Neuropilins (NRPs) are transmembrane receptors that bind class 3 semaphorins and VEGF family members to regulate axon guidance and angiogenesis. Although expression of NRP1 by vascular smooth muscle cells (SMCs) has been reported, NRP function in smooth muscle (SM) in vivo is unexplored. Using Nrp2+/LacZ and Nrp2+/gfp transgenic mice, we observed robust and sustained expression of Nrp2 in the SM compartments of the bladder and gut, but no expression in vascular SM, skeletal muscle, or cardiac muscle. This expression pattern was recapitulated in vitro using primary human SM cell lines. Alterations in cell morphology after treatment of primary visceral SMCs with the NRP2 ligand semaphorin-3F (SEMA3F) were accompanied by inhibition of RhoA activity and myosin light chain phosphorylation, as well as decreased cytoskeletal stiffness. Ex vivo contractility testing of bladder muscle strips exposed to electrical stimulation or soluble agonists revealed enhanced tension generation of tissues from mice with constitutive or SM-specific knockout of Nrp2, compared with controls. Mice lacking Nrp2 also displayed increased bladder filling pressures, as assessed by cystometry in conscious mice. Together, these findings identify Nrp2 as a mediator of prorelaxant stimuli in SMCs and suggest a novel function for Nrp2 as a regulator of visceral SM contractility.
Neuropilins 1 and 2 (NRP1 and NRP2) are 130-kDa transmembrane receptors expressed on a range of cell types that mediate the effects of two independent ligand families: class 3 semaphorins (SEMA3) and members of the vascular endothelial growth factor (VEGF) family. Interaction of neuropilins with SEMA3 family members regulates axonal guidance in the central and peripheral nervous systems (reviewed by Bagri et al
1
). Neuropilins have been shown to regulate angiogenesis by acting as coreceptors for VEGF on endothelial cells. Although both neuropilins bind SEMA3 proteins, they display distinctive ligand binding preferences. SEMA3A binds preferentially to NRP1, whereas SEMA3F binds with high affinity to NRP2. Similarly, the VEGF binding profiles of NRP1 and NRP2 are distinct: whereas both receptors bind VEGF-A, only NRP2 binds VEGF-C and VEGF-D (reviewed by Bielenberg and Klagsbrun2
). Members of the plexin family, in particular plexins A1 through A4, are also necessary for SEMA3-mediated neuropilin-dependent signaling. Binding of SEMA3 ligands to neuropilins initiates a cascade of signals leading to marked changes in cell shape, which in turn mediate repulsion of NRP-expressing cells from the SEMA3 source. Such alterations in cell phenotype underlie the well-characterized growth cone collapse that drives axon guidance,3
, 4
as well as the antiangiogenic effects of SEMA3 ligands described previously.5
, 6
The expression patterns for NRP1 and NRP2 are largely nonoverlapping in vivo, suggesting discrete functions for the two proteins. Transgenic overexpression of Nrp1 resulted in embryonic lethality arising from excess and dilated vessels, hemorrhage, and malformation of the heart and limbs.
7
Conversely, genetic ablation strategies in mice and other experimental organisms revealed functions for Nrp1 in neuron guidance and cardiovascular development.8
, 9
, 10
The demonstration that both overexpression and deletion of Nrp1 evoke profound developmental abnormalities illustrates the requirement for Nrp1 function during essential developmental processes, but also suggests that NRP1 activity is dose-dependent. Mice with targeted mutation of Nrp2 survive to adulthood, but display aberrant development and/or organization of cranial and spinal nerves, as well as a profound decrease in lymphatic capillaries.11
, 12
, 13
, 14
Mice lacking both Nrp1 and Nrp2 die early in embryonic development (embryonic day 8.5) as a result of impaired angiogenesis of the yolk sac and other structures.15
- Takashima S.
- Kitakaze M.
- Asakura M.
- Asanuma H.
- Sanada S.
- Tashiro F.
- Niwa H.
- Miyazaki Ji J.
- Hirota S.
- Kitamura Y.
- Kitsukawa T.
- Fujisawa H.
- Klagsbrun M.
- Hori M.
Targeting of both mouse neuropilin-1 and neuropilin-2 genes severely impairs developmental yolk sac and embryonic angiogenesis.
Proc Natl Acad Sci USA. 2002; 99: 3657-3662
Analyses of signaling pathways that underlie neuropilin-mediated effects, such as collapse and inhibition of migration, have identified Rho family G proteins as important regulatory nodes.
16
, 17
, 18
We recently demonstrated that SEMA3F signals via a NRP2-plexin A1 complex to mediate cytoskeletal collapse in tumor and endothelial cells.17
In that study, we identified ABL2/ARG and RhoA/Rho kinase (ROCK) as components of a signaling cascade leading to cofilin-mediated actin depolymerization and an ensuing reduction in cell contractility and migration. RhoA and ROCK are also known regulators of smooth muscle (SM) function, namely, the maintenance of tone and contractility in vessels and other hollow organs (reviewed by Puetz et al19
). Although several reports have described expression of NRP1 in vascular smooth muscle cells (SMCs) and have identified a role for NRP1 in regulation of SMC migration,20
, 21
, 22
, 23
the function of neuropilins in SM in vivo has been unexplored. SEMA3-dependent activation of neuropilins elicits effects on the cytoskeleton similar to those seen with inhibition of the Rho-ROCK axis. These observations imply that neuropilins are important regulators of the SM phenotype and provide a compelling rationale for determining their function in SM-rich hollow organs.With the present study, we identify visceral SM as a major site of expression of Nrp2 in vivo. Alterations in SM phenotype contribute to various pathologies affecting hollow organs, such as fibrosis and obstruction. Using a combination of biochemical and functional evaluation of cytoskeletal integrity, together with mouse models of Nrp2 deficiency, we show that loss of Nrp2 in vivo enhances agonist-induced force generation in muscle strips and leads to major alterations in hollow organ function. Collectively, these findings implicate Nrp2 as a novel regulator of visceral SM contractility.
Materials and Methods
Ethics Statement
All animal studies were performed with approval from the Children's Hospital Boston Animal Care and Use Committee and with strict adherence to US Public Health Service and Office of Laboratory Animal Welfare guidelines.
Nrp2 Transgenic Mice
Nrp2+/LacZ mice were generated as described previously.
15
Briefly, a targeting vector was constructed in which the translated portion of the first coding exon and the proximal part of the next intron of Nrp2 were replaced with a promoterless Escherichia coli β-galactosidase gene to produce Nrp2+/LacZ mice. These mice were backcrossed to the C57BL/6 strain for >10 generations. Pups from heterozygous offspring mated to produce Nrp2LacZ/LacZ (effectively, Nrp2−/−) die soon after birth. Nrp2+/gfp mice (also known as Nrp2tm1.2Mom/MomJ) were purchased from the Jackson Laboratory (Bar Harbor, ME; stock no. 006700) and maintained in the C57BL/6 background. Nrp2gfp/gfp mice were viable and fertile and were used for contractility experiments. SM-specific deletion of Nrp2 after birth was accomplished by breeding mice harboring an inducible Cre recombinase (CreERT2) under control of the SM22α promoter (hereafter referred to as SM22α-CreERT2)- Takashima S.
- Kitakaze M.
- Asakura M.
- Asanuma H.
- Sanada S.
- Tashiro F.
- Niwa H.
- Miyazaki Ji J.
- Hirota S.
- Kitamura Y.
- Kitsukawa T.
- Fujisawa H.
- Klagsbrun M.
- Hori M.
Targeting of both mouse neuropilin-1 and neuropilin-2 genes severely impairs developmental yolk sac and embryonic angiogenesis.
