www.symbiosisonline.org Symbiosis www.symbiosisonlinepublishing.com Research article SOJ Immunology Open Access The PDGFC CUB Domain Enhances Survival in PDGFC Mutant Mice Brian J. Hayes1, Kimberly J. Riehle1,2, Debra G. Gilbertson3, Wendy R. Curtis3, Edward J. Kelly4, Piper M. Treuting5 and Jean S. Campbell1* 1Department of Pathology, University of Washington, Seattle, WA 98195, USA 2Department of Surgery, University of Washington, Seattle, WA 98195, USA 3Zymogenetics Inc, a Bristol-Myers Squibb Company, Seattle, WA 98102, USA 4Department of Pharmaceutics, University of Washington, Seattle, WA 98195, USA 5Department of Comparative Medicine, University of Washington, Seattle, WA 98195, USA Received: May 01, 2015; Accepted: August 27, 2015; Published: September 14, 2015 *Corresponding author: Jean S. Campbell, Executive Director of Research and Development, Oncosec Medical, 434 N. 34th St, Suite 100; Seattle, WA 98103, USA, Tel: +(206) 221-5422; E-mail: [email protected] PDGF-AA, PDGF-BB, and PDGF-AB are secreted as active Absract dimers [8], while PDGF-CC and PDGF-DD are secreted as Platelet-derived growth factor (PDGF) signaling pathways are inactive homodimers requiring extracellular cleavage of an necessary for normal development. Here we report a homozygous N-terminal complement components C1r/C1s sea urchin EGF Pdgf-c mutant mouse that is viable. In this mouse, alternative splicing bone morphogenic protein 1 (CUB) domain to allow receptor gives rise to a truncated transcript containing the entire coding binding (Figure 1a)[9-12]. All PDGFs share a common growth region of the complement components C1r/C1s sea urchin EGF bone factor domain (GFD), which is 110 amino acids long and contains majority of the Growth Factor Domain (GFD). A mutant protein is 8 conserved cysteines that facilitate intra- and intermolecular translatedmorphogenetic from theprotein truncated 1 (CUB) sequence domain in vitro.of PDGFC However, but thelacking viability the of our homozygous mutant Pdgf-c mice depends on the presence of through the GFD, and mutations of these cysteines and loss of both Pdgfrα alleles. disulfide bonds [13]. PDGF signal transduction is mediated Keywords: [14,15]. Furthermore, mutation of sequences in loops I, II, or III their disulfide bonds cause a lack of PDGFR signal transduction PDGFC; Cleavage, Growth factor; Knockout Introduction signalof the transductionGFD, the domain by Pdgf that-c thusphysically should interacts require thewith conserved PDGFRs, decreases the GFD’s ability to bind PDGFRs [16-18]. Effective The platelet-derived growth factor (PDGF) signal transduction cysteines and loops I-III of the GFD, as well as cleavage of the CUB domain. We hypothesized that deletion of the majority of the GFD would thus eliminate receptor binding and downstream andpathway deletion has ofimportant Pdgf-a, -b, functions or –c results in development. in perinatal Mice lethality lacking in signal transduction. Contrary to our hypothesis, we describe a PDGF receptor (PDGFR) -α or –β are embryonic lethal [1,2], homozygous Pdgf-c mutant mouse wherein most of the GFD is in Pdgf-a and Pdgf-c In vitro studies demonstrate indeed deleted, but the expression of the CUB domain in these thathomozygous PDGF-AA, Knockout -BB, and -CC(KO) induce animals, Pdgf thoughrα dimerization, live births PDGF-BB, are seen and PDGF-DD induce KO dimerization mice [3-5]. of Pdgfrβ, and -AB, -BB, and Pdgfrα mice is sufficient for viability. Viability, however, depends on in vitro mutations [19,20]. genetic background, as has been described for other binding studies, Pdgf-b Pdgf-b -CC induce αβ -receptor dimerization [6]. As predicted by Methods KO mice have a phenotype similar to Mouse Models KO mice, with defects in kidney and hematologic Pdgfdevelopmentrα, which [2,4]. Mice lacking PDGFA have a defect in lung development The Pdgf-c locus was isolated from a 129S5/SvEvBrd genomic [3], and interestingly do not phenocopyPdgf deletion-c of causes abnormalitiestm1nagy in skeletal development and neural crest to as Pdgfc , have a defect in palate formation, resulting from BAC library. The targeting construct replaced a 5 kb genomic migration [1,7]. On the other hand, KO animals, referred abnormal neural crest cell migration and proliferation [5]. Ding, region on chromosome 3 with an IRES LacZ MC1 Neo cassette. et al. [5], further reported that Pdgf-a, Pdgf-c The homologous arms consisted of a 3 kb 5’ fragment and a 2.1 kb a phenotype similar to Pdgfrα that PDGF-C is 3’ fragment. Lex-1 129S5/SvEvBrd ES cells were transfected withc-2J double KO mice have generate chimeras. Chimeras were bred to albino C57BL/6J-Tyr a major activator of Pdgfrα signal transduction in the context of the targeting construct and injected into C57BL/6 blastocysts to