HUMAN MUTATION Mutation in Brief #886(2006) Online
MUTATION IN BRIEF
© 2006 WILEY-LISS, INC.
Received 21 October 2005; accepted revised manuscript 21 December 2005.
MPL Mutations in 23 Patients Suffering from
Congenital Amegakaryocytic Thrombocytopenia: The
Type of Mutation Predicts the Course of the Disease
Manuela Germeshausen*, Matthias Ballmaier, and Karl Welte
Department of Pediatric Hematology and Oncology, Hannover Medical School, Hannover, Germany
*Correspondence to: Manuela Germeshausen, Pediatric Hematology and Oncology, Hannover Medical School,
Carl-Neuberg-Str. 1, D-30625 Hannover , Germany; E-mail: Germeshausen.Manuela@mh-hannover.de
Communicated by Sergio Ottolenghi
Congenital amegakaryocytic thrombocytopenia (CAMT) is a rare inherited bone marrow
failure syndrome. Mutations in the gene for the thrombopoietin receptor MPL were defined
as the molecular cause in CAMT patients. Extending our sequence analyses from eight to a
total of now 23 CAMT patients we could verify our hypothesis of genotype-phenotype
correlation in CAMT. Seven different mutations predicted to lead to a complete loss of
function of the thrombopoietin receptor were found in 13 patients belonging to group
CAMT I with persistently low platelet counts and a fast progression into pancytopenia. Nine
different missense mutations were detected in 10 patients of group CAMT II, characterized
by a transient increase in platelet counts over 50 nl-1 during the first years of life. Using in
vitro assays with hematopoietic progenitors from patients of both patient groups we could
provide experimental evidence for a residual activity of the thrombopoietin receptor in
CAMT II patients. © 2006 Wiley-Liss, Inc.
KEY WORDS: congenital amegakaryocytic thrombocytopenia, bone marrow failure; CAMT; MPL
Congenital amegakaryocytic thrombocytopenia (CAMT, MIM# 604498) is a rare bone marrow failure
syndrome presenting with absent or severely reduced megakaryocytes in the bone marrow, low platelet counts and
development of pancytopenia during the course of disease. Recently it became evident that mutations in the gene
MPL (MIM# 159530) coding for the thrombopoietin receptor cause CAMT. Up to now, data on MPL mutations
were published for 15 different CAMT patients (Ballmaier et al., 2001; Ihara et al., 1999; van den Oudenrijn 2000
and 2002; Gandhi et al., 2005). Beside these mutations in patients with CAMT which lead to a loss or restricted
function of the thrombopoietin receptor, Moliterno et al. (2004) described a functional gain-of-function
polymorphism in MPL associated with the clinical phenotype of thrombocytosis.
In a previous report (Ballmaier et al., 2001) we proposed to define two subgroups of CAMT patients based on
the clinical course: Group CAMT I with a more severe type of thrombocytopenia with constantly low platelet
counts and an early onset of pancytopenia and group CAMT II which is characterized by a transient increase of
platelet counts during the first year of life and a later or no development of pancytopenia.
We hypothesized that the type of MPL mutation determines the clinical course of CAMT: A total loss of the
TPO receptor due to homozygous nonsense mutations, deletions and frame shift mutations causes the more severe
2 Germeshausen et al.
course of disease found in CAMT patients of subgroup I. In contrast, we assumed that the transient increase of
platelet counts and the later development of pancytopenia in patients with homozygous or compound heterozygous
missense mutations are due to a residual function of the TPO-receptor. In this study we present the results of MPL
sequence analysis of 15 new CAMT patients and their relation to the clinical type of CAMT. Using in vitro
cultures of hematopoietic progenitors of patients of both groups we could demonstrate a residual TPO reactivity in
cells from patients of group CAMT II.
