Cell Stem Cell
Neuroprotection of Host Cells by Human Central
Nervous System Stem Cells in a Mouse Model
of Infantile Neuronal Ceroid Lipofuscinosis
Stanley J. Tamaki,1Yakop Jacobs,1Monika Dohse,1Alexandra Capela,1Jonathan D. Cooper,2Michael Reitsma,1
Dongping He,1Robert Tushinski,1Pavel V. Belichenko,3Ahmad Salehi,3William Mobley,3Fred H. Gage,5Stephen Huhn,1
Ann S. Tsukamoto,1Irving L. Weissman,4and Nobuko Uchida1,*
1StemCells, Inc., 3155 Porter Drive, Palo Alto, CA 94304, USA
2Department of Neuroscience, Centre for the Cellular Basis of Behaviour, MRC Centre for Neurodegeneration Research,
Institute of Psychiatry, King’s College London, 125 Coldharbour Lane, London SE5 9NU, UK
3Department of Neurology and Neurological Sciences
4Institute of Stem Cell Biology and Regenerative Medicine, Departments of Pathology and Developmental Biology
Stanford University School of Medicine, Stanford, CA 94305, USA
5Laboratory of Genetics, The Salk Institute for Biological Studies, 10010 North Torrey Pines Road, La Jolla, CA 92037, USA
Infantile neuronal ceroid lipofuscinosis (INCL) is a
fatal neurodegenerative disease caused by a defi-
ciency in the lysosomal enzyme palmitoyl protein
thioesterase-1 (PPT1). Ppt1 knockout mice display
hallmarks of INCL and mimic the human pathology:
accumulation of lipofuscin, degeneration of CNS
neurons, and a shortened life span. Purified non-
genetically modified human CNS stem cells, grown
as neurospheres (hCNS-SCns), were transplanted
into the brains of immunodeficient Ppt1?/?mice
where they engrafted robustly, migrated extensively,
and produced sufficient levels of PPT1 to alter host
neuropathology. Grafted mice displayed reduced
autofluorescent lipofuscin, significant neuroprotec-
tion of host hippocampal and cortical neurons, and
delayed loss of motor coordination. Early interven-
tion with cellular transplants of hCNS-SCns into the
brains of INCL patients may supply a continuous
and long-lasting source of the missing PPT1 and
provide some therapeutic benefit through protection
of endogenous neurons. These data provide the
experimental basis for human clinical trials with
these banked hCNS-SCns.
Neuronal ceroid lipofuscinosis (NCL), commonly referred to as
Batten disease, belongs to a group of hereditary neurodegener-
ative lysosomal storage diseases (LSD) (reviewed in Mole,
2004; Wisniewski et al., 2001). Among the nine identified forms
of NCL, the infantile (INCL), late infantile (LINCL), and congenital
(CNCL) forms are each caused by autosomal-recessive muta-
tions of the genes that encode lysosomal enzymes. INCL results
from mutations in the CLN1 gene, which encodes for palmitoyl
protein thioesterase-1 (PPT1) (Das et al., 1998; Hofmann et al.,
of lipofuscin, an autofluorescent lipoproteinaceous storage
material that characterizes the pathology of this disease. Clinical
symptoms, which begin as early as 6 months of age, manifest
as a decline in cognitive and motor skills, visual impairment,
seizures, and premature death. The development of a PPT1
ulates many aspects of its human disease counterpart, INCL,
provides an experimental model for testing the potential thera-
peutic benefits of human neural stem cell transplant for the
treatment of this disease.
Enzyme replacement therapy (ERT), which ameliorates non-
CNS symptoms for some LSDs, is not beneficial in treating
CNS neurodegeneration, because intravenously administered
enzymes can not cross the blood-brain barrier. Currently, there
is no effective treatment available for any form of NCL. Here
we take a novel approach of transplanting normal, nontumori-
genic, and non-genetically modified banked human neural
stem cells to provide secreted lysosomal targeted enzymes to
a mouse model of NCL.
We have generated banks of expanded cells from prospec-
tively isolated human central nervous system stem cells
(hCNS-SC). Purification of cells by surface marker expression
yields a highly purified neural stem cell population (Tamaki
et al., 2002; Uchida et al., 2000) that is grown as neurospheres
(hCNS-SCns, also known as HuCNS-SC) in serum-free defined
media. We hypothesize that transplantation of hCNS-SCns
directly into the brains of Ppt1?/?mice will provide the deficient
enzyme through secretion of PPT1 by these cells. PPT1 is
captured via mannose-6-phospate receptors on the cell surface
and is transported to lysosomes of target cells, where it partici-
pates in degrading accumulated toxic metabolites; this process
is referred to as ‘‘cross-correction.’’ To test our hypothesis,
Ppt1?/?mice were backcrossed into immunodeficient NOD-
SCID mice to generate the xenotransplant INCL mouse model
(Ppt1?/?/NSCID) and a series of human neural stem cell trans-
plantation studies were conducted.
In this study we describe the engraftment, neurogenesis,
migration, and differentiation characteristics of hCNS-SCns
310 Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc.
transplanted into Ppt1?/?/NSCID mice. These transplanted
human hCNS-SCns migrated extensively, provided PPT1,
reduced endogenous lipofuscin storage material, delayed the
loss of motor coordination, and led to broad neuroprotection of
host cells in the hippocampus and cortex.
In Vitro, hCNS-SCns Secrete the PPT1 that Is
Internalized by Mutant CLN1 Fibroblasts via
the Mannose 6-Phosphate Receptor
To determine whether our cross-correction hypothesis would
for 7 days in a transwell system with fibroblasts derived from the
Ppt1?/?mouse and CLN1 patient. Secreted proteins can pass
through the transwell without direct contact between hCNS-
SCns and fibroblasts. After coculture with hCNS-SCns, a signifi-
cant increase and accumulation of intracellular PPT1 were
observed in mutant fibroblasts derived from Ppt1?/?mice (p <
0.0005) or a CLN1 patient (p < 0.05). Addition of a competitive
inhibitor, mannose 6-phosphate, reduced PPT1 activity to basal
levels in both Ppt1?/?and CLN1 mutant fibroblasts (Figures 1A
and 1B). Thus, hCNS-SCns produce and secrete lysosomal
PPT1 via the expected biochemical pathways for secretion
and endocytosis in vitro. We expect that hCNS-SCns will exhibit
this same mechanism of action to cross-correct enzyme-defi-
cient host cells in vivo.
hCNS-SCns Engraft and Migrate in the Brain
of Ppt1?/?/NSCID Mice
entiate into different neural cell types in Ppt1?/?/NSCID mice.
after transplant (almost at the end of their life span). Human cells
specific antigen, respectively. Sagittal brain sections showed
that the grafted cells were present in the subventricular zone
(SVZ), the endogenous mouse stem/progenitor niche. In addi-
tion, a high density of human cells was detected in the rostral
migratory stream (RMS), the normal migration route of endoge-
nous neurogenesis in the olfactory system, and additionally in
the cortex (Figures 1C and 1D).
Stereological counts of transplanted human cells were made
from one hemisphere of five Ppt1?/?/NSCID mice and six
NOD-SCID mice. On average, 2 3 105cells were detected in
Ppt1?/?/NSCID mice, and NOD-SCID mice showed an average
of 8.5 3 104cells. In the hippocampus of Ppt1?/?/NSCID mice,
analyses showed that transplanted hCNS-SCns can survive in
the neurodegenerative environment of the Ppt1-deficient cortex
hCNS-SCns Differentiate in a Neurodenegerative
Environment and Integrate into the Architecture
of the Host Brain
We examined whether hCNS-SCns retained their differentiation
potential in the microenvironment of the degenerating Ppt1?/?
brain by histologically characterizing these engrafted cells.
