Common deleted genes in the 5q? syndrome: thrombocytopenia and reduced erythroid
colony formation in SPARC null mice
S Lehmann1,2, J O’Kelly1, S Raynaud3, SE Funk4, EH Sage4and HP Koeffler1
1Division of Hematology/Oncology, Cedars-Sinai Medical Center, School of Medicine, University of California-Los Angeles,
Los Angeles, CA, USA;2Department of Hematology, Karolinska University Hospital, Stockholm, Sweden;3Departement
d’Hematologie Biologique, Hopital de l’Archet, Nice, France and4Hope Heart Program, The Benaroya Institute at
Virginia Mason, Seattle, WA, USA
The commonly deleted region (CDR) for the 5q? syndrome has
been identified as a 1.5-megabase interval on human chromo-
some 5q32. We studied, by real-time reverse-transcription (RT)–
PCR, the expression of 33 genes within the CDR that are known
to be expressed in CD34þ hematopoietic stem cells. Genes in
the 5q? samples that showed the most pronounced decrease in
expression compared to non-5q? samples were: solute carrier
family 36, member 1 (SLC36A1; 89% downregulated), Ras-
GTPase-activating protein SH3 domain-binding (G3BP; 79%),
antioxidant protein 1 (ATOX1; 76%), colony-stimulating factor-1
receptor precursor (CSF1R; 76%), ribosomal protein S14
(PDGFRB; 73%), Nef-associated factor 1 (TNIP1; 72%), secreted
protein, acidic and rich in cysteine (SPARC; 71%), annexin VI
(ANAX6; 69%), NSDT (66%) and TIGD (60%). We further studied
the hematopoietic system in SPARC-null mice. These mice
showed significantly lower platelet counts compared to wild-
type animals (P¼0.008). Although hemoglobin, hematocrit and
mean corpuscular volume (MCV) were lower in mice lacking
SPARC, differences were not statistically significant. SPARC-
null mice showed a significantly impaired ability to form
erythroid burst-forming units (BFU-E). However, no significant
differences were found in the formation of erythroid colony-
forming units (CFU-E), granulocyte/monocyte colony-forming
units (CFU-GM) or megakaryocyte colony-forming units (CFU-
Mk) in these animals. We conclude that many of the genes
within the CDR associated with the 5q? syndrome exhibit
significantly decreased expression and that SPARC, as a
potential tumor suppressor gene, may play a role in the
pathogenesis of this disease.
Leukemia (2007) 21, 1931–1936; doi:10.1038/sj.leu.2404852;
published online 12 July 2007
Keywords: myelodysplastic syndrome; 5q? syndrome; SPARC
Myelodysplastic syndromes (MDS) are characterized by periph-
eral blood cytopenias affecting one or more hematopoietic
lineages and a predisposition to develop acute myeloid
leukemia (AML).1Symptoms in MDS are caused by insufficient
numbers of circulating erythrocytes, leukocytes and/or platelets.
Approximately 50% of MDS patients have cytogenetic abnorm-
alities in their malignant clone.2,3Deletion of 5q is the most
common cytogenetic abnormality in MDS, with 10–15% of the
patients exhibiting this deletion.3,4It may occur as a sole
abnormality or in combination with other cytogenetic abnorm-
alities. The 5q? syndrome, a subgroup of MDS, is usually
characterized by an elderly female who has refractory anemia,
high or normal platelet numbers in the peripheral blood, bone
marrow with erythroid dysplasia, hypolobulated megakaryo-
cytes, and a low risk to transform to AML.5,6The 5q? syndrome
is a hematopoietic stem cell disease, typically affecting 490%
of the CD34(þ)CD38(?) hematopoietic stem cells in the bone
marrow of these patients.7
The commonly deleted region (CDR) in the 5q? syndrome
has been narrowed to a 1.5-megabase interval at 5q32 flanked
by DNA segment number D5S413 and the GLRA1 gene.8This
region is distinctly different and more telomeric compared to the
CDR in cases of MDS and AML with loss of 5q and often with
other karyotypic changes.9,10As the 5q deletion is the sole
karyotypic abnormality in the 5q? syndrome, it can be assumed
that the genes within the region play a part in the pathogenesis
of the disease. However, no causative gene has been found
although several within the region have been proposed to play a
role, for example, Ras-GTPase-activating protein SH3 domain-
binding (G3BP), interleukin 17B precursor (IL17B), platelet-
derived growth factor receptor-b (PDGFRB), colony-stimulating
factor-1 receptor precursor (CSF1R) and secreted protein, acidic
and rich in cysteine (SPARC).
