Downloaded from http://jmg.bmj.com/ on July 7, 2015 - Published by group.bmj.com
JMG Online First, published on July 3, 2015 as 10.1136/jmedgenet-2014-102931
New loci
ORIGINAL ARTICLE
Whole exome sequencing identifies LRP1 as a pathogenic gene in autosomal recessive keratosis pilaris atrophicans Joakim Klar,1 Jens Schuster,1 Tahir Naeem Khan,2 Muhammad Jameel,2 Katrin Mäbert,1 Lars Forsberg,1 Shehla Anjum Baig,3 Shahid Mahmood Baig,2 Niklas Dahl1 1
Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Biomedical Centre, Uppsala, Sweden 2 Human Molecular Genetics Laboratory, Health Biotechnology Division, National Institute for Biotechnology and Genetic Engineering (NIBGE), Pakistan Institute of Engineering and Applied Sciences (PIEAS), Faisalabad, Pakistan 3 Department of Pathology, Children’s Hospital, Pakistan Institute of Medical Sciences, (PIMS), Islamabad, Pakistan Correspondence to Dr Niklas Dahl, Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, BMC, Box 815, Uppsala 752 37, Sweden; [email protected] JK and JS contributed equally. Received 12 December 2014 Revised 27 May 2015 Accepted 14 June 2015
ABSTRACT Background Keratosis pilaris atrophicans (KPA) is a group of rare genodermatoses characterised by perifollicular keratosis and inflammation that progresses to atrophy and scars of the facial skin. Keratosis pilaris of extensor areas of limbs is a common associated finding. Most cases with KPA are sporadic and no consistent inheritance pattern has been documented. Methods A large consanguineous Pakistani pedigree segregating autosomal recessive KPA of a mixed type was subject to autozygosity mapping and whole exome sequencing. Quantification of mRNA and protein levels was performed on fibroblasts from affected individuals. Cellular uptake of the low-density lipoprotein (LDL) receptor-related protein 1 (LRP1) ligand α2macroglobulin (α2M) was quantified using fluorescence confocal microscopy. Results Genetic analyses identified a unique homozygous missense variant (K1245R) in the LRP1 in all affected family members. LRP1 encodes the LRP1, a multifunctional cell surface receptor with endocytic functions that belongs to the LDL receptor family. The LRP1 mRNA and LRP1 protein levels in fibroblasts of affected individuals were markedly reduced when compared with controls. Similarly, the LRP1-mediated cellular uptake of α2M was reduced in patient fibroblasts. Conclusions This is the first report on LRP1 as a pathogenic gene for autosomal recessive KPA and keratosis pilaris. The inflammatory characteristics of the KPA entity in our family suggest a link to the immuneregulatory functions of LRP1.
INTRODUCTION
To cite: Klar J, Schuster J, Khan TN, et al. J Med Genet Published Online First: [please include Day Month Year] doi:10.1136/ jmedgenet-2014-102931
Keratosis pilaris atrophicans (KPA) is a small group of follicular syndromes characterised by inflammation and atrophy.1 The morphological hallmarks are variable degrees of perifollicular inflammation, secondary scarring and/or alopecia. Three clinical entities of KPA are described: keratosis pilaris atrophicans faciei (KPAF), keratosis follicularis spinulosa decalvans (KFSD) and atrophoderma vermiculatum (AV).1–3 However, affected individuals may present with features that overlap between the three entities. The onset of KPAF is usually in infancy with erythema, follicular keratotic papules on the lateral parts of the eyebrows that
may extend to the forehead as well as scarring alopecia.2 KFSD also presents in infancy with papules primarily in the malar area, patchy scarring alopecia of the scalp as well as the absence of eyebrows and eyelashes.3 AV (MIM 209700) is considered a more severe form of KPA with a typical onset in late childhood or adolescence. The AV entity is characterised by erythema and follicular keratotic papules located on the cheeks, preauricular regions and forehead that may extend to the lips and earlobes.4 The papules progress to pitted atrophic and depressed scars in a reticular or honeycomb pattern. Eyebrows, eyelashes and scalp are usually spared but eyebrows may sometimes be involved. The course of the three KPA entities is usually slowly progressive with poor response to treatment. Typical histopathological findings in KPA include follicular hyperkeratosis, various degrees of perifollicular inflammation and atrophic hair follicles. The primary defects appear to be an abnormal keratinisation of the upper part of the hair follicle leading to follicular plugs which may obstruct the growing hair shaft to produce a chronic inflammatory infiltrate with scarring below that level.5 An abnormal keratinisation may precede follicular dystrophy in and around the pilosebaceous follicle. Keratosis pilaris (MIM 604093) with follicular plugging and atrophy of hair follicles of extensor areas of the limbs is a common associated finding.2 4 The genetics behind KPA is unclear and most reported cases are sporadic. However, a few familial cases have been reported consistent with either autosomal dominant, autosomal recessive or X linked inheritance.2–4 6 7 No gender or ethnical predilection has been reported for the disease. Here, we report findings from a large consanguineous family segregating isolated and autosomal recessive KPA of a mixed type and with similarities to both AV and KFSD. Furthermore, we show that a missense mutation in a highly conserved residue of the extracellular domain of low-density lipoprotein (LDL) receptor-related protein 1 (LRP1) segregates with the disease. The gene variant is associated with reduced levels of the protein as well as with reduced cellular uptake of α2-macroglobulin (α2M). Based on the role of LRP1 in inflammation, we suggest a pathogenic mechanism in our family caused by LRP1-mediated dysregulation of the inflammatory response.
