Getting to the Core of Inherited Bone Marrow Failures

EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS

1. Department of Medicine, Duke Getting to the Core of Inherited Bone Marrow University Medical Center, Failures Durham, USA; 2. King Abdulaziz University, Jeddah, KSA; 3. Institute 1,2* 3* 4 of Genetic Medicine, Newcastle SOHEIR ADAM , DARIO MELGUIZO SANCHIS , GHADA EL‐KAMAH , 5 6 4 University, UK; 4. Division of Hu‐ SUJITH SAMARASINGHE , SAMEER ALHARTHI , LYLE ARMSTRONG 4# man Genetics & Genome Research, AND MAJLINDA LAKO National Research Center, Cairo, Egypt; 5. Department of Hematol‐ Key words. inherited bone marrow failures  hESC  hiPSC  animal models ogy, Great Ormond Street Hospital for Children NHS Foundation Trust, ABSTRACT London, UK; 6. Princess Al Jawhara Al‐Brahim Center of Excellence in Bone marrow failure syndromes (BMFS) are a group of disorders with Research of Hereditary Disorders, complex pathophysiology characterized by a common phenotype of pe‐ King Abdulaziz University, KSA ripheral cytopenia and/or a hypoplastic bone marrow. Understanding ge‐ # to whom correspondence should netic factors contributing to the pathophysiology of BMFS has enabled the be addressed: Prof. Majlinda Lako, identification of causative genes and development of diagnostic tests. To Newcastle University, Institute of date more than 40 mutations in genes involved in maintenance of genomic Genetic Medicine, International stability, DNA repair, ribosome and telomere biology have been identified. Centre for Life, Newcastle NE1 3BZ, In addition, pathophysiological studies have provided insights into several UK, phone: 00 44 191 241 8688, biological pathways leading to the characterization of geno‐ Email: [email protected]; type/phenotype correlations as well as the development of diagnostic ap‐ *JOINT FIRST AUTHORS proaches and management strategies. Recent developments in bone mar‐ row transplant techniques and the choice of conditioning regimens have Received August 29, 2016; accept‐ helped improve transplant outcomes. However, current morbidity and ed for publication October 28, mortality remain unacceptable underlining the need for further research in 2016; available online without sub‐ this area. Studies in mice have largely been unable to mimic disease phe‐ scription through the open access notype in humans due to difficulties in fully replicating the human muta‐ option. tions and the differences between mouse and human cells with regard to telomere length regulation, processing of reactive oxygen species and ©AlphaMed Press lifespan. Recent advances in induced pluripotency have provided novel 1066‐5099/2016/$30.00/0 insights into disease pathogenesis and have generated excellent platforms for identifying signaling pathways and functional mapping of haplo‐ This article has been accepted for insufficient genes involved in large‐scale chromosomal deletions– publication and undergone full associated disorders. In this review we have summarized the current state peer review but has not been of knowledge in the field of BMFS with specific focus on modelling the in‐ through the copyediting, typeset‐ herited forms and how to best utilize these models for the development of ting, pagination and proofreading targeted therapies. STEM CELLS 2016; 00:000–000 process which may lead to differ‐ ences between this version and the SIGNIFICANCE STATEMENT: Version of Record. Please cite this article as doi: 10.1002/stem.2543 Bone marrow failure syndromes are characterized by a common phenotype of peripheral cytopenia and/or a hypoplastic bone marrow. Great strides have been made in the last twenty years both scientifically and clinically re‐ sulting in identification of more than 40 causative genes and improved transplant outcomes. In this review we summarise the most recent findings achieved through application of animal models and stem cells which have led to important insights in disease physiopathology and improved patient care.

STEM CELLS 2016;00:00‐00 www.StemCells.com ©AlphaMed Press 2016 2

INTRODUCTION DBA, SCN & TAR present with single cytopenias that rarely become aplastic but have increased risks of leu‐ Bone marrow failure syndromes (BMFS) are rare dis‐ kemia. Solid tumors like head & neck and anogenital eases characterized by peripheral cytopenias and/or a squamous cell carcinoma are associated with FA & DC hypoplastic bone marrow and can either be inherited or and osteogenic sarcoma with DBA (4, 5). A summary of acquired (1‐3). The purpose of this review is to discuss the clinical features of these diseases together with the methods by which we may create in vitro models of associated mutations and therapeutic options is given in these conditions, therefore we will focus upon inherited Table 1 (6‐23). BMFS since although it is possible to induce bone mar‐ Although rare, the clinical impact of BMFS is un‐ row failure in experimental animals by administration of doubtedly significant. Experimental approaches to in‐ specific chemicals (1), the development of methods to crease our understanding of these disorders are thus replicate the phenotypes of acquired syndromes using essential. Moreover, since many of the causative genes cellular models is currently difficult. In essence, two play important roles in the development and mainte‐ broad mechanisms may be used to generate useful nance of the hematopoietic system, studying their dys‐ models of these diseases. Animal models can work well functions may provide further insights into the produc‐ in cases where the genetic causes of the disease are tion mechanisms of blood and immune cells. Thus in‐ known at the level of DNA sequence since it is feasible vestigations using animal and cellular models of the to engineer experimental animals in which the relevant group of diseases reviewed herein are of great value. mutations play analogous role to their human homo‐ logues. A significant problem with this approach is that Animal models of the bone marrow failure the physiology of experimental animal species is often a syndromes recapitulate only some disease poor match to that of humans and genes known to car‐ features ry disease causing mutations in humans may not always Several strategies may be applied to the generation of have exactly the same mechanistic role in animals such animal models of BMFS with murine models being most as mice or rats. An increasingly attractive alternative is typically applied. Genetically modified mice can be gen‐ the use of pluripotent stem cell technology to create in erated by either direct pronuclear injection of exoge‐ vitro models of disease. Typically, this will involve “re‐ nous DNA into fertilized zygotes or injection of genet‐ programming” patient somatic cells followed by differ‐ ically‐modified murine embryonic stem cells (ESC) into a entiation to the types of cells most affected by the dis‐ blastocyst. Direct pronuclear injection is technically ease. The behavior of these cells in vitro may replicate demanding and often results in multiple, random inte‐ many features of the disease making them a valuable grations of the injected DNA into the genome, and the tool for increasing our understanding and identifying resulting disease phenotypes can vary depending on the potential treatments. We will discuss both of these expression level of the injected transgene. Mouse ESCs modelling options in turn. have the advantage that they can be genetically modi‐ Bone marrow failure syndromes have a broad clini‐ fied by means of homologous recombination, a process cal spectrum, sharing the failure of hematopoietic stem by which a fragment of genomic DNA introduced into a cells (HSCs) to produce functional blood cells and can mammalian cell can recombine with the endogenous affect patients of all ages (1‐3). More than 30 inherited homologous sequence. This process is known as “gene BMFS have been described and although they are gen‐ targeting.” When such genetically modified ES cells are erally rare, conditions such as Fanconi Anemia (FA), introduced into a pre‐implantation embryo, they can Dyskeratosis Congenita (DKC), Diamond‐Blackfan Ane‐ contribute to all cell lineages of the resulting chimeric mia (DBA), Shwachman‐Diamond Syndrome (SDS), Con‐ animal. If this contribution also comprises germ cells genital Amegakaryocytic Thrombocytopenia (CAMT), and the chimeras are capable of breeding, it is possible Severe Congenital Neutropenia (SCN) and Thrombocy‐ to establish lines of animals that are both heterozygous topenia Absent Radii (TAR) are among the most com‐ and homozygous for the genetic alteration introduced mon types. To date, more than 40 mutations in genes into the ESCs. This process can be used to add DNA se‐ involved in maintenance of genomic stability, DNA re‐ quences to specific genomic loci (knock‐ins), a protocol pair and telomere biology have been identified in inher‐ most often used to generate cell lines with gene‐specific ited BMFS. In addition, pathophysiological studies have reporter systems (24), or to create point mutations with provided insights into several biological pathways un‐ pinpoint accuracy. The technology for generating raveling genotype/phenotype correlations, diagnostic mouse models using gene targeting is well developed approaches, and management strategies (1). Given the and while discussion of this technology is outside the association between BMFS and genes involved in DNA scope of this article, excellent reviews on the subject repair mechanisms, it is perhaps unsurprising that many are available (25‐27). of the BMFS have a high predisposition towards malig‐ nancy (4, 5). FA, DKC, SDS and CAMT often present with aplastic anemia and may evolve into myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). www.StemCells.com ©AlphaMed Press 2016 3