MATERIALS AND METHODS
Peripheral blood and/or bone marrow as well as clinical data from 23 patients suffering from CAMT were
collected after giving informed consent under a protocol approved by the local ethical committee of the Hannover
Medical School. Inclusion criteria were severe hypomegakaryocytic thrombocytopenia from birth on without
physical malformations as well as exclusion of Fanconi's anemia. Subdivision of patients into the categories
CAMT I and II resulted primarily from the course of platelet counts during the first year of life (CAMT I: platelet
counts always ≤50 nl-1, CAMT II: transient increase of platelet counts > 50 nl-1). If no sufficient data from the first
years of life were available, patients were classified according to the age of development of pancytopenia (age ≤ 3
years: CAMT I, age > 6 years: CAMT II). MPL sequence analysis was performed as previously described
(Ballmaier, et al., 2001). Reference sequences were GenBank accession numbers M90102.1 for cDNA and
U68159.1, U68160.1, U68161.1 U68162.1 for gDNA, respectively. For the detection of colony forming units from
hematopoietic progenitors we used a serum-free, collagen-based culture system as previously described for CFU-
Mk (Ballmaier et al., 2001). 103 CD34+ cells, obtained by flow cytometric sorting from bone marrow mononuclear
cells were cultured in the presence of the following growth factor combinations: 1) rhTPO (50 ng/ml), 2) SCF, IL-
6, IL-3 (10 ng/ml each), 3) SCF, IL-6, IL-3 (10 ng/ml each) rhTPO (50 ng/ml). Total colony numbers were
counted at day 12 of culture, CFU-Mk were counted after immunocytochemical staining with CD41 as already
described (Ballmaier et al., 2001).
Figure 1. MPL mutations in CAMT patients. Allocation of mutations from 30 CAMT patients described by us (upper row) and
others (lower row, see text). Each symbol represents a mutated allele in a CAMT patient. The exon structure of the MPL gene
as well as the derived protein with the functional domains are depicted below. Filled diamonds: nonsense mutations; open
diamonds: missense mutations; S: splice site mutations; filled triangles: frame shift deletions.
RESULTS AND DISCUSSION
By screening of 15 new CAMT patients for mutations in the 12 exons of the MPL gene, in the promoter and the
adjacent exon-intron boundaries we could verify our presumptions from the first series of 8 patients (Table 1): we
found homozygous or compound heterozygous missense mutations in 7 patients of group CAMT II with a transient
MPL Mutations in CAMT 3
increase in platelet counts over 50 nl-1 during the first years of life, 7 of these mutations are newly described here.
In contrast, we detected homozygous or compound heterozygous mutations predicted to lead to total c-Mpl
deficiency in 8 CAMT I patients with persistently low platelet counts and a fast progression into pancytopenia, 3 of
these mutations are newly described here. In all cases in which family studies were possible, we could demonstrate
that the mutations found in the patients with CAMT were inherited from their heterozygous parents. None of the
mutations could be detected in a control group of 50 healthy donors (100 alleles).
In the total group of 23 CAMT patients analyzed in our laboratory, we detected 9 different missense mutations
in ten patients of group CAMT II. Eight mutations were found in one single patient each, only mutation c.305G>C
was found in 5 patients from unrelated families. This mutation has also been detected heterozygously in 2 patients
by Oudenrijn et al. (2000 and 2002) and is the most frequent mutation reported for CAMT patients so far.
Table 1. MPL mutations in CAMT patients and correlation to clinical course
Patient ID Exon/
rise / max
- / 20
- / 38
- / 31
- / 40
- / 25
- / 42
- / 46
- / n.a.
- / 21
- / 28
- / 20
- / 49
- (HSCT: 9)
- (HSCT: 25)
I 4 - / n.a. 30
+ / 90
+ / 100
- (age at last exam.: 168)
+ / 163
+ / 110
+ / 110
+ / 100
+ / 154
- (HSCT: 87)
- (HSCT: 35)
- (age at last exam.: 36)
CAMT 67 II 10 n.a. 85
*: Patients previously described in our previous study; HSCT: hematopoietic stem cell transplantation.