Staining with SC121 revealed that cells in the olfactory bulb
exhibited neuron-like morphology (Figure 2A), appeared astro-
cytic near the SVZ and corpus callosum (Figure 2B), and resem-
bled oligodendrocytes in white matter tracts (Figure 2C). Cells in
the cortex exhibited an undifferentiated morphology (Figure 2D),
similar to the morphology of human cells in the cortex of normal
Ppt1+/+/NOD-SCID mice. Lineage-specific staining was con-
ducted, with Nestin as a marker for neural stem/progenitor cells
and Doublecortin (DCX) and MAP2 for immature and mature
P < 0.05
Human CLN1 fibroblast
P < 0.0005P < 0.0001
Figure 1. PPT1 Secreted by hCNS-SCns Is Internalized by Mutant
(A and B) Extracellular uptake of PPT1 can be blocked by addition of free
mannose-6-phosphate to cultures. PPT1 activity of mutant of fibroblasts
derived from (A) mouse Ppt1?/?or (B) CLN1 patients. In each histogram
mean ± SEM PPT1 activity is plotted from mutant fibroblasts alone (green),
transwell cocultures with hCNS-Sns (red), and cocultures with hCNS-SCns
plus mannose-6-phosphate (blue).
(C and D) Engraftment of hCNS-SCns transplanted into the Ppt1?/?/NSCID
(C) Immunoperoxidase staining with the human-specific mAb SC101 (brown).
Neonatal Ppt1?/?/NSCID mice received transplants into the anterior cortex,
the lateral ventricle, and cerebellum. Atotalof 83 105hCNS-SCns weretrans-
planted, i.e., each hemisphere received 4 3 105hCNS-SCns, and was
analyzed at 175 days post-neonatal transplant. The progeny of human cells
were detected by SC101, which recognized human nuclear antigen. Anterior
and posterior labels indicate orientation of this sagittal section.
(D) Immunoperoxidase staining of an adjacent section with a second human-
specific mAb SC121 (brown) reveals a similar distribution of grafted hCNS-
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc. 311
neurons, respectively. As expected, human-specific Nestin
(hNestin) identified a densely packed region of positive cells in
the SVZ (Figure 3A). In the olfactory bulb, human neuronal cells
double stained with SC121 and DCX (Figure S4D) or MAP2
(Figure 3B) were detected (see below for more details).
The SVZ is a neurogenic niche composed of stem/progenitor
cells undergoing cell division that can be detected with Ki67
antibody. Serial brain sections of transplanted Ppt1?/?/NSCID
mice revealed that, although there are many human cells in
and around the SVZ (Figure 3C), only dispersed Ki67-positive
cells are present (Figure 3D). The origin of these Ki67-positive
cells (whether they are human or mouse) is not known because
the Ki67 antibody used reacts with both species and human
cells in the SVZ are too densely packed to evaluate individual
cells. A quantitative assessment was performed to determine
the proportion of proliferating human cells (i.e., SC121+Ki67+)
in the cortex. The assessment revealed that only rare human
cells were Ki67 positive (2.6% ± 0.88% [SEM]) (Figures S2A
and S2B, Table S2). These data suggest that proliferation of
transplanted human cells in the Ppt1?/?/NSCID mouse CNS is
Astrocytes play multiple roles in the formation of brain archi-
tecture and function. They associate with blood vessels to
form the blood-brain barrier and form tunnels of cells to allow
migration of neuroblasts in the RMS (Alvarez-Buylla and Garcia-
Verdugo, 2002; Kandel et al., 2000); they can also proliferate as
reactive astrocytes that can induce gliosis or glial scarring
(reviewed in Pekny and Nilsson, 2005). The Alvarez-Buylla group
has suggested that GFAP-positive cells in the SVZ are multipo-
tent neural stem cells in both humans (Sanai et al., 2004) and
adult mice (Doetsch etal., 1999). Wetested whether human cells
of astrocytic lineage, defined by expression of GFAP, would
participate in the same fashion as mouse astrocytes in our
transplanted Ppt1?/?/NSCID mice. Immunohistochemical stain-
ing of serial brain sections with SC121 revealed robust human
engraftment (Figure 3E), whereas a human-specific GFAP anti-
body (hGFAP;SC123) showed limited staining and the few posi-
tive cells were restricted to the SVZ, RMS, and corpus callosum,
with only occasional cells detected in the cortex (Figure 3F).
When sections were costained with SC123 and an antibody to
b-dystroglycan, a marker for mouse blood vessels, human
D Undifferentiated neural cells
Figure 2. Engraftment of hCNS-SCns Transplanted into Ppt1?/?/
NSCID Mouse Brains
Immunoperoxidase staining with mAb SC121 (brown) counterstained with
methyl green (turquoise). SC121 immunostaining of human cells, which
morphologically appear to be: (A) neurons in the granule layer of the olfactory
bulb (arrowhead), (B) fibrous astroctyes (arrow) between the striatum and
corpus callosum, (C) oligodendrocytes in the white matter of the cerebellum,
and (D) undifferentiated neural cells in the cortex.
12 μ μm
SCns within the Ppt1?/?/NSCID Mouse CNS
(A) Engrafted human cells express Nestin in the SVZ.
(B) Confocal image of the olfactory bulb stained with SC121 (red) and anti-
MAP2 (green) revealing that the progeny of human cells differentiate along
a neuronal lineage and extend both axons and dendrites.
(C and D) The proliferative status of cells transplanted into the lateral ventricle
and the anterior cortex of a neonatal Ppt1?/?/NSCID mouse was examined
140 days post-transplant using a mAb to Ki67.
(C) Robust engraftment of human cells was detected in the SVZ.
(D) In an adjacent section from the same brain, small numbers of Ki67-positive
cells were detected in the SVZ.
(E and F) Detection of human cells of astrocyte lineage in the Ppt1?/?/NSCID
(E) Immunostaining with human-specific SC121 mAbs reveals robust human
cell engraftment in SVZ, RMS, corpus callosum, and cortex.
(F) Immunostaining with the human-specific GFAP (SC123) mAb reveals
human GFAP-positive cells in the SVZ and RMS of hCNS-SCns with a few
positive cells in the corpus callosum.
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
312 Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc.
GFAP-positive cells with astrocytic endfeet processes appeared
in contact with a mouse b-dystroglycan-positive blood vessel
(Figure 4A; Figures S3A and S3B). This morphology is character-
istic of astrocytes contributing to blood-brain barrier formation
Further, we tested whether the transplanted human cells
immunofluorescence staining was performed with SC121 and
a GFAP antibody that reacts with both mouse and human
GFAP. We counted approximately 1000 GFAP-positive cells
and 200 SC121-positive human cells in the same cortical field
of the cortex of the transplanted Ppt1?/?/NSCID brain, revealing
only a very small proportion of cells that were double positive
(SC121+GFAP+) human astrocytes (Table S3), with a few
double-positive cells also detected in the fimbria and the SVZ.
