Figure 1.
Genomic organization of the mouse Lrig2 gene and the generation and molecular analyses of Lrig2-deficient mice.
(A) Schematic drawing of the wild-type, conditional, and disrupted Lrig2 alleles. A PKG-neo selection cassette was inserted downstream of exon 12. Exon 12 and the PKG-neo cassette were flanked by loxP sites and were deleted together, in a single step, by mating with OzCre mice. Color coding is as in C. (B) Southern blot using tail DNA isolated from the offspring of an Lrig2E12+/- × Lrig2E12+/- mating. The expected sizes are as follows: wild-type (Lrig2E12+) allele, 11.0 kb; and Lrig2 exon 12-ablated (Lrig2E12-) allele, 6.6 kb. Lane B13 is from Lrig2E12-/-, lanes B10-12 and B14-15 are from Lrig2E12+/-, and the remaining lanes are from Lrig2E12+/+ mice. (C) Genomic organization of the mouse Lrig2 gene. Gene structure of Mus musculus Lrig2 is shown. Gene organization was deduced from the sequence of the mouse genome and the Lrig2 mRNA (GenBank accession number NM_001025067). Exons are indicated with boxes and are drawn to scale. Exon numbers are indicated with numbers. The Lrig2 gene is approximately 58 kb, and it is located on mouse chromosome 3 F2. Color coding depicts the encoded protein domains, including a signal peptide (red), a leucine-rich repeats domain (yellow), three immunoglobulin-like domains (blue), a transmembrane domain (orange), and a cytosolic domain (green). (D) Western blot of MEF cell lines of different Lrig2 genotypes. Top, anti-Lrig2 polyclonal; Bottom, anti-actin. (E) Lrig mRNA levels in mice of different genotypes. The mRNA levels of Lrig1, exon 12-containing Lrig2 (Lrig2E12), exon 17-18 boundary-containing Lrig2 (Lrig2E17-18), and Lrig3 in brains of 3-week old mice were analyzed using quantitative real-time RT-PCR. The Lrig/Rn18s ratio was calculated and normalized to the corresponding ratio in reference RNA from Stratagene. Shown are the means of wild-type (n=6), Lrig2E12+/- (n=5), and Lrig2E12-/- (n=8) mice, with error bars indicating the standard deviations. Significant differences compared with the wild-type mice are indicated with asterisks (*p<0.01 and **p<0.001).
Figure 2.
Lrig2-deficient mice have increased spontaneous mortality and show transiently reduced body weight compared with wild-type and heterozygous littermates.
Kaplan-Meier curves for the survival of male (A) and female (B) C57BL/6 mice of the different Lrig2 genotypes based on 153 male and159 female mice. (C) There were no significant differences in embryonic weight at E13.5 or at birth, but by 5 days of age, the Lrig2E12-/- mice were significantly lighter than the Lrig2E12+/+ and Lrig2E12+/- mice (* p<0.05). Male (D) and female (E) Lrig2E12-/- mice had a lower body weight than the Lrig2E12+/+ and Lrig2E12+/- mice from 3 to 10 weeks of age. At 12 weeks of age, the male (F) (p = 0.019), but not the female (G) (p = 0.051), Lrig2E12-/- mice were significantly smaller than Lrig2E12+/+ and Lrig2E12+/- mice, and at 15 weeks or older, all genotypes had a similar body weight. At least eight mice were included per genotype and sex for all ages.
Figure 3.
Histological and in situ hybridization analyses of PDGFB-induced mouse brain tumors.
Newborn mice were transduced intracranially with PDGFB-encoding RCAS retroviruses. At 12 weeks of age, the mice were sacrificed, and the histological sections were prepared. (A) A hematoxylin-eosin-stained section showing a tumor resembling human oligodendroglioma, WHO grade II/III (low grade). The tumor tissue showed moderate cellularity and was composed of monomorphic cells with uniform, round nuclei. Original magnification 200X. (B) A hematoxylin-eosin-stained section showing a tumor resembling human glioblastoma, WHO grade IV (high grade), with pseudopalisading necroses. Original magnification 100X. (C) A hematoxylin-eosin stained section showing a high grade glioma with pseudopalisading necroses (arrow) and frequent mitoses. Original magnification 200X. (D) Magnification of C, mitoses are shown with arrows. (E) In situ hybridization analysis of Lrig2 in high grade glioma. The in situ hybridization signals are shown as red dots. Original magnification 400X. (F) In situ hybridization analysis of the same tumor as in E, using a negative control probe, showing low non-specific background staining.