Proc Natl Acad Sci USA. 2002; 99: 3657-3662
24
to those expressing a floxed allele of Nrp2 (Nrp2fl/fl), and treating offspring positive for both genotypes with 0.5 mg 4-hydroxytamoxifen (4-OHT) for 3 to 5 days. Tissues from mice were harvested for analysis at selected times after initiation of recombination.Histological Analysis and IHC
After euthanasia, whole Nrp2+/LacZ newborn pups (postnatal day P1) or organs (bladder, heart, and skeletal muscle) from Nrp2+/LacZ mice (P28 or P56) were directly embedded in optimal cutting temperature compound (OCT: Sakura, Tokyo, Japan) and frozen with liquid nitrogen. For detection of β-galactosidase activity, cryosections (8 to 10 μm thick) were fixed in cold methanol for 10 minutes,
25
rinsed with PBS (pH 7.2), and incubated overnight at 37°C in X-gal reagent [1 mg/mL 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside diluted in dimethyl sulfoxide, 5 mmol/L K3Fe(CN)6, 5 mmol/L K4Fe(CN)6, and 2 mmol/L MgCl2 in PBS (pH 6.5)]. The next day, sections were rinsed in PBS, counterstained with eosin Y solution alcoholic with phloxine (Sigma-Aldrich, St. Louis, MO), and mounted with Permount medium (Thermo Fisher Scientific, Waltham, MA). For IHC staining, cryosections were fixed in 2% paraformaldehyde in PBS for 10 minutes (for mNrp2 staining), rinsed with PBS, incubated in 3% H2O2 in methanol for 12 minutes to block endogenous peroxidases, rinsed with PBS, incubated in protein blocking solution (3% goat serum, 2% sheep serum in PBS) for 15 minutes, and then incubated overnight at 4°C in primary antibody (rabbit monoclonal (D39A5) anti-mouse Nrp2 IgG antibody (3366; Cell Signaling Technology, Danvers, MA) diluted in protein blocking solution. The next day, for Nrp2 staining, sections were rinsed with PBS, incubated for 1 hour in Alexa Fluor-488-conjugated goat anti-rabbit IgG (H+L) antibody (Invitrogen-Life Sciences, Carlsbad, CA) diluted in protein blocking solution, rinsed with PBS, counterstained with Hoechst 33258 dye (bisbenzimide; Sigma-Aldrich), and mounted with Fluoro-Gel medium (Electron Microscopy Sciences, Hatfield, PA).For α-smooth muscle actin (α-SMA) IHC staining, cryosections were stained in a single-day procedure according to the manufacturer's instructions in a M.O.M. (Mouse-on-Mouse) basic immunodetection kit (BMK-2202; Vector Laboratories, Burlingame, CA) using mouse monoclonal anti-α-SMA primary antibody (no. M0851, clone 1A4; Dako, Carpinteria, CA). Sections were rinsed with PBS, incubated for 1 hour in M.O.M. biotinylated anti-mouse IgG (Vector Laboratories) diluted in protein blocking solution, rinsed with PBS, incubated in horseradish peroxidase-conjugated avidin (Vectastain Elite ABC kit, PK-6100; Vector Laboratories) for 30 minutes, washed with PBS, visualized with a 3,3′-diaminobenzidine peroxidase chromogenic substrate kit (SK-4100; Vector Laboratories), rinsed with water, counterstained with Gill no. 3 hematoxylin (Sigma-Aldrich), and mounted with Permount medium. Control specimens exposed to each secondary antibody alone showed no specific staining.
Photomicrographs were obtained using Optronics Engineering image analysis software version 1.0.0.4 (Bioscan, Seattle, WA) and Adobe Photoshop software version 5.0.2 installed in a PC microcomputer with a frame grabber and with a Sony 3-CCD camera mounted on a Nikon Eclipse E600 microscope. Scans of stained sections of newborn (postnatal day P1) mice were obtained using a Polaroid SprintScan 35 Plus microscope slide scanner (Meyer Instruments).
Cell Culture, Treatments, and Biochemical Analyses
Human primary SMC cultures obtained from bladder (HBSMC), colon (HCSMC), and bronchus (HBrSMC) were purchased from ScienCell (San Diego, CA) and propagated in smooth muscle cell medium (SMCM; ScienCell). Human primary vascular SMCs obtained from pulmonary artery (HPASMC), coronary artery (HCASMC), or umbilical artery (HUASMC) were purchased from Lonza (Walkersville, MD) and cultured in smooth muscle growth medium (SmGM-2; Lonza). Spontaneously immortalized porcine aortic endothelial (PAE) cells were provided by Dr. Lena Claesson-Welsh (Uppsala University, Uppsala, Sweden) and maintained in F12 Ham's medium (Invitrogen-Life Technologies) supplemented with 10% fetal bovine serum (FBS). PAE cells overexpressing NRP1 (PAE/NRP1) or NRP2 (PAE/NRP2) were used as controls and described previously.
5
, 26
Human bladder urothelial cells were propagated in ProstaLife medium (both from Lifeline Cell Technology, Frederick, MD), according to the manufacturer's guidelines.For protein analysis, primary SMCs were seeded in six-well plates at a density of 105 cells/well in complete growth medium. For comparison of visceral and vascular SMC protein content, lysates were harvested when cells reached approximately 80% to 90% confluency in complete growth medium. For analysis of protein content after growth factor treatment, cells were switched after 24 hours in complete growth medium to serum-depleted medium for 24 hours, at which time growth factors were added to a final concentration of 1 nmol/L for a further 24 hours. At the end of the incubation period, cells were rinsed twice with ice-cold PBS and harvested in 1× lysis buffer for protein analysis as described previously.
27
In selected experiments, membrane-enriched fractions were prepared. Briefly, cells were rinsed twice with ice-cold PBS, scraped in PBS and 2 mmol/L EDTA, and pelleted at approximately 250 × g at 4°C. Cell pellets were resuspended in buffer M (50 mmol/L HEPES pH 7.4, 10 mmol/L NaCl, 5 mmol/L MgCl2, and 0.1 mmol/L EDTA supplemented with protease inhibitors) and incubated for 10 minutes on ice, to promote cell swelling. Cells were disrupted mechanically using a Potter-Elvehjem homogenizer; the resulting homogenates were centrifuged at 16,000 × g for 10 minutes at 4°C, and pellets were resuspended in buffer M containing 1% Triton X-100 and incubated on ice for 30 minutes. Samples were centrifuged at 16,000 × g for 15 minutes at 4°C and supernatants were assayed for protein content. For comparison of protein content between various mouse organs, a wild-type C57BL/6 mouse (age P56) was euthanized, and organs were snap-frozen in liquid nitrogen. Frozen organs were ground to a powder, and protein lysates were isolated with radioimmunoprecipitation assay buffer (50 mmol/L Tris-HCl pH 7.5, 150 mmol/L NaCl, 1% NP40, 0.5% sodium deoxycholate, and 0.1% SDS; Boston Bioproducts, Ashland, MA) supplemented with Complete Mini protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). Protein concentrations were determined using a Bio-Rad (Hercules, CA) DC protein assay kit.RhoA activity assays were performed using a RhoA activation assay kit based on rhotekin pull-down, according to the manufacturer's instructions (Cytoskeleton, Denver, CO) and as described previously.