These results are similar to those observed in the hCNS-SCns-
transplanted NOD-SCID mice (Guzman et al., 2007); based
on these observations, it does not appear that transplanted
hCNS-SCns cells contribute to gliosis in the brains of Ppt1?/?/
NSCID. Oligodendrocytes, another type of glial cell, were
present in the white matter tracts of the corpus callosum and
the anterior commissure, as defined by double fluorescence
staining with an oligodendrocytes marker CC-1APC (Figure 4B).
to examine whether they contributed to the architecture of the
migratory pathway to the olfactory bulb. Sections were stained
with SC121, SC123 (hGFAP), and DCX. Confocal analysis
of mouse astrocytes in the RMS where we observed migrating
neuroblasts, as defined by DCX staining (Figure 4C; Figures
S4A–S4C). Some of the migrating DCX-positive neuroblasts
were SC121+human cells (Figure 4D). When these migrating
human cells were examined in the olfactory bulb, a quantitative
analysis showed that approximately 17.4% (105 of 603 cells
within three mouse brains) of the human cells expressed DCX
(Figure S4D). In contrast, we did not observe any human cells
that expressed DCX in the cortex of these samples. Moreover,
some of the cells within the olfactory system differentiated
further to mature neurons, as defined by MAP2 staining (Fig-
ure 3B). In conclusion, even within a neurodegenerative environ-
and human astrocytes contributed to the architecture of both
the blood-brain barrier and the RMS.
hCNS-SCns Do Not Induce Additional Inflammation
in the Brain of Ppt1?/?/NSCID Mice
Inflammation has been identified as an important early event in
the pathological cascade that eventually leads to the degenera-
tion observed in the brains of Ppt1-deficient mice (Bible et al.,
2004; Kielar et al., 2007). Transplants of human neural cells
tion in inflammation in these brains (Lee et al., 2007), and we
wanted to see whether hCNS-SCns transplants exerted a similar
effect on inflammation within the Ppt1?/?/NSCID mouse brain.
Iba-1 staining is commonly used to monitor the presence of
microglia and their activation. As expected, more activated
microglia were present in the brains of the nontransplanted
Ppt1?/?/NSCID mice (Figure 4E) compared to the brains of
heterozygous Ppt1+/?/NSCID controls (Figure 4G). Qualitative
comparisons between age-matched, nontransplanted (Fig-
ure 4E) and transplanted (Figure 4F) Ppt1?/?/NSCID brains
showed there were no obvious changes in morphology or distri-
cells and therefore cannot trigger the immune cascade. Never-
theless, these mice still have monocyte/macrophages, neutro-
phils, or low levels of NK cells, all parts of the innate immune
system that can exert an inflammatory process. However, our
data reveal that transplantation of human cells into NOD-SCID
A hGFAP/β β-dystroglycan
E Not transplanted F Transplanted
4.6 μ μ
5.8 μ μm
.8 μ μm
5.8 μ μm
Figure4. Site-AppropriateEngraftmentandDifferentiation ofhCNS-
SCns in Ppt1?/?/NSCID Mice
(A) Confocal image of a blood vessel stained with SC123 (red), anti-mouse
b-dystroglycan (green), and Hoechst 33345 counter staining (Hoechst, blue).
Human GFAP-positive astrocytes associate with mouse blood vessel as their
end-feet apposed to mouse b-dystroglycan-positive blood vessels.
(B) Human cells of oligodendrocyte lineage (arrow) stained with SC121 (red),
CC1-APC (green), and Hoechst (blue).
(C) Confocal image of the RMS stained with DCX (green), SC121 (red), and
SC123 (magenta). Progeny of human cells express GFAP and form a tunnel
(arrows) for neuroblast migration (DCX-positive).
(D) Some of the DCX-positive neuroblasts (green) are also SC121 positive
(red), indicating they are of human origin (yellow, arrows).
(E–G) Immunoperoxidase staining with Iba-1 (brown) to detect host microglia
in brain sections from (E) nontransplanted Ppt1?/?/NSCID mice, (F) hCNS-
SCns transplanted Ppt1?/?/NSCID, and (G) Ppt1 heterozygous NSCID control
mice. No obvious difference in Iba-1 staining was observed between treat-
ment groups. Even at the bolus of the human injection core, there was no
evidence of a high density of Iba-1-positive cells with the morphology of acti-
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc. 313
mice did not result in any appreciable activation of host micro-
glia, demonstrating that hCNS-SCns do not induce additional
inflammation in the brains of Ppt1?/?/NSCID mice.
Transplanted hCNS-SCns Express Intracellular PPT1
in the Brains of Ppt1?/?/NSCID Mice
Having tested our cross-correction hypothesis in vitro, we next
examined whether the hCNS-SCns can provide PPT1 to the
brain of Ppt1?/?/NSCID mice in a cell dose-dependent manner.
As expected, at 26 weeks of age, the Ppt1?/?/NSCID brain dis-
played widespread accumulation of lipofuscin deposits, mostly
in neurons. Characteristically, these deposits displayed auto-
fluorescence, which can be easily identified (Figures 5A–5C,
arrowhead). SC121-positive human cells (red), which do not
contain any autofluorescent storage material, were present in
the vicinity of host neurons (arrow, Figures 5A–5C). These trans-
planted cells expressed PPT1 (green), which was distributed in
small granules presumably in the lysosomes of their cell bodies
and processes. The engrafted human cells appear to sustain
PPT1 expression after transplantation. However, because
double immunofluoresce staining is hampered by neighboring
autofluorescent cells, detection of PPT1 at the single cell level
is challenging. Additional examples of PPT1-positive human
cells are shown in Figure S5A as well as immunoperoxidase
staining with the PPT1 antibody (Figure S5B). Nevertheless, we
could not evaluate whether PPT1 could be taken up by host
cells in vivo.
The residual level of PPT1 activity present in cell lines from
CLN1 patients correlates directly with the age of onset, rate of
progression, and severity of their disease phenotype (Das et al.,
1998). For example, CLN1 mutations associated with ‘‘juvenile
onset’’ showed detectable enzyme activity ?1%–2% of normal
(Das et al., 2001). Although a limited data set, the results suggest
that low levels of PPT1 delay the onset of disease manifestation.
Based on this study, we hypothesized that providing some level
of PPT1 into the brains of Ppt1?/?/NSCID mice could alter their
To assess human cell engraftment and PPT1 activity, mice
transplanted bilaterally with hCNS-SCns were sacrificed and
their brains were dissected into hemispheres. The left hemi-
spheres were processed for immunohistological staining and
the right hemispheres were processed to measure PPT1 levels.
