|Home | About | Journals | Submit | Contact Us | Français|
Plg-RKT is a novel integral membrane plasminogen receptor that binds plasminogen via a C-terminal lysine exposed on the cell surface and promotes plasminogen activation on the cell surface by both tissue plasminogen activator and urokinase plasminogen activator.
To evaluate the role of Plg-RKT in vivo we generated Plg-RKT-/- mice using a homologous recombination technique.
We characterized the effect of Plg-RKT deletion on reproduction, viability, health, and spontaneous thrombosis and inflammation.
Plg-RKT-/- mice were viable and fertile. Survival of Plg-RKT-/- mice and Plg-RKT+/+ littermates was not significantly different. However, quite strikingly, all pups of Plg-RKT-/- females died within 2 days of birth, consistent with a lactation defect in Plg-RKT-/- mothers. Additionally, there was a significant effect of Plg-RKT deficiency on growth rates of female, but not male mice. In experimental peritonitis studies, Plg-RKT-/- mice exhibited a marked defect in macrophage recruitment. As a contributing mechanism, the capacity of Plg-RKT-/- macrophages for plasminogen binding was markedly decreased.
These studies demonstrate that Plg-RKT is required for plasminogen binding and macrophage migration in vivo. In addition, Plg-RKT deficiency is not compatible with survival of the species, due to death of all offspring of Plg-RKT-/- females. This new mouse model will be important for future studies aimed at delineating the role of cell surface plasminogen activation in challenge and disease models in vivo.
The activation of plasminogen to the broad spectrum serine protease, plasmin, is markedly enhanced when plasminogen is co-localized with plasminogen activators on cell surfaces [reviewed in ]. Plasmin remains bound to the cell surface where it is relatively protected from inactivation by α2-antiplasmin [reviewed in ]. Thus, a local nidus of cell membrane-associated broad spectrum proteolytic activity is achieved, a key feature of pathological processes in which plasmic degradation of extracellular matrices is required for efficient cell migration [reviewed in ].
Studies with plasminogen null (Plg-/-) mice have demonstrated a major physiologic role for plasminogen in functions consistent with a role for the interaction of plasminogen with cell surfaces. These include wound healing, tissue remodeling, tumor growth and dissemination, and inflammation [reviewed in ]. In particular, these studies have demonstrated a key role for cell surface binding of plasminogen via proteins exposing a C-terminal basic residue in the inflammatory cell recruitment response [3-5].
The plasminogen receptor, Plg-RKT, is a recently discovered integral membrane protein that binds plasminogen via a C-terminal lysine exposed on the cell surface [6,7]. Plg-RKT also binds tissue plasminogen activator (tPA) and promotes tPA-dependent plasminogen activation [6,8]. Furthermore, Plg-RKT is highly colocalized with the urokinase plasminogen activator (uPA) receptor (uPAR) [6,8] and promotes uPA-dependent plasminogen activation . Notably, Plg-RKT is highly conserved with high interspecies homology, high identity, no gaps in the sequences and the presence of a C-terminal lysine for all 36 mammalian orthologs for which sequence information is available [see , Table 2 for alignment of 20 representative orthologs].
In order to assess the physiological function of Plg-RKT we disrupted the PLGRKT gene in mice, and characterized the effect of Plg-RKT deletion on reproduction, viability, health, spontaneous thrombosis, and inflammation.
Human Glu-plasminogen was purified as described [10,11] and labeled with fluorescein isothiocyanate (FITC) using a FITC labeling kit (EMD, EMD Millipore, Darmstadt, Germany). Phycoerythrin labeled annexin V was from R&D Systems, Minneapolis, MN. Antibodies used were: polyclonal rabbit anti-casein (Abbiotec, San Diego, CA), anti-β-actin MAb (LI-COR, Lincoln, NE), and HRP-conjugated secondary antibody (Biosource, San Diego, CA) or appropriate fluoresceinated secondary antibodies (LI-COR, Lincoln, NE).
