A double hit implicates DIAPH3 as an autism risk gene.

Recent studies have shown that more than 10% of autism cases are caused by de novo structural genomic rearrangements. Given that some heritable copy number variants (CNVs) have been observed in patients as well as in healthy controls, to date little attention has been paid to the potential function of these non-de novo CNVs in causing autism. A normally intelligent patient with autism, with non-affected parents, was identified with a maternally inherited 10 Mb deletion at 13q21.2. Sequencing of the genes within the deletion identified a paternally inherited nonsynonymous amino-acid substitution at position 614 of diaphanous homolog 3 (DIAPH3) (proline to threonine; Pro614Thr). This variant, present in a highly conserved domain, was not found in 328 healthy subjects. Experiments showed a transient expression of Diaph3 in the developing murine cerebral cortex, indicating it has a function in brain development. Transfection of Pro614Thr in murine fibroblasts showed a significant reduction in the number of induced filopodia in comparison to the wild-type gene. DIAPH3 is involved in cell migration, axon guidance and neuritogenesis, and is suggested to function downstream of SHANK3. Our findings strongly suggest DIAPH3 as a novel autism susceptibility gene. Moreover, this report of a 'double-hit' compound heterozygote for a large, maternally inherited, genomic deletion and a paternally inherited rare missense mutation shows that not only de novo genomic variants in patients should be taken seriously in further study but that inherited CNVs may also provide valuable information.

Disruption of ETV6 in intron 2 results in upregulatory and insertional events in childhood acute lymphoblastic leukaemia.

We describe four cases of childhood B-cell progenitor acute lymphoblastic leukaemia (BCP-ALL) and one of T-cell (T-ALL) with unexpected numbers of interphase signals for ETV6 with an ETV6-RUNX1 fusion probe. Three fusion negative cases each had a telomeric part of 12p terminating within intron 2 of ETV6, attached to sequences from 5q, 7p and 7q, respectively. Two fusion positive cases, with partial insertions of ETV6 into chromosome 21, also had a breakpoint in intron 2. Fluorescence in situ hybridisation (FISH), array comparative genomic hybridization (aCGH) and Molecular Copy-Number Counting (MCC) results were concordant for the T-cell case. Sequences downstream of TLX3 on chromosome 5 were deleted, leaving the intact gene closely apposed to the first two exons of ETV6 and its upstream promoter. qRT-PCR showed a significant upregulation of TLX3. In this study we provide the first incontrovertible evidence that the upstream promoter of ETV6 attached to the first two exons of the gene was responsible for the ectopic expression of a proto-oncogene that became abnormally close as the result of deletion and translocation. We have also shown breakpoints in intron 2 of ETV6 in two cases of insertion with ETV6-RUNX1 fusion.

Detailed analysis of 22q11.2 with a high density MLPA probe set.

The presence of chromosome-specific low-copy repeats (LCRs) predisposes chromosome 22 to deletions and duplications. The current diagnostic procedure for detecting aberrations at 22q11.2 is chromosomal analysis coupled with fluorescence in situ hybridization (FISH) or PCR-based multiplex ligation dependent probe amplification (MLPA). However, there are copy number variations (CNVs) in 22q11.2 that are only detected by high-resolution platforms such as array comparative genomic hybridization (aCGH). We report on development of a high-definition MLPA (MLPA-HD) 22q11 kit that detects copy number changes at 37 loci on the long arm of chromosome 22. These include the 3-Mb region commonly deleted in DiGeorge/velocardiofacial syndrome (DGS/VCFS), the cat eye syndrome (CES) region, and more distal regions in 22q11 that have recently been shown to be deleted. We have used this MLPA-HD probe set to analyze 363 previously well-characterized samples with a variety of different rearrangements at 22q11 and demonstrate that it can detect copy number alterations with high sensitivity and specificity. In addition to detection of the common recurrent deletions associated with DGS/VCFS, variant and novel chromosome 22 aberrations have been detected. These include duplications within as well as deletions distal to this region. Further, the MLPA-HD detects deletion endpoint differences between patients with the common 3-Mb deletion. The MLPA-HD kit is proposed as a cost effective alternative to the currently available detection methods for individuals with features of the 22q11 aberrations. In patients with the relevant phenotypic characteristics, this MLPA-HD probe set could replace FISH for the clinical diagnosis of 22q11.2 deletions and duplications.

A de novo (1;2;3;15;18) chromosome rearrangement with six nonreciprocal translocations.

