Anosmin-1-Like Effect of UMODL1/Olfactorin on the Chemomigration of Mouse GnRH Neurons and Zebrafish Olfactory Axons Development

Abstract

The impairment of development/migration of hypothalamic gonadotropin-releasing hormone (GnRH) neurons is the main cause of Kallmann's syndrome (KS), an inherited disorder characterized by hypogonadism, anosmia, and other developmental defects. Olfactorin is an extracellular matrix protein encoded by the UMODL1 (uromodulin-like 1) gene expressed in the mouse olfactory region along the migratory route of GnRH neurons. It shares a combination of WAP and FNIII repeats, expressed in complementary domains, with anosmin-1, the product of the ANOS1 gene, identified as the causative of KS. In the present study, we have investigated the effects of olfactorin in vitro and in vivo models. The results show that olfactorin exerts an anosmin-1-like strong chemoattractant effect on mouse-immortalized GnRH neurons (GN11 cells) through the activation of the FGFR and MAPK pathways. In silico analysis of olfactorin and anosmin-1 reveals a satisfactory similarity at the N-terminal region for the overall arrangement of corresponding WAP and FNIII domains and marked similarities between WAP domains' binding modes of interaction with the resolved FGFR1-FGF2 complex. Finally, in vivo experiments show that the down-modulation of the zebrafish z-umodl1 gene (orthologous of UMODL1) in both GnRH3:GFP and omp ^2k :gap-CFP ^rw034 transgenic zebrafish strains leads to a clear disorganization and altered fasciculation of the neurites of GnRH3:GFP neurons crossing at the anterior commissure and a significant increase in olfactory CFP + fibers with altered trajectory. Thus, our study shows olfactorin as an additional factor involved in the development of olfactory and GnRH systems and proposes UMODL1 as a gene worthy of diagnostic investigation in KS.

Keywords: GnRH (gonadotropin-releasing hormone); UMODL1; anosmia; anosmin-1; development; hypothalamus and neuroendocrinology; olfactorin; olfactory axons.

FIGURE 1

Structural domains of olfactorin and anosmin-1 and analysis of their N-terminal region. (A) The schematic comparison of the structures and of the main protein domains of anosmin-1 and olfactorin. (B) The ribbon representation of the domain solution structure model of human (green) and mouse (blue) olfactorins showing a similarity with human anosmin-1 (grey) in the N-terminal region of the molecules with common WAP (whey acidic protein-like) and FNIII (fibronectin type III) domains. (C) The ribbon representation and comparison of the WAP domains; the corresponding cysteine (C) residues forming disapplied bridges are evidenced. Abbreviations: EGF, calcium-binding EGF-like domain; FNIII, fibronectin type 3 domain; H, histidine-rich regions; SEA, sea urchin sperm protein; SP, signal; TM, transmembrane domain; WAP, whey acidic protein domain; ZP, zona pellucida domain.

FIGURE 2

Identification of olfactorin in COS-7 cells transfected with the pCMV Sport6.1-FLAG-BQ887653 mouse Umodl1 expression vector. (A) Immunofluorescence detection of olfactorin with an anti-FLAG antibody in COS-7 cells transfected with the pCMV Sport6.1-FLAG-BQ887653 vector. (B) Olfactorin expression in COS-7 cells was evaluated by WB after 48 h of transfection. In protein extracts (olfactorin_FLAG), a full-length olfactorin (148 kDa) detected with olfactorin-serum (olfactorin-Ab) was overexpressed; (C) in the CM of transfected cells the secreted form of olfactorin (62 kDa) was detected by both anti-olfactorin-serum and anti-FLAG-Ab. Experiments were performed independently three times, and a representative blot is shown. C: untrasfected cells; lipo: cells transfected with an empty vector.

FIGURE 3

Effects of the exposure to the olfactorin-enriched CM (olfactorin_FLAG) on the chemomigration of GN11 cells. Microchemotaxis experiments were performed in the Boyden's chamber using the CM from COS-7 cells transfected with the empty vector (CM) or with pCMV SPORT6.1 PPT-FLAG-UMODL1 (the olfactorin-enriched CM; olfactorin_FLAG) or pMT21myc-ANOS1 (the anosmin-1-enriched CM; anosmin-1). (A) Comparison of the chemotactic effects exerted by olfactorin- and anosmin-1-enriched CMs on GN11 neurons. Internal controls for the procedure are represented by the chemotaxis to culture medium alone (DMEM) or culture medium with 0.1% Fetal Bovine Serum (FBS) as general chemoattractant. The results are expressed as the number of migrated cells per square mm of the porous membrane in 3 h. Mean ± SD from more than 4–6 independent experiments. *p < 0.05 vs respective CMs; **p < 0.05 vs DMEM. (B) Effect of immunoneutralization with a monoclonal anti-FLAG antibody on GN11 cell chemomigration induced by olfactorin_FLAG. (C) Involvement of HSPG in the chemotaxis of GN11 cells induced by olfactorin_FLAG. The CM was preincubated at 37°C with 30 μg/ml heparin for 30 min, prior the Boyden’s chamber assay. The results (mean ± SD; n. 6) are expressed as the relative chemotactic response of GN11 cells with respect to the stimulus exerted by 0.1% FBS. *p < 0.05 vs respective CM.

