View a list of NIH investigators and their contact information.
Volunteer for NIAID-funded clinical studies related to ALPS on ClinicalTrials.gov.
Autoimmune Lymphoproliferative Syndrome: a Disorder of Apoptosis
Christine E. Jackson, Ph.D. and Jennifer M. Puck, M.D.
Genetics and Molecular Biology Branch
National Human Genome Research Institute
National Institutes of Health
On this page:
Autoimmune Lymphoproliferative Syndrome (ALPS) is a recently recognized disease in which a genetic defect in programmed cell death, or apoptosis, leads to breakdown of lymphocyte homeostasis and normal immunologic tolerance. Some authors have referred to ALPS as Canale-Smith syndrome or lymphoproliferative syndrome with autoimmunity. Patients with ALPS have chronic enlargement of the spleen and lymph nodes, various manifestations of autoimmunity, and elevation of a normally rare population of "double negative T-cells" (DNTs), T lymphocytes bearing ab T-cell receptors and expressing neither CD4 nor CD8 surface antigens. When lymphocytes from ALPS patients are cultured in vitro, they are resistant to apoptosis, or programmed cell death, as compared to cells from healthy controls. Most ALPS patients have mutations in a gene now named tumor necrosis factor receptor gene superfamily member (TNFRSF6). This gene, previously known as APT1, encodes the cell surface receptor for the major apoptosis pathway in mature lymphocytes; this receptor has also had many names, including Fas (to be used here), CD95 and APO-1. ALPS is subdivided into type Ia, ALPS with mutant Fas; type Ib, systemic lupus erythematosus with lymphadenopathy and mutations in the ligand for Fas; type II, ALPS with mutant caspase 10 (see below and Wang et al.); and type III, ALPS as yet without any defined genetic cause.
Since the original description of ALPS in 1995, many affected kindreds have been reported, permitting better definition of clinical spectrum and inheritance of ALPS. Most ALPS patients have heterozygous, dominant mutations of one of their two copies of TNFRSF6 on chromosome 10q24.1. Rare individuals with both gene copies mutated, or homozygous TNFRSF6 mutations, may display a more severe ALPS phenotype. Analysis of kindreds in which multiple individuals carry a mutation has shown a strong correlation between the type of mutation (genotype) and the risk of mutation-bearing relatives also having ALPS features (penetrance and phenotype). There is also mounting evidence that at least some ALPS genotypes are associated with an increased risk for developing lymphoma. Functional studies of mutant Fas proteins and identification of ALPS-related mutations in another apoptotic pathway protein, caspase 10, have provided insights into normal mechanisms of apoptosis and the roles of apoptotic pathways in immune homeostasis and immune disease.
ALPS is a recently characterized syndrome. Patients with chronic, non-malignant lymphadenopathy, splenomegaly, and hemolytic anemia or thrombocytobenia have been known for decades; several were described by Canale and Smith in 1967. In 1992, Sneller et al. found that two such patients had elevated DNTs, suggesting similarity to the lpr and gld mouse models of autoimmunity. These mice, often referred to as lupus mice, have lymphoproliferation, elevated DNTs, and autoimmune nephritis (Theofilopoulos et al, Cohen et al). The lpr and gld mouse strains have been shown to harbor homozygous mutations of Fas and Fas ligand respectively, and these mice were critical to the discovery of lymphocyte apoptosis, or programmed cell death (Watanabe-Fukunaga et al, Takahashi et al). Reports by Fisher et al. and Rieux-Laucat et al. in 1995 described apoptotic defects and Fas mutations in ALPS patients.
