Molecular Genetics and Metabolism 85 (2005) 203–212 www.elsevier.com/locate/ymgme
Over-expression of human lysosomal -mannosidase in mouse embryonic stem cells A.J. Robinson ¤, A.C. Crawley, J.J. Hopwood Lysosomal Diseases Research Unit, Department of Genetic Medicine, Women’s and Children’s Hospital, 72 King William Road, North Adelaide, SA 5006, Australia Department of Pediatrics, University of Adelaide, SA 5005, Australia Received 9 December 2004; received in revised form 8 March 2005; accepted 9 March 2005 Available online 26 April 2005
Abstract -Mannosidosis is a lysosomal storage disorder characterised by the lysosomal accumulation of mannose-containing oligosaccharides and a range of pathological consequences, caused by a deWciency of the lysosomal enzyme -mannosidase. One of the major features of -mannosidosis is progressive neurological decline, for which there is no safe and eVective treatment. Implantation of stem cells into the central nervous system has been proposed as a potential therapy for these disorders. We report the construction and characterisation of mouse embryonic stem cell lines for the sustained over-expression of recombinant human lysosomal -mannosidase (rhM). Two vectors (involving recombinant human -mannosidase expression driven by either the chicken -actin promoter/ CMV enhancer or by the elongation factor 1- promoter) were constructed and used to transfect mouse D3 embryonic stem cells. Selected clonal cell lines were isolated and tested to evaluate their expression of recombinant human -mannosidase. Stem cell clones transfected with the chicken -actin promoter/CMV enhancer maintained rhM expression levels throughout diVerentiation. This expression was not markedly elevated above background. In contrast, the vector incorporating the elongation factor 1- promoter facilitated substantial over-expression of -mannosidase when analysed out to 21 days of diVerentiation in stably transfected cell lines. The highest expressing cell line was found to qualitatively retain a similar diVerentiation potential to untransfected cells, and to secrete -mannosidase that could mediate a reduction in the level of oligosaccharides stored by human -mannosidosis skin Wbroblasts. These results suggest potential for the use of this cell line for investigation of a stem cell therapy approach to treat -mannosidosis. 2005 Elsevier Inc. All rights reserved. Keywords: Lysosomal storage disorder; -Mannosidase; Mannosidosis; Embryonic stem cells; Therapy; Clonal; DiVerentiation
Introduction Lysosomal storage disorders (LSD) are a group of more than 45 diVerent heritable disorders characterised by accumulation of a variety of substrates in the lysosomes of aVected individuals. LSD occur due to a functional and/or quantitative deWciency of an enzyme or enzymes required for the degradation of these substrates,
*
Corresponding author. Fax: +61 8 8161 7100. E-mail address:
[email protected] (A.J. Robinson).
1096-7192/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2005.03.005
or transport of the degradation products out of the lysosome, with broad pathological consequences [1]. -Mannosidosis is a LSD of glycoprotein catabolism (part of a subset of LSD collectively referred to as the glycoproteinoses) resulting in abnormal levels and excretion of mannose-rich oligosaccharides [2,3]. -Mannosidosis occurs in approximately 1 per 1 million live births in Australia [4]. This autosomal recessive disease is caused by a mutation(s) in the -mannosidase gene that results in a decrease or loss of lysosomal -mannosidase (E.C. 3.2.1.24) activity in the lysosomes of aVected individuals. -Mannosidase referred to henceforth is the lysosomal form of this enzyme.
