Matrix metalloproteinase-9 deficiency impairs host defense against abdominal sepsis

Matrix metalloproteinase (MMP)-9 is involved in extracellular matrix degradation and leukocyte migration. To determine the role of MMP-9 in the innate immune response to peritonitis, MMP-9 gene-deficient (MMP-9(-/-)) and normal wild-type mice were i.p. infected with Escherichia coli. MMP-9 mRNA and pro-MMP-9 protein levels increased rapidly upon induction of peritonitis. Although MMP-9(-/-) neutrophils showed a normal phagocytosis of E. coli in vitro, MMP-9(-/-) mice displayed a reduced resistance against E. coli peritonitis, as indicated by an enhanced bacterial outgrowth in the peritoneal cavity and increased dissemination of the infection. Furthermore, the cytokine response to LPS was not influenced by MMP-9 deficiency. However, during E. coli peritonitis, MMP-9(-/-) mice showed much higher peritoneal chemokine and cytokine levels compared with wild-type mice. Despite the increased local chemokine concentrations, MMP-9(-/-) mice displayed a diminished recruitment of leukocytes to the site of infection, indicating that cellular migration was impaired. Moreover, MMP-9(-/-) mice developed more severe distant organ damage during infection. These data suggest that MMP-9 is an essential component of an effective host response to E. coli peritonitis.     

Endogenous tissue-type plasminogen activator is protective during Escherichia coli-induced abdominal sepsis in mice

Sepsis is associated with enhanced production of tissue-type plasminogen activator (tPA). We investigated the function of endogenous tPA in the immune responses to Escherichia coli-induced abdominal sepsis using tPA gene-deficient (tPA(-/-)) and normal wild-type (WT) mice. tPA(-/-) mice demonstrated an impaired defense against E. coli peritonitis as indicated by higher bacterial loads at the primary site of the infection, enhanced dissemination, and reduced survival. The protective function of tPA was independent of plasmin since plasminogen gene-deficient (Plg(-/-)) mice were indistinguishable from WT mice. Relative to WT mice, tPA(-/-) mice demonstrated similar neutrophil counts in the peritoneal cavity despite much higher bacterial loads and higher local concentrations of neutrophil attracting chemokines, suggesting a reduced migratory response. In line, tPA(-/-) mice demonstrated a reduced thioglycolate-induced neutrophil influx into the peritoneal cavity and i.p. injection of WT mice with a replication-defective adenoviral vector expressing tPA caused an enhanced cell migration to the peritoneal cavity during E. coli peritonitis. These findings identify a novel protective function of tPA in abdominal sepsis caused by E. coli that seems independent of its role in the generation of plasmin.     

Axl deficiency does not affect adipogenesis or adipose tissue development

To evaluate a potential role of Axl, the high-affinity receptor of growth arrest-specific protein 6 (GAS6) in adiposity, murine embryonic fibroblasts (MEF) derived from mice with genetic deficiency of Axl (Axl(-/-)) or wild-type littermates (Axl(+/+)) were differentiated into mature adipocytes. In addition, Axl(-/-) and Axl(+/+) mice were kept on standard fat diet (SFD) or on high-fat diet (HFD) for 15 weeks. Deficiency of Axl in MEF did not affect differentiation, as shown by a similar uptake of Oil Red O and expression of the adipogenic markers aP2 and peroxisome proliferator activator receptor γ (PPARγ) at the end of the differentiation. In the first 7 weeks of HFD feeding, Axl(-/-) mice gained less weight than their wild-type littermates. Weight gain for both genotypes on either SFD of HFD over 15 weeks was, however, not significantly different, resulting in comparable body weights, as well as subcutaneous (s.c.) and gonadal (GON) fat mass. Adipocyte size in the fat tissues was not affected by Axl deficiency. Gene expression analysis indicated that the absence of Axl in vivo may be compensated for by the other TAM family members Mer and Tyro3. Glucose and insulin tolerance tests (ITT) in Axl(-/-) and Axl(+/+) mice did not reveal significant differences in glucose homeostasis. Thus, Axl deficiency had no significant effect on adipogenesis in vitro or in vivo.     

Biochemical properties of conjugates of urokinase-type plasminogen activator with a monoclonal antibody specific for cross-linked fibrin

Equimolar mixtures of recombinant single chain urokinase-type plasminogen activator (rscu-PA) and a murine monoclonal antibody (MA-15C5) directed against fragment-D dimer of human cross-linked fibrin were conjugated, using the cross-linking agent N-succinimidyl 3-(2-pyridyldithio)propionate (PySSProSu). The conjugate (rscu-PA/MA-15C5), purified by immunoadsorption on a urokinase antibody and affinity chromatography on fibrin fragment-D dimer with a yield of 42 +/- 15% (mean +/- SD, n = 3), contained an average of 1.2 +/- 0.3 IgG molecules/rscu-PA molecule. On non-reduced SDS/PAGE it migrated as a main band with apparent Mr of 200,000. Specific amidolytic activities expressed/mass of u-PA were less than 250 IU/mg for rscu-PA/MA-15C5 and rscu-PA, 140,000 +/- 13,000 IU/mg and 100,000 +/- 17,000 IU/mg for their plasmin-generated two chain derivatives rtcu-PA/MA-15C5 and rtcu-PA respectively. Specific activities on fibrin plates were 100,000 +/- 24,000 IU/mg and 130,000 +/- 49,000 IU/mg for rscu-PA/MA-15C5 and rtcu-PA/MA-15C5 respectively, as compared to 180,000 +/- 15,000 IU/mg for both rscu-PA and rtcu-PA. Activation of plasminogen with rscu-PA/MA-15C5 (Km = 0.37 +/- 0.16 microM, k2 = 0.0063 +/- 0.0030 s-1 or rtcu-PA/MA-15C5 (Km = 19 +/- 3.0 microM, k2 = 2.0 +/- 0.10 s-1) in purified systems followed Michaelis-Menten kinetics with Km and k2 values comparable to those of rscu-PA and rtcu-PA. In an in vitro system composed of a 125I-fibrin-labeled whole human plasma clot immersed in citrated human plasma, dose- and time-dependent lysis was obtained; 50% lysis in 2 h required 1.4 microgram/ml of rscu-PA or 0.33 microgram/ml of rtcu-PA, but only 0.22 microgram u-PA/ml of rscu-PA/MA-15C5 or 0.15 microgram u-PA/ml of rtcu-PA/MA-15C5. Addition of purified fragment-D dimer reversed the increased fibrinolytic potency of rscu-PA/MA-15C5 in a concentration-dependent way (50% inhibition at 7.2 micrograms fragment-D dimer/ml). Thus, conjugation of u-PA moieties with the fibrin-specific antibody MA-15C5 targets the plasminogen activator to the clot, resulting in a significant increase of their fibrinolytic potencies as compared to their unconjugated counterparts: 6.4-fold for rscu-PA and 2.2-fold for rtcu-PA.     