Plg-RKT-/- mice were generated as described in Fig.1 and backcrossed ten generations into the C57Bl/6J background.
Plg-RKT-/- and Plg-RKT+/+ sex matched mice (12 to 14 weeks of age) were injected intraperitoneally with 4% thioglycollate (Sigma-Aldrich, St. Louis, MO). At specific time points, mice were killed and peritoneal leukocytes were collected by peritoneal lavage. All animal experiments were approved by the Institutional Animal Care and Use Committee of The Scripps Research Institute.
Quantitative flow cytometry was performed and binding parameters were analyzed as described 
Blood cell counts and differentials were determined using a Vet ABC hematology analyzer (SCIL Animal Company, Gurnee, IL).
Prothrombin Time (PT) and Activated Partial Thromboplastin Time (APTT) were determined in duplicate with an ST4 semi-automated mechanical coagulation instrument (Diagnostica Stago, Parsippany, NJ).
Plasma plasminogen and fibrinogen levels were determined using Mouse Plasminogen Assay Kit (IMPLGKT-TOT) and Mouse Fibrinogen Antigen ELISA kit (MFBGNKT-1212), respectively, from Innovative Research, Inc, Novi, MI.
Harvested mouse tissues were fixed in 4% paraformaldehyde, paraffin processed, and embedded. The slides were either stained with hematoxylin & eosin or immunostained for the presence of fibrin(ogen) using goat polyclonal anti-fibrinogen (GAM/Fbg Batch #4951, Nordic Immunological Labs, Eindhoven, The Netherlands) as described .
To study the role of Plg-RKT in murine health and reproduction, we generated a Plg-RKT null mouse strain using a homologous recombination-based technique as described in Fig. 1. Heterozygous mice were crossed and viable Plg-RKT+/-, Plg-RKT-/- and Plg-RKT+/+ offspring were born. Plg-RKT gene disruption was confirmed by PCR genotyping with specific primers (Fig. 1C) and Plg-RKT protein was not detected in Plg-RKT-/- cells (Fig. 1D).
Among 2,340 littermates resulting from heterozygous matings, 611 were Plg-RKT+/+ (26%), 1,136 were Plg-RKT+/- (49%) and 593 were Plg-RKT-/- (25%). This distribution was not significantly different (Chi-square test) from the expected Mendelian 1:2:1 ratios. Additionally, the distribution of genotypes among male and female mice was not significantly different from the expected Mendelian ratios. Thus, Plg-RKT+/+, Plg-RKT+/- and Plg-RKT-/- mice of both sexes were equally viable and there was no evidence of fetal loss of Plg-RKT-/- mice.
Male Plg-RKT-/- mice (age 10 weeks, n=5), mated for 12-19 weeks to female wild-type C57Bl/6J female mice (age 6 weeks, n=5) produced 3 to 6 litters each of 6.5±0.5 offspring per litter. These values were not significantly different from the breeding characteristics of Plg-RKT+/+ male littermates (age, 10 weeks, n=6) that produced 3 to 5 litters each of 6.7±0.7 offspring per litter (P=0.769).
Female Plg-RKT-/- mice (age 8 weeks, n=6), mated for 15-19 weeks to male wild-type C57Bl/6J male mice (age 8 weeks, n=6) produced 4 to 5 litters each of 5.1±0.5 offspring per litter. These values were not significantly different from the breeding characteristics of Plg-RKT+/+ female littermates that produced 3-8 litters of 5.2±0.8 offspring per litter (P=0.923). Notably, plasminogen deficiency does not affect ovulation, when follicular wall rupture and ovulation efficiency are analyzed directly . Thus, our results exclude a role for Plg-RKT in regulating ovulation via other proteases.