204
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
One of the most diYcult aspects of -mannosidosis to treat is central nervous system (CNS) pathology, a primary feature of this disease. Although bone marrow transplantation has been shown to eVectively treat CNS pathology in some -mannosidosis patients [5], it requires a compatible donor and carries with it considerable risk as a procedure [6]. As such, there is presently no safe and eVective treatment available for all -mannosidosis patients. Although studies investigating enzyme replacement therapy in a -mannosidosis knockout mouse have shown a reduction in oligosaccharide levels in the brain [7], it is unclear whether this reduction reXects a decrease in lysosomal storage in neurons and glial cells in the -mannosidosis brain. We have a guinea pig model of this disease that closely models human mannosidosis, displaying signiWcant neurological abnormalities and CNS pathology as part of the course of their disease [8]. Enzyme replacement therapy studies in the -mannosidosis guinea pig have shown no decrease in lysosomal storage of mannosides in the capillary depleted brain following treatment (Barbara King, personal communication). Previous studies have demonstrated proof of principle for the use of stem cell-based approaches for treatment of neurological pathology in LSD [9–12]. These studies achieved therapeutic gene expression in the brain in mouse models of various LSD by using genetically modiWed adult stem cells as a production and delivery vector for the deWcient enzyme, mediating a reduction in the lysosomal storage characteristic of these diseases. Two critical issues remain with the use of adult stem cells such as neural stem cells (NSC). First, is a requirement for ongoing derivation (and the potential lack of consistency that may arise as a result), and second the availability of source stem cell material. Although other more readily obtainable adult stem cells (such as mesenchymal stem cells from bone marrow and cells from umbilical cord blood) have shown potential for treatment of neurodegenerative diseases, issues concerning plasticity and amenability to expansion and manipulation in culture still need to be resolved [13]. At present, embryonic stem (ES) cells represent one of the most promising source populations of cells for widespread therapeutic application. Theoretically, ES cells are inWnitely expandable in culture, highly amenable to genetic modiWcation, and able to produce a wide variety of clinically relevant phenotypes, including NSC and neural progenitors [14–17]. With the further development of reproducible and eYcient in vitro diVerentiation protocols for the generation of homogeneous cell populations for transplantation, ES cell-based therapies show great potential for treatment of neurological disease in LSD. Previously, mouse ES cells have been stably transfected to facilitate the over-expression and secretion of the lysosomal enzyme sulphamidase [18]; generating a useful
cell line for investigation of cell-based therapies for mucopolysaccharidosis type IIIA, another LSD with CNS pathology. Towards a similar cell-based approach for therapy in -mannosidosis, this study involved the creation of a mouse D3 ES cell line over-expressing human lysosomal -mannosidase.
Methods Construction of expression vectors The vectors pENTR11 and pCAGINeoGW were provided by Dr. Stephen Wood (Child Health Research Unit, Adelaide). The vector pcDNA3.1ATG2, containing the human -mannosidase cDNA [19], was provided by Dr. Thomas Berg (University of Tromso). All restriction enzymes were purchased from New England Biolabs, USA. pCAG-mann was created by subcloning the 3.2 kb fragment of the KpnI/SalI digested pcDNA3.1ATG2 into the KpnI/XhoI linearised pENTR11 (Gateway entry vector, Invitrogen, Australia). The resultant pENTR-mann was then combined with pCAGINeoGW (Gateway destination vector, Invitrogen, Australia) in a recombination reaction (as per the manufacturer’s instructions) to create pCAG-mann (Fig. 1A). The insert orientation was conWrmed by separate KpnI and ScaI digests followed by agarose gel electrophoresis. The vector pEFIRESpuro-6 was provided by Dr. Dan Peet (University of Adelaide). pcDNA3.1ATG2 was digested with XbaI to provide the 3.2 kb human -mannosidase cDNA fragment before subcloning into the XbaI site of pEFIRESpuro-6, to create the vector pEFIRES-mann (Fig. 1B). Insert orientation was conWrmed by digestion with XbaI and NotI followed by agarose gel electrophoresis. Transfection of pCAG-mann and pEFIRES-mann into D3 ES cells The constructs pCAG-mann and pEFIRES-mann (5 g of each) were transfected into D3 ES cells ([20], provided by Dr. Joy Rathjen from the University of
Fig. 1. Schematic diagram of the expression cassettes of the vectors pCAG-mann (A) and pEFIRES-mann (B). CMV, human CMV immediate early enhancer; chA, chicken -actin promoter; rhM, human -mannosidase cDNA; IRES, internal ribosomal entry site; neo, neomycin resistance gene; puro, puromycin resistance gene.
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
205
Adelaide) using methods described previously [18]. Transfected cells were selected in media containing 1.5 g/ml puromycin (for pEFIRES-mann transfected ES cells; Gibco-BRL, Invitrogen, Australia) or 250 g/ml G418 (for pCAG-mann transfected ES cells; GibcoBRL, Invitrogen, Australia).
described above. EBMs were also fed on day 11, 13, 15, and 18 of the protocol. On day 21, media were harvested by aspiration, and EBM cell outgrowths were harvested by incubation in trypsin/versene (0.05% w/v; CSL Limited, Australia) for 5 min, followed by trituration to detach and harvest, to give EBM-21 samples.