Klar J, et al. J Med Genet 2015;0:1–8. doi:10.1136/jmedgenet-2014-102931
Copyright Article author (or their employer) 2015. Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from http://jmg.bmj.com/ on July 7, 2015 - Published by group.bmj.com
New loci METHODS Patients Members of an extended Pakistani family were referred to National Institute for Biotechnology and Genetic Engineering (NIBGE) for clinical investigation. After the recruitment of family members to the study, we obtained venous blood samples for DNA extraction from 13 family members and skin biopsies from two affected individuals and two unrelated, age-matched and healthy individuals. Affected individuals were evaluated by a dermatologist.
Genetic analysis Autozygosity mapping was performed using the GeneChip Mapping 250K Array. Analysis and sorting of homozygous genomic regions were performed as described previously with the dedicated software AutoSNPa.8 9 A cut-off of >127 consecutive homozygous SNPs was used for the selection of candidate regions. Two-point logarithm of odds (LOD) scores were calculated using the MLINK program,10 assuming an autosomal recessive inheritance, equal male to female recombination rate, full penetrance and a disease allele frequency of 0.00001. Equal allele frequencies of the genotyped markers were used in the calculations. Whole exome sequencing (WES) was performed on 50 ng of genomic DNA from two affected individuals. Fragment libraries were created from the sheared samples using the AB Library Builder System (Life Technologies) and target enrichment was performed according to the manufacturer’s protocols (Agilent SureSelect Human All Exon v4 kit). Captured DNA was amplified followed by emulsion PCR (EZ Bead System, Life Technologies) and single-end sequenced on the SOLiD5500xl system, generating over 100 million reads of 75 bp length for each of the samples. Alignment of reads to the human reference sequence (hg19 assembly) and variant detection was performed using v2.1 of the LifeScope software. SNPs and indel data were stored in an in-house exome database together with variant annotation information obtained from ANNOVAR11 and dbSNP135. Custom R scripts were used to identify potentially damaging variants that were shared between the patients while not present in any of the other ∼900 exomes in the in-house database. Segregation analysis of polymorphic markers on chromosomes 3, 10, 11 and 12 as well as the LRP1 variant c.3734A>G was performed using Sanger sequencing. To investigate a possible effect of the variant on exonic splicing regulatory sequences, we used the two on-line prediction tools, Rescue Exonic Splice Enhancers (ESE) (http://genes.mit.edu/ burgelab/rescue-ese/) and Human Splicing Finder (http://www. umd.be/HSF3/), respectively.12 13 Fibroblasts of two patients and the two age-matched control individuals were cultured in parallel under similar conditions in dulbecco’s modified eagle medium (DMEM) and 10% feta calf serum (FCS). The cells were harvested between passages 6 and 10. Total RNA was extracted using the PureLink RNA Mini Kit (Invitrogen), treated with DNA-free (Ambion) and then converted into cDNA by using RevertAid H Minus First Strand cDNA Synthesis Kit (Fermentas). Quantitative real-time PCR was performed with the Platinum SYBR Green qPCR SuperMix-UDG Kit (Invitrogen) and the MxPro Real-Time PCR System (Stratagene). The primers for LRP1 mRNA quantification were designed to amplify exon–intron boundaries between exons 14 and 15 (c.2772-c.2904) and exons 35 and 36 (c.6217-c.6330), respectively, to ensure that the amplicons were cDNA specific. All reactions were performed three times and in triplicates and normalised to β-actin. Student’s two-tailed t test 2
was used for statistical analysis. Primer sequences used to amplify the LRP1 cDNA and LRP1 exon 23 on genomic DNA were designed using Primer 3 Plus software ( primer3plus.com/) and are available upon request.