17
Briefly, 1.5 × 106 cells HBSMCs were plated in full growth medium and allowed to adhere overnight. Next, cells were incubated with 320 ng/mL SEMA3F for 0 to 30 minutes, then were washed with PBS and lysed with cell lysis buffer (25 mmol/L Tris-HCl pH 7.5, 150 mmol/L NaCl, 5 mmol/L MgCl2, and 1% Triton X-100). Whole-cell lysates were incubated with rhotekin-bound beads for 1.5 hours at 4°C and washed with low-salt buffer (25 mmol/L Tris-HCl pH 7.5, 40 mmol/L NaCl, 15 mmol/L MgCl2). Proteins were dissociated from beads by incubation in reducing sample buffer and were analyzed by Western blotting.Western Blotting
Proteins were diluted in reducing sample buffer and resolved on 7.5% polyacrylamide gels or 4% to 20% gradient polyacrylamide gels. Proteins were transferred to nitrocellulose membranes and blocked with 5% nonfat milk in Tris-buffered saline/Tween (50 mmol/L Tris-HCl pH 7.5, 150 mmol/L NaCl, and 0.05% Tween-20) for 1 hour. For mouse proteins, the membrane was incubated overnight at 4°C with rabbit monoclonal (D39A5) anti-mouse Nrp2 IgG antibody (no. 3366; Cell Signaling Technology), or mouse monoclonal (6C5) anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) IgG1 antibody (no. MAB374; Millipore, Billerica, MA). For human proteins, the membrane was incubated overnight at 4°C with one of the following antibodies: mouse monoclonal (C-9) anti-human NRP2 IgG2b antibody (no. SC-13117; Santa Cruz Biotechnology, Santa Cruz, CA), mouse monoclonal (A-12) anti-human NRP1 IgG1 antibody (SC-5307; Santa Cruz Biotechnology), rabbit anti-plexin A1 polyclonal antibody (3813; Cell Signaling Technology), rabbit monoclonal (67B9) anti-RhoA IgG antibody (2117; Cell Signaling Technology), mouse monoclonal (MY-21) anti-myosin light chain (MLC) IgM antibody (M4401; Sigma-Aldrich), rabbit anti-phospho-MLC (pSer19) polyclonal antibody (M6068; Sigma-Aldrich), mouse monoclonal (AC-15) anti-β-actin IgG1 antibody (A5441; Sigma-Aldrich), or anti-calnexin polyclonal antibody (Abcam, Cambridge, MA). Membranes were washed with Tris-buffered saline/Tween and incubated in the appropriate secondary antibody, either horseradish peroxidase-linked donkey anti-rabbit IgG whole antibody or horseradish peroxidase-conjugated sheep anti-mouse IgG whole antibody (both from GE Healthcare, Piscataway, NJ), for 1 hour. Membranes were washed, exposed to enhanced chemiluminescence (Pierce SuperSignal West Pico chemiluminescent substrate; Thermo Fisher Scientific, Rockford, IL) and signals were visualized after exposure of membranes to X-ray film.
RT-PCR
RNA was isolated from semiconfluent dishes of HBSMCs or human bladder urothelial cells using TRIzol reagent (Invitrogen-Life Technologies) or from SM and urothelial compartments acquired by laser capture microdissection from cryosections of bladders of C57BL/6 mice using Arcturus PicoPure RNA isolation reagent (Life Technologies). RNA was reverse-transcribed using a high-capacity cDNA synthesis kit (Applied Biosystems-Life Technologies, Foster City, CA), as described previously.
27
cDNAs were amplified using gene-specific primers for mouse and human neuropilin 2 and semaphorin 3F (SABiosciences-Qiagen, Valencia, CA), and relative mRNA levels were determined with normalization to GAPDH, as described previously.27
Purification of SEMA3 Proteins
SEMA3F protein was purified as described previously.
28
Briefly, HEK293 cells were transfected with human SEMA3F plasmid (provided by Dr. Marc Tessier-Lavigne, Rockefeller University) using FuGENE 6 transfection reagent (Roche Applied Science). After 16 hours, the medium was replaced with serum-free CD293 medium (Invitrogen-Life Technologies). Conditioned medium containing SEMA3F protein was collected after 48 hours, centrifuged, filtered, adjusted to pH 7.4, purified over a HiTrap high-performance nickel chelating column (GE Healthcare) using fast protein liquid chromatography, eluted with imidazole-containing buffer, and desalted using a PD-10 gel filtration column (GE Healthcare). The resulting proteins were diluted in PBS (pH 7.2), and protein concentrations were established using a Bio-Rad DC protein assay.Cell Collapse Assay
The SEMA3-induced cytoskeletal collapse assay was modified from a previously reported assay
28
for use with HBSMCs or HCSMCs. Briefly, acid-washed glass coverslips (18 × 18 mm) were placed into six-well dishes. HBSMCs or HCSMCs were plated onto the coverslips in full growth medium. After 16 hours, purified SEMA3 proteins (320 ng/mL) were added to half of the wells and were incubated at 37°C. After 0 to 60 minutes, cells were fixed in 4% paraformaldehyde in PBS for 10 minutes, washed with PBS, permeabilized with 0.2% Triton X-100 (Sigma-Aldrich) in PBS for 5 minutes, washed with PBS, stained with Alexa Fluor 488 phalloidin (1:250 dilution; Invitrogen-Life Technologies) for 1 hour, washed with PBS, counterstained with Hoechst 33258 dye, and mounted with Fluoro-Gel medium (Electron Microscopy Sciences). The actin cytoskeleton of the SMCs was visualized and imaged using a Leica TCS SP2 AOBS confocal laser scanning system attached to a Leica DM IRE2 inverted microscope (Leica Microsystems, Wetzlar, Germany) and equipped with a 63× oil objective (NA 1.4), a 1.6× optivar, a 488-nm argon ion laser (for F-actin), and a 405-nm diode (for nuclei). Differential interference contrast images were also obtained. Leica confocal software LCS and ImageJ software (NIH, Bethesda, MD) were used to scale recorded images. In selected experiments, cells were pretreated with 10 µmol/L Rho kinase inhibitor Y27632 (EMD Chemicals, Darmstadt, Germany), 1 µg/mL RhoA inhibitor C3 transferase (Cytoskeleton Inc.), or 1 nmol/L HB-EGF.Optical Magnetic Twisting Cytometry
Primary HBSMCs were seeded in 96-well plates at a density of 10,000 cells/well; when confluent, cells were switched to serum-depleted medium 24 hours before analysis. On the day of measurement, cells were incubated with Arg-Gly-Asp (R-G-D)-coated beads (4.5 μm in diameter) for 20 minutes, to allow bead binding to integrins on the cell surface. Vehicle, SEMA3F (0.2 or 0.6 μg/mL), or Y27632 (10 μM) was then added at the indicated concentrations for an additional 40 minutes, after which plates were mounted on a microscope stage custom fitted for twisting cytometry.
29
, 30
, 31
Briefly, beads were magnetized horizontally and then twisted vertically in an oscillatory magnetic field with a frequency of 0.75 Hz. From the ratio of the applied mechanical torque to the measured lateral bead displacement, we calculated the cell elastic modulus G′, a measure of cell stiffness. For each experimental condition, data were pooled from thousands of beads attached to hundreds of cells from at least eight separate wells. Data are presented as the median G′ in pascals per nanometer ± SEM.Assessment of Bladder Function by Cystometry
To analyze bladder function in mice with constitutive knockout of Nrp2, we performed conscious cystometry, essentially as described previously.
32
Briefly, Nrp2-null mice and wild-type littermate controls were anesthetized with isoflurane and the bladder was exposed with a lower midline abdominal incision. A polyethylene catheter (PE10) was placed in the dome of the bladder and secured; the free end of the catheter was tunneled subcutaneously to the back of the mouse, where it was exteriorized. The abdominal incision was closed in two layers, and animals were allowed to recover. The end of the catheter was connected via a stopcock to an infusion pump (Harvard Apparatus, Holliston, MA) and physiological pressure transducer, and the bladder was filled with normal saline solution at a rate of 12.5 μL/minute. After an equilibration period of 20 to 30 minutes, intravesical pressure was monitored over multiple cycles of filling and voiding during a 60-minute time frame. From these data, parameters such as peak voiding pressure, intercontraction interval, and bladder compliance were determined.Contractility Testing
Functional analysis of muscle strips was performed essentially as described previously.
