alpha) IL-1beta but not lipopolysaccharide (LPS) stimulation caused a significant increase in gelatinase B production in zymography analysis. Both types of stimulation failed to increase net gelatinase activity in FASC assay. Interestingly, interferon-beta (IFN-beta) significantly diminished both the total zymolytic production and the net bioactive gelatinase produced by DC in a dose-dependent manner. We conclude that human monocyte-derived DC secrete bioactive gelatinase B and that IFN-beta inhibits this production.

Gelatinase A secretion and its control in peritubular and Sertoli cell cultures: effects of hormones, second messengers and inducers of cytokine production

Extracellular matrix components as well as enzymes and enzyme-inhibitors controlling the turn-over of these components play an important role in the local control of testicular function. Zymographic analysis was used to study the secretion and the control of the secretion of gelatinase A (MMP-2) and B (MMP-9) by primary cultures of rat Sertoli cells and by subcultures of peritubular cells. Data on gelatinase A were complemented by measurement of the corresponding mRNA by Northern blot analysis. The agonists investigated included hormones (FSH, testosterone), second messengers (dbcAMP, phorbolester and a Ca(2+)- ionophore), interleukin-1 beta (IL-1 beta) and inducers of cytokine production (Concanavalin A: ConA; lipopolysaccharide: LPS; double stranded RNA: PIC). It is demonstrated that Sertoli cells originally secrete both gelatinase A and B. When maintained in serum-free medium, however, they rapidly lose the ability to secrete gelatinase B. After 3 days of culture gelatinase A remains the only measurable gelatinase in both Sertoli and peritubular cell cultures. The production in peritubular cells, however, exceeds that in Sertoli cells some 25-fold. This was confirmed by a 30-fold difference in the level of steady-state gelatinase A mRNA levels. Gelatinase A secretion and gelatinase A mRNA were stimulated by ovine FSH in Sertoli cells and by dbcAMP and ConA in both Sertoli and peritubular cells. IL-1 beta displayed measurable but limited stimulatory effects in both cell types. Interestingly, in peritubular cells but not in Sertoli cells, ConA stimulated the production of a lower MW species probably representing an activated form of gelatinase A. It is concluded that both the amounts of gelatinase A produced, the levels of the corresponding mRNA and the regulation differ in cultured peritubular cells and Sertoli cells. The lectin concanavalin A is a novel and potent inducer of gelatinase A. It resembles cytochalasin D in selectively inducing an activated form of gelatinase A in peritubular cells. The mechanism responsible for this selective effect warrants further investigation.

Gelatinase B in proliferative vitreoretinal disorders.

PURPOSE To investigate whether gelatinases A and B are involved in the pathogenesis of proliferative vitreoretinal disorders. METHODS In a prospective study of 101 consecutive patients, vitreous and paired serum samples were obtained from 38 patients with rhegmatogenous retinal detachment complicated by proliferative vitreoretinopathy, 25 patients with rhegmatogenous retinal detachment with no proliferative vitreoretinopathy, and 38 patients with proliferative diabetic retinopathy. Gelatinase activities were determined by quantitative zymography. RESULTS All vitreous samples contained comparable levels of the constitutive gelatinase A. Inducible gelatinase B was detected in eight (32%) of 25 vitreous samples from patients with rhegmatogenous retinal detachment with no proliferative vitreoretinopathy (mean +/- SD, 319.5 +/- 521.0 scanning units), in 17 (44.7%) of 38 vitreous samples from patients with proliferative vitreoretinopathy (560.6 +/- 718.9 scanning units), and in 34 (89.5%) of 38 vitreous samples from patients with proliferative diabetic retinopathy (1,707.2 +/- 1,220.3 scanning units). The incidence of detection of gelatinase B in proliferative diabetic retinopathy cases was significantly higher than it was in rhegmatogenous retinal detachment with no proliferative vitreoretinopathy and proliferative vitreoretinopathy cases (P < .001). Gelatinase B levels in the vitreous samples of patients with proliferative diabetic retinopathy were higher than the levels found in patients with rhegmatogenous retinal detachment with no proliferative vitreoretinopathy and in patients with proliferative vitreoretinopathy (P = .0152). Gelatinase A was detected in all the tested sera, whereas none of the tested paired serum samples contained detectable gelatinase B activity. CONCLUSIONS Gelatinase B may play an important role in extracellular matrix degradation associated with neovascularization in proliferative diabetic retinopathy.

Adenylyl cyclase-associated protein-1/CAP1 as a biological target substrate of gelatinase B/MMP-9

Matrix metalloproteinases (MMPs) are classically associated with the turnover of secreted structural and functional proteins. Although MMPs have been shown to process also a kaleidoscope of membrane-associated substrates, little is known about the processing of intracellular proteins by MMPs. Physiological and pathological cell apoptosis, necrosis and tumor lysis by chemotherapy, radiotherapy or immunological cytotoxicity, are examples of conditions in which an overload of intracellular proteins becomes accessible to the action of MMPs. We used a model system of dying human myelomonocytic cells to study the processing of intracellular protein substrates by gelatinase B/MMP-9 in vitro. Adenylyl cyclase-associated protein-1 or CAP1 was identified as a novel and most efficient substrate of gelatinase B/MMP-9. The presence of CAP1 in the extracellular milieu in vivo was documented by analysis of urine of patients with systemic autoimmune diseases. Whereas no active MMP-9 could be detected in urines of healthy controls, all urine samples of patients with clinical parameters of renal failure contained activated MMP-9 and/or MMP-2. In addition, in some of these patients indications of CAP1 cleavage are observed, implying CAP1 degradation in vivo. The high turnover rate of CAP1 by MMP-9, comparable to that of gelatin as the natural extracellular substrate of this enzyme, may be critical to prevent pathological conditions associated with considerable cytolysis.