SPARC is a matricellular protein with complex roles in various
biological processes. It helps to regulate the composition of
extracellular matrix and binds to many of its resident proteins.11
It can also affect cellular proliferation, differentiation, migration
and cell adhesion.12SPARC-null mice exhibit early cataract
formation,13increased deposition of subcutaneous and abdom-
inal fat14and accelerated wound healing.12Interestingly, bone
marrow-derived SPARC can inhibit the binding of thrombos-
pondin to the surface of platelets, and, thereby, inhibit platelet
aggregation.15,16Recently, SPARC was shown to inhibit breast
cancer cell invasion, platelet aggregation and metastasis.17
Increased methylation of SPARC in lung and pancreatic tumors
is consistent with its role as a potential tumor suppressor gene
that is silenced in selected cancers.18,19However, the role of
SPARC in hematopoiesis remains unknown.
In this study, we measured the expression of genes within the
CDR of the 5q? syndrome by real-time reverse-transcription
(RT)–PCR. The genes with the most prominent decrease in
expression were solute carrier family 36, member 1 (SLC36A1),
G3BP, antioxidant protein 1(ATOX1), CSF1R, ribosomal protein
S14 (RPS14), PDGFRB, Nef-associated factor 1 (TNIP1), annexin
VI (ANAX6) and SPARC. We studied the role of SPARC in
hematopoiesis and found a significant decrease in circulating
platelets in SPARC-null mice compared to their wild-type
counterparts. The capacity of bone marrow cells to form
Received 28 January 2007; revised 28 April 2007; accepted 29 May
2007; published online 12 July 2007
Correspondence: Dr S Lehmann, Department of Hematology M54,
Karolinska University Hospital, Huddinge, Stockholm 141 86,
Leukemia (2007) 21, 1931–1936
& 2007 Nature Publishing Group All rights reserved 0887-6924/07 $30.00
erythroid burst-forming units (BFU-E) was equally impaired;
however, no difference was noted between wild-type and
SPARC-null mice in the ability to form erythroid colony-forming
units (CFU-E), granulocyte/monocyte colony-forming units
(CFU-GM) or megakaryocyte colony-forming units (CFU-Mk).
We conclude that SPARC affects hematopoiesis, although it
cannot account for the entire clinical phenotype of the 5q?
syndrome. The 5q? syndrome is likely to be a result of the
insufficiency of several genes within the deleted region, one of
which is SPARC.
Materials and methods
Cell lines and patient cells
The KG-1 cell line has previously been described;20the TP-1
cell line was a gift from Dr K Kitamura (Tokyo, Japan). The NB-4
cell line was from Dr Lanotte (Paris, France), HEL cells were
from Dr Papayannopoulou (Seattle, WA, USA) and the HL-60
cells were provided by Dr Collins (Seattle, WA, USA). The U937
cell line was purchased from American Type Culture Collection
(Manassas, VA, USA).
Mononuclear cells from bone marrow of patients were
purified by Ficoll gradient. MDS patients were diagnosed
according to the FAB and WHO classifications. Twelve with
the 5q? syndrome were studied; each had 5q deletion as a sole
chromosomal abnormalities and characteristic clinical and
morphologic features. All but one patient were women. Of the
14 non-5q? MDS patients, 10 were men and 4 were women.
The protocol was approved by the Local Ethical Committee
(CPPRB) of the Centre Hospitalier Universitaire de Nice.
C57BL/6?129SVJ SPARC-null mice were generated as de-
scribed.21SPARC-null and wild-type mice were housed in a
modified, pathogen-free facility. The mice were euthanized at
either 6 to 8 weeks or at 12 months of age by anesthetized
cervical dislocation (isofluorane inhalation) or by CO2overdose,
followed by cervical dislocation. The femurs were dissected and
kept on ice before they were flushed, and the cells were
resuspended in MEM (Invitrogen, Carlsbad, CA, USA), contain-
ing 2% fetal calf serum (FCS). Mononuclear cells were separated
on Ficoll (Cellgro, Herndon, VA, USA). For peripheral blood
counts, blood was aspirated from the heart at the time of
euthanization. Samples were analyzed according to standard
procedures by Phoenix Central Laboratory (Everett, WA, USA).