33
Briefly, bladders from female Nrp2 wild-type and null mice were harvested with minimal handling into ice-cold Krebs solution (120 mmol/L NaCl, 5.9 mmol/L KCl, 25 mmol/L NaHCO3, 1.2 mmol/L Na2H2PO4, 1.2 mmol/L MgCl2 · 6H2O, 2.5 mmol/L CaCl2, and 11.5 mmol/L dextrose). The bladders were incised longitudinally, and the mucosa was removed by dissection under a stereomicroscope. Tissue strips were attached to a force transducer (Grass Technologies, Quincy, MA) and were suspended in an organ bath maintained at 37°C and bubbled with a mixture of 95% O2 and 5% CO2. Bladder tissue was stretched to a force of 0.5 × g and equilibrated for 45 minutes. Contractile responses to 1 nmol/L to 10 µmol/L of the cholinergic agonist carbachol, 120 mmol/L KCl, and 10 µmol/L α-β-methyl-ATP and to electrical field stimulation (1 to 64 Hz, 40 V, 0.5-ms pulse width, 10-second duration) were measured in separate strips. Data were calculated as force (millinewtons) normalized by tissue cross-sectional area and are expressed as means ± SEM.Statistical Analysis
Differences in contractile responses between Nrp2 wild-type and Nrp2gfp/gfp mice were determined by Student's t-test. P < 0.05 was considered significant. For analysis of cystometric data, a generalized estimating equation (GEE) model was used to analyze the differences between Nrp2gfp/gfp mice and Nrp2 wild-type mice and also to handle correlation of data obtained from the same mouse. Four voids were obtained per mouse, for a total of 16 voids in Nrp2gfp/gfp mice and 20 voids in Nrp2 wild-type mice.
Results
Neuropilin 2 Is Expressed in Visceral Smooth Muscle in Vivo
To determine sites of expression of Nrp2 in vivo, we used a genetic mouse model in which the gene encoding bacterial β-galactosidase (LacZ) replaces exon 1 and the proximal part of intron 1 of the Nrp2 gene.
15
Neonatal (P1) Nrp2+/LacZ mice were euthanized and frozen in optimal cutting temperature compound. Cryosections were stained with X-gal reagent to reveal robust staining of the bladder and gastrointestinal tract (Figure 1A). Localization of Nrp2 signal to SM (Figure 1, B and D) was verified by coincident immunostaining of serial sections with an antibody to α-SMA (Figure 1, C and E). This staining pattern was reminiscent of that reported by Chen et al,- Takashima S.
- Kitakaze M.
- Asakura M.
- Asanuma H.
- Sanada S.
- Tashiro F.
- Niwa H.
- Miyazaki Ji J.
- Hirota S.
- Kitamura Y.
- Kitsukawa T.
- Fujisawa H.
- Klagsbrun M.
- Hori M.
Targeting of both mouse neuropilin-1 and neuropilin-2 genes severely impairs developmental yolk sac and embryonic angiogenesis.
Proc Natl Acad Sci USA. 2002; 99: 3657-3662
4
who noted high Nrp2 mRNA levels in the presumptive bladder and gut during development. Consistent with previous reports,11
, 12
, 13
X-gal signal corresponding to Nrp2 was also detected in the brain (Figure 1A). Expression of neuropilins is regulated developmentally and is known to be down-regulated after birth in vessels14
and neurons.34
To determine the postnatal expression pattern of Nrp2, tissues from neonate and 4- and 8-week old Nrp2+/LacZ mice were stained with X-gal reagent, which revealed intense staining of bladder and gut SM at P1, 4 weeks, and 8 weeks (Figure 1, A, B, and D; see also Supplemental Figure S1, A–C, at http://ajp.amjpathol.org). This observation suggests that regulation of Nrp2 in SM differs from that in vessels and neurons and highlights the robust expression of Nrp2 in this tissue type. We confirmed expression of Nrp2 protein in the bladder wall and gut by IHC staining and demonstrated localization to SM, as described above (Figure 1F; see also Supplemental Figure S1D at http://ajp.amjpathol.org); importantly, no protein was detected in homozygous null Nrp2LacZ/LacZ mice (Figure 1G).
Figure 1Neuropilin 2 is expressed in bladder and intestine. A: Cryosection of a newborn (P1) Nrp2+/LacZ mouse stained with X-gal reagent (blue) and counterstained with eosin (pink) shows strong Nrp2 expression in the brain and visceral organs. The bladder is marked by an asterisk. B–E: Cryosections of an 8-week-old Nrp2+/LacZ mouse from bladder (B and C) or intestine (D and E) stained with X-gal reagent [blue (B and D)] or anti-SMA [brown (C and E)] and counterstained with eosin [pink (B and D) or hematoxylin [blue (C and E)]. In bladder and intestine, Nrp2 expression was strongest in the muscle layer, but was also found in endothelial cells (arrows). F and G:Neuropilin 2 localization was assessed in cryosections of bladder from an 8-week-old heterozygous Nrp2+/LacZ mouse (F) and knockout Nrp2LacZ/LacZ mouse (G) using anti-Nrp2/Hoechst 33258 (green/blue) staining. H: Western blot of whole-cell lysates from organs of wild-type C57BL/6 mouse. Of the organs examined (liver, lung, bladder, kidney, skin, heart, brain, and colon), the bladder expressed the highest relative amount of Nrp2 and sNrp2 protein. Brain served as a positive control for Nrp2, and GAPDH was used as a loading control. I–K: X-gal staining [blue (I and J)] or anti-SMA staining [brown (K)] of cryosections of 8-week-old Nrp2+/LacZ bladder (I) and a large artery and vein in the limb (J and K). Nrp2 is expressed in visceral SM but not in vascular SM, and in venous but not arterial endothelial cells. Scale bars: 1 mm (A); 0.1 mm (B–G); 0.1 mm (I–K).
Nrp2 expression in different organs was also assessed by immunoblotting. In agreement with findings from X-gal and Nrp2 immunostaining, this analysis revealed the bladder and colon to display the highest levels of full-length Nrp2, compared with expression in other tissues (Figure 1H). To identify the predominant site of expression of Nrp2 within the bladder wall, we performed laser capture microdissection coupled with real-time RT-PCR of bladder tissue from three different strains of mice. We observed significant enrichment of Nrp2 mRNA in SM, relative to urothelium, whereas the Nrp2-selective ligand Sema3F was enriched in urothelium, relative to SM (see Supplemental Figure S1E at http://ajp.amjpathol.org).
Nrp2 was not detected in cardiac or skeletal muscle (Figure 1A; see also Supplemental Figure S1, F–G at http://ajp.amjpathol.org), although it was present in α-SMA-positive arrector pili muscle (see Supplemental Figure S1H at http://ajp.amjpathol.org). Notably, Nrp2 was absent from vascular SM (Figure 1J); however, Nrp2 was expressed in endothelial cells of capillaries and veins (Figure 1J). Together, these data identify visceral SM as the predominant site of Nrp2 expression in vivo.
Neuropilin 2 Is Expressed by Primary Visceral SMCs in Vitro and Mediates Cytoskeletal Rearrangement
To determine whether the expression pattern of Nrp2 observed in mice was recapitulated in humans, we interrogated a series of primary human SMCs isolated from different sites, including bladder, colon, airways, and both low- and high-resistance arteries. NRP2 expression in vitro was evident in SMCs from bladder and colon, but was weak in both pulmonary and coronary artery SMCs (Figure 2A), which is in agreement with our in vivo observations (Figure 1, I versus J). Next, we determined the expression level of plexins, a family of NRP2 coreceptors, in human bladder and colon SMCs in vitro. Using RT-PCR, we observed expression of plexin A1, but not plexin A2 or A4, in bladder SMCs (data not shown). Plexin A1 and NRP2 protein expression were verified in membrane preparations of bladder and colonic SMCs (Figure 2B). We also examined expression of the NRP2 ligand SEMA3F and found that, in contrast to NRP2 (which is enriched in SMCs), SEMA3F was enriched in urothelial cells (Figure 2C), consistent with our observations in mouse bladder tissue (see Supplemental Figure S1E at http://ajp.amjpathol.org).
