Blood Cells, Molecules and Diseases 54 (2015) 53–55
Contents lists available at ScienceDirect
Blood Cells, Molecules and Diseases journal homepage: www.elsevier.com/locate/bcmd
Short Communication
A late onset sickle cell disease reveals a mosaic segmental uniparental isodisomy of chromosome 11p15 Isabelle Vinatier a, Xavier Martin b, Jean-Marc Costa a, Anne Bazin a, Stéphane Giraudier c, Philippe Joly d,⁎ a
Laboratoire CERBA, 95066 Cergy-Pontoise cedex 9, France Service de médecine polyvalente, Centre hospitalier Antoine Gayraud, 11890 Carcassonne cedex 9, France Service d'hématologie biologique, GH Henri-Mondor, 51 avenue du Maréchal de Lattre de Tassigny, 94010 Créteil Cedex, France d Unité de Pathologie Moléculaire du Globule Rouge, Laboratoire de Biochimie et Biologie moléculaire, Hôpital Edouard Herriot, Hospices Civils de Lyon & Université Claude Bernard-Lyon 1, Lyon, France b c
a r t i c l e
i n f o
Article history: Submitted 20 May 2014 Accepted 28 July 2014 Available online 20 August 2014 (Communicated by M. Narla, DSc, 28 July 2014) Keywords: Late-onset SCD Mosaicism Loss of heterozygosity 11p15 isoparental isodisomy
a b s t r a c t We report, in a 78-year old man constitutionally heterozygous for the sickle cell trait, a late onset sickle cell disease (SCD) caused by a mosaic segmental uniparental isodisomy of chromosome 11p15. The mosaic loss of heterozygosity (LOH) of the HBB gene was suggested in front of an unusually weak βA peak at Sanger direct sequencing and a semi-quantitative FRET Light Cycler method which showed a low expression of the βA allele compared to the βS allele. A SNP array analysis then revealed a 45.9 Mb LOH on almost the whole short arm of chromosome 11 without any copy loss number and with an estimated level of mosaicism of 80%. Culture and genotyping of erythroblastic burst forming units confirmed the presence of AS and SS hematopoietic cells in the proportions of 2/3 and 1/3, respectively. Such a late-onset SCD had already been described but for a much younger patient (a 14-year-old boy). This discrepancy could be explained either by a much lower degree of mosaicism at birth in our proband (and thus a much more delayed clinical expression) or by inter-individual variations (modifier genes for example) that could have slowed down the positive selection of S/S clones. © 2014 Elsevier Inc. All rights reserved.
Loss of heterozygosity (LOH) of chromosome 11p15 by deletion or mosaic segmental uniparental isodisomy (MSUI) has already been involved in many cases of late onset beta-thalassemia major (TM) or intermedia (TI) and in one case of moderate sickle cell disease (SCD) syndrome (Table 1). We report a similar mosaicism for SS and AS erythrocytes revealed in a 78-year-old man suffering from a corticoresistant hemolytic anemia (6.7 g/dL). The proband is a French Caucasian man who addressed to a hematology practitioner at 70-years old to support a fortuitously discovered anemia. His main medical records were a non-treated type 2 diabetes mellitus, an acute pancreatitis following cholecystectomy and a previous shingles episode. A hemolytic anemia (Hb 10.4 g/dL) with a large splenomegaly (14 cm) was found and not initially treated because of the negativity of the direct Coombs test (IgG, C3d, IgA). However, in the following years, the Hb level continued to drop (8 g/dL) and the diagnosis of auto-immune hemolytic anemia was made by default, leading to some corticosteroids treatments that all revealed unsuccessful. At 78 years old, the Hb level had fall to 6.7 g/dL and the diagnosis was completely reevaluated. Besides many other biological tests (myelogram, medullar karyotype, nocturnal paroxysmal hemoglobinuria clone, auto-antibodies, viral serologies, etc.) which all revealed negative, a hemoglobinopathy screening ⁎ Corresponding author. E-mail address:
[email protected] (P. Joly).