Gene editing of murine ESCs has been used to cre‐ these data that loss of function mutations of single ate a number of genetically modified animals carrying genes of the FA pathway in mice do not compromise mutations associated with some of the more common short term survival but rather restrict the capacity of forms of BMFS. Despite the effectiveness of gene edit‐ mice to repair damage induced by environmental in‐ ing, these models do not always demonstrate all the sults or DNA damaging agents. This implies that loss of mechanistic or symptomatic problems shown by hu‐ additional genes might be needed to recapitulate the mans. However, they have generated useful data on the characteristics of human FA. Thus, several double mu‐ mechanisms of BMFS. An overview of animal models tant mouse models have been created to analyze pro‐ created to date is provided in Table 2 (28‐59). We here‐ cesses that may enhance the development of FA. This by discuss individual examples of this approach: approach is exemplified by the observation that while Fanconi anemia (FA): The hallmark of FA is hyper‐ Fancc‐/‐ mice do not develop bone marrow hypocellular‐ sensitivity to DNA cross‐linking agents, intolerance to ity, the Fancc‐/‐ and Sod1‐/‐ double mutants develop this oxidative stress and frequent chromosomal aberrations feature and go on to develop anemia and leucopenia, pointing to a DNA damage response defect, all of which providing some evidence that oxidative stress contrib‐ are used as diagnostic tests. To date 17 complementa‐ utes to bone marrow failure in FA (61). More recently, tion groups have been identified and the genes encod‐ double mutants of Fancd2‐/‐ and Aldh2‐/‐ have been ing these groups A (FANCA), B (FANCB), C (FANCC), D1 generated and these exhibit unusual sensitivity to en‐ (FANCD1/BRCA2), D2 (FANCD2), E (FANCE), F(FANCF), G dogenous aldehydes in utero (34, 35). Ethanol (a source (FANCG), I (FANCI/KIAA1794), J (FANCJ/BRIP1), L of exogenous of acetyldehyde) exposure by postnatal (FANCL), M (FANCM), N (FANCN/PALB2), P double‐deficient mice rapidly precipitates BMFS and (FANCP/SLX4/BTBD12), O (FANCO/RAD51C), S results in spontaneous development of acute leukemia, (FANCS/BRCA1) and T (FANCT/UBE2T) have been cloned suggesting that the FA pathway counteracts acetalde‐ (6‐12,60). The FA core complex (FANCA, B, C, E, F, G, L hyde induced toxicity. Other promising models include and M), functions as an E3 monoubiquitin ligase that the Btbd12 knockout mouse, the orthologue of Slx4 responds to DNA damage or replication stress. The FA (Fancp) which mimics many features of FA including core complex ubiquitinates FANCD2 and FANCI (the ID peripheral cytopenia, reduced fertility, dysmorphic fea‐ complex) which then recruits FAN1 nuclease to help tures, ocular abnormalities, hydrocephalus, chromoso‐ process and cleave the DNA damage, followed by fur‐ mal instability, accumulation of damaged chromo‐ ther recruitment of FANCD1/BRCA2, FANCJ and FANCN somes, hypersensitivity to DNA crosslinking agents and to sites of DNA damage. Once the mono‐ubiquitinated abnormal lymphopoeisis (40). Whilst the data generat‐ FANCD2 and its associated partners (FANCI, FANCD1, ed from such models are interesting, the need to create FANCJ and FANCN) are correctly localized on sites of double knockouts to recapitulate, even in part the phe‐ DNA double strand breaks, they associate with other notype which in humans results solely from mutations DNA repair and checkpoint proteins such as NBS1, ATR, in the FANC genes remains a significant problem. The CHK1 and 2, γ‐H2AX, RAD51 and BRCA1 to initiate DNA potential greater susceptibility of mice to sustain and repair and arrest the cell cycle (4,5). FANC proteins also retain DNA damage or the presence of alternate regula‐ unwind DNA triplexes, remodel DNA structures such as tory mechanisms for FANC proteins in humans, indicate Holliday junctions, D‐loops and replication forks and that murine FA models may not be optimal tools to un‐ channel the DNA lesions towards error free DNA repair derstand the pathophysiology of FA and develop novel achieved by homologous recombination. treatments. Furthermore, the nature of mutations in Targeted single deletions in mouse of various genes various types of FA is extremely heterogeneous, includ‐ such as Fanca, Fancc, Fancd2 and Fancg exhibit de‐ ing point mutations, small insertions/deletions, splicing creased long term HSC repopulating activity and germ mutations and large intragenic deletions, which makes cell loss in addition to cellular sensitivity to DNA inter‐ it difficult to replicate exactly all human mutations strand crosslinks and oxidative stress, but lack the clini‐ through targeted gene knock‐ins/outs in the mouse cal characteristic of FA including; marrow aplasia, he‐ system. matological abnormalities and early life tumorigenesis Dyskeratosis Congenita (DKC): DKC is the first dis‐ (28‐33). Cells cultured from all FA mouse models show order to be etiologically linked to mutations in the te‐ accumulation of chromosomal aberrations when ex‐ lomere pathway (62). About 70% of DKC patients have posed to DNA crosslinking agents, suggesting some de‐ identifiable germ‐line mutations affecting genes re‐ gree of functional conservation of the FA DNA repair sponsible for regulation and maintenance of telomeres pathway between species. Cells present in the spleens (2). To date, nine genes have been associated with DKC of the mutant mice are highly susceptible to accumula‐ phenotype; DKC1, TERT, TERC, TINF2, WRAP 53, NOP10, tion of unrepaired chromosomal aberrations following NHP2, CTC1, and RTEL1 (63‐64). Two categories of mu‐ exposure to DNA crosslinking agents and abnormal sen‐ tations are found in DKC: mutations that decrease te‐ sitivity to IFNγ. Moreover, Fancc‐/‐mice are particularly lomerase activity including those affecting Dyskerin sensitive to the action of the DNA cross linking agent, (DKC1), TERC and TERT, and mutations that impair te‐ Mitomycin C, administration of which causes bone mar‐ lomerase recruitment in genes such as TIN2 (3). The row failure within 3‐8 weeks. A key inference from resultant telomere shortening leads to cell senescence www.StemCells.com ©AlphaMed Press 2016 4 and stem cell exhaustion. Moreover, telomere attrition show enhanced HSC apoptosis but to our knowledge; results in chromosomal fusion and genetic instability, however this has not been attempted in mice. which is at the root of development of secondary ma‐ A further complication with disease modelling of lignancies in affected individuals (4). DKC using targeted gene knockdown in mice is the na‐ Gene editing of murine ESC has been used to model ture of human mutations. In most documented cases to loss of function for most of these genes; however con‐ date, partial loss of function as well as haploinsufficien‐ cerns about differences in telomere maintenance be‐ cy (for example DKC1 in de novo dyskeratosis congenita) tween mouse and human can limit the utility of these and in some cases dominant negative mutations (for models. Although the telomeric DNA sequence is iden‐ example TINF2 in autosomal dominant dyskeratosis tical in both species, abnormalities in telomere mainte‐ congenita) have been reported in addition to loss of nance and in telomerase function do not coincide in function mutations. Whilst, loss of function mutations phenotype in humans and mice. Most strains of labora‐ can be modelled with targeted gene approaches in tory bred mice have telomeres 5‐10 times longer than mice, partial loss of function and haploinsufficiency are humans, whereas absence of telomerase activity is only difficult to mimic unless the human mutation is intro‐ phenotypically present over several generations in mice duced into the mouse germ line or ESC using the most and even heterozygous mutations affecting the te‐ recent gene editing technique (for example Crispr/Cas9 lomerase reverse transcriptase subunit (hTERT) cause method). This however can be easily superseded by the defects in stem cell proliferation, organ regeneration iPSC disease modelling approach which enables the and incidence of cancer in humans. Patients with te‐ assessment of human mutations in a dish by repro‐ lomerase dysfunction including DKC, frequently develop gramming of patient specific somatic cells. aplastic anemia whereas telomerase‐null murine mod‐ The severity of the mutations is also an important els display only modest hematopoietic defects. aspect which cannot always be matched between Early models of DKC (hypomorphic DKC1) display a mouse models and human patients. This is best exem‐ DKC‐like phenotype reflected in increased evidence of plified by a rare and severe form of DKC, Hoyeraal‐ tumors in the mammary glands and lungs (not in gut Hreidarsson syndrome, which is caused by mutations in and skin as in human DKC patients), splenomegaly, dys‐ a subset of genes, including DKC1, TINF2, TPP1 and keratosis of the skin, anemia, yet in the early genera‐ RTEL1 and manifests early in childhood. Human patients tions there is no obvious telomerase dysfunction (43). with two mutations in the TPP1 gene (single amino acid Similar findings have been reported for Tert‐/‐ and Tr‐/‐ deletion + missense mutation) require regular platelet early generations where classical degeneration pheno‐ and red blood cell transfusions for bone marrow failure types and telomere dysfunction could only be achieved (66); however homozygous mouse mutants of Tpp1 after successive mattings which results in substantial were embryonic lethal (67), suggesting that the “blunt” erosion of telomeres and fusion and loss of chromo‐ loss of function created by targeted gene knock‐out in somes (44, 45). These observations are consistent with mice enables creation of much more severe phenotypes the probability that mouse telomeres are too long to compares to human but also result in loss of viability erode sufficiently in a single mouse lifespan to generate preventing further disease modelling. This also raises all the DKC symptoms observed in humans. Deletion of the intriguing question as to why the mouse homozy‐ the RNA template subunit (mTERC) in mice that already gous mutants are lethal if telomeres in mice are much have short telomeres is a more effective means of repli‐ longer than humans, which suggests that the simple cating the DKC phenotype. Similarly, double knockout view of telomere length in stem cell renewal is not as of Pot1b (an orthologue of the “protection of telomeres important as telomere protection from degradation protein” encoding gene) and Terc results in enhanced which is essential for maintaining the ends of chromo‐ telomere degradation (rather than progressive telo‐ somes and genomic stability in proliferating stem and mere shortening) and results in premature death, BMF, progenitor cells. significant anemia, leukopenia and thrombocytopenia. Diamond‐Blackfan anemia (DBA): Mutations in ri‐ That such double knockout is necessary to create a DKC‐ bosomal protein genes that encode structural compo‐ like phenotype calls into question once more the validi‐ nents of the ribosome responsible for the correct as‐ ty of murine models of human disease. Further, the sembly of the ribosomal subunits, are associated with question of how stem cell failure occurs at all in these the abnormal pre‐rRNA maturation patterns in DBA models also arises. Loss or functional failure of HSC patients (68). RPS19 gene mutations are found in 25% clearly occurs after significant erosion of telomeric DNA of DBA patients (69). Mutations resulting in haploinsuf‐ but whether this is due to induction of senescence by ficiency or loss‐of‐function in all genes identified so far critically short telomeres is not yet clear. Other mecha‐ include missense mutations, nonsense mutations, splice nisms to account for the dysfunctional HSC of DKC pa‐ mutations, insertions, deletions and rearrangements tients have been proposed such as defects in ribosome (5). These mutations prevent the assembly of ribosomal biogenesis leading to defects in the processing of 18s protein to form pre‐ribosomal particles, which in turn rRNA (65). This process has been modelled in zebrafish activates nucleolar stress pathways that are at the cen‐ by knockdown of the gene Nop10 and this model does ter of the pathophysiology of DBA. Some of the muta‐ tions in RPS19 can affect the synthesis of the ribosomal www.StemCells.com ©AlphaMed Press 2016 5 protein by altering transcription, splicing, or translation phenotype in the bone marrow of SDS patients remains (70‐72). RPS19 function is essential for correct pro‐ to be identified. cessing within ITS1 and subsequent maturation of the 3’ Severe Congenital Neutropenia (SCN): SCN is char‐ end of 18S rRNA. Inactivation of both copies of RPS19 acterized by a defective neutrophil maturation leading genes in Saccharomyces Cerevisiae yeast leads to com‐ to severe infections. Human SCN‐associated mutations plete arrest of small ribosomal subunit synthesis (73). in ELA2, GFI1 and WASP appear not to be functionally Accumulation of large amounts of 21S pre‐rRNA was null in humans and this creates some difficulty when detected in DBA patients carrying RPS19 mutations. This comparing the phenotype of SCN in humans and animal defect in pre‐RNA maturation was also observed in models. A clear example of this is provided by the Gfi1 CD34‐ hematopoietic precursors, skin fibroblasts, and knockout model which shows a much more severe phe‐ lymphoblastoid cell lines. AS shown in Table 2, Rsp19‐/‐ notype compared to patients with SCN (75). Similarly homozygous mice are embryonic lethal, whilst hetero‐ Ela2‐/‐ mice display impaired neutrophil function, con‐ zygous mice either are normal or show a mild macrocyt‐ sistent with a role for elastase 2 in bacterial killing; ic anaemia (depending on the nature of targeted gene however heterozygous Ela2 mice do not show neutro‐ event), indicating that these models are unable to truly penia. This suggests that to date there is no suitable recapitulate the RPS19 haploinsufficiency described in animal model for mimicking ELA2 causing SCN in man. the human DBA patients. Mice heterozygous for mis‐ For HAX1, all three known human mutations lead to sense mutations of Rsp19 and a similar ribosomal com‐ premature stop codons during the translation stage and ponent Rsp20 show mild macrocytic anemia indicating therefore it was hoped that Hax1‐/‐ knockout models possible HSC dysfunction. Other models including anti‐ would mimic the disease phenotype. Unfortunately, sense oligonucleotide knockdown of Rpl11 in Zebrafish Hax1‐/‐ mice fail to survive past 14 weeks of age, thus show significant effects on early hematopoietic devel‐ obscuring functional studies of this gene in adulthood. opment and defects in adult hematopoiesis such as re‐ duced formation and maturation of erythroid cells The advantages of in vitro modelling which is similar to the phenotype of DBA. Another in‐ From the range of studies described so far, it should be formative model is the conditional Rsp6 mouse where clear that animal models of BMFS are at best only ap‐ homozygous deletions of Rsp6 were achieved through proximations of the etiological mechanisms in humans. CD4‐driven Cre recombinase resulting in abrogation of T Thus, there is a clear and largely unmet need to create cell development. In contrast to homozygous deletions, alternative disease models to further our mechanistic Rps6 haploinsufficiency although it did not impact T cell understanding of BMFS and to explore management maturation, it affected their proliferation. Together approaches. Despite huge efforts, the last decades these data suggest that blood cells with Rsp6 haploin‐ have witnessed an alarming failure to develop new sufficiency may cope with ribosome synthesis under low treatments. At the center of this dilemma is the fact cell proliferation; however this is severely compromised that animal disease models do not mirror human dis‐ in tissues characterized by rapid proliferation (such as ease, thus the drug is either ineffective or leads to un‐ the hematopoietic system). acceptable toxicities in humans. Drug discovery based Shwachman‐Bodian‐Diamond (SDS): Knockouts of on affected human models would yield superior results, two genes implicated in Shwachman‐Bodian‐Diamond but unfortunately affected tissues are often accessible syndrome (Sbds) and ribosomal protein 18 (Rps18) only on deceased patients. Such tissues are unsuitable which are central to ribosome function are unsurpris‐ for testing drug toxicity and efficacy as they reflect end ingly embryonic lethal (74). Approximately 90% of SDS stages of the disease. This is particularly true for inher‐ patients have mutations in the SBDS gene which shares ited BMFS since ex vivo expansion of HSC from aplastic 97% homology with an adjacent pseudogene, SBDSP bone marrow is challenging (76). which contains deletions and nucleotide changes pre‐ The advent of human pluripotent stem cell technol‐ venting generation of a functional protein. 