Figure 2NRP2 is functional in human SMCs. A: Immunoblot of whole-cell lysates from primary cultures of human SMCs isolated from different organs. Visceral SMCs from bladder and colon expressed relatively higher levels of NRP2 than vascular SMCs from pulmonary or coronary artery. PAE/NRP2 and PAE/NRP1 served as a positive and negative control, respectively; actin was used as a loading control. B: Human bladder and colon SMCs have functional SEMA3F receptors. Immunoblot of membrane fractions of HBSMCs and HCSMCs express the SEMA3F receptor NRP2 and the coreceptor plexin A1. Calnexin protein expression was used as a loading control for membrane protein. C: Relative expression of NRP2 and SEMA3F in HBSMCs or human bladder urothelial cells (HBUC) was determined by semiquantitative RT-PCR. NRP2 and SEMA3F mRNA levels in human bladder urothelial cells were set to a value of 1. D–G: HBSMCs undergo cytoskeletal rearrangement after treatment with SEMA3F (600 ng/mL). Phalloidin staining (green) shows F-actin stress fibers (D and E) and corresponding differential interference contrast images illustrate thinning of the cells (F and G). Nuclei were stained with Hoechst 33258 dye (blue). Original magnification, ×100. HxSMC indicates human SMCs from bladder (B), colon (C), umbilical artery (UA), pulmonary artery (PA), bronchus (Br), and coronary artery (CA). PAE, porcine aortic endothelial cells.
Previous observations from our research group demonstrated profound morphological changes in tumor cells or endothelial cells after challenge with SEMA3F.
17
To assess whether SMCs undergo similar changes, cells seeded on glass coverslips were exposed to SEMA3F and cell morphology was assessed by staining with the actin-binding agent phalloidin. SEMA3F elicited a marked cytoskeletal rearrangement in SMCs, characterized by thinning of cells (Figure 2, E and G), compared with vehicle-treated controls (Figure 2, D and F). These observations indicate that NRP2 in human SMCs is functional.Signals regulating NRP2 expression are largely undefined. To investigate NRP2 regulation in SMCs, we exposed cells to agents that are known to regulate SM phenotype. These include the SMC mitogens heparin-binding EGF-like growth factor (HB-EGF)
35
and platelet-derived growth factor (PDGF),36
as well as the prodifferentiation factor transforming growth factor β-1 (TGF-β-1).37
HB-EGF and PDGF markedly down-regulated NRP2 protein levels in bladder and colonic SMCs, but up-regulated expression of NRP1 (Figure 3A). In each case, PDGF caused greater NRP2 down-regulation than HB-EGF, likely reflecting its higher activity on SMCs.38
Conversely, TGF-β-1 stimulated NRP2 expression in SMCs, but decreased NRP1 levels, thus demonstrating differential regulation of neuropilins in SMCs.
Figure 3HB-EGF down-regulates NRP2 and inhibits the effects of SEMA3F. A: HBSMCs and HCSMCs were treated with various growth factors, and whole-cell lysates were examined by immunoblotting for NRP1 and NRP2. PDGF and HB-EGF increased NRP1 but decreased NRP2. TGF-β-1 decreased NRP1 but slightly increased NRP2. B–E: HBSMCs pretreated with vehicle (B and C) or HB-EGF (D and E) were exposed to SEMA3F (C and E) or not (B and D). Phalloidin staining (green) shows F-actin stress fibers; nuclei were visualized with Hoechst 33258 dye (blue). Original magnification, ×100.
To assess the functional consequence of growth factor-mediated down-regulation of NRP2, we pretreated SMCs with PDGF or HB-EGF and determined the effect of SEMA3F on cell shape. Although HB-EGF alone did not alter SMC morphology (Figure 3, B versus D), HB-EGF pretreatment prevented SEMA3F-induced cytoskeletal rearrangement (Figure 3, C versus E). The effect of PDGF on SEMA3F-induced changes could not be assessed in this experiment, because PDGF pretreatment elicited significant changes in cell shape, which masked the effect of SEMA3F. These findings indicate that stimuli known to regulate SMC behavior modulate NRP function.
Neuropilin 2 Regulates SMC Contractility via RhoA/ROCK
Next, we interrogated signals downstream of the SEMA3F-NRP2 axis that mediate alterations in SMC morphology. The cytoskeletal rearrangement observed after SEMA3F treatment (Figure 4, A and B) was phenocopied by the addition of either the RhoA inhibitor C3 transferase (Figure 4C) or the ROCK inhibitor Y27632 (Figure 4D). Consistent with these observations, SEMA3F treatment led to inhibition of RhoA activity (Figure 4E), as well as a reduction in myosin light chain (MLC) phosphorylation (Figure 4F), corresponding to decreased MLC activity. To determine whether the cytoskeletal changes induced by SEMA3F were associated with altered cytoskeletal stiffness, we performed optical magnetic twisting cytometry.
29
, 30
Exposure of human SMCs to SEMA3F led to a dose-dependent reduction in the cell elastic modulus G′, to an extent similar to that obtained with the ROCK inhibitor Y27632 (Figure 4G). These observations indicate that, in response to SEMA3F binding, NRP2 transduces signals that alter intracellular tension in SMCs in vitro through the Rho-ROCK axis.
Figure 4SEMA3F-induced changes in cell morphology are mediated via RhoA, ROCK and MLC. A–D: Collapse assays using HBSMCs treated with vehicle (A), SEMA3F (B), RhoA inhibitor (C3 transferase) (C), or ROCK inhibitor (Y27632) (D). Phalloidin staining (green) shows F-actin stress fibers. Compounds inhibiting RhoA or ROCK evoked similar changes in cell shape as SEMA3F. Original magnification, ×100. E: SEMA3F inhibits RhoA activity in HBSMC. F: SEMA3F treatment leads to decreased phosphorylation of myosin light chain (MLC) in human bladder (HBSMC) and colonic (HCSMC) SMCs. Cells were exposed to vehicle or SEMA3F (600 ng/mL) for 30 minutes before immunoblotting with antibodies. G: SEMA3F treatment decreases the elastic modulus G′ of HBSMCs in a dose-dependent manner. Cells treated with 10 μmol/L Y27632 (Y27) were included as a positive control. Data are representative of at least two independent trials.
Neuropilin 2 Regulates Smooth Muscle Contractility in Vivo
In view of the visceral SM-enriched expression pattern for Nrp2 in mice and its ability to transduce prorelaxant signals in vitro, we next explored its potential function in vivo. Because the bladder wall harbors the highest levels of Nrp2 expression in vivo, we chose to analyze bladder function as a representative system. For these analyses, we used two complementary experimental approaches: cystometry (for quantitative assessment of parameters such as voiding pressure and presence/frequency of nonvoiding contractions in intact, conscious mice) and ex vivo contractility testing of bladder muscle strips (to measure force generation by tissue strips suspended under isometric tension and exposed to contractile agonists). Initially, we analyzed mice with constitutive knockout of Nrp2 (Nrp2gfp/gfp) and compared them with age- and sex-matched wild-type littermate controls (Nrp2+/+) (see Supplemental Figure S2A at http://ajp.amjpathol.org). For cystometry, PBS was infused at a defined rate (12.5 μL/minute) into the bladders of unrestrained wild-type and Nrp2gfp/gfp mice via a catheter placed in the bladder 3 days before measurement. Bladder pressures were determined over a period of approximately 60 minutes. Overall, the voiding patterns of Nrp2-deficient mice did not differ significantly from wild-type animals (Figure 5, A and B). Notably, however, Nrp2gfp/gfp mice showed a greater increase in intravesical (ie, bladder) pressure during the filling phase of the cystometrogram (Figure 5B), compared with wild-type mice (P = 0.01) (Figure 5A). The estimated mean compliance for Nrp2gfp/gfp mice was 6.34 ± 1.86, whereas for wild-type mice it was 12.69 ± 1.67; the difference was statistically significant (P = 0.01). The reduced compliance in Nrp2gfp/gfp mice suggested an abnormality in bladder wall elasticity and/or contractility. Heterozygous Nrp2+/gfp mice also displayed an increased pressure change during filling, compared with wild-type mice, but this did not reach statistical significance (P = 0.15) (data not shown).