Immunohistochemistry, fluorescent immunostaining and confocal microscopy Formalin-fixed paraffin-embedded skin punch biopsy from the cheek of an affected family member was sectioned and treated according to standard procedures. Staining was performed with a mouse monoclonal anti-LRP1 (dilution 1:200, L2420, Sigma-Aldrich). Intracellular LRP1 and α2M protein levels were analysed and quantified in fibroblasts from two affected individuals and two age-matched controls at passage 8–12 using a Laser scanning Microscope (LSM510, ZEISS). Cellular uptake of α2M was determined according to Galliano et al,14 after seeding of 10 000 fibroblasts per well from patients and controls into Lab-Tek-II chamber slides coated with fibronectin. Cells were serum starved in OptiMEM medium (Life Technologies). Next day, to assay LRP1-mediated uptake of α2M, fresh OptiMEM supplemented with 10 ng/mL hrPDGF-BB (PeproTech) and 4 nM α2MA (R&D Systems) was added and cells were incubated for 45 min at 37°C. Cells were washed twice in ice-cold dulbecco’s phosphate buffered saline (DPBS) and fixed in 4% paraformaldehyde (PFA). Primary antibodies used for immunostaining and quantification were mouse anti-LRP1 (clone 8G1; Progen Biotechnik) and goat anti-α2M (R&D Systems). The secondary antibodies were anti-mouse IgG AlexaFluor488 (Invitrogen) and anti-goat IgG AlexaFluor633 (Invitrogen). Cells were imaged and high-throughput image analysis was performed using the CellProfiler system software.15 Attempts to transfect the w.t. LRP1-EGFP expressing construct into fibroblast cell lines from patients and controls were performed with the FuGENE 6 reagent (Roche, Basel).16
Statistical analysis of immunostaining For analysis of images of LRP1 and α2M stainings, we used a generalised linear model in the statistical software R (versions 3.0.1) to determine the relative amounts of LRP1 in the experiments.17 When modelling differences between patients and controls, we adjusted for a significant interexperimental variation (p127 consecutive homozygous SNPs. The regions were on chromosome 3 (hg19:86062188—98031540), chromosome 10 (hg19:35255101—71974469), chromosome 11 (hg19: 17627613—24853326) and chromosome 12 (hg19:63359011 —67072557). These selected regions were further investigated by genotyping of available family members over three generations with microsatellite markers derived from polymorphic repeats. Haplotype and segregation analysis excluded three out of the four homozygous regions, whereas homozygosity for the chromosome 12q region was confirmed in all four affected individuals (figure 1A) with a maximum two-point LOD score of 2.55 (MLINK).10 The homozygous chromosome 12q region contains 1093 consecutive homozygous SNPs spanning a 12 Mb region and 215 protein-coding genes. Considering the large number of genes in the autozygous region, we then performed WES on DNA samples from two affected family members (ind. IV-6 and IV-8). An average of 97% of the exonic baits was covered at least 1×, and 90% were covered >10×. Common variants were excluded by filtering against dbSNP135 (mean allele frequency (MAF) >0.01) and 900 in house exomes. Within the candidate region on chromosome 12q, only one homozygous variant could be identified from the exome sequencing data. The missense variant is a c.3734A>G (K1245) transition in exon 23 of LRP1 (NM_002332.2; MIM107770) encoding the LRP1. Segregation analysis using Sanger sequencing confirmed homozygosity for the missense variant in the four affected family members, whereas unaffected first-degree relatives were heterozygous or homozygous for the wild-type allele (figures 1A and 2A). The disease haplotype was on one side traced back to ind. II:1 who is related to her husband through the grandparents of ind. I:2. The missense variant was excluded in 200 Swedish and 200 Pakistani control chromosomes and it was not present in LRP1 sequences from 900 in house exomes, the dbSNP (http://www.ncbi.nlm.nih.gov/SNP/) or the EVS data release (ESP6500SI-V2) on the Exome Variant Server, NHLBI GO Exome Sequencing Project (ESP), Seattle, Washington, USA (URL: http://evs.gs.washington.edu/EVS/) when accessed November 2014. It was further excluded in the 60552 exome sequenced individuals in the Exome Aggregation Consortium (ExAC) data (http://exac.broadinstitute.org/). The LRP1 residue K1245 is located in the sixth epidermal growth factor (EGF)-like domain (SMART: SM00181; a.a. 1227-1263; figure 2B). The entire EGF-like domain is highly conserved and K1245 is completely conserved along the phylogenic scale from human to zebrafish (PhyloP score 2.0 GERP score 4.9; figure 2C). Furthermore, the amino acid change is predicted to be damaging by both PolyPhen2 and SNP&GO. This suggests that the K1245 residue plays a crucial functional role for the protein. Furthermore, the variant is located in an 4