75% of pa‐ ogy may offer a solution to these problems. Human tients with SDS have SBDS mutations which are ac‐ embryonic stem cells (hESCs) and more latterly, induced quired as result of gene conversion events from the pluripotent stem cells (iPSCs), are capable of indefinite pseudogene SBDSP. Patients homozygous for SBDS mu‐ in vitro expansion while retaining the ability to differen‐ tations have not been identified, suggesting that com‐ tiate into cell types characteristics of most tissues found plete loss of SBDS is likely to be lethal in humans. This is in the developing embryo. Reprogramming somatic corroborated by animal studies which have shown that cells to pluripotency by ectopic expression of four tran‐ loss of Sbds gene in mice results in early embryonic le‐ scription factors expressed by ESC was first reported in thality (Table 2), whilst heterozygous mice are com‐ 2006 (77). The products of such reprogramming were pletely normal, indicating once more the inability of termed “induced pluripotent stem cells” or iPSC by their heterozygous mouse mutants to mimic the precise na‐ originator, Shinya Yamanaka. In the following year, ture of human SBDS mutations. Whilst studies in yeast human iPSC (hiPSCs) were reported by Yamanaka’s and have clearly shown disruption of ribosome biogenesis as two other groups (78, 79). The pluripotency of iPSCs result of SBDS dysfunction, how this leads to a specific confers properties on these cells particularly interesting for clinical research since iPSCs can be patient‐specific www.StemCells.com ©AlphaMed Press 2016 6 and disease‐specific, thus providing a promising plat‐ lower compared to fibroblasts taken from unaffected form for investigating the pathophysiology of a specific controls (94). These hiPSCs showed frequent chromo‐ disease and testing the effectiveness and toxicity of somal abnormalities and were unable to generate bona‐ drugs in the cell of interest. Despite the need to over‐ fide teratomae with contribution to cell types repre‐ come some potential hurdles such as the acquisition of sentative of all three embryonic germ layers. Despite de novo genetic alterations during reprograming, possi‐ this, the FA‐C hiPSCs were able to differentiate into ble retention of a small number epigenetic marks pre‐ hematopoietic progenitors albeit ones with significantly sent in the somatic parent cells and aberrant DNA reduced clonogenicity and enhanced levels of apopto‐ methylation acquired during the reprogramming pro‐ sis. A greater percentage of the phenotypic hematopoi‐ cess, which may affect the cell differentiation capabili‐ etic progenitors from FA patient specific iPSC were in S‐ ties, there is an increasing awareness of the value of phase of the cell cycle indicating a higher rate of prolif‐ iPSC based disease modelling as a tool to further drug eration and the possibility that this is a compensatory discovery (80‐83). response to the higher levels of apoptosis. Replenish‐ Differentiation of pluripotent stem cells towards pa‐ ment of hematopoietic progenitors at this higher rate tient specific hematopoietic cell types is the basis of contributes to the development of more chromosomal modelling inherited BMFS. However, the major chal‐ abnormalities and replicative exhaustion of the stem lenge is the generation of fully functional HSCs. Several cell population. Whilst the increased apoptosis of hem‐ studies have described the generation of hematopoietic atopoietic progenitors and accumulation of genomic progenitors, but on further differentiation in both in instabilities were features already described in mouse vitro and in vivo experimental models, such progenitors models, the role of FA pathway during somatic cell‐ show a preference towards myeloid differentiation so induced reprogramming and the availability of FA‐ high long term engraftment efficiency and multi‐lineage specific cells for drug discovery are novel aspects which differentiation of functional HSC in the bone marrow of could only be achieved through the hiPSCs disease immune compromised mice remains challenging (84‐ modelling approach. For regenerative medicine, caution 90). Several patient specific hiPSCs lines from patients should be exercised to obtain patient specific cells as with inherited BMFS have now been established. To early as possible and to ensure they are genetically maintain clarity with respect to our discussion of animal complemented with the functional alleles before the models we will describe hiPSCs based models of BMFS reprogramming and differentiation process in order to in the same order before discussing additional in vitro maintain genomic stability. modelling attempts. Dyskeratosis congenita: hiPSCs have been derived Fanconi anemia: The processes of reprogramming from DKC patients with mutations in DKC1, TERC, TERT needed to generate hiPSCs is dependent on enhanced and TCAB1 which display normal karyotypes and hall‐ cellular proliferation during the initial stages of repro‐ marks of pluripotency as described earlier (95‐98). Since gramming which requires the activation of DNA repair telomerase activity is normally increased during the pathways to ensure maintenance of genomic stability. reprogramming process, by upregulation of the catalytic Thus, it is quite difficult to derive hiPSCs from somatic subunit hTERT, one might expect that DKC specific iPSC cells of patients suffering from DNA repair disorders, would be less useful as disease models. This is probably which renders the development of a hiPSCs based mod‐ true for DKC cases involving mutations in the RNA tem‐ el of FA difficult to achieve. Initial studies suggested plate component (TERC) since hTERT reactivation ap‐ that reprogramming of FA patient specific hiPSCs was pears to be sufficient to overcome the reduction in te‐ only possible after genetic complementation of the do‐ lomerase activity resulting from the TERC mutations. nor fibroblasts prior to reprogramming or when repro‐ However, attempted upregulation of an hTERT gene gramming was performed under hypoxic conditions (91, carrying heterozygous mutations merely produces dys‐ 92). The increased incidence of unrepaired DNA double functional hTERT protein, which cannot function effec‐ strand breaks in FA patients was thought to contribute tively within the telomerase holoenzyme complex. For to this. An alternative approach to modelling FA is to this reason, hiPSCs lines derived from patients with such knockdown the expression of FA genes in otherwise hTERT mutations display shortened telomeres following healthy hESCs and some data were generated by Tul‐ extended culture which ultimately prevents their self‐ pule et al by RNA interference mediated knockdown of renewal. This phenotype is more pronounced in hiPSCs FANCD2 and FANCA (93). Differentiation of hESCs sub‐ lines derived from patients with X‐linked DKC where the jected to this approach yielded hematopoietic progeni‐ mutation of DKC1 blocks telomerase assembly and dis‐ tors with reduced ability to mature into functional rupts telomere elongation. The severity of the telomer‐ CD45+ cells. Furthermore, these FA‐deficient hemato‐ ase dysfunction depends upon the precise nature of the poietic progenitors displayed an abnormal transition of DKC1 mutation; hiPSCs lines carrying Q31E and ΔL37 embryonic to adult globin expression, indicating an im‐ mutations maintained the same telomere length as the paired hematopoietic development with a bias towards parental fibroblasts; however the A353V mutant DKC‐ primitive populations. Yung and colleagues succeeded hiPSCs presented shorter telomeres compared to the to reprogram FA‐C fibroblasts under normoxic condi‐ parental fibroblasts suggesting that this A353V muta‐ tions, although the reprogramming efficiency was much www.StemCells.com ©AlphaMed Press 2016 7 tion has a more severe effect on the telomere mainte‐ This phenotype could be rescued by retroviral transduc‐ nance process. tion of HAX1 and ELA2 in the respective isogenic hiPSCs Diamond‐Blackfan anemia: DBA‐hiPSCs have been lines. In a recent study, Nayak et al identified the mislo‐ generated from patients with mutations in RPS19 and calization of the ELA2 gene encoding protein neutrophil RPL5 genes (99). These DBA‐hiPSCs lines exhibited de‐ elastase (NE) as the inductor of the UPR/ER stress, dys‐ fective assembly of ribosomal subunits and production functional differentiation and apoptosis and demon‐ of ribosomal RNA as well as impaired generation of strated that the SCN phenotype could be corrected by hematopoietic progenitors with erythroid lineages be‐ addition of the NE inhibitor called sivelestat to the cul‐ ing the most affected. These features recapitulate the ture media of the hiPSC‐derived myeloid progenitors. phenotype observed in DBA patients, thus providing an Together these studies indicate that SCN‐hiPSCs provide excellent tool to investigate DBA. Genetic correction of an excellent platform for high‐throughput screening of the affected genes restored the expression of deficient drugs to reverse various congenital neutrophil disorders ribosomal proteins resulting in an increase of the ribo‐ as well as better understanding of disease pathogenesis some biogenesis, normal generation of hematopoietic which can be clinically exploited to achieve therapeutic progenitors and erythroid cells. Using patient specific responses using lower doses of G‐CSF combined with lines and hematopoietic differentiation as a model, the targeting to correct NE mislocalization. same authors described for the first time dysregulation Recent progress in the field of genome editing is giv‐ of non‐canonical TGFβ signaling pathway mediated by ing iPSC technology a solid framework for disease mod‐ p‐JNK which may be the underlying cause for abnormal elling and for future clinical therapies. Previous gene hematopoiesis in DBA‐hiPSCs, for TGFβ signaling is in‐ therapies carried out via transient or constitutive ex‐ hibitory to hematopoietic commitment (100). These pression of transgenes were undermined by the possi‐ studies clearly indicate that patient specific hiPSCs can bility of transgene integration into loci that conferred provide novel insights into the disease pathogenesis as uncontrolled cellular proliferation resulting in tumor‐ well as a platform for regulating critical pathways in‐ igenesis. To date the development robust and efficient volved in hematopoiesis and erythroid differentiation human genome engineering tools including zinc‐finger via the use of small molecules. nucleases (ZFNs), transcription activator‐like effector Shwachman‐Bodian‐Diamond: hiPSCs have been nuclease (TALEN) and Crispr‐Cas9 systems which permit generated from patients with reduced SBDS expression in situ gene editing with reduced risk of off‐site muta‐ which display dysfunctional ribosome assembly (101). genesis have propelled forward the genome engineer‐ Upon differentiation to pancreatic cell lineages, the ing approaches (104). In particular, the Crispr/Cas9 SDS‐hiPSCs and hESCs showed increased cell death and system stands out from the other gene editing tools for impaired organization of the acinar‐like structures, its precision, versatility and simplicity. Early applications which led to progressive loss of exocrine tissue. SDS‐ of in situ gene editing approaches involved introduction hiPSCs were able to differentiate to mesodermal line‐ of specific gene mutations for disease modelling studies ages; however a reduced percentage of hematopoietic in various species ranging from fruit flies to Man (105‐ cells and colony forming potential was noted together 107). Currently genome editing tools are being used to with increased levels of proteases in culture superna‐ correct genetic defect in iPSC harboring disease‐causing tant. The residual SBDS expression in some patient spe‐ mutations providing insights on the pathophysiology of cific lines led to absence of the pancreatic phenotype hematological disorders (such as β‐thalassemia or he‐ and presence of hematopoietic defects only, similar to mophilia) and confirming disease causality of the cor‐ previous clinical observations in SDS patients, thus fur‐ rected genetic defect (108‐112). ther corroborating the genotype‐phenotype correlation. To date, a handful of studies have reported the suc‐ Despite this observed variability, both the pancreatic cessful use of genome editing tools for the study of and/or hematopoietic phenotype could be rescued ei‐ BMFS. For example, ZFN mediated disruption of the ther by the forced expression of the SBDS transgene or FANCA gene in wildtype hESC to create a FA disease by the use of protease inhibitors in the culture media, model and restoration of specific mutations in the thus providing a clear example of drug‐reversible phe‐ FANCC gene in wild type fibroblasts via the Crispr/Cas9 notype using the hiPSC model for these patients and system have been reported (113, 114). Recently, Blu‐ providing novel therapeutic insights. teau et al inactivated the REV7 using the Crispr/Cas9 Severe congenital neutropenia: Several groups have approach and showed that this resulted in increased reported the generation of patient specific hiPSCs carry‐ cellular hypersensitivity to DNA inter‐strand cross‐link ing mutations in genes involved in SCN such as ELA2 and drugs (ICL) as well as an impaired ability of the mouse HAX1 (102‐103). Upon differentiation to neutrophils hematopoietic progenitor cells to form hematopoietic using a defined feeder free method, SCN‐hiPSCs lines colonies in CFU assay, resulting in characterization of a show developmental arrest at the myeloid progenitor new gene in the FA pathway (115). Combination of ge‐ stage resulting in decreased pool of myeloid progeni‐ nome editing tools with the iPSC technology has the tors, impaired response to G‐CSF stimuli as well as in‐ potential to result in generation of disease free iPSC creased levels of apoptosis in neutrophils, reminiscent derived haematopoetic cells that can be used in poten‐ of abnormal granulopoiesis observed in SCN patients. tial cell based therapies. Rio et al reported in 2014 the www.StemCells.com ©AlphaMed Press 2016 8 successful phenotypic correction of FANCA defective MK, platelets and erythrocytes to similar levels as in fibroblasts and later generation of disease‐free iPSC control hiPSCs. Interestingly, excessive expression of which upon haematopoetic differentiation could gener‐ MPL transgene led to a dysregulation of thrombopoiesis ate similar number of hematopoietic colonies compared and defective megakaryopoiesis suggesting that MPL to healthy cord blood progenitor cells (116). Likewise, signaling is finely regulated and levels of expression the disease‐free iPSC‐derived haematopoetic progeni‐ must fall within a precise expression window for normal tors proved to be resistant to exposure to DNA ICL development of thrombopoiesis. These insights gained drugs thus demonstrating the reversal of disease phe‐ through the hiPSC approach are particularly important notype using novel and safe gene editing tools. for designing curative gene correction strategies which Application of iPSC technology as a platform for should aim to restore an appropriate level of MPL ex‐ gene correction and drug screening was also reported pression. for DKC (117). By using CRISP/Cas9 genome editing sys‐ MDS‐hiPSC: MDS is the most common form of pri‐ tem Woo et al. reported the genetic correction of muta‐ mary BMF. The somatic loss of the long arm of chromo‐ tion in DKC1 in patient specific IPSC as well as introduc‐ some 7 (del (7q) is one of the most characteristic chro‐ tion of DKC1 mutation in wild type iPSC. As expected, mosomal abnormalities in MDS. Recently, Kotani et al. the corrected DKC1‐iPSC showed high telomerase activi‐ derived hiPSCs clones that displayed normal karyotype ty, whereas DKC1 mutant‐ iPSC showed lower telomer‐ as well as clones with del(7q) from MDS patients (119). ase activity than wildtype controls. Interestingly, this Whereas normal isogenic hiPSCs were able to generate study also provided data that supported the link be‐ hematopoietic progenitors, MDS‐hiPSCs displayed a tween DKC1 and Wnt signaling pathway by showing reduced hematopoietic differentiation potential and restoration of telomere length, telomere capping and clonogenic capacity in all myeloid lineages. Further‐ reduction of p53BP1 telomere foci upon application of more, MDS‐hiPSCs showed an increase in cell death rate the Wnt agonist CHIR99021 in DKC‐iPSC, indicating that during differentiation. These observations are con‐ both gene editing and manipulation of signalling path‐ sistent with the phenotype observed in primary MDS way provide potential avenues for treatment of BMFS. cells. Interestingly, the authors observed a spontaneous To date, allogeneic HSC transplantation from HLA‐ dosage correction in one of the clones which acquired a identical healthy donors remains the only curative ther‐ 30 Mb region of a telomeric part of chromosome 7q apy for BMFS patients. The possibility of performing (chr7q) and upon differentiation showed a fully re‐ gene correction approaches in patients HSCs together stored hematopoietic potential to a level similar to with an improved yield of HSC expansion so that suffi‐ normal hiPSCs. This suggests that the defect responsible cient numbers of cells for autologous transplantation for the abnormal hematopoietic phenotype is localized can be achieved, offers a revolutionary prospect for the in this region. Gene expression comparison between field of BMFS. This will require a reduction in off‐target del(7q) MDS clones and chr7q dosage corrected clones mutagenesis; however with the pace the field is pro‐ allowed the authors to identify 4 haploinsuficient can‐ gressing, one would hope that this is within our reach didate genes whose forced expression could partially very soon in the future. rescue the hematopoietic defects in del(7q) MDS iPSC. This study offers an interesting approach for functional Generation of in vitro cellular models that mapping and identification of haplo‐insufficient genes have not been modelled in animals involved in large‐scale chromosomal deletions– associated disorders such as MDS using the hiPSC tech‐ CAMT‐hiPSC: CAMT is characterized by the loss of func‐ nology. tion or deletion of the thrombopoietin (TPO) receptor encoded by the MPL gene, resulting in loss of megakar‐ SUMMARY yocytes (MK) in the bone marrow, severe thrombocyto‐ penia and development of fatal BMF later in life. Hirata In the last twenty years, great progress has been et al investigated the mechanisms involved in the MPL achieved in identifying new genetic caus‐ signaling and the development of MK/Erythrocyte pro‐ es/susceptibilities of inherited BMFS, which have re‐ genitor (MEP) by generating hiPSCs from CAMT patients sulted in better genotype‐phenotype correlations, im‐ carrying MPL mutations (118). The resultant CAMT‐ proved diagnosis and clinical treatments. A large num‐ hiPSCs failed to generate MKs and platelets after hema‐ ber of mouse models have been created enabling gen‐ topoietic differentiation, thus recapitulating the typical eration of insights into disease mechanisms and pathol‐ feature of human disease: thrombocytopenia. Likewise, ogy. Nevertheless, the inability of gene knock‐ins/out to CAMT‐hiPSCs displayed impaired colony formation ca‐ mimic the precise nature of human mutations and es‐ pability especially with regard to erythrocytes and pecially happloinsufficiency together with differences megakaryocytes, demonstrating a critical role for MPL DNA damage tolerance, telomere length and short life signaling in the formation of the common MEP progeni‐ spans have undermined their utility in mimicking the tor which is able to generate both lineages. Retroviral full spectrum of BMFS (Figure 1). The advent of in‐ transduction of MPL transgene in CAMT‐hiPSCs restored duced pluripotency enabling generation of patient spe‐ the differentiation potential of the CAMT to generate cific hematopoietic cells carrying the genetic change www.StemCells.com ©AlphaMed Press 2016 9 responsible for the disease has revolutionized the field ACKNOWLEDGMENTS and has provided a functional platform for uncovering new signaling pathways and small molecules that can The authors are grateful to Newcastle University, ERC be explored therapeutically. The tremendous advances (614620) and Deanship of Scientific Research (DSR) King made in the genetic engineering field has enabled in Abdul Aziz University (grant number 1343‐287‐1‐HiCi) situ gene corrections in patients fibroblasts, hiPSC and for their technical and financial support. HSCs opening new avenues towards generation of au‐ tologous cell replacement therapies for BMFS. Several AUTHOR CONTRIBUTIONS improvements need to be achieved towards the gener‐ ation of long term reconstituting HSC from hiPSC, effi‐ S.H.: Conception and design, manuscript writing, final cient ex vivo expansion of HSCs obtained from adult approval of manuscript; D.M.S.: Conception and design, bone marrow and reducing the off‐site targets of in situ manuscript writing, final approval of manuscript; G.E.K.: gene editing; nevertheless the fast pace of progress manuscript writing, final approval of manuscript; S.S.: made in the field of stem cells and genetic engineering Conception and design, manuscript writing, final ap‐ boasts for a bright and promising future for treatment proval of manuscript, fund raising; S.A.: manuscript of BMFS which could have not been anticipated at the writing, final approval of manuscript; L.A.: Conception beginning of this century! and design, manuscript writing, final approval of manu‐ script, fund raising; M.L.: Conception and design, manu‐ script writing, final approval of manuscript, fund raising

