From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
High-Level Reconstitution of Respiratory Burst Activity in a Human XLinked Chronic Granulomatous Disease (X-CGD) Cell Line and Correction
of Murine X-CGD Bone Marrow Cells by Retroviral-Mediated Gene
Transfer of Human gp91ph""
By Chunjin Ding, Akihiro Kume, Helga Bjorgvinsdottir, Robert G. Hawley, Nancy Pech, and Mary C. Dinauer
The X-linked form ofchronic granulomatous disease (XCGD) results from mutations in thegene encoding gp9lPhox,
a 91-kD membrane glycoprotein that is the larger subunit of
the respiratory burst oxidase cytochrome b. In this study, a
new retroviral vectorforexpressionof
human gp91PhoN,
MSCV-h91Neo. based on murine stem cell virus vectors,was
evaluated using a human X-CGD myeloid cell line (X-CGD
PLB-985 cells) and murine bone marrow cells. Expression of
recombinant gp9IPho"in transduced X-CGD PLB-985cells
was substantially improved compared with levels achieved
previously using a different retroviral construct, and respiratory burst oxidase activity was fully reconstituted in themajority of clones analyzed. Expression of gp9lPho"transcripts
was also observed in primary and secondary murine colony-
forming unit-spleen derived from transduced bone marrow
cells. Furthermore, respiratory burst activity was restored to
granulocyte-monocyte progeny of transduced X-CGD mice
bone marrow cells cultured in vitro. This observation is the
first reported use of gene transfer to correct the enzymatic
defect in murine CCD phagocytes and is also consistent with
the high conservation of the oxidase complex among different species. Taken together, these data suggest that the
MSCV-h91Neo vector may be useful for gene replacement
therapy in X-linked CGD, in which high-level reconstitution
of phagocyte oxidase activity may be important for full correction of phagocyte microbicidal function.
0 1996 by The American Societyof Hematology.
C
Because the genetic defect in X-CGD affects cells of the
hematopoietic system, retrovirus-mediated gp91Ph"' gene
transfer into hematopoietic stem cells
is one strategy that
may provide a lifelong cure. Complete correction of respiratoryburstactivity
in even as few as 10% of circulating
phagocytes might ameliorate the increased susceptibility to
infection in CCD, asinferred from studiesof female carriers
of X-linkedCCD.' The relativelevel of respiratoryburst
oxidase activity within individual phagocytes, however, may
also be an important determinant in the outcome of gene
replacement therapy. "Variant" patients with X-linkedCGD
who have some residual respiratory burst activity
can still
develop severe infectious complications,'.' and in vitro killing of Staphylococcus aureus by neutrophils from such patients has been shown to be markedly abnormal.' These observationssuggestthatapartialcorrectionofrespiratory
burst oxidase activity may not be sufficient to restore full
microbicidal function to the phagocyte.
Reconstitution of phagocyte respiratory burst activity in
the gp9lP""', ~ 2 2 / ' ~ "and
, p47"*" deficient forms of CGD has
beenachievedingranulocytesand
monocytes cultured in
vitro after transduction of peripheral blood stedprogenitor
cells by simplified retroviral vectors.'".'' The relative level
of superoxide generation by phagocytes derived from transduced cells could not be measured directly in these studies.
Our laboratoryhas alsodeveloped asimplifiedretroviral
vector, ZipPGKgp9 1, for expression of the human gp9lPh""
cDNA and characterizeditsactivity
in clonalpopulations
isolatedaftertransduction
of a human X-CGD phagocyte
cell line." This cultured cell model of X-CGD was derived
from PLB-985 myeloid leukemia cells by disruption of the
gp91Ph"xgene using gene targeting." Expression of recombinant gp91''h" proteinin ZipPGKgp91-transduced X-CCD
PLB-985 cells was, on average, 5% of wild-type levels, and
respiratory burst oxidaseactivity was only partially reconstituted."
In the present study, wereport the functionalcharacterization of a new retroviral vector, MSCV-h91Neo, which contains the cDNA for humangp91Ph"'along with a downstream
expression cassette for neomycinphosphotransferase. The
HRONIC granulomatous disease (CCD) is an inherited
disease in which any one
of four subunitsof the phagocyte respiratoryburst oxidase (NADPH-oxidase) isdefective, resulting in life-threatening bacterial and fungal infections due toa deficiency in microbicidaloxidants.' Mutations
in the X-linked gene for gp9IPhox,
a 91-kD membrane glycoprotein that is the larger subunit of the oxidase cytochrome
b, account for two thirds
of cases (X-CCD).A rare autosomal
recessive form of CGD results fromdefectsin
the gene
encoding p22"*", a 22-kD peptide that is the smaller subunit
of cytochrome b. The remaining two autosomal
recessive
subgroups involve mutations in one of two soluble proteins
of the oxidase complex, ~47"""" and
~ 6 7 ~ " "Current
".
management of CGD includes the prophylactic use of antibiotics'
andinterferon-y,3 butthedisorder
can still be associated
with considerable morbidity,especially in the X-linked form
of the d i ~ e a s e Murine
.~
models of X-linked' and autosomal
recessive' CCD have recently been developed that should
prove useful in investigating new therapeutic approaches to
this disorder.
