Supplementary data

Patients and Methods Case index

At evaluation of clinical data collected during admissions made for epilepsy in the infancy of the index case (III-11), we observed that every episode of unconsciousness was associated with hypoglycemia lower than 50 mg/dl. Similar symptoms had been observed since the first months of life in her sister III-12. Both subjects, however, showed a normal insulin response after glucose load, thus excluding juvenile diabetes. Repeated inter-ictal electroencephalography (EEG) was inconsistent for epilepsy. Anti-epileptic drugs were removed and a balanced diet was started, thus preventing further episodes of hypoglycemia. Interestingly, III-12, at age 23, suffered from two novel episodes of loss of consciousness associated with hypoglycemia, after intake of alcoholic drinks with high sugar content. Blood analyses revealed a substantial increase of baseline levels of cortisol with insulin level within the normal range during the 36-48 h after each episode.

The index case underwent a study protocol which included blood and urine analyses (comprising in particular dosage of vitamin D, muscle enzymes, markers of inflammatory, rheumatic and autoimmune and celiac diseases, thyroid, parathyroid and pancreas function, and research of occult blood in the stool), ophthalmologic examination comprising fundus, mesopic vision test and electroretinogram, cardiology with electrocardiogram (ECG) and cardiac ultrasound evaluation, entire skeletal X-ray, basic EEG with activation test and brain magnetic resonance imaging (MRI).

The most relevant pathological findings were creatine phosphokinase (CPK) increase at 250 IU/L

(n.v.=190IU/L), erythrocyte sedimentation rate (ESR), and high sensitivity C-reactive protein (hs-

CRP) increased at 40 mm/h (n.v.=3-20 mm/h) and 6 mg/L (n.v.<2mg/L) ,respectively. Fasting blood glucose ranging 95-105 mg/dl (n.v.=60-110mg/dl) in repeated measurements was reported.

Sideremia at 30-35 g/dl (n.v.= 37-147g/dl), 1.25-OH vitamin D lower than 20 ng/ml (n.v.=30-

100 ng/ml) and vitamin B12 lower than 180 ng/L (n.v.=180-914 ng/L) were also found at repeated

1 measurements. Plasma glucose and insulin levels after glucose load were normal. TSH increase at

6.1 mIU/L (n.v.=0.5-4.0 mIU/L) with normal levels of T3 and T4. Thyroid sonography demonstrated goiter with micro nodules.

Blood platelet aggregation and secretion tests were conducted to evaluate the presence of coagulation defects that could explain the wound healing alteration and bleeding observed in the affected individual. A physiological activation of platelets was observed excluding the presence of a platelet disorder.

At clinical history, 18 out of 21 members of the family reported one or more symptoms similar to those complained by the index case (Table S1). A clinical and instrumental study protocol was proposed like the one conducted in the proband. This latter and all the individuals who accepted the investigation have been followed up for twenty years by a team of specialists, including rheumatologist, ophthalmologist, hematologist, endocrinologist, orthopedist, abdominal surgeon and cardiologist. The following procedures have been conducted when appropriate, after signed informed consent. Muscle biopsies of the vastus lateralis of quadriceps and skin biopsies from the lateral aspect of the right thigh were processed according to standard protocols for histology, immunofluorescence (IF), biochemistry and expression studies. Skin specimens served also to establish fibroblast primary cell cultures. Peripheral venous blood samples were collected to conduct platelet aggregation and secretion studies and genomic DNA and RNA extraction [37].

Muscle and skin samples of 6 unrelated females (age ranging 25-50 years) from the tissues bank of our Department, as resulting normal at microscopy and biochemistry, were used as control tissues.

The study was approved by ethics review committee at the relevant Institution and all participants provided informed consent.

Gene identification strategy

2

To identify the responsible gene we used a combined approach of whole-exome sequencing (WES) and total blood RNA sequencing to detect mutations in coding regions, exon-intron splicing sites, untranslated regions (UTRs) as well as large insertions or deletions which could generate aberrant transcripts.

For WES strategy, we analyzed three affected samples from the pedigree (IDs II-7, III-4 and III-11).

In particular, the affected family members II-7 and III-11 manifested the complete phenotype; instead, the individual III-4 was not affected by severe myopia (Table S1). Genomic DNA was isolated from whole blood, using the QIAamp DNA Blood Midikit (Qiagen), according to the manufacturer’s protocol. Exonic regions of genomic DNA were enriched using the Agilent

Haloplex Exome kit based on DNA digestion and capture. Exomes were barcoded and sequenced at multiple sites on the Illumina HiSeq1000 platform. Average coverage for all the experiments was

70x and at least 20x for 89% of the target. Paired sequencing reads were aligned to the reference genome (UCSC, hg19 build) using BWA and sorted with SAMtools and Picard. Post-alignment processing (local realignment around insertions-deletions and base recalibration), SNV, and small insertions-deletions (ins-del) calling were performed with Genome Analysis Toolkit (GATK) with parameters adapted to the haloplex-generated sequences. The called SNV and ins-del variants produced with both platforms were annotated using ANNOVAR [38-41]. In a first step, we performed a thorough survey of all previously identified genes associated with connective disorders, in order to definitively exclude their involvement in the disease phenotype. Sequencing data were analyzed as in our previous works [42, 43]. Data were filtered to eliminate common variants (MAF > 1%), neutral variants, variants with low quality score, and variants not shared by all analyzed affected subjects, when covered. Additional frequency filters were used by comparing internal databases of whole exome sequencing data (n = 300). Prioritization was also made based on

MAF frequency. Considering a dominant mode of inheritance, we matched data from the three affected family members to obtain sixteen shared variants and in a second step we only compared the affected individuals II-7 and III-11 showing the severe myopia and III-4 was considered as 3 healthy. Only the c.9418G>A change, causing the missense variant p.V3140M in the LAMA5 gene, showed complete segregation with the disease in the family, was not found in 600 chromosomes from unrelated healthy subjects from the same geographical origin and was reported in Ensembl database with the rs number “rs369572769” and a minor allele frequency (MAF)<0.00007

(ESP6500: 1 allele A/8581 alleles G; EXAC: 5A/120359G). We analysed the p.V3140M variant with SIFT, PolyPhen2 and PMut software and confirmed a deleterious effect of this variant (Tables

S2 and S3).

