Date Published: May 23, 2012
Publisher: Hindawi Publishing Corporation
Author(s): Anja Geiselhart, Amelie Lier, Dagmar Walter, Michael D. Milsom.
Fanconi anemia (FA) is the most common inherited bone marrow failure syndrome. FA patients suffer to varying degrees from a heterogeneous range of developmental defects and, in addition, have an increased likelihood of developing cancer. Almost all FA patients develop a severe, progressive bone marrow failure syndrome, which impacts upon the production of all hematopoietic lineages and, hence, is thought to be driven by a defect at the level of the hematopoietic stem cell (HSC). This hypothesis would also correlate with the very high incidence of MDS and AML that is observed in FA patients. In this paper, we discuss the evidence that supports the role of dysfunctional HSC biology in driving the etiology of the disease. Furthermore, we consider the different model systems currently available to study the biology of cells defective in the FA signaling pathway and how they are informative in terms of identifying the physiologic mediators of HSC depletion and dissecting their putative mechanism of action. Finally, we ask whether the insights gained using such disease models can be translated into potential novel therapeutic strategies for the treatment of the hematologic disorders in FA patients.
Fanconi anemia (FA) is a rare, autosomal recessive and X-linked hereditary disorder, which is characterized by progressive bone marrow failure (BMF), congenital developmental defects, and an early onset of cancers such as leukemia and some solid tumors . In general, the hematologic manifestations of FA remain the primary cause of morbidity and mortality, with patients suffering from a markedly increased risk of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). In addition, FA patients are also predisposed towards various forms of solid tumor such as squamous cell carcinoma of the head and neck, esophagus, and gynecologic area [2, 3].
Hematologic abnormalities, which are found in virtually all FA patients, include cytopenias such as thrombocytopenia (abnormally low platelet counts in the peripheral blood), neutropenia (low neutrophil counts), and progressive pancytopenia (abnormality in two or three blood cell lineages) . At birth, FA patients usually do not show any signs of these defects and have normal blood cell counts, but, as the patient grows older, the hematologic complications start to develop, mainly within the first decade of life. Macrocytosis (enlargement of red blood cells) is usually the first to be detected, followed by thrombocytopenia and aplastic anemia (insufficient production of red blood cells, leukocytes, and platelets in the BM), finally resulting in the characteristic progressive BMF phenotype [3, 32, 33]. Unless treated, BMF represents the primary cause of morbidity in FA patients.
Since all hematopoietic lineages are compromised in FA patients, it would seem reasonable to assume that a defective FA signaling pathway may negatively impact upon the biology of HSCs, which comprise the top of the hematopoietic system hierarchy. However, while it is relatively straightforward to assess the depletion of mature hematopoietic cells and myeloid progenitors in FA patients, it is more difficult to directly examine HSC function. Nonetheless, there are several lines of evidence in FA patients that suggest that the HSC pool is compromised.
Clinical observations from FA patients provide some evidence that allows us to implicate a defect at the level of HSCs in driving the BMF disease phenotype. Nonetheless, experimental model systems must be employed to directly interrogate the function of HSCs defective in the FA signaling pathway in a reproducible manner. The “gold standard” for assessing the capacity of HSCs to be able to differentiate to form all mature lineages of the hematopoietic system, while also being capable of self-renewal, is to perform BM transplantation and measure long-term multilineage engraftment within the recipient. Ideally, this assay would be performed with a limiting dilution of HSCs and involve at least one serial transplantation into a secondary recipient. While this rigorous assessment clearly cannot be performed in patients, a number of surrogate assays have been developed in order to dissect human HSC biology.
Studying the etiology of human disease ideally involves the use of human model systems. However, given the previously discussed constraints that are associated with studying HSC biology in FA, namely, the lack of an abundant source of patient HSCs to act as a starting material and the absence of a transplantation system to assess HSC function, it has been necessary to develop animal model systems for this disease. Fortunately, the FA signaling pathway has been well conserved throughout evolution; thus, there are several potential model systems available (Table 1).
There are a number of vertebrate model systems that have been used to interrogate HSC biology. In the zebrafish, transplantation of whole kidney marrow cells into lethally irradiated recipient fish was shown to be radioprotective, specifically rescuing the ablation of the hematopoietic system that is observed in nontransplanted fish . This demonstrated that transplantable zebrafish HSCs were to be found in the adult kidney. The recently identified existence of histocompatibility genes in the zebrafish has allowed the further improvement of this transplant system . Although zebrafish contain the full complement of FA family members found in humans, loss of function models have only been described for a few complementation groups . The knockdown of the zebrafish ortholog of FANCD2 using an antisense morpholino approach leads to similar developmental defects as those observed in some FA patients, including decreased body size, microcephaly, and microphthalmia . This suggests that the FA pathway plays a similar role in zebrafish and humans. While the morpholino approach is particularly useful for the study of a gene product during development in the zebrafish, it is not appropriate for the ongoing evaluation of gene function in the adult organism. To date, zebrafish mutant lines for FANCL and FANCD1 have been described . Although these fish have an interesting defect in sex determination, they have no documented defects in hematopoiesis. Nonetheless, it is possible that a more detailed assessment of the HSC function of these mutant fish using the transplantation assay described above may yield a phenotype.
The observation that none of the currently available models of FA fully phenocopies the progressive BMF observed in patients may relate to the lack of an environmental and/or endogenous factor that drives HSC loss. Since FA cells are invariably hypersensitive to DNA crosslinking agents, resulting in cell cycle arrest and apoptosis, it is not unreasonable to focus upon the identification of potential crosslinking agents in FA patients in the search for a physiologic mediator of HSC depletion (reviewed in ).