Mutations numbered with +1 as A of ATG initiation codon. GenBank M90102.1
II 14 + / 170 - (age at last exam.: 8)
4 Germeshausen et al.
Figure 2. Highly conserved amino acids are affected from missense mutations. Amino acid alignment of human c-Mpl with
homologous of mouse, rat, cattle, dog, chicken, zebrafish and with the intracytoplasmatic portion of the myeloproliferative
leukemia virus MPLV (accession numbers: human: AAB08424.1, mouse: NP_034953.1, rat: XP_345573.2, cattle:
XP_604266.1, dog: XP_853442.1, chicken: AAT45555.1, zebrafish: AAQ82785.1, MPLV: P40931). The positions affected by
missense mutations are framed and displayed above the aligned sequences. Alignment was done using the BioEdit alignment
editor (Hall, 1999).
Seven different nonsense, frame-shift or splice-site mutations, all predicted to lead to a complete loss of
function of the thrombopoietin receptor were found in 13 patients, all could be assigned to group CAMT I. An
accumulation was observed for the nonsense mutation c.127C>T (homozygous in 5 patients from 4 unrelated
families) and the single-base deletion c.378delT (heterozygous in 2 and homozygous in 3 patients from 4 unrelated
families). There was no overlap with mutations described by other authors so far.
Mutations spread nearly over the entire gene with a clear accumulation in exons 2 and 3 (Fig. 1). Except for two
missense mutations in exon 8 and exon 12 coding for the second cytokine receptor homology domain and for the
distal half of the intracytoplasmic domain, respectively, all mutations were found in exons 2-5 coding for the first
cytokine receptor homology domain of the thrombopoietin receptor. The mutations found in group CAMT I
patients were located in exons 2, 3 and in the splice sites between exons 1-3. Missense mutations found in patients
of group CAMT II spread over exons 3, 4, 5, 8 and 12. Only highly conserved amino acids were affected from the
missense mutations described here (Fig. 2).
The predominant occurrence of mutations in the first exons was not so clear for the 7 patients described by
other authors (Ihara et al. 1999; van den Oudenrijn et al. 2000 and 2002, Gandhi et al. 2005) (Fig. 1): nonsense
mutations and deletions were also found in the regions coding for the second cytokine receptor homology domain
and the transmembrane domain of the receptor. The clinical course of the patient described by Ihara et al. (1999;
Muraoka et al., 1997 and 2005) also proves our rule of genotype-phenotype correlation in CAMT: the type of
mutations as well as the clinical course meets our criteria of group CAMT I. The clinical information about the
patients provided by Oudenrijn et al. (2000 and 2002) did not allow a correlation between the type of mutations
and the course of the disease in this patient group. Interestingly, Gandhi et al. (2005) recently described a family
with three siblings affected from a milder form of CAMT due to a splicing defect which results in diminished c-
Mpl expression. The clinical course of these patients seems to meet the criteria of group CAMT II.
We hypothesized, that the transient increase in platelet counts and the delayed development of pancytopenia
could be due to a residual activity of the thrombopoietin receptor. However, in our first report we did not observe
any growth of megakaryocytic colonies with TPO as a single growth factor, furthermore we were not able to detect
any synergism between TPO and ADP in the activation of CAMT patients' platelets (Ballmaier et al., 2001). We
MPL Mutations in CAMT 5
assumed that the assay systems used were not sensitive enough to detect a small residual activity of the receptor.
We therefore performed a series of colony assays with CD34+ cells from bone marrow samples of patients from
both groups (CAMT I: 8 samples from 8 patients, CAMT II: 7 samples from 4 patients) with a cytokine cocktail
containing SCF, IL-3 and IL-6 with and without TPO (Fig. 3). As previously described, the total number of colony
forming cells was significantly lower in samples from CAMT patients of both groups compared to healthy donors.
However, in samples from normal donors and patients of group CAMT II the addition of TPO led to a significant
higher number of total colonies (p < 0.016 using the non-parametric Wilcoxon Matched-Pairs Signed-Ranks Test).
In contrast, TPO did not show an effect on the total colony number in samples from patients of group CAMT I.
Figure 3. TPO reactivity of bone marrow progenitors.