ESE site with the sequence gtgaAg or tgaAgt (the variant position is capitalised) according to both prediction tools used (Rescue ESE and Human Splicing Finder, respectively). This is predicted to result in a loss of the ESE according to both prediction tools.
mRNA analyses, image-based quantification of LRP1 protein and ligand uptake To investigate the functional significance of the missense mutation, we first quantified the expression of LRP1 mRNA in fibroblast cultures from two affected family members (ind. IV-1 and ind. IV-8) as well as from two aged-matched control individuals using real-time qRT-PCR. The results indicated a fivefold reduction in LRP1 mRNA expression in patient-derived fibroblasts when compared with control fibroblasts (figure 3A). Next, we quantified the relative LRP1 protein amounts in fibroblast cell lines from patients and controls using immunostaining and fluorescence confocal microscopy. The intracellular LRP1 protein contents were determined in patient cells (n=1660) and control cells (n=2808), respectively, using the CellProfiler analysis.15 The analysis confirmed significantly reduced LRP1 levels in cells of affected individuals when compared with controls (p
JMG Online First, published on July 3, 2015 as 10.1136/jmedgenet-2014-102931
New loci
ORIGINAL ARTICLE
Whole exome sequencing identifies LRP1 as a pathogenic gene in autosomal recessive keratosis pilaris atrophicans Joakim Klar,1 Jens Schuster,1 Tahir Naeem Khan,2 Muhammad Jameel,2 Katrin Mäbert,1 Lars Forsberg,1 Shehla Anjum Baig,3 Shahid Mahmood Baig,2 Niklas Dahl1 1
Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Biomedical Centre, Uppsala, Sweden 2 Human Molecular Genetics Laboratory, Health Biotechnology Division, National Institute for Biotechnology and Genetic Engineering (NIBGE), Pakistan Institute of Engineering and Applied Sciences (PIEAS), Faisalabad, Pakistan 3 Department of Pathology, Children’s Hospital, Pakistan Institute of Medical Sciences, (PIMS), Islamabad, Pakistan Correspondence to Dr Niklas Dahl, Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, BMC, Box 815, Uppsala 752 37, Sweden; [email protected] JK and JS contributed equally. Received 12 December 2014 Revised 27 May 2015 Accepted 14 June 2015
ABSTRACT Background Keratosis pilaris atrophicans (KPA) is a group of rare genodermatoses characterised by perifollicular keratosis and inflammation that progresses to atrophy and scars of the facial skin. Keratosis pilaris of extensor areas of limbs is a common associated finding. Most cases with KPA are sporadic and no consistent inheritance pattern has been documented. Methods A large consanguineous Pakistani pedigree segregating autosomal recessive KPA of a mixed type was subject to autozygosity mapping and whole exome sequencing. Quantification of mRNA and protein levels was performed on fibroblasts from affected individuals. Cellular uptake of the low-density lipoprotein (LDL) receptor-related protein 1 (LRP1) ligand α2macroglobulin (α2M) was quantified using fluorescence confocal microscopy. Results Genetic analyses identified a unique homozygous missense variant (K1245R) in the LRP1 in all affected family members. LRP1 encodes the LRP1, a multifunctional cell surface receptor with endocytic functions that belongs to the LDL receptor family. The LRP1 mRNA and LRP1 protein levels in fibroblasts of affected individuals were markedly reduced when compared with controls. Similarly, the LRP1-mediated cellular uptake of α2M was reduced in patient fibroblasts. Conclusions This is the first report on LRP1 as a pathogenic gene for autosomal recessive KPA and keratosis pilaris. The inflammatory characteristics of the KPA entity in our family suggest a link to the immuneregulatory functions of LRP1.