Total RNA was extracted by the use of QIAprep Spin Miniprep
Kit (QIAGEN, Valencia, CA, USA) according to the manufac-
turer’s instructions. For cDNA synthesis, RNA was reverse-
transcribed using Superscript II RT as described by the
manufacturer (Invitrogen). Expression levels were measured by
real-time RT–PCR. Levels of b2-microglobulin and b-actin were
internal reference standards. Reactions were performed with
Platinum Taq DNA polymerase (Invitrogen) and SYBRGreen I
(Molecular Probes, Eugene, OR, USA). Reactions were per-
formed in triplicate with an iCycler iQ system (Biorad, Hercules,
CA, USA). For each sample, levels of the target gene and
reference gene were determined from standard curves.
Mononuclear bone marrow cells from wild-type and SPARC-
null mice were resuspended in methylcellulose medium M3234
for CFU-E (2?105cells/ml), and in M3434 for CFU-GM and
BFU-E (2?104cells/ml; StemCell Technologies Inc., Vancou-
ver, Canada). EPO (Ortho Biotech, Bridgewater, NJ, USA) was
added to the CFU-E and BFU-E colonies at a concentration of
3U/ml. Cells were plated in six-well plates in a volume of 1ml
and were incubated at 371C in a humidified atmosphere
containing 5% CO2. CFU-E were counted after 3 days, and
CFU-GM and BFU-E were counted after 8 days, with an inverted
microscope. Colony type was established by morphology. To
ensure accuracy, we plucked representative colonies from the
plates, centrifuged them onto slides, stained the cells with
Wright–Giemsa stain, and, subsequently, examined them by
light microscopy. BFU-E colonies were stained with benzidine.
For the CFU-Mk assay, bone marrow cells (1?105) were
cultured with MegaCult-C media (StemCell Technologies) in the
presence of 10ng/ml recombinant murine IL-3 (Calbiochem,
San Diego, CA, USA), 20ng/ml recombinant human IL-6
(Calbiochem) and 50ng/ml recombinant human TPO (Calbio-
chem). After culture for 7 days, colonies were stained with
acetylcholinesterase (AchE). CFU-Mk colonies were defined as
at least three megakaryocytes in a cluster.
Evaluation of mononuclear bone marrow cells from wild-type
and SPARC-null mice was performed by standard flow
cytometry on a FACScalibur (Becton Dickinson, Franklin Lakes,
NJ, USA) instrument with identical parameters between
samples, using appropriate fluorochrome-conjugated monoclo-
nal antibodies against cell surface markers to TER119, CD43,
CD4, CD8, CD41, and an appropriate isotype control (Calbio-
chem, La Jolla, CA, USA). Flow cytometry data were analyzed
with FlowJo software (Tree Star, Ashland, OR, USA).
Expression of genes within the CDR
The mRNA levels of the 33 genes within the CDR that are
known to be expressed in CD34-positive hematopoietic stem
Average relative expression (%)
(CDR). The graph shows the most downregulated genes within the
CDR. Mononuclear bone marrow cells from 12 patients with the 5q?
syndrome and 16 myelodysplastic syndrome (MDS) patients with
normal karyotype were analyzed with real-time reverse-transcription
(RT)–PCR. The average expression in the MDS samples with normal
karyotype was set as 100%, expression of all samples was normalized
Expression of genes within the commonly deleted region
Common deleted genes in the 5q? syndrome
S Lehmann et al
Cell lines5q- non-5q-
Cell lines5q- non-5q-
Cell lines5q- non-5q-
antioxidant protein 1 (ATOX1) and ribosomal protein S14 (RPS14) in cell lines and patient cells. Seven human myeloid leukemia cell lines (NB-4
(no. 1), K562 (no. 2), TP-1 (no. 3), KG-1 (no. 4), HEL (no. 5), U937 (no. 6) and HL-60 (no. 7) from left to right) and bone marrow mononuclear cells
from patients with the 5q? syndrome, as well as, from myelodysplastic syndrome (MDS) patients with normal karyotype, were analyzed by real-
time reverse-transcription (RT)–PCR. b2-microglobulin was used as an internal control for all samples, and the average expression in MDS samples
with normal karyotype was set as 1.0. The panels show the mRNA expression of SPARC (a), G3BP (b), ATOX1 (c) and RPS14 (d).
