Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Aug 27;9(1):12421.
doi: 10.1038/s41598-019-48741-w.

SerpinB2 Inhibits Migration and Promotes a Resolution Phase Signature in Large Peritoneal Macrophages

Affiliations
Free PMC article

SerpinB2 Inhibits Migration and Promotes a Resolution Phase Signature in Large Peritoneal Macrophages

Wayne A Schroder et al. Sci Rep. .
Free PMC article

Abstract

SerpinB2 (plasminogen activator inhibitor type 2) has been called the "undecided serpin" with no clear consensus on its physiological role, although it is well described as an inhibitor of urokinase plasminogen activator (uPA). In macrophages, pro-inflammatory stimuli usually induce SerpinB2; however, expression is constitutive in Gata6+ large peritoneal macrophages (LPM). Interrogation of expression data from human macrophages treated with a range of stimuli using a new bioinformatics tool, CEMiTool, suggested that SerpinB2 is most tightly co- and counter-regulated with genes associated with cell movement. Using LPM from SerpinB2-/- and SerpinB2R380A (active site mutant) mice, we show that migration on Matrigel was faster than for their wild-type controls. Confocal microscopy illustrated that SerpinB2 and F-actin staining overlapped in focal adhesions and lamellipodia. Genes associated with migration and extracellular matrix interactions were also identified by RNA-Seq analysis of migrating RPM from wild-type and SerpinB2R380A mice. Subsequent gene set enrichment analyses (GSEA) suggested SerpinB2 counter-regulates many Gata6-regulated genes associated with migration. These data argue that the role of SerpinB2 in macrophages is inhibition of uPA-mediated plasmin generation during cell migration. GSEA also suggested that SerpinB2 expression (likely via ensuing modulation of uPA-receptor/integrin signaling) promotes the adoption of a resolution phase signature.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
CEMiTool and IPA analyses of mRNA microarray data. (a) Microarray data sets from human monocytes/macrophages cultured under 65 different conditions were analyzed using CEMiTool. CEMiTool identified 129 genes (Table S1a) that were either (i) co-regulated with SerpinB2 (i.e. genes usually up-regulated when SerpinB2 was up-regulated) or (ii) counter-regulated with SerpinB2 (i.e. genes usually down-regulated when SerpinB2 was up-regulated). Expression z scores for SerpinB2 and the 129 genes after the 65 different treatments are shown. (b) Co-regulated genes were given a nominal expression value of 2 and counter-regulated genes were give a nominal expression value of −2, and the gene list analyzed using IPA (Direct only); 117 of the 129 genes were recognized by IPA. IPA identified Cellular Movement as the top scoring Molecular and Cellular Function; 51/117 genes appeared in 1 or more of the 96 Categories/Diseases or Functions Annotations that comprise Cellular Movement (Table S1b). The box plots are provided by IPA for the indicated function; each of the many (e.g. 96) internal boxes represents one annotation, with the box size indicating log10 p value and the color indicating the z-score.
Figure 2
Figure 2
LPM in SerpinB2−/− and SerpinB2+/+ mice. (a) Peritoneal lavage cells from SerpinB2−/− and SerpinB2+/+ mice were stained with CD11b and F4/80 and analyzed by FACS. Numbers (±SD) show the percentage of SPM and LPM (gates indicated) as a proportion of all peritoneal lavage cells (n = 3 mice per group). The F4/80hi lines indicate the cut-off used to FACS sort LPM. (b) Immunofluorescent antibody staining of FACS-sorted LPM using anti-murine SerpinB2 antibody and FITC-phalloidin (F-actin) imaged using confocal microscopy. All three images for SerpinB2+/+ LPM and SerpinB2−/− LPM are of the same field. (c) Immunoblotting of LPM from SerpinB2+/+ and SerpinB2−/− mice using anti-murine SerpinB2 and anti-GAPDH antibodies. Full length gels are shown in Fig. S15.
Figure 3
Figure 3
RPM migration and length of membrane protrusions. (a) Migration into “scratch wounds” of ex vivo SerpinB2−/− and SerpinB2+/+ RPM seeded onto Matrigel-coated wells analyzed using the standard IncuCyte scratch wound assay. Data was generated (for both SerpinB2−/− and SerpinB2+/+ RPM) from 3 pools of RPM, each derived from 3 mice, with each pool seeded into 12 wells (i.e. n = 9 SerpinB2−/− and n = 9 SerpinB2+/+ mice). The mean of the 3 pools is shown (n = 3). Statistics by repeat measures ANOVA. (b) As for (a) but using ex vivo RPM from SerpinB2R380A and C57BL/6J mice, one pool of 3 mice (for each strain) and 23 Matrigel coated wells. (c) Western blot of RPM from the indicated mice strains, stained with the murine anti-SerpinB2 antibody. (d) Length of cellular protrusions of RPM within the scratch areas for mice described in (a). The number of cells interrogated is provided within each bar (n), with the longest protrusion from each cell used (i.e. one value for each cell). (e) As for (d) for mice described in (b). (f,g) Representative phase images of the RPM described in (d) at 25 hours post seeding. See also Fig. S3d,e.
Figure 4
Figure 4
Confocal immunofluorescence microscopy. Adherent LPM from wild-type mice were dual labeled with FITC-phalloidin and anti-SerpinB2 antibody. (a) Actin; FITC-phalloidin (F-actin) staining showing actin concentrations at focal adhesions (green). SerpinB2; SerpinB2 staining localized around and in the focal adhesions (red). Merge; overlapping staining (yellow); see Fig. S5 for enlargements and quantitation of overlap. (b) Actin; FITC-phalloidin staining showing F-actin concentration along the leading edge of a lamellipodia (bottom left, green). SerpinB2; SerpinB2 staining (red) also staining the leading edge of a lamellipodia (bottom left, red). Merge; overlapping staining at the leading edge of a lamellipodia (yellow, arrowheads). (c) The two clusters of focal adhesions indicated in b (arrows) are enlarged to show overlapping localization of actin and SerpinB2 staining. The images shown are representative of 2 independent experiments where 5 images were examined containing ≈10 cells per image with ≈50 cells examined per experiment. Approximately 30% of cells examined showed overlapping staining at focal adhesions and/or lamellipodia.
Figure 5
Figure 5
Summary of RNA-Seq data and pre-ranked GSEA for resolution phase macrophages. (a) RPM from SerpinB2R380A and C57BL/6 mice were seeded onto Matrigel for 24 h and were then analyzed by RNA-Seq; 1481 DEGs were identified (using a filter of q < 0.01); 624 DEGs were up-regulated, and 857 were down-regulated in SerpinB2R380A RPM. These DEGs were analyzed by Enrichr and IPA. (b) Up-regulated DEGs from resolution phase macrophages (Table S1g) were analyzed by GSEA against the complete RNA-Seq gene list (Table S1c), with genes pre-ranked by p-value. The NES score (normalized enrichment score) and false discovery rate (FDR) q values are indicated. (c) Heat map of core enriched genes from b, illustrating that genes up-regulated in wild-type (WT) RPM (relative to SerpinB2R380A RPM) are also up-regulated in resolution phage macrophages (relative to naïve and inflammatory macrophages). The core enriched gene list is also provided in Table S1g.
Figure 6
Figure 6
GSEA using microarray data sets from Gata6−/− LPM. The 624 up-regulated and the 857 down-regulated DEGs in SerpinB2R380A RPM were compared by GSEA with 3 microarray data sets that compared LPM from Gata6+/+ and Gata6−/− mice; (a) GSE47049, (b) GSE56684 and (c) GSE37448. GSEAs showing significant counter-regulation (i.e. up in SerpinB2R380A and down in Gata6−/− or down in SerpinB2R380A and up in Gata6−/−) are indicated with red p values. Heat maps of core enriched genes showing counter-regulation are shown for (a,b) and for Down DEGs in (c), which showed counter-regulation that did not reach significance (q = 0.094). Up DEGs in (c) showed co-regulation (positive NES score) indicating no significant counter-regulation. The core enriched genes illustrated by the heat maps are listed in Table S1h.