Inapilot study,neonatal Ppt1?/?/NSCIDmice weretransplanted
with 1.5 3 105cells per hemisphere (3.0 3 105/total per brain)
and evaluated at 18 weeks after transplant. Brain extracts from
four of these animals had ?0.8% ± 0.16% (SEM) of normal
enzyme activity. In this study, the small size of the neonatal
mouse brain (6 mm between anterior end of cortex and the
posterior end of colliculus) limited the volume of cells that could
be physically introduced at one time. Therefore to determine the
effects of grafts with increasing numbers of cells, we designed
a strategy that allowed administration of hCNS-SCns at multiple
locations at different stages of disease progression. In the
following quantitative studies, mice targeted for a low cell dose
received transplants twice, first as neonates (P0-P1) and sub-
sequently as juveniles. Mice receiving high cell doses were
transplanted three times, as neonates, postnatally (P7), and
as juveniles (see Supplemental Experimental Procedures and
Table S4 for cell doses and transplantation scheme).
at 160–188 days after transplant, and PPT1 enzyme levels were
quantified. Consistent with the original description of these mice
(Gupta et al., 2001), no detectable PPT1 was measured from
control nontransplanted Ppt1?/?/NSCID mice. In contrast,
a significantly elevated level of PPT1 activity was detected in
mice receiving either a low or high dose of cells (p < 0.001
between the nontransplanted and the low dose, and p < 0.01
between low and high cell doses). Mice receiving a low trans-
plant dose (15–20 3 105; double transplant, n = 7) showed an
average of approximately 4.4% of normal enzyme levels; mice
receiving a high cell dose (28 3 105; triple transplant, n = 11)
Storage material (area of autofluorescence)
% PPT1 activity in transplanted brain
5.8 μ μm
Figure 5. Progeny of Transplanted Human Cells Produce PPT-1 and
Reduce Autofluorescent Lipofuscin Deposits
(A–C) Confocal images of Ppt1?/?/NSCID mouse brain sections stained with
PPT1 antibody (green) (A) and SC121 (red) (B) and merged (C), with Hoechst
of lipofuscin deposits, mostly in neurons (arrowheads). As expected, these
lipofuscin deposits autofluoresce in both green and red channels. Human
cells, which do not display autofluorescence, engraft in the vicinity of host
neurons and express PPT1 (C) in a punctate pattern, compatible with lyso-
somal expression (arrows).
(D) PPT1 activity in hCNS-SCs transplanted Ppt1?/?/NSCID mouse brains
correlates with the hCNS-SCns cell dose. PPT1 activity was measured from
whole brain extracts from 160- to 188-day-old Ppt1?/?/NSCID mice. hCNS-
SCns were transplanted in either low (red, n = 7) or high (green, n = 11) cell
doses. Enzyme activities are expressed as the mean ± SEM percentage of
PPT1 activity of wild-type brain, which was set to 100%. In untreated Ppt1?/?/
found in wild-type mouse brain (n = 12). This background was subtracted from
all samples to set the PPT1 activity in the nontransplanted Ppt1?/?/NSCID to
zero. Nontransplanted versus low-dose transplanted, p < 0.001; low versus
high dose, p < 0.01.
(E) Quantification of the area of autofluorescent deposits (mm2) as measured
by confocal microscopy and thresholding image analysis expressed as
mean ± SEM area of autofluorescence. The area of autofluorescence was
reduced in five different brain regions of Ppt1?/?/NSCID mice after hCNS-
SCns transplantation. Four nontransplanted and three transplanted age-
matched Ppt1?/?/NSCID animals were analyzed. *p < 0.0001.
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
314 Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc.
had a correspondingly higher level, up to 6.9% PPT1 at 160–
188 days after transplant (Figure 5D). These data demonstrate
that hCNS-SCns transplants provide measurable and long-
lasting levels of PPT1 activity in Ppt1?/?/NSCID mice. These
data do not prove that clinically meaningful levels of PPT1
reached host cells, but nevertheless these results encouraged
us to undertake a histopathological analysis to evaluate the
biological consequences of enzyme delivery via hCNS-SCns
transplants in this preclinical model of disease.
Transplanted hCNS-SCns Contribute to a Decrease
in the Autofluorescent Lipofuscin Material in the Brains
of Ppt1?/?/NSCID Mice
One hallmark of both the Ppt1?/?/NSCID mice and human INCL
pathology is the accumulation of lipofuscin deposits throughout
thebrain. Inmice,autofluorescence (AF)loadisroutinelyusedas
a marker of disease progression (e.g., Griffey et al., 2004), and
we investigated whether transplanted hCNS-SCns would lead
to a reduction in the host AF load. Quantitative confocal analysis
of AF load in untransplanted Ppt1?/?/NSCID mice (n = 4 at 162–
169 day of age) was compared to that in transplanted age-
matched low-dose (1.5–1.8 3 106cells; n = 3 at 167–170 days
of age) mice. AF load was quantified as the average area that
was autofluorescent per image field in scanned confocal images
from the rostral and caudal portions of the cortex, the thalamus,
the CA1 region of the hippocampus, and the cerebellum (see
schematic sampling and analysis in Figure S5C and Supple-
mental Experimental Procedures). A significant reduction in AF
load was observed in the low-dose group of mice for all areas
of the brain measured (p = 0.0001) (Figure 5E). The reduction
of storage material ranged from 31% in the cortex (CCtx) to
37% in the thalamus and more than 50% in the hippocampus
and cerebellum when compared to the nontransplanted group.
AF load was also measured in the high-dose cohort (2.8 3 106
cells; n = 3 at 188 days of age). No age-matched untransplanted
controls were available because Ppt1?/?/NSCID become mori-
of mice aged 162–169 days was used for control. Significant
reductions in storage deposits were also observed in all brain
regions in the high-dose cohort, except for the thalamus (see
individual p values in Figure S5D). We know that disease
progression and lipofuscin accumulation continue into the
terminal stage of nontransplanted Ppt1?/?/NSCID mouse, so it
is likely that this comparison represents an underestimate of
the AF load reduction in the high-dose cohort, because these
mice were 17–26 days older that the control group. Neverthe-
less, transplantation of hCNS-SCns into Ppt1?/?/NSCID signifi-
cantly reduced AF storage material. Although this reduction of
storage material was only partial, compared to control normal
mice, we nevertheless explored whether hCNS-SCns trans-
plants resulted in any downstream biological effects by exam-
ining neuronal survival.
Transplanted hCNS-SCns Prolong the Survival
of Host Ppt1-Deficient Neurons and Delay
the Loss of Motor Coordination
Having demonstrated in vivo production of PPT1 by hCNS-SCns
and the concomitant reduction of AF storage material, we next
examined whether these findings correlated with a neuroprotec-
tive effect on Ppt1-deficient host cells. Host neuronal survival
was examined in nontransplanted control and transplanted
Ppt1?/?/NSCID mice ranging in age from 165 to 188 days, via
staining for markers expressed by mature neurons, including
calbindin, calretinin, and neural nuclear antigen (NeuN). NeuN
is a DNA-binding protein expressed in the nuclei and perinuclear
cytoplasm of most postmitotic neurons of mice, with the excep-
tion of Purkinje cells, mitral cells, and photoreceptors. NeuN
expression was widely distributed in the cortex and hippo-
campus, and we identified an antibody to NeuN that preferen-
tially stained mouse neurons over human neurons. SC121 and
NeuN antibodies were used to double label brain sections
analyzed by confocal microscopy. As expected, none of the
SC121-positive human cells found in the cortex or hippocampus
colocalized with NeuN staining (Figure S6A), confirming that,
under these conditions, NeuN staining is restricted to mouse
cells. This NeuN staining of mouse cells was used to assess
mice (Figure 6A) and from nontransplanted (NT, Figure 6B),
low-dose (Figure 6C), and high-dose (Figure 6D) Ppt1?/?/NSCID
mice. There was significant loss of CA1 hippocampal neurons in
Ppt1-deficient mice (Bible et al., 2004), and, as expected, the
CA1 of nontransplanted Ppt1?/?/NSCID mice displayed greatly
reduced NeuN staining compared to NOD-SCID control mice.