Expression of secreted protein, acidic and rich in cysteine (SPARC), Ras-GTPase-activating protein SH3 domain-binding (G3BP),
wt 1 wt 2wt 3 wt 4null 1 null 2 null 3 null 4
wt 1wt 2wt 3 wt 4null 1 null 2 null 3 null 4
BFU-E (mean +/- SD)
wt 1wt 2wt 3wt 4null 1 null 2 null 3 null 4
wt 1wt 2 wt 3wt 4null 1 null 2 null 3 null 4
CFU-E (mean +/ SD)
CFU-GM (mean +/- SD)
CFU-Mk (mean +/- SD)
erythroid burst-forming units (BFU-E) (a), erythroid colony-forming units (CFU-E) (b), granulocyte/monocyte colony-forming units (CFU-GM) (c)
and megakaryocyte colony-forming units (CFU-Mk) (d) from wild-type (wt) and SPARC-null mice. Bone marrow mononuclear cells were grown in
methylcellulose medium or MegaCult-C (CFU-Mk) with cytokines as described in Materials and methods, number of colonies per well was
determined at the end of the assay. Each panel shows one representative experiment. The results represent the mean7s.d. from four age- and sex-
matched mice, performed in triplicate. Four independent experiments were performed with similar results (data not shown). The BFU-E assay
showed a statistically significant difference between wild-type and SPARC-null mice (P¼0.03).
Clonogenic growth or marrow cells from wild-type versus secreted protein, acidic and rich in cysteine (SPARC)-null mice. Numbers of
Common deleted genes in the 5q? syndrome
S Lehmann et al
cells8were examined in mononuclear cells from 12 patients
with the 5q? syndrome. The expression levels were compared
to those of the mononuclear bone marrow cells from patients
with MDS, but having a normal karyotype. Figure 1 shows the
genes with the most pronounced downregulation. The solute
carrier SLC36A1 had the lowest average expression, with an
89% decrease compared to non-5q? MDS samples. Other
strongly downregulated genes included G3BP (79% decreased),
CSFR1 (76% decreased), ATOX1 (76% decreased), RPS14 (75%
decreased), PDGFRB (73% decreased), TNIP (72% decreased)
and SPARC (71% decreased). From these, as well as, from
examination of expression of the RNAs in different tissues,
including variation in hematopoietic cells, as shown in the
SymAtlas (http://symatlas.gnf.org), and the known functions of
the genes in the context of the pathogenesis of MDS, we
selected several genes for further studies.
SPARC is a matricellular protein with diverse functions.11It
regulates extracellular matrix formation, inhibits cell prolifera-
tion, has de-adhesive properties, and is methylated in lung and
pancreatic cancer cells, consistent with its role as a tumor
suppressor gene.12,18,19Except for an anti-adhesive effect on
platelets, its effects on hematopoiesis are largely unknown.
Figure 2a shows the expression of SPARC in each sample from
12 5q? patients and 14 MDS patients with normal karyotype,
together with its levels in seven leukemic cell lines. The mean
expression ratio of SPARC was significantly lower in the 5q?
samples compared to the non-5q? samples (Po0.05). The
expression levels in the cell lines were lower than in both of the
MDS patient groups. HEL, K562 and TP-1 cells showed higher
levels of SPARC mRNA than in the NB-4 and KG-1 cells,
whereas levels in U937 and HL-60 cells were only slightly
above the level of detection.
G3BP has diverse effects on cellular processes, including its
role as a downstream effector of Ras signaling.22G3BP was
strongly downregulated in the 5q? samples compared to the
control samples (Po0.005). In contrast to SPARC, the expression
of G3BP in the myeloid cell lines was higher compared to both
MDS patient groups (Figure 2b). K562 and HL-60 cells exhibited
especially high expression of G3BP.