A de novo complex chromosome rearrangement (CCR) found in a phenotypically abnormal boy was characterized by G-bands, FISH with subtelomere probes, and M-FISH. The G-banding analysis revealed involvement of chromosomes 1, 2, 3, 15, and 18 with (at least) eight breakpoints, five nonreciprocal translocations (1q --> 2q --> 8q --> 15q --> 2p --> 1q), and a 3p insertion into the der(2); there was also a presumptive deletion of 1q41. The 5 derivatives were described as follows: der(1)(1pter --> 1q32.3?::2p21--> 2pter),der(2)(1qter --> 1q42?::2q24.2 --> 2p21::3p13 --> 3p26::15q15 --> 15qter),der(3)(3qter --> 3p13:),der(15)(15pter --> 15q15::18q11 --> 18qter),der(18)(18pter --> 18q11::2q24.2 --> 2qter). The molecular assays confirmed the segmental composition of each derivative and documented the localization of most relevant telomeres. In addition to the novelty of the 1, 2, 3, 15 and 18 combination, this CCR may also be unique in the sense that it represents a cluster of 6 nonreciprocal transpositions regardless of the occurrence (or lack thereof) of secondary unbalances. Finally, there appears to be an excess of CCRs in fetuses conceived by intracytoplasmic sperm injection.

MLPA: a rapid, reliable, and sensitive method for detection and analysis of abnormalities of 22q.

In this study, essential test characteristics of the recently described multiplex ligation-dependent probe amplification (MLPA) method are presented, using chromosome 22 as a model. This novel method allows the relative quantification of approximately 40-45 different target DNA sequences in a single reaction. For the purpose of this study, MLPA was performed in a blinded manner on a training set containing over 50 samples, including typical 22q11.2 deletions, various atypical deletions, duplications (trisomy and tetrasomy), and unbalanced translocations. All samples in the training set have been previously characterized by fluorescence in situ hybridization (FISH) with cosmid or BAC clones and/or cytogenetic studies. MLPA findings were consistent with cytogenetic and FISH studies, no rearrangement went undetected and repeated tests gave consistent results. At a relative change in comparative signal strength of 30% or more, sensitivity and specificity values were 0.95 and 0.99, respectively. Given that MLPA is likely to be used as an initial screening method, a higher sensitivity, at the cost of a lower specificity, was deemed more appropriate. A receiver operator characteristic (ROC) curve analysis was performed to calculate the most optimal threshold range, with associated sensitivity and specificity values of 0.99 and 0.97, respectively. Finally, performance of each individual probe was analyzed, providing further useful information for the interpretation of MLPA results. In conclusion, MLPA has proven to be a highly sensitive and accurate tool for detecting copy number changes in the 22q11.2 region, making it a fast and economic alternative to currently used methods. The current study provides valuable and detailed information on the characteristics of this novel method.

Derivative chromosome 9 deletions are a significant feature of childhood Philadelphia chromosome positive acute lymphoblastic leukaemia.

Deletions from the derivative chromosome 9, der(9), of the translocation, t(9;22)(q34;q11), at the site of the ABL/BCR fusion gene, have been demonstrated by fluorescence in situ hybridisation (FISH), in both Philadelphia chromosome (Ph)-positive chronic myeloid leukaemia (CML) and acute lymphoblastic leukaemia (ALL). In CML they occur in 10-15% of cases and appear to indicate a worse prognosis, whereas in ALL, the situation is unclear. This study presents the findings of dual fusion FISH used to detect such deletions in a series of 27 BCR/ ABL-positive childhood ALL patients. Metaphase FISH was essential for the accurate interpretation of interphase FISH signal patterns. Three cases (11%) had a single fusion signal, resulting from deletions of the der(9). Three other patients with variant translocations and one with an insertion, also had a single fusion, but with no evidence of deletions. Gain of a fusion in approximately one-third of patients indicated a second Ph, which appears to be a diagnostic marker of Ph-positive ALL. This study shows that the incidence of deletions from the der(9) in childhood ALL is at least as high as that reported for CML.

Amplification of AML1 on a duplicated chromosome 21 in acute lymphoblastic leukemia: a study of 20 cases.

This study identifies multiple copies of the AML1 gene on a duplicated chromosome 21, dup(21), as a recurrent abnormality in acute lymphoblastic leukemia (ALL). Clusters of AML1 signals were visible at interphase by fluorescence in situ hybridization (FISH). In metaphase, they appeared tandemly duplicated on marker chromosomes of five distinct morphological types: large or small acrocentrics, metacentrics, submetacentrics or rings. The markers comprised only chromosome 21 material. Karyotypes were near-diploid and, besides dup(21), no other established chromosomal changes were observed. A total of 20 patients, 1.5 and <0.5% among consecutive series of childhood and adult ALL respectively, showed this phenomenon. Their median age was 9 years, white cell counts were low and all had a pre-B/common immunophenotype. Although this series is not the first report of this abnormality, it is the largest, permitting a detailed description of the variety of morphological forms that duplicated chromosome 21 can assume.