REFERENCES FANCG with BRCA2/FANCD1. Hum Mol Genet. Proceedings of the National Academy of 2003;12(19):2503‐10. Sciences of the United States of America.

10 Moynahan ME, Pierce AJ, Jasin M. 1999;96(6):3132‐6. 1 Chen J. Animal models for BRCA2 is required for homology‐directed 19 Welte K, Zeidler C. Severe congenital acquired bone marrow failure repair of chromosomal breaks. Molecular cell. neutropenia. Hematology/oncology clinics of syndromes. Clin Med Res. 2001;7(2):263‐72. North America. 2009;23(2):307‐20. 2005;3(2):102‐8. 11 Folias A, Matkovic M, Bruun D, Reid S, 20 Dale DC, Person RE, Bolyard AA, Hejna J, Grompe M, et al. BRCA1 interacts Aprikyan AG, Bos C, Bonilla MA, et al. 2 Tamary H, Nishri D, Yacobovich J, Zilber directly with the Fanconi anemia protein Mutations in the gene encoding neutrophil R, Dgany O, Frequency and natural history of FANCA. Hum Mol Genet. 2002;11(21):2591‐7. elastase in congenital and cyclic neutropenia. inherited bone marrow failure syndromes: 12 Sommers JA, Rawtani N, Gupta R, Blood. 2000;96(7):2317‐22. the Israeli Inherited Bone Marrow Failure Bugreev DV, Mazin AV, Cantor SB, et al. 21 Albers CA, Paul DS, Schulze H, Freson K, Registry.Haematologica. 2010 FANCJ uses its motor ATPase to destabilize Stephens JC, Smethurst PA, et al. Compound Aug;95(8):1300‐7. protein‐DNA complexes, unwind triplexes, inheritance of a low‐frequency regulatory 3 Blanche P. Alter, Neelam Giri, Sharon A. and inhibit RAD51 strand exchange. The SNP and a rare null mutation in exon‐junction et al. Malignancies and survival patterns in Journal of biological chemistry. complex subunit RBM8A causes TAR the National Cancer Institute inherited bone 2009;284(12):7505‐17. syndrome. Nature genetics. 2012;44(4):435‐9, marrow failure syndromes cohort study. Br J 13 Vulliamy T, Dokal I. Dyskeratosis S1‐2. Haematol. 2010 Jul; 150(2): 179–188 congenita. Seminars in hematology. 22 Castillo‐Caro P, Dhanraj S, Haut P, 4 Parikh S, Bessler M. Recent insights into 2006;43(3):157‐66. Robertson K, Dror Y, Sharathkumar AA. inherited bone marrow failure syndromes. 14 Vulliamy TJ, Knight SW, Mason PJ, Dokal Proximal radio‐ulnar synostosis with bone Current opinion in pediatrics. 2012;24(1):23‐ I. Very short telomeres in the peripheral marrow failure syndrome in an infant without 32. blood of patients with X‐linked and autosomal a HOXA11 mutation. Journal of pediatric 5 Rosenberg PS, Greene MH, Alter BP. dyskeratosis congenita. Blood cells, molecules hematology/oncology. 2010;32(6):479‐85. Cancer incidence in persons with Fanconi & diseases. 2001;27(2):353‐7. 23 Gagne KE, Ghazvinian R, Yuan D, Zon RL, anemia. Blood. 2003;101(3):822‐6. 15 Konno Y, Toki T, Tandai S, Xu G, Wang R, Storm K, Mazur‐Popinska M, et al. Pearson 6 Meetei AR, Medhurst AL, Ling C, Xue Y, Terui K, et al. Mutations in the ribosomal marrow pancreas syndrome in patients Singh TR, Bier P, et al. A human ortholog of protein genes in Japanese patients with suspected to have Diamond‐Blackfan anemia. archaeal DNA repair protein Hef is defective Diamond‐Blackfan anemia. Haematologica. Blood. 2014;124(3):437‐40. in Fanconi anemia complementation group 2010;95(8):1293‐9. 24 Collin J, Mellough CB, Dorgau B, M. Nature genetics. 2005;37(9):958‐63. 16 Boocock GR, Morrison JA, Popovic M, Przyborski S, Moreno‐Gimeno I, Lako M. 7 Levitus M, Waisfisz Q, Godthelp BC, de Richards N, Ellis L, Durie PR, et al. Mutations Using Zinc Finger Nuclease Technology to Vries Y, Hussain S, Wiegant WW, et al. The in SBDS are associated with Shwachman‐ Generate CRX‐Reporter Human Embryonic DNA helicase BRIP1 is defective in Fanconi Diamond syndrome. Nature genetics. Stem Cells as a Tool to Identify and Study the anemia complementation group J. Nature 2003;33(1):97‐101. Emergence of Photoreceptors Precursors genetics. 2005;37(9):934‐5. 17 Germeshausen M, Ballmaier M, Welte K. During Pluripotent Stem Cell Differentiation. 8 Meetei AR, Levitus M, Xue Y, Medhurst MPL mutations in 23 patients suffering from Stem Cells. 2016;34(2):311‐21. AL, Zwaan M, Ling C, et al. X‐linked congenital amegakaryocytic 25 Bouabe H, Okkenhaug K. Gene targeting inheritance of Fanconi anemia thrombocytopenia: the type of mutation in mice: a review. Methods in molecular complementation group B. Nature genetics. predicts the course of the disease. Human biology (Clifton, N J ). 2013;1064:315‐36. 2004;36(11):1219‐24. mutation. 2006;27(3):296. 26 Zimmermann AG, Sun Y. Conventional 9 Hussain S, Witt E, Huber PAJ, Medhurst 18 Ihara K, Ishii E, Eguchi M, Takada H, murine gene targeting. Methods in molecular AL, Ashworth A, Mathew CG. Direct Suminoe A, Good RA, et al. Identification of biology (Clifton, N J ). 2013;1031:1‐18. interaction of the Fanconi anaemia protein mutations in the c‐mpl gene in congenital 27 Hall B, Limaye A, Kulkarni AB. Overview: amegakaryocytic thrombocytopenia. generation of gene knockout mice. Curr www.StemCells.com ©AlphaMed Press 2016 10