ATOX1 is an antioxidant protein and a copper chaperone that
transports copper intracellularly and detoxifies the cells from
excessive amounts of this metal.23Even though the mean
ATOX1 expression in the 5q? samples was just 24% of that of
the MDS samples with normal karyotype, the levels were very
low compared to the myeloid leukemic cell lines; the cell lines
had almost 100-fold increased levels (Figure 2c). HEL and HL-60
cells showed the highest expression. Finally, RPS14, a ribosomal
protein and a component of the S40 ribosomal subunit, had a
mean expression level that was 25% of that of the non-5q?
samples (Po0.05; Figure 2d). mRNA levels in the myeloid cell
lines were slightly higher compared to those in the MDS
Hematopoietic clonogenic growth in SPARC-null mice
To study the importance of SPARC in the hematopoietic system,
we performed colony-forming assays on bone marrow cells from
wild-type and SPARC-null mice. Figure 3 shows the results
(mean7s.d.) of one representative experiment from four wild-
type and four SPARC-null mice. In the case of BFU-E, a
significant difference occurred, with the bone marrow from the
null mice forming fewer colonies (P¼0.03; Figure 3a). No
significant difference in CFU-E formation occurred between
SPARC-null and wild-type mice, although a tendency to form
fewer colonies was observed in SPARC-null mice (Figure 3b).
No difference was found in the formation of CFU-GM or CFU-
Mk, although the latter tended to be lower in the SPARC-null
mice (Figures 3c and d). These results indicate a role for SPARC
in erythroid development.
Blood counts, flow cytometry and bone marrow
morphology in SPARC-null mice
Peripheral blood from 1-year-old wild-type and SPARC-null
mice was analyzed (Table 1). Hemoglobin, hematocrit, mean
corpuscular volume (MCV) and red blood cell (RBC) count were
slightly lower in the null mice, but the differences were not
statistically significant. The only statistically significant differ-
ence between wild-type and SPARC-null mice was their platelet
counts. The mice-lacking SPARC had an average platelet count
of 448K/ml, ranging from 391 to 537K/ml, whereas wild-type
animals had an average of 600K/ml, ranging from 553 to 627K/
ml (P¼0.008). The white blood cell (WBC) count tended to be
higher in SPARC-null mice.
Bone marrow mononuclear cells from wild-type and SPARC-
null mice were analyzed by flow cytometry for hematopoietic
subpopulations, including erythroid marker (TER119), stem cell
marker (CD34), T-cell markers (CD8 and CD4), as well as, the
platelet and megakaryocyte marker (CD41; Table 2). None of
the markers showed a statistically significant difference, but the
difference in the erythroid population (TER119-positive) was
Table 1Peripheral blood counts
Platelet count (L/ml)
4. 0 (1.2)
Abbreviations: ND, not determined; NS, not significant; SPARC,
secreted protein, acidic and rich in cysteine.
Mean values for hemoglobin, hematocrit, mean corpuscular volume
(MCV), red blood cell (RBC) counts, white blood cell (WBC) counts and
platelet counts in wild-type and SPARC-null mice (four sex- and age-
matched mice in each group).
Table 2Subtypes of bone marrow mononuclear cells
Wild type (7s.d.) SPARC null (7s.d.)
Abbreviations: NS, not significant; SPARC, secreted protein, acidic
and rich in cysteine.
Percentage of positive mononuclear bone marrow cells from wild-type
and SPARC-null mice. The values are mean7s.d. of four wild-type and
four SPARC-null mice (12 months old, sex-matched). TER119 is an
erythroid marker; CD34 a hematopoietic stem cell marker; CD4 and
CD8 are T-cell markers.
Common deleted genes in the 5q? syndrome
S Lehmann et al
close to significant (P¼0.07); nevertheless, the difference was
modest, with 11.1% (71.3%) erythroid progenitor cells in wild-
type mice and 9.1 % (71.0%) in SPARC-null mice.
Cytospin slides were also made from the bone marrow of the
mice, and morphology was examined after staining with May–
Giemsa reagent. No differences were noted, neither in the
percentage of megakaryocytes, erythroid or myeloid cells, nor in
their morphologic feature. Eosin-hematoxylin staining was also
performed on sections of the femurs; no differences in cellularity
or morphology were observed (data not shown).