The severe myopia was associated with a novel mutation in a different gene (manuscript in preparation), thus excluding this defect from the syndrome manifestations.

RNA sequencing strategy. Blood total RNA from the affected family members II-7, III-4 and III-11 and from two unrelated controls was isolated by using the Tempus Spin RNA Isolation Kit (Life

Technology), according to manufacturer’s protocol. Integrity was assessed by Experion (RNA

StdSens Chip, Bio-Rad, Hercules, CA, USA). Total RNA was reverse transcribed using

SuperScriptTM II Reverse transcriptase (Life Technology) according to manufacturer’s protocol. cDNA libraries were prepared using TruSeqTM RNA Sample Preparation kit (Illumina, San Diego,

CA, USA) and paired-end reads were sequenced on Illumina HiSeq2000 platform. Mapping was performed using TopHat v2.0. Gene expression and alternative splicing were analyzed using

Cufflinks v2.1.1 [44]. RNA sequencing confirmed the presence of the LAMA5 mutation and excluded the presence of gene rearrangements and mutations producing altered RNA processing both in LAMA5 and other genes.

RNA sequencing data were analyzed to assess gene specific mRNA amount, by comparing data between cases and controls to search for those genes showing a difference >1.5-fold change. By using the “Advanced search tool” of the Gene Cards database was performed an in silico analysis to retrieve a list of genes encoding proteins involved in inflammation and wound healing (Tables S4 and S5).

4

Mutation analysis

Validation and segregation analysis of the selected variants was performed by Sanger sequencing.

PCR amplification and direct sequencing protocols have been previously described [42, 43].

LAMA5 oligonucleotides for mutation analysis were:

LAMA5 F: 5’ - TGGGCAAGTATGTGGACCTC - 3’

LAMA5 R: 5’ - ACTCATTCCAGACACCCCAG - 3’

Fibroblast and myoblast primary cell cultures

Skin biopsies were obtained from the affected family members (II-7, II-9, III-2, III-4, III-11 and III-

13), from healthy family members III-3 and III-10 and unrelated healthy individuals. The tissues pieces were enzymatically digested for 24 hours at 37°C by using Collagenase/Dispase kit (Roche), followed by a mechanically disaggregation with knife. Cells were grown in Dulbecco’s modified

Eagle’s medium (DMEM) supplemented with 4.5 g/L D-Glucose, 0.11 g/L Sodium Pyruvate, 10%

FBS, 1% NEAA, 1% L-glutamine, Gentamycin 1:200, 1% Pen-Strep and 1% Amphotericin B at

37°C in a humidified CO2 5% air for about 4-5 weeks. Fibroblast cell lines were maintained and were used in all the experiments in sub-confluent mono-layers.

LAMA5 synthetic peptides design

The LG modules of the laminin alpha chains consist of 14 β strands labeled A−N arranged in two sheets resulting in a  sandwich structure (figure S1B and C) [45]. Recently have been showed that short synthetic peptides (less than 20 amino acids long) designed on E and F  strands of laminin alpha chains, manifest biological activities being able to promote cell attachment and to bind specific receptors [22]. Based on these previous studies, we synthesized two short LAMA5 synthetic peptides designed on the B  strand which houses the V3140M mutation (China Peptides

Co., Ltd) (figure S1F). Specifically, the two peptides labeled A5G3B-V and A5G3B-M, 13 amino acids long, cover the B β strand and the B-C connecting loop from residues 3131 to 3143. The

5

A5G3B-V peptide carries the 3140 valine and the A5G3B-M peptide carries the 3140 methionine.

The amino acid sequence of each peptide is reported:

A5G3B-3140V :3131 GFLRLALSNVAPL 3143

A5G3B-3140M :3131 GFLRLALSNMAPL 3143

Cell adhesion assay

Two soluble synthetic peptides, designed on the LG3 domain of the laminin alpha 5, were used to perform the cell attachment assay. Peptides diluted in phosphate buffer saline (PBS) 1x were coated at growing concentrations (2 µg, 5 µg and 10 µg) to the tissue culture 24-well plates by incubating overnight at 4°C. The plates were air-dried before use and blocked with 1% heat-denatured bovine serum albumin (BSA, Sigma) in DMEM at 37°C for 1h and then washed two times with 0.1% BSA in DMEM. Equal numbers of fibroblast cells, derived from the affected and healthy members of the pedigree, were plated and grown in DMEM supplemented with 4.5 g/L D-Glucose, 0.11 g/L

Sodium Pyruvate, 10% FBS, 1% NEAA, 1% L-glutamine, Gentamycin 1:200, 1% Pen-Strep and

1% Amphotericin B at 37°C for 24 h in 5% CO2. The medium was removed by aspiration and the adherent cells were fixed with 4% paraformaldehyde (PFA) and stained with 0.5% crystal violet solution. The adherent cells were photographed using a Leica MZFLIII Stereo Microscope. Cells were quantified measuring the absorbance at 540 nm in each plate well. The assay was carried out in triplicate and each experiment was repeated at least three times.

Cytokines and peptides treatment

Fibroblast cells were seeded from passage 2 to passage 6 in 60 mm plates. Before each treatment the cells were washed with PBS 1x and starved for 24 h in DEMEM, without FBS, supplemented with 1% L-glutamine, 1% Penicillin/Streptomycin, Gentamicin 1:200, 1% Amphotericin B. For cytokines treatment, cultures were stimulated with recombinant human IL1β (10ng/ml, Peprotech,

UK) or recombinant human TGFβ1 (10ng/ml, Peprotech, UK) for 48 h. Each experiment was performed in triplicate. For LAMA5 specific peptides treatment, cultures were stimulated with 6 increasing amount (10 ng/ml, 100 ng/ml, 1 g/ml, 10 g/ml) of A5G3B-V and A5G3B-M peptides for 48 h together or separately. Total RNA was extracted and used for qPCR assays, as described in expression studies.