Cell culture
-Mannosidase activity in clones throughout diVerentiation
Routine culture of D3 ES cells was as described [21]. All conditioned media for neural induction of ES cells (MEDII) used in this study were produced and supplied by BresaGen Pty. Ltd, as described [21] and stored frozen at ¡70 °C prior to use. BrieXy, HepG2 cells [22] were trypsinised to a single cell suspension and seeded at 5 £ 104 cells/cm2 in Dulbecco’s modiWed Eagle’s medium (DMEM; Gibco-BRL, Invitrogen, Australia) suppleme nted with 10% fetal calf serum (FCS; CSL Limited, Australia) and 1 mM L-glutamine (CSL Limited, Australia) to give a ratio of 1.75 £ 105 cells/ml medium. Conditioned medium was collected after 4 days of culture, sterilised by Wltration through a 0.20 m membrane (Sartorius, Germany), and supplemented with 0.1 mM -mercaptoethanol (-ME; Sigma Chemical, USA) before use. Generation of embryoid bodies All cell aggregates (embryoid bodies or EBMs) grown in MEDII were formed by seeding 1 £ 106 ES cells (as a single cell suspension) into 10 ml of 50% (v/v) MEDII (DMEM with 10% FCS, 1 mM L-glutamine, and 0.1 mM -ME (Sigma, USA), supplemented with 50% (v/v) MEDII) in sterile bacterial petri dishes (to prevent attachment). EBMs were grown under 5% CO2 in a humidiWed incubator. EBMs were divided 1:2 (into two new dishes with fresh media, per plate) on day 2 and 4 of the protocol. EBMs harvested on day 4 for analysis were labelled EBM-4 (the number following “EBM” denoting the number of days in culture following commencement of the diVerentiation protocol). Fresh medium was added on day 5 and 6 of the protocol to generate more mature EBMs. On day 7, EBMs were transferred to serum-free medium [50% (v/v) DMEM and 50% (v/v) Ham’s F12 (Gibco-BRL, Invitrogen, Australia)]; supplemented with 1£ insulin–transferrin–sodium selenite (ITSS; Boehringer–Mannheim, Germany) and 10 ng/ml Wbroblast growth factor-2 (FGF-2; Peprotech, USA). EBM-9s were harvested 2 days later for analysis. For generation of EBM-12s, fresh serum-free medium was added on day 9 and 11 of the protocol, before harvesting of EBM-12s the following day. Later stage EBM populations (i.e., EBM-21) were generated by seeding EBMs onto gelatin-treated tissue culture-grade plastic-ware (BD Falcon, USA) on day 9 of the protocol and feeding with serum-free medium as
Samples from each clone were harvested as undiVerentiated ES cells (prior to commencement of the diVerentiation protocol), EBM-4, EBM-12, and EBM-21. Adherent cell populations were trypsinised to detach, EBMs were harvested by collection into a 10 ml centrifuge tube. Following harvest, cells or embryoid bodies were washed in phosphate-buVered saline (PBS; CSL Limited, Australia) and pelleted before resuspension in 500 l (per 10 cm plate or Xask) PBS lysis buVer [1% sodium deoxycholate (w/v; BDH, UK), 0.1% sodium dodecyl sulphate (w/v; BDH, UK), and 0.5% Nonidet P40 (v/v; Sigma Chemical, USA), in PBS]. Lysed samples were incubated at 4 °C overnight before microcentrifugation and storage at ¡20 °C prior to analysis of the supernatant. Total -mannosidase activity for each sample was measured by incubation with the Xuorogenic substrate 4-methylumbelliferyl--D-mannopyranoside [23] (Sigma Chemical, USA) prior to analysis on a Xuorometer (Perkin Elmer Luminescence Spectrophotometer, LS 50B). Human speciWc -mannosidase activity was measured in a similar fashion, incorporating an immunocapture stage into the protocol using a mouse monoclonal antibody speciWc for human -mannosidase (4C5 monoclonal [24]). This antibody was generated to facilitate immunopuriWcation of recombinant human -mannosidase in a previous study [24], and shows very low cross-reactivity with -mannosidase from other species (0.1 and 0.04% from guinea pig and cat, respectively, unpublished observations). Protein concentration of all samples was assayed using the Bio-Rad Protein Assay Dye reagent concentrate, according to the manufacturer’s speciWcations (Bio-Rad, USA). All activities and proteins were measured in triplicate, and activity was normalised to total cell protein. Cross-correction of lysosomal storage Uptake of -mannosidase from conditioned media samples was assessed using a similar method to a previous study [24]. -Mannosidosis (SF39) and normal (SF48 and SF49) human skin Wbroblasts were used for cross-correction studies. Conditioned media were obtained from populations of seeded embryoid bodies, harvested at the EBM-21 stage (such that media had been conditioned for 72 h). Conditioned media were Wltered using a 0.20 m Wlter (Sartorius, Germany) prior