ETV6/AML1 fusion by FISH in adult acute lymphoblastic leukemia.

Dual-color interphase fluorescence in situ hybridization (FISH) with ETV6 and AML1 probes was used for the first time on a series of 159 adult patients with acute lymphoblastic leukemia (ALL), for detection of the t(12;21)(p13;q22) translocation. Seven patients (4.4%) were found, with 50-100% of positive cells, of whom one of two tested, proved negative for the fusion product by RT-PCR. Two of them, aged 43 and 50 years, are the oldest patients so far confirmed to have the translocation. Three who relapsed at 10, 11 and 24 months, suggest that adults may not enjoy the good short-term prognosis reported for t(12;21)-positive children. Thirty-one-negative cases had signal numbers differing from the two expected for each gene. In 15 cases these results were consistent with the karyotype. In nine cases with uninformative cytogenetics, the numbers were consistent with those for centromeres and indicated a hidden aneuploidy. Loss of ETV6 genes in two cases and AML1 amplification in three others were not suspected from the cytogenetics. In conclusion, FISH proved to be reliable in defining ETV6/AML1 positivity in this group of patients as well as providing valuable insights into negative cases.

Molecular tracking of leukemogenesis in a triplet pregnancy.

The occurrence of childhood acute lymphoblastic leukemia (ALL) in 2 of 3 triplets provided a unique opportunity for the investigation of leukemogenesis and the natural history of ALL. The 2 leukemic triplets were monozygotic twins and shared an identical, acquired TEL-AML1 genomic fusion sequence indicative of a single-cell origin in utero in one fetus followed by dissemination of clonal progeny to the comonozygotic twin by intraplacental transfer. In accord with this interpretation, clonotypic TEL-AML1 fusion sequences could be amplified from the archived neonatal blood spots of the leukemic twins. The blood spot of the third, healthy, dizygotic triplet was also fusion gene positive in a single segment, though at age 3 years, his blood was found negative by sensitive polymerase chain reaction (PCR) screening for the genomic sequence and by reverse transcription-PCR. Leukemic cells in both twins had, in addition to TEL-AML1 fusion, a deletion of the normal, nonrearranged TEL allele. However, this genetic change was found by fluorescence in situ hybridization to be subclonal in both twins. Furthermore, mapping of the genomic boundaries of TEL deletions using microsatellite markers indicated that they were individually distinct in the twins and therefore must have arisen as independent and secondary events, probably after birth. These data support a multihit temporal model for the pathogenesis of the common form of childhood leukemia.

Analysis of ETV6/AML1 abnormalities in acute lymphoblastic leukaemia: incidence, alternative spliced forms and minimal residual disease value.

The t(12;21)(p13;q22) translocation, resulting in the fusion of the ETV6 and AML1 genes, occurs in 20-25% of paediatric B-lineage acute lymphoblastic leukaemias (ALL). The identification of the fusion product has important prognostic and therapeutic implications as the translocation has been associated with a favourable clinical outcome. The aim of this study was threefold: (i) to assess the frequency and clinical association of the fusion gene in patients with and without a cytogenetically detectable chromosome 12 and/or 21 abnormality or failed cytogenetic results, (ii) to characterize alternative forms of ETV6/AML1 transcripts, and (iii) to use ETV6/AML1 as a molecular marker for the investigation of minimal residual disease (MRD). ETV6/AML1 fusion was detected in 22 (39%) of 56 cases studied by reverse transcriptase polymerase chain reaction (RT-PCR). ETV6/AML1 fusion was found in nine out of 16 (56%) cases with a cytogenetically visible chromosome 12 abnormality, but also in nine out of 29 patients (31%) without a chromosome 12 abnormality or patients with failed cytogenetics (four out of 11 patients, 36%), making this the most common cytogenetic abnormality in childhood ALL. Alternatively spliced ETV6/AML1 forms were investigated in 14 of the positive patients. Exon 5 of ETV6 was fused in frame to all AML1 exons, except exon 4. Fusion to exon 6 of AML1 resulted in one amino acid change. The presence of ETV6/AML1 was associated with a lower white blood cell count (Student's t-test; P = 0.009) and common (c)ALL phenotype (chi(2) test; P > 0.001), but no better disease-free survival. Our study shows that (i) RT-PCR is the most effective approach for the detection of t(12;21) in childhood ALL, (ii) the association of ETV6/AML1 and chromosome 12 and/or 21, seen in 56% of our cases, further confirms existing data, (iii) overall survival of patients with t(12;21) was not better than other cytogenetics groups, and (d) MRD analysis using ETV6/AML1 fusion is specific, but not sensitive enough to avoid false negative results.

Human cytotrophoblastic cells absorb the NK blocking activity of monoclonal BA11.