Protoc Cell Biol. 2009;Chapter 19:Unit 19.2 41 The Mouse Genome Database Group. 55 Grenda DS, Johnson SE, Mayer JR, .2.1‐7. 2015. The Mouse Genome Database (MGD): McLemore ML, Benson KF, Horwitz M, et al. 28 Wong JCY, Alon N, McKerlie C, Huang JR, facilitating mouse as a model for human Mice expressing a neutrophil elastase Meyn MS, Buchwald M. Targeted disruption biology and disease. [database on the mutation derived from patients with severe of exons 1 to 6 of the Fanconi Anemia group Internet]. June 28, 2015 [cited September 18, congenital neutropenia have normal A gene leads to growth retardation, strain‐ 2015]. Available from: granulopoiesis. Blood. 2002;100(9):3221‐8. specific microphthalmia, meiotic defects and http://www.informatics.jax.org/marker/MGI: 56 Chao J‐R, Parganas E, Boyd K, Hong CY, primordial germ cell hypoplasia. Hum Mol 2150020. Opferman JT, Ihle JN. Hax1‐mediated Genet. 2003;12(16):2063‐76. 42 Gowen LC, Johnson BL, Latour AM, Sulik processing of HtrA2 by Parl allows survival of 29 Noll M, Battaile KP, Bateman R, Lax TP, KK, Koller BH. Brca1 deficiency results in early lymphocytes and neurons. Nature. Rathbun K, Reifsteck C, et al. Fanconi anemia embryonic lethality characterized by 2008;452(7183):98‐102. group A and C double‐mutant mice: neuroepithelial abnormalities. Nature 57 Karsunky H, Zeng H, Schmidt T, Zevnik B, functional evidence for a multi‐protein genetics. 1996;12(2):191‐4. Kluge R, Schmid KW, et al. Inflammatory Fanconi anemia complex. Exp Hematol. 43 He J, Navarrete S, Jasinski M, Vulliamy T, reactions and severe neutropenia in mice 2002;30(7):679‐88. Dokal I, Bessler M, et al. Targeted disruption lacking the transcriptional repressor Gfi1. 30 Chen M, Tomkins DJ, Auerbach W, of Dkc1, the gene mutated in X‐linked Nature genetics. 2002;30(3):295‐300. McKerlie C, Youssoufian H, Liu L, et al. dyskeratosis congenita, causes embryonic 58 Westerberg LS, Meelu P, Baptista M, Inactivation of Fac in mice produces inducible lethality in mice. Oncogene. Eston MA, Adamovich DA, Cotta‐de‐Almeida chromosomal instability and reduced fertility 2002;21(50):7740‐4. V, et al. Activating WASP mutations reminiscent of Fanconi anaemia. Nature 44 Jaskelioff M, Muller FL, Paik J‐H, Thomas associated with X‐linked neutropenia result in genetics. 1996;12(4):448‐51. E, Jiang S, Adams AC, et al. Telomerase enhanced actin polymerization, altered 31 Pulliam‐Leath AC, Ciccone SL, Nalepa G, reactivation reverses tissue degeneration in cytoskeletal responses, and genomic Li X, Si Y, Miravalle L, et al. Genetic disruption aged telomerase‐deficient mice. Nature. instability in lymphocytes. J Exp Med. of both Fancc and Fancg in mice recapitulates 2011;469(7328):102‐6. 2010;207(6):1145‐52. the hematopoietic manifestations of Fanconi 45 Lee HW, Blasco MA, Gottlieb GJ, Horner 59 Davis AP, Witte DP, Hsieh‐Li HM, Potter anemia. Blood. 2010;116(16):2915‐20. JW, 2nd, Greider CW, DePinho RA. Essential SS, Capecchi MR. Absence of radius and ulna 32 Suzuki A, de la Pompa JL, Hakem R, Elia role of mouse telomerase in highly in mice lacking hoxa‐11 and hoxd‐11. Nature. A, Yoshida R, Mo R, et al. Brca2 is required for proliferative organs. Nature. 1995;375(6534):791‐5. embryonic cellular proliferation in the mouse. 1998;392(6676):569‐74. 60 Rickman KA, Lach FP, Abhyankar Genes Dev. 1997;11(10):1242‐52. 46 Frescas D, de Lange T. A TIN2 A, Donovan FX, Sanborn EM et al. Deficiency 33 Zhang Q‐S, Marquez‐Loza L, Eaton L, dyskeratosis congenita mutation causes of UBE2T, the E2 Ubiquitin Ligase Necessary Duncan AW, Goldman DC, Anur P, et al. telomerase‐independent telomere shortening for FANCD2 and FANCI Ubiquitination, Causes Fancd2‐/‐ mice have hematopoietic defects in mice. Genes Dev. 2014;28(2):153‐66. that can be partially corrected by resveratrol. 47 Ding H, Schertzer M, Wu X, Gertsenstein FA‐T Subtype of Fanconi Anemia. Cell Blood. 2010;116(24):5140‐8. M, Selig S, Kammori M, et al. Regulation of Rep. 2015 Jul 7;12(1):35‐41. 34 Langevin F, Crossan GP, Rosado IV, murine telomere length by Rtel: an essential 61 Hadjur S, Ung K, Wadsworth L, Dimmick Arends MJ, Patel KJ. Fancd2 counteracts the gene encoding a helicase‐like protein. Cell. J, Rajcan‐Separovic E, Scott RW, et al. toxic effects of naturally produced aldehydes 2004;117(7):873‐86. Defective hematopoiesis and hepatic in mice. Nature. 2011;475(7354):53‐8. 48 Gu P, Min J‐N, Wang Y, Huang C, Peng T, steatosis in mice with combined deficiencies 35 Garaycoechea JI, Crossan GP, Langevin F, Chai W, et al. CTC1 deletion results in of the genes encoding Fancc and Cu/Zn Daly M, Arends MJ, Patel KJ. Genotoxic defective telomere replication, leading to superoxide dismutase. Blood. consequences of endogenous aldehydes on catastrophic telomere loss and stem cell 2001;98(4):1003‐11. mouse haematopoietic stem cell function. exhaustion. The EMBO journal. 62 Mitchell JR, Wood E, Collins K. A Nature. 2012;489(7417):571‐5. 2012;31(10):2309‐21. telomerase component is defective in the 36 Yang Y, Kuang Y, Montes De Oca R, Hays 49 Sulic S, Panic L, Barkic M, Mercep M, human disease dyskeratosis congenita. T, Moreau L, Lu N, et al. Targeted disruption Uzelac M, Volarevic S. Inactivation of S6 Nature. 1999;402(6761):551‐5. of the murine Fanconi anemia gene, ribosomal protein gene in T lymphocytes 63 Ballew BJ, Yeager M, Jacobs K, Giri N, Fancg/Xrcc9. Blood. 2001;98(12):3435‐40. activates a p53‐dependent checkpoint Boland J, Burdett L, et al. Germline mutations 37 The Mouse Genome Database Group. response. Genes Dev. 2005;19(24):3070‐82. of regulator of telomere elongation helicase 2015. The Mouse Genome Database (MGD): 50 Watkins‐Chow DE, Cooke J, Pidsley R, 1, RTEL1, in Dyskeratosis congenita. Hum facilitating mouse as a model for human Edwards A, Slotkin R, Leeds KE, et al. Genet. 2013;132(4):473‐80. biology and disease [database on the Mutation of the diamond‐blackfan anemia 64 Savage SA, Giri N, Baerlocher GM, Orr N, Internet]. June 28, 2015 [cited September 18, gene Rps7 in mouse results in morphological Lansdorp PM, Alter BP. TINF2, a component 2015]. Available from: and neuroanatomical phenotypes. PLoS of the shelterin telomere protection complex, http://www.informatics.jax.org/marker/MGI: genetics. 2013;9(1):e1003094. is mutated in dyskeratosis congenita. Am J 2384790. 51 Matsson H, Davey EJ, Draptchinskaia N, Hum Genet. 2008;82(2):501‐9. 38 Rantakari P, Nikkila J, Jokela H, Ola R, Hamaguchi I, Ooka A, Leveen P, et al. 65 Pereboom TC, van Weele LJ, Bondt A, Pylkas K, Lagerbohm H, et al. Inactivation of Targeted disruption of the ribosomal protein MacInnes AW. A zebrafish model of Palb2 gene leads to mesoderm differentiation S19 gene is lethal prior to implantation. Mol dyskeratosis congenita reveals hematopoietic defect and early embryonic lethality in mice. Cell Biol. 2004;24(9):4032‐7. stem cell formation failure resulting from Hum Mol Genet. 2010;19(15):3021‐9. 52 McGowan KA, Li JZ, Park CY, Beaudry V, ribosomal protein‐mediated p53 stabilization. 39 Bakker ST, van de Vrugt HJ, Rooimans Tabor HK, Sabnis AJ, et al. Ribosomal Blood. 2011;118(20):5458‐65. MA, Oostra AB, Steltenpool J, Delzenne‐ mutations cause p53‐mediated dark skin and 66 Kocak H, Ballew BJ, Bisht K, Eggebeen Goette E, et al. Fancm‐deficient mice reveal pleiotropic effects. Nature genetics. R, Hicks BD, et al. Hoyeraal‐Hreidarsson syn‐ unique features of Fanconi anemia 2008;40(8):963‐70. drome caused by a germline mutation in the complementation group M. Hum Mol Genet. 53 Zhang S, Shi M, Hui C‐C, Rommens JM. TEL patch of the telomere protein TPP1. 2009;18(18):3484‐95. Loss of the mouse ortholog of the 40 Crossan GP, van der Weyden L, Rosado shwachman‐diamond syndrome gene (Sbds) Genes Dev. 2014 Oct 1;28(19):2090‐ IV, Langevin F, Gaillard P‐HL, McIntyre RE, et results in early embryonic lethality. Mol Cell 102. al. Disruption of mouse Slx4, a regulator of Biol. 2006;26(17):6656‐63. 67 Kibe T, Osawa GA, Keegan CE, de Lange structure‐specific nucleases, phenocopies 54 Gurney AL, Carver‐Moore K, de Sauvage T. Telomere protection by TPP1 is mediated Fanconi anemia. Nature genetics. FJ, Moore MW. Thrombocytopenia in c‐mpl‐ by POT1a and POT1b.Mol Cell Bi‐ 2011;43(2):147‐52. deficient mice. Science (New York, N Y ). 1994;265(5177):1445‐7. ol. 2010 Feb;30(4):1059‐66. www.StemCells.com ©AlphaMed Press 2016 11

