A single amino acid change A91V in perforin: a novel, frequent predisposing factor to childhood acute lymphoblastic leukemia?

Abstract

We screened 100 children with acute lymphoblastic leukemia (ALL) to assess the incidence of single amino acid change A91V in perforin. Heterozygous A91V was found in 12/100 patients and 5/127 controls (OR, 3.4; 95%CI: 1.15-9.95; p=0.014). A91V is a novel and frequent predisposing factor for childhood ALL.

Full text

Page 1

Letters to the Editor
haematologica/the hematology journal | 2005; 90(5) | 697 |
Acute Lymphoblastic Leukemia
A single amino acid change A91V in perforin: a
novel, frequent predisposing factor to childhood
acute lymphoblastic leukemia?
We screened 100 children with acute lympho-
blastic leukemia (ALL) to assess the incidence of
single amino acid change A91V in perforin.
Heterozygous A91V was found in 12/100 patients
and 5/127 controls (OR, 3.4; 95%CI: 1.15-9.95;
p=0.014). A91V is a novel and frequent predispos-
ing factor for childhood ALL.
haematologica 2005; 90:697-698
Perforin plays a pivotal role in the cytotoxicity of natu-
ral killer (NK) and cytotoxic T lymphocytes (CTL).
Perforin gene (PRF1) mutations have been associated
with familial hemophagocytic lymphohistiocytosis (FHL,
HLH).1The pathogenic role of the DNA variant C272T,
resulting in a single amino acid change, A91V, has been
recently questioned by reports of 4 and 17% prevalences
among controls, suggesting this is only a neutral poly-
morphism.2,3To address this issue, we studied the preva-
lence of A91V in our control population.
Furthermore, a possible link between PRF1 mutations
and lymphomagenesis was suggested by reports of: lym-
phoma or HLH in two siblings with PRF1 mutations;4
autoimmune lymphoproliferative syndrome and lym-
phoma in a patient with heterozygous Fas and PRF1
mutations;5and 29 patients with lymphoma sharing clin-
ical characteristics with HLH harboring PRF1 mutations.6
To assess whether the impaired cytotoxic function
resulting from A91V PRF1 might be associated with
increased risk of developing childhood ALL, we screened
100 consecutive, unselected patients with ALL (<18
years). We also analyzed constitutional DNA from 127
Caucasian controls (61 healthy subjects and 66 consecu-
tive neonates). To search for the C272T DNA variant by
minisequencing, we amplified PFR1 exon 2. Polymerase
chain reaction products were used to perform a primer
single base extension reaction (SNaPshot Kit Applied
Biosystems). The reaction is based on a specific minise-
quencing primer, which is exactly one base short of the
272 position, and fluorescent ddNTP. The product
obtained is separated by electrophoresis and the color of
the peaks obtained makes it possible to identify the geno-
type (GeneScan Analysis Software, Applied Biosystems).
Heterozygous A91V was found in 2 of 61 healthy con-
trols (3.2%) and 3 of 66 (4.5%) neonates. Thus, overall,
heterozygous A91V was observed in 5 of 127 controls,
with a prevalence of 3.9%. Five 272T alleles were
observed in 254 chromosomes. Of the 100 children with
ALL, 11 were heterozygous, and one homozygous, for
A91V. This incidence was significantly superior to that in
controls (OR, 3.4; 95% CI: 1.15-9.95). Thirteen 272T
alleles were observed in 192 chromosomes (p=0.014)
(Table 1). The main presenting features of the 12 ALL
patients with A91V are described in Table 2.
Among 22 patients enrolled in the National HLH reg-
istry,7in whom a PRF1 mutation was identified, A91V
was present in 11 (50%), always in the heterozygous
state, associated with another mutation (Aricò M., unpub-
lished data). These data suggest that A91V is the most rel-
http://www.haematologica.org/journal/2005/5/697.html)
evant single PRF1 protein change in Italian patients with
HLH. The pathogenic role of A91V was questioned by
two recent reports: ZurStadt found that 15 of 86 (17.5%)
healthy Caucasians were heterozygous for A91V,3while
Molleran Lee found the amino acid change in 7 of 202
controls (3%).2Both suggested that A91V should be con-
sidered only a neutral polymorphism. Our finding of
A91V in 3.9% of controls is in keeping with the results of
Molleran Lee.2Is A91V only a neutral polymorphism? At
least two recent studies suggest that this is not the case.
Voskoboinik reported that A91V perforin was expressed
at lower levels in rat basophil leukemia cells than in wild-
type cells, resulting in partial loss of lytic capacity.8
Trambas et al. demonstrated that A91V perforin from a
patient with HLH is not recognized using an antibody
raised against native perforin (dG9), but is readily detect-
ed using an antibody raised against a peptide epitope
(2d4), suggesting that the epitope recognised by dG9 is
destroyed by A91V. A91V perforin undergoes conforma-
tional changes and impaired cleavage, likely explaining
the reduced cytotoxicity in CTL and NK cells contribut-
ing to the pathogenesis of HLH.9Altogether these data
suggest that A91V cannot be considered only as a neutral
polymorphism but rather as a functional polymorphism –
or a mutation – which affects perforin function.
Is HLH the only clinical condition in which A91V has
Table 1. Allele distribution for the 272T DNA variant in children
with leukemia and in control subjects.
CategoryNumber Heterozygous
of subjects
Number
of alleles
272T alleles
(%)subjects
(%)
Controls
Neonates
Healthy controls
127
66
61
5 (3.9)
3 (4.5)
2 (3.2)
254
132
122
5 (1.9)
3 (2.2)
2 (1.9)
Children with acute
lymphoblastic leukemia
10012* (12.0)°200 13 (6.5)#
*One patient homozygous; °Odds Ratio: 3.4; 95% CI: 1.15-9.95; #c2of Pearson
without Yates’ correction: 6,03 (p=0.014).
Table 2. Presenting features of 12 patients with childhood acute
lymphoblastic leukemia and the single amino acid change A91V.
PatientAge/SexWBC count
(mm3)
Immunophenotype
1.
2.
3.
4.
5.*
6.
7.
8.
9.
10.
11.
12.
5/F
9/F
4/M
9/M
2/M
3/F
0.7/M
12/M
4/F
2/F
14/M
11/F
3.200
3.430
43.890
3.390
10.980
6.660
243.000
9.310
82.000
4.360
151.000
4.600
Common
Common
T
Common
Common
Common
Pre-B°
B Mature
Common
Pre-B
Common
Pre-B
*This patient was homozygous for A91V. °t(9;11) at diagnosis; prednisone good
responder. At relapse immunophenotype shifted from Pre-B to Pro-B.
©Ferrata Storti Foundation