Expression studies

Total RNAs from tissues and cell lines were isolated by using TRIreagent® (Sigma-Aldrich) protocol. 1 µg of total RNA was reverse transcribed with the RevertAid RT Reverse Transcription

Kit (Thermo Fisher). qPCR reactions were performed in triplicate, as described in previous works

[42, 43], using gene specific primers and sybr select master mix for CFX (Life Technology) following the manufacturer’s directions. qPCR expression values of the specific genes were normalized versus the expression of the glyceraldehydes-3-phosphate dehydrogenase (GAPDH) gene. Primers used are listed in Table S6.

The mean values ± standard deviations of three independent experiments were analyzed as multiple data sets with ANOVA test for multiple comparisons with Bonferroni's corrections. A value of p<0.05 was considered to be statistically significant.

Protein extraction and Western blot analysis

Fibroblast cells were collected and washed with cold PBS. Cytoplasmic proteins were obtained in

100–1,000 L of hypotonic buffer (10 mM Tris-HCl [pH 7.5], 2mM MgCl2, 10 mM CaCl2, 30 mM

Sucrose, 1% NP40) and 1x proteinase inhibitor mixture (Sigma Aldrich). The lysis was performed with a syringe with a narrow-gauge (no. 27) hypodermic needle and was observed by the addition of a Trypan Blue solution. Then, the disrupted cells were centrifuged for 20 min at 10,000–11,000 rcf. Nuclear extracts were sequentially obtained in 140 L of extraction buffer (0.42 MNaCl, 0.2m

MEDTA, 25% Glycerol, 1% NP40) and 1x proteinase inhibitor mixture. Crude nuclei were shaken gently and then incubated for 10 min on ice. Finally, nuclear extracts were obtained through

7 centrifugation for 5 min at 20,000–21,000 rcf. Conditioned medium proteins (normalized versus the number of cells) were precipitated over night by using pre-cooled acetone. Proteins were recovered after centrifugation at 1500 g at 4° C. Proteins were separated on SDS-PAGE and transferred onto

0.45- m-pore size Immobilon-P membranes (Millipore, Bedford, MA). The expression of LAMA5,

-actin and nucleolin proteins was determined by immunoblot analysis using anti- Laminin alpha5

(sc-20145, Santa Cruz Biotechnology, 1:1000/Goat anti Rabbit HRP-conjugated Biorad, 1:2000); anti-Nucleolin (C23, Santa Cruz Biotechnology, 1:1000/ Goat anti-mouse HRP-conjugated Biorad,

1:2000); anti- β-Actin (sc-47778, Santa Cruz Biotechnology, 1:1000/Goat anti-mouse HRP- conjugated Biorad, 1:2000).

Densitometry analysis was performed using Scion Image (Ver. 4.0.2; Scion Corporation, USA).

The signals obtained for each protein were normalized versus -actin (cytoplasmic proteins), nucleolin (nucleus proteins) and ponceau (medium proteins). The mean ± SD of three independent experiments were analyzed as multiple data sets with ANOVA test for multiple comparisons with

Bonferroni's corrections and were plotted as histogram representation. A value of p<0.05 was considered to be statistically significant.

Immunofluorescence (IF) on fibroblast cells

IF staining was performed on cells seeded on sterile glass coverslips in 24 well plates. At desired confluence, the cells were fixed with 4% paraformaldehyde in PBS pH 7.4 for 20 min at room temperature (RT). To prevent non-specific binding, the cells were blocked with 1% FBS for 1h at

RT then incubated with anti-LAMA5 (sc-20145, Santa Cruz Biotechnology, 1:200)/ Donkey anti-

Rabbit IgG (H+L) Alexa Fluor® 488 conjugate (A21206, Life technology, 1:400) and anti-ITGA6

(sc-6597, Santa Cruz Biotechnology, 1:200)/ Donkey anti-Goat IgG (H+L), Alexa Fluor® 594 conjugate (A-11058, Life technology, 1:400).

8

The nuclei were detected with DAPI (Thermo Scientific, 1:1000), the actin of the cytoskeleton was stained with rhodamine phalloidin (Sigma, 1:100). The images were captured with the Nikon

Confocal/Nis-C software.

Histology and IF on tissues

Hematoxylin and eosin (HE) analysis and IF staining were performed on serial sections of mice and human tissues.

Human tissues. We used 6 µm thick cryostatic sections of skin biopsies from the lateral aspect of the right thigh and skeletal muscle biopsies of the vastus lateralis of quadriceps of three affected family members (II-7, II-9, III-11), one healthy member of the family (III-10) and two unrelated individuals with neither connective tissues diseases nor LAMA5 mutations, matched for age and sex. Cryostatic sections were fixed in pre-cooled acetone. 0,9 µm thick epoxy embedded semithin sections were obtained as described by Fei et al 2009 [46]. The endogenous peroxidase was blocked by treating the sections with 0.3% H2O2 and methanol for 5 min at RT. After, the sections were incubated in a humidified chamber with blocking buffer (BSA/PBS1x) for 1 h and then incubated with anti-Laminin α5 (clone 4C7, MAB1924 Millipore, 1:200)/Sheep anti-Mouse IgG–Cy3 (C2181,

Sigma, 1:200), anti-Integrin α6 (sc-6597, Santa Cruz Biotechnology, 1:200)/Rabbit anti-Goat IgG–

FITC (F2016, Sigma, 1:100), anti- COL4A (sc-59814, Santa Cruz Biotechnology,

1:200)/Fluorescein Horse Anti-Mouse IgG (FI-2000, Vector, 1:200).

HE analysis was performed using standard procedures. Nuclei were detected with DAPI (Thermo

Scientific) at the dilution 1:1000. Positive and negative controls were included for each esperiment.

Tissues images were captured using DMI6000 LEICA inverted microscope/LAS AF software.