The CDR in the 5q? syndrome has been defined as a 1.5-Mb
region on human 5q32.8It spans approximately 40 genes, of
which 33 are expressed in CD34-positive hematopoietic stem
cells. This study was initiated by measuring the expression of
these 33 genes in samples from patients with the 5q? syndrome,
in comparison to the expression in samples from patients with
MDS with normal karyotype. Genes such as SLC36A1, G3BP,
ATOX1, CSF1R, RPS14, PDGFRB, TNIP1, SPARC and ANAX6
showed the most pronounced decrease in expression, all
exhibiting more than 50% decrease. Several of these genes
may be of interest for the development of MDS, such as, G3BP,
a downstream effector of the Ras-signaling pathway;22ATOX1,
an anti-oxidant copper chaperone;23CSFR1 (also known as
either fms or M-CSF receptor), a tyrosine kinase receptor;24and
RPS14, a component of the 40S ribosomal subunit. Interestingly,
RPS19, also a component of the S40 subunit, is often mutated in
Diamond–Blackfan anemia and when its expression is inhibited
in mice, this causes impaired erythropoiesis.25
SPARC is a secreted matricellular protein with diverse
Several identified functions of SPARC could
influence neoplastic disease, such as its effects on proliferation,
differentiation, cell migration, invasiveness of tumors and
A recent study showed that SPARC inhibits
metastasis, as well as, platelet aggregation and tumor-induced
thrombocytopenia in a breast cancer model.17
Accumulated data in other cancers suggested that SPARC
might be a tumor suppressor gene. Exogenous SPARC inhibited
proliferation and induced apoptosis of ovarian cancer cells.26
Other investigators showed that forced expression of SPARC
inhibited growth of human ovarian cancer cells in immunodefi-
cient mice and in breast cancer cells in vitro.27SPARC-null mice
exhibited increased proliferation of mesangial cells,28as well as,
increased insulin growth factor signaling.29The protein also
suppressed the growth of pancreatic cancer cells and glioma
cells.30Further studies showed that the SPARC gene was
methylated in both pancreatic18and lung cancers.19Taken
together, SPARC may be a tumor suppressor gene, and blunting
of its expression in the 5q? syndrome may give the cells a
The expression of SPARC, ATOX1, G3BP and RPS14 was also
studied in leukemic cell lines. ATOX1 had the highest
expression, followed by G3BP and RPS14. SPARC was the only
gene that exhibited lower expression in the leukemic cell line
compared to the MDS samples. These results confirm that G3BP
is associated with proliferation. However, the 100-fold higher
expression of ATOX1 in these AML cell lines is surprising, but
little is known about the role of ATOX1 in tumors. ATOX1 might
be linked to proliferation. In contrast, the low level of SPARC
expression in the leukemic cell lines strengthens the hypothesis
that its expression might be associated with control of growth of
hematopoietic progenitor cells. The AML cell line data could
lead to the speculation that the low levels of SPARC mRNA that
we noted in the bone marrow of 5q? syndrome represent cells
at a different stage of differentiation compared to the bone
marrow of the non-5q? MDS patients or normal individuals.
However, the array-based expression data online (http://
www.ncbi.nlm.nih.gov/geo) showed no major differences in
SPARC expression between CD34þ, CD33þ, CD3þ and
CD19þ cell subpopulations from normal marrow, suggesting
that, in normal hematopoiesis, SPARC does not greatly vary.
We also studied the hematologic phenotype and function of
SPARC in SPARC-null compared to wild-type mice. Interest-
ingly, the number of BFU-E formed by bone marrow cells from
the null mice was significantly lower compared to that of wild-
type mice. However, the capacity to form CFU-E was not
significantly impaired, and the CFU-GM or CFU-Mk numbers
were not different between wild-type and SPARC-null mice.
CFU-E represent late erythroid progenitors, and BFU-E are the
primitive erythroid cells more closely related to hematopoietic
stem cells.31The difference in CFU-E and BFU-E formation in
the SPARC-null mice may reflect a defect that affects very early
erythroid differentiation, and deserves further study. Differences
in the levels of BFU-E versus CFU-E in other mutated mice have
previously been noted.32
Peripheral blood values from SPARC-null versus wild-type
mice showed significantly lower levels of platelets. A possible
explanation relates to the known loss of the anti-aggregation
effect of SPARC on platelets;15–17this could result in their
clumping and rapid clearance from the circulation. However,
no increased incidence of thrombosis has been reported in
mice-lacking SPARC. Hemoglobin, hematocrit, RBC and MCV
exhibited somewhat lower values in the SPARC-null mice.
However, none of the differences were statistically significant.