development. KO mice, suggesting mice (The Jackson Laboratory) to generate F1 progeny. F1 hybrids were crossed to generate initial Mendelian ratios (Table 1). Mice Symbiosis Group *Corresponding author email: [email protected] The PDGFC CUB Domain Enhances Survival in PDGFC Mutant Mice Copyright: © 2015 Campbell et al. Figure 1: Design of targeting constructs to inactivate PDGF-C. a) A linear diagram of the PDGF-C protein showing the CUB (orange) and GFD (green) above the exons that correspond to these protein domains. b) The targeting strategy for Pdgfctm1lex mice, which removes exons 4 and 5. c) The target- ing strategy for the Pdgftm1nagy complement components C 1r/C1s sea urchin EGF b strain, which removes exon 2 Ding, et al. [5] Abbreviations : splice acceptor (SA), internal ribosomal entry site (IRES), Neomycin). one morphogenic protein 1 (CUB), beta galactosidase neomycin fusion protein ( β GEO), fusion polyprotein (A) tail (pA), Lac Z gene encoding beta galactosidase (Lac Z), enhancer, promoter, and neomycin phosphotransferase (MC1/ HSVtk and approved facility, and all experiments were performed with the were further backcrossed to a C57BL/6 background six times Universityfor the Assessment of Washington and Accreditation Institutional of LaboratoryAnimal Care Animal and CareUse Centersto generate (http://www.mmrrc.org/index.php the Mendelian ratio for C57BL/6) supported (Table 1). by These the Committee approval. NIH.mice canHemizygous be obtained Pdgf fromrα mice, the Mutant Pdgfra tm11(egfp)sorMouse Regional, were purchased Resource Necropsy and Histology housed at the University of Washington, which is an Association Homozygote Pdgfctm1lex from The Jackson Laboratory (stock # 007669). Animals were mice (n = 4) and WT C57BL/6 Citation: Page 2 of 11 Hayes BJ, Riehle KJ, Gilbertson DG, Curtis WR, Kelly EJ, et al. (2015) The PDGFC CUB Domain Enhances Survival in PDGFC Mutant Mice. SOJ Immunol 3(3): 1-11. DOI: http://dx.doi.org/10.15226/soji/3/3/00133 The PDGFC CUB Domain Enhances Survival in PDGFC Mutant Mice Copyright: © 2015 Campbell et al. Table 1: Heterozygous Pdgfctm1lex No. of Cohort breeding +/+ pairs in two backgrounds produce Pdgfcoffspringtm1lex/+ with normal Mendelian Pdgfctm1lex distribution. / Pdgfctm1lex Chi sq value animals 129Sv/EvBrd observed 99 50 0.19 expected 2625 50 2325 C57BL/6 observed 55 147 male 265 63 3.7 observed 29 expected 146 37 8373 3437 3.1 female observed 119 29 expected 2630 6459 30 0.83 Pdgfctm1lex/+ mice were bred and the genotypes of the resulting offspring were determined from DNA extracted from Male and female heterozygous tail snips by PCR. Two different genetic backgrounds, 129SvBrd and C57BL/6, were analyzed. The expected number of pups with a specific genotype was calculated based on the total number of pups analyzed. Analysis of normal Mendelian distribution was determined by Chi square analysis with two degrees of freedom. A Chi square number greater than 5.99 is statistically significant. ethan 2 asphyxiation followed by complete necropsy. Blood was collectedlittermates for (n chemistry = 4) aged and 5 andcomplete 8 weeks blood were counts euthanized performed by ol precipitation and resuspension in Tris-HCl 10mM EDTA CO 1mM pH 7.4 (TE). Polymerase chain reaction was carried out with 0.5 U GemTaq (MGQuest), supplied 5x buffer, and dNTPs to a final by Phoenix Central Laboratories (Everett, WA). Tissues were concentration of 0.2mM with primers 5’ CCTGGTCAAGCGCTGTGG collected for histological analysis with immersion fixation in 10% 3’(200nM), 5’ TCTGGATTCATCGACTGTGG 3’(200nM), 5’ phosphate-buffered formalin, 4-6 µm sections were made and ACGGCTAACATGGAGCACG 3’ (100 nM). All primers were adrenalstained withglands, hematoxylin gallbladder, and brown eosin. Tissuesand white examined adipose included: tissue, designed using Oligo Calc [21]. Cycling conditions were 95°C for lungs, esophagus, trachea, liver, kidneys, heart and great vessels, exocrine and endocrine pancreas, spleen, thyroid, submandibular, 3 min, and 35 cycles of 95°C for 20 sec, 60°C for 30 sec, and 72°C parotid and sublingual salivary glands, mesenteric lymph nodes, forSouthern 30 sec with blotting a final extension at 72°C for 10 min. mesentery, preputial or clitoral glands, skeletal muscle, bladder, - and small intestine, glandular and non-glandular stomach, The targeting construct was verified using standard methods uterusmale accessory and cross sex section glands, of testesthe head or (eyes,ovaries, middle haired and skin, internal large as described [22]. Briefly, genomic DNA was extracted as de ears, oral and nasal cavities, cerebrum, cerebellum, olfactory scribed above and purified DNA was restriction digested using lobes, brain stem, pituitary, tongue and teeth). All tissues were from Pdgfctm1lex