Mouse tissues samples. 7-10 µm thick paraffin sections of dorsal skin and small intestine biopsies of 7 knocked-in homozygous Lama5V3144M, 7 heterozygous and 7 wild-type female mice at 28 post natal days were used. The entire small intestine was divided into sections of 1.5 cm which were embedded in paraffin and cut into serial longitudinal sections. Paraffin slides were deparaffinized in 9 xylene and rehydrated in graded alcohols and water at RT. For antigen detection, the slides were immersed in alkaline buffer pH 9 and boiled in microwave vessel for 30 minutes. Then the sections were incubated with blocking buffer (PBS1X/5% serum) for 1 h and with primary and fluorescent secondary antibodies as described for the cryostatic sections. For HE staining, the sections were stained with hematoxylin and eosin and then dehydrated in increasing gradient of alcohols up to xylene. To measure the height, the basal diameter and the density of the villi, a quantitative study was carried out with the program NIS (Nikon LTD) on about 100 sections for wt and mutant animals. The mean ± SD of all measurements were analyzed as multiple data sets with ANOVA test for multiple comparisons with Bonferroni's corrections. A value of p<0.05 was considered to be statistically significant.

Knock-in mouse model generation

The knock-in mouse model was generated by introducing the Lama5 missense mutation c.9430

G>A p.V3144M which is syntenic to the human p.V3140M mutation. The short and long arms of the vector, carrying the mouse genomic region, were assembled through intermediate steps of cloning and sequencing. The nucleotide change was introduced at the end of the long arm of the targeting vector by PCR-based site-directed mutagenesis (Quick Change II XL, Agilent). LoxP- flanked Neo cassette was used as a selectable marker. The vector was sequenced to confirm the mutagenesis and to exclude undesired mutations. The construct was linearized with PacI restriction enzyme and was introduced in the E14Tg2a4 ESC line by electroporation (figure S9). The screening of homologous recombinant ESC clones was performed by PCR, followed by Southern blot to confirm the identification of positive ES cell clones. Mice carrying the knock-in

Lama5V3144M allele were generated from ES cells by standard procedures [47].

Ethics Statements

10

The research involving human participants has been approved by the Institutional Review Board

(IRB) (Università degli Studi della Campania L. Vanvitelli), and clinical investigation has been conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent has been obtained from the participants.

The project for the generation of a Lama5V3144M knock-in mouse model was approved by Italian

Health Ministry, Department for Veterinary, Public Health Nutrition and Food Safety, Directorate-

General for animal health and veterinary medicinal products.

Web Resources

The URLs for data presented herein are as follows: 1000 Genomes, http://browser.1000genomes.org; ExAC Browser, http://exac.broadinstitute.org/; GenBank, http://www.ncbi.nlm.nih.gov/genbank/; OMIM, http://www.omim.org/; UCSC Genome Browser, http://genome.ucsc.edu; Picard, http://broadinstitute.github.io/picard/; SIFT, http://sift.bii.a- star.edu.sg/www/SIFT_seq_submit2.html; PolyPhen2, http://genetics.bwh.harvard.edu/pph2/; Gene

Cards database, http://www.genecards.org; RaptorX, http://www.raptorx.uchicago.edu.

11

References

37. Zhou L, Schmaier AH. Platelet aggregation testing in platelet-rich plasma: description of

procedures with the aim to develop standards in the field. Am J Clin Pathol 2005; 123:172-183.

38. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform.

Bioinformatics 2009; 25:1754-1760.

39. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R;

1000 Genome Project Data Processing Subgroup. The sequence alignment/map format and

SAMtools. Bioinformatics 2009; 25:2078-2079.

40. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, Hartl C, Philippakis AA, del

Angel G, Rivas MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, Sivachenko AY,

Cibulskis K, Gabriel SB, Altshuler D, Daly MJ. A framework for variation discovery and

genotyping using next-generation DNA sequencing data. Nat Genet 2011; 43:491-498.

41. Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from

high-throughput sequencing data. Nucleic Acids Res 2010; 38:e164.

42. Esposito T, Sampaolo S, Limongelli G, Varone A, Formicola D, Diodato D, Farina O,

Napolitano F, Pacileo G, Gianfrancesco F, Di Iorio G. Digenic mutation inheritance of integrin

alpha 7 and myosin heavy chain 7B genes causes congenital myopathy with left ventricular non

compact cardiomyopathy. Orphanet J Rare Dis 2013; 8:91.

43. Esposito T, Lea RA, Maher BH, Moses D, Cox HC, Magliocca S, Angius A, Nyholt DR, Titus

T, Kay T, Gray NA, Rastaldi MP, Parnham A, Gianfrancesco F, Griffiths LR. Unique X-linked

familial FSGS with co-segregating heart block disorder is associated with a mutation in the

NXF5 gene. Hum Mol Genet 2013; 22:3654-3666.

44. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL,

Pachter L. Differential gene and transcript expression analysis of RNA-Seq experiments with

TopHat and Cufflinks. Nat Protoc 2012; 7:562–578.

12

45. Timpl R, Tisi D, Talts JF, Andac Z, Sasaki T, Hohenester E. Structure and function of laminin

LG modules. Matrix Biol 2000;19(4):309-17.

46. Fei L, del Genio G, Rossetti G, Sampaolo S, Moccia F, Trapani V, Cimmino M, del Genio A.

Hiatal hernia recurrence: surgical complication or disease? Electron microscope findings of the

diaphragmatic pillars. J Gastrointest Surg. 2009;13(3):459-64.

47. Acampora D, Mazan S, Lallemand Y, Avantaggiato V, Maury M, Simeone A Brûlet P.

Forebrain and midbrain regions are deleted in Otx2-/- mutants due to a defective anterior

neuroectoderm specification during gastrulation. Development 1995; 121:3279-3290.