206
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
to further use. Media samples were diluted in unconditioned serum-free media where required. Mannose-6phosphate (M6P; Sigma Chemical, USA) was added to conditioned media at a concentration of 5 mM where indicated. Fibroblasts were incubated with conditioned media samples for 48 or 96 h before harvesting in PBS lysis buVer for analysis. The level of mannose-containing oligosaccharides Man2GlcNAc and Man3GlcNAc was quantitated by Dr. Maria Fuller (Women’s and Children’s Hospital, Adelaide) as described previously [24]. ImmunoXuorescence protocols for lineage analysis Thermanox coverslips (Nunc, Rochester, NY, USA) were coated by sequential overnight incubation (in 24 well-plates) in a 20 g/ml solution of polyornithine (Sigma Chemical, USA) and then a 1 g/ml solution of laminin (Sigma Chemical, USA) at 37 °C. Coverslips were rinsed three times in sterile water and stored in PBS prior to use. Approximately, two to four EBM-9 bodies per well were transferred into serum-free media in 24-well dishes containing coverslips, cultured until the EBM-21 stage, and then Wxed in 4% (w/v) paraformaldehyde (Sigma Chemical, USA) in PBS for 10 min. Coverslips were washed in PBS and stored at 4 °C in PBS/0.1% (w/v) sodium azide (Sigma Chemical, USA) prior to immunoXuorescence. For Oct4 immunoXuorescence, Wxation was carried out in 4% paraformaldehyde in 70% (v/v) ethanol for 10 min, prior to washing and storage in 70% (v/v) ethanol. Seeded EBMs were washed in a buVer consisting of PBS/0.3% (v/v) Triton X-100 (Sigma Chemical, USA), incubated in dimethyl sulfoxide for 10 min, then washed again before blocking in wash buVer containing 10% (v/ v) normal donkey serum (Jackson Immunoresearch Laboratories, USA) for 30 min. Incubations in primary antisera (diluted in wash buVer by a factor speciWed below) were carried out overnight at room temperature, before washing as above. Samples were then incubated in secondary antisera (diluted in wash buVer by a factor speciWed below) overnight at room temperature, wrapped in foil to prevent photobleaching. Samples were then washed before mounting in Vectashield mounting media containing the nuclear dye 4⬘-6-diamidino-2-phenylindole (DAPI; #H1200, Vector Laboratories, USA). Primary antisera and dilution factors used were: mouse -Nestin (a marker of neural progenitors [25]; Rat 401 monoclonal IgG1, Yale, USA), 1:500; rabbit -NF200 (neuroWlament protein, 200 kDa subunit, a marker for mature neuronal processes [26]; N52 monoclonal, Sigma Chemical, USA), 1:500; rabbit -GFAP (glial Wbrillary acidic protein, a marker expressed by both glial precursors and diVerentiated astrocytes [27]; G9269, Sigma Chemical, USA), 1:200; goat -Oct4
(Octamer-4, a marker of pluripotent cells [28]; sc-8628 (N19) polyclonal, Santa Cruz Biotechnology, USA), 1:1000; mouse -SMA (smooth muscle actin, a marker for smooth muscle cells; [29]; A2547 clone 1A4 monoclonal, Sigma Chemical, USA), 1:2000. Secondary antisera and dilution factors used were: sheep -mouse FITC (AQ325F, Chemicon, USA), 1:300; donkey -rabbit FITC (711-095-152, Jackson Immunoresearch Laboratories, USA), 1:100; donkey -mouse CY3 (715-167-003, Jackson Immunoresearch Laboratories, USA), 1:300; donkey -goat FITC (705-095-003, Jackson Immunoresearch Laboratories, USA), 1:200. Fluorescence microscopy was carried out using an Olympus AX70 Wtted with epiXuorescence and a video camera connected to a PC computer. For Cy3, a Chroma 31002 Wlter block (beam splitter: 565 nm, excitation Wlter: 515–50 nm, and barrier Wlter: 575–615 nm); for FITC, a Chroma 31002 Wlter block (beam splitter: 505 nm, excitation Wlter: 465–95 nm, barrier Wlter: 515– 55 nm); and for DAPI, a Chroma 41008 Wlter block (beam splitter: 660 nm, excitation Wlter: 590–650 nm, and barrier Wlter: 665–740 nm) were used.