From the Herman B WellsCenter for Pediatric Research, Departments of Pediatrics (Hematology/Oncology) and Medical and MOlecular Genetics, James Whitcomb Riley Hospital for Children, Indiana University Medical Center, Indianapolis; and Oncology
Research Laboratories, The TorontoHospital, Universiv of Toronto,
Ontario,Canada.
Submitted January 25, 1995; accepted April 16, 1996.
Supported by the National Institutes of Health, a Center for Excellence in Hematology award (1P.50 DK48218), a Program Project
Grant (POIHL53586), a Clincal Research Award from the March
of Dimes, and the National Cancer Institute of Canada.
Address reprint requests to Mary C. Dinauer, MD, PhD, Herman
B Wells Center for Pediatric Research, James Whitcomb Riley HOSpita1 for Children, Room 2600, 702 Barnhill Dr, Indianapolis, IN
46202.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
0006-4971/96/8805-0002$3.00/0
1834
Blood, Vol 88, No 5 (September I ) , 1996: pp 1834-1840
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
GENE TRANSFER OF HUMAN gp91ph"Ox
MSCV-LTR
KPI
h9 1
KpnI
WO1
PGKpr
Ne0
MSCV-LTR
WO1
W
A)
B)
1835
panded. Virus-containing supernatant was collected by adding 4 mL
of a-minimal essential medium (@"EM) containing 109 FCS to
IO-cm confluent plates overnight. Obtained mediumwasfiltered
(0.45 pm) and stored at -80°C until use.
To make an amphotropic MSCV-h9lNeo packaging line. Am12
packaging cells were infected with the high titer ecotropic MSCVh91Neo vector by supernatant "ping-pong." Briefly, Am12 cells
were incubated with MSCV-h9INeo retroviral supernatant in the
presence of polybrene (X pg/mL) (Aldrich Chemical CO,Milwaukee,
WI) for 2 hours on three successive days. A clone with a titer of 1
X 10' cfu/mL, identified as described above, was used for transduction of PLB-985 cells.
To confirm the(3418-resistancetiter and packaging of intact retrovirus by the ecotropic and amphotropic lines, RNA titering was done
by incubating SO% confluent 3T3 cells with retroviral supernatant
at a 1:10 dilution for 2 hours in the presence of 8 pg/mL polybrene,
extracting RNA from these cells 2 days later and performing Northern blot assay using the neo cDNA as a probe. Both ecotropic and
amphotropic producer cells contained intact retrovirus as shown by
the appearence of neo-containing transcripts of the expected sizes
(data not shown).
Trunsrluction of X-CCD PLB-985 cells. X-CCD PLB-985 cells
were transduced using a 2-day coculture infection protocolwith
amphotropic producer cells pretreated withmitomycin c as described." The transduced X-CGD PLB-985 cells were then plated
onto 96-well plates at limiting dilution in the presence of1.5 mgl
mL G418. At day 12, (3418-resistant clones were randomly picked
and expanded. Nine clones were assayed for the expression of provirus RNA, gp91"""' protein, and respiratory burst activity after granulocytic differentiation with dimethylformamide as described." In
most cases, data from all three assays were obtained on each clone,
except as noted.
Trmsduction qf wild-type murine CFU-S. Wild-type C57B 1/65
mice were injected intraperitoneally with IS0 mg/kg S-fluororacil
(S-FU: Solo Pak Laboratory Inc,Elk Grove Village. IL). and BM
cells were obtained from mice 2 days postinjection. The cells were
plated at a density of 2 X IO7 cellsllO-cm dish and prestimulated
for 2 days in a-MEM containing 30% heat-inactivated fetal bovine
serum, glutamine, P/S, 100 UlmL recombinant human interleukin6 (rhlL-h) (PeproTech, Rock Hill, NJ). and 1 0 0 n@mL recombinant
murine stem cell factor (rmSCF) (Genzyme, Cambridge. MA). On
day 3.4 X IO6 cells were plated on a IO-cm dish containing mitomycin c-treated viral producer cells plated the previous day at 3 X 10"
cells per dish, and cocultured for 2 additional days under the same
growth factor conditions in thepresenceof 4 pg/mL polybrene.
Lethally irradiated ( 1 I Cy) (cesium-137 using a gamma cell 40;
Nordion International Inc. Kanata, Ontario, Canada) wild-type
CS7B116J recipient mice weregiven I X IO' to 2 X IO' cells per
mouse intravenously to obtain individual colony-forming unit spleen
(CFU-S) foci. For secondary CFU-S-12 foci, approximately one half
of the cells of single CFU-S-14 colonies were transplanted into each
lethally irradiated secondary recipient. Individual CFU-S colonies
were dissected on days 12 or 14 posttransplantation and examined
for provirus integration and expression by Southern and Northern
blot analyses. respectively.