Stromelysin-2 (MMP-10) deficiency does not affect adipose tissue formation in a mouse model of nutritionally induced obesity

Adipose tissue development is associated with angiogenesis, adipogenesis and extracellular matrix degradation. The class of matrix metalloproteinases contributes to these processes, but little information is available on the role of individual proteinases. We report that stromelysin-2 (MMP-10) deficiency has no significant effect on total body weight or on subcutaneous (SC) or gonadal (GON) adipose tissue mass of mice kept on a high fat diet for 15 weeks. The adipocyte size and density in SC and GON adipose tissues were also comparable in MMP-10 deficient and wild-type control mice. Similarly, blood vessel size and density in obese SC and GON adipose tissues was not affected by MMP-10 deficiency.Metabolic parameters and blood cell composition were similar for both genotypes. Stromelysin-1 (MMP-3) expression was significantly reduced in adipose tissues of the deficient mice as compared to the wild-type controls.These data indicate that MMP-10 does not significantly contribute to adipose tissue development and associated angiogenesis in a mouse model of nutritionally induced obesity.     

Blood Vessel Density in De Novo Formed Adipose Tissue Is Decreased Upon Overexpression of TIMP‐1

Matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) play a role in the development of obesity by contributing to adipogenesis, angiogenesis, and extracellular matrix degradation. We have evaluated a potential functional role of TIMP‐1, which inhibits most MMPs, in in vivo adipogenesis. Therefore, human (h) TIMP‐1 was overexpressed by injection in the tail vein of NUDE mice of an adenoviral vector 3 days before injection of 3T3‐F442A preadipocytes in the back. After 4 weeks of high‐fat diet, the de novo formed fat was analyzed. Overexpression of hTIMP‐1 had no effect on de novo formed fat pad mass. However, the blood vessel density of the fat pads from mice overexpressing hTIMP‐1 was significantly lower than in controls (587 ± 11 mm^?2 vs. 806 ± 20 mm^?2, P < 0.0001) whereas the adipocytes were somewhat larger (1,477 ± 44 µm² vs. 1,285 ± 32 µm², P = 0.03). Thus, in vivo hTIMP‐1 overexpression did not significantly affect the extent of de novo adipose tissue formation, but was associated with significantly lower blood vessel density.     

Prostromelysin-1 (proMMP-3) stimulates plasminogen activation by tissue-type plasminogen activator.

Matrix metalloproteinase-3 (MMP-3 or stromelysin-1) specifically binds to tissue-type plasminogen activator (t-PA), without however, hydrolyzing the protein. Binding affinity to proMMP-3 is similar to single chain t-PA, two chain t-PA and active site mutagenized t-PA (Ka of 6.3 x 106 to 8.0 x 106 M-1), but is reduced for t-PA lacking the finger and growth factor domains (Ka of 2.0 x 106 M-1). Activation of native Glu-plasminogen by t-PA in the presence of proMMP-3 obeys Michaelis-Menten kinetics; at saturating concentrations of proMMP-3, the catalytic efficiency of two chain t-PA is enhanced 20-fold (kcat/Km of 7.9 x 10-3 vs. 4.1 x 10-4 microM-1.s-1). This is mainly the result of an enhanced affinity of t-PA for its substrate (Km of 1.6 microM vs. 89 microM in the absence of proMMP-3), whereas the kcat is less affected (kcat of 1.3 x 10-2 vs. 3.6 x 10-2 s-1). Activation of Lys-plasminogen by two chain t-PA is stimulated about 13-fold at a saturating concentration of proMMP-3, whereas that of miniplasminogen is virtually unaffected (1.4-fold). Plasminogen activation by single chain t-PA is stimulated about ninefold by proMMP-3, whereas that by the mutant lacking finger and growth factor domains is stimulated only threefold. Biospecific interaction analysis revealed binding of Lys-plasminogen to proMMP-3 with 18-fold higher affinity (Ka of 22 x 106 M-1) and of miniplasminogen with fivefold lower affinity (Ka of 0.26 x 106 M-1) as compared to Glu-plasminogen (Ka of 1.2 x 106 M-1). Plasminogen and t-PA appear to bind to different sites on proMMP-3. These data are compatible with a model in which both plasminogen and t-PA bind to proMMP-3, resulting in a cyclic ternary complex in which t-PA has an enhanced affinity for plasminogen, which may be in a Lys-plasminogen-like conformation. Maximal binding and stimulation require the N-terminal finger and growth factor domains of t-PA and the N-terminal kringle domains of plasminogen.