FIGURE 4

Effect of the ERK1/2 signaling pathway inhibitor on anaosmin-1- and olfactorin_FLAG-induced GN11 chemomigration. GN11 cells were pretreated with ERK1/2 inhibitor U0126 (1 µM) and then to the exposed control CM, anosmin-1 (A) or olfactorin_FLAG (B). The results (mean ± SD; n. 4) are expressed as the relative chemotactic response of GN11 cells with respect to the stimulus exerted by 0.1% FBS. *p < 0.05 vs. respective CM.

FIGURE 5

Effect of growth factor receptor inhibitors on olfactorin_FLAG-induced GN11 chemomigration. GN11 cells were pretreated with FGFR1-3 inhibitor BGJ398 (10 and 50 µM) (A), VEGFR inhibitor AMG-706 (1 and 0.1 µM) (B), HGFR inhibitor SU11274 (10 and 50 µM) (C) and exposed to the control CM or olfactorin_FLAG. The results (mean ± SD; n. 4) are expressed as the relative chemotactic response of GN11 cells with respect to the stimulus exerted by 0.1% FBS. *p < 0.05 vs. respective CM; **p < 0.05 vs. response to the CM of cells not pretreated with blockers.

FIGURE 6

Transcriptional analysis of olfactorin effects on GN11 cells. (A) GN11 neurons were transfected with OSE luciferase reporter and then exposed 24 h to increasing doses of FGF2 (at 10 and 40 ng), used as internal assay control, to the control CM and to olfactorin_FLAG. (B) Immunoneutralization of the olfactorin_FLAG by preincubation with the anti-FLAG antibody. All treatments were done in triplicate, with cultures repeated three times in total; NT and control untreated cells. Data are presented as fold induction vs. CM (mean ± SD, n. 4). *p < 0.05 vs NT; **p < 0.05 vs CM.

FIGURE 7

Potential interaction between the modeled human olfactorin structures with the resolved FGFR1-FGF2 complex (PDB code: 1FQ9) and its comparison with anosmin-1. As evidenced (red circles) WAP domains of olfactorin and anosmin-1 interact with the D2 domain of FGFR1 by assuming comparable arrangements. Greater differences are seen in the pose of the FNIII.1 domain; however, in both complexes it appears to conveniently approach the D2 domain of FGFR1. As shown in the left panel, the WAP-FGFR1 interactions appear to be mostly stabilized by ionic contacts.

FIGURE 8

Depletion of z-umodl1 affects fasciculation, trajectory, and connectivity of GnRH3+ and OMP + neurons of the zebrafish olfactory system. (A). Scheme of exon/intron junctions of the z-umodl1 gene. The normal splicing is shown on the top, the abnormal splicing due to the Splice-Site (SS) MO is on the bottom. This altered splicing leads to a frame-shift mutation. Black arrows indicate the primers used for the mRNA analysis to detect the aberrant splicing.(B). RT-PCR analysis of RNA from mismatch and SS MO-injected embryos, amplified with the primers indicated in panel a. In the sample from SS MO-treated embryos an additional band lacking exon 2 is evident that is not present in the control sample. The z-gapdh mRNA was also detected for reference. (C). Scheme showing the positions of the GnRH3:GFP + neurons (in green), the OMP: CFP + neurons (in blue) and the Trcp2: Venus + neurons (in yellow, not used in this study) relative to the eyes, the olfactory placodes (OPL), the olfactory bulbs (OB) and the olfactory nerves, in a frontal view. The anterior commissure (AC) is shown at the basis of the OB. (D). Scheme illustrating the view planes for the images of the GnRH + neurons (anterior-ventral) and for the OMP + neurons (frontal). GL, Glomeruli. (E–G). Micrographs of gnrh3:GFP zebrafish embryos that were either not injected (D) or injected with the control mismatched MO (E) or with anti-z-umodl1 SS (F) MO. In control injected embryos, no significant alteration was observed compared to not injected. White arrows indicate the normal position of the GFP + neurons. Red arrows and asterisks indicate, respectively, misrouted GFP + fibers and altered fasciculation or absent fibers. (H). Whole-mount bright field micrographs of embryos corresponding to the micrographs above, showing normal embryonic morphology and growth. (I). Quantification of the observed phenotype (altered fasciculation/trajectory), expressed as a percent of the injected GFP + embryos showing the indicated phenotype, over the total number of GFP + embryos examined. Open boxes, not injected; grey-shaded boxes, mismatch MO; solid black boxes, anti-z-umodl1 SS MO. * indicates p < 0.05 (J–L). Micrographs of omp:CFP zebrafish embryos either not-injected (i), injected with the control mismatch MO (j) or injected with anti-z-umodl1 SS (K) MO at 400 μM concentration. In control injected embryos, no significant alteration was observed compared to not injected. White arrows indicate the normal position of the olfactory placode (OPL) and of the glomeruli (GL). Red arrows and asterisks indicate, respectively, misrouted or mis-fasciculated OMP + fibers and absent glomeruli. (M, N). Same as in i-k, but the anti-z-umodl1 control and SS (m and n) MO were used at 800 μM concentration. (P). Whole-mount bright field micrographs of embryos corresponding to the micrographs in j-n, showing normal embryonic morphology and growth. (Q). Quantification of the observed phenotypes (altered trajectory/glomerulogenesis), expressed as a percent of the injected CFP + embryos showing the indicated phenotype, over the total number of CFP + embryos examined. Open boxes, not injected; grey-shaded boxes, mismatch MO; solid black boxes, anti-z-umodl1 SS MO. *p < 0.05.