Figure 5Constitutive deletion of neuropilin 2 enhances SM contractility in vivo and ex vivo. A and B: Representative cystometrograms showing three voiding cycles of conscious unrestrained mice in which Nrp2 is intact (wild-type; Nrp2+/+) (A) or deleted (Nrp2gfp/gfp) (B). Data analysis revealed decreased compliance in Nrp2-deficient animals (P = 0.01). C–F: Contractile responses of bladder muscle strips from Nrp2-expressing and Nrp2-deficient mice were determined in response to electrical field stimulation (C), carbachol (D), the ATP analog α,β-me-ATP (E), and KCl (F). In each case, force generation was higher in Nrp2gfp/gfp mice. *P < 0.05 (C and D); *P = 0.025 (E); *P = 0.031 (F). n = 5 mice per group (A); n = 4 mice per group (B); n = 6 mice per group (C–F).
We proceeded to investigate the contractile responses of bladder tissue strips from wild-type or Nrp2gfp/gfp mice to electric field stimulation or to exogenous soluble agonists. Tissue strips from Nrp2gfp/gfp mice showed a significantly higher contractile response to electric field stimulation at all frequencies tested, compared with those from wild-type mice (P < 0.05) (Figure 5C). Because the bladder SM response to electric field stimulation consists of both cholinergic and purinergic components, we investigated the responses to carbachol and α-β-methylene-ATP. The dose response to carbachol was also significantly higher in Nrp2gfp/gfp mice at concentrations >30 nmol/L (P < 0.05) (Figure 5D); however, the EC50 was similar between groups (0.48 ± 0.09 μmol/L versus 0.46 ± 0.07 μmol/L), suggesting no alteration in receptor density or sensitivity to agonist with Nrp2 deficiency. Purinergic stimulation also induced a greater contraction from Nrp2gfp/gfp mice, compared with wild-type mice (P = 0.025) (Figure 5E). Similarly, in bladder strips from Nrp2gfp/gfp mice, contractions generated by direct stimulation of SM with extracellular KCl were substantially greater than in tissue from wild-type mice (P = 0.031) (Figure 5F). These observations suggest that neuropilin 2 regulates SM contractility and that its loss enhances bladder tissue contractility ex vivo.
To extend this analysis, we used Cre-LoxP technology to delete Nrp2 specifically in SM, using the tamoxifen-inducible SM22α-CreERT2 strain.
24
At selected time points after the last injection of 4-OHT, mice were euthanized and bladders were harvested for isometric tension testing. Loss of Nrp2 expression in SM was highly efficient, with little or no protein evident in SM tissue from targeted animals, as determined by immunoblot analysis of tissues (see Supplemental Figure S2B at http://ajp.amjpathol.org) and by IHC staining (see Supplemental Figure S2, C versus D and E versus F, at http://ajp.amjpathol.org). Contractility testing showed that bladder muscle strips from mice with SM-specific Nrp2 deletion (SM-Cre:Nrp2fl/fl) displayed enhanced force generation in response to electric field stimulation, carbachol, α-β,methylene-ATP, or KCl, compared with tissue from nondeleted controls (Nrp2fl/fl) (P < 0.05) (Figure 6). These findings are consistent with the increased contractility of bladder specimens from mice with constitutive Nrp2 knockout (Figure 5) and support the conclusion that Nrp2 regulates visceral SM contractility.
Figure 6SM-specific deletion of neuropilin 2 enhances SM contractility ex vivo. Contractile responses of SM strips from bladders of SM22α-Cre:Nrp2fl/fl mice (black symbols or bars) or Nrp2fl/fl mice (gray symbols or bars), treated with 4-OHT, were determined in response to electrical field stimulation (A), carbachol (B), the ATP analog α,β-me-ATP (C), and KCl (D). In each case, force generation was greater in SM22α-Cre:Nrp2fl/fl mice. *P < 0.05 (A and B); *P = 0.033 (C); *P = 0.008 (D). n = 10 mice per group, SM22α-Cre:Nrp2fl/fl; n = 5 mice per group, Nrp2fl/fl.
Discussion
With this study, we provide the first demonstration of a role for Nrp2 in regulation of SM contractility. Although previous studies have reported expression of neuropilins in isolated vascular SMCs and have demonstrated expression of Nrp2 mRNA in the developing bladder,
4
none described a role for these proteins in SM in vivo. Using a series of complementary mouse models, including those with constitutive or tissue-specific deficiency in Nrp2 expression, we show that the SM compartments of the bladder and gastrointestinal tract are among the sites with highest expression of Nrp2 in vivo. Loss of expression of Nrp2 in the bladder wall, either constitutively or in SM, led to profound functional changes, marked by enhanced force generation in response to a range of contractile agonists. In addition, we observed a significant effect on the gastrointestinal tract, characterized by alterations in intestinal transit and evidence of obstruction (data not shown).Neuropilins are expressed in multiple cell types, including neurons, vascular cells, dendritic cells, epithelial cells, and SMCs. Several recent studies have described expression of both Nrp1 and Nrp2 in bladder epithelium and submucosa in vivo and have demonstrated changes in their expression in response to inflammatory stimuli.
39
, - Saban M.R.
- Backer J.M.
- Backer M.V.
- Maier J.
- Fowler B.
- Davis C.A.
- Simpson C.
- Wu X.R.
- Birder L.
- Freeman M.R.
- Soker S.
- Hurst R.E.
- Saban R.
VEGF receptors and neuropilins are expressed in the urothelial and neuronal cells in normal mouse urinary bladder and are upregulated in inflammation.
Am J Physiol Renal Physiol. 2008; 295: F60-F72
40
, 41
, 42
Our analysis of Nrp2 expression in vivo using the Nrp2+/LacZ and Nrp2+/gfp reporter mice, together with immunofluorescence imaging of wild-type mouse tissues, revealed robust Nrp2 expression in SM but failed to detect significant Nrp2 expression in either bladder or gut epithelia. Expression of neuropilins in many cell types is typically down-regulated after birth, once the axonal guidance and vascular development functions have been fulfilled.- Saban M.R.
- Sferra T.J.
- Davis C.A.
- Simpson C.
- Allen A.
- Maier J.
- Fowler B.
- Knowlton N.
- Birder L.
- Wu X.R.
- Saban R.
Neuropilin-VEGF signaling pathway acts as a key modulator of vascular, lymphatic, and inflammatory cell responses of the bladder to intravesical BCG treatment.
Am J Physiol Renal Physiol. 2010; 299: F1245-F1256
14
, 34
However, we observed sustained expression of Nrp2 in visceral SM in the postnatal period, consistent with a function for Nrp2 in this location that is distinct from its neurovascular roles.The function of hollow organs such as the bladder and gastrointestinal tract is to store waste products at low pressure and expel them under volitional control. Serious consequences can result from sustained increases in intraluminal pressure, especially in the case of the bladder, with initial hypertrophy of the bladder wall leading to fibrosis, transmission of elevated pressure to the kidneys and ensuing renal damage. Our initial findings with tissues from mice with constitutive knockout of Nrp2 revealed increased contractility of bladder muscle strips exposed to a range of contractile agonists, as well as alterations in voiding behavior in conscious mice. In particular, analysis of voiding function in conscious mice indicated a reduction in bladder compliance in Nrp2-deficient animals, compared with controls. Decreased compliance in hollow organs can result from remodeling and thickening of SM in response to elevated pressures, as occurs secondary to bladder outlet obstruction, hypertension, or asthma, and is characterized by aberrant deposition and turnover of the extracellular matrix.
43
, 44
, 45
Although we cannot exclude the possibility, our observation that neither peak voiding pressures nor bladder weight in Nrp2-null mice differed significantly from wild-type controls argues against the possibility that the reduced compliance observed in Nrp2-null mice results from bladder outlet obstruction.Because Nrp2 is expressed in multiple cell types of relevance to bladder wall contractility, including SMCs and neurons, initially it was not possible to ascribe the enhanced force generation in tissues lacking Nrp2 to a specific cell of origin. To address the contribution of SM-expressed Nrp2, we deleted the gene selectively in this tissue type, using SM22α to drive Cre recombinase. The demonstration that Nrp2 deficiency specifically in SM also led to enhanced contractility in muscle strips strongly supports a myogenic, as opposed to neurogenic, mechanism for increased force generation. Our analysis of laser capture-microdissected SM and urothelial compartments from mouse bladders, as well as analysis of human bladder SMCs and urothelial cells in culture, revealed robust expression of SEMA3F in epithelial cells, consistent with paracrine activation of NRP2 expressed on SMCs. It is important to note, however, that the tension testing analysis was conducted in the absence of exogenous SEMA3F or VEGF.