Page 2

implications? We found A91V in 12% of a cohort of
Caucasians children with ALL. This prevalence exceeded
that of 3.9% observed in our fully comparable control
population from the same geographic area. Since the
patients with ALL were not selected for any additional
criteria, including HLH-like clinical features, this finding
suggests that the single amino acid change A91V in the
perforin is significantly associated with the risk of devel-
oping childhood ALL. The presenting features of children
with A91V and ALL were not different from those of the
remaining ALL patients. In conclusion, we suggest that
impaired function of the cytotoxic machinery, as induced
by the A91V transition, may result in different clinical
manifestations and may predispose to ALL, the most
common type of childhood cancer.
Alessandra Santoro,* Sonia Cannella,° Antonino Trizzino,°
Luca Lo Nigro,#Giovanni Corsello,@Maurizio Aricò°
*Ematologia I, A.O. Cervello, Palermo; °Onco Ematologia
Pediatrica, Ospedale dei Bambini ”G. Di Cristina”, Palermo;
#Onco Ematologia Pediatrica, Policlinico di Catania;
@Clinica Pediatrica, Dipartimento Materno-Infantile,
©Ferrata Storti Foundation
University of Palermo, Italy
Funding: this research was partly supported by AIRC (MA, AS);
Associazione Ricerca Sindromi Emofagocitiche – ARSE (MA);
and Ministero della Salute, Progetto di Ricerca Finalizzata 2004
“Istiocitosi e Tumori” (MA, AS).
Key words: A91V, childhood acute lymphoblastic leukemia,
perrforin.
Correspondence: Maurizio Aricò, Onco Ematologia Pediatrica,
Ospedale dei Bambini “G. Di Cristina”, via Benedettini 1, 90134
Palermo, Italy. Phone: international +39.091.6666131. Fax: inter-
national +39.091.6666001. E-mail: arico@ospedalecivicopa.org
References
1. Stepp SE, Dufourcq-Lagelouse R, Le Deist F, Bhawan S,
Certain S, Mathew PA, et al. Perforin gene defects in familial
hemophagocytic lymphohistiocytosis. Science 1999;286:1957-
9
2. Molleran Lee S, Villanueva J, Sumegi J, Zhang K, Kogawa K,
Davis J, et al. Characterisation of diverse PRF1 mutations lead-
ing to decreased natural killer cell activity in North American
families with haemophagocytic lymphohistiocytosis. J Med
Genet 2004;41:137-44.
3. Zur Stadt U, Beutel K, Weber B, Kabisch H, Schneppenheim R,
Janka G, A91V is a polymorphism in the perforin gene not
causative of an FHLH phenotype. Blood 2004;104:1909-10.
4. Clementi R, Emmi L, Maccario R. Adult onset and atypical
presentation of hemophagocytic lymphohistiocytosis in sib-
lings carrying PRF1 mutations. Blood 2002;100:2266-7.
5. Clementi R, Dagna L, Dianzani U, Dupre L, Dianzani I,
Ponzoni M, et al. Inherited perforin and Fas mutations in a
patient with autoimmune lymphoproliferative syndrome and
lymphoma. N Engl J Med 2004;351:1419-24.
6. Clementi R, Locatelli F, Duprè L, Garaventa A, Emmi L, Bregni
M, et al. A proportion of patients with lymphoma may harbor
mutations of the perforin gene. Blood 2005;22:[Epub ahead of
print]
7. Aricò M, Janka G, Fischer A, Henter JI, Blanche S, Elinder G, et
al. Hemophagocytic lymphohistiocytosis. Report of 122 chil-
dren from the International Registry. FHL Study Group of the
Histiocyte Society. Leukemia 1996;10:197-203.
8. Voskoboinik I, Thia MC, Trapani JA. A functional analysis of
the putative polymorphisms A91V and N252S, and 22 mis-
sense perforin mutations associated with hemophagocytic
lymphohistiocytosis. Blood 2005;8:[Epub ahead of print].
9. Trambas C, Gallo F, Pende D, Marcenaro S, Moretta L, De