Transduction ef X-CCD mouse BM and culture in vitro. Fibronectin 30/3S fragments (kindly supplied byP.Dutt and D.A. Williams, Indiana University School of Medicine) were immobilized
on35-mmPetri
dishes (Falcon Labware, LincolnPark. NJ) at a
concentration of 4 pg/cm' as previously described." X-CGD donor
mouse BM was harvested from 6- to 8-week-old mice, and I X 10"
cells/mL nonadherent BM cells were exposed to ecotropic retroviral
supernatant for I to 2 days on fibronectin-coated dishes in a-MEM
containing 30% FCS (Hyclone Laboratories. Logan.UT), I O 0 U/
mL rhIL-6. and 100 nglmL rmSCF. BM cells were then collected
I
C)
Fig 1. MSCV-h91Neo retrovirus.The open boxes are MSCV-LTRs,
the shaded box is thehuman gp9lPhox
cDNA (h91). the dottedbox is
the murine PGK promoter (PGKpr), and the hatched box is the neomycin phosphotransferase cDNA(Neo). The arrows indicate possible
transcripts: (A) unsplicedviral genomic RNA derived from 5'-LTR (-7
kb before polyadenylation); (B)spliced viral genomic RNA derived
from 5'-LTR (-4 kb before polyadenylation);(C) PGK promoter-driven
mRNA ( 4 . 4 kb). Kpn I and Xho I represent restriction enzyme sites,
respectively.
murine stem cell virus (MSCV) vectors were designed for
efficient transduction and expression of genes in primitive
hematopoietic cells and their progeny and contain modified
myeloproliferative sarcoma virus LTRs and an optional selectable marker." We found that MSCV-h91Neo substantially improved recombinant gp91'""' expression in transduced X-CGD PLB-985 cells, which led to high levels of
respiratory burst oxidase activity in this model system. Furthermore, the MSCV-h91Neo vector also conferred functional expression of gp9Ip"'" in progeny of transduced XCGD murine bonemarrow (BM) cells cultured in vitro,
which is the first report demonstrating genetic correction of
the enzymatic defect in murine CGD phagocytes. These data
suggest that the MSCV-h91Neo vector maybe useful for
therapeutic retroviral-mediated gene transfer in X-CGD.
MATERIALS AND METHODS
NIH 3T3 fibroblasts (provided byD.A. Williams.
Indiana University School of Medicine, Indianapolis) were cultured
in Dulbecco's modified Eagle's medium (GIBCO, Grand Island,
NY) with 10% newborn bovine serum (Sigma, St Louis, MO), 100
U/mL penicillin (Sigma), and 1 0 0 pg/mL streptomycin (Sigma) (P/
S). Retrovirus packaging lines GP+ E86 (E86) and GP+ envAm12
(Am12) were obtained fromA.Bank (Columbia University, New
York, NY) and cultured in DMEF-I2 medium with 10% fetal calf
serum (FCS) (Sigma), 1% P/S, I % glutamine (GIBCO), and 0 . 1
molL HEPES (GIBCO). Wild-type and X-CCD PLB-985 cells were
cultured in RPM1mediumwith 10% FCS, I % P/S, and l % glutamine, granulocytic differentiation induced with dimethyl-formamide
as described."
Retroviral vector and producer cells. The human gp9 I~'""' cDNA
was cloned into an Xho I restriction site downstream of the LTR in
the MSCVneoEB shuttle plasmid. A phosphoglyero kinase (PGK)
promoter-controlled expression cassette for neomycin phosphotransferase (neo) is located 3' of hgp9Ip'"" cDNA (Fig l ) . This construct,
MSCV-h9INeo, was transfected into E86 packaging cells using the
DOTAP transfection reagent (Boehringer-Manheim, Indianapolis,
IN). Transfected packaging cells were selected in 0.8 mg/mL G418
(dry powder) (GIBCO). The ecotropic producer clone that had highest G418-resistance titer on 3T3 cells (2 X l@ cfu/mL) was exCell lines.
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
DING ET AL
1836
by cell dissociation buffer and plated for high-proliferative potential
(HPP) progenitors. As controls. mock infections of X-CCD and
wild-type BM were done under the same conditions a s above except
using a-MEM medium instead of rctroviral supernatant.
HPP progenitor colonies weregrown in 35-mm tissue culture
dishes using double-layer agar as described previously."'Briefly.
growth factors ( 100 ng rmSCF. 1.600 U human macrophage colonystimulating factor (hMCSF), 200 U rmlL-3, and 1 , 0 0 0 U rmlL-la)
(hMCSF and rmlL-la from Genzyme: rmIL-3 from PeproTech)
were incorporated into I-mL underlays with 0.5% agar and a-medium with 20% FCS. Retrovirus supernatant-infected andmockinfected BM cells were plated at 2.5 X 10' to S X IO' cells per dish
in a 0.5-mL 0.3% agar overlay with medium and FCS. Plates were
incubated in 5% 0 2 .1 0 % CO, at37°C for 14 days. To analyze
transduction efficiency and reconstitution of respiratory burst activity, an in situ nitroblue tetrazolium (NBT) testwas performed by
overlaying 0.4 mL of a saturated solution of NBT with S00 nglmL
phorbol myristate acetate (PMA) onto each day-14 HPPdishand
incubating for 2 hours at 37°C. Those colonies in which a portion
of the cells accumulated dark formazan deposits from reduction of
NBT by superoxide were scored positive. Wild-type and transduced
or mock-infected X-CCD BM cells were also expanded in a liquid
culture system at I X IO' cellslmL in a-MEM with20%FCSand
300 units recombinant human granulocyte colony-stimulating factor
(rhG-CSF) (Amgen Biologicals) for an additional 1 I days, refeeding
with an additional 300 U G-CSF every 2 days. The NBT slide test"
was performed on the expanded cells, which contained a mixture of
differentiating granulocytes and macrophagelmonocytes. The oxidase was activated by phorbol myristate acetate in the presence of
NBT. which is converted into insoluble dark purple formazan deposits by superoxide. After 40 minutes at 37°C. slides were fixed and
counterstained with safranin.
DNA. RNA, cInc1 immlrnohlor crricr1wi.s. DNAwas extracted using
lsoquick (Microprobe Corporation, Bothell. WA) according to the
manufacturer's instructions. Genomic DNAwas cutwith K p ~ zI to
assess provirus copy number or Xho I to examine the number of
junctional fragments, using the neo cDNA as a probe (see Fig I ) .
Total RNA was extracted by guanidine thiocyanate and separated
on a formaldehyde agarose gel as described.'' Southern and Northern
transfers to Magnacharge nylon membrane (Micron Separations,
Westboro. MA) were performed using standard protocols supplied
by the manufacturers. Blots wcre probed with cDNA fragments derivedfrom either the gp91"""' or neo cDNAs. as previously described." Cellular proteins were extracted using a buffered I % Triton X- IO0 solution. and electrophoresis, immunohlotting, and
scanning densitometry were performed as previously described." A
gp9I"""' monoclonal antibody was kindly providcd byDrD. Roos
and A. Verhoeven (Central Laboratory of the Netherlands Red Cross
Blood Transfusion Service, Amsterdam. The Netherlands).
Asscry qf .s~~pero.~ide
fonncrticm i n PLB-985 cells. Respiratory
burst activity in PMA-stimulated granulocytic PLB-985 cells was
assessed bythe NBT assay and. for quantitative measurement of
superoxide formation. a continuous cytochrome c reduction assay
as described.''
RESULTS
Fw~-tionnlc~xpressionof ~p91p'"'"in kttr?~nnX-CCD PLR985 myloid cells transdlrced with MSCV-h9INeo. After
transduction of X-CGD PLB-985 cells with MSCV-h91Neo
retrovirus. G418-resistant clones were isolated by limiting
dilution. Twenty-four randomly picked clones exhibited reconstitution of respiratory activity a s assessed by the NBT
test performed after granulocyte differentiation. This indicated that both the neo gene and human
gp91/""" gene were
L
I
X-CGD +
0 MSCV-h91Neo
+ a
,
I
I
I
I
I
I
l
l
l
l
l
28s -
18s -
Fig 2. Northern blot analyses of MSCV-h91Neo-transducedXCGD myeloid cells. Total cellular RNAs (20 p g per lane) were probed
with radiolabeled h91 cDNA (top) and reprobed with radiolabeled
actin cDNA (bottom). Results from one of t w o separate experiments
are shown. M,wild-type PLB-985 cells; +DMF indicates cells after
granulocyte differentiation; X-CGD,
X-CGD
PLB-985
cells;
XCGD+MSCV-h91Neo. representative retrovirally transduced clones
(numbered 2 through 9, which also correspond to clones 2 through
9 in Fig 3 and Table 1). The position of 28s and 18s ribosomal RNAs
(top) and actin RNA (bottom) are indicated. The arrows indicate the
position of the unspliced and spliced LTR-driven transcripts.
functionally expressed from the provirus in the transduced
in one to two
X-CGD PLB-985 cells, whichwaspresent
copies per cell in each clonal population (data not shown).
Nine of 2 4 NBT-positive clones were randomly chosen for
detailed analysis. A gp91/'"''' transcript of approximately 4.9
kb was detected in undifferentiated X-CGD PLB-985 cells
by Northern blot (Fig 2 ) . which was the size predicted for
the unspliced viral S'-LTR transcript (Fig
I). Very little. if
any, of thesplicedtranscriptwasdetectable.
A clone-toclone variation in the level of vector derived RNA was observed. which was unrelated
to copy number but did correlate
with gp9lP""' protein levels (see below). To examine expression of gp91/""" in the transduced X-CGD PLB-985 clones,
proteinwasextractedfromgranulocyte-induced
cells and
analyzedbysodiumdodecylsulfate-polyacrylamidegel
electrophoresis and immunoblotting with a monoclonal antibody for gp91'""' (Fig 3). Before transduction, no gp9Ip""'
wasdetectedinX-CGDPLB-985cells.Immunoreactive
gp9Ir""" protein was detected in all transduced clones, with
the level in different clones ranging from 6% to 48% (mean,
25%) of that present in wild-type PLB-985
cells (Table I),
as estimated by scanning densitometry of the
gp91/"'"' signal.
Respiratoryburstoxidaseactivity.asmeasured
by the
quantitative cytochrome c assay. was partially or completely
reconstituted in X-CGDPLB-985 cells aftertransduction
with MSCV-h9INeo (Table I ) . The relative levels of super-
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
GENETRANSFEROF
1837
HUMAN gp9lPho"
n
Fig 3. lmmunoblot analyses of MSCV-h91Neotransduced X-CGD myeloid cells. Cellular proteins
were extracted from granulocyte-differentiated
PLB985 cells and derivatives and analyzed for gp91Ph0"
expression by immunoblot. Data shown are from
one of two separate experiments.
wild-type
PLB-985
cells;
X-CGD,
X-CGD
PLB-985
cells;
XCGD+MSCV-hSlNeo, retrovirally transduced clones
(numbered 2 through 9). Twenty micrograms of extracted cellular protein was loaded in each lane, except for wild-type PLB-985, for which 10 @g was
loaded, as noted. The two faint bands on theX-CGD
lane are presumed t o represent proteins that bind
t o antibody nonspecifically as they are not detected
with other gp91Ph0"specific antibodies. The blot was
probed with a gp9lPhox monoclonal
antibody.
