ARTICLE Ataxia-Pancytopenia Syndrome Is Caused by Missense Mutations in SAMD9L Dong-Hui Chen,1 Jennifer E. Below,2,10 Akiko Shimamura,3,4,5,6,11 Sioban B. Keel,4 Mark Matsushita,7 John Wolff,7 Youngmee Sul,7 Emily Bonkowski,1 Maria Castella,6 Toshiyasu Taniguchi,6 Deborah Nickerson,2 Thalia Papayannopoulou,4 Thomas D. Bird,1,7,8,* and Wendy H. Raskind7,8,9,* Ataxia-pancytopenia (AP) syndrome is characterized by cerebellar ataxia, variable hematologic cytopenias, and predisposition to marrow failure and myeloid leukemia, sometimes associated with monosomy 7. Here, in the four-generation family UW-AP, linkage analysis revealed four regions that provided the maximal LOD scores possible, one of which was in a commonly microdeleted chromosome 7q region. Exome sequencing identified a missense mutation (c.2640C>A, p.His880Gln) in the sterile alpha motif domain containing 9-like gene (SAMD9L) that completely cosegregated with disease. By targeted sequencing of SAMD9L, we subsequently identified a different missense mutation (c.3587G>C, p.Cys1196Ser) in affected members of the first described family with AP syndrome, Li-AP. Neither variant is reported in the public databases, both affect highly conserved amino acid residues, and both are predicted to be damaging. With time in culture, lymphoblastic cell lines (LCLs) from two affected individuals in family UW-AP exhibited copy-neutral loss of heterozygosity for large portions of the long arm of chromosome 7, resulting in retention of only the wild-type SAMD9L allele. Newly established LCLs from both individuals demonstrated the same phenomenon. In addition, targeted capture and sequencing of SAMD9L in uncultured blood DNA from both individuals showed bias toward the wild-type allele. These observations indicate in vivo hematopoietic mosaicism. The hematopoietic cytopenias that characterize AP syndrome and the selective advantage for clones that have lost the mutant allele support the postulated role of SAMD9L in the regulation of cell proliferation. Furthermore, we show that AP syndrome is distinct from the dyskeratoses congenita telomeropathies, with which it shares some clinical characteristics.

Introduction In 1978, Frederick Pei Li reported a familial cancer susceptibility disease, ataxia-pancytopenia (AP) syndrome, also called myelocerebellar disorder (MIM: 159550).1 The father and all five of his children had onset of ataxia in the first to third decade of their lives and hematologic cytopenias of variable severity. Two of the children died from bleeding and/or sepsis and two died of acute myeloid leukemia (AML) or myelomonocytic leukemia (AMMoL).2 Somatic loss of chromosome 7 was identified in marrow from the child with AMMoL and one of the children with marrow failure.1,2 Stable mild hypogammaglobulinemia was documented in the child who developed AMMoL. Cells from three of the children did not display increased sensitivity to UV or ionizing radiation (IR) or mitomycin C (MMC). Ataxia was progressive in both the father and his surviving affected daughter. Decreased vibration sense was documented in the father, but none of the children were reported to have sensory impairments. Computed tomographic (CT) scans of the father and two of the children showed cerebellar atrophy, and in both siblings in whom the cerebellum was examined at autopsy, there was marked cerebellar atrophy with moderate to marked diffuse loss of Purkinje

cells. In addition, one child had moderate neuronal loss in the inferior olives and central nuclei. Mild microcephaly was documented in all the affected children. One child had minor dysmorphic features, and the father and four children had strabismus. Cognitive ability was normal in all. To our knowledge, one other family that appears to fit the distinct AP syndrome phenotype has been reported. A mother and her daughter and son manifested sensory and motor neuropathy in addition to early childhood onset of cerebellar ataxia.3 The only CT scan described showed mild cerebellar atrophy in the 5-year-old affected boy. The mother’s hematologic status appeared to be normal, but both children had cytopenias. In one child, monosomy 7 and trisomy 21 were evident in the marrow karyotype before the development of AML and persisted afterward. Ataxia, pancytopenia, and a predisposition to myeloid malignancies are also observed in patients with dykeratosis congenita (DC), a genetically heterogeneous group of at least 13 disorders often associated with the triad of reticulated skin hyperpigmentation, nail dystrophy, and oral leukoplakia.4–6 Hoyeraal-Hreidarsson syndrome (HHS) is a severe variant of DC, with the additional manifestations of cerebellar hypoplasia, microcephaly, and psychomotor

1 Department of Neurology, 2Department of Genome Sciences, 3Department of Pediatrics, 4Division of Hematology University of Washington School of Medicine, Seattle, WA 98195, USA; 5Seattle Children’s Hospital, Seattle, WA 98105, USA; 6Fred Hutchinson Cancer Research Center, Seattle, WA 98109, USA; 7Division of Medical Genetics, University of Washington School of Medicine, Seattle, WA 98195, USA; 8Geriatric Research, Education, and Clinical Center, Veterans Administration Puget Sound Veterans Health Care Center, Seattle, WA 98108, USA; 9Division of Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, WA 98195, USA 10 Present address: The University of Texas School of Public Health, P.O. Box 20186, Houston, TX 77225, USA 11 Present address: Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115 *Correspondence: [email protected] (T.D.B.), [email protected] (W.H.R.) http://dx.doi.org/10.1016/j.ajhg.2016.04.009. Ó 2016 American Society of Human Genetics.

1146 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

Figure 1. Identification of SAMD9L Mutations in Two Families Affected by Autosomal-Dominant AP Syndrome (A) Family UW-AP, with the c.2640C>A mutation. The segregation of microsatellitedefined haplotypes on 7q21–7q31 is shown for all pedigree members from whom a blood sample was obtained. The location of SAMD9L is indicated on the STRP schematic, shown per the Marshfield map. Recombination events in individuals III-6 and IV-2 define a 32 Mb region on the red haplotype; missing genotypes from connecting relatives preclude use of individual III-1 to further narrow the region. Exome sequencing was performed on individuals IV-1 and IV-3. (B) Family Li-AP, with the c.3587G>C mutation. Filled-in quadrants indicate the variable manifestations in both pedigrees; d. denotes age at death; c. denotes current age.

retardation.4–6 Most patients with DC have very short telomeres, a consequence of mutations in genes that encode components of the telomerase complex or shelterin, important in the elongation and protection of the telomeric end, respectively.6,7 Mutations in five of these genes can cause HHS—dyskerin (DKC1 [MIM: 300126]8), telomerase reverse transcriptase (TERT [MIM: 187270]9), regulator of telomere elongation helicase 1 (RTEL1 [MIM: 608833]10), adrenocortical dysplasia homolog (ACD [MIM: 609377]11), and trf1interacting nuclear factor 2 (TINF2 [MIM: 604319]12). It is well recognized that genetically confirmed DC patients might not manifest all the characteristic phenotypic findings of the disease. The status of Purkinje cells—preserved in HHS13,14 and degenerated in AP syndrome—can only be evaluated on autopsy. Thus, it can be difficult to distinguish AP syndrome from DC on the basis of clinical findings alone. For example, the categorization of AP syndrome in two other reported families15,16 is probably incorrect. The presence of a TINF2 mutation in one of the families described as having AP syndrome confirms a diagnosis of HHS, and indeed this patient had other characteristics indicative of DC, specifically oral leukoplakia and very short telomere lengths in multiple leukocyte subsets.12,16 Intellectual disability, dysmorphic features, and cell sensitivity to bleomycin, in addition to cerebellar ataxia and pancytopenia, argue against AP syndrome in a sporadic case.15 The overlapping hematologic and neurologic manifestations of HHS and AP syndrome create some diagnostic uncertainty, given the variable clinical phenotypes in both disorders. Here, we describe another multigenerational family with AP syndrome and present evidence

that missense mutations in the sterile alpha motif domain containing 9-like gene (SAMD9L [MIM: 611170]) are the cause of the syndrome in this family and the one reported by Li et al.1 and document that HHS and AP syndrome are distinct disorders.

Subjects and Methods Subjects and Samples Approval for the recruitment and genetic analysis of these families was granted by the respective institutional review boards of the University of Washington, the VA Puget Sound Health Care System, and Harvard University, and all living subjects provided informed consent. The pedigree of a new family, UW-AP, affected by familial cerebellar ataxia and pancytopenia is shown in Figure 1A. Blood samples were collected from 11 individuals across three generations, and EBV-transformed lymphocyte lines (LCLs) were established by standard methods. Autopsy material was obtained from one deceased individual. Stored fibroblasts from individuals I-1, I-2, and II-5 in the Li-AP family2 were cultured and fibroblasts were grown from a newly obtained skin biopsy from individual Li-AP II-4 (Figure 1B) by standard methods. Genomic DNA was extracted from these samples by standard methods.