CD34+ cells from CAMT patients of both groups were
cultured in semi-solid medium for the determination of
colony forming units. We did not observe any colony
growth using TPO as a single growth factor in patients
with CAMT. For a comparison of the colony stimulating
activity of two cytokine combinations with and without
TPO in different patients the relative numbers of total
colonies are plotted (100% = total colony number
obtained with growth factor combination IL-3, IL-6 and
SCF). Statistical significance was determined using the
non-parametric Wilcoxon matched-pairs signed-ranks
This is a first experimental verification of our hypothesis that the type of mutation has a strong influence on the
course of the disease and that the transient increase of platelet counts and the delayed development of pancytopenia
in CAMT II patients is due to a residual activity of the TPO receptor.
We are indebted to the patients and families who participated in this study and the referring physicians for their
cooperation. We would especially like to thank the following physicians for providing us with samples and clinical
data from their patients: Ninette N. Amariglio (Tel Aviv), Phil Ancliff (London), Chaim Churi (Tel Hashomer),
Yigal Dror (Toronto), Paul Fields (London), Rupert Handgretinger (Memphis), Reinhard Kolb (Oldenburg), Uwe
Kordes (Hamburg), Bernhard Kremens (Essen), Rolf Ljung (Malmø), Helen New (London), Arnulf Pekrun
(Göttingen), Manfred Rister (Koblenz), Jörg Ritter (Münster), Hans-Joachim Spaar (Bremen), Gabriele Strauß
(Berlin), Catherine Trichet (Paris), Regina Wieland (Essen), Jochen Wulff (Datteln).
Ballmaier M, Germeshausen M, Schulze H, Cherkaoui K, Lang S, Gaudig A, Krukemeier S, Eilers M, Strauss G, Welte K.
2001. c-mpl mutations are the cause of congenital amegakaryocytic thrombocytopenia. Blood 97:139-146.
Gandhi MJ, Pendergrass TW, Cummings CC, Ihara K, Blau CA, Drachman JG. 2005. Congenital amegakaryocytic
thrombocytopenia in three siblings: molecular analysis of atypical clinical presentation. Exp Hematol 33:1215-1221
Ihara K, Ishii E, Eguchi M, Takada H, Suminoe A, Good RA, Hara T. 1999. Identification of mutations in the c-mpl gene in
congenital amegakaryocytic thrombocytopenia. Proc Natl Acad Sci U S A 96:3132-3136.
Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT.
Nucl Acids Symp Ser 41:95-98
6 Germeshausen et al.
Moliterno AR, Williams DM, Gutierrez-Alamillo LI, Salvatori R, Ingersoll RG, Spivak JL. 2004. Mpl Baltimore: a
thrombopoietin receptor polymorphism associated with thrombocytosis. Proc Natl Acad Sci U S A 101:11444-11447
Muraoka K, Ishii E, Ihara K, Imayoshi M, Miyazaki S, Hara T, Hamasaki Y. 2005. Successful bone marrow transplantation in a
patient with c-mpl-mutated congenital amegakaryocytic thrombocytopenia from a carrier donor. Pediatr Transplant 9:101-
Muraoka K, Ishii E, Tsuji K, Yamamoto S, Yamaguchi H, Hara T, Koga H, Nakahata T, Miyazaki S. 1997. Defective response
to thrombopoietin and impaired expression of c-mpl mRNA of bone marrow cells in congenital amegakaryocytic
thrombocytopenia. Br J Haematol 96:287-292.
van den Oudenrijn S, Bruin M, Folman CC, Bussel J, de Haas M, von dem Borne AE. 2002. Three parameters, plasma
thrombopoietin levels, plasma glycocalicin levels and megakaryocyte culture, distinguish between different causes of
congenital thrombocytopenia. Br J Haematol 117:390-398.
van den Oudenrijn S, Bruin M, Folman CC, Peters M, Faulkner LB, de Haas M, von dem Borne AE. 2000. Mutations in the
thrombopoietin receptor, Mpl, in children with congenital amegakaryocytic thrombocytopenia. Br J Haematol 110:441-448.