X-CGD +
MSCV-h91Neo
WT.
c
oxide production in transduced clones ranged from 57% to
119% (mean.88%) ofwild-typelevels.
Clones 3 and 5 ,
which contained the lowest levels of gp91'""" protein, exhibited rates of superoxide formation that were approximately
50% of wild-type. with the remaining clones having oxidase
activity in the wild-type range.
E.rpression nrtcl.fitrlctiorrof MSCV-h9INeo in priman murine kentaropietic cells. Expression
of
vector-derived
mRNAs in primary hematopoieticcellswas
assessed by
Northern blot analyses of CFU-S- I2and CFU-S- 14 derived
from MSCV-h91 Neo-transduced murine BM cells. Genomic
Southern blot showed 7 of 7 (100%)CFU-S- 12 colonies that
contained oneto two copiesof the integrated provirus,except
for one clone, which had four copies (data not shown). In
two additional experiments, 24 of 28 (86%) CFU-S-12/14
and 2 of 12 secondary CFU-S-12 colonies expressed provirus-derived transcripts as shown by Northern blot (Fig 4).
The unspliced LTR-driven message was the most abundant
vector-derived transcript in the majority of CFU-S. The level
of this transcript varied in different colonies and was relatively lower in the two RNA-positive secondaryCFU-S com-
---
"
pared with the range seen in primary CFU-S and in transduced X-CGD PLB-985 cells.
To assess the functional expression of gp91"""' in primary
hematopoietic cells after MSCV-h91 Neo transduction, XCGD mouse BM was transduced with virus supernatant and
plated for HPPcolonies. Colonies were not selected for G41 8
resistance, because cells derived from the X-CGD mice contain a functional PGK-neo transgene.5 Respiratory burst activity, as monitored by the NBT test, was detected in 6% to
21% of day-l4 HPP derived from transduced X-CGD BM,
whereas nopositive colonieswere noted in HPP derived
from mock-infected X-CGD BM (Table 2). Functional expression of human gp91"""' in progeny of murine X-CGD
BM cells transduced with MSCV-h91Neo was confirmed by
thereconstitution of respiratoryburstactivity,
as assayed
by the NBTslide test, in a subpopulationofneutrophils
differentiated in a liquid culture system (FigS ) . In this experiment, 8% of X-CGD neutrophils derived from transduced
X-CGD BM cellswere NBT-positive, with a degree of
formazan staining that was comparable in intensity with that
observed for neutrophils differentiated from wild-type mu-
Table 1. Superoxide Generation by Human PLB-985 Cell Lines After Granulocytic Differentiation
Vm""
0;
Cell line
(nmol O2/min/1O7cells)
lnmol/30 minl10' cells)
n
Relative Superoxide-Generating
Activity (%)
Relative Level of
gp91'"O' (%)
Wild-type
X-CGD
X-CGD+MSCV-h91Neo
Clone l
Clone 2
Clone 3
Clone 5
Clone 6
Clone 7
Clone 8
Clone 9
43 lr 11
0
610 lr 135
0
4
4
100
0
100
0
27 -c 12
45 2 13
14 t 4
19 -c 11
77 lr 15
52 f 21
74 2 6
40 -c 6
547
727
287
320
644
600
593
593
90
119
47
52
106
98
97
97
9
24
6
6
42
28
48
38
2 208
2 47
2 9
f 302
2 102
2 94
2 85
2 66
PLB-985 cell lines were induced to differentiate with 0.5% DMF for 6 days. Superoxide formation was measured by a continuous cytochrome
c reduction assay. Data for superoxide formation are the mean t SD.The relative superoxide-generating was calculated from the mean nmol/
0;/30 minll0' cells. The relative level of gp91PhoX
was estimated using scanning densitometry of a representative immunoblot probed with a
gp9lPho"antibody.
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
DING ET AL
1838
A
X-CGD
PLB-985
12
CFU-S14
CFU-S
ff-
28s -
18s
-
fddlk
Actin -
A
“A
a
B
J 1
Secondarv CFU-S 12
X-CG D
PLB-985
a
I
i
e
28s -
Actin -
in the two
positive
secondary
were
CFU-S-12 that
for LTR-derived transcripts.
rine BM, 87% of which were NBT-positive. Although the
NBT test is not a quantitative measurement of respiratory
burst activity, the density of formazan deposits in activated
phagocytes does reflect the relative level of superoxide generation as measured by the cytochrome c assay.” For
Table 2. Respiratory Burst Oxidase Activity in HPP Colonies
Derived From Wild-Type and X-CGD Murine BM
Wild-type
5/58
X-CGD mock
X-CGD
251391
transduced
3/49
n
Fig 4. Northern
blot
analyses of CFU-S colonies
obtained after transduction of wild-type mouse BM
cells with MSCV-h91Neo. Total cellular RNAs (20 p g
per lane) from primary(A) and secondary CFU-S colonies (B) were probed with radiolabeled neo cDNA
(top)and reprobed with radiolabeled with actin
cDNA (bottom). Results from one of two experiments are shown. CFU-S-12 and CFU-S-14 represent
CFU-S colonies dissected on day 12 and day 14 posttransplantation, respectively; X-CGD
PLB-985,
XCGD PLB-985 cells transduced with MSCV-h91Neo
(clones 6 and 7 in Figs 2 and 3). PLB-985 cells have
a higher actin content thanCFU-S, and the amounts
of RNA loaded in the X-CGDPLB-985 lanes were
comparable with the CFU-Ssamples, based on the
rRNA signal. The position of 28s and 18s ribosomal
RNAs (top) and actin RNA (bottom) are indicated.
The arrows indicated the position of the unspliced
and spliced LTR-driven transcripts and thePGK promoter-driven transcript; the latter was not
detected
Experiment 1
Experiment 2
Experiment 3
70175t
(93)
On0
22/22
1151142
(81)
01139
NA
7/34
(21)
(6)
NA
(6.3)
example, neutrophils from “variant” X-CGD patients with
residual respiratory burst activity exhibit weakly positive
NBT-staining.X.’X
Again, no NBT-positive neutrophils were
observed in mock-infected murine X-CGD BM cells. The
apparent lower rates of gene transfer in the above experiments, as assessed by the frequency of NBT-positive cells,
compared with the CFU-S 12 studies described above, are
likely to be related to the differences in transduction protocols (see Materials and Methods).
Experiment 4’
DISCUSSION
(100)
0122
(9)
Values in parentheses are percents.
Experiment 4 used 5-FU day-2 murine bone marrow cells,
whereas experiments 1 to 3 used BM from mice not treated with 5FU.
t The number of NET-positive HPP coloniesltotal HPP colonies analyzed.
These studies demonstrate that retroviral-mediated gene
transfer of human g ~ 9 1 ~ ” ’the
” . ~protein
,
that is defective in
the X-linkedform of chronic granulomatous disease, can
confer high-level reconstitution of respiratory burst activity
in a cultured cell model of human X-CGD phagocytes and
restore superoxide production to murine X-CGD phagocytes.
The vector used, MSCV-h91Neo, isbasedontheMSCV
vector series developed by Hawley et all4 from the murine
embryonic stem cells virus (MESV) and the LN retroviral
vectors. A number of features make this vector design attrac-
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
1839
GENETRANSFER OF HUMAN gp9lPho”
A
C
Fig 5. NBT assay of respiratory burst oxidase activity in murine neutrophils. Neutrophils were derived from in vitro liquid culture of murine BM cells
after incubation on fibronectin fragments in the absence (mock) or presenceof MSCV-h91Neo retroviral
supernatant. (A) Wild-type (mock); (B) X-CGD
(mock). Only safranin staining is visible. (C) X-CGD
IMSCV-h91Neo). Formazan deposits are visiblein
the center cell.
tive. First, its variant LTR from the MESV directs efficient
expression of downstreamsequences in primitive cells.”
Second, it contains the extended packaging signal ($+) of
LN vectors with absence of all env sequence, and a mutagenized gag region minimizes the generation of replication competent viruses. Third, a downstream selectable marker cassette facilitates the generation and
titering of packaging lines
and can be potentially used to concentrate transduced target
cells. Previous studies have shown that MSCV vectors conferred long-term in vivo expression of the IL-6 and IL-l I
cDNAsafter transductionof murinehematopoieticstem
As the first step in assessing MSCV-h91 Neofunction, we
used a human X-CGD PLB-985 myeloid cell line
as an in
vitro model of X-CGD phagocytes. In transduced X-CGD
PLB-985 cells,the level of recombinant gp91/”””protein was,
on average, 25% of wild-type levels, which was a fivefold
improvement over what had been previously observed using
the ZipPGK-gp91 retrovirus.” The reason forthehigher
levelsof gp91”””‘ protein expression is uncertain. In Zip/
PGK-gp91 -transduced cells, vector gp9lP”‘” transcripts were
derived almost entirely from the Maloney murine leukemia
virus LTR and wcrc comparablc in abundance to wild-type
PLB-985 granulocyte gp91””“‘ mRNA. However, recombinant gp9IP””,‘protein levels were low, suggesting inefficient
translation,possibly due to secondary structure in a PGK
promotersequence immediatelyupstreamof
the gp91””‘”
cDNA. In the MSCV-h91Neo vector, a modified myeloproliferative sarcoma virus LTR drives gp91p”””expression. The
level of vector-derived gp9IP””‘ transcripts varied
in different
transduced clones and, in some cases, exceeded expression
of the endogenous gp91p””.rmRNA in wild-type PLB-985
granulocytes. The reason for the clonal variation in expression did not appear to be related to the number of proviral
integrations, suggesting that integration into some chromosomal sites may be less favorable for transcription. Amounts
ofrecombinant gp91”””‘ protein correlated withthe abundance of the proviral transcript but were still below wildtype proteinlevels,again
suggestingsome inefficiency in
translation.
Importantly, the improved levels of recombinant gp9lP”””
protein expression resulted in substantialrespiratoryburst
oxidase activity in MSCV-h91Neo-transduced X-CGD PLB985 cells, which was ir? the normal range in the majority of
clones analyzed. Even clones with as little as =6% of wild-
type protein levels exhibited almost 50% of wild-type levels
of superoxide formation. These data are consistent with our
previous results indicating that relatively small amounts of
gp91p’”” expression can reconstituteconsiderablesuperoxide-generating activity.“,’” Theability to restore high levels
of enzyme activity to deficient phagocytes may be critical
to the success of future clinical gene transfer applications,
as X-CGD patients with low residual levels of oxidase activity can still experience a severe clinical course.’.‘’
Because vector function in permanent cell lines does not
always predict expression in primaryhematopoieticcells,
we also evaluated MSCV-h91 Neofunction in wild-type and
X-CGD murine BM cells. Expression of vector-derived transcripts was observed in both primary and secondary
CFUS - l 2 colonies, whichexhibited a clone-to-clonevariation
in abundance similar to that observed for MSCV-h91Neotransduced PLB-985 cells. Therelative level of LTR-driven
mRNAwaslower
in the secondary CFU-S-12 compared
with that seen in primary CFU-S-12 and in X-CGD PLB985 cells. Whether this reflects a consistent decrease in LTR
function in moreprimitivehematopoietic
cellscannotbe
determined in these studies. Promoter suppression in retrovirus vectors that have two promoters has been reported, particularlyaftertransductionofprimarycells.”
Both LTRderived gp91’”‘” transcriptsand PGK promoter-driven neo
transcripts were detected in the majority of MSCV-h9 1 Neotransduced CFU-S colonies. However, further studies evaluating long-term expression in vivo will be required to determine whether suppression of the LTR-derived transcript in
bone marrow cells occurs over time.
MSCV-h9 1Neo-mediated expression of human gp9 Ir””‘
also reconstitutesthe respiratoryburst enzymedefect in
granulocyte-monocyte cells of the X-CGD mouse, which is
the first reported use ofgene transfer to correct the functional
defect in murine CGD phagocytes. Superoxide formation in
progeny of transduced X-CGD BM cells, as assayed by the
NBT test, was observed in bothneutrophilsdifferentiated
from progenitors in a liquid culture system and in HPP colonies. The ability of the human gp9lP”“ protein to complement the murine oxidase complexis consistent with the conservednatureof
theenzymesubunitsamong
different
species.6.?3-?5The murineg ~ 9 1 “ ” ”has
. ~ previously been shown
to reconstitute superoxide formation in human X-CGD PLB985 cells.26 Theaminoacidsequence
of human gp9Ir’””.‘
differs in only 39 of 570 residues from its murinehomo-
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
1840
DING ET AL
logue, with the majority of differences being conservative
substitutions that occur outside of putative heme-binding and
flavoprotein domains.26
In conclusion, the MSCV-h91Neo retrovirus is a promising vector for gene therapy of X-linked CGD. High levels
of respiratory burst oxidase activity were reconstituted in a
human X-CGD phagocyte cell line, a factor thatmay be
important in full restoration of phagocyte microbicidal activityand correction of the defect inhost defense in vivo.
Transduction of BM cells obtained from X-CGD mice also
resulted in expression of functional gp9lPhoX
in granulocytemonocyte progeny cultured in vitro. An important remaining
question from the standpoint of gene replacement therapy is
the longevity of provirus expression in vivo after transduction of reconstituting hematopoietic stem cells with MSCVh91Ne0, which is presently under investigation in the XCGD mouse.
ACKNOWLEDGMENT
We thank Ling Lin Li for construction of the MSCV-h91Neo
plasmid, Ling Zhen for performing the densitometry analysis of
immunoblots, Mary Gifford and Jim Cummings for genotyping and
maintenance of the X-CGD mouse colony, David Williams for rat
SCF used in pilot studies on murine BM, Xangli Xiao for providing
advice on performing the HPP assays, and Donna Fischer for assistance with report preparation.
REFERENCES
1. Curnutte J, Orkin S, Dinauer M: Genetic disorders of phago-
cyte function, in Stamoyannopoulos G (ed): The Molecular Basis of
Blood Diseases (ed 2). Philadelphia, Saunders, 1994, p 493
2. Gallin J, Buescher E, Seligmann B, Nath J, Gaither T, Katz
P: Recent advances in chronic chronic granulomatous disease. Ann
Intern Med 99:657, 1983
3. The International Chronic Granulomatous Disease Cooperative
Study Group: A controlled trial of interferon gamma to prevent
infection in chronic granulomatous disease. N Engl J Med 324:509,
1991
4. Weening R, Adriaansz L, Weemaes C, Lutter R, Roos D:
Clinical differences in chronic granulomatous disease inpatients with
cytochrome b-negative or cytochrome b positive neutrophils. J Pediatr 107:102, 1985
5. Pollock J, Williams D, Gifford M, Li L, Du X, Fisherman J,
Orkin S, Doerschuk C, Dinauer M: Mouse model of X-linked chronic
granulomatous disease, an inherited defect in phagocytes superoxide
production. Nat Genet 9:202, 1995
6. Jackson S, Malech H, Kozak C, Lomax K, Gallin J, Holland
S: Cloning and functional expression of the mouse homologue of
p47ph"". Immunogenetics 39:272, 1994
7. Woodman R, Teoh D, Johnston F, Rae J, Sim V, Urton T,
Wright B, Curnutte J: X-linked carriers of gp9'-ph0x
chronic granulomatous disease (CGD): Clinical and potential therapeutic implications of extreme unbalanced X chromosome inactivation. Blood
86:27a, 1995
8. Bu-Ghanim H, Segal A, Keep N. Casimir C: Molecular analysis in three cases of X91- variant chronic granulomatous disease.
Blood 86:3575, 1995
9. Roos D, de Boer M, Borregard N, Bjermm 0, Valerius N,
Seger R, Muhlebach T, Belohradsky B, Weening R: Chronic granulomatous disease with partial deficiency of cytochrome b558and
incomplete respiratory burst: Variants of the X-linked, cytochrome
b558-negative form of the disease. J Leukoc Biol 51:164, 1992
IO. Sekhsaria S, Gallin J, Linton G, Mallory R, Mulligan R,
Malech H: Peripheral blood progenitors as a target for genetic correctionof ~47~~""-deficient
chronic granulomatous disease. Proc Natl
Acad Sci USA 90:7446, 1993
11. Li F, Linton G, Sekhsaria S, Whiting-Theobald N, Katkin J,
Gallin J, Malech H: CD34+ peripheral blood progenitors as a target
for genetic correction of the two flavocytochrome bSS8defective
forms of chronic granulomatous disease. Blood 84:53, 1994
12. Kume A, Dinauer M: Retrovirus-mediated reconstitution of
respiratory burst activity in X-linked chronic granulomatous disease
cells. Blood 84:3311, 1994
13. Zhen L, King A, Xiao Y, Chanock S, Orkin S, Dinauer M:
Gene targeting of X-linked chronic granulomatous disease locus in
a human myeloid leukemia cell line and rescue by expression of
recombinant gp9Iph"".Proc Natl Acad Sci USA 90:9832, 1993
14. Hawley R, Lieu F, Fong A, Hawley T: Versatile retroviral
vectors for potential use in gene therapy. Gene Ther 1:136, 1994
15. Moritz T, Patel V, Williams D: Bone marrow extracellular
matrix molecules improve gene transfer into human hematopoietic
cells via retroviral vectors. J Clin Invest 93:1451, 1994
16. Lowry P, Deacon D, Whitefield P, Rao S, Quesenbeny M,
Quesenbeny P: The high-proliferative-potential megakaryocyte
mixed (HPP-Meg-Mix) cell: A trilineage murine hematopoietic progenitor with multiple growth factor responsiveness. Exp Hematol
23:1135, 1995
17. Ochs H, Igo R: The NBT slide test: A simple screening
method for detecting chronic granulomatous disease and female carriers. J Pediatr 83:77, 1973
18. Curnutte J: Disorders of Granulocyte Function and Granulopoiesis, vol. 1 (ed 4). Philadelphia, Saunders, 1992
19. Grez M, Akgun E, Hilberg F, Ostertage W: Embryonic stem
cell virus, a recombinant murine retrovirus with expression in embryonic stem cells. Proc Natl Acad Sci USA 87:9202, 1990
20. Hawley R, Fong A, Bums B, Hawley T: Transplantable myeloproliferative disease induced in mice by an interleukin 6 retroviNS. J Exp Med 176:1149, 1992
21. Hawley R, Fong A, Ngan B, de Lanux V, Clark S, Hawley
T: Progenitor cell hyperplasia withrare development of myeloid
leukemia in interleukin 11bonemarrow
chimeras. J Exp Med
178:1175, 1993
22. Williams DA: Expression of introduced genetic sequences
in hematopoietic cells following retroviral-mediated gene transfer.
Human Gene Ther 1:229, 1990
23. Zhou Y, Murtaugh M: Cloning and expression of the gene
encoding the porcine NADPH oxidase light-chain subunit (p22phox). Gene 148:363, 1994
24. Sumimoto H, Nozaki M, Sasaki H, Takeshige K, Sakaki Y,
Minakami S: Complementary DNA for the mouse homolog of the
small subunit of human cytochrome b55s. Biochem Biophys Res
Commun 165:902, 1989
25. Knoller S, Shpungin S, Pick E: The membrane-associated
component of the amphiphile-activated, cytosol-dependent superoxide-forming NADPH oxidase of macrophages is identical to cytochrome bI19. J Biol Chem 266:2795, 1991
26. Bjorgvinsdbttir H, Zhen L, Dinauer M: Cloning of murine
gp91Ph"'cDNAand
functional expression in a human X-linked
chronic granulomatous disease cell line. Blood 87:2005, 1996
From www.bloodjournal.org by guest on July 28, 2017. For personal use only.
1996 88: 1834-1840
High-level reconstitution of respiratory burst activity in a human Xlinked chronic granulomatous disease (X-CGD) cell line and
correction of murine X-CGD bone marrow cells by retroviral-mediated
gene transfer of human gp91phox
C Ding, A Kume, H Bjorgvinsdottir, RG Hawley, N Pech and MC Dinauer
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/5/1834.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.