Results Creation of ES cell lines All transfected ES cell clones were found to generate colonies within the selection period described, whereas untransfected ES cells did not generate G418 resistant colonies in this protocol. The pCAG-mann transfected clones only had marginally elevated levels of total mannosidase activity relative to levels observed in untransfected ES cells, in select clones (Fig. 2A). pCAGmann transfected clones analysed at the EBM-12 stage of diVerentiation did not show increased total -mannosidase expression relative to endogenous -mannosidase expression levels measured in untransfected D3 embryoid bodies (Fig. 2A). Further analysis of human -mannosidase activity (Fig. 2B) indicated that -mannosidase activity resulting from transgene expression (rhM) was below endogenous -mannosidase expression levels, and clone G8 was the only cell line that maintained rhM expression out to day 21 of the diVerentiation protocol. In contrast, pEFIRES-mann transfected clones demonstrated a dramatic increase above expression levels observed in untransfected cells (Fig. 3A), which was veriWed by analysis of rhM activities (Fig. 3B). Clone EF3 produced the highest levels of rhM at the EBM-21 stage, resulting in total -mannosidase activity observed to be approximately fourfold above endogenous levels. Subjective analysis of embryoid bodies generated from diVerent clones was also undertaken in both diVerentiation experiments to conWrm that transfected clones were all able to generate bodies not morphologically dis-
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
207
Fig. 2. -Mannosidase activity in pCAG-mann transfected clones. Activities expressed as nmol/min/mg of total cell protein, +1SD, measured in triplicate. D3 ES cells shown were untransfected (wild-type). All other clones described are D3 ES cells transfected with pCAG-mann. (A) Total -mannosidase activity in pCAG-mann transfected clones, (B) speciWcally human -mannosidase activity in pCAG-mann transfected clones.
similar to those generated by untransfected cells. All clones were observed to generate morphologically similar embryoid bodies and similar outgrowths following seeding (Table 1), with the notable exception of clones G5 and G9. These two clones appeared to respond poorly to the neurectoderm induction of the MEDII protocol in both diVerentiation experiments, and appeared to have fewer neural outgrowths following seeding. Clone EF3 was chosen for further characterisation. Testing of clone EF3: lineage analysis A qualitative assessment of a number of lineage markers was carried out to conWrm that transfected ES cells retained a similar diVerentiation potential to untransfected ES cells. There was no diVerence between undiVerentiated ES cells (from both untransfected D3 cells and transfected clone EF3 cells) with regard to their expression of Oct4, with almost all cells from both sources found to be immunoreactive for this marker. At the EBM-21 stage, small pockets of Oct4 immunoreactive cells were observed within some embryoid bodies, although the frequency of these pockets was the same for both D3 and EF3 cell populations. Both D3 and EF3 cell lines strongly expressed Nestin at the EBM-21 stage, and similarly diVerentiated into GFAP and SMA posi-
tive cells. D3-derived EBMs had a greater prevalence of NF200 positive outgrowths (Fig. 4), although this probably reXects variation between diVerent seeded cell populations. Overall, there were no qualitative diVerences observed in immunoreactivity between D3- and EF3derived embryoid bodies. Cross-correction of lysosomal storage in -mannosidosis skin Wbroblasts The addition of 4 £ 10¡4 U/ml -mannosidase (produced by the cell line EF3 and consisting of both endogenous mouse and recombinant human enzyme) to the culture media of -mannosidosis skin Wbroblasts reduced the level of oligomannoside storage (Fig. 5). This reduction was found to be dose-dependent, with the addition of a lower dose of enzyme (8.6 £ 10¡6 U/ml mannosidase) resulting in a lesser reduction in oligomannosides. Additionally, this reduction was found to be greater 48 h following addition of the conditioned media sample than at 96 h, which suggested that at 96 h, the lower dose of enzyme is unable to maintain reduced levels of these stored oligomannosides, resulting in further accumulation. The addition of 8.6 £ 10¡6 U/ml mannosidase in the form of conditioned media derived from untransfected D3 embryoid bodies (endogenous mouse enzyme only) similarly reduced the level of stored
208
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
Fig. 3. -Mannosidase activity in pEFIRES-mann transfected clones. Activities expressed as nmol/min/mg of total cell protein, +1SD, measured in triplicate. D3 ES cells shown were untransfected (wild-type). All other clones described are D3 ES cells transfected with pEFIRES-mann. (A) Total -mannosidase activity in pEFIRES-mann transfected clones, (B) speciWcally human -mannosidase activity in pEFIRES-mann transfected clones.
oligomannosides in these manosidosis cell lines (Fig. 5). The addition of 5 mM M6P to the conditioned media with 4 £ 10¡4 U/ml -mannosidase (from cell line EF3) lessened the reduction of oligomannosides Man2GlcNAc and Man3GlcNAc.