13

Legends to supplementary figures

Figure S1. LAMA5 protein structure and modeling of the wt and mutated proteins. (A)

Structure of the laminin alpha 5 chain. The LG globular domains (LG1-LG5), the localization of the p.V3140M mutation in the LG3 domain and the regions of binding with specific receptors (e.g. integrins, dystroglycan, luteran) are reported. (B and C) Modeling of the LG3 domain carrying the

3140V (B) and 3140M (C). The 14-stranded beta-sheet sandwich structure is reported with yellow arrows. Secondary structure modeling of the wild type (3140V) and mutant (3140M) proteins predicts the alteration of the alpha helix structure (in red) and of the linker spaces (gray lines) when the valine is substituted with methionine, and a global derangement of the tridimensional conformation of the entire LG1-LG5 globular domain (D, E). (F) The amino acid sequence of the

LG3 domain of LAMA5 is reported. The 14  strands labeled A-N were predicted with raptorX software (E-strands below the amino acid sequence) and indicated with blue arrows. The peptide is reported with a red bar. The valine 3140 is indicated with arrowhead.

Figure S2. LAMA5 and COL4A expression in human skin. 6 µm thick cryostatic sections of skin biopsies from the lateral aspect of the right thigh of three affected family members (II-7, II-9,

III-11) (20x), one healthy member of the family (III-10) (20x) and two unrelated individuals (Cnt-1 and Cnt-2, 20x) were stained with anti-LAMA5 (MAB1924, clone 4C7, 1:200) and anti-COL4A

(sc-59814, Santa Cruz Biotechnology, 1:200). In affected and healthy individuals the same distribution, intensity and continuity of LAMA5 and COL4A was seen at the dermal-epidermal interface, in the wall of the intradermal blood vessels with cross caliber ranging 10-20 m

(indicated with a white arrow) as well as in basement membranes of hair follicles, sebaceous and exocrine sweat glands (indicated with a white asterisk).

Figure S3. LAMA5 and ITGA6 expression in human skin. 6 µm thick cryostatic sections of skin biopsies from the lateral aspect of the right thigh of the affected individual III-11 (20x) and her healthy cousin (III-10) (20x) were stained with anti-LAMA5 (sc-20145, Santa Cruz Biotechnology,

1:200) and anti-ITGA6 (sc-6597, Santa Cruz Biotechnology, 1:200). Thinning of the epidermis 14 layer (stained with DAPI in blue) with poorly developed dermal papillae (DP) is evident in the proband compared with her healthy cousin. The same distribution, intensity and continuity of

LAMA5 and ITGA6 was observed in both affected and healthy individuals.

Figure S4. LAMA5 and COL4A expression in human skeletal muscle. 6 µm thick cryostatic sections of skeletal muscle biopsies of the vastus lateralis of quadriceps of the affected family member (III-11) and the unrelated control (CNT) were stained with anti-LAMA5 (A-B, 20x, C-D,

40x, control; E, 20x, F-G-H 40x, III-11; MAB1924, clone 4C7, 1:200) and anti-COL4A (I, J, 20x, control; K, 20x, L, 40x III-11; sc-59814, Santa Cruz Biotechnology, 1:200). In affected and healthy individuals the same distribution, intensity and continuity of LAMA5 and COL4A was seen at the endomysium and perimysium and in the wall of the blood vessels. LAMA5 antibody also stains the nuclei (indicated with white asterisk) and muscle fibers in regeneration (indicated with white arrow).

Figure S5. LAMA5 expression in human fibroblast and myoblast cell lines. Immunolocalization of LAMA5 in fibroblast and myoblast cells derived from the affected family member (III-11) and her healthy cousin (III-10). LAMA5 protein was stained with anti-LAMA5 (sc-20145 Santa Cruz

Biotechology, 1:200)/fluorescent secondary antibody (Alexa fluor 488, 1:100). F-actin fibers were stained with rhodamine phalloidin and nuclei were stained with DAPI. The localization of the

LAMA5 protein is indicated by green spots. Intense reaction is visible at the level of the entire nucleus and of cell membrane, while weak and scattered blots of positivity are observed into the cytoplasm. 60x images were acquired by Confocal Microscopy (Nikon).

Figure S6. LAMA5 and ITGA6 expression in human fibroblast cell lines. Immunolocalization of LAMA5 and ITGA6 in fibroblast cells derived from the affected family member (III-11) and her healthy cousin (III-10). LAMA5 protein is stained with anti-LAMA5 (sc-20145 Santa Cruz

Biotechnoly, 1:200)/fluorescent secondary antibody (Alexa fluor 488, 1:100) and is indicated by green spots. ITGA6 protein is stained with anti-ITGA6 (sc-6597 Santa Cruz Biotechnoly,

15

1:200)/fluorescent secondary antibody (Texas red, 1:100) and is indicated by red spots. LAMA5 and ITGA6 co-localize at cell membrane (white arrow).

Figure S7. Expression of ECM genes in human fibroblast cells treated with TGF1.

Representation of the qPCR performed on RNA isolated from fibroblast cell lines of the affected family members (II-7, II-9 and III-11) and healthy individuals (CNT1-3: 1 healthy family member

III-10 and 2 unrelated controls) stimulated with recombinant human TGFβ1 (10 ng/ml, Peprotech,

UK) for 48 h or not. Each experiment was performed in triplicate. N3 indicates that the bar is a medium value of the experiments carried out in the three affected family members (II-7, II-9 and

III-11) and three healthy individuals (CNT1-3). The diagrams show the fold of relative expression

(treated cells versus untreated) of a set of genes involved in ECM remodeling. qPCRs were performed in triplicate and normalized to GAPDH mRNA levels. The mean ± SD of three independent experiments were analyzed with ANOVA test for multiple comparisons with

Bonferroni's corrections and were plotted as histogram representation. A value of p<0.05 was considered to be statistically significant and was indicated with the asterisk (*). A value of p<0.01 was indicated with two asterisks (**). Note that under TGF1 treatment a significant up regulation of GLI1 and COL1A1 is observed.

Figure S8. Expression of ECM genes in human fibroblast cells treated with IL1.