Exclusion and Linkage Analyses Exclusion analyses were done by evaluation of short tandem repeat polymorphic (STRP) markers17,18 flanking genes involved in multiple ataxias and hematologic disorders (Table S1). A microsatellite genome scan using Linkage Mapping Set version 2 was done as previously described.18 Power and multipoint linkage analyses for this genome scan were performed with SLINK19,20 and GENEHUNTER version 1.2,21 respectively. The single nucleotide polymorphic (SNP)-based genome-wide linkage analysis was performed on a subset of SNP genotypes generated from the Illumina Human Omni1-Quad v.1-0 array, which covers 1,140,419 total SNPs. A principal component analysis of these results indicated that reference allele frequencies from the CEU (Utah residents

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1147

with ancestry from northern and western Europe) HapMap population were appropriate for this family. Summary statistics for quality control, including sex checks, tests for individual- and SNP-level missingness, heterozygosity, and identity-by-descent (IBD) analyses, were performed with PLINK. Mendelian errors were identified with Pedcheck. Parametric LOD scores were obtained with Allegro.22

Whole-Exome Sequencing DNAs isolated from peripheral blood of individuals UW-AP IV-1 and IV-3 (Figure 1A) was used for exome sequencing. Target enrichment and exome sequencing were performed as previously described.23,24 NimbleGen solution capture arrays, versions 1 and 2, were used for individuals IV-3 and IV-1, respectively, given the time gap between sequencing of the two samples. Variants were prioritized on the basis of effect on protein sequence, absence from the sequence databases dbSNP, Exome Variant Server (EVS), Exome Aggregation Consortium (ExAC), and 1000 Genomes Project, evolutionary conservation (Genomic Evolutionary Rate Profiling [GERP]25 and phastCons26), and predicted effect on protein structure and function (Sorting Intolerant from Tolerant [SIFT]27 and Polymorphism Phenotyping v.2 [PolyPhen-2]).28,29

treated with IR with a JL Shepherd Mark I Cesium Irradiator (JL Shepherd & Associates) at various exposures up to 16 Gy or with MMC (Sigma, M4287) with increasing doses up to 200 ng/mL. Cells were incubated for 5 days. Cell survival was measured with a Cell Proliferation Kit II (XTT) (Roche), according to the manufacturer’s instructions. BD0952,32,33 an LCL from an individual with the FANCI subgroup of Fanconi anemia (FA [MIM: 611360]), was used as a positive control and LCL PD734 was a normal control. To evaluate induction of the FA pathway and ataxia-telangiectasia (ATM [MIM: 607585]) and Rad3-related (ATR [MIM: 601215]) signaling, the relevant proteins were quantified via western blotting. Cells were lysed in 0.05 M Tris-HCl (pH 6.8), 2% SDS, and 6% b-mercaptoethanol and boiled for 5 min. SDS-PAGE electrophoresis was done with NuPAGE 3% to 8% Tris-acetate gels (Invitrogen), and proteins were transferred to a nitrocellulose membrane. Primary antibodies were diluted in blocking buffer (5% milk in TBS-Tween 20) and incubated overnight. The following primary antibodies were used: anti-FANCD2 (Abcam, ab2187), anti-pCHK1 S345 (Cell Signaling, 23415), and antipATM S1981 (Cell Signaling, 4526). Horseradish-peroxidaseconjugated anti-mouse and anti-rabbit immunoglobulin G (IgG) (Amersham) were used as secondary antibodies. Images were acquired via an ImageQuant LAS4000 system (GE Healthcare).

Variant Validation by Sanger Sequencing Capillary sequencing with customized primers was performed to confirm the unique coding variants in the linkage regions, to investigate co-transmission with the AP syndrome phenotype in all members of the UW-AP family, and to evaluate SAMD9L in the Li-AP family. Genomic DNA was PCR amplified with Taq Polymerase (QIAGEN) with annealing temperatures shown in Table S1 and sequenced as previously described.23,24

Analysis of Mosaicism and Loss of Heterozygosity The initially established LCLs from seven affected subjects and one unaffected at-risk subject in the UW-AP family, as well as from three of their spouses who married in to the family and served as control subjects, were thawed for reculture. To confirm results from these lines, in two affected subjects, LCLs were again generated from viably frozen mononuclear stocks. DNA was extracted at 2 week to 1 month intervals for up to 1 year and evaluated for presence of the mutation. Allele-specific PCR amplification of the SAMD9L c.2640C>A variant was performed with primers containing either the reference nucleotide or the mutant nucleotide in the 30 position and a mismatched nucleotide, T, in the preceding position. A common reverse primer was used for both mutant and reference alleles (Table S2). PCR products were visualized and sequenced. The mechanism and extent of loss of heterozygosity was investigated in one subject with the Illumina Human Omni1-Quad v.1-0 array and in another subject with the Illumina Multi-Ethnic Genotyping Array (MEGA), in accordance with the manufacturer’s instructions (Illumina). Previously described best practices metrics30 were used to determine the quality of results. DNA from buffy coat and LCLs at early and later stages of culture in all available subjects were also analyzed for read depth of the variant and reference alleles by target capture and sequencing via a bone marrow failure and myelodysplasia syndromes (MDS) panel as described.31

Analyses of DNA Repair Response To measure cellular sensitivity to DNA damaging agents, cells were seeded in 96-well plates at a density of 15,000 cells/well and

Results Clinical History The four-generation UW-AP family is of Irish, German, and Native American ancestry and includes at least nine affected members, seven men and two women, who have a disorder characterized by cerebellar ataxia and variable hematologic cytopenias. The pedigree demonstrates an autosomal-dominant pattern of inheritance with maleto-male transmission and variable expressivity and no evidence of incomplete penetrance or genetic anticipation (Figure 1A). Clinical characteristics of affected family members are summarized in Table 1. A branch of the family descended from a sibling in generation II (not shown) is also reported to contain affected individuals, but records were not available for review. Onset of neurologic problems reported by patients, most often imbalance as the initial symptom, ranged from 10 to 62 years, but manifestations are detectable much earlier by exam. Horizontal and vertical nystagmus and dysmetria are evident in most individuals. Deep tendon reflexes are usually increased, ankle clonus is easily elicited, and some affected individuals have extensor plantar responses. Strength and sensation to touch, vibration, and proprioception are preserved. Hematologic abnormalities are variable and, in some patients, intermittent. Given the phenotypic variability, clinical evaluations of four affected siblings are described here in more detail. Individual III-3 presented with purpura at 3 months of age, with pancytopenia characterized by a severe normocytic anemia (hemoglobin 6.1 g/dL), mild neutropenia (absolute neutrophil count [ANC] 1,200/mL), and mild thrombocytopenia (platelet count 100,000/mL), suggesting trilineage involvement. He lacked a history of recurrent infections

1148 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

Table 1.

Clinical Features of Affected Individuals in the UW-AP Family

Hematologic Evaluation (Abnormal Measurements/ Total Measurements Available for Review)

Neurologic Evaluation

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1149

Subj HCT

MCV

WBC

Platelets 3 103

Onset of Symptoms DTRs

Babs

Ankle Clonus Dysmet

Dysarth Nystag Severity

Imaging: Pathology

Other

II-3

anemia by history

NR

NR

NR

45–50y

NE

NE

NE

NE



NE

assistance with walking at ~65 y

ND

d. 80 y; lower GI bleed

II-4

48 (0/1)

91 (0/1)

4.92 (0/1)

148.9–152 (1/2)

62 y

[

þ



þ

þ

HþV

moderate

ND

d. 85 y

III-3 20–43.8 (2/10)

95–101.4 (3/6)

2.3–4.5 (11/12)

58–201 (3/11)

30 y

[

þ/

þ

slight

mild

HþV

mild; gait normal at 47 y

ND

none noted

III-5 19–36 (2/3)

NR

1.7–6.0 (4/5)a

5.8–150 (9/10)a

unknown

NE

NE

NE

NE



NE

neurologically normal

autopsy: severe loss of PC

mild coarctation of aorta; d. 15 y; retroperitoneal bleed

III-6 45 (0/1)

95 (0/1)

4.18 (1/1)

132–179 (1/2)

25 y

[



þ

moderate þ

HþV

severe; walker by 38 y

MRI: marked cerebellar degeneration; pronounced in the midline

none noted

III-8 36.5–45.2 (0/20)

94–103.9 (16/19)

3.6–7.8 (9/20)

41–216 (6/23)

unknown

[

þ left only þ

slight



HþV

mild; gait normal at 38 y

MRI: moderate cerebellar none noted atrophy, mild white matter abnormalities

IV-1

44 (0/1)

87 (0/1)

9.1 (0/1)

326 (0/1)

18 y

[



þ

þ



HþV

gait normal at 18 y ND

none noted

IV-2

29–40 (4/5)

89–108 (1/2)

3.59–5.9 (1/2)

36–444 (1/2)

6y

normal 



mild



HþV

gait normal at 16 y ND

sagittal suture craniosynostosis

IV-3

22.2–38.2 (4/12)

81.7–85 (0/9)

2.7–5.9 (8/11)

13–346 (3/11)

10 y

[

þ

þ



H

gait normal at 11 y ND

myoclonus, KT normal when pancytopenic



Subj, subject; Hct, hematocrit; MCV, mean corpuscular volume; WBC, white cell count; DTRs, deep tendon reflexes; Babs, Babinski sign; Dysmet, dysmetria; Dysarth, dysarthria; Nystag, nystagmus; NR, no records available; NE, not examined by an author and not included in a reviewed neurologic exam; ND, not done; d., age at death; PC, Purkinje cells; H, horizontal; V, vertical; KT, karyotype. a Estimated from hospital notes. Terminal hospital admission hematocrit was 19, platelet count was 5,800, and white cell count was 1,700 with no neutrophils.

Figure 2. Imaging and Pathology of the AP Syndrome (A) Cerebellar section of individual UW-AP III-5 after death from a retroperitoneal bleed at age 16 years, stained with anti vimentin (scale bar, 100 mm). This demonstrates marked loss of Purkinje cells and milder loss of granule cells. (B) T-1-weighted sagittal MRI of the brain of individual UW-AP III-8 at age 38, showing moderate midline cerebellar atrophy despite mild clinical manifestations. (C) T-1-weighted sagittal MRI of individual Li-AP II-4 at age 52, showing marked midline cerebellar and pontine atrophy. (D) T-2-weighted coronal MRI of individual Li-AP II-4 at age 52, showing diffuse bilateral abnormal white matter signal.

in childhood; however, as an adult, he had chronic gingivitis requiring gum surgery and frequent dental cleanings. His platelet counts between ages 11 to 47 ranged from 58,000/mL to 177,000/mL, and neutropenia (defined as ANC < 1,500/mL) was persistent. Bone marrow aspirate at 11 years of age showed hypoplasia with normal maturation but some more immature forms in both the erythroid and myeloid series and focal plasmacytosis; marrow biopsy was not done. Cytogenetic studies were not done. He developed slowly progressive imbalance at about 30 years old. At age 47 years, neurologic examination was notable for nystagmus that was more pronounced with horizontal than vertical gaze, mild dysarthria, mild finger-to-nose dysmetria, and hyperreflexia with unsustained ankle clonus. He had normal gait but difficulty with tandem walking. The oropharynx was normal without thrush or aphthous ulcers. There was no angular stomatitis. At age 1 year, his brother, individual III-5, manifested ecchymoses and petechiae. Marrow aspirate showed markedly reduced megakaryocytes. At age 8.5 years, for the possible diagnosis of thrombopoietin deficiency, plasma transfusion was administered without evidence of response. Platelet counts oscillated over the next 20 days, with a mean of 48,500/mL (range 34,000–64,000/ml). Hemoglobins were 11 - 12 g/dl and white cell counts ranged from 2,600/ml to 6,000/ml (average 3,500/ml). The marrow aspirate at that time was hypocellular, with a myeloid to erythroid ratio of 1:1 and only 1% megakaryocytes.

There was no frank morphologic or immunophenotypic dysplasia. Cytogenetic study was not done. He had persistent thrombocytopenia (platelet counts ranging from 50,000 to 150,000/mL) with petechiae and epistaxis. Ataxia was not mentioned in his medical records or reported by his family members. At 16 years of age, after a 2 week prodrome of pallor, fatigue, and increased petechiae, he was hospitalized with severe pancytopenia (hematocrit [HCT] 19%, ANC 0, and platelets 5,800/mL). Two bone marrow aspirates were markedly hypocellular for his age, with a relative erythroid hyperplasia and absent megakaryocytes. There was no morphologic or immunophenotypic dysplasia. Marrow biopsy and cytogenetic evaluation were not done. Despite multiple red cell and platelet transfusions, he developed a retroperitoneal bleed and rectal necrosis, and after subsequent colostomy he died of cardiac failure. Laboratory studies documented normal protime and partial thromboplastin time until the day of his death, when they were increased. Fibrinogen and factors V and VII levels were normal. The autopsy report stated that there was ‘‘marked panhypoplasia of the bone marrow with an increase in plasma cells suggesting a hyperergic form of suppression.’’ Although there was no history of neurologic dysfunction, cerebellar and inferior olivary atrophy were seen, along with severe loss of Purkinje cells (Figure 2A). In general, white matter was well myelinated, but extensive gliosis of white matter was noted in the hippocampus. Individual III-6 noted unsteady gait at 25 years that progressed to severe ataxia, requiring a walker by age 38. Horizontal and vertical nystagmus, hyperreflexia, ankle clonus, dysmetria, and moderate dysarthria were all apparent at age 31; the dysarthria was severe by age 41. There was no sensory abnormality or cognitive decline. MRI at age 32 showed marked cerebellar degeneration that was most pronounced in the midline region. No pontine atrophy was identified. Two small foci of high signal intensity in the frontal white matter were noted. Only three hematologic laboratory results were available for review. A complete blood count at the age of 43 years, performed when the family was being evaluated at the University of Washington, was normal except for a mildly decreased white cell count of 4.18 3 103/mL with 2.18 3 103/mL neutrophils. Mild to moderate thrombocytopenia was found at ages 6 and 14 years: 58,000/mL and 132,000/mL, respectively. Clinical molecular genetic testing was negative for pathologically expanded repeats in the genes for spinocerebellar ataxia (SCA) types 1 (ataxin 1 [MIM: 601556]), 2 (ataxin 2 [MIM: 601517]), 3 (ataxin 3 [MIM: 607047]), 6 (calcium channel, voltage-dependent, p/q type, alpha-1a subunit [MIM: 601011]), and 7 (ataxin 7 [MIM: 607640]). Because of the family history, their sister, individual III-8, had a neurologic evaluation at age 25 years. Her only symptom was ‘‘jumpy legs.’’ Physical exam was remarkable for horizontal and vertical nystagmus and ankle clonus. An MRI was reported to be normal. She first noticed some balance problems with bike riding at age 36. At that time, her exam also revealed slight finger-to-nose

1150 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

dysmetria and normal gait and tandem walking. Reflexes were hyperactive with sustained ankle clonus. Two years later she demonstrated mild difficulty with tandem walking and a right Babinski reflex was elicited. MRI at 38 years showed moderate cerebellar atrophy and a few subcortical and periventricular white matter hyperintensities (Figure 2B). Her complete blood counts and red cell indices were normal until the age of 45 years, when mild macrocytosis (100–104 fL) without anemia and intermittent mild neutropenia (white blood cell count [WBC] 3.6 3 103/mL and ANC 1.6 3 103/mL) and thrombocytopenia (platelet count 82,000–131,000/ml) became apparent. Serum folate and B12 levels were normal. At age 50 years, telomere lengths in hematopoietic cell subsets were measured by Repeat Diagnostics. Unfractionated lymphocytes, naive T cells (CD45RAþ), memory T cells (CD45RA), and NK cells (CD57þ) were in the normal range. Granulocytes and B cells (CD20þ) had low median telomere lengths at 6.3 kb and 6.2 kb, respectively, as compared to the medium telomere lengths of 8.0 kb and 7.7 kb, respectively, for her age, but both were above the sixth percentile. For comparison, dyskeratosis congenita telomere lengths are typically below the first percentile in multiple leukocyte subsets.35 The severities of the hematologic and neurologic abnormalities are not concordant. Individual III-3 had severe hematologic manifestations but only very slowly progressive neurologic disability, in contrast to his brother III-6, who had less severe hematologic manifestations but much more progressive neurologic manifestations. Several other clinical features warrant mention. Individual II-3 suffered from dementia for 10 years and died of a gastrointestinal hemorrhage at 79. Unfortunately, blood counts were not available for review. At 6.5 years, individual IV-3 was given a presumptive diagnosis of aplastic anemia, supported by results of a bone marrow aspirate. Cytogenetic analysis showed a normal male karyotype with a constitutional non-pathogenic polymorphic pericentric inversion, 46,XY,inv(9)(p11q13), and no clonal findings. In addition, two affected individuals had birth defects—mild coarctation of the aorta in III-5 and sagittal suture craniosynostosis in IV-2. Candidate-Gene-Exclusion, Linkage, and Sequence Analyses in the UW-AP Family Given the small size of the family, which has a maximum estimated LOD < 2, candidate genes were initially selected on the basis of phenotype, and flanking markers were genotyped to evaluate all the SCAs localized at the time and multiple genes implicated in hematopoiesis (Table S1). By haplotype analyses, close linkage to all 26 candidate genes and regions was excluded, except for peptidase, mitochondrial processing, beta (PMPCB [MIM: 603131]), which encodes a binding partner of frataxin. Sequence analysis of the coding exons of PMPCB revealed no mutation. During quality control of the SNP-based linkage analysis, more than 6,000 Mendelian inconsistencies were detected,

almost all of which mapped to the long arm of chromosome 7 and nearly all of which involved subject III-3, whose DNA was from an LCL. Of 3,225 SNPs that showed Mendelian incompatibility with III-3 and his parents, 3,222 were incompatible with his affected father, II-4. This excess of Mendelian errors prompted further analysis of the genotype data of III-3, revealing a region of complete lack of heterozygosity for almost the entire long arm of chromosome 7 (Figure 3A). Given that there was no decrease in signal intensities as would be seen for a deletion (Figure 3B), this most likely reflects uniparental isodisomy (UPD) for this portion of the chromosome in the LCL. This individual was omitted from the linkage analysis without loss of power because genotypes from his affected father and two affected children, none of whom exhibited a similar stretch of homozygosity, allowed recovery of all missing data. In this analysis, LOD scores greater than 1.5 were obtained on chromosomes 2, 5, 7, 10, and 11, spanning a cumulative total of 38 megabases (Figure 3E). Except for chromosome 5, the regions overlapped with results from a previous microsatellite genome scan that obtained a multipoint LODmax of 1.63 at chromosome 7q21.11–7q31.2 (data not shown). Whole-exome sequence data were obtained for individuals UW-AP IV-1 and IV-3. Sequences that failed Genome Analysis Toolkit (GATK) quality filters other than strand bias were not further considered. Ranking of the remaining variants was based on (1) location within the five SNP linkage regions; (2) coding region sequence effect— missense, nonsense, coding indels, and splice sites; (3) heterozygosity, given the autosomal-dominant inheritance pattern; and (4) variant frequency—absence from dbSNP131, the 1000 Genomes Project database, and the Exome Variant Server (EVS); and (5) was shared by both exomes or not read (specified as ‘‘N’’ in the exome sequence data) in one of the exomes. The very small region on chromosome 2 contained no genes. In the regions of interest, c.2640C>A in SAMD9L (GenBank: NM_152703.3) on chromosome 7q was the only previously unreported variant that was shared by both exomes and that cosegregated completely with disease in the UW-AP family, including individual III-5, whose DNA was later obtained from a post-mortem kidney paraffin section, and III-3, when DNA from blood was used (Figure 1C). This variant predicts substitution of glutamine for histidine at residue 880 (p.His880Gln), affects an evolutionarily conserved residue (GERP25 3.94, phastCons26 0.993, and PhyloP36 1.03), is predicted to be deleterious by SIFT (0.01)27 and probably damaging by PolyPhen-2 (0.993),28,29 and is not present in the ExAC database. We also scanned the exome data for genes responsible for ATM, DC/HHS,9–12 and autosomaldominant ataxias,37,38 but no potentially pathogenic variants were present. Hematopoietic Mosaicism in the UW-AP Family After recognition that the SAMD9L variant allele was absent in the LCL from individual III-3, LCLs from six

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1151

Figure 3. Genotyping and Linkage Analysis in the UW-AP Family (A) SNPs on chromosome 7 demonstrating LOH for most of the long arm in a cell line from III-3. Minor-allele, b, frequency (MAF) in III-3 (red) is compared to MAF in an affected relative whose DNA was analyzed at the same time (black). (B) Log R ratio of SNP signals show equivalent signal intensities across chromosome 7 for individual III-3 and his relative, consistent with diploidy. (C) SNPs on chromosome 7 demonstrating LOH for most of the long arm in an LCL from UW-AP individual IV-3. MAF in the LCL (red) is compared to MAF in his buffy coat DNA (black), analyzed at the same time. (D) Log R ratio of SNP signals in individual IV-3 shows equivalent signal intensities across chromosome 7 for both of his samples, consistent with diploidy. (E) SNP genome LOD plot of UW-AP. The LCL sample from individual III-3 was omitted from the analysis with Allegro without loss of power because genotypes from his parents and affected children were included. There were five regions of interest with LOD scores greater than 1.5.

additional affected individuals (II-4, III-6, III-8, IV-1, IV-2, and IV-3) were thawed and cultured for up to 1 year and evaluated periodically for the presence of the mutant allele. All lines, including those from III-3, were initially heterozygous, but the lines from III-3 and IV-3 exhibited loss of heterozygosity (LOH) for the mutant allele after 3 weeks to 6 months. A SNP array performed on IV-3 demonstrated copy-neutral LOH (Figures 3C and 3D), as was observed in the LCL from III-3 during the linkage analysis. Figure 3 shows that the borders of LOH in the two individuals are not identical, as had also been seen by microsatellite genotyping. For III-3, the proximal breakpoint was defined by rs11979094 (7q11.22; chr7:70,111,420), and the area of homozygosity extended to at least D7S2465 (7q36.3; chr7:156,114,685), at or very close to the terminus of the chromosome. For IV-3, the region of LOH extended from rs76004043 (7q11.23; chr7:77,154,883) to rs4255049 (chr7:149,286,695), and the span was broken up by small stretches of heterozygosity. The UCSC

Genome Browser (hg19) was used to define these breakpoints. The observation of LOH for 7q in LCLs from two patients raised the possibility of somatic mosaicism. To investigate this hypothesis, we regenerated LCLs from viably frozen mononuclear stocks from III-3 and IV-3. In these new LCLs, LOH was again detectable after 4–16 weeks. The occurrence of LOH for 7q in independently established LCLs strongly suggests that both patients have somatic mosaicism for the mutant allele in their hematopoietic systems. We then used a next-generation sequencing panel for marrow failure and myelodysplasia syndrome to estimate the proportion of buffy coat and LCL cells that do not carry the mutant allele of SAMD9L. The sample from affected subject IV-1 failed. The mutant allele comprised 30% and 42% of total reads in buffy coat from subjects III-3 and IV-3, respectively, strongly suggestive of somatic mosaicism for the mutant allele (Table S3). LCL DNA from IV-3 initially showed the same proportion of mutant reads as blood, but after 2 months in culture the variant was substantially decreased to 5%. The early culture sample from III-3 failed, but the variant also decreased to 5% by 6 months. By allelespecific amplification, we were also able to monitor the progressive loss of the mutant allele during culture of LCLs (Figure 4). Although not quantitative, band intensities after allele-specific amplification were lower for the mutant allele in buffy coat DNA from III-3 and IV-3, as

1152 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

Figure 4. Allele-Specific PCR Amplification of the c.2640C>A Mutation in Blood and LCLs in the UW-AP Family The mutant allele was amplified with primers containing the mutant nucleotide, A, in the 30 position and a mismatched nucleotide, T, in the preceding position. In unaffected individual III-2 no mutant band was visualized. In blood from all three affected individuals, IV-1, III-3 and IV-3, a mutant band is seen. In culture of LCLs from III-3 and IV-3, the mutant band is diminished, as shown for time points T1 to T3.

expected for mosaicism. Furthermore, in contrast to the equivalent intensities for the wild-type allele throughout long-term LCL culture, mutant allele bands early in culture from these two subjects were less intense than those obtained for other affected subjects and were completely absent late in culture. We observed the same progressive loss of the mutant allele in newly generated LCLs from these same subjects, again demonstrating that cells that have lost the variant allele have a selective growth advantage in culture. In two other affected individuals, a downward shift in the proportion of mutant allele reads from buffy coat to LCL was observed: 51% to 43% in II-4 and 54% to 44% in IV-2. These results provide additional support for a selective advantage in culture of cells without the mutant allele, either lost during culture or reflecting low level hematopoietic mosaicism (Table S3). Normal Response to DNA Damage and Proficiency of Repair Pathways Bone marrow failure and ataxia are hallmarks of FA and ATM, autosomal-recessive disorders caused by mutations in DNA repair genes.39–41 To assess DNA repair proficiency in the AP syndrome, we tested cellular sensitivity to MMC and IR. No increased sensitivity was observed in LCLs derived from affected individual UW-AP IV-3 in comparison to LCLs from unaffected UW-AP III-2 and PD7, a wild-type control individual34 (Figures 5A and 5B). In contrast to the FA LCL, normal induction of FANCD2 ubiquitination, CHK1 phosphorylation at Ser345, and ATM phosphorylation at Ser1981 by DNA-damaging agents IR and MMC confirmed efficient activation of the FA pathway and ATR and ATM signaling in cells from UW-AP IV-3 (Figure 5C). Analysis of SAMD9L in the Li-AP Family We had the opportunity to evaluate the one surviving affected sibling (II-4) from the Li et al. family2 (Figure 1B). She developed mild gait ataxia as an adolescent, was noted to have nystagmus at age 17, and has had persistent mild anemia. Her occipitofrontal circumfer-

ence was 52 cm (C, p.Cys1196Ser) was identified in all three affected individuals but not in the unaffected mother (Figure 1D). This mutation affects a conserved residue (GERP 4.88, phastCons 1), is not present in the dbSNP, 1000 Genomes, or ExAC databases and is predicted to be pathogenic by SIFT (0.06). The high conservation scores reflect the presence of cysteine at this position in all primates. The fact that a glycine residue is present at the homologous position of all other vertebrate SAMD9L orthologs or paralogs is likely responsible for the low, predicted benign, PolyPhen score of 0.009 for the serine substitution. Loss of the variant allele was not observed in cultured fibroblasts from the subjects in this family.

Discussion The AP syndrome was first described by Frederick Li, most noted for his discovery of the disorder known eponymously as Li-Fraumeni syndrome (MIM: 151623).42 Herein, we provide evidence that missense mutations in SAMD9L are responsible for the AP syndrome in two multigenerational families. SAMD9L-related AP is an autosomal-dominant ataxia with prominent hematopoietic manifestations.37,38 It differs from two other disorders with combined neurologic and hematologic features. X-linked sideroblastic anemia and ataxia (XLSA/A [MIM: 301310]) also demonstrates both cerebellar degeneration and hematologic abnormalities, but the associated mild anemia is usually asymptomatic.43,44 ATM, an autosomalrecessive disorder, confers sensitivity to IR and a high risk for leukemia and lymphoma but not pancytopenia,45 and heterozygotes do not manifest neurologic impairments.46 As was previously reported for the AP-Li family,2 in our study of AP-UW, we observed no increased sensitivity to IR or MMC or impairment of the repair pathways of FA and ATM.

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1153

Figure 5. Response of UW-AP LCLs to DNA Damaging Agents Individual UW-AP III-2 is unaffected, UW-AP IV-3 is affected, NC is a normal control, and FA is a patient with Fanconi anemia. (A) The surviving fraction of treated cells compared to untreated cells after 5 days of exposure to MMC (n ¼ 3, mean 5 SEM). (B) The surviving fraction of treated cells compared to untreated cells 5 days after a dose of IR (n ¼ 3, mean 5 SEM). (C) Quantification by Western blot of proteins involved in induction of the FA pathway or in ATM or ATR signaling. Cells were harvested 8 hr after treatment with IR (10 Gy) or 24 hr after continuous exposure to MMC (60 ng/mL), as indicated. In contrast to the increased sensitivity to DNA damaging agents manifested by FA, response of IV-3 was not different from that of III-2 or NC.

The two families with most likely pathogenic mutations in SAMD9L demonstrate marked variation in ages of onset, rates of progression, and noncoordinate severity of hematologic and neurologic manifestations. In UW-AP, the two individuals with the most severe cytopenias had relatively mild ataxia (III-3 and III-5), whereas the one with the most severe ataxia had mild intermittent cytopenias (III-6). Persistent moderate neutropenia did not lead to infectious consequences. Thrombocytopenia was the cause of most morbidity, with epistaxis, petechiae, and fatal hemorrhage in at least one affected man (III-5). In contrast to the Li-AP family and the one other family affected by AP syndrome (Daghistani-AP),3 no one had dysplastic marrow changes and no one developed myelodysplasia or leukemia. We also note that substantial brain pathology might be present before neurologic signs are manifest. Individual III-6 had moderate cerebellar atrophy but very mild neurologic impairment, and III-5 already had substantial Purkinje cell loss without any observable neurologic problems. The affected woman in the Li-AP family had slowly progressive neurologic disease over five decades despite pontine and cerebellar atrophy and marked diffuse white matter disease mimicking leukodystrophy. White matter disease was observed to a mild degree in three affected members of the UW-AP family. Although three autosomalrecessive ataxias have associated white matter disease

(cerebrotendinous xanthomatosis [MIM: 213700], Niemann-Pick type C [MIM: 257220], and SPAX3 [MIM: 611390]), this is not common in autosomal-dominant SCAs. Therefore, white matter disease might be a distinguishing feature of the autosomal-dominant AP syndrome. SAMD9L-related AP further expands the category of inherited marrow failure syndromes. Patients with classic presentations of DC, HHS, or AP will fall into distinct categories, but those with only one hematologic feature can pose a diagnostic dilemma. The mucocutaneous abnormalities are not present in many cases of DC,47,48 and mutations in DC genes have been found in patients with only one of the hematologic manifestations.49,50 A subset of those without mucocutaneous involvement, with or without obvious cerebellar dysfunction, might have AP, especially if normal telomere length can be documented. Therefore, molecular verification will be necessary for an accurate diagnosis in some of these cases. In addition, it is possible that AP syndrome might encompass an even broader clinical spectrum. The proband in the Li-AP family had microcephaly, strabismus, micrognathia and flattened malar areas, and clinodactyly; other affected relatives had mild microcephaly and strabismus.1 The molecular etiology in the Daghistani-AP family has not been reported, so it is not known whether lower extremity weakness and sensory and motor neuropathy in that family3 are

1154 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

part of the SAMD9L-related AP syndrome. The range of manifestations of SAMD9L-related AP syndrome and investigation of genotype effects on the phenotype must be more fully defined by identification of additional families with germline mutations in this gene. The role of SAMD9L in human hematopoietic dysplasias and malignancies is not known. Human SAMD9L is contiguous to a highly conserved paralogous gene, sterile alpha motif (SAM) domain containing 9 (SAMD9 [MIM: 610456]51). These genes each contain one SAM domain that has been shown to homo- and hetero-oligomerize, associate with both SAM-containing and non-SAMcontaining proteins, and bind RNA.52–54 SAMD9L and SAMD9 are widely expressed in adult and fetal tissues, including hematopoietic cells and CNS (Human Protein Atlas), at moderate to high levels. Mutations in SAMD9 are responsible for the autosomal-recessive disorder normophosphatemic familial tumoral calcinosis (NFTC [MIM: 610455]),55–57 and there is evidence that SAMD9 is involved in control of cell proliferation and functions as a tumor suppressor.58,59 The function of SAMD9L is not well characterized, and to date it had not been causally related to an inherited human disease. However, evidence has accumulated that SAMD9L also has a role in cell proliferation, most likely as a tumor suppressor. It is a key antiproliferative gene repressed by DNp63a, an oncogene in squamous cell carcinoma.60 SAMD9L is expressed at lower levels in breast cancer tissue and hepatocellular carcinomas than in normal breast or liver tissue from the same patients.58,61 Somatic mutations in SAMD9L that cluster in the coding region have been identified in hepatitis-Brelated hepatocellular carcinomas.61,62 Knockdown of SAMD9L by siRNA or shRNA resulted in enhanced proliferation and colony formation of multiple hepatocellular carcinoma cell lines and increased tumorigenicity in nude mice.61 Relevant to the marrow failure and propensity to myeloid malignancy observed in the AP-syndrome, studies in mice heterozygous or homozygous null for Samd9l, the ortholog of SAMD9L,63 suggest that SAMD9L might function as a tumor suppressor for myelodysplasia and myeloid leukemias.64 At approximately 12 months of age, these mice develop a myeloid disorder akin to human MDS characterized by variable cytopenias, marrow hypercellularity, and morphologic dysplasia, and, rarely, myeloid leukemia. In comparison to marrow cells from Samd9lþ/þ mice, those from Samd9l/ mice produced more colonies after more serial transfers. In competitive repopulation transplant studies into lethally irradiated recipients, Samd9l-deficient cells preferentially expand compared to Samd9lþ/þ cells,64 further suggesting enhanced reconstitution ability of progenitors in Samd9l-deficient mice. In contrast to the haploinsufficiency mechanism for myelodysplasia in Samd9l-deficient mice, in patients with AP-syndrome with heterozygous mutations of SAMD9L, marrow hypocellularity without dysplastic features could reflect a toxic gain of function with suppression of precur-

sor cell divisions. It is possible that the neurologic component of AP syndrome results from a toxic gain of function as well, given that neurologic deficits were not reported and presumably not present in Samd9l-knockout mice. We can speculate that conversion to marrow hypercellularity and myelodysplasia associated with monosomy 7 in the Li-AP family resulted from loss of the proliferation-repressive mutant SAMD9L allele, leaving a single wild-type allele and a haploinsufficient state analogous to what was seen in mice lacking one copy or both copies of Samd9l. Consistent with a multistep pathogenesis of cancer, progression to myeloid leukemia likely requires additional mutational events. Appropriate samples from these patients are not available to evaluate this hypothesis. Monosomy for all or part of chromosome 7 is a recurrent karyotypic abnormality in development and progression of myelodysplasia and myeloid leukemias,65–67 frequently characterizes treatment-associated hematologic disorders68–70 and inherited marrow failure syndromes,71–75 and is associated with poor prognosis. There is support for involvement of several non-overlapping segments on chromosome 7q in this oncogenic process.67 In a study of juvenile myelomonocytic leukemia with a normal karyotype, 8 of 21 patients had a submicroscopic deletion of chromosome 7q21.2–7q21.3;51 SAMD9L is one of only three genes in the shared deletion area. In a targeted analysis of SAMD9L in sporadic MDS and AML, no SAMD9L mutations were detected, but 28% of the patients carried only one copy of the gene.64 A recent comprehensive study identified one or more genes in band 7q32.3–7q36.1 as crucial for hematopoietic differentiation in induced pluripotential stem cells from patients with MDS.76 This study noted a proliferative advantage in culture for lines that acquired diploidy of the distal 7q region by segmental UPD. These observations suggest that SAMD9L plays a role in sporadic MDS and myeloid leukemia. Our study of LCLs in two subjects with AP syndrome provides evidence that cells that are diploid for wildtype SAMD9L have a selective culture advantage over those that are heterozygous for mutant SAMD9L. The most parsimonious explanation for our observation of mosaic LOH for chromosome 7q in these subjects’ blood is that hematopoietic precursor clones that lost the mutant SAMD9L allele and duplicated the wild-type allele through UPD persist in vivo because of a selective advantage. In summary, we show that missense mutations in SAMD9L are responsible for autosomal-dominant AP syndrome, possibly through a gain-of-function proliferationsuppressive effect. The development of myeloid leukemia and/or MDS in some patients with AP syndrome and in Samd9l-knockout mice, together with findings in sporadic human MDS, strongly suggest that SAMD9L is one of the crucial genes in the development of sporadic myelodysplastic syndromes and myeloid leukemias, including those characterized by deletions of chromosome 7q.

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1155

Accession Numbers The accession number for the exome sequences reported in this paper is dbGaP: phs000707.

Supplemental Data Supplemental Data include three tables and can be found with this article online at http://dx.doi.org/10.1016/j.ajhg.2016.04.009.

Acknowledgments We dedicate this manuscript to Dr. Frederick Li, who astutely recognized the AP syndrome as a distinct inherited disorder. We are grateful to the families for their participation in this study. Drs. Raj Kapur and Harvey Sarnat, Department of Pathology, Seattle Children’s Hospital, were extremely helpful in obtaining tissues and records and in reviewing the neuropathology in case III-5 in the UW-AP family. This work was supported by the US Department of Veterans Affairs (GRECC and VISN-20 MIRECC VA Puget Sound Health Care System and VA Merit Award to T.D.B. and W.H.R.), the American Recovery and Reinvestment Act funds (through RC2 HG005608 from the NIH to D.N. and W.H.R.), the National Institute of Neurologic Diseases and Stroke (R01NS069719 to W.H.R.), the National Institute of Diabetes and Digestive and Kidney Diseases (R24DK099808-01 to A.S.), and the National Human Genome Research Institute and the National Heart, Lung, and Blood Institute (NHLBI) University of Washington Center for Mendelian Genomics (U54HG006493 to Drs. D.N., Michael Bamshad, and Suzanne Leal). We would also like to recognize the ongoing studies that produced and provided exome variant calls for comparison: the NHLBI GO Cohort Sequencing Project (HL1029230), Women’s Health Initiative Sequencing Project (HL102924), Broad GO Sequencing Project (HL102925), Seattle GO Sequencing Project (HL102926), and Heart GO Sequencing Project (HL103010). Received: January 6, 2016 Accepted: April 14, 2016 Published: June 2, 2016

Web Resources 1000 Genomes, http://www.1000genomes.org dbSNP, http://www.ncbi.nlm.nih.gov/projects/SNP/ ExAC Browser, http://exac.broadinstitute.org/ GenBank, http://www.ncbi.nlm.nih.gov/genbank/ NHLBI Exome Sequencing Project (ESP) Exome Variant Server, http://evs.gs.washington.edu/EVS/ OMIM, http://www.omim.org/ PLINK, http://pngu.mgh.harvard.edu/~purcell/plink/ PolyPhen-2, http://genetics.bwh.harvard.edu/pph2/ SIFT, http://sift.bii.a-star.edu.sg/ The Human Protein Atlas, http://www.proteinatlas.org/ UCSC Genome Browser, http://genome.ucsc.edu

References 1. Li, F.P., Potter, N.U., Buchanan, G.R., Vawter, G., Whang-Peng, J., and Rosen, R.B. (1978). A family with acute leukemia, hypoplastic anemia and cerebellar ataxia: association with bone marrow C-monosomy. Am. J. Med. 65, 933–940.

2. Li, F.P., Hecht, F., Kaiser-McCaw, B., Baranko, P.V., and Potter, N.U. (1981). Ataxia-pancytopenia: syndrome of cerebellar ataxia, hypoplastic anemia, monosomy 7, and acute myelogenous leukemia. Cancer Genet. Cytogenet. 4, 189–196. 3. Daghistani, D., Curless, R., Toledano, S.R., and Ayyar, D.R. (1989). Ataxia-pancytopenia and monosomy 7 syndrome. J. Pediatr. 115, 108–110. 4. Savage, S.A., and Bertuch, A.A. (2010). The genetics and clinical manifestations of telomere biology disorders. Genet. Med. 12, 753–764. 5. Ruggero, D., and Shimamura, A. (2014). Marrow failure: a window into ribosome biology. Blood 124, 2784–2792. 6. Townsley, D.M., Dumitriu, B., and Young, N.S. (2014). Bone marrow failure and the telomeropathies. Blood 124, 2775– 2783. 7. Glousker, G., Touzot, F., Revy, P., Tzfati, Y., and Savage, S.A. (2015). Unraveling the pathogenesis of Hoyeraal-Hreidarsson syndrome, a complex telomere biology disorder. Br. J. Haematol. 170, 457–471. 8. Knight, S.W., Heiss, N.S., Vulliamy, T.J., Greschner, S., Stavrides, G., Pai, G.S., Lestringant, G., Varma, N., Mason, P.J., Dokal, I., and Poustka, A. (1999). X-linked dyskeratosis congenita is predominantly caused by missense mutations in the DKC1 gene. Am. J. Hum. Genet. 65, 50–58. 9. Marrone, A., Walne, A., Tamary, H., Masunari, Y., Kirwan, M., Beswick, R., Vulliamy, T., and Dokal, I. (2007). Telomerase reverse-transcriptase homozygous mutations in autosomal recessive dyskeratosis congenita and Hoyeraal-Hreidarsson syndrome. Blood 110, 4198–4205. 10. Le Guen, T., Jullien, L., Touzot, F., Schertzer, M., Gaillard, L., Perderiset, M., Carpentier, W., Nitschke, P., Picard, C., Couillault, G., et al. (2013). Human RTEL1 deficiency causes Hoyeraal-Hreidarsson syndrome with short telomeres and genome instability. Hum. Mol. Genet. 22, 3239–3249. 11. Kocak, H., Ballew, B.J., Bisht, K., Eggebeen, R., Hicks, B.D., Suman, S., O’Neil, A., Giri, N., Maillard, I., Alter, B.P., et al.; NCI DCEG Cancer Genomics Research Laboratory; NCI DCEG Cancer Sequencing Working Group (2014). Hoyeraal-Hreidarsson syndrome caused by a germline mutation in the TEL patch of the telomere protein TPP1. Genes Dev. 28, 2090– 2102. 12. Savage, S.A., Giri, N., Baerlocher, G.M., Orr, N., Lansdorp, P.M., and Alter, B.P. (2008). TINF2, a component of the shelterin telomere protection complex, is mutated in dyskeratosis congenita. Am. J. Hum. Genet. 82, 501–509. 13. Mahmood, F., King, M.D., Smyth, O.P., and Farrell, M.A. (1998). Familial cerebellar hypoplasia and pancytopenia without chromosomal breakages. Neuropediatrics 29, 302–306. 14. Akaboshi, S., Yoshimura, M., Hara, T., Kageyama, H., Nishikwa, K., Kawakami, T., Ieshima, A., and Takeshita, K. (2000). A case of Høyeraal-Hreidarsson syndrome: delayed myelination and hypoplasia of corpus callosum are other important signs. Neuropediatrics 31, 141–144. 15. Gonza´lez-del Angel, A., Cervera, M., Go´mez, L., Pe´rez-Vera, P., Orozco, L., Carnevale, A., and Del Castillo, V. (2000). Ataxiapancytopenia syndrome. Am. J. Med. Genet. 90, 252–254. 16. Tsangaris, E., Adams, S.L., Yoon, G., Chitayat, D., Lansdorp, P., Dokal, I., and Dror, Y. (2008). Ataxia and pancytopenia caused by a mutation in TINF2. Hum. Genet. 124, 507–513. 17. Fernandez, M., Raskind, W., Wolff, J., Matsushita, M., Yuen, E., Graf, W., Lipe, H., and Bird, T. (2001). Familial dyskinesia and

1156 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016

18.

19. 20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

facial myokymia (FDFM): a novel movement disorder. Ann. Neurol. 49, 486–492. Raskind, W.H., Matsushita, M., Peter, B., Biberston, J., Wolff, J., Lipe, H., Burbank, R., and Bird, T.D. (2009). Familial dyskinesia and facial myokymia (FDFM): Follow-up of a large family and linkage to chromosome 3p21-3q21. Am. J. Med. Genet. B. Neuropsychiatr. Genet. 150B, 570–574. Ott, J. (1989). Computer-simulation methods in human linkage analysis. Proc. Natl. Acad. Sci. USA 86, 4175–4178. Weeks, D., Ott, J., and Lathrop, G. (1990). SLINK: a general simulation program for linkage analysis. Am. J. Hum. Genet. 47, A204. Kruglyak, L., Daly, M.J., Reeve-Daly, M.P., and Lander, E.S. (1996). Parametric and nonparametric linkage analysis: a unified multipoint approach. Am. J. Hum. Genet. 58, 1347–1363. Gudbjartsson, D.F., Jonasson, K., Frigge, M.L., and Kong, A. (2000). Allegro, a new computer program for multipoint linkage analysis. Nat. Genet. 25, 12–13. Chen, Y.Z., Matsushita, M.M., Robertson, P., Rieder, M., Girirajan, S., Antonacci, F., Lipe, H., Eichler, E.E., Nickerson, D.A., Bird, T.D., and Raskind, W.H. (2012). Autosomal dominant familial dyskinesia and facial myokymia: single exome sequencing identifies a mutation in adenylyl cyclase 5. Arch. Neurol. 69, 630–635. Korvatska, O., Strand, N.S., Berndt, J.D., Strovas, T., Chen, D.H., Leverenz, J.B., Kiianitsa, K., Mata, I.F., Karakoc, E., Greenup, J.L., et al. (2013). Altered splicing of ATP6AP2 causes X-linked parkinsonism with spasticity (XPDS). Hum. Mol. Genet. 22, 3259–3268. Cooper, G.M., Stone, E.A., Asimenos, G., Green, E.D., Batzoglou, S., and Sidow, A.; NISC Comparative Sequencing Program (2005). Distribution and intensity of constraint in mammalian genomic sequence. Genome Res. 15, 901–913. Siepel, A., Bejerano, G., Pedersen, J.S., Hinrichs, A.S., Hou, M., Rosenbloom, K., Clawson, H., Spieth, J., Hillier, L.W., Richards, S., et al. (2005). Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res. 15, 1034–1050. Ng, P.C., and Henikoff, S. (2003). SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res. 31, 3812–3814. Sunyaev, S., Ramensky, V., Koch, I., Lathe, W., 3rd, Kondrashov, A.S., and Bork, P. (2001). Prediction of deleterious human alleles. Hum. Mol. Genet. 10, 591–597. Ramensky, V., Bork, P., and Sunyaev, S. (2002). Human nonsynonymous SNPs: server and survey. Nucleic Acids Res. 30, 3894–3900. Grove, M.L., Yu, B., Cochran, B.J., Haritunians, T., Bis, J.C., Taylor, K.D., Hansen, M., Borecki, I.B., Cupples, L.A., Fornage, M., et al. (2013). Best practices and joint calling of the HumanExome BeadChip: the CHARGE Consortium. PLoS ONE 8, e68095. Zhang, M.Y., Keel, S.B., Walsh, T., Lee, M.K., Gulsuner, S., Watts, A.C., Pritchard, C.C., Salipante, S.J., Jeng, M.R., Hofmann, I., et al. (2015). Genomic analysis of bone marrow failure and myelodysplastic syndromes reveals phenotypic and diagnostic complexity. Haematologica 100, 42–48. Levitus, M., Rooimans, M.A., Steltenpool, J., Cool, N.F., Oostra, A.B., Mathew, C.G., Hoatlin, M.E., Waisfisz, Q., Arwert, F., de Winter, J.P., and Joenje, H. (2004). Heterogeneity in Fanconi anemia: evidence for 2 new genetic subtypes. Blood 103, 2498–2503.

33. Smogorzewska, A., Matsuoka, S., Vinciguerra, P., McDonald, E.R., 3rd, Hurov, K.E., Luo, J., Ballif, B.A., Gygi, S.P., Hofmann, K., D’Andrea, A.D., and Elledge, S.J. (2007). Identification of the FANCI protein, a monoubiquitinated FANCD2 paralog required for DNA repair. Cell 129, 289–301. 34. Yamashita, T., Barber, D.L., Zhu, Y., Wu, N., and D’Andrea, A.D. (1994). The Fanconi anemia polypeptide FACC is localized to the cytoplasm. Proc. Natl. Acad. Sci. USA 91, 6712– 6716. 35. Alter, B.P., Giri, N., Savage, S.A., and Rosenberg, P.S. (2015). Telomere length in inherited bone marrow failure syndromes. Haematologica 100, 49–54. 36. Pollard, K.S., Hubisz, M.J., Rosenbloom, K.R., and Siepel, A. (2010). Detection of nonneutral substitution rates on mammalian phylogenies. Genome Res. 20, 110–121. 37. Hersheson, J., Haworth, A., and Houlden, H. (2012). The inherited ataxias: genetic heterogeneity, mutation databases, and future directions in research and clinical diagnostics. Hum. Mutat. 33, 1324–1332. 38. Bird, T.D. (1998). Hereditary ataxia overview. In GeneReviews, R.A. Pagon, M.P. Adam, and H.H. Ardinger, et al., eds. (Seattle: University of Washington), pp. 1993–2016. 39. Taniguchi, T., Garcia-Higuera, I., Xu, B., Andreassen, P.R., Gregory, R.C., Kim, S.T., Lane, W.S., Kastan, M.B., and D’Andrea, A.D. (2002). Convergence of the fanconi anemia and ataxia telangiectasia signaling pathways. Cell 109, 459–472. 40. Taniguchi, T., and D’Andrea, A.D. (2006). Molecular pathogenesis of Fanconi anemia: recent progress. Blood 107, 4223–4233. 41. Ambrose, M., and Gatti, R.A. (2013). Pathogenesis of ataxiatelangiectasia: the next generation of ATM functions. Blood 121, 4036–4045. 42. Li, F.P., and Fraumeni, J.F., Jr. (1969). Rhabdomyosarcoma in children: epidemiologic study and identification of a familial cancer syndrome. J. Natl. Cancer Inst. 43, 1365–1373. 43. Allikmets, R., Raskind, W.H., Hutchinson, A., Schueck, N.D., Dean, M., and Koeller, D.M. (1999). Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSA/A). Hum. Mol. Genet. 8, 743–749. 44. D’Hooghe, M., Selleslag, D., Mortier, G., Van Coster, R., Vermeersch, P., Billiet, J., and Bekri, S. (2012). X-linked sideroblastic anemia and ataxia: a new family with identification of a fourth ABCB7 gene mutation. Eur. J. Paediatr. Neurol. 16, 730–735. 45. Perlman, S.L., Boder Deceased, E., Sedgewick, R.P., and Gatti, R.A. (2012). Ataxia-telangiectasia. Handb. Clin. Neurol. 103, 307–332. 46. Ahmed, M., and Rahman, N. (2006). ATM and breast cancer susceptibility. Oncogene 25, 5906–5911. 47. Vulliamy, T.J., Kirwan, M.J., Beswick, R., Hossain, U., Baqai, C., Ratcliffe, A., Marsh, J., Walne, A., and Dokal, I. (2011). Differences in disease severity but similar telomere lengths in genetic subgroups of patients with telomerase and shelterin mutations. PLoS ONE 6, e24383. 48. Vulliamy, T., Beswick, R., Kirwan, M.J., Hossain, U., Walne, A.J., and Dokal, I. (2012). Telomere length measurement can distinguish pathogenic from non-pathogenic variants in the shelterin component, TIN2. Clin. Genet. 81, 76–81. 49. Vulliamy, T.J., Walne, A., Baskaradas, A., Mason, P.J., Marrone, A., and Dokal, I. (2005). Mutations in the reverse transcriptase

The American Journal of Human Genetics 98, 1146–1158, June 2, 2016 1157

50.

51.

52.

53. 54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

component of telomerase (TERT) in patients with bone marrow failure. Blood Cells Mol. Dis. 34, 257–263. Yamaguchi, H., Baerlocher, G.M., Lansdorp, P.M., Chanock, S.J., Nunez, O., Sloand, E., and Young, N.S. (2003). Mutations of the human telomerase RNA gene (TERC) in aplastic anemia and myelodysplastic syndrome. Blood 102, 916–918. Asou, H., Matsui, H., Ozaki, Y., Nagamachi, A., Nakamura, M., Aki, D., and Inaba, T. (2009). Identification of a common microdeletion cluster in 7q21.3 subband among patients with myeloid leukemia and myelodysplastic syndrome. Biochem. Biophys. Res. Commun. 383, 245–251. Peterson, A.J., Kyba, M., Bornemann, D., Morgan, K., Brock, H.W., and Simon, J. (1997). A domain shared by the Polycomb group proteins Scm and ph mediates heterotypic and homotypic interactions. Mol. Cell. Biol. 17, 6683–6692. Kim, C.A., and Bowie, J.U. (2003). SAM domains: uniform structure, diversity of function. Trends Biochem. Sci. 28, 625–628. Aviv, T., Lin, Z., Lau, S., Rendl, L.M., Sicheri, F., and Smibert, C.A. (2003). The RNA-binding SAM domain of Smaug defines a new family of post-transcriptional regulators. Nat. Struct. Biol. 10, 614–621. Topaz, O., Indelman, M., Chefetz, I., Geiger, D., Metzker, A., Altschuler, Y., Choder, M., Bercovich, D., Uitto, J., Bergman, R., et al. (2006). A deleterious mutation in SAMD9 causes normophosphatemic familial tumoral calcinosis. Am. J. Hum. Genet. 79, 759–764. Chefetz, I., Ben Amitai, D., Browning, S., Skorecki, K., Adir, N., Thomas, M.G., Kogleck, L., Topaz, O., Indelman, M., Uitto, J., et al. (2008). Normophosphatemic familial tumoral calcinosis is caused by deleterious mutations in SAMD9, encoding a TNF-alpha responsive protein. J. Invest. Dermatol. 128, 1423–1429. Hershkovitz, D., Gross, Y., Nahum, S., Yehezkel, S., Sarig, O., Uitto, J., and Sprecher, E. (2011). Functional characterization of SAMD9, a protein deficient in normophosphatemic familial tumoral calcinosis. J. Invest. Dermatol. 131, 662–669. Li, C.F., MacDonald, J.R., Wei, R.Y., Ray, J., Lau, K., Kandel, C., Koffman, R., Bell, S., Scherer, S.W., and Alman, B.A. (2007). Human sterile alpha motif domain 9, a novel gene identified as down-regulated in aggressive fibromatosis, is absent in the mouse. BMC Genomics 8, 92. Ma, Q., Yu, T., Ren, Y.Y., Gong, T., and Zhong, D.S. (2014). Overexpression of SAMD9 suppresses tumorigenesis and progression during non small cell lung cancer. Biochem. Biophys. Res. Commun. 454, 157–161. Gallant-Behm, C.L., Ramsey, M.R., Bensard, C.L., Nojek, I., Tran, J., Liu, M., Ellisen, L.W., and Espinosa, J.M. (2012). DNp63a represses anti-proliferative genes via H2A.Z deposition. Genes Dev. 26, 2325–2336. Wang, Q., Zhai, Y.Y., Dai, J.H., Li, K.Y., Deng, Q., and Han, Z.G. (2014). SAMD9L inactivation promotes cell proliferation via facilitating G1-S transition in hepatitis B virus-associated hepatocellular carcinoma. Int. J. Biol. Sci. 10, 807–816. Huang, J., Deng, Q., Wang, Q., Li, K.Y., Dai, J.H., Li, N., Zhu, Z.D., Zhou, B., Liu, X.Y., Liu, R.F., et al. (2012). Exome sequencing of hepatitis B virus-associated hepatocellular carcinoma. Nat. Genet. 44, 1117–1121. Li, Q., Guo, H., Matsui, H., Honda, H., Inaba, T., Sundberg, J.P., Sprecher, E., and Uitto, J. (2012). Mouse Samd9l is not a functional paralogue of the human SAMD9, the gene mutated in

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74. 75.

76.

normophosphataemic familial tumoral calcinosis. Exp. Dermatol. 21, 554–556. Nagamachi, A., Matsui, H., Asou, H., Ozaki, Y., Aki, D., Kanai, A., Takubo, K., Suda, T., Nakamura, T., Wolff, L., et al. (2013). Haploinsufficiency of SAMD9L, an endosome fusion facilitator, causes myeloid malignancies in mice mimicking human diseases with monosomy 7. Cancer Cell 24, 305–317. Hussain, F.T., Nguyen, E.P., Raza, S., Knudson, R., Pardanani, A., Hanson, C.A., Van Dyke, D., and Tefferi, A. (2012). Sole abnormalities of chromosome 7 in myeloid malignancies: spectrum, histopathologic correlates, and prognostic implications. Am. J. Hematol. 87, 684–686. Jerez, A., Sugimoto, Y., Makishima, H., Verma, A., Jankowska, A.M., Przychodzen, B., Visconte, V., Tiu, R.V., O’Keefe, C.L., Mohamedali, A.M., et al. (2012). Loss of heterozygosity in 7q myeloid disorders: clinical associations and genomic pathogenesis. Blood 119, 6109–6117. Honda, H., Nagamachi, A., and Inaba, T. (2015). -7/7q- syndrome in myeloid-lineage hematopoietic malignancies: attempts to understand this complex disease entity. Oncogene 34, 2413–2425. Leone, G., Pagano, L., Ben-Yehuda, D., and Voso, M.T. (2007). Therapy-related leukemia and myelodysplasia: susceptibility and incidence. Haematologica 92, 1389–1398. Navarro, J.T., Feliu, E., Grau, J., Espinet, B., Colomer, D., Ribera, J.M., Oriol, A., Granada, I., Junca`, J., and Milla´, F. (2007). Monosomy 7 with severe myelodysplasia developing during imatinib treatment of Philadelphia-positive chronic myeloid leukemia: two cases with a different outcome. Am. J. Hematol. 82, 849–851. McLaughlin, P., Estey, E., Glassman, A., Romaguera, J., Samaniego, F., Ayala, A., Hayes, K., Maddox, A.M., Preti, H.A., and Hagemeister, F.B. (2005). Myelodysplasia and acute myeloid leukemia following therapy for indolent lymphoma with fludarabine, mitoxantrone, and dexamethasone (FND) plus rituximab and interferon alpha. Blood 105, 4573–4575. Shimamura, A., and Alter, B.P. (2010). Pathophysiology and management of inherited bone marrow failure syndromes. Blood Rev. 24, 101–122. Maciejewski, J.P., Risitano, A., Sloand, E.M., Nunez, O., and Young, N.S. (2002). Distinct clinical outcomes for cytogenetic abnormalities evolving from aplastic anemia. Blood 99, 3129– 3135. Huang, J.N., and Shimamura, A. (2011). Clinical spectrum and molecular pathophysiology of Shwachman-Diamond syndrome. Curr. Opin. Hematol. 18, 30–35. Haase, D. (2008). Cytogenetic features in myelodysplastic syndromes. Ann. Hematol. 87, 515–526. Maserati, E., Minelli, A., Pressato, B., Valli, R., Crescenzi, B., Stefanelli, M., Menna, G., Sainati, L., Poli, F., Panarello, C., et al. (2006). Shwachman syndrome as mutator phenotype responsible for myeloid dysplasia/neoplasia through karyotype instability and chromosomes 7 and 20 anomalies. Genes Chromosomes Cancer 45, 375–382. Kotini, A.G., Chang, C.J., Boussaad, I., Delrow, J.J., Dolezal, E.K., Nagulapally, A.B., Perna, F., Fishbein, G.A., Klimek, V.M., Hawkins, R.D., et al. (2015). Functional analysis of a chromosomal deletion associated with myelodysplastic syndromes using isogenic human induced pluripotent stem cells. Nat. Biotechnol. 33, 646–655.

1158 The American Journal of Human Genetics 98, 1146–1158, June 2, 2016