INTRODUCTION
To cite: Klar J, Schuster J, Khan TN, et al. J Med Genet Published Online First: [please include Day Month Year] doi:10.1136/ jmedgenet-2014-102931
Keratosis pilaris atrophicans (KPA) is a small group of follicular syndromes characterised by inflammation and atrophy.1 The morphological hallmarks are variable degrees of perifollicular inflammation, secondary scarring and/or alopecia. Three clinical entities of KPA are described: keratosis pilaris atrophicans faciei (KPAF), keratosis follicularis spinulosa decalvans (KFSD) and atrophoderma vermiculatum (AV).1–3 However, affected individuals may present with features that overlap between the three entities. The onset of KPAF is usually in infancy with erythema, follicular keratotic papules on the lateral parts of the eyebrows that
may extend to the forehead as well as scarring alopecia.2 KFSD also presents in infancy with papules primarily in the malar area, patchy scarring alopecia of the scalp as well as the absence of eyebrows and eyelashes.3 AV (MIM 209700) is considered a more severe form of KPA with a typical onset in late childhood or adolescence. The AV entity is characterised by erythema and follicular keratotic papules located on the cheeks, preauricular regions and forehead that may extend to the lips and earlobes.4 The papules progress to pitted atrophic and depressed scars in a reticular or honeycomb pattern. Eyebrows, eyelashes and scalp are usually spared but eyebrows may sometimes be involved. The course of the three KPA entities is usually slowly progressive with poor response to treatment. Typical histopathological findings in KPA include follicular hyperkeratosis, various degrees of perifollicular inflammation and atrophic hair follicles. The primary defects appear to be an abnormal keratinisation of the upper part of the hair follicle leading to follicular plugs which may obstruct the growing hair shaft to produce a chronic inflammatory infiltrate with scarring below that level.5 An abnormal keratinisation may precede follicular dystrophy in and around the pilosebaceous follicle. Keratosis pilaris (MIM 604093) with follicular plugging and atrophy of hair follicles of extensor areas of the limbs is a common associated finding.2 4 The genetics behind KPA is unclear and most reported cases are sporadic. However, a few familial cases have been reported consistent with either autosomal dominant, autosomal recessive or X linked inheritance.2–4 6 7 No gender or ethnical predilection has been reported for the disease. Here, we report findings from a large consanguineous family segregating isolated and autosomal recessive KPA of a mixed type and with similarities to both AV and KFSD. Furthermore, we show that a missense mutation in a highly conserved residue of the extracellular domain of low-density lipoprotein (LDL) receptor-related protein 1 (LRP1) segregates with the disease. The gene variant is associated with reduced levels of the protein as well as with reduced cellular uptake of α2-macroglobulin (α2M). Based on the role of LRP1 in inflammation, we suggest a pathogenic mechanism in our family caused by LRP1-mediated dysregulation of the inflammatory response.
Klar J, et al. J Med Genet 2015;0:1–8. doi:10.1136/jmedgenet-2014-102931
Copyright Article author (or their employer) 2015. Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from http://jmg.bmj.com/ on July 7, 2015 - Published by group.bmj.com
New loci METHODS Patients Members of an extended Pakistani family were referred to National Institute for Biotechnology and Genetic Engineering (NIBGE) for clinical investigation. After the recruitment of family members to the study, we obtained venous blood samples for DNA extraction from 13 family members and skin biopsies from two affected individuals and two unrelated, age-matched and healthy individuals. Affected individuals were evaluated by a dermatologist.