http://dx.doi.org/10.1016/j.bcmd.2014.07.021 1079-9796/© 2014 Elsevier Inc. All rights reserved.
was performed and showed the presence of HbS at the atypical level of 44.5% at cation-exchange high performance liquid chromatography (CE-HPLC) (Biorad-Variant II apparatus — β-thalassemia dual kit) later confirmed (47%) by capillary electrophoresis (Sebia — Capillarys 2 Flex Piercing — Hemoglobin kit). After written agreement, Sanger sequencing of the HBB gene was carried out using DNA obtained from blood leucocytes with the Chemagen® kit (Janus® apparatus — PerkinElmer). Except the β6 GAG N GTG (HBB:c.20A N T) mutation in homozygous condition, no anomaly was found but, exactly as described by Joly et al. [1], a weak βA peak was also characterized suggesting the presence of the βA allele in low dose. This was confirmed by a more sensitive in-house semiquantitative fluorescence resonance energy transfer (FRET) method (Light Cycler® 480 — Roche) which allowed us to evaluate at around 16% and 84% the respective proportions of the βA and βS alleles (Fig. 1). Culture of erythroblastic burst forming units could be practiced and revealed a mosaicism of A/S and S/S hematopoietic cells: among the 40 obtained colonies, 2/3 were S/S and 1/3 A/S. In other words, in blood cells DNA, 5 βS alleles are expected for a single βA allele, that is to say a proportion of βA alleles of about 16.7% that confirms the FRET Light cycler results. Even if a parental study was not possible, we highly suspected that our proband was constitutionally A/S but with a mosaic uniparental isodisomy of the βS allele, as described a few years ago [2]. A SNP array analysis (CytoScan HD SNP Affymetrix) confirmed this hypothesis as it revealed a 45.9 Mb LOH on almost the whole short arm of
54
I. Vinatier et al. / Blood Cells, Molecules and Diseases 54 (2015) 53–55
Table 1 Literature review for the late-onset hemoglobinopathies cases of non-Mendelian transmission. Year
First author (reference)
Hemoglobinopathy
Genetic mechanism
Parental allele implied in hemoglobinopathy
LOH length
Degree of Sex (M/F) mosaicism and age (%) when describing
1992 Beldjord [7]
Beta-thalassemia major + neonatal Beckwith– Wiedemann syndrome
Paternal
ND
NA
M — 1st year of life
2002 Badens [8]
Beta-thalassemia intermedia
Paternal
ND
75%
M — 10 years
2004 Galanello [9]
Beta-thalassemia intermedia (2 cases) Beta-thalassemia major
Isoparental isodisomy of chromosome 11p (germline event) Mosaicism deletion of chromosome 11p Mosaicism deletion(*) of chromosome 11p MSUI of chromosome 11p MSUI of chromosome 11p MSUI of chromosome 11p
Paternal
ND 90% ND
M — 42 years F — 23 years F — 28 years
84%
M — 14 years
Mosaicism deletion of chromosome 11p MSUI chromosome 11p MSUI of chromosome 11p
Maternal
8 Mb ND ~26 Mb (11p14.3 to 11p15.5) ~27 Mb (11p14.2 to 11p15.5) 47.7 Mb 47.2 Mb 48.9 Mb ND
65% 80% 70% ND
F — 30 years M — 43 years M — 51 years F — 39 years
Paternal ND
8.8 Mb 45 Mb
80% 80%
F — 21 years M — 78 years
2008 Chang [4] 2010 Swensen [2] 2012 Harteveld [6]
2012 Joly (7) 2013 Bento [5] 2014 Vinatier (this paper)
Moderate sickle cell disease syndrome (14-year-old boy) Beta-thalassemia major (3 cases)
Unusual low sickle cell trait (12% HbS) Beta-thalassemia major Moderate sickle cell disease syndrome (78-year-old man)
Paternal Paternal Paternal
ND: not determined; MSUI: mosaic segmental uniparental isodisomy; NA: not applicable; LOH: loss of heterozygosity; and (*): deletion mentioned but potential MSUI.