Figure 4.
The transduction efficiency of RCAS-PDGFB-HA in Ntv-a cells of different Lrig2 genotypes.
Neural cells from brains of newborn Ntv-a transgenic mice were transduced with RCAS-PDGFB-HA in vitro. After 5 days of infection, cells were stained with fluorescent anti-HA antibodies (green) and analyzed using fluorescence microscopy. Cell nuclei were counter-stained with DAPI (blue). (A) Wild-type (Lrig2E12+/+) cells. (B) Heterozygous (Lrig2E12+/-) cells. (C) Lrig2-defecient (Lrig2E12-/-) cells. (D) Quantifications of transduction efficiencies. Shown are the means from three independent experiments, including wild-type (n=7), heterozygous (n=12), and Lrig2-deficient (n=7) cell lines from three different litters, with standard deviations indicated by error bars. There were no apparent differences in the transduction efficiency between cells of different Lrig2 genotypes.
Figure 5.
Pdgfr levels in serum-starved primary MEFs.
Wild-type (Lrig2E12+/+), heterozygous (Lrig2E12+/-), or Lrig2-deficient (Lrig2E12-/-) MEF cell lines were serum-starved for the indicated times followed by analysis of Pdgfr levels using immunofluorescence microscopy (A–H) or immunoblotting (I–L). (A–H) Cells were serum starved for 24 hours followed by analysis of Pdgfrα (A–D) or Pdgfrβ (E–H) levels using immunofluorescence microscopy. Pdgfr immunoreactivity was visualized with Alexa-conjugated secondary antibodies (red), and the nuclei were counter-stained with DAPI (blue). (D, H) Quantifications of the Pdgfrα and Pdgfrβ immunofluorescence levels, respectively. Shown are the means from three independent experiments, including wild-type (n=8), heterozygous (n=9), and Lrig2-deficient (n=6) cell lines from three different litters, with standard deviations indicated by error bars. There were no differences observed in Pdgfrα or Pdgfrβ protein levels, as determined by quantitative immunofluorescence microscopy, between cells of different Lrig2 genotypes. (I–L) Cells were serum starved for 0, 6, 12 or 24 hours followed by cell lysis and Western blot analyses with antibodies against Pdgfrα, Pdgfrβ, or, as a loading control, actin. Representative immunoblots of Pdgfrα (I) and Pdgfrβ (J) in wild-type (Lrig2E12+/+), heterozygous (Lrig2E12+/-), and Lrig2-deficient (Lrig2E12-/-) cells after 0 or 24 hours of serum starvation. (K–L) Quantifications of Pdgfrα (K) and Pdgfrβ (L) immunoblots in wild-type, heterozygotes and Lrig2-deficient cells after 0, 6, 12 and 24 h starvation. Shown are the means from three independent experiments including wild-type (n=8), heterozygous (n=9), and Lrig2-deficient (n=6) cell lines from three different litters, with standard deviations indicated by error bars. There were no differences observed in Pdgfrα or Pdgfrβ protein levels as determined by immunoblotting, between cells of different Lrig2 genotypes.
Figure 6.
Expression level analyses of co-transfected LRIG1, LRIG2, and PDGFRα.
HEK-293T cells were co-transfected with expression vectors encoding myc-LRIG1 or FLAG-LRIG2 and PDGFRα. The numbers indicate amount of plasmid (μg) used in the respective transfection. Empty pcDNA 3.1 and p3XFLAG-CMV-13, respectively, were used to bring the total amount of plasmid DNA to the same amount (2 µg) in each transfection. Cell extracts were analyzed by Western blotting with antibodies against LRIG1, PDGFRα, FLAG (recognizing FLAG-LRIG2), or, as a loading control, actin. (A, C) Representative Western blots. Two specific PDGFRα bands were observed, here called PDGFRα-upper and PDGFRα-lower, respectively. (B, D) Quantification of PDGFRα immunoblots. Shown are the means of four independent co-transfection experiments, with error bars indicating the standard deviations. Significant differences compared with the empty vector control are indicated with asterisks (*p<0.05).
Figure 7.
The length of the primary cilia in growth-arrested primary MEFs.
Wild-type, heterozygous, and Lrig2-deficient cells were serum starved for 24 hours followed by staining of primary cilia with antibodies against acetylated tubulin (red). The cell nuclei were counter-stained with DAPI (blue). Representative confocal immunofluorescence micrographs of: (A) wild-type (Lrig2E12+/+) cells, (B) heterozygous (Lrig2E12+/-) cells, and (C) Lrig2-defecient (Lrig2E12-/-) cells. (D) Quantification of primary cilia length. Shown are the means from three independent experiments including wild-type (n=8), heterozygous (n=9), and Lrig2-deficient (n=6) cell lines from three different litters, with standard deviation indicated by error bars. There were no differences observed in the abundance (data not shown) or the length of primary cilia in cells of different Lrig2 genotypes.