In contrast, Ppt1?/?/NSCID mice that received either a low or
high cell dose displayed markedly more NeuN+CA1 neurons,
indicating thatthe humancells wereprotecting thehost neuronal
we used image analysis to calculate the area of NeuN immuno-
reactivity in the CA1 region of each group and normalized these
values to data from NOD-SCID mice (Figure 6E; Table 1). Only
8% of host CA1 region neurons survived in the nontransplanted
Ppt1?/?/NSCID controls, compared to NOD-SCID animals. In
contrast, all hCNS-SCns transplanted mice had a significantly
higher level of NeuN-positive neurons, with an average of 33%
and 57% CA1 neurons surviving in the low and high cell dose
groups, respectively (Figure 6E, p < 0.001).
To determine whether these neuroprotective effects extended
yses of the survival of hippocampal CA2 and CA3 neurons and
cortical neurons (Figure S6B). In Ppt1?/?/NSCID mice receiving
a low cell dose, NeuN-positive cells were found to be at 92%
of normal levels and, for the high cell dose at 97% of normal
(p < 0.001) (Table 1), with significantly more (p < 0.05) NeuN-
positive cortical neurons in these high cell dose animals (77%
of normal, Table 1). A clear trend was observed correlating
higher survival of host neurons with increased hCNS-SCns cell
dose. Thus, hCNS-SCns can protect host neurons from degen-
eration in brains of Ppt1?/?/NSCID mice, presumably by cross
ciated motor deficits and reveals a progressively impaired
performance in Ppt1?/?mice (Griffey et al., 2006). An acceler-
ating rotarod protocol was used to assess the performance of
Ppt1?/?/NSCID mice either nontransplanted (NT?/?) or trans-
planted (Txp?/?) with hCNS-SCns. Heterozygous sibling groups,
NT+/?and Txp+/?, showed no decline in performance at all
time points. Repeated-measures ANOVA yielded a significant
interaction effect between these groups over time (p < 0.001,
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc. 315
F = 7.35) (Figure 6F). Bonferroni posttests demonstrated signifi-
cance differences between NT?/?and Txp?/?mice at 17 (p <
0.001) and 18 (p < 0.05) weeks of age. Moreover, a significant
deficit on the rotarod was observed in NT?/?compared to NT+/?
mice by week 17, whereas the Txp?/?group did not show a
significant decline until week 18, indicating that hCNS-SCns
transplantation delayed the loss of motor coordination.
primarily peripheral organs that can be treated effectively by
soluble adjuvant ERT. Unfortunately, more than half of the
LSDs affect the CNS and these are recalcitrant to ERT because
of the inability of the enzyme to cross the blood-brain barrier
(Sly and Vogler, 2002). To overcome this problem, one possible
treatment strategyisdirect administration of themissing enzyme
into the CNS by either transplantation of donor cells that secrete
the enzyme or by gene therapy targeting of the host cells. We
have chosen to test transplants of long-term self-renewing
human neural stem cell in this paradigm because these trans-
plants can theoretically provide life-long production of the
missing enzymes. Here, we provide proof of principle for this
approach with hCNS-SCns supplying long-lasting delivery of
PPT1 and neuroprotection in the Ppt1?/?/NSCID mouse model
of INCL, which is a profoundly disabling LSD.
Banked human neural stem cells are ideal cell candidates for
treatment of LSD, because they can be cryopreserved and
tested to ensure their quality for clinical use. Highly purified
CD133+, CD24?/lohCNS-SC directly isolated from fetal brain
were characterized at the clonal level for their ability to self-
renew and differentiate into neurons and glia (Uchida et al.,
2000). These cells are selectively purified further by expansion
in serum-free defined conditions to generate cell banks. These
nontumorigenic, non-genetically modified hCNS-SCns have
been transplanted into the brains of NOD-SCID mice and re-
sulted in reproducible engraftment and extensive migration
only within the CNS (Uchida et al., 2000). Unlike embryonic
stem cells, which can form teratomas, these brain-tissue-
derived somatic stem cells do not form tumors upon transplan-
tation into the CNS (i.e., orthotopic sites) of NOD-SCID mice,
immunosuppressed mice, rats, and nonhuman primates. We
have transplanted more than 3000 immunodeficient mice to
date and never found tumor formation from these cells (N.U.,
S.J.T., D.H., E. Apilado, K. Eckert, unpublished data). Also, no
evidence of tumor formation was observed when hCNS-SCns
were transplanted into the spinal cord of NOD-SCID mice or
subcutaneously into flanks of nude mice. Taken together, our
data reveal that hCNS-SCns can safely deliver therapeutic levels
of PPT1 throughout the CNS, suggesting that this approach
might be broadly applicable to other LSDs associated with
enzyme deficiency, such as LINCL (CLN2).
The neuroprotection we observed in Ppt1?/?/NSCID mice
most probably resulted from local secretion of PPT1 by trans-
planted cells and its uptake by neighboring Ppt1-deficient cells.
Both PPT1 activity and neuroprotection were correlated with the
dose of hCNS-SCns and/or number of graft sites. Moreover,
transplanted Ppt1?/?/NSCID mice retained significantly better
coordinated motor performance than nontransplanted mutant
mice. However, we cannot exclude alternate mechanisms for
neuroprotection, including elaboration of some neurotrophic
factors, and/or modulating the inflammation associated with
neurodegeneration, or replacement of host neurons and/or glia.
% Host Neurons in CA1
B Ppt1-/-/NScid Not transplanted
D High Cell Dose
A Ppt1+/+ (NOD-Scid)
C Low Cell Dose
Time on Rotarod (sec)
1415 1617 181920
Figure 6. hCNS-SCns Protect Host Cell Neurons in Ppt1?/?/NSCID
Mice and Delay the Loss of Motor Coordination
Human cells were transplanted into brains of Ppt1?/?/NSCID mice as
described in the Supplemental Data. Brain sections were stained with mAb
(A) The hippocampus of Ppt1+/+Nod-SCID mice shows representative NeuN
staining of the unaffected CA subfields (CA1-CA3).
(B) At 6 months of age, the hippocampus of nontransplanted Ppt1?/?/NSCID
mice display marked neuron loss that is most pronounced in CA1.
(C and D) Transplanted hCNS-SCns provided neuroprotection in each hippo-
campal subfield with more pronounced neuroprotection after high cell dose
transplants (D) than low cell dose transplants (C).
(E) Quantification of NeuN-immunoreactivity by SIS image analysis in the CA1
subfield of the hippocampus. Results of the mean ± SEM area of NeuN immu-
noreactivity are plotted in individual experimental groups. Abbreviations: NT,
nontransplanted (blue, n = 9); low, low cell dose transplanted (red, n = 6);
high, high cell dose transplanted (green, n = 6). Percentages are normalized
against untreated NOD-SCID mice.
(Txp?/?, red, n = 14) was significantly better than nontransplanted Ppt1?/?/
NSCID mice (NT?/?, blue n = 8) at 17 weeks (p < 0.001) and 18 weeks (p <
0.05) of age. Sibling controls of transplanted (Txp+/?, orange, n = 12) and non-
transplanted (NT+/?, black, n = 3) Ppt+/?/NSCID mice performed similarly (n.s.)
at all the time points evaluated and demonstrated no evidence of impaired
performance on this task.