chromosome 11 (Fig. 2) without any copy loss number and with an estimated level of mosaicism of 80% [3]. Genomic DNA from oral mucosal cells was also analyzed with our semi-quantitative FRET Light Cycler
method: respective proportions of 29% and 71% of βA and βS alleles were found, allowing us to conclude that the mosaicism concerned other tissues than the hematopoietic lineage but to a different extent.
Fig. 1. Estimation of the proportion of βA alleles (versus βS alleles) in blood and oral mucosal cells with an in-house semi-quantitative FRET Light Cycler method. (A) Calibration graphs obtained by mixing in appropriate proportions βA/βA and βS/βS DNA. The Tm of the βA allele is around 62 °C and the peaks height is directly proportional to the βA allele proportion in the mix. (B) Graphs obtained with DNA extracted from blood and oral mucosal cells of the propositus and showing respective proportions of 16 and 29% for the βA allele (calibration plot not shown).
I. Vinatier et al. / Blood Cells, Molecules and Diseases 54 (2015) 53–55
55
Fig. 2. SNP array data analysis showing a mosaic segmental (45.9 Mb) uniparental isodisomy of chromosome 11p. (A) The log R ratio is consistent with normal copy number; (B) The B allele frequency shows a 45.9 Mb loss of heterozygosity of chromosome 11p15 as well as a level of mosaicism of about 80%. Taken together, these observations lead to the conclusion of MSUI of chromosome 11p15.
Such a non-Mendelian SCD transmission has already been described by Swensen et al. but for a much younger patient (a 14-year-old boy) [2]. In both cases, the SCD syndrome was revealed by acute and chronic hemolytic anemia with an associated splenomegaly and not with the other typical symptoms of the disease (painful vaso-occlusive crisis, acute chest syndrome, priapism, etc.). We can hypothesize that the presence of AS red cells in the circulation is sufficient to prevent the vaso-occlusion process to occur, as it is the case for most of βS/β+-thal patients. Very interestingly, if we consider that AS red cells contain 35 to 40% HbS, we can estimate the proportion of AS cells in the blood flow to about 70% whereas it has been shown to be of one third only in BFU-E colonies. This discrepancy reveals a very short lifespan of SS red cells that indirectly confirms the importance of the hemolytic process. We have no proof that the βs allele is of paternal origin but this is the highest probability since all late onset β-globin disorders caused by MSUI described so far were paternally inherited due to the parental imprinting genes IGF2 (paternal expression) and H19 (maternal expression) located on the 11p15 region. Indeed, overexpression of IGF2 confers a selective advantage to the homozygote or hemizygote clones bearing the paternal 11p15 region only [4]. Provided that this paternal allele also bears a β-thal or a βS allele, it thus gives rise to a late onset β-thalassemia major or SCD syndrome since homozygous or hemizygous βX cells progressively replace the heterozygous βX/βA cells. On the contrary, if the βX mutation takes place on the maternal allele, heterozygous βX/βA cells progressively disappear from the hematopoietic lineage [1]. Our description of late-onset SCD is thus very similar to that of Swensen except one thing: the age of diagnosis was 78 versus 14. Maybe our proband had a much lower degree of mosaicism at birth and thus a much more delayed clinical expression (even if the previous cholecystectomy was possibly the consequence of a sub-clinical hemolysis). We can also imagine that the degree of mosaicism was the same at birth in both cases but that inter-individual variations and modifier genes have slowed