68 Ellis SR, Gleizes P‐E. Diamond Blackfan reprogramming to pluripotency. Nature. hematopoietic progenitors. Stem Cells. anemia: ribosomal proteins going rogue. 2011;471(7336):58‐62. 2013;31(5):1022‐9. Seminars in hematology. 2011;48(2):89‐96. 83 Gore A, Li Z, Fung H‐L, Young JE, Agarwal 95 Gu BW, Apicella M, Mills J, Fan JM, 69 Draptchinskaia N, Gustavsson P, S, Antosiewicz‐Bourget J, et al. Somatic Reeves DA, French D, et al. Impaired Andersson B, Pettersson M, Willig TN, coding mutations in human induced Telomere Maintenance and Decreased Dianzani I, et al. The gene encoding ribosomal pluripotent stem cells. Nature. Canonical WNT Signaling but Normal protein S19 is mutated in Diamond‐Blackfan 2011;471(7336):63‐7. Ribosome Biogenesis in Induced Pluripotent anaemia. Nature genetics. 1999;21(2):169‐75. 84 Wang L, Menendez P, Shojaei F, Li L, Stem Cells from X‐Linked Dyskeratosis 70 Leger‐Silvestre I, Caffrey JM, Dawaliby R, Mazurier F, Dick JE, et al. Generation of Congenita Patients. PLoS One. Alvarez‐Arias DA, Gas N, Bertolone SJ, et al. hematopoietic repopulating cells from human 2015;10(5):e0127414. Specific Role for Yeast Homologs of the embryonic stem cells independent of ectopic 96 Batista LF, Pech MF, Zhong FL, Nguyen Diamond Blackfan Anemia‐associated Rps19 HOXB4 expression. J Exp Med. HN, Xie KT, Zaug AJ, et al. Telomere Protein in Ribosome Synthesis. The Journal of 2005;201(10):1603‐14. Epub 2005/05/11. shortening and loss of self‐renewal in biological chemistry. 2005;280(46):38177‐85. 85 Vodyanik MA, Thomson JA, Slukvin, II. dyskeratosis congenita induced pluripotent 71 Idol RA, Robledo S, Du H‐Y, Crimmins DL, Leukosialin (CD43) defines hematopoietic stem cells. Nature. 2011;474(7351):399‐402. Wilson DB, Ladenson JH, et al. Cells depleted progenitors in human embryonic stem cell 97 Agarwal S, Loh YH, McLoughlin EM, for RPS19, a protein associated with Diamond differentiation cultures. Blood. Huang J, Park IH, Miller JD, et al. Telomere Blackfan Anemia, show defects in 18S 2006;108(6):2095‐105. Epub 2006/06/08. elongation in induced pluripotent stem cells ribosomal RNA synthesis and small ribosomal 86 Tian X, Woll PS, Morris JK, Linehan JL, from dyskeratosis congenita patients. Nature. subunit production. Blood cells, molecules & Kaufman DS. Hematopoietic engraftment of 2010;464(7286):292‐6. diseases. 2007;39(1):35‐43. human embryonic stem cell‐derived cells is 98 Wang F, Yin Y, Ye X, Liu K, Zhu H, Wang 72 Flygare J, Aspesi A, Bailey JC, Miyake K, regulated by recipient innate immunity. Stem L, et al. Molecular insights into the Caffrey JM, Karlsson S, et al. Human RPS19, Cells. 2006;24(5):1370‐80. Epub 2006/02/04. heterogeneity of telomere reprogramming in the gene mutated in Diamond‐Blackfan 87 Risueno RM, Sachlos E, Lee JH, Lee JB, induced pluripotent stem cells. Cell Res. anemia, encodes a ribosomal protein Hong SH, Szabo E, et al. Inability of human 2012;22(4):757‐68. required for the maturation of 40S ribosomal induced pluripotent stem cell‐hematopoietic 99 Garcon L, Ge J, Manjunath SH, Mills JA, subunits. Blood. 2007;109(3):980‐6. derivatives to downregulate microRNAs in Apicella M, Parikh S, et al. Ribosomal and 73 Choesmel V, Bacqueville D, Rouquette J, vivo reveals a block in xenograft hematopoietic defects in induced pluripotent Noaillac‐Depeyre J, Fribourg S, Cretien A, et hematopoietic regeneration. Stem stem cells derived from Diamond Blackfan al. Impaired ribosome biogenesis in Diamond‐ Cells.30(2):131‐9. Epub 2011/12/02. anemia patients. Blood. 2013;122(6):912‐21. Blackfan anemia. Blood. 2007;109(3):1275‐ 88 Narayan AD, Chase JL, Lewis RL, Tian X, 100.Ge J, Apicella M, Mills JA, Garcon L, 83. Kaufman DS, Thomson JA, et al. Human French DL, Weiss MJ, et al. Dysregulation of 74 de Oliveira JF, Sforca ML, Blumenschein embryonic stem cell‐derived hematopoietic the Transforming Growth Factor beta TMA, Goldfeder MB, Guimaraes BG, Oliveira cells are capable of engrafting primary as well Pathway in Induced Pluripotent Stem Cells CC, et al. Structure, dynamics, and RNA as secondary fetal sheep recipients. Blood. Generated from Patients with Diamond interaction analysis of the human SBDS 2006;107(5):2180‐3. Epub 2005/11/10. Blackfan Anemia. PLoS One. protein. J Mol Biol. 2010;396(4):1053‐69. 89 Ledran MH, Krassowska A, Armstrong L, 2015;10(8):e0134878. 75 Zhu J, Jankovic D, Grinberg A, Guo L, Dimmick I, Renstrom J, Lang R, et al. Efficient 101.Tulpule A, Kelley JM, Lensch MW, Paul WE. Gfi‐1 plays an important role in IL‐2‐ hematopoietic differentiation of human McPherson J, Park IH, Hartung O, et al. mediated Th2 cell expansion. Proceedings of embryonic stem cells on stromal cells derived Pluripotent stem cell models of Shwachman‐ the National Academy of Sciences of the from hematopoietic niches. Cell Stem Cell. Diamond syndrome reveal a common United States of America. 2008;3(1):85‐98. Epub 2008/07/03. mechanism for pancreatic and hematopoietic 2006;103(48):18214‐9. 90 Suzuki N, Yamazaki S, Yamaguchi T, dysfunction. Cell Stem Cell. 2013;12(6):727‐ 76 Walasek MA, van Os R, de Haan G. Okabe M, Masaki H, Takaki S, et al. 36. Hematopoietic stem cell expansion: Generation of engraftable hematopoietic 102.Hiramoto T, Ebihara Y, Mizoguchi Y, challenges and opportunities. Ann N Y Acad stem cells from induced pluripotent stem Nakamura K, Yamaguchi K, Ueno K, et al. Sci. 2012;1266:138‐50. cells by way of teratoma formation. Wnt3a stimulates maturation of impaired 77 Takahashi K, Yamanaka S. Induction of Molecular therapy : the journal of the neutrophils developed from severe congenital pluripotent stem cells from mouse embryonic American Society of Gene Therapy. neutropenia patient‐derived pluripotent stem and adult fibroblast cultures by defined 2013;21(7):1424‐31 cells. Proceedings of the National Academy of factors. Cell. 2006;126(4):663‐76. 91 Raya A, Rodriguez‐Piza I, Guenechea G, Sciences of the United States of America. 78 Yu J, Vodyanik MA, Smuga‐Otto K, Vassena R, Navarro S, Barrero MJ, et al. 2013;110(8):3023‐8. Antosiewicz‐Bourget J, Frane JL, Tian S, et al. Disease‐corrected haematopoietic 103.Nayak RC, Trump LR, Aronow BJ, Myers Induced pluripotent stem cell lines derived progenitors from Fanconi anaemia induced K, Mehta P, Kalfa T, et al. Pathogenesis of from human somatic cells. Science (New York, pluripotent stem cells. Nature. ELANE‐mutant severe neutropenia revealed N Y ). 2007;318(5858):1917‐20. 2009;460(7251):53‐9. by induced pluripotent stem cells. The Journal 79 Takahashi K, Tanabe K, Ohnuki M, Narita 92 Muller LUW, Milsom MD, Harris CE, Vyas of clinical investigation. 2015;125(8):3103‐16. M, Ichisaka T, Tomoda K, et al. Induction of R, Brumme KM, Parmar K, et al. Overcoming 104.Collin J, Lako M. Concise review: put‐ pluripotent stem cells from adult human reprogramming resistance of Fanconi anemia ting a finger on stem cell biology: zinc finger fibroblasts by defined factors. Cell. cells. Blood. 2012;119(23):5449‐57. nuclease‐driven targeted genetic editing in 2007;131(5):861‐72. 93 Tulpule A, Kelley JM, Lensch MW, human pluripotent stem cells. Stem Cells. 80 Nishizawa M, Chonabayashi K, Nomura McPherson J, Park IH, Hartung O, et al. 2011;29(7):1021‐33. M, Tanaka A, Nakamura M, Inagaki A, et al. Pluripotent stem cell models of Shwachman‐ 105.Highly efficient genome modi‐ Epigenetic Variation between Human Induced Diamond syndrome reveal a common fications mediated by CRISPR/Cas9 Pluripotent Stem Cell Lines Is an Indicator of mechanism for pancreatic and hematopoietic Differentiation Capacity. Cell Stem Cell. dysfunction. Cell Stem Cell. 2013;12(6):727‐ in Drosophila. Yu, Z., Ren, M., Wang, 81 Kim K, Zhao R, Doi A, Ng K, Unternaehrer 36. Z., Zhang, B., Rong, Y.S., Jiao, R., and J, Cahan P, et al. Donor cell type can influence 94 Yung SK, Tilgner K, Ledran MH, Gao, G. Genetics. 2013; 195: 289– the epigenome and differentiation potential Habibollah S, Neganova I, Singhapol C, et al. 291 of human induced pluripotent stem cells. Brief report: human pluripotent stem cell Nature biotechnology. 2011;29(12):1117‐9. models of fanconi anemia deficiency reveal 106.Multiplex genome engineering 82 Hussein SM, Batada NN, Vuoristo S, an important role for fanconi anemia proteins using CRISPR/Cas systems. Cong, L., Ching RW, Autio R, Narva E, et al. Copy in cellular reprogramming and survival of Ran, F.A., Cox, D., Lin, S., Barretto, number variation and selection during www.StemCells.com ©AlphaMed Press 2016 12

R., Habib, N., Hsu, P.D., Wu, X., with CRISPR/Cas9. Stem Cells Transl. Med. 5, Targeted gene therapy and cell reprogram‐ Jiang, W., Marraffini, L.A., and 8–19 (2016). ming in Fanconi anemia.EMBO Mol 112.Park CY, Kim DH, Son JS, Sung JJ, Lee J, Med. 2014 May 23;6(6):835‐48. Zhang, F. Science. 2013; 339: 819–82 Bae S, Kim JH, Kim DW, Kim JS. 2015. Func‐ 117.Woo DH, Chen Q, Yang TL, Glineburg 107.Correction of a genetic disease in tional Correction of Large Factor VIII Gene mouse via use of CRISPR‐Cas9. Wu Y, Liang D, Chromosomal Inversions in Hemophilia A MR, Hoge C, et al. Enhancing a Wnt‐Telomere Wang Y, Bai M, Tang W, Bao S, Yan Z, Li D, Li J. Patient‐Derived iPSCs Using CRISPR‐Cas9. Cell Feedback Loop Restores Intestinal Stem Cell Cell Stem Cell. 2013 Dec 5;13(6):659‐62. Stem Cell 17(2): 213‐20. Function in a Human Organotypic Model of Dyskeratosis Congenita. Cell Stem Cell. 2016 doi: 10.1016/j.stem.2013.10.016. 113.Vanuytsel K, Cai Q, Nair N, Khurana 108.Song, B. et al. Improved hematopoietic Sep 1;19(3):397‐405 S, Shetty S et al. FANCA knockout in human differentiation efficiency of gene‐corrected 118.Hirata S, Takayama N, Jono‐Ohnishi R, embryonic stem cells causes a severe growth beta‐thalassemia induced pluripotent stem Endo H, Nakamura S, Dohda T, et al. disadvantage. Stem Cell Res. 2014 cells by CRISPR/Cas9 system. Stem Cells Dev. Congenital amegakaryocytic Sep;13(2):240‐50. 24, 1053–1065 (2015). thrombocytopenia iPS cells exhibit defective 114.Osborn MJ, Gabriel R, Webber BR, 109.Xie, F. et al. Seamless gene correction MPL‐mediated signaling. The Journal of DeFeo AP, McElroy AN, et al. Fanconi anemia of β‐thalassemia mutations in patient‐specific clinical investigation. 2013;123(9):3802‐14. gene editing by the CRISPR/Cas9 system. Hum iPSCs using CRISPR/Cas9 and piggyBac. Ge‐ 119.Kotini AG, Chang C‐J, Boussaad I, Gene Ther. 2015 Feb;26(2):114‐26. nome Res. 24, 1526–1533 (2014). Delrow JJ, Dolezal EK, Nagulapally AB, et al. 110.Xu, P. et al. Both TALENs and 115.Bluteau D, Masliah‐Planchon Functional analysis of a chromosomal CRISPR/Cas9 directly target the HBB IVS2–654 J, Clairmont C, Rousseau A, Ceccaldi R et al. deletion associated with myelodysplastic (C>T) mutation in β‐thalassemia‐derived Biallelic inactivation of REV7 is associated syndromes using isogenic human induced iPSCs. Sci. Rep. 5, 12065 (2015). with Fanconi anemia. J Clin Invest. 2016 Sep pluripotent stem cells. Nature biotechnology. 111.Yang, Y. et al. Naïve induced pluripo‐ 1;126(9):3580‐4 2015;33(6):646‐55. tent stem cells generated from β‐thalassemia 116.Rio P, Baños R, Lombardo A, Quintana‐ fibroblasts allow efficient gene correction Bustamante O, Alvarez L, Garate Z, et al.

www.StemCells.com ©AlphaMed Press 2016 13

Figure 1: A schematic summary of advantages and disadvantages of animal and hiPSC based disease modelling ap‐ proaches.

www.StemCells.com ©AlphaMed Press 2016 14

Table 1: A summary of clinical features of IBMF together with causative genes and current therapies. IST= immunosuppressive therapies, HSCT = hematopoietic stem cell transplantation, TPO = thrombopoietin, G-CSF = granulocyte colony stimulating factor, MDS = myelodysplastic syndrome, AML = acute myeloid leukaemia, AA = aplastic anemia, Epo= erythropoietin.

Bone mar- Hematopoiet- Other clinical Genes Biologi- Disease Ref- row failure ic abnormali- features identified cal fea- manage- eren syndrome ties tures ment and en- treatment ces FA: genomic pancytopenia congenital ab- FANCA, hyper- hematolog- (6- instability and bone mar- normalities, FANCB, sensitivi- ic monitor- 12) disorder row failure growth retarda- FANCC, ty to ing and sol- caused by tion, bone mar- FANCD1, DNA id tumor alterations in row failure, in- FANCD2, cross- surveil- genes increased risk of FANCE, linking lance, an- volved in haematological FANCF, agents/in drogen replication- malignancies FANCG, tolerance therapy, dependant- and other solid FANCI, to oxida- antioxi- repair and tumours (mostly FANCJ, tive dants, G- removal of head and neck), FANCL, stress CSF com- DNA cross- radiosensitivity FANCM, and fre- bined with links and premature FANCN, quent Epo and/or ageing FACNP, chromo- androgens, FANCO, somal allogeneic FANCS aberra- HSCT tions pointing to a DNA damage response defect DKC: multi- pancytopenia solid tumours DKC1, acceler- hematolog- (13- system disor- and bone mar- (head, neck and TERT, TR, ated te- ic monitor- 14) der caused by row failure colorectal), ab- TINF2, lomere ing and sol- defective te- normal skin RTEL1, shorten- id tumor lomere pigmentation, NOP10, ing in all surveil- maintenance nail dystrophy, NHP2, leuko- lance, an- and/or ribo- mucosal leuco- WRAP53, cyte drogen some func- plakia, perio- C16orf57 subsets therapy, G- tion dontal disease, and CTC1 resulting CSF com- premature gray- in cell bined with ing, osteoporo- loss or Epo, al- sis mental re- dysfunc- logeneic tardation, and tion and HSCT pulmonary dis- genomic ease instability DBA: selec- red blood cell craniofacial, RPS24, defective steroid (15) tive reduction aplasia: mac- skeletal, cardiac RPS17, ribo- therapy, red in erythroid rocytic ane- and/or genitou- RPS7, some cell transfu- precursors mia, reticulo- rinary abnor- RPS10, synthesis sion, iron www.StemCells.com ©AlphaMed Press 2016 15 with macro- cytopenia, and malities RPS19, chelation cytic anemia nearly absent RPS26, allogeneic erythroid pro- RPS27, HSCT genitors in the RPS29, bone marrow RPL5, RPL11, RPL26, RPL15, RPL27, RPL35 SDS: exo- bone marrow exocrine pan- SBDS hemato- transfu- (16) crine pancre- failure, neu- creatic insuffi- poietic sions, pan- atic insuffi- tropenia, ciency, short progeni- creatic en- ciency and anemia, pan- stature, met- tors have zymes, an- hematologic cytopenia, aphyseal dysos- faulty tibiotics, G- abnormalities MDS and leu- tosis, rib and prolifer- CSF, HSCT kaemia thoracic cage ative abnormalities properties and increased apoptosis linked to hyper activation of the Fas signalling pathway CAMT: iso- thrombocyto- no specific so- MPL isolated Platelet (17- lated throm- penia with matic abnormal- throm- transfusion, 18) bocytopenia reduced or ities bocyto- and anti- and absent penia fibrinolytic megakaryo- megakaryo- and ab- agents for cytopenia cytes in the sence of bleeding with no phys- marrow pro- megakar red cell ical anoma- gression to yocytes transfusion lies pancytopenia in the for anemia, and marrow bone allogeneic hypoplasia marrow HSCT can occur caused by a defective response to TPO SCN: very severe neu- ELA2, matura- G-CSF, al- (19- low neutro- tropenia , my- HAX1, tion ar- logeneic 20) phil count eloid series AK2, rest of HSCT often less maturation GFI1, granulo- than arrest, WASP, poiesis www.StemCells.com ©AlphaMed Press 2016 16