The observation that tissue strips from Nrp2-null mice showed enhanced responses to contractile agonists suggests that Nrp2 may also interact functionally with canonical regulators of contractility, such as muscarinic and purinergic receptors and/or their effectors, independently of ligand. In support of this possibility, the cytoplasmic domain of both Nrp1 and Nrp2 harbors the amino acids Ser-Glu-Ala (S-E-A) at the C-terminus. This S-E-A motif possesses PDZ domain binding activity and has been shown to mediate interaction with the protein, neuropilin-interacting protein (NIP), also known as synectin or GIPC.
46
, 47
GIPC in turn interacts with a number of proteins associated with the contractile apparatus, including transgelin/SM22α, which is highly enriched in SM.48
Notably, Nrp1, acting through GIPC, was found to elicit effects on endothelial cell spreading by modulating the internalization and adhesion function of α5β1 integrin.- Ewing R.M.
- Chu P.
- Elisma F.
- Li H.
- Taylor P.
- Climie S.
- McBroom-Cerajewski L.
- Robinson M.D.
- O'Connor L.
- Li M.
- Taylor R.
- Dharsee M.
- Ho Y.
- Heilbut A.
- Moore L.
- Zhang S.
- Ornatsky O.
- Bukhman Y.V.
- Ethier M.
- Sheng Y.
- Vasilescu J.
- Abu-Farha M.
- Lambert J.P.
- Duewel H.S.
- Stewart I.I.
- Kuehl B.
- Hogue K.
- Colwill K.
- Gladwish K.
- Muskat B.
- Kinach R.
- Adams S.L.
- Moran M.F.
- Morin G.B.
- Topaloglou T.
- Figeys D.
Large-scale mapping of human protein-protein interactions by mass spectrometry.
Mol Syst Biol. 2007; 3: 89
49
Importantly, the ability of Nrp1 to affect α5β1 integrin occurred independently of interactions with SEMA3 ligand and of its function as a VEGF coreceptor. Taken together, these findings suggest that Nrp2 may regulate SM contractility through both ligand-dependent and ligand-independent processes.Activation of cholinergic and purinergic receptors, as well as exposure of tissues to electrical field stimulation or depolarizing agents such as KCl, can evoke a range of different intracellular signals. All of them, however, are known to activate the RhoA-ROCK axis to promote SM contraction.
50
, 51
Stimuli transduced by neuropilin 2 also impinge on the RhoA-ROCK cascade. Data from primary culture SMCs (Figure 3), as well as previous work from our research group,17
showed that SEMA3F-mediated signaling via NRP2 inhibited RhoA activity. In SMCs, this led to decreased phosphorylation of MLC and a reduction in cytoskeletal stiffness as measured by optical magnetic twisting cytometry, consistent with stimulation of relaxation. In contrast, inhibition of SEMA3F function in vivo by genetic ablation of Nrp2 resulted in loss of relaxation (ie, increased contractility), compared with tissues from controls with intact Nrp2.Together, these observations are consistent with a model in which Nrp2, through binding of SEMA3F, maintains appropriate SM tone, and loss of Nrp2 enhances agonist-induced SM contractility by relieving inhibition on RhoA/ROCK signaling.
In summary, we provide the first evidence to support a role for Nrp2 in regulating contractility of visceral SM. The ability of Nrp2 to promote SM relaxation, along with its robust expression in hollow organs such as the bladder and intestine, implies that this axis could be exploited for therapeutic benefit in conditions characterized by inappropriate SM contractility. These include bladder and bowel hyperreflexia secondary to spinal cord injury, as well as gastrointestinal tract motility disorders such as intestinal pseudo-obstruction and paralytic ileus. The existence of a natural ligand for Nrp2, as well as pharmacological agents that target components of the Nrp2 signaling axis, suggests the potential for relatively rapid translation of this novel finding to the clinical setting.
Acknowledgments
We thank Ricardo Sanchez for histological sections and consultations; Michelle Mulone for assistance with genotyping, tissue processing, and voiding analyses; Kristin Johnson for graphic design and figures; Marc Tessier-Lavigne (Rockefeller University) for the SEMA3F construct; and Dr. Michael Freeman and members of the Urological Diseases Research Center at Children's Hospital Boston for helpful discussions.
Supplementary data
- Supplemental Figure S1
Neuropilin 2 expression in bladder muscle does not diminish with age. A–C: Cryosections of bladders from Nrp2+/LacZ mice at age P1 (A), P28 (B), and P56 (C) were stained with X-gal reagent (blue) and counterstained with eosin (pink). D: Neuropilin 2 protein expression was verified in a cryosection of colon from an 8-week-old heterozygous Nrp2+/LacZ mouse, using anti-Nrp2/Hoechst 33258 dye (green/blue) staining. E: Semiquantitative RT-PCR comparing Nrp2 and Sema3F mRNA levels from urothelium (Uro) and SM tissues, separated by laser capture microdissection, of mouse bladders from female and male C57BL/6 mice. The SM compartment consistently expressed higher levels of Nrp2 than the Uro compartment, whereas Sema3F expression was enriched in the Uro compartment, relative to the SM compartment. In each case, mRNA levels were normalized to GAPDH, and the level in urothelium was set to a value of 1. F–K: Cryosections of tissue from 8-week-old Nrp2+/LacZ mice were stained with X-gal reagent (blue) and counterstained with eosin (pink) (F–H). Serial cryosections were stained with anti-α-SMA to detect SM (brown) and counterstained with hematoxylin (blue) (I–K). Nrp2 is absent from cardiac muscle (F) and skeletal muscle (G), but is expressed in vein and capillary endothelial cells in each tissue type. Nrp2 is also expressed in α-SMA–positive arrector pili muscle (H, arrows). Scale bars: 0.2 mm (A–C); 0.1 mm (D, F–K).
- Supplemental Figure S2
A: Dose-dependent reduction in neuropilin 2 protein expression in bladder, lung, and brain of wild-type (WT), heterozygous [Nrp2+/gfp (HET)], and homozygous null [Nrp2gfp/gfp (KO)] mice, as determined by Western blotting. B: Nrp2 protein is efficiently deleted in bladder tissue from 4-OHT-treated SM22α-Cre:Nrp2fl/fl mice (lanes 5–12), whereas protein remains intact in 4-OHT-treated Nrp2fl/fl mice (lanes 1–4). C–F: IHC staining of cryosections from Nrp2fl/fl (C and E) or SM22α-Cre:Nrp2fl/fl mice (D and F) for Nrp2 (brown color), after 4-OHT treatment. Sections were counterstained with hematoxylin (blue color). Mice harboring deletion of Nrp2 showed little or no staining for Nrp2 in SM. Nrp2 expression is retained in capillaries (E, arrows) and neurovascular bundles (D and F, arrows) after 4-OHT treatment. Epithelium (e) lacks Nrp2. Scale bar = 0.1 mm.