Since DNA synthesis can be considered a form of DNA damage, replicative stress may also be a candidate for HSC depletion in FA. Indeed, in FA competent cell lines, upon induction of replicative stress, FANCD2, FANCM, and the Blooms complex are localized to discrete fragile sites on sister chromatids during mitosis [118, 119]. These fragile sites comprise common chromosomal break points and are also the location at which stress-induced ultrafine DNA bridges form. In human and murine FA-deficient cells, including FA HSC/progenitors, there is an increased number of ultrafine DNA bridges compared to their wild-type counterparts . This correlates with an increased frequency of cytokinesis failure, as assessed by the number of binucleated HSC/progenitors, and an increased rate of apoptosis. Thus, it is hypothesized that the FA pathway is involved in the resolution of these spontaneously occurring ultrafine bridges and that the absence of a functional FA pathway leads to cytokinesis failure followed by programmed cell death, or to genetic instability. Such a mechanism would be an attractive explanation for the progressive BMF seen in FA patients, as HSCs would potentially be depleted as they were induced into cycle.
In addition to an inability to resolve some forms of DNA damage, FA cells are also hypersensitive to the inhibitory action of certain proinflammatory cytokines. Proinflammatory cytokines are potential physiologic mediators of BMF in FA, since HSCs are routinely exposed to a range of proinflammatory cytokines during either infection or as part of the etiology of diseases with an inflammatory component, such as rheumatoid arthritis.
Hematopoietic stem cell transplantation (HSCT) derived from the BM, mobilized peripheral blood, or umbilical cord blood of a human-leukocyte-antigen-(HLA-) matched donor currently remains the only curative treatment option for the hematologic abnormalities in FA . In fact, FA was the first disease that was successfully treated by transplantation using cord blood from an unaffected HLA-identical sibling as a starting material . Significant barriers to successful transplantation of FA patients include the challenge of creating a satisfactory preparative regimen in light of the patient’s acute sensitivity to chemotherapeutics and radiotherapy [39, 141, 142]; the availability of an HLA-matched donor who does not suffer from the disease. Most patients are dependent on alternative donor grafts from an HLA-unmatched donor as only a very small number of patients have an unaffected, matched sibling donor (less than 25%). However, advances in the conditions used in preparative regimens allow the achievement of almost similar transplant outcomes for both cases, with survival rates of 52–88% for mismatched family members or matched unrelated donors and 69–93% for HLA-identical sibling donor HSCTs [39, 143, 144]. While FA patients are waiting for a suitable HSC donor, supportive care can be provided via red blood cell and platelet transfusions; oral administration of androgens such as oxymetholone, methyltestosterone, the androgen analogue danazol; or the direct injection of growth factors such as granulocyte-colony forming factor (G-CSF) [145, 146]. However, androgen application can also lead to adverse sideeffects like masculinization of female patients, acne, hyperactivity and diverse problems associated with the liver such as deranged liver enzymes, hepatic adenomas and the potential risk of hepatic adenocarcinoma . While hematopoietic growth factors such as G-CSF and GM-CSF are capable of enhancing peripheral blood neutrophil counts, and, in some cases platelets [148, 149], they should be avoided in patients with clonal cytogenetic abnormalities because of the risk of inducing leukemia. In any case, they are only effective for the short-term treatment and HSCT is ultimately the definite therapy.
Since the lack of availability of HLA-matched disease-free donor HSCs is a major limitation in HSCT, one attractive novel therapeutic modality is the genetic correction of autologous patient HSCs via the reintroduction of the defective FA cDNA using a delivery system such as a retroviral vector. Recent major advances have been made in the field of gene therapy, which has allowed the correction of a range of different inherited genetic disorders with a hematologic basis via the retroviral-mediated delivery of correcting cDNAs into patient HSCs [150, 151]. Since the input cells are patient derived, there should be no issues with immunologic rejection of the graft unless the vector system or transgene payload is immunogenic. The phenomenon of reverse mosaicism, that we have previously discussed, would appear to indicate that FA would be an ideal candidate disease for treatment via gene therapy, since correction of an individual HSC can result in sustained reversal of BMF . Indeed, this finding has been recapitulated in murine models of FA . Unfortunately, FA also presents some unique problems, which means that it may be an incredibly difficult disease to treat with gene therapy using existing technologies. These include the extremely low yield of CD34+ cells that can be collected for gene modification relative to those routinely achieved in non-FA patients; the extreme sensitivity of FA cells to ex vivo culture [37, 92]. To date, the clinical gene therapy trials for FA have all failed to achieve robust engraftment of corrected patient HSCs, although advances have been made in the ability of clinicians to transduce FA CD34+ cells with retroviral vectors [37, 152, 153]. Fortunately, some of the model systems that have been developed for FA have been able to assist in the formulation of new strategies that may help overcome the barriers to effective gene therapy of FA.
In addition to gene therapy, data obtained from experimental models of FA have been used to devise other alternative novel therapeutic modalities for FA (summarized in Figure 2). Metabolism is predicted to generate reactive by-products, such as ROS and aldehydes, which are believed to be one potential physiological source of DNA damage that precipitates the FA phenotype. While healthy cells protect against these threats through the combined action of enzymatic detoxification and DNA repair, FA cells clearly show a defect in this respect.
FA is a fatal inherited disorder, which almost universally results in severe defects of the hematopoietic system, likely as a direct consequence of defective HSC biology. Advances in our ability to model the HSC defect in FA patients have not only enhanced our understanding of the underlying etiology of this disease but have also highlighted novel targets for therapeutic intervention. One challenge for the immediate future is to determine whether the defects that have so far been identified in FA HSCs can be extrapolated to explain the abnormal biology of other tissues that are commonly impacted upon by a defect in the FA signaling pathway. This is of particular importance since there are very limited treatment options for the serious nonhematologic complications observed in FA patients such as the increased predisposition towards solid tumors.