PROBLEM: R80K is a polymorphic alloantigeneic protein present on human placental trophoblast and on paternal B lymphocytes and monocytes. This protein, unlike the former candidate TLX antigen, stimulates a protective maternal immune response in vivo. A murine monoclonal BA11 antibody, directed against R80K, prevents abortion in three murine pregnancy-failure models and inhibits human and murine NK activity. We attempted to define the target of BA11 in the human NK assay system. METHODS: A CELISA method was used to detect R80K antigen on the surface of different cells using the BA11 antibody. The effect, on human peripheral blood NK activity against K562, by BA11 before and after absorption by different cells, including the K562 target, was determined. RESULTS: R80K was detected on term placental syncytio and cytotrophoblast and on BeWo cells, by CELISA. BA11 suppressed NK lysis of K562 cell sin a dose-dependent manner. Absorption of the BA11 by BeWo and by cytotrophoblastic cells significantly decreased the NK-inhibitory activity. There was minimal absorption by K562 and BA11-pretreateed K562 cells remained susceptible to NK lysis. By contrast, BA11-pretreated peripheral blood cells lost all NK activity. CONCLUSIONS: The inhibition of NK killing of K562 cells by BA11 is more complex than simple masking of a trophoblast cell-associated molecule in K562 necessary for recognition in NK cells.

Immunosuppressive properties of monoclonal antibodies and human polyclonal alloantibodies to the R80K protein of trophoblast.

PROBLEM: The R80K protein on human trophoblast is antigenically polymorphic, and in all placentae of successful pregnancies, the protein is covered by maternal alloantibody. Alloantibody eluted from human placenta has been shown to inhibit killing by human NK cells. Do those antibodies to R80K that inhibit NK killing also affect the murine abortion models? METHODS: We made three murine monoclonal antibodies to conserved epitopes, on human R80K, all of which also reacted with the homologous murine molecule. One antibody only, BA11, suppressed NK cytotoxicity to K562 and of mouse spleen NK cells to murine trophoblast. All three were tested in mouse models of abortion: the CBA x DBA/2 model with a high resorption rate of F1 embryos compared with the parental strains, an endotoxin induced abortion/resorption model and a third model in which the pregnant mouse is subject to sonic stress. CONCLUSION: Those IgG antibodies eluted from microvesicles which bound to K562, and one of the three monoclonals, BA11, inhibited NK killing. The antibodies react with the murine molecule, and BA11 inhibited abortion in all three mouse abortion models. This reinforces the thesis that interference with NK killing can influence abortion/resorption in mice, and the BA11 antibody may effect similar results in analogous human situations.

Anomalous inheritance of a paternally derived trophoblast antigen.

PROBLEM: Recurrent spontaneous abortion occurs in 1 in 500 random matings and usually results in abortion of all pregnancies. If absence of antibody to a paternally derived antigen caused abortion, the woman would be expected to make antibody to the other paternal antigen and abort only half her pregnancies. METHODS: Microvesicles were prepared from equine placentae. Acid-eluted IgG antibody was eluted from the polymorphic R80K antigen and used to type the residual R80K antigen on vesicles or on peripheral blood leucocytes. RESULTS: In several equine sibships all the half-sibs had the same paternal R80K alloantigen. In the extended horse family descended from the stallion Nearco, three allotypes were found. The allele present was usually the grandpaternal one, but exceptions are seen. Whichever allele is transmitted all the progeny have the same alloantigen (probability of this occurring by chance = 2(-45)). CONCLUSION: Because only one paternal allotype is present in all progeny, lack of antibody to the R80K antigen would result in loss of all pregnancies, not one half.

An 80-kDa syncytiotrophoblast alloantigen bound to maternal alloantibody in term placenta.

PROBLEM: We have shown that most of the IgG present on term syncytiotrophoblast, membrane, microvesicles is bound to an 80 kDa protein antigen (R80K). METHODS: Microvesicles were prepared from term human placenta, and the IgG eluted at pH3. RESULTS: When IgG antibody was eluted at pH3 and reacted with acid-treated vesicles of other placentae, the alloantibody always bound to the preparation from which it was obtained, but only to about 10% of acid-treated preparations from other placentae. A similar polymorphic protein found in association with IgG antibody was found in term horse placentae. Cross-reactivity of the antibodies between species was not found. Using binding of labelled antibody, complement dependent cytotoxicity and FACS two-color analysis, the human polymorphic antigen was present on peripheral blood monocytes and B-lymphocytes. The R80k antigen on intact microvesicles was resistant to trypsin, but after acid elution of IgG, trypsin released a soluble 50 kDa fragment which reacted with the acid-eluted IgG antibody. CONCLUSION: The presence of antibodies to R80K in all term placentae studied, including first pregnancies, suggests that development of this alloantibody may be a normal requirement for successful pregnancy.