Figure 8.
The effect of Lrig2 on the induction of immediate-early gene expression.
Wild-type (Lrig2E12+/+), heterozygous (Lrig2E12+/-), or Lrig2-deficient (Lrig2E12-/-) MEFs were serum starved for 24 hours followed by stimulation with PDGF-BB for 0, 10, 20, 40, 60, or 120 minutes. Thereafter, total RNA was prepared, and gene expression was quantified using real-time RT-PCR. Samples were run in triplicate, and the specific mRNA levels were normalized to respective Rn18s levels. Shown are the specific mRNA/Rn18s ratios on arbitrary scales. (A) Kinetics of relative Fos expression in cells stimulated with 50 ng/ml PDGF-BB. Shown are the means from three independent experiments, including wild-type (n=8), heterozygous (n=9), and Lrig2-deficient (n=6) cell lines from three different litters, with standard error of the means indicated by error bars. (B) Kinetics of relative Egr2 expression levels in cells stimulated with 10 ng/ml PDGF-BB. Shown are the means from two independent experiments, including wild-type (n=6), heterozygous (n=6), and Lrig2-deficient (n=4) cell lines from two different litters, with standard error of the means indicated by error bars. Compared with wild-type cells, the Lrig2-deficient cells showed altered and faster kinetics of induction of both Fos and Egr2 expression in response to PDGF-BB stimulation. Significant differences compared with the wild-type cell lines are indicated with asterisks (*p<0.05 and **p<0.01).
Figure 9.
PDGF induced phosphorylation events in cells of different Lrig2 genotypes.
Wild-type, heterozygous, or Lrig2-deficient MEFs were serum starved for 24 hours followed by stimulation with 50 ng/ml PDGF-BB for different times. (A–N) Cells were untreated or stimulated with 50 ng/ml PDGF-BB for 10 minutes followed by cell fixation and analysis of the phosphorylation status of respective Pdgfr by in situ proximity ligation assay (PLA). Phosphorylated Pdgfr was visualized using fluorescence (red spots). Cell nuclei were counter-stained with DAPI (blue). (A–F) Representative PLA images of phosphorylated Pdgfrα (red spots) in un-stimulated (A–C) or PDGF-BB stimulated (D–F) cells of the indicated Lrig2 genotypes. (G–L) Representative PLA images of phosphorylated Pdgfrβ (red spots) in non-stimulated (G–I) or PDGF-BB-stimulated (J–L) cells of the indicated Lrig2 genotypes. (M) Quantification of PLA spots for phosphorylated Pdgfrα. Shown are the means from three independent experiments, including wild-type (Lrig2E12+/+, n=8), heterozygous (Lrig2E12+/-, n=9), and Lrig2-deficient (Lrig2E12-/-, n=6) cell lines from three different litters, with standard deviations indicated by error bars. (N) Quantification of PLA spots for phosphorylated Pdgfrβ. Shown are the means from three independent experiments, including wild-type (Lrig2E12+/+, n=8), heterozygous (Lrig2E12+/-, n=9), and Lrig2-deficient (Lrig2E12-/-, n=5) cell lines from three different litters, with standard deviations indicated by error bars. There were no differences observed in the levels of activated Pdgfrα or Pdgfrβ between cells of different genotypes. (O–Q) Cell lysates from cells that had been untreated or treated with 50 ng/ml PDGF-BB for 15 minutes were analyzed through Western blotting with antibodies against the indicated proteins. (O) A representative Western blot is shown of Akt, Erk1/2, phosphorylated Akt (pAkt), and phosphorylated Erk1/2 (pErk1/2) using cell lysates from cells of the indicated Lrig2 genotypes that had been untreated (-) or treated (+) with PDGF-BB. (P) Quantification of pAkt/Akt-ratios for non-stimulated and stimulated cells, respectively. Shown are the means from three independent experiments, including wild-type (n=8), heterozygous (n=9), and Lrig2-deficient (n=5) cell lines from three different litters, with standard deviations indicated by error bars. (Q) Quantification of pErk1/2/Erk1/2-ratios for non-stimulated and stimulated cells, respectively. Shown are the means and corresponding standard deviations as for P.