Although the null mice had lower levels of platelets, no
differences were noted in either bone marrow cellularity or
morphology between SPARC-null and wild-type mice, including
number and appearance of the megakaryocytes, in either
cytospin preparations or stained sections of femurs. This result
is not surprising, as morphology is a relatively insensitive
assessment of the function of the hematopoietic system, given
the comparatively slight impact on colony-forming ability and
platelet counts in the SPARC-null mice. Flow cytometry for
TER119, a murine erythroid marker, tended to be lower in the
bone marrows from SPARC-null mice, with a borderline P-
value. This finding might also be an indication of the moderate
impact that SPARC has on hematopoiesis. Clearly, the hemato-
logic picture in the SPARC-null mice is not that of a full 5q?
MDS presentation. Still, the hematologic abnormalities in these
animals indicate that SPARC could play a role in MDS.
Our results show that a number of genes within the CDR in
patients with the 5q? syndrome are significantly downregu-
lated. For the first time, we show that SPARC affects
hematopoiesis, shown by the lower platelet counts in SPARC-
null mice, as well as an impaired ability to form BFU-E.
However, SPARC-null mice did not present with clinically overt
anemia. This observation supports the concept that the 5q?
syndrome could be an accumulative lack of normal expression
of several genes within the CDR, including SPARC.
1 Catenacci DV, Schiller GJ. Myelodysplasic syndromes: a compre-
hensive review. Blood Rev 2005; 19: 301–319.
2 Le Beau MM, Albain KS, Larson RA, Vardiman JW, Davis EM,
Blough RR et al. Clinical and cytogenetic correlations in 63
patients with therapy-related myelodysplastic syndromes and
Common deleted genes in the 5q? syndrome
S Lehmann et al
acute nonlymphocytic leukemia: further evidence for character- Download full-text
istic abnormalities of chromosomes no. 5 and 7. J Clin Oncol
1986; 4: 325–345.
3 Sole ´ F, Espinet B, Sanz GF, Cervera J, Calasanz MJ, Lun ˜o E et al.
Incidence, characterization and prognostic significance of chromo-
somal abnormalities in 640 patients with primary myelodysplastic
syndromes. Grupo Cooperativo Espanol de Citogenetica Hemato-
logica. Br J Haematol 2000; 108: 346–356.
4 Heim S, Mitelman F. Chromosome abnormalities in the myelo-
dysplastic syndromes. Clin Haematol 1986; 4: 1003–1021.
5 Van den Berghe H, Cassiman JJ, David G, Fryns JP, Michaux JL,
Sokal G. Distinct haematological disorder with deletion of long
arm of no. 5 chromosome. Nature 1974; 251: 437–438.
6 Boultwood J, Lewis S, Wainscoat JS. The 5q? syndrome. Blood
1994; 84: 3253–3260.
7 Nilsson L, Astrand-Grundstrom I, Anderson K, Arvidsson I,
Hokland P, Bryder D et al. Involvement and functional impairment
of the CD34(+)CD38(?)Thy-1(+) hematopoietic stem cell pool in
myelodysplastic syndromes with trisomy 8. Blood 2002; 100:
8 Boultwood J, Fidler C, Strickson AJ, Watkins F, Gama S, Kearney L
et al. Narrowing and genomic annotation of the commonly deleted
region of the 5q?syndrome. Blood 2002; 99: 4638–4641.
9 Horrigan SK, Westbrook CA, Kim AH, Banerjee M, Stock W,
Larson RA. Polymerase chain reaction-based diagnosis of del
(5q) in acute myeloid leukemia and myelodysplastic syn-
drome identifies a minimal deletion interval. Blood 1996; 88:
10 Zhao N, Stoffel A, Wang PW, Eisenbart JD, Espinosa R, Larson RA
et al. Molecular delineation of the smallest commonly deleted
region of chromosome 5 in malignant myeloid diseases to
1–1.5Mb and preparation of a PAC-based physical map. Proc
Natl Acad Sci USA 1997; 94: 6948–6953.
11 Bradshaw AD, Sage EH. SPARC, a matricellular protein that
functions in cellular differentiation and tissue response to injury.
J Clin Invest 2001; 107: 1049–1054.
12 Framson PE, Sage EH. SPARC and tumor growth: where the seed
meets the soil? J Cell Biochem 2004; 92: 679–690.