Similar articles

See all similar articles

Cited by 2 articles

References

    1. Ritchie H, Robbie LA, Kinghorn S, Exley R, Booth NA. Monocyte plasminogen activator inhibitor 2 (PAI-2) inhibits u-PA-mediated fibrin clot lysis and is cross-linked to fibrin. Thromb Haemost. 1999;81:96–103. doi: 10.1055/s-0037-1614425. - DOI - PubMed
    1. Corsetti JP, et al. Influences on plasminogen activator inhibitor-2 polymorphism-associated recurrent cardiovascular disease risk in patients with high HDL cholesterol and inflammation. Atherosclerosis. 2016;250:1–8. doi: 10.1016/j.atherosclerosis.2016.04.017. - DOI - PubMed
    1. Kruithof EK, Baker MS, Bunn CL. Biological and clinical aspects of plasminogen activator inhibitor type 2. Blood. 1995;86:4007–4024. - PubMed
    1. Medcalf RL. Plasminogen activator inhibitor type 2: still an enigmatic serpin but a model for gene regulation. Methods Enzymol. 2011;499:105–134. doi: 10.1016/B978-0-12-386471-0.00006-7. - DOI - PubMed
    1. Al-Horani RA. Serpin regulation of fibrinolytic system: implications for therapeutic applications in cardiovascular diseases. Cardiovasc Hematol Agents Med Chem. 2014;12:91–125. doi: 10.2174/1871525712666141106095927. - DOI - PubMed

Publication types

LinkOut - more resources

Feedback