Genetic analysis Autozygosity mapping was performed using the GeneChip Mapping 250K Array. Analysis and sorting of homozygous genomic regions were performed as described previously with the dedicated software AutoSNPa.8 9 A cut-off of >127 consecutive homozygous SNPs was used for the selection of candidate regions. Two-point logarithm of odds (LOD) scores were calculated using the MLINK program,10 assuming an autosomal recessive inheritance, equal male to female recombination rate, full penetrance and a disease allele frequency of 0.00001. Equal allele frequencies of the genotyped markers were used in the calculations. Whole exome sequencing (WES) was performed on 50 ng of genomic DNA from two affected individuals. Fragment libraries were created from the sheared samples using the AB Library Builder System (Life Technologies) and target enrichment was performed according to the manufacturer’s protocols (Agilent SureSelect Human All Exon v4 kit). Captured DNA was amplified followed by emulsion PCR (EZ Bead System, Life Technologies) and single-end sequenced on the SOLiD5500xl system, generating over 100 million reads of 75 bp length for each of the samples. Alignment of reads to the human reference sequence (hg19 assembly) and variant detection was performed using v2.1 of the LifeScope software. SNPs and indel data were stored in an in-house exome database together with variant annotation information obtained from ANNOVAR11 and dbSNP135. Custom R scripts were used to identify potentially damaging variants that were shared between the patients while not present in any of the other ∼900 exomes in the in-house database. Segregation analysis of polymorphic markers on chromosomes 3, 10, 11 and 12 as well as the LRP1 variant c.3734A>G was performed using Sanger sequencing. To investigate a possible effect of the variant on exonic splicing regulatory sequences, we used the two on-line prediction tools, Rescue Exonic Splice Enhancers (ESE) (http://genes.mit.edu/ burgelab/rescue-ese/) and Human Splicing Finder (http://www. umd.be/HSF3/), respectively.12 13 Fibroblasts of two patients and the two age-matched control individuals were cultured in parallel under similar conditions in dulbecco’s modified eagle medium (DMEM) and 10% feta calf serum (FCS). The cells were harvested between passages 6 and 10. Total RNA was extracted using the PureLink RNA Mini Kit (Invitrogen), treated with DNA-free (Ambion) and then converted into cDNA by using RevertAid H Minus First Strand cDNA Synthesis Kit (Fermentas). Quantitative real-time PCR was performed with the Platinum SYBR Green qPCR SuperMix-UDG Kit (Invitrogen) and the MxPro Real-Time PCR System (Stratagene). The primers for LRP1 mRNA quantification were designed to amplify exon–intron boundaries between exons 14 and 15 (c.2772-c.2904) and exons 35 and 36 (c.6217-c.6330), respectively, to ensure that the amplicons were cDNA specific. All reactions were performed three times and in triplicates and normalised to β-actin. Student’s two-tailed t test 2
was used for statistical analysis. Primer sequences used to amplify the LRP1 cDNA and LRP1 exon 23 on genomic DNA were designed using Primer 3 Plus software ( primer3plus.com/) and are available upon request.
Immunohistochemistry, fluorescent immunostaining and confocal microscopy Formalin-fixed paraffin-embedded skin punch biopsy from the cheek of an affected family member was sectioned and treated according to standard procedures. Staining was performed with a mouse monoclonal anti-LRP1 (dilution 1:200, L2420, Sigma-Aldrich). Intracellular LRP1 and α2M protein levels were analysed and quantified in fibroblasts from two affected individuals and two age-matched controls at passage 8–12 using a Laser scanning Microscope (LSM510, ZEISS). Cellular uptake of α2M was determined according to Galliano et al,14 after seeding of 10 000 fibroblasts per well from patients and controls into Lab-Tek-II chamber slides coated with fibronectin. Cells were serum starved in OptiMEM medium (Life Technologies). Next day, to assay LRP1-mediated uptake of α2M, fresh OptiMEM supplemented with 10 ng/mL hrPDGF-BB (PeproTech) and 4 nM α2MA (R&D Systems) was added and cells were incubated for 45 min at 37°C. Cells were washed twice in ice-cold dulbecco’s phosphate buffered saline (DPBS) and fixed in 4% paraformaldehyde (PFA). Primary antibodies used for immunostaining and quantification were mouse anti-LRP1 (clone 8G1; Progen Biotechnik) and goat anti-α2M (R&D Systems). The secondary antibodies were anti-mouse IgG AlexaFluor488 (Invitrogen) and anti-goat IgG AlexaFluor633 (Invitrogen). Cells were imaged and high-throughput image analysis was performed using the CellProfiler system software.15 Attempts to transfect the w.t. LRP1-EGFP expressing construct into fibroblast cell lines from patients and controls were performed with the FuGENE 6 reagent (Roche, Basel).16
Statistical analysis of immunostaining For analysis of images of LRP1 and α2M stainings, we used a generalised linear model in the statistical software R (versions 3.0.1) to determine the relative amounts of LRP1 in the experiments.17 When modelling differences between patients and controls, we adjusted for a significant interexperimental variation (p127 consecutive homozygous SNPs. The regions were on chromosome 3 (hg19:86062188—98031540), chromosome 10 (hg19:35255101—71974469), chromosome 11 (hg19: 17627613—24853326) and chromosome 12 (hg19:63359011 —67072557). These selected regions were further investigated by genotyping of available family members over three generations with microsatellite markers derived from polymorphic repeats. Haplotype and segregation analysis excluded three out of the four homozygous regions, whereas homozygosity for the chromosome 12q region was confirmed in all four affected individuals (figure 1A) with a maximum two-point LOD score of 2.55 (MLINK).10 The homozygous chromosome 12q region contains 1093 consecutive homozygous SNPs spanning a 12 Mb region and 215 protein-coding genes. Considering the large number of genes in the autozygous region, we then performed WES on DNA samples from two affected family members (ind. IV-6 and IV-8). An average of 97% of the exonic baits was covered at least 1×, and 90% were covered >10×. Common variants were excluded by filtering against dbSNP135 (mean allele frequency (MAF) >0.01) and 900 in house exomes. Within the candidate region on chromosome 12q, only one homozygous variant could be identified from the exome sequencing data. The missense variant is a c.3734A>G (K1245) transition in exon 23 of LRP1 (NM_002332.2; MIM107770) encoding the LRP1. Segregation analysis using Sanger sequencing confirmed homozygosity for the missense variant in the four affected family members, whereas unaffected first-degree relatives were heterozygous or homozygous for the wild-type allele (figures 1A and 2A). The disease haplotype was on one side traced back to ind. II:1 who is related to her husband through the grandparents of ind. I:2. The missense variant was excluded in 200 Swedish and 200 Pakistani control chromosomes and it was not present in LRP1 sequences from 900 in house exomes, the dbSNP (http://www.ncbi.nlm.nih.gov/SNP/) or the EVS data release (ESP6500SI-V2) on the Exome Variant Server, NHLBI GO Exome Sequencing Project (ESP), Seattle, Washington, USA (URL: http://evs.gs.washington.edu/EVS/) when accessed November 2014. It was further excluded in the 60552 exome sequenced individuals in the Exome Aggregation Consortium (ExAC) data (http://exac.broadinstitute.org/). The LRP1 residue K1245 is located in the sixth epidermal growth factor (EGF)-like domain (SMART: SM00181; a.a. 1227-1263; figure 2B). The entire EGF-like domain is highly conserved and K1245 is completely conserved along the phylogenic scale from human to zebrafish (PhyloP score 2.0 GERP score 4.9; figure 2C). Furthermore, the amino acid change is predicted to be damaging by both PolyPhen2 and SNP&GO. This suggests that the K1245 residue plays a crucial functional role for the protein. Furthermore, the variant is located in an 4
ESE site with the sequence gtgaAg or tgaAgt (the variant position is capitalised) according to both prediction tools used (Rescue ESE and Human Splicing Finder, respectively). This is predicted to result in a loss of the ESE according to both prediction tools.
mRNA analyses, image-based quantification of LRP1 protein and ligand uptake To investigate the functional significance of the missense mutation, we first quantified the expression of LRP1 mRNA in fibroblast cultures from two affected family members (ind. IV-1 and ind. IV-8) as well as from two aged-matched control individuals using real-time qRT-PCR. The results indicated a fivefold reduction in LRP1 mRNA expression in patient-derived fibroblasts when compared with control fibroblasts (figure 3A). Next, we quantified the relative LRP1 protein amounts in fibroblast cell lines from patients and controls using immunostaining and fluorescence confocal microscopy. The intracellular LRP1 protein contents were determined in patient cells (n=1660) and control cells (n=2808), respectively, using the CellProfiler analysis.15 The analysis confirmed significantly reduced LRP1 levels in cells of affected individuals when compared with controls (p