However, quite strikingly, no offspring of Plg-RKT-/- mothers survived to weaning (dying within 48 hours after birth) (Fig. 2A). No milk was observed in the stomachs of offspring of Plg-RKT-/- females. There was a significant maternal gene dosage effect on pup survival, with pups of Plg-RKT+/- female mice, having survival rates intermediate between offspring of Plg-RKT-/- and Plg-RKT+/+ female mice (Fig. 2A). These results are consistent with a requirement for Plg-RKT for lactational competence. A less severe lactation defect is present in Plg-/- mice [14,15]. Milk production (assessed by casein content) was severely and significantly decreased in mammary glands harvested 2 days postpartum from Plg-RKT-/- females (Fig. 2B). Notably, milk production by Plg-/- females is not affected at either 2 or 7 days postpartum . Because milk production by secretory epithelium requires a correctly oriented extracellular matrix (ECM) [16-19], it is possible that Plg-RKT may regulate proteases in addition to plasmin that are involved in ECM remodeling in the mammary gland.
Survival curves for Plg-RKT-/-, Plg-RKT+/-, and Plg-RKT+/+ offspring of heterozygous matings (showing plots for mice that survived to be weaned) were not significantly different [P=0.297, Log-rank (Mantel Cox) test] and there was no significant trend (P=0.318, Log-rank test for trend) (Fig. 2C). No consistencies in causes of death were apparent for the Plg-RKT-/- and Plg-RKT+/- in these cohorts. Furthermore, all Plg-RKT-/- mice in our cohort of 28 mice survived for at least 12 months.
Total body weights were determined weekly for 60 weeks (beginning at 4.5 weeks). Plg-RKT deficiency significantly affected the growth rate of female mice (Fig. 2D) but there was no significant effect on growth rates of male mice (Fig. 2E). Effects of Plg-RKT deletion on growth rates of female mice may be due to effects on mammary gland growth and development. Plg-/- mice of both sexes exhibit decreased growth rates compared with wild type littermates [20,21].
Spontaneous dermatitis developed by 4 months and was observed 4-fold more frequently in Plg-RKT-/- mice [16/838 (1.9%)] compared with Plg-RKT+/+ littermates [6/1164 (0.5%)] and 2-fold more frequently in Plg-RKT+/- mice [(22/1884 (1.2%)] compared with Plg-RKT+/+ littermates (X2=8.497, df=2, P=0.014). Extensive non-healing skin ulcerations are also observed in mice doubly deficient for tPA and uPA , consistent with a requirement for plasminogen activation for maintenance of skin health.
Upon dissection, gross examination, and examination of H&E-stained tissues including liver, lung, brain, heart muscle, skeletal muscle, lymph nodes, thymus, spleen, stomach, small intestine, large intestine, colon, cecum, rectum, pancreas, adrenal, kidney, uterus, ovaries, vas deferens, testes, epididymis, (collected from 12 Plg-RKT+/+ and 12 Plg-RKT-/- littermates sacrificed at 8-9 months), no differences were detected between the genotypes. Furthermore, Plg-RKT deficiency did not have a significant effect on either spontaneous fibrin deposition or spontaneous inflammation. In contrast, Plg-/- mice survive to adulthood but exhibit severe spontaneous thrombotic lesions in multiple organs [23,24]. This results in retarded postnatal growth after weaning , severe wasting and high mortality [23,24]. Notably, removal of fibrinogen eliminates these effects of plasminogen deficiency on survival, growth and health . Thus, surveillance against spontaneous thrombosis may be receptor independent or, alternatively, may be accomplished by redundant functions of additional plasminogen receptors. Notably, mice null for components of the annexin A2 heterotetramer (a phosphatidyl-serine-linked cell surface plasminogen binding profibrinolytic complex) also do not exhibit extensive fibrin-dependent organ damage and there is no reported effect of A2 heterotetramer deletion on murine health and survival [21,26].
There was no effect of Plg-RKT deletion on general hematological parameters with regard to platelet counts, leukocyte counts, erythrocyte counts, hemoglobin and hematocrit or on specific coagulation and fibrinolytic parameters (Table 1). Notably, higher plasma fibrinogen concentrations are present in both Plg-/- mice [24,27] and mice expressing plasminogen with a plasmin-inactivating site mutation , which has been attributed to an acute-phase response in the spontaneous disease resulting from the absence of active plasmin .