Fusco C, et al. A single amino acid change, A91V, leads to con-
formational changes which can impair processing to the active
form of perforin. Blood 2005;[Epub ahead of print].
n with and without granulocyte colony-stimulating factor
leukemia. Biol Blood Marrow Transplant 2002;8:261-7
Chronic Lymphoproliferative Disorders
Reversal of bone marrow angiogenesis in chronic
lymphocytic leukemia following fludarabine
therapy
We evaluated bone marrow (BM) angiogenesis in
12 responding patients with Binet stage B chronic
lymphocytic leukemia who received either chlo-
rambucil or fludarabine. Microvessel density
(MVD) was compared between pre-treatment mar-
row samples and those obtained after at least 4
cycles of chemotherapy. BM angiogenesis decreas-
es in all patients but one, although subset analysis
revealed the decrease of BM angiogenesis was sig-
nificant only in patients who received fludarabine.
Even if based on a small number of patients, these
results point out the potential in vivo anti-angio-
genic activity of fludarabine.
haematologica 2005; 90:698-700
Research over the past two decades has unravelled the
importance of abnormal, increased angiogenesis in the
development and spread of both solid and hematologic
tumors.1-6In the context of hematologic malignancies,
increased bone marrow (BM) angiogenesis has been
demonstrated in multiple myeloma, chronic myeloid
leukemia, acute myeloid or lymphocytic leukemia,
chronic lymphocytic leukemia (CLL) as well as myelodys-
plastic syndromes and has been found to have prognos-
tic value.1-6However, changes in BM angiogenesis after
therapy have not been systematically evaluated.
With this in mind we analyzed the effect of chemother-
apy on BM angiogenesis in responder B-cell CLL patients.
The study included 12 symptomatic Binet stage B
patients (median age 54 year; range, 28-67; M,7; F,5) who
received intermittent chlorambucil (5 patients) or fludara-
bine (7 patients) as up-front therapy. Paraffin-embedded
BM biopsy blocks were used to prepare slides for
microvessel density (MVD) determination according to
previously described methods.6BM evaluations were per-
formed at the time of diagnosis and after at least 4 cours-
es of chemotherapy (median, 8; range 4-12). According to
the criteria proposed by the National Cancer Institute
(NCI)77 patients were considered in complete remission
(CR) and 5 in good partial remission (G-PR). Minimal
residual disease (MRD), assessed in flow cytometry as
the percentage of CD19+/CD5+BM cells, closely reflected
the type of therapy: chlorambucil, 21.8% (range, 11-
29.9%); fludarabine, 5.2% (range, 0.1-11%; p=0.006;
Mann-Whitney test). BM MVD decreased in all but one
patient (Table 1). Specifically, the median microvessel
area was 2.616 mm2¥10-2(range, 0.545-4.126) before
therapy and 0.644 mm2¥10-2(range, 0.383-1.914) after
therapy (p=0.003; Mann-Whitney test). Finally, we won-
dered whether different therapies could affect the
amount of MVD reduction. Separate comparisons carried
out in patients treated with chlorambucil and fludarabine
revealed a significant decrease of MVD only in patients
who received the latter treatment (p=0.04). Interestingly,
no correlation was found between the amount of MVD
reduction related to the therapy and MRD as defined by
the percentage of BM CD5+/CD19+cells (Spearman r= -
0.042; p=0.907).
(http://www.haematologica.org/journal/2005/5/698.html)
Letters to the Editor
| 698 | haematologica/the hematology journal | 2005; 90(5)