Discussion This study describes the generation and characterisation of the mouse ES cell line EF3, which expresses rhM at levels approximately 15-fold above endogenous expression levels (at ES cell stage). These expression levels decrease throughout 21 days of diVerentiation to approximately fourfold above endogenous levels. Through qualitative analysis of various lineage markers in undiVerentiated ES cells and their derivatives, this cell line was found to have retained a similar diVerentiation potential to untransfected ES cells, suggesting that the genetic modiWcation carried out to facilitate rhM over-
expression did not noticeably aVect other characteristics of this ES cell line. The CMV enhancer/chicken -actin promoter (CAG) and the EF1 promoter were evaluated for these cell lines, as they have previously been used to facilitate transgene expression in ES cells and their derivatives [30]. This previous study reported a higher activity for the EF1 promoter compared to the CAG promoter in transiently transfected ES cells, and showed that the activity of the promoter EF1 was both reduced and found to be more variable through 6 days of diVerentiation. To date, the creation of stably transfected mouse ES cell lines for the expression of a lysosomal enzyme has been reported only once [18]. In this study, the CAG vector was evaluated for achieving expression of the enzyme sulphamidase, and was shown to eVectively facilitate sustained (out to 15 days of diVerentiation) overexpression in ES cells and their derivatives, possibly due to relatively low endogenous sulphamidase levels in this system [18]. To our knowledge, the present study details
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
209
Table 1 Morphological evaluation of embryoid body appearance throughout diVerentiation Clone D3
G5
G8
G9
G12
G15
EF1
EF3
EF7
EF8
EF15
EBM-4 Smoothness Neurectoderm Necrosis Beating Maturity
+++ +++ + + ++
¡¡ + ¡¡ + ¡
+++ + + + +++
¡¡ + ¡¡ + ¡
+++ + + + +++
¡ + ¡¡ + +
+++ +++ + + ++
¡ +++ + + +++
+++ +++ + + +++
+++ +++ + + +
+++ +++ + + ++
EBM-12 Smoothness Neurectoderm Necrosis Beating Maturity
+++ ++ ¡ + +
¡¡ + ¡¡ + +
+++ ++ ¡ + +
¡¡ + ¡¡ + +
+++ +++ ¡ + +
+++ +++ ¡ + +
+++ +++ ¡ + +
+++ ++ ¡ + +
+++ + ¡ + +
+++ + ¡ + +
++ ++ ¡¡ + +
EBM-21 Neurites Fibroblasts Density
+++ +++ +++
¡ +++ ++
+++ +++ +++
++ ++ ¡
+++ +++ +++
+++ +++ +++
¡ +++ ++
+++ +++ +++
+++ +++ +++
+++ +++ +++
+++ +++ +++
EBM-4 and EBM-12: Smoothness. Surface texture of embryoid bodies (positive score indicates smooth surfaced embryoid bodies, negative score indicates areas of outgrowth from surface of bodies). Neurectoderm. Cavitation/neurectoderm formation in embryoid bodies (positive score indicates degree of neurectoderm formation). Necrosis. Necrosis within embryoid bodies (positive score indicates absence of necrotic areas within bodies, negative score indicates prevalence of necrotic areas). Beating. Beating areas within embryoid bodies (positive score indicates lack of beating muscle in bodies). Maturity. Maturity of embryoid bodies (positive score indicates relative speed at which cultures matured to morphologically diverse bodies, negative score indicates the presence of immature, less diVerentiated bodies). EBM-21: Neurites. Filamentous outgrowths extending out from embryoid body. Fibroblasts. Fibroblastic cells growing around embryoid body. Density. Density of outgrowths from embryoid body.
the Wrst report of stable transfection of ES cells with both of these vectors, followed by isolation of multiple clonal cell lines and evaluation through 21 days of diVerentiation (directed using the MEDII diVerentiation protocol). In this study, pCAG-mann was found to facilitate sustained expression of rhM in select clones (such as clone G8, shown in Fig. 3A), although only at moderate levels relative to endogenous -mannosidase expression levels. Despite the observation that rhM expression levels in pEFIRES-mann decline throughout diVerentiation, this vector was found to be most eVective for facilitating sustained over-expression of rhM, as overexpression was still noted after three weeks of in vitro diVerentiation. These results suggest the need to evaluate appropriate expression vectors for each individual gene where over-expression is the ultimate aim of constructing the cell line. Morphological analysis of embryoid bodies also showed variable results between transfected clones. The highest expressing clone (EF3) was found to generate embryoid bodies of similar morphology to those generated from untransfected cells, while some other transfected clones appeared to generate embryoid bodies of abnormal morphology. Although not a deWnitive method
of analysing diVerentiation potential, this observation suggests the need for careful evaluation of stably transfected clones prior to use in further studies, to ensure that they retain their ability to respond to extrinsic diVerentiation factors in a similar fashion to untransfected ES cells. In order for a cell therapy to be successful for treatment of LSD, secreted enzyme from implanted cells must be able to be taken up by host cells. -Mannosidase has been shown to enter cells via M6P receptor-mediated endocytosis in a similar fashion to other lysosomal enzymes [24,31]. The mannose-6-phosphorylation of mannosidase is thus required for delivery of deWcient enzyme to aVected tissues via the mechanism of receptor-mediated endocytosis. Enzyme secreted by diVerentiated derivatives of the cell line EF3 (both rhM and endogenous mouse mannosidase) was found to mediate a reduction in the levels of stored oligosaccharides in human -mannosidosis skin Wbroblasts in a dose- and M6P-dependent manner. Although present at lower concentrations than rhM secreted by the cell line EF3, endogenous mouse -mannosidase (produced and secreted by untransfected D3 ES cells) also showed a cross-correction eVect. The results of this experiment suggest that for a given dose of -mannosidase activity provided to
210
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
Fig. 4. Immunoreactivity of D3 and EF3 embryoid bodies. (A, C, and E) Embryoid bodies and cell outgrowths (EBM-21 stage) derived from untransfected D3 ES cells. (B, D, and F) An embryoid bodies and cell outgrowths (EBM-21 stage) derived from EF3 ES cells. (A and B) Blue, DAPI (nuclear Xuorescence); red, nestin (precursor cells). (C and D) Blue, DAPI; green, GFAP (glia). (E and F) Blue, DAPI; green, NF200 (neurons).