Representation of the qPCR performed on RNA isolated from fibroblast cell lines of the affected family members (II-7, II-9 and III-11) and healthy individuals (CNT1-3: 1 healthy family member

III-10 and 2 unrelated controls) stimulated with recombinant human IL1β (10 ng/ml, Peprotech,

UK) for 48 h. Each experiment was performed in triplicate. N3 indicates that the bar is a medium value of the experiments carried out in the three affected family members (II-7, II-9 and III-11) and healthy individuals (CNT1-3). The diagrams show the fold of relative expression (treated cells versus untreated) of a set of genes involved in ECM remodeling. qPCRs were performed in triplicate and normalized to GAPDH mRNA levels. The mean ± SD of three independent

16 experiments were analyzed with ANOVA test for multiple comparisons with Bonferroni's corrections and were plotted as histogram representation. A value of p<0.05 was considered to be statistically significant and was indicated with the asterisk (*). A value of p<0.01 was indicated with two asterisks (**). Note that under IL1 treatment a significant up regulation of LAMA5,

MMP1 and MMP3 is observed.

Figure S9. Schematic representation of the generation of the Lama5V3144M knock-in mouse model. (A) Chromosomal localization, genomic structure and BglII restriction map of the mouse

Lama5 gene. The position of the mutation is reported in red. (B) Engineered PGND vector containing the genomic region at 5’ (long arm) and 3’ (short arm) of the mutation. Schematic representation of the two crossing over events which generated the recombinant cells. (C) Southern blot analysis to identify recombinant clones. A single immune band of 4500 bp is expected in the presence of recombination. (D) Animals cross schematic representation.

3144M -/- Figure S10. Skin morphology of Lama5 knock-in mouse model. 7 µm thick paraffin sections of dorsal skin biopsies of knock-in Lama5V3144M homozygous and wild-type female mice at

28 post natal days were analyzed. HE staining (10x) of homozygous mice (A-F) and wt animals (G-

I) was reported. Scattered hair and reduced or absent number of mature hair follicles along the dermal-epidermal interface were observed in mutated animals compared to wild‐type mice.

V3144M +/- Figure S11. Skin morphology of Lama5 knock-in mouse model. 7 µm thick paraffin sections of dorsal skin biopsies of knock-in Lama5V3144M heterozygous and wild-type female mice at 28 post natal days were used. (A) HE staining (10x) of three heterozygous mice (A-C) and three wt animals (D-F) were reported. Histological appearance did not show significant morphological differences between mutated and wt mice. In the heterozygous mice the largest numbers of the hair follicles were mature and tidy along the dermal-epidermal junction and regular hystoarchitecture of the dorsal skin was reported similarly to wt animals.

17

Figure S12. Small intestine morphology of Lama5V3144M +/- knock-in mouse model. 7 µm thick paraffin sections of small intestine biopsies of knock-in Lama5V3144M heterozygous female mice and wild-type female mice at 28 post natal days were analyzed. HE staining (4, 10 and 20x) of heterozygous (A-F), homozygous (G-I) and wt (J-L) mice were reported. In the wt animals (J-L) a regular histology of the small intestine was observed; the villi extended from the muscularis mucosae vertically with the same length and caliber. Compared to homozygous mice (G-I), in heterozygous animals (A-F), only few villi showed club-shaped form and were inclined from the lamina muscularis mucosa which in some points appeared more lax and wider.

Table S1. Clinical symptoms of the affected family members.

Table S2. Summary of the whole exome sequencing of Italian patients. Autosomal shared: autosomal variants shared among the three analyzed patients; Aut+segregation: autosomal variants remaining after segregation analysis in the extended pedigree.

Table S3. Segregation analysis of the 16 variants identified in WES. For each gene is reported the chromosomal localization (Chr), the symbol name, the nucleotide and amino acid change, the minor allele frequency (MAF), the disease relevance (Dis), dominance or recessive inheritance of the reported diseases (D/R), tolerated or deleterious effect determined through bioinformatics tools

(T/D), the evolutionary conservation (C) and the pedigree number of each analyzed subject (in bold are reported the affected family members).

Table S4. The list of genes showing increased levels of expression. Chr=Chromosome; fold: indicates fold of expression which is obtained by comparison to healthy subjects; inflam.=inflammation. In the table are reported the genes showing increased levels of expression which are involved in specific functions or diseases: wound-healing and inflammation.

Table S5. The list of genes showing decreased levels of expression. Chr=Chromosome; fold: indicates fold of expression which is obtained by comparison to healthy subjects; inflam.=inflammation. In the table are reported the genes showing decreased levels of expression which are involved in specific functions or diseases: wound-healing and inflammation. 18

Table S6. Primers used for qPCR analysis. For all primer pairs used in qPCR analysis forward and reverse sequences as well as Melting Temperature (MT) and PCR sizes are reported.

19

A B C N-terminus

α6β4 integrins LN

LFx

L4

LG3-3140V LG3-3140M D E

Lutheran α3β1, α6β1 integrins LG1 LG2 V3140M LG3 α6β4 integrin LG4 LG5 α-dystroglycan Syndecan-4 C-terminus LG1-LG5-3140V LG1-LG5-3140M

F A B C D E F G H b-sheet a5LG3 3124 3200 peptide b-sheet

I L M N O P b-sheet a5LG3 3201 3292

b-sheet

Figure S1. LAMA5 protein structure and modeling of the wt and mutated proteins. CNT-1 CNT-2 III-10 healthy 100 μm 100 μm 100 μm

LAMA5 *

A B C II-7 II-9 III-11 100 μm 100 μm 100 μm

LAMA5

D E F CNT-1 CNT-2 III-10 healthy 100 μm 100 μm 100 μm

COL4A

G H I II-7 II-9 III-11

100 μm 100 μm 100 μm

COL4A

J K L

Figure S2. LAMA5 and COL4A expression in human skin LAMA5 ITGA6 DAPI MERGE

100 μm 100 μm 100 μm 100 μm

III-11 Proband

100 μm 100 μm 100 μm 100 μm

III-10 healthy

Figure S3. LAMA5 and ITGA6 expression in human skin. LAMA5

100 μm 100 μm 200 μm 200 μm

CNT

A B C D

100 μm 200 μm 200 μm 200 μm

11

- III *

E F G H

CNT III-11

100 μm 100 μm 100 μm 200 μm

COL4A

I J K L

Figure S4. LAMA5 and COL4A expression in human skeletal muscle. FIBROBLAST CELL LINES Phalloidin LAMA5 Merge