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
316 Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc.
been made into a mouse model of Sandhoff disease (Lee et al.,
2007). These transplanted neural cells migrated extensively
and appeared to cross-correct the Hexb?/?mouse brain and
reduced CNS inflammation (Lee et al., 2007). These data, taken
together with our data from Ppt1-deficient mice, indicate that
cell transplantation is a viable treatment for LSDs with CNS
The potential ofa genetherapy strategy with adenoassociated
virus (AAV) to treat LSD with CNS involvement has been tested
extensively in different animal models, including MPSIIV (Frisella
et al., 2001), Niemann-Pick (Passini et al., 2007), CLN1 (Griffey
et al., 2004, 2005, 2006), and CLN2 (Cabrera-Salazar et al.,
viding cells, lifelong provision of lysosomal enzymes in the CNS
is possible, depending on the life span of the cells infected and/
or whether AAV also infects stem cells. One limitation of this
approach is that enzyme production is restricted to the AAV
injection sites, because the infected cells (e.g., neuronal cells)
do not migrate.
In patients with Krabbe disease, another LSD with CNS
involvement, it has been demonstrated that performing cord
blood transplants at infancy, before disease manifestation, is
crucial for obtaining clinical benefit (Escolar et al., 2005, 2006).
Taken together, these studies provide sufficient evidence to
believe that CLN1 patients could be treated most effectively by
administrating the missing PPT1 via hCNS-SCns grafts into the
affected brain early in disease progression. This early interven-
tion would most likely result in better neuroprotection. In this
paper, all groups were transplanted as neonates with hCNS-
SCns, although some mice received additional transplants
1 week later or as young adults. It is yet to be determined
whether hCNS-SCns have a similar neuroprotective effect
when they are transplanted into adult INLC mice.
The engrafted human cells of the glial lineage were found to
have a normal tissue distribution and did not overgrow or
contribute to astrocytic gliosis. Reactive astrocytosis has classi-
cally been thought to accompany or follow neuron loss, but in
long before the onset of neuron loss (reviewed in Castaneda
et al., 2008). This phenotype is also evident in multiple forms of
NCL (Oswald et al., 2005; Pontikis et al., 2004), although the
precise nature and timing of these events differs between forms
of this disorder. In the Ppt1-deficient mice used in this study,
localized astrocytosis precedes the progressive loss of relay
neurons within individual thalamic nuclei, which was itself
followed by a wave of microglial activation (Kielar et al., 2007).
protective response, but these data are consistent with the idea
that astrocytosis may also provide a sensitive marker of ongoing
hCNS-SCns did not promote further inflammatory changes, as
evaluated qualitatively by comparing the endogenous microglia
between nontransplanted and transplanted brains. In addition,
we demonstrated that a significant reduction in lipofuscin
accumulation (by AF) was observed in the thalamus of trans-
planted animals. An interesting possibility would be if human
PPT1 could also be provided by anterograde axonal transport
(Griffey et al., 2005), although it remains unclear how secreted
PPT1 may be transported to host cells.
This study extends the surprising finding that transplanted
hCNS-SCns integrate into the mouse brain and respond to
the mouse brain structure with site-appropriate migration and
differentiation. For example, in this INCL mouse model, many
human astrocytes were found in the RMS, where they integrated
and formed an astrocytic tunnel surrounding migrating mouse
and human neuroblasts. Taken together, these observations
suggest that hCNS-SCns can respond to local microenviron-
mental cues, even during neurodegeneration or injury.
ical models. In stroke models, hCNS-SCns preferentially migrate
toward the lesion border and differentiate toward the neuronal
lineage (Cummings et al., 2005; Guzman et al., 2007; Kelly
et al., 2004), whereas in myelin-deficient shiverer mice, they
differentiate into myelinating oligodendrocytes (Cummings
et al., 2005). In models of spinal cord injury, hCNS-SCns inte-
grate, differentiate to oligodendrocytes, and remyelinate host
axons or differentiate to form synapses. Most important is the
Table 1. hCNS-SCns Transplants Provide Neuroprotection to Host Hippocampal and Cortical Neurons in Ppt1?/?/NSCID Mice
Not transplanted mean2,902 ± 34117,035 ± 1,024 400,606 ± 24,449
% 8% 47%59%
Low cell dose mean12,186 ± 784 33,270 ± 2,222500,524 ± 16,144
High cell dosemean 21,002 ± 1,03534,947 ± 2,226 523,952 ± 16,988
*p < 0.001, **p < 0.05 by ANOVA, compared to untransplanted Ppt1?/?.
The Soft Imaging System image analysis software was used to quantify the area of NeuN-immunoreactivity in the CA1, CA2/3 subfields of the hippo-
campus and cortex. Results of the mean ± SEM area (mm2) of NeuN staining in nontransplanted (n = 9), low cell dose (n = 6), and high cell dose (n = 6)
CA2/3 36,183 ± 2,220; and cortex 679,231 ± 55,656.
Quantitative analyses of the CA2/CA3 subfields of nontransplanted Ppt1?/?/ NOD-SCID mice show only ?47% of the number of NeuN-positive cells
comparedto wild-typemice.92%and97%ofnormal levelsofNeuN-positivecellswere detectedintheCA2/3subfieldsofthehippocampus inthelow
and high cell dose transplanted Ppt1?/?/NSCID mice, respectively. In the cortex, the number of NeuN-positive cells was also analyzed. Nontrans-
planted Ppt1?/?/NSCID mice showed a reduction in cortical neurons to 59% of the levels found in Ppt1+/+/NOD-SCID mice. Ppt1?/?/NSCID mice
that received a high cell dose of hCNS-SCns had significantly more (p < 0.05) NeuN-positive cells (77% of normal).
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc. 317
demonstration that locomotion improvement directly correlates
with the survival of human cells; loss of human cells results in
loss of locomotor function (Cummings et al., 2005). Further
apeutic potential of hCNS-SCns in NCL disease and other
diseases ortraumaticinjuries. FordiseasessuchasLSDs,where
a genetic mutation leads to a specific deficiency resulting in
tissue destruction in peripheral organs and the brain, both
systemic and CNS treatments will be required. Combination
therapies might include enzyme infusion throughout life to both
sites, hematopoietic cell transplants for the body and hCNS-
SCns transplants for the brain, viral-mediated gene therapy to
both, or some combination of these approaches. Whatever
methods in cell, gene, and protein therapies prove to be effica-
cious, these approaches will hopefully bring medical break-
throughs for these previously untreatable diseases. As a first
step toward this goal, our data support our rationale for a phase I
clinical trial in patients with INCL and LINCL.
hCNS-SCns were prospectively isolated from fetal brain tissue (16–20 gesta-
tional weeks) by flow cytometry and grown as neurospheres (hCNS-SCns), as
described previously(Cummingsetal.,2005; Uchidaet al.,2000).hCNS-SCns
at passages 8–10 were used for the transplantation studies, having been
demonstrated as karyotypically normal.
Generation of Ppt1?/?/NSCID Mice and Transplantation
All animal procedures were approved by the Institutional Animal Care and Use
Committee at StemCells, Inc. Ppt1-deficient mice, described previously (Bible
et al., 2004; Gupta et al., 2001), were provided by Sandra Hofmann (University
of Texas Southwestern, TX). Ppt1?/?mice were backcrossed onto the NOD-
SCID background (i.e., Ppt1?/?/NSCID), with all mice used in this study being
from six backcross generations (N6) in most of the experiments, with the
exception of rotarod testing, which used N10 generations of backcrosses.
To achieve the desired cell dose levels, animals received transplants either
solely as neonates or at multiple times. Details of the transplantation scheme,
dose, and schedule are described in the Supplemental Experimental Proce-
dures and Table S4.