the deWcient Wbroblasts, endogenous enzyme (derived from untransfected D3 ES cells) was able to mediate a greater reduction in oligosaccharide levels than EF3derived rhM (see Fig. 5). One possible explanation for this observation is that rhM may be less eYciently mannose-6-phosphorylated than endogenous mouse mannosidase, resulting in less eYcient delivery of this enzyme to the lysosomes of deWcient Wbroblasts. Previous studies have also demonstrated a greater level of mannose-6-phosphorylation of mouse -mannosidase, compared to rhM, but showed that rhM was still able to mediate a signiWcant reduction in lysosomal storage in treated mice [7].
At present, it is unknown exactly how much graftderived -mannosidase would be required in vivo to see a therapeutic eVect on neurological pathology. The observation of clinical beneWt to neurological disease in -mannosidosis from BMT alone [5,32] suggests that very high expression levels may ultimately not be required to achieve the desired reduction in lysosomal storage, and in turn, clinical improvement. However, therapeutic beneWt as a result of stem cell implantation would be expected to depend on the dose of -mannosidase delivered to the aVected CNS. Thus, the generation of the -mannosidase over-expressing mouse ES cell line described in this study is expected to
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
211
Fig. 5. Cross-correction of -mannosidosis skin Wbroblasts: Man2GlcNAc levels. The levels of oligomannosides are expressed as a ratio of intensity (counts per second) relative to the internal standard methyllactose. (A) Man2GlcNAc levels, (B) Man3GlcNAc levels. “Low dose” was 8.6 £ 10¡6 U/ ml total -mannosidase. “High dose” was 4 £ 10¡4 U/ml total -mannosidase. “Transfected” and “Untransfected” indicate source of conditioned media, EF3 or D3 ES cells, respectively. M6P was added to a Wnal concentration of 5 mM, where indicated.
improve the prospects for stem cell-based therapies for -mannosidosis.
Acknowledgments The authors thank Drs. Dan Peet, Steve Wood, and Thomas Berg for providing the basic vectors used in this study and for helpful discussions; Dr. Joy Rathjen and Professor Peter Rathjen for their technical advice and
training with ES cells; Dr. Tom Litjens for his helpful advice regarding vector construction; Adeline Lau and Drs. Adrian Meedeniya, Bruce Davidson, and Kim Hemsley for technical advice and helpful discussions regarding cell line characterisation; and Dr. Maria Fuller for mass spectrometry analysis. We also thank Bionomics Ltd and BresaGen Ltd for the use of their facilities. This work was supported by the National Health and Medical Research Council of Australia and an Australian Postgraduate Award (A.J.R.).