III-11

50 μm 50 μm 50 μm

III-10 healthy

50 μm 50 μm 50 μm

MYOBLAST CELL LINE Phalloidin LAMA5 Merge

III-11

50 μm 50 μm 50 μm

III-10 healthy

50 μm 50 μm 50 μm

Figure S5. LAMA5 expression in human fibroblast and myoblast cell lines. LAMA5 ITGA6 MERGE

III-11

50 μm 50 μm 50 μm

III-10 healthy

50 μm 50 μm 50 μm

Figure S6. LAMA5 and ITGA6 expression in human fibroblast cell lines. **

**

GLI1 LAMA5

* * * *

MMP1 MMP3

**

**

COL1A1 Figure S7. Expression of ECM genes in human fibroblast cells treated with TGFb1. **

**

GLI1 LAMA5

** **

**

**

MMP1 MMP3

**

COL1A1

Figure S8. Expression of ECM genes in human fibroblast cells treated with IL1b. A ATG STOP

V3144M

660 49913

chr2 PGND B p.V3144M

Wild type allele

Mutated allele

C Probes short arm D Long arm Short arm

M 1 2 3 4 5 6 7 8 9 10 C

4500 bp

Southern blotting short arm

Figure S9. Schematic representation of the generation of the Lama5V3144M knock-in mouse model. 3 Lama5 3144M -/- 1 2

100 μm 100 μm 100 μm

A CC -1 A Het-2 B Het-3 C 4 5 6 100 μm 100 μm 100 μm

D E F Wt 1 2 3 100 μm 100 μm 100 μm

G H I 4 5 6

100 μm 100 μm 100 μm

J K L

Figure S10. Skin morphology of Lama53144M -/- knock-in mouse model. Lama5 V3144M +/- 1 2 3

100 μm 100 μm 100 μm

A B C

Wt 1 2 3

100 μm 100 μm 100 μm

D E F

Figure S11. Skin morphology of Lama5 V3144M +/- knock-in mouse model. Lama5 V3144M +/-

100 μm 100 μm 100 μm

1

A B C

100 μm 100 μm 100 μm

2

D E F Lama53144M -/-

100 μm 100 μm 100 μm

G H I

Wt

100 μm 100 μm 100 μm

J K L

Figure S12. Small intestine morphology of Lama5V3144M +/- knock-in mouse model. Table S1: Clinical symptoms of the affected family members. LAMA5 MULTISYSTEM SYNDROME Pedigree ID II-4 II-7 II-9 III-1 III-2 III-4 III-5 III-8 III-9 III-11 III-12 Sex M F F M F F F M F F F Age° 61 y 60 y 56 y 43 y 42 y 33 y 35 y 39 y 36 y 37 y 33 y ORGAN/ CLINICAL FEATURES SYSTEM Skin striae; delayed wound healing ; atrophic scars; mild alopecia with dry and brittle SKIN hair Dental crowding; gingival recessions, temporo-mandibular joint algo-dysfunctional STOMATOGNATHIC syndrome Night blindness. (Severe myopia*) EYE Myopia absent in III-1, 2, 4, 5, 12 Goiter and hypothyroidism ENDOCRINE Absent in III-1 SKELETAL Laxity of osteo-articular joints; joint pain; serum negative osteoarthritis GASTRO- Cardial incontinence and hiatus hernia; celiac-like malabsorption syndrome INTESTINAL CONNECTIVE Dislocation and prolapse of abdominal and pelvic organs TISSUE AND Pelvic prolapse absent in II-4, III-1, 8 LIGAMENTS Reduced muscle endurance**; myalgias and cramps after work and at rest; diffuse hypotonia; increased CPK; degenerative myopathy (muscle biopsy performed in II-7, 9 MUSCLE and III-11). CPK increase absent in III-1, 5 Slightly reduced left ventricle ejection fraction (EF 55-62%) HEART Absent in II-4, III-8

° Current age *Not associated to LAMA5 mutations **Cycle ergometry test stopped at sub-maximal effort

Table S2. Summary of the whole exome sequencing of Italian patients. Data II-7 III-11 III-4 II-7+III-11+III-4 coding variants 23.052 24.060 22.120 Non-synonym. 820 600 610 (NS)Autosomal shared 16 Aut+segregation 1 Autosomal shared: autosomal variants shared among the three analyzed patients; Aut+segregation: autosomal variants remaining after segregation analysis in the extended pedigree.

Table S3. Segregation analysis of the 16 variants identified in WES.

Chr Gene bp aa MAF Dis D/R T/D C II-4 II-7 II-8 II-9 II-10 III-1 III-2 III-3 III-4 III-5 III-8 III-9 III-10 III-11 III-12 III-13 1 LDLRAP1 T>A L31Q ARH R D C TA TA TT TA TT TA TT TA TA TT TT TA TA TA TA TT 1 CEP85 C>T A686V D C CT CT CC CT CC CC CC CT CT CC CC CT CT CT CT CC 1 NAV1 G>A E1435K D C GG GA GG GA GA GA GA GA GG GA GG GA GG GA 1 AGMAT T>C L318P D C TC TC TT TC TC TT TT TC TC TT TT TC TC TC TT TT 1 PTPRU A>G S487G 0 T AG AG AA AG AA AA AA AG AG AA AA AG AG AG AG AA 4 ADAD1 A>G S3G 0 T AG AG AA AG AG AG AG AA AG AA AA AA AA AG AA AA 4 GAK C>T R164Q 0.001 PD D T CT CT CC CC CT CT CT CC CT CT CC CT CT CT CC CC 5 IL6ST C>T P446L 0 T CC CT CC CT CT CC CT CC CT CC CT CC CC CT CC CC 11 CD3E G>C D71H 0,00 D/T GG GC GG GC GG GG GG GG GC GC GC GC GG GC GG GG 14 SAMD15 C>A N388K 0 T CA CA CC CA CC CC CC CA CA CA CC CA CC CA CA CA 16 ITGAX G>A V604M 0 D C GA GA GG GA GA GG GG GA GA GG GA GA GG GA GG GA 16 TMEM208 G>A Splicing 0 GA GA GG GA GG GA GG GG GA GG GA GG GG GA GA GG 18 SALL3 G>T A185L 0,03 T/D GG GT GG GT GG GG GT GG GT GG GT GG GG GT GC GC 18 LOXHD1 G>A E693K 0,002 T/D GA GA GG GG GA GG GA GG GA GG GG GA GG GA GG GG 18 ZNF407 T>G F1024L 0,001 ID R D C TT TG TT TG TT TT TG TG TT TT TG TG TG TG TG 20 LAMA5 G>A V3140M 0,00007 D C GA GA GG GA GG GA GA GG GA GA GA GA GG GA GA GG For each gene is reported the chromosomal localization (Chr), the symbol name, the nucleotide and amino acid change, the minor allele frequency (MAF), the disease relevance (Dis), dominance or recessive inheritance of the reported diseases (D/R), tolerated or deleterious effect determined through bioinformatics tools (T/D), the evolutionary conservation (C) and the pedigree number of each analyzed subject (in bold are reported the affected family members). Table S4. The list of genes showing increased levels of expression.

Gene Chr Fold Gene product Wound- Inflam Healing HLA-DRB6 6 63 major yes histocompatibility complex, class II, DR beta 6

IFITM3 1 15.2 interferon induced yes transmembrane protein 3 C2 6 8.89 complement yes component 2 CRP 1 5.9 C-reactive protein, yes pentraxin-related

SERPING1 11 5.2 serpin peptidase yes inhibitor, clade G (C1 inhibitor), member 1 HMCN1 1 5.08 hemicentin 1 yes

CTSL2 9 4.80 cathepsin L2 yes

IL6 7 4.24 interleukin 6 yes yes

MIF 22 2.88 macrophage yes yes migration inhibitory factor (glycosylation- inhibiting factor) FBLN5 14 2.75 fibulin 5 yes

ITGA1 5 2.71 integrin alpha 1 yes

IL6ST 5 2.32 interleukin 6 signal yes yes transducer CD40LG X 2.17 CD40 ligand yes

MAPK3 16 2.04 mitogen-activated yes protein kinase 3 (ERK1) EGF 4 1.93 epidermal growth yes factor IL1B 2 1.93 interleukin 1 beta yes yes

ITGA6 2 1.93 integrin alpha 6 yes

CXCL5 4 1.84 chemokine (C-X-C yes motif) ligand 5

TGFA 2 1.79 transforming growth yes yes factor, alpha

PECAM1 17 1.61 platelet and yes yes endothelial cell adhesion molecule 1 F13A1 6 1.60 coagulation factor yes XIII, A1 polypeptide

Chr=Chromosome; fold: indicates fold of expression which is obtained by comparison to healthy subjects; inflam.=inflammation. In the table are reported the genes showing increased levels of expression which are involved in wound-healing and inflammation. Table S5. The list of genes showing decreased levels of expression.

Gene Chr Fold Gene Product Wound- Inflam Healing HBEGF 5 0.14 heparin-binding EGF-like growth yes factor VEGFA 6 0.30 vascular endothelial growth factor yes A

IFNG 12 0.38 interferon, gamma yes yes

COL4A1 13 0.42 collagen, type IV, alpha 1 yes

COL4A3 2 0.54 collagen, type IV, alpha 3 yes

PDGFA 7 0.59 platelet-derived growth factor alpha yes yes polypeptide

CTSK 1 0.60 cathepsin K yes

SERPINE1 7 0.62 serpin peptidase inhibitor, clade E yes (nexin, plasminogen activator inhibitor type 1), member 1

TAGLN 11 0.63 transgelin yes

TGFBR3 1 0.64 transforming growth factor, beta yes yes receptor III ITGB5 3 0.68 Integrin beta 5 yes

CTNNB1 3 0.73 catenin (cadherin-associated yes protein), beta 1

Chr=Chromosome; fold: indicates fold of expression which is obtained by comparison to healthy subjects; inflam.=inflammation. In the table are reported the genes showing decreased levels of expression which are involved in wound-healing and inflammation.

Table S6. Primers used for qPCR analysis Human Genes Forward sequence 5’-3’ Reverse Sequence 5’-3’ MT Size

COL1A1 GGGATTCCCTGGACCTAAAG GGAACACCTCGCTCTCCA 60 °C 165bp

GLI1 CCACATCAACAGCGAGCACA GTCTTCAGGTTTTCGAGGCG 60 °C 198bp

LAMA5 ACTTTCCATGGCCACGGCTT ACTCCCGTGGCATTGCTGTA 60 °C 263bp

LEF1 GCCAGACAAGCACAAACCTC GGATGGGTGGAGAAAGAGAT 60 °C 190bp

MMP1 GCAAACACATCTGACCTACA CGATGATCTCCCCTGACAAA 60 °C 177bp

MMP3 GGAGATGCCCACTTTGATGA AGTGTTGGCTGAGTGAAAGA 60 °C 120bp

VEGFA165 GTTTGTACAAGATCCGCAGACGT GTTCCCGAAACCCTGAGGG 60 °C 159bp

SHH CTACAAGCAGTTTATCCCCA CCTTACACCTCTGAGTCATC 60 °C 182bp

GAPDH GGAAGGTGAAGGTCGGAGT GACAAGCTTCCCGTTCTCAG 60 °C 200bp

Mouse Genes Forward sequence 5’-3’ Reverse Sequence 5’-3’

Lama5 AAGACATCAGCATTGGTGGG ATCGGTCACAAGAGCCTCCA 60 °C 132bp

Gapdh TGGAGAAACCTGCCAACTAT CATACCAGGAAATGAGCTTG 60 °C 198bp

For all primer pairs used in qPCR analysis forward and reverse sequences as well as Melting Temperature (MT) and PCR sizes are reported.