Determination of PPT1 Activities
PPT1 assay were performed as described previously (Page et al., 1993; Sohar
et al., 2000; Vines and Warburton, 1998; Voznyi et al., 1999). PPT1-specific
quenched fluorogenic substrates were used to assess the enzyme activity
within protein extracts prepared from cell lines or tissues. To measure PPT1
activity in brains of transplanted mice, homogenates from the right hemi-
spheres of Ppt1?/?/NSCID mice were processed.
Immunohistochemical Analysis of Transplanted Mouse Brains
Transplanted mice were anesthetized and perfused with PBS followed by 4%
paraformaldehyde (PFA). Mouse brains were sectioned sagittally with a micro-
tome(Leica SM2400,Nussloch,Germany)at40mmthicknessand stainedwith
various antibodies to reveal the distribution of transplanted cells and subse-
For confocal immunofluorescence microscopy, imaging was performed on
a Leica SP2 AOBS microscope (Leica Microsystems, Wetzlar, Germany). For
characterization of lineage analysis, the resulting image stacks were analyzed
with Volocity Software (Improvision, Coventry, UK). To confirm the colocaliza-
tionofthelabeled antigenswithinacell, thefluorescentstainingwasinspected
in the z-dimension via the orthogonal view tool.
Quantification of Autofluorescent Lipofuscin Deposits
As described previously, confocal microscopy was used to quantify the
amount of autofluorescent storage material in Ppt1-deficient mice (Griffey
et al., 2004, 2005, 2006). All image acquisition and analyses were performed
blind to treatment (Supplemental Experimental Procedure). Data were
analyzed via one-way ANOVA followed by post-hoc Bonferroni analysis
(SPSS, V 12.0). Differences between control and treated (transplanted) groups
were considered significant if p < 0.05.
Quantitation of NeuN-Positive Neurons in PPT1?/?/NSCID
Every sixth 40 mm sagittal section was stained immunohistochemically with
mAb against NeuN. All histological sections utilized in this study were imaged
with an Olympus BX61 microscope. Quantitative image analysis was per-
formed with the Soft Imaging System (SIS) GmbH Biological Suite with Scope-
view software. Detailed procedures for image acquisition, defining region of
interest, and quantitative procedures are described in the Supplemental
Data. All data points were analyzed by one-way ANOVA followed by the Bon-
Ppt1?/?/NSCID and their sibling Ppt1+/?/NSCID NSCID mice from N10 gener-
ations of backcrosses received transplants of hCNS-SCns as described in
Table S4. Motor coordination was tested with a Rotarod apparatus (Accuscan
from 0 to 15 rpm) with performance scored as time stayed on the rotarod (max
60s)inthree trialspersession.Individual micewithknowncohortsweretested
weekly until Ppt1 mutant mice could no longer perform the test. Data were
analyzed by repeated-measures ANOVA followed by post-hoc Bonferini
post t test comparisons of individual time points of rotarod testing for Ppt1/
Supplemental Data include Supplemental Experimental Procedures, six
figures, and four tables and can be found with this article online at http://
We would like to thank Drs. Sandy Hofmann, Mark Sands, Krystyna Wisniew-
ski, Aileen Anderson, Brian Cummings, and Tonya Bliss for the generous gifts
of the Ppt1?/?mice, the antibody against PPT1, and fibroblasts from Ppt1?/?
mice CLN1 and CLN2 patients and expert advice. S.J.T., Y.J., M.D., A.C.,
M.R., D.H., R.T., S.H., A.S.T., and N.U. are employees of, J.D.C. and W.M.
are past consultants of, and F.H.G. and I.L.W. are scientific founders of
Received: November 28, 2007
Revised: September 23, 2008
Accepted: May 20, 2009
Published: September 3, 2009
Abbott, N. (2002). Astrocyte-endothelial interactions and blood-brain barrier
permeability. J. Anat. 200, 527.
Alvarez-Buylla, A., and Garcia-Verdugo, J.M. (2002). Neurogenesis in adult
subventricular zone. J. Neurosci. 22, 629–634.
Bible, E., Gupta, P., Hofmann, S.L., and Cooper, J.D. (2004). Regional and
cellular neuropathology in the palmitoyl protein thioesterase-1 null mutant
mouse model of infantile neuronal ceroid lipofuscinosis. Neurobiol. Dis. 16,
Cabrera-Salazar, M.A., Roskelley, E.M., Bu, J., Hodges, B.L., Yew, N., Dodge,
J.C., Shihabuddin, L.S., Sohar, I., Sleat, D.E., Scheule, R.K., et al. (2007).
Timing of therapeutic intervention determines functional and survival
outcomes in a mouse model of late infantile batten disease. Mol. Ther. 15,
Castaneda, J.A., Lim, M.J., Cooper, J.D., and Pearce, D.A. (2008). Immune
system irregularities in lysosomal storage disorders. Acta Neuropathol.
(Berl.) 115, 159–174.
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
318 Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc.
Cummings, B.J., Uchida, N., Tamaki, S.J., Salazar, D.L., Hooshmand, M., Download full-text
Summers, R., Gage, F.H., and Anderson, A.J. (2005). Human neural stem cells
differentiate and promote locomotor recovery in spinal cord-injured mice.
Proc. Natl. Acad. Sci. USA 102, 14069–14074.
Das, A.K., Becerra, C.H., Yi, W., Lu, J.Y., Siakotos, A.N., Wisniewski, K.E., and
Hofmann, S.L. (1998). Molecular genetics of palmitoyl-protein thioesterase
deficiency in the U.S. J. Clin. Invest. 102, 361–370.
Das, A.K., Lu, J.Y., and Hofmann, S.L. (2001). Biochemical analysis of muta-
tions in palmitoyl-protein thioesterase causing infantile and late-onset forms
of neuronal ceroid lipofuscinosis. Hum. Mol. Genet. 10, 1431–1439.
Doetsch, F., Caille, I., Lim, D.A., Garcia-Verdugo, J.M., and Alvarez-Buylla, A.
(1999). Subventricular zone astrocytes are neural stem cells in the adult
mammalian brain. Cell 97, 703–716.
Escolar, M.L., Poe, M.D., Martin, H.R., and Kurtzberg, J. (2006). A staging
system for infantile Krabbe disease to predict outcome after unrelated umbil-
ical cord blood transplantation. Pediatrics 118, e879–e889.
Escolar, M.L., Poe, M.D., Provenzale, J.M., Richards, K.C., Allison, J., Wood,
S., Wenger, D.A., Pietryga, D., Wall, D., Champagne, M., et al. (2005). Trans-
plantation of umbilical-cord blood in babies with infantile Krabbe’s disease.
N. Engl. J. Med. 352, 2069–2081.
Frisella, W.A., O’Connor, L.H., Vogler, C.A., Roberts, M., Walkley, S., Levy, B.,
Daly, T.M., and Sands, M.S. (2001). Intracranial injection of recombinant
adeno-associated virus improves cognitive function in a murine model of
mucopolysaccharidosis type VII. Mol. Ther. 3, 351–358.
Griffey, M., Bible, E., Vogler, C., Levy, B., Gupta, P., Cooper, J., and Sands,
M.S. (2004). Adeno-associated virus 2-mediated gene therapy decreases
autofluorescent storage material and increases brain mass in a murine model
of infantile neuronal ceroid lipofuscinosis. Neurobiol. Dis. 16, 360–369.
Griffey, M., Macauley, S.L., Ogilvie, J.M., and Sands, M.S. (2005). AAV2-medi-
ated ocular gene therapy for infantile neuronal ceroid lipofuscinosis. Mol. Ther.