0.5×109/L progression to CSF3R, at the MDS and G6PC3, level of AML may GATA1, promye- occur in some JAGN1, locytes patients VPS45 with peripheral blood absolute neutrophil counts below 0.5 x 109/l and early onset of severe bacterial infections TAR: new- reduction in characteristic RBM8A abnor- platelet (21) borns present the number of bilateral absent mal dif- transfusions with throm- platelets, re- radii (unilateral ferentia- bocytopenia duction in in ~ 2%), with tion number of abnormal but mecha- megakaryo- present thumbs, nism of cytes in bone facial dys- megakar marrow fre- morphism, car- yocyte quent bleed- diac defects, and ing episodes and genitouri- platelet in the first nary malfor- produc- year of life mations tion that diminish in frequency and severity with age RS: thrombocyto- limited prona- HOXA11 Certain supportive (22) amegakaryo- penia or AA tion/supination cyto- transfusions cytic throm- of the arms due kines and alloge- bocytopenia to proximal ra- stimulate neic HSCT dioulnar the mat- synostosis uration of megakar yocytic progenitor cells, other signals as PF4, CXCL5, CXCL7 www.StemCells.com ©AlphaMed Press 2016 17

& CCL5 inhibit platelet formation Pearson Syn- transfusion exocrine pan- contiguous mito- supportive (23) drome dependant creas, liver , and gene dele- chondri- transfusions macrocytic renal tubular tion/duplic al DNA with blood anemia, vari- defects ation syn- abnor- and plate- able neutro- drome in- malities lets as penia and volving needed, thrombocyto- several treatment penia vacu- mtDNA with Epo oles in mar- genes and G-CSF row precur- for severe sors, ringed neutropenia sideroblasts, anemia

Table 2. Summary of key murine models of inherited BMFS.

Disease Affected systems References phenotype/ Gene name FA/Fanca Homozygotes displayed FA-like phenotypes including 28,29 growth retardation, microphthalmia, craniofacial malformations and hypogonadism. Homozygous females demonstrate premature reproductive senescence and an increased incidence of ovarian cysts. Homozygous males exhibit an elevated frequency of mis-paired meiotic chromosomes and increased apoptosis in germ cells, implicating a role for Fan- ca in meiotic recombination. Fancc-/- Fanca-/- display the same phenotype as the single mutants suggesting that these two genes are epistatic. FA/Fancc Homozygotes do not show developmental abnormalities or 30-31 haematological defects till 9 -12 months of age. Male and female mutant mice have reduced numbers of germ cells and females have markedly impaired fertility. The CFC capacity of hematopoietic progenitors is abnormal and the cells are hypersensitive to gamma-interferon. Fancc-/-Tert-/- double mutant mice have exacerbate telomere attrition when murine bone marrow cells experience high cell turnover after serial transplantation and increase in the incidence of telomere sis- ter chromatid exchange. Fancc-/-Fancg-/- double-mutant mice develop spontaneous hematologic sequelae, including bone marrow failure, acute myeloid leukaemia, myelodysplasia and complex random chromosomal abnormalities. FA/Fancd1 Homozygous null mutants are embryonic lethal with abnor- 32 malities including growth retardation, neural tube defects, and mesoderm abnormalities; conditional mutations cause genetic instability and enhanced tumor formation; mutants with truncated BRCA2 protein survive, are small, infertile, show improper tissue differentiation and develop lymphomas and carcinomas FA/Fancd2 Homozygous mutant mice exhibit meiotic defects and germ 33-35 cell loss. In addition, mutant mice display perinatal lethality, susceptibility to epithelial cancer and microphthalmia. Ho- mozygous mice have smaller hematopoietic stem cell pool and reduced lymphoid progenitor frequency. Fancd2-/-Aldh2- /- double homozygous mice are unusually sensitive to ethanol exposure in utero, and ethanol consumption by postnatal double-deficient mice rapidly precipitates bone marrow failure and spontaneously developed acute leukaemia. Aged Aldh2-/-Fancd2-/- mutant mice which do not develop leukaemia, spontaneously develop aplastic anemia, with concomi- tant accumulation of damaged DNA within the hematopoietic stem and progenitor cell pool. FA/Fancg Females and males homozygous for targeted null mutations 36 exhibit hypogonadism and reduced fertility. Cytogenetic analysis shows somatic chromosome aberrations occurrence www.StemCells.com ©AlphaMed Press 2016 19

at a higher spontaneous rate. Cells are also more sensitive to mitomycin C. FA/Fanci These mice show craniofacial, vision and eye abnormalities. 37 FA/Fancn Homozygotes display embryonic lethality with impaired in- 38 ner cell mass proliferation, impaired gastrulation, absence of the amnion, somites and tail bud and general improper or- ganogenesis. FA/Fancm Homozygotes exhibit reduced female transmission, hy- 39 pogonadism, premature death and increased incidence of tumours. FA/Fancp Homozygotes display exhibit preweaning lethality, reduced 40 fertility, abnormal eye morphology, abnormal skeletal morphology, hydrocephalus, chromosomal instability, early cellular replicative senescence and abnormal lymphopoeisis. Mutant mice are characterised by blood cytopenia, premature senescence, accumulation of damaged chromosomes and hypersensitivity to DNA cross linking agents. FA/Fanco Mice homozygous for a null mutation display embryonic le- 41 thality. Mice carrying a null and a hypomorphic allele have partial penetrance of male and female infertility due to defects in meiosis. FA/Fancs Homozygous null mutants are embryonic lethal with abnor- 42 malities including growth retardation, neural tube defects, and mesoderm abnormalities; conditional mutations cause genetic instability and enhanced tumor formation; mutants with truncated BRCA1 protein survive, have a kinky tail, pigmentation anomalies, male infertility and increased tumor incidence. DKC/Dkc1 Early generation male mice hemizygous for a hypomorphic 43 allele exhibit bone marrow failure, dyskeratosis, extramedul- lary hematopoieis, splenomegaly, lung and kidney abnormalities, increased tumor incidence, altered ribosome function. Decreased telomere length is noted only in later generations. DKC/Tert In spite of impaired telomerase function, homozygous mu- 44 tant mice are overtly normal in early generations. Impaired fertility has been reported in later generations for homozygotes of at least one knockout allele. Homozygous Tert mice display short dysfunctional telomeres and sustained increased DNA damage signaling and classical degenerative phenotypes upon successive generational mattings and ad- vancing age. DKC/Tr Early generation mice homozygous for a null allele have in- 45 tact telomeres and appear grossly unaffected and healthy, whereas late generation mutants exhibit premature death, shortened and dysfunctional telomeres, apoptotic and proliferative defects, infertility, and multi-organ degenerative de- cline. Late-generation animals exhibit defective spermato- genesis, with increased programmed cell death (apoptosis) and decreased proliferation in the testis. Proliferative capacity of hematopoietic cells in the bone marrow and spleen is also compromised. These progressively adverse effects coin- www.StemCells.com ©AlphaMed Press 2016 20

cide with substantial erosion of telomeres and fusion and loss of chromosomes. DKC/Tinf2 Targeted disruption of this gene results in embryonic lethali- 46 ty prior to E7.5 through a mechanism that is independent of telomerase function. Second and third generation heterozygotes develop mild pancytopenia, consistent with hematopoietic dysfunction in DKC, as well as diminished fecundity. DKC/Rtel1 Homozygous null mice display embryonic lethality with ab- 47 normal development of the neural tube, brain, heart, vascula- ture, placenta, and allantois and chromosomal abnormalities in differentiating cells. DKC/Ctc1 Mice homozygous for a targeted allele exhibit defective te- 48 lomere replication that leads to stem cell exhaustion, bone marrow failure and premature death. DBA/Rps6 Conditional Rps6 mice using CD4-Cre abolishes T cell de- 49 velopment. DBA/Rps7 Rps7 disruption results in decreased body size, abnormal 50 skeletal morphology, mid-ventral white spotting, and eye malformations. Rps7 mutants display overt malformations of the developing central nervous system and deficits in work- ing memory; however they do not show anemia or hyper- pigmentation. DBA/Rps1 Homozygous null embryos die prior to the formation of a 51 9 blastocyst. Mice heterozygous for some point mutations show pigment defects affecting the feet and tail. However the heterozygotes show a normal development of the hematopoietic system. Heterozygous missense mutations of Rps19 show a mild macrocytic anaemia reflecting the fact that mutations causes a hypomorphic allele rather than true happloinsufficiency. DBA/Rps2 Heterozygous missense mutations of Rps20 show a mild 52 0 macrocytic anaemia reflecting the fact that mutations causes a hypomorphic allele rather than true happloinsufficiency. SDS/Sbds Loss of Sbds gene results in early embryonic lethality, with 53 homozygotes showing histological abnormalities of the liver and accumulation of free cytoplasmic 40 S and 60 S subunits. Heterozygotes have a normal phenotype. CAMT/Mpl Mice homozygous for targeted mutations at this locus are 54 unable to produce normal numbers of megakaryocytes and platelets and display HSC deficiencies that are not limited to megakaryocytic lineages. These mice also have increased concentrations of circulating TPO. SCN/Ela2 Homozygotes for a null allele show impaired neutrophil 55 physiology, susceptibility to Gram (-) bacterial infection, reduced sensitivity to xenobiotics and abnormal local Schwartzman responses. Homozygotes for a knock-in allele show susceptibility to fungal infection and resistance to en- dotoxic shock. Heterozygous mice do not show neutropenia. www.StemCells.com ©AlphaMed Press 2016 21

SCN/HAX1 Mice homozygous for deletion of this gene fail to survive 56 beyond 14 weeks of age. Apoptosis of neurons in the stria- tum and cerebellum occurs as does loss of lymphocytes and neutrophils. SCN/Gfi1 Homozygotes are severely neutropenic and accumulate im- 57 mature monocytes in blood and bone marrow. Their myeloid precursors cannot differentiate into granulocytes upon stimu- lation with G-CSF; however they can develop into macro- phages. Conditional knockouts indicate defects in Th2 cell expansions and enhanced IFN production. SCN/Wasp Homozygous mutant females and hemizygous mutant males 58 exhibit reduced numbers of peripheral blood lymphocytes and platelets, but increased numbers of neutrophils. RS/Hoxa11 Homozygotes for targeted null mutations exhibit homeotic 59 transformations affecting thoracic and sacral vertebrae, and forelimb defects. Mutants are sterile due to malformed vas deferens and cryptorchism in males, and defective uteri in females.