down the selection of the βS/βS clones in our case. Anyway, the lower degree of mosaicism in oral mucosal cells compared to the hematopoietic lineage would indicate a tissue-dependent kinetic for the clonal selection. A comparative study of the seven cases of late onset TM demonstrates no correlation between percentage of mosaicism and (i) age of onset for transfusion-dependency (21 years to 51 years) or (ii) LOH extent (~8 Mb to ~50 Mb). Sex only could eventually be linked with a more rapid evolution in TM: before 30 years for women versus after 40 years for men. Interestingly, despite several
authors [4–6] have described the clinical worsening of the disease, none has yet been able to document the evolution of mosaicism in the same time. We can conclude that, when clinical signs are present, the diagnosis of SCD or ΤΜ/ΤΙ must always be excluded by hemoglobin analyses whatever the age or the family context. Now that the molecular principles of late onset β-hemoglobinopathies have been well established, it remains to determine which parameters influence both clinical severity and age of expression and this study brings a new stone to do so. Authorship and disclosures IV was the principal investigator who coordinated the research and takes primary responsibility for the paper. AB performed the SNParray analysis and JMC the allelic dosage. SG realized the BFU-E culture. XM was the hematologist physician who addressed us the propositus. IV and PJ wrote the paper. The authors reported no potential conflicts of interest. References [1] P. Joly, C. Schluth-Bolard, P. Lacan, C. Barro, S. Pissard, A. Labalme, D. Sanlaville, C. Badens, HBB loss of heterozygosity in the hemopoietic lineage gives rise to an unusual sickle-cell trait phenotype, Haematologica 98 (2013) e7–e8. [2] J.J. Swensen, A.M. Agarwal, J.M. Esquilin, S. Swierczek, A. Perumbeti, D. Hussey, M. Lee, C.H. Joiner, G. Pont-Kingdon, E. Lyon, J.T. Prchal, Sickle cell disease resulting from uniparental disomy in a child who inherited sickle cell trait, Blood 116 (2010) 2822–2825. [3] J.R. Gonzalez, B. Rodriguez-Santiago, A. Caceres, R. Pique-Regi, N. Rothman, S.J. Chanock, L. Armengol, L.A. Perez-Jurado, A fast and accurate method to detect allelic genomic imbalances underlying mosaic rearrangements using SNP array data, BMC Bioinforma. 12 (2011) 166. [4] J.G. Chang, W.C. Tsai, I.W. Chong, C.S. Chang, C.C. Lin, T.C. Liu, {beta}-thalassemia major evolution from {beta}-thalassemia minor is associated with paternal uniparental isodisomy of chromosome 11p15, Haematologica 93 (2008) 913–916. [5] C. Bento, T.M. Maia, J.D. Milosevic, I.M. Carreira, R. Kralovics, M.L. Ribeiro, Beta thalassemia major due to acquired uniparental disomy in a previously healthy adolescent, Haematologica 98 (2013) e4–e6. [6] C.L. Harteveld, C. Refaldi, A. Giambona, C.A. Ruivenkamp, M.J. Hoffer, J. Pijpe, P. De Knijff, C. Borgna-Pignatti, A. Maggio, M.D. Cappellini, P.C. Giordano, Mosaic segmental uniparental isodisomy and progressive clonal selection: a common mechanism of late onset beta-thalassemia major, Haematologica 98 (2012) 691–695. [7] C. Beldjord, I. Henry, C. Bennani, D. Vanhaeke, D. Labie, Uniparental disomy: a novel mechanism for thalassemia major, Blood 80 (1992) 287–289. [8] C. Badens, M.G. Mattei, A.M. Imbert, C. Lapoumeroulie, N. Martini, G. Michel, D. LenaRusso, A novel mechanism for thalassaemia intermedia, Lancet 359 (2002) 132–133. [9] R. Galanello, L. Perseu, C. Perra, L. Maccioni, S. Barella, M. Longinotti, A. Cao, M. Cazzola, Somatic deletion of the normal beta-globin gene leading to thalassaemia intermedia in heterozygous beta-thalassaemic patients, Br. J. Haematol. 127 (2004) 604–606.