Griffey, M.A., Wozniak, D., Wong, M., Bible, E., Johnson, K., Rothman, S.M.,
Wentz, A.E., Cooper, J.D., and Sands, M.S. (2006). CNS-directed AAV2-medi-
ated gene therapy ameliorates functional deficits in a murine model of infantile
neuronal ceroid lipofuscinosis. Mol. Ther. 13, 538–547.
Gupta, P., Soyombo, A.A., Atashband, A., Wisniewski, K.E., Shelton, J.M.,
Richardson, J.A., Hammer, R.E., and Hofmann, S.L. (2001). Disruption of
PPT1 or PPT2 causes neuronal ceroid lipofuscinosis in knockout mice. Proc.
Natl. Acad. Sci. USA 98, 13566–13571.
Guzman, R., Uchida, N., Bliss, T.M., He, D., Christopherson, K.K., Stellwagen,
D., Capela, A., Greve, J., Malenka, R.C., Moseley, M.E., et al. (2007). Long-
term monitoring of transplanted human neural stem cells in developmental
and pathological contexts with MRI. Proc. Natl. Acad. Sci. USA 104, 10211–
Haskell, R.E., Hughes, S.M., Chiorini, J.A., Alisky, J.M., and Davidson, B.L.
(2003). Viral-mediated delivery of the late-infantile neuronal ceroid lipofuscino-
sis gene, TPP-I to the mouse central nervous system. Gene Ther. 10, 34–42.
Hofmann, S.L., Das, A.K., Yi, W., Lu, J.Y., and Wisniewski, K.E. (1999). Geno-
type-phenotype correlations in neuronal ceroid lipofuscinosis due to palmi-
toyl-protein thioesterase deficiency. Mol. Genet. Metab. 66, 234–239.
Kandel, E.R., Schwartz, J., and Jessell, T. (2000). Principles of Neural Science
(New York: McGraw-Hill).
Kelly, S., Bliss, T.M., Shah, A.K., Sun, G.H., Ma, M., Foo, W.C., Masel, J.,
Yenari, M.A., Weissman, I.L., Uchida, N., et al. (2004). Transplanted human
fetal neural stem cells survive, migrate, and differentiate in ischemic rat
cerebral cortex. Proc. Natl. Acad. Sci. USA 101, 11839–11844.
Kielar, C., Maddox, L., Bible, E., Pontikis, C.C., Macauley, S.L., Griffey, M.A.,
Wong, M., Sands, M.S., and Cooper, J.D. (2007). Successive neuron loss in
the thalamus and cortex in a mouse model of infantile neuronal ceroid lipofus-
cinosis. Neurobiol. Dis. 25, 150–162.
Lee, J.P., Jeyakumar, M., Gonzalez, R., Takahashi, H., Lee, P.J., Baek, R.C.,
Clark, D., Rose, H., Fu, G., Clarke, J., et al. (2007). Stem cells act through
multiple mechanisms to benefit mice with neurodegenerative metabolic
disease. Nat. Med. 13, 439–447.
Mole, S.E. (2004). The genetic spectrum of human neuronal ceroid-lipofusci-
noses. Brain Pathol. 14, 70–76.
Oswald, M.J., Palmer, D.N., Kay, G.W., Shemilt, S.J., Rezaie, P., and Cooper,
J.D. (2005). Glial activation spreads from specific cerebral foci and precedes
neurodegeneration in presymptomatic ovine neuronal ceroid lipofuscinosis
(CLN6). Neurobiol. Dis. 20, 49–63.
Page, A.E., Fuller, K., Chambers, T.J., and Warburton, M.J. (1993). Purification
and characterization of a tripeptidyl peptidase I from human osteoclastomas:
evidence for its role in bone resorption. Arch. Biochem. Biophys. 306, 354–
Passini, M.A., Bu, J., Fidler, J.A., Ziegler, R.J., Foley, J.W., Dodge, J.C., Yang,
W.W., Clarke, J., Taksir, T.V., Griffiths, D.A., et al. (2007). Combination brain
and systemic injections of AAV provide maximal functional and survival bene-
fits in the Niemann-Pick mouse. Proc. Natl. Acad. Sci. USA 104, 9505–9510.
Pekny, M., and Nilsson, M. (2005). Astrocyte activation and reactive gliosis.
Glia 50, 427–434.
Pontikis, C.C., Cella, C.V., Parihar, N., Lim, M.J., Chakrabarti, S., Mitchison,
H.M., Mobley, W.C., Rezaie, P., Pearce, D.A., and Cooper, J.D. (2004). Late
onset neurodegeneration in the Cln3?/? mouse model of juvenile neuronal
ceroid lipofuscinosis is preceded by low level glial activation. Brain Res.
Raivich, G., Bohatschek, M., Kloss, C.U., Werner, A., Jones, L.L., and Kreutz-
berg, G.W. (1999). Neuroglial activation repertoire in the injured brain: graded
response, molecular mechanisms and cues to physiological function. Brain
Res. Brain Res. Rev. 30, 77–105.
Sanai, N., Tramontin, A.D., Quinones-Hinojosa, A., Barbaro, N.M., Gupta, N.,
Kunwar, S., Lawton, M.T., McDermott, M.W., Parsa, A.T., Manuel-Garcia
Verdugo, J., et al. (2004). Unique astrocyte ribbon in adult human brain
contains neural stem cells but lacks chain migration. Nature 427, 740–744.
Sly, W.S., and Vogler, C. (2002). Brain-directed gene therapy for lysosomal
storage disease: Going well beyond the blood-brain barrier. Proc. Natl.
Acad. Sci. USA 99, 5760–5762.
Sohar, I., Lin, L., and Lobel, P. (2000). Enzyme-based diagnosis of classical
late infantile neuronal ceroid lipofuscinosis: comparison of tripeptidyl pepti-
dase I and pepstatin-insensitive protease assays. Clin. Chem. 46, 1005–1008.
Tamaki, S., Eckert, K., He, D., Sutton, R., Doshe, M., Jain, G., Tushinski, R.,
Reitsma, M., Harris, B., Tsukamoto, A., et al. (2002). Engraftment of sorted/
expanded human central nervous system stem cells from fetal brain. J. Neuro-
sci. Res. 69, 976–986.
Uchida, N., Buck, D.W., He, D., Reitsma, M.J., Masek, M., Phan, T.V., Tsuka-
moto, A.S., Gage, F.H., and Weissman, I.L. (2000). Direct isolation of human
central nervous system stem cells. Proc. Natl. Acad. Sci. USA 97, 14720–
Vines, D., and Warburton, M.J. (1998). Purification and characterisation of
a tripeptidyl aminopeptidase I from rat spleen. Biochim. Biophys. Acta 1384,
Voznyi, Y.V., Keulemans, J.L., Mancini, G.M., Catsman-Berrevoets, C.E.,
Young, E., Winchester, B., Kleijer, W.J., and van Diggelen, O.P. (1999). A
ceroid lipofuscinosis (INCL) and its variants. J. Med. Genet. 36, 471–474.
Wisniewski, K.E., Kida, E., Golabek, A.A., Kaczmarski, W., Connell, F., and
Zhong, N.(2001). Neuronalceroid lipofuscinoses:classification and diagnosis.
Adv. Genet. 45, 1–34.
Cell Stem Cell
Therapeutic Potential of Human Neural Stem Cells
Cell Stem Cell 5, 310–319, September 4, 2009 ª2009 Elsevier Inc. 319