The Fas apoptotic pathway is important for getting rid of excess T cells after they have been activated and also eliminating antigen-driven and autoreactive T-cell clones (Brunner et al., Dhein et al., Ju et al.). Fas is a functional trimer residing at the cell membrane that, when engaged by trimeric Fas ligand (FasL), initiates a proteolytic cascade leading to chromosomal DNA degradation and cell death. The Fas protein consists of an extracellular domain with cysteine-rich regions that bind Fas ligand, a transmembrane domain, and an intracellular domain. The intracellular domain includes a conserved module known as the death domain, which interacts with Fas-associated death domain protein (FADD) to transduce the death signal. Heterozygous mutations of Fas have been demonstrated to be dominant negative (Fisher et al., Jackson et al.). The mechanism for dominant interference of a mutant Fas death domain can be understood by considering the trimer structure of the functional Fas receptor. Assuming equal synthesis of mutant and wild type Fas protein chains, random assembly of Fas trimers would lead to only one out of eight trimers composed entirely of normal Fas alone.
The clinical and laboratory findings for 62 unrelated index cases of ALPS are shown in Clinical and Laboratory Findings. All patients have enlarged spleen, lymph nodes, or both, and almost all have overt autoimmune disease, most frequently autoimmune blood cytopenias and rashes are very common. Other more rarely observed autoimmune diseases include glomerulonephritis, Guillain-Barré syndrome, autoimmune hepatitis, uveitis and vasculitis. Hydrops fetalis occurred in two severely affected infants with homozygous Fas mutations.
Laboratory diagnosis of ALPS, as summarized in Table 1, depends on demonstration of elevated numbers of DNTs; this may most readily be suspected when there is a gap between the sum of percentages of CD4 plus CD8 bearing cells as compared to the total T-cell percentage as measured by CD3. In addition elevations of B and T-cell numbers, high concentrations of one or more immunoglobulin isotypes, and significant titers of autoantibodies support the diagnosis. Serum IL-10 levels are commonly elevated. Definitive diagnosis requires demonstration in vitro of defective lymphocyte apoptosis and or deleterious mutation in an apoptosis pathway gene, most commonly Fas.
The average age at presentation in ALPS probands is 1.8 years, but abnormalities have been noted at birth and may first arise at any age. Autoimmunity may be diagnosed only after years of lymphoid cell accumulation. For example, one patient described by Infante et al. first developed autoimmune hemolytic anemia at age 54. There can be a wide spectrum of disease signs and symptoms even within a single family where several individuals share the same Fas defect. One report described 11 subjects with the same Fas mutation, a four-generation kindred followed for over 25 years (Infante et al). All had features of ALPS, but expression was highly variable. Some subjects had only a subclinical in vitro apoptosis defect and slight elevation in DNTs. Others had mild to massive lymphadenopathy and splenomegaly. Overt autoimmune disease was absent in some, but others had Coombs positive hemolytic anemia and autoimmune thrombocytopenia. One individual developed lymphoma (see below). Accumulation of lymphoid cells tends to resolve as affected children reach adolescence and early adulthood (Infante, Rieux-Laucat et al., Sneller).
Peters et al. reported a 5-year-old girl with ALPS and a T-cell lymphoma, whose uncle also had Hodgkin’s disease. Both of these individuals had a heterozygous death domain Fas mutation. Within the NIH ALPS cohort, six instances of B-cell lymphoma have been documented in five ALPS families, all in persons with death domain Fas mutations. Histologically the lymphomas were diverse: one classical Hodgkin’s disease (HD), one nodular lymphocyte predominant HD, one T cell-rich large B-cell lymphoma, two Burkitt lymphomas, and one follicular lymphoma. Combined data from centers specializing in ALPS indicates that a Fas mutation confers a 15-fold increased risk of lymphoma over that in the general population (Straus et al).
There have also been reports of somatic Fas mutations associated with various lymphoid and non-lymphoid malignancies. These include non-Hodgkin’s lymphoma, T-cell leukemias, multiple myeloma, melanoma, and non-small cell lung cancer (Grønbæk et al, Beltinger et al and Maeda et al, Landowski et al, and Shin et al, Lee et al). Of note, one report showed Fas mutations in 28 percent (12 out of 43) of transitional cell carcinomas of the urinary bladder. 8 of the 10 mutations that occurred in the death domain were identical, suggesting a mutational hot spot in bladder cancers at Fas cDNA nucleotide 993.