There was no genotype effect of Plg-RKT deletion on plasma plasminogen levels (Table 1). However, there was a gender effect on circulating plasminogen, with males having significantly lower plasminogen concentrations (P=0.003). Notably, female sex is also associated with higher plasminogen levels in humans .
There was no effect of Plg-RKT deletion on general blood chemistry values (glucose, blood urea nitrogen, creatinine, electrolytes, and liver function panel (data not shown).
Because Plg-RKT is a plasminogen binding protein with a C-terminal lysine, we compared thioglycollate-elicited leukocyte recruitment in Plg-RKT-/- mice compared to wild type littermate controls. Leukocyte recruitment at 72 hr was substantially decreased in Plg-RKT-/- mice compared with Plg-RKT+/+ littermates (Fig. 3A).
Because monocyte recruitment peaks at 72 hr in this model , we further purified macrophages from the recruited peritoneal leukocyte population by adherence. Macrophage recruitment in Plg-RKT-/- null mice was markedly decreased (by 76%) compared with wild type littermate controls (Fig. 3B). The reduction in levels of recruited macrophages was not due to decreased levels of resident peritoneal cells in Plg-RKT-/- mice (Fig. 3A, see 0 time point). The time course of the effect on leukocyte recruitment was similar to that reported for Plg-/- mice .
Because macrophage recruitment in response to thioglycollate requires plasminogen binding , we assessed plasminogen binding to Plg-RKT-/- macrophages. Plasminogen binding to thioglycollate elicited macrophages was evaluated using quantitative fluorescence activated cell sorting as described . Macrophages were identified by their distinct forward scatter/side scatter properties and annexin V was used to gate for cells that were not apoptotic [i.e., did not express phosphatidyl serine, which binds annexin V, on their surfaces]. Specific binding of FITC-plasminogen to the cells was calculated by subtracting binding in the presence of 0.2 M EACA (non-specific binding) from total binding.
Specific binding isotherms for the interaction of FITC-plasminogen with viable, non-apoptotic, Plg-RKT+/+ and Plg-RKT-/- macrophages were saturable (Fig. 3C). Apparent Kd values of 2.2±0.5 μM and 0.4±0.1 μM and Bmax values of=17,547±2,065 molecules/cell and 3,452±291 molecules/cell were calculated for Plg-RKT+/+ (R2=0.993) and Plg-RKT-/- (R2=0.967) macrophages, respectively (Fig. 3D). Thus, the capacity of Plg-RKT-/- macrophages for plasminogen was only 20% of that of Plg-RKT+/+ macrophages. This is consistent with the established requirement for binding of plasminogen to the cell surface via C-terminal basic residues as a contributing mechanism in inflammatory cell recruitment in response to sterile peritonitis induced by thioglycollate [3-5], and demonstrates a major role for Plg-RKT in this process.
In summary, Plg-RKT-/- mice mimic some, but not all, spontaneous phenotypes exhibited by Plg-/- mice , consistent with a role for this plasminogen receptor in specific physiological and pathological functions. Notably, although Plg-RKT deficiency was compatible with survival of individual mice, Plg-RKT deficiency is not compatible with survival of the species due to death of all offspring of Plg-RKT-/- females. This new mouse model will be important for future studies aimed at delineating the role of cell surface plasminogen activation in challenge and disease models in vivo.
This work was supported by the National Institutes of Health (grants HL 081046 to L.A.M. and CA166473 to B.M.M. and L.A.M. and 2P30CA023100, NIH/NCI to N.M.V.) and by Merit Review Award #5I01BX002026 from the U.S. Department of Veterans Affairs (to R.J.P.).
Authors state no conflicts of interests.
Addendum: L. A. Miles and R. J. Parmer designed research studies, analyzed data and wrote the manuscript; N. Baik, S. Lighvani, S. Khaldoyanidid and H. Bai conducted experiments, acquired data and analyzed data; N. M. Varki and B. M. Mueller analyzed data.