13 Yan Q, Clark JI, Wight TN, Sage EH. Alterations in the lens capsule
contribute to cataractogenesis in SPARC-null mice. J Cell Sci 2002;
14 Bradshaw AD, Graves DC, Motamed K, Sage EH. SPARC-null mice
exhibit increased adiposity without significant differences in
overall body weight. Proc Natl Acad Sci USA 2003; 100:
15 Clezardin P, Malaval L, Ehrensperger AS, Delmas PD, Dechavanne
M, McGregor JL. Complex formation of human thrombospondin
with osteonectin. Eur J Biochem 1988; 175: 275–284.
16 Clezardin P, Malaval L, Ehrensperger AS, Delmas PD, Dechavanne
M, McGregor JL. Osteonectin is an a-granule component involved
with thrombospondin in platelet aggregation. J Bone Miner Res
1991; 6: 1059–1070.
17 Koblinski JE, Kaplan-Singer BR, VanOsdol SJ, Wu M, Engbring JA,
Wang S et al. Endogenous osteonectin/SPARC/BM-40 expression
inhibits MDA-MB-231 breast cancer cell metastasis. Cancer Res
2005; 65: 7370–7377.
18 Suzuki M, Hao C, Takahashi T, Shigematsu H, Shivapurkar N,
Sathyanarayana UG et al. Aberrant methylation of SPARC in
human lung cancers. Br J Cancer 2005; 92: 942–948.
19 Sato N, Fukushima N, Maehara N, Matsubayashi H, Koopmann J,
Su GH et al. SPARC/osteonectin is a frequent target for
aberrant methylation in pancreatic adenocarcinoma and a
mediator of tumor–stromal interactions. Oncogene 2003; 22:
20 Koeffler HP, Golde DW. Human myeloid leukemia cell lines: a
review. Blood 1980; 56: 344–350.
21 Norose K, Clark JI, Syed NA, Basu A, Heber-Katz E, Sage EH et al.
SPARC deficiency leads to early onset cataractogenesis. Invest
Ophthalmol Vis Sci 1998; 39: 2674–2680.
22 Irvine K, Stirling R, Hume D, Kennedy D. Rasputin, more
promiscuous than ever: a review of G3BP. Int J Dev Biol 2004;
23 Harris ED. Basic and clinical aspects of copper. Crit Rev Clin Lab
Sci 2003; 40: 547–586.
24 Roussel MF, Sherr CJ. Oncogenic potential of the c-FMS proto-
oncogene (CSF-1 receptor). Cell Cycle 2003; 2: 5–6.
25 Ebert BL, Lee MM, Pretz JL, Subramanian A, Mak R, Golub TR
in Diamond–Blackfan anemia recapitulates defective hematopoi-
esis and rescue by dexamethasone: identification of dexa-
methasone-responsive genes by microarray. Blood 2005; 105:
26 Yiu GK, Chan WY, Ng SW, Chan PS, Cheung KK, Berkowitz RS
et al. SPARC (secreted protein acidic and rich in cysteine)
induces apoptosis in ovarian cancer cells. Am J Pathol 2001;
27 Dhanesuan N, Sharp JA, Blick T, Price JT, Thompson EW.
Doxycycline-inducible expression of SPARC/Osteonectin/BM40
in MDA-MB-231 human breast cancer cells results in growth
inhibition. Breast Cancer Res Treat 2002; 75: 73–85.
28 Bradshaw AD, Francki A, Motamed K, Howe C, Sage EH. Primary
mesenchymal cells isolated from SPARC-null mice exhibit altered
morphology and rates of proliferation. Mol Biol Cell 1999; 10:
29 Francki A, Motamed K, McClure TD, Kaya M, Murri C, Blake DJ
et al. SPARC regulates cell cycle progression in mesangial cells via
its inhibition of IGF-dependent signaling. J Cell Biochem 2003; 88:
30 Schultz C, Lemke N, Ge S, Golembieski WA, Rempel SA.
Secreted protein acidic and rich in cysteine promotes glioma
invasion and delays tumor growth in vivo. Cancer Res 2002; 62:
31 Gregory CJ, Eaves AC. Three stages of erythropoietic progenitor
cell differentiation distinguished by a number of physical and
biologic properties. Blood 1978; 51: 527–537.
32 Chui DH, Liao SK, Walker K. Fetal erythropoiesis in steel mutant
mice. III. Defect in differentiation from BFU-E to CFU-E during
early development. Blood 1978; 51: 539–547.
Common deleted genes in the 5q? syndrome
S Lehmann et al