The associations of both germline Fas mutations with lymphomas and somatic Fas mutations with a variety of malignancies suggest an important role for Fas-mediated apoptosis as a physiologic surveillance mechanism for eliminating transformed cells.
Lim et al. have described the histopathological and immunophenotypic findings distinctive in ALPS. Most notable is the striking paracortical expansion of T cells in lymph nodes. Many of the T cells in the paracortical areas are DNTs. The extent of expansion may suggest a diagnosis of immunoblastic lymphoma, with many cells expressing the Ki-67 antigen indicative of active proliferation (Straus et al, Puck et al). Other features in ALPS lymph nodes include follicular hyperplasia, prominent vascularity of interfollicular areas and florid plasmocytosis. No histological differences have differentiated ALPS patients with or without Fas mutations.
Avila et al. have reported computed tomographic and ultrasonographic studies on 18 individuals with ALPS. Included are findings from lymph nodes, liver, spleen, thymus, and thyroid. The authors suggest that consistency of the clinical findings over time and the high echo density relative to liver of the ultrasonographic pattern of abdominal lymph nodes may assist in the diagnosis of ALPS.
In a review of 26 ALPS patients from the NIH cohort, Sneller et al. remarked that splenectomy has often been required in ALPS patients due to severe hypersplenism. Although autoimmune disease subsequently improved in some, post-splenectomy sepsis ensued in many patients despite prescription of prophylactic antibiotics. Episodes of autoimmune hemolytic anemia and thrombocytopenia are generally responsive to short courses of high dose glucocorticosteroids. Monthly administration of dexamethasone was effective in interruption of recurrent thrombocytopenia in several patients. In two patients with neutropenia and recurrent infections, recombinant granulocyte colony-stimulating factor produced sustained increases in neutrophil counts. A transient decrease in lymphadenopathy has been seen after treatment with prednisone or intravenous immunoglobulin in many patients. However, lymph nodes consistently became enlarged after cessation of treatment. It is recommended not to institute treatment aimed at decreasing the size of lymph nodes unless they are causing anatomic obstruction.
Two ALPS patients with severe phenotype due to homozygous or compound heterozygous mutations of Fas have required bone marrow transplantation. Both patients are reported to be doing well more than two years post transplant.
TNFRSF6 mutations causing ALPS in single patients or entire kindreds with ALPS have been reported from many investigators worldwide. Of these, 69 percent affect the intracellular domain of Fas. Most are changes, additions, or deletions of one or a few nucleotides in the coding exons or splice sites of TNFRSF6. Because almost all mutations are heterozygous, models for their function are based on co-expression of mutated and wild-type Fas within a cell. Martin et al. and Vaishnaw et al. have helped to clarify the mechanism whereby defective Fas death domain proteins interfere with apoptosis via normal Fas. All 15 of the mutant death domains they introduced into laboratory constructs of apoptosis-mediating complexes were unable to bind FADD to activate the intracellular apoptotic cascade. Nuclear magnetic resonance (NMR) studies by Huang et al. and Martin et al. showed that some death domain mutations cause global disruption of the Fas death domain, while others seem to act locally in the area of the molecule that normally contacts FADD. All interfered with formation of a functional death-inducing signaling complex.
The mechanism whereby Fas proteins bearing extracellular mutations interfere with apoptosis has been more difficult to define. The studies of Vaishnaw et al. suggest that extracellular missense mutations may interfere with the ability of Fas to be engaged by FasL. Mutations predicting synthesis of truncated fragments of the extracellular Fas domain are potentially secreted in soluble form and could interfere with FasL; alternatively, such fragments may be associated with the extracellular portionsof normal Fas chains, preventing a functional trimeric Fas complex from engaging FasL. Another alternative for the reason why extracellular domain mutations cause ALPS is that they are not expressed, and therefore the total amount of Fas produced by the cell is lower than necessary for effective apoptosis (haploinsufficiency). However, studies by Jackson et al. indicate some degree of dominant interference when mutant Fas molecules with extracellular truncation are co-expressed with wild type Fas. Further studies are needed to determine which of these mechanisms apply in vivo.