212
A.J. Robinson et al. / Molecular Genetics and Metabolism 85 (2005) 203–212
References [1] J.J. Hopwood, D.A. Brooks, An introduction to the basic science and biology of the lysosome and storage diseases, in: D.A. Applegarth, J.E. Dimmick, J.G. Hal (Eds.), Organelle Diseases, Chapman & Hall, London, 1997. [2] G.H. Thomas, A.L. Beaudet, Disorders of glycoprotein degradation and structure: -mannosidosis, -mannosidosis, fucosidosis, sialidosis, Aspartylglucosaminuria, and carbohydrate-deWcient glycoprotein syndrome, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Basis of Inherited Disease, McGraw-Hill, New York, 1995, pp. 2529–2561. [3] J.C. Michalski, A. Klein, Glycoprotein lysosomal storage disorders: alpha- and beta-mannosidosis, fucosidosis and alpha-N-acetylgalactosaminidase deWciency, Biochim. Biophys. Acta 1455 (2–3) (1999) 69–84. [4] P.J. Meikle, J.J. Hopwood, A.E. Clague, W.F. Carey, Prevalence of lysosomal storage disorders, JAMA 281 (3) (1999) 249–254. [5] D.A. Wall, D.K. Grange, P. Goulding, M. Daines, A. Luisiri, S. Kotagal, Bone marrow transplantation for the treatment of mannosidosis, J. Pediatr. 133 (2) (1998) 282–285. [6] P.M. Hoogerbrugge, O.F. Brouwer, P. Bordigoni, O. Ringden, P. Kapaun, J.J. Ortega, A. O’Meara, G. Cornu, G. Souillet, D. Frappaz, S. Blanche, A. Fischer, Allogenic bone marrow transplantation for lysosomal storage diseases, Lancet 345 (1995) 1398–1402. [7] D.P. Roces, R. Lullmann-Rauch, J. Peng, C. Balducci, C. Andersson, O. Tollersrud, J. Rogh, A. Orlacchio, T. Beccari, P. Saftig, K. von Figura, EYcacy of enzyme replacement therapy in alphamannosidosis mice: a preclinical animal study, Hum. Mol. Genet. 13 (18) (2004) 1979–1988. [8] A.C. Crawley, M.Z. Jones, L.E. Bonning, J.W. Finnie, J.J. Hopwood, -Mannosidosis in the Guinea Pig: a new animal model for lysosomal storage disorders, Pediatr. Res. 46 (5) (1999) 501– 509. [9] E.Y. Snyder, R.M. Taylor, J.H. Wolfe, Neural progenitor cell engraftment corrects lysosomal storage throughout MPS VII mouse brain, Nature 374 (6520) (1995) 367–370. [10] H.D. Lacorazza, J.D. Flax, E.Y. Snyder, M. Jendoubi, Expression of human beta-hexosaminidase alpha-subunit gene (the gene defect of Tay–Sachs disease) in mouse brains upon engraftment of transduced progenitor cells, Nat. Med. 2 (4) (1996) 424–429. [11] H.K. Jin, J.E. Carter, G.W. Huntley, E.H. Schuchman, Intracerebral transplantation of mesenchymal stem cells into acid sphingomyelinase-deWcient mice delays the onset of neurological abnormalities and extends their life span, J. Clin. Invest. 109 (9) (2002) 1183–1191. [12] X.L. Meng, J.S. Shen, T. Ohashi, H. Maeda, S.U. Kim, Y. Eto, Brain transplantation of genetically engineered human neural stem cells globally corrects brain lesions in the mucopolysaccharidosis type VII mouse, J. Neurosci. Res. 74 (2) (2003) 266–277. [13] G.Q. Daley, M.A. Goodell, E.Y. Snyder, Realistic prospects for stem cell therapeutics., Hematology (Am. Soc. Hematol. Educ. Program) 2003 (2003) 398–418. [14] V. Tropepe, S. Hitoshi, C. Sirard, T.W. Mak, J. Rossant, D. van der Kooy, Direct neural fate speciWcation from embryonic stem cells: a primitive mammalian neural stem cell stage acquired through a default mechanism, Neuron 30 (1) (2001) 65–78. [15] J. Rathjen, B.P. Haines, K.M. Hudson, A. Nesci, S. Dunn, P.D. Rathjen, Directed diVerentiation of pluripotent cells to neural lineages: homogeneous formation and diVerentiation of a neurectoderm population, Development 129 (11) (2002) 2649–2661. [16] M.P. Stavridis, A.G. Smith, Neural diVerentiation of mouse embryonic stem cells, Biochem. Soc. Trans. 31 (Pt 1) (2003) 45–49.
[17] T. Barberi, P. Klivenyi, N.Y. Calingasan, H. Lee, H. Kawamata, K. Loonam, A.L. Perrier, J. Bruses, M.E. Rubio, N. Topf, V. Tabar, N.L. Harrison, M.F. Beal, M.A. Moore, L. Studer, Neural subtype speciWcation of fertilization and nuclear transfer embryonic stem cells and application in parkinsonian mice, Nat. Biotechnol. 21 (10) (2003) 1200–1207. [18] A.A Lau, K.M. Hemsley, A. Meedeniya, J.J. Hopwood, In vitro characterisation of genetically modiWed embryonic stem cells as a therapy for murine mucopolysaccharidosis type IIIA, Mol. Genet. Metab. 81 (2) (2004) 86–95. [19] O. Nilssen, T. Berg, H.M. Riise, U. Ramachandran, G. Evjen, G.M. Hansen, D. Malm, Tranebjaerg, O.K. Tollersrud, -Mannosidosