Point mutations in KAL1 and the mitochondrial gene MT-tRNAcys synergize to produce Kallmann syndrome phenotype

Kallmann syndrome (KS) is an inherited developmental disorder defined as the association of hypogonadotropic hypogonadism and anosmia or hyposmia. KS has been shown to be a genetically heterogeneous disease with different modes of inheritance. However, variants in any of the causative genes identified so far are only found in approximately one third of KS patients, thus indicating that other genes or pathways remain to be discovered. Here, we report a large Han Chinese family with inherited KS which harbors two novel variants, KAL1 c.146G>T (p.Cys49Phe) and mitochondrial tRNAcys (m.5800A>G). Although two variants can’t exert obvious effects on the migration of GnRH neurons, they show the synergistic effect, which can account for the occurrence of the disorder in this family. Furthermore, the disturbance of the mitochondrial cysteinyl-tRNA pathway can significantly affect the migration of GnRH cells in vitro and in vivo by influencing the chemomigration function of anosmin-1. Our work highlights a new mode of inheritance underlay the genetic etiology of KS and provide valuable clues to understand the disease development.


Protein structure imitation and pathogenicity Prediction
To analyze the possible impacts of the amino acid substitution caused by p.Cys49Phe mutation on the three-dimensional protein structures and its consequence on protein functions, we used the PolyPhen program (http://genetics.bwh.harvard.edu/pph/) and the SIFT program (http://sift.jcvi.org/www/SIFT_enst_submit.html), which allow to predict whether an amino acid change is likely to be deleterious to protein function. Secondary structure of wild-type and mutated anosmin-1 had been predicted using PHYRE2 (http://www.sbg.bio.ic.ac.uk/phyre2).

Mutation Screening
Different tissues were collected from the family members, and genomic DNA was extracted using standard methods. Fourteen exons of the KAL1 gene were amplified and analyzed by Sanger sequencing (Invitrogen, USA). For analysis of the entire mitochondrial genome, eight overlapping fragments were amplified by PCR using 8 sets of primers; these fragments were purified and subsequently analyzed by direct sequencing (Invitrogen, USA). The assembled sequence was compared to the revised consensus Cambridge sequence (rCRS, GenBank accession number: NC-012920). PCRbased restriction fragment length polymorphism (PCR-RFLP) analyses were used to screen for single nucleotide polymorphisms (SNPs) in the KAL1 gene and the mitochondrial DNA site at position 5800 using standard methods. The uniqueness of the KAL1 and mtDNA variations in this family was confirmed by an exhaustive search in the relevant databases (dbSNP, 1000 Genome Project, NHLBI Exome Variant Server and human mitochondrial database) and by screening at least 2,000 control individuals of the same origin with no history of KS. Nucleotide numbering used +1 as the A of the ATG translation initiation codon in the reference sequence, with the initiation codon as codon 1. For mtDNA, nucleotide numbering referred to rCRS. transferred into 15-ml conical centrifuge tubes. A Ficoll-Hypaque gradient was used to separate the mononuclear cells, and the cells were resuspended in transformation medium, which consisted of RPMI-1640 medium containing harvested EBV stock, 15% fetal calf serum, and 0.03 mg/ml cyclosporin A. The suspension cultures were incubated 7 to 10 days in 25-cm 2 tissue culture flasks and were periodically fed by replacing one third of the medium. The transformed B-cell lines emerged after approximately 6 weeks, after which they rapidly became the only viable elements in the cultures.

Cell culture, cell proliferation, and immunofluorescence
Mouse GnRH neuronal NLT cells were kindly provided by Xiao-juan Zhu (Northeast Normal University, China) and were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin (Invitrogen, Carlsbad, CA); the cells were incubated under standard conditions (5% CO2, 37°C). To examine the subcellular localization of anosmin-1, HEK293T cells were transfected with pEGFP-N1/KAL1 or KAL1 G146T for 48 h. The cells were then fixed with 4% paraformaldehyde for 30 min, and the nuclei were stained with Hoechst 333342 (Sigma). The fluorescent signals were examined under a Nikon epifluorescence microscope (Nikon Eclipse 80i; Nikon, Tokyo, Japan).

Whole-mount in situ hybridization
Embryos were fixed in phosphate-buffered 4% paraformaldehyde. Whole-mount in situ hybridization was performed. cDNA fragments for the gnrh2, gnrh3, kal1a, and kal1b sequences were cloned into the pESAY-T3 vector (Transgen) and were used to synthesize antisense RNA probes with digoxigenin-UTP.

Wound healing assay
NLT cells were incubated in 12-well plates and wounded in the confluent monolayer using a thick plastic cell scraper (Corning). Then, the cells were rinsed two times with PBS and fresh culture medium was added. Cells were captured under a microscope (Eclipse Ti-S, Nikon) at 0, 2, 4, 6, 9, 12 and 24 hours after scrape. Total distance migrated by cells in indicated time was determined as the difference of the cell front relative to the 0 hour timepoint. Experiments were repeated three times, with three distance measurements per field of view.

Supplementary figures
Supp. Figure S1 X chromosome region Xp22.32 was closely linked to the KS phenotype.
Two polymorphic STR markers on X chromosome region Xp22.32, DXS9895 and DXS9906, flanking KAL1 gene were used for linkage analysis. The genotypes of the family members were shown by numbers. The haplotype ' ' was closely linked to the KS phenotype.

Supp. Figure S3 Prediction of structures of anosmin-1.
Structure prediction of wild type (up panel) and mutant type anosmin-1 (down panel). The sticks of mutated residue was circled. Electrostatic potential mapped onto the molecular surface of anosmin-1 and mutant anosmin-1 with the two bound potentials were contoured from -5 (red) to +5 kT/e (blue).

Supp. Figure S4 The nonsynonymous amino acid substitution in anosmin-1 might not affect the subcellular localization and expression of anosmin-1. (A and B). Both the anosmin-1-GFP and mutated anosmin-1-GFP proteins were highly expressed in 293T cells (A) and localized in the endoplasmic reticulum (ER) and in the Golgi region (B). GM-130, a cis-Golgi matrix protein, as the
Golgi complex marker to demonstrate the localization of anosmin-1 in the cytoplasm. Red Arrow indicated anosmin-1-GFP accumulated in ER and Golgi region. Red arrow heads indicated anosmin-1-GFP localized in Golgi region. When 293T cells were transfected with wild type or mutated KAL1 overexpression vectors for 24 h and treated with 10 μg/ml brefeldin A, a protein transport inhibitor which blocks ER-Golgi membrane trafficking, for 30min, accumulations of both wild type and mutated anosmin-1-GFP were detected in the cytoplasmic region (A). Blue, Hoechst; green, anosmin-1-GFP; Red, GM-130. Scale bar: 50 μm. (C and D). Western blotting analysis of expression level of anosmin-1-GFP protein, produced by transfected 293T cells (down panel) and presented in their culture media conditioned for 16-18 h (up panel) (C). The results were also shown no significant difference between them (D).
Supp. Figure S5 CARS2 was localized to the mitochondrion. (A). The phylogeny tree of CARS2 proteins with human CARS protein as an out-group. The tree was constructed by the neighbor-joining method with MEGA5.2. Hs, Homo sapiens, Mm, Mus musculus, Dr, Danio rerio, Dm, Drosophila melanogaster, Rn, Rattus norvegicus, Pt, Pan troglodytes, Xl, Xenopus laevis. (B). CARS2 was a mitochondrial protein. The 293T cells were transfected with GFP-tagged CARS2 and purified mitochondria were isolated by standard methods. Western blotting analysis indicated the distribution of CARS2 between cytoplasm and mitochondrion . Cytoplasmic protein GAPDH was used as the control.

Supp. Figure S6 Overexpression of CARS2 did not affect the migration distance of NTL cells. (A).
Serum starvation blocked NLT cells motility. Addition of serum induced robust migration of NLT cells that were transfected with pcDNA3.1 or pcDNA3.1/CARS2. Scale bar：50 μm. (B). Quantification shows no significant difference in the migration distance of CARS2 overexpressed cells (•) compared with cells transfected with control vectors (■).
Supp. Figure S7 ATP levels, mitochondrial translation, structure and cell growth were not affected in KS patient. (A). ATP level of B cell lines established from KS patients was similar as B cell lines from controls. (B). Expression levels of mitochondrion-encoded and nuclear-encoded mitochondrial proteins were not affected in B cell lines of KS patients. CYTB, a mitochondrion-encoded protein, is a component of the ubiquinol-cytochrome c reductase complex (complex III or cytochrome b-c1 complex), which is a respiratory chain that generates an electrochemical potential coupled to ATP synthesis. Mitofusin-1 (MFN1, NP_284941.2), a nuclear-encoded mitochondrial essential transmembrane GTPase, which mediates mitochondrial fusion. Western blotting analysis of expression level of B cell lines established from KS patients was similar as B cell lines from controls.
(C). The media from wild type and mutated anosmin-1 transfected control or KS B cell lines did not affect the growth of NLT cells. B cell lines from control or KS patients were transfected with pEGFP-N1, pEGFP-N1 KAL1 or pEGFP-N1 KAL1 G146T for 24 h and different groups of media were collected to treat 293T cells. After 24 h, cell growth was detected by CCK-8 assay.
(B). Toxicities caused by cars2-i2e3 MO in zebrafish embryos. 4 ng of cMO or cars2-i2e3 MO was injected into 1-cell stage WT embryos. One day after injection, normal, deformed, unfertilized and dead embryos were recorded. The results indicated three independent experiments.

Supplementary Tables
Supp.