The extreme variability in clinical features of family members with the same Fas mutation is not understood, but three recent studies have clarified why some families have higher penetrance, that is more mutation-bearing members with clincal problems. Jackson et al analyzed sixty Fas mutation-bearing relatives of ALPS probands and found that 18 percent (3/17) of relatives with extracellular mutations versus 88 percent (38/43) of relatives with intracellular mutations (p<.001) displayed one or more features of ALPS (adenopathy, significant titers of autoantibodies, overt autoimmunity, or elevated DNTs). Additionally, penetrance of ALPS-related significant morbidity (splenectomy, autoimmune disease requiring treatment, or lymphoma) was 44 percent (19/43) for relatives with intracellular mutations versus 0 percent (0/17) for relatives with extracellular mutations (p<.001).
Vaishnaw et al. similarly found penetrance of the full-blown disease to be 60 percent (6 out of 10) for relatives with intracellular mutations and 0 percent (0 out of 5) for relatives with extracellular mutations. Rieux-Laucat et al. reported penetrance of clinical features to be 100 percent (7 out of 7) for relatives with intracellular mutations and 38 percent (3 out of 8) for relatives with extracellular mutations.
Clearly, increased penetrance of ALPS features and ALPS-related significant morbidity is associated with mutations that disrupt the intracellular portion of the Fas protein containing the death domain. However, the existence of healthy mutation-bearing relatives in ALPS kindreds suggests that additional factors influence the development of the ALPS phenotype. These might include either protective or deleterious genetic changes, or might be related to infections or other environmental agents.
Wang et al. have described mutations in the gene CASP10, encoding the apoptotic cysteine protease caspase 10 in two ALPS patients, who have been designated as having ALPS type II. One of the two patients had typical ALPS and a heterozygous dominant mutation of caspase 10 which caused interference with the cell death pathway. The other patient had the same mutation in both caspase 10 alleles. The physiologic role of caspase 10 has not been completely defined, but this study provides evidence that caspase 10 may interact with caspase 8 in the Fas apoptotic pathway and that caspase 10 is also important in the death pathway mediated by the ligand TRAIL. The patients had defects in apoptosis mediated by multiple death receptors. Interestingly, a specific apoptotic defect in dendritic cells was demonstrated for the first patient, whose lymph nodes showed abnormally numerous dendritic cells. Accumulation of dendritic cells which would normally be eliminated by apoptosis would be predicted to lead to excessive antigen presentation, a new mechanism to explain development of autoimmunity in ALPS type II.
ALPS is the first pediatric syndrome described in which the primary defect is in programmed cell death. Defective apoptosis in lymphocytes (and in ALPS Type II, dendritic cells) leads to accumulation these cells in the lymphoid organs after they would normally be eliminated. As a result, cells with autoimmune potential are unchecked and can induce a variety of autoimmune diseases, and the risk for malignant transformation to lymphoma is increased. Most patients with ALPS have dominant interfering mutations in the apoptosis mediator Fas, but additional genes in the Fas pathway are already known to be mutated in some ALPS patients. Further efforts will be needed to uncover additional genetic defects underlying ALPS and elucidate why the clinical spectrum is highly variable.
The authors thank past and current members of the NIH ALPS group, Stephen E. Straus, Michael J. Lenardo, Youngnim Choi, Janet K. Dale, Roxanne E. Fischer, Galen H. Fisher, Thomas A. Fleisher, Ivan J. Fuss, Amy P. Hsu, Elaine S. Jaffe, Megan S. Lim, David A. Martin, Lindsay A. Middelton, Julie Niemela, Richard M. Siegel, Michael C. Sneller, Warren Strober, and Jin Wang.
back to top
Last Updated June 14, 2009