References
- Neuropilins as Semaphorin receptors: in vivo functions in neuronal cell migration and axon guidance.Adv Exp Med Biol. 2002; 515: 13-31
- Targeting endothelial and tumor cells with semaphorins.Cancer Metastasis Rev. 2007; 26: 421-443
- Neuropilin is a receptor for the axonal chemorepellent Semaphorin III.Cell. 1997; 90: 739-751
- Neuropilin-2, a novel member of the neuropilin family, is a high affinity receptor for the semaphorins Sema E and Sema IV but not Sema III [Erratum appeared in Neuron 1997, 19:559].Neuron. 1997; 19: 547-559
- Semaphorin 3F, a chemorepulsant for endothelial cells, induces a poorly vascularized, encapsulated, nonmetastatic tumor phenotype.J Clin Invest. 2004; 114: 1260-1271
- Successful inhibition of tumor development by specific class-3 semaphorins is associated with expression of appropriate semaphorin receptors by tumor cells.PLoS One. 2008; 3: e3287
- Overexpression of a membrane protein, neuropilin, in chimeric mice causes anomalies in the cardiovascular system, nervous system and limbs.Development. 1995; 121: 4309-4318
- A requirement for neuropilin-1 in embryonic vessel formation.Development. 1999; 126: 4895-4902
- Neuropilin-1 is required for vascular development and is a mediator of VEGF-dependent angiogenesis in zebrafish.Proc Natl Acad Sci USA. 2002; 99: 10470-10475
- Functional knockdown of neuropilin-1 in the developing chick nervous system by siRNA hairpins phenocopies genetic ablation in the mouse.Dev Dyn. 2004; 230: 299-308
- Neuropilin-2 is required in vivo for selective axon guidance responses to secreted semaphorins.Neuron. 2000; 25: 29-41
- Neuropilin-2 regulates the development of selective cranial and sensory nerves and hippocampal mossy fiber projections.Neuron. 2000; 25: 43-56
- Aberrant sensory innervation of the olfactory bulb in neuropilin-2 mutant mice.J Neurosci. 2002; 22: 4025-4035
- Abnormal lymphatic vessel development in neuropilin 2 mutant mice.Development. 2002; 129: 4797-4806
- Targeting of both mouse neuropilin-1 and neuropilin-2 genes severely impairs developmental yolk sac and embryonic angiogenesis.Proc Natl Acad Sci USA. 2002; 99: 3657-3662
- Semaphorin3A enhances endocytosis at sites of receptor-F-actin colocalization during growth cone collapse.J Cell Biol. 2000; 149: 411-422
- ABL2/ARG tyrosine kinase mediates SEMA3F-induced RhoA inactivation and cytoskeleton collapse in human glioma cells.J Biol Chem. 2008; 283: 27230-27238
- Semaphorin-mediated axonal guidance via Rho-related G proteins.Curr Opin Cell Biol. 2001; 13: 619-626
- Regulation of smooth muscle contraction by small GTPases.Physiology (Bethesda). 2009; 24: 342-356
- Expression of vascular endothelial growth factor receptors in smooth muscle cells.J Cell Physiol. 2001; 188: 359-368
- Upregulation of neuropilin-1 by basic fibroblast growth factor enhances vascular smooth muscle cell migration in response to VEGF.Cytokine. 2005; 32: 206-212
- VEGF-A165 induces human aortic smooth muscle cell migration by activating neuropilin-1-VEGFR1-PI3K axis.Biochemistry. 2008; 47: 3345-3351
- Neuropilin-1 mediates PDGF stimulation of vascular smooth muscle cell migration and signalling via p130Cas.Biochem J. 2011; 435: 609-618
- Temporally controlled somatic mutagenesis in smooth muscle.Genesis. 2000; 28: 15-22
- An optimized protocol for detection of E. coli beta-galactosidase in lung tissue following gene transfer.Histochem Cell Biol. 2005; 124: 77-85
- Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor.Cell. 1998; 92: 735-745
- An Akt- and Fra-1-dependent pathway mediates platelet-derived growth factor-induced expression of thrombomodulin, a novel regulator of smooth muscle cell migration.Am J Pathol. 2010; 177: 119-131
- Semaphorin-induced cytoskeletal collapse and repulsion of endothelial cells.Methods Enzymol. 2008; 443: 299-314
- Scaling the microrheology of living cells.Phys Rev Lett. 2001; 87: 148102
- Universal physical responses to stretch in the living cell.Nature. 2007; 447: 592-595
- Fluidization and resolidification of the human bladder smooth muscle cell in response to transient stretch.PLoS One. 2010; 5: e12035
- Urodynamic properties and neurotransmitter dependence of urinary bladder contractility in the BK channel deletion model of overactive bladder.Am J Physiol Renal Physiol. 2005; 289: F604-F610
- Smooth muscle caveolae differentially regulate specific agonist induced bladder contractions.Neurourol Urodyn. 2007; 26: 71-80
- Age-dependent effects of secreted Semaphorins 3A, 3F, and 3E on developing hippocampal axons: in vitro effects and phenotype of Semaphorin 3A (−/−) mice.Mol Cell Neurosci. 2001; 18: 26-43
- Heparin-binding EGF-like growth factor.Biochim Biophys Acta. 1997; 1333: F179-F199
- Myocardin is a key regulator of CArG-dependent transcription of multiple smooth muscle marker genes.Circ Res. 2003; 92: 856-864
- Transforming growth factor-beta1-induced expression of smooth muscle marker genes involves activation of PKN and p38 MAPK.J Biol Chem. 2005; 280: 31172-31181
- Signaling through PI3K/Akt mediates stretch and PDGF-BB-dependent DNA synthesis in bladder smooth muscle cells.J Urol. 2003; 169: 2388-2393
- VEGF receptors and neuropilins are expressed in the urothelial and neuronal cells in normal mouse urinary bladder and are upregulated in inflammation.Am J Physiol Renal Physiol. 2008; 295: F60-F72
- Urothelial expression of neuropilins and VEGF receptors in control and interstitial cystitis patients.Am J Physiol Renal Physiol. 2008; 295: F1613-F1623
- Upregulation of vascular endothelial growth factor isoform VEGF-164 and receptors (VEGFR-2, Npn-1, and Npn-2) in rats with cyclophosphamide-induced cystitis.Am J Physiol Renal Physiol. 2008; 295: F826-F836
- Neuropilin-VEGF signaling pathway acts as a key modulator of vascular, lymphatic, and inflammatory cell responses of the bladder to intravesical BCG treatment.Am J Physiol Renal Physiol. 2010; 299: F1245-F1256
- The bladder extracellular matrix.Nat Rev Urol. 2009; 6: 596-611
- Extracellular matrix in vascular morphogenesis and disease: structure versus signal.Trends Cell Biol. 2003; 13: 51-56
- Role of human airway smooth muscle in altered extracellular matrix production in asthma.Clin Exp Pharm Physiol. 2001; 28: 233-236
- Cloning and characterization of neuropilin-1-interacting protein: a PSD-95/Dlg/ZO-1 domain-containing protein that interacts with the cytoplasmic domain of neuropilin-1.J Neurosci. 1999; 19: 6519-6527
- Synectin, syndecan-4 cytoplasmic domain binding PDZ protein, inhibits cell migration.J Cell Physiol. 2000; 184: 373-379
- Large-scale mapping of human protein-protein interactions by mass spectrometry.Mol Syst Biol. 2007; 3: 89
- Neuropilin-1/GIPC1 signaling regulates alpha5beta1 integrin traffic and function in endothelial cells.PLoS Biol. 2009; 7: e25
- KCl evokes contraction of airway smooth muscle via activation of RhoA and Rho-kinase.Am J Physiol Lung Cell Mol Physiol. 2004; 287: L852-L858
- Participation of Rho-associated kinase in electrical stimulated and acetylcholine-induced contraction of feline esophageal smooth muscle.Eur J Pharmacol. 2009; 607: 220-225
Article Info
Publication History
Published online: June 11, 2012
Accepted:
April 5,
2012
Footnotes
Supported in part by grants from the NIH ( P50 DK065298 to D.R.B., M.K., and R.M.A.; R01 DK077195 to R.M.A.; K01 CA118732 to D.R.B.; and R01 CA037392 to M.K.).
Supplemental material for this article can be found at http://ajp.amjpathol.org or at http://dx.doi.org/10.1016/j.ajpath.2012.04.013.
Identification
Copyright
© 2012 American Society for Investigative Pathology. Published by Elsevier Inc.
User License
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
00350-1&id=gr1.jpg){kind=link}
00350-1&id=gr2.jpg){kind=link}
00350-1&id=gr3.jpg){kind=link}
00350-1&id=gr4.jpg){kind=link}
00350-1&id=gr5.jpg){kind=link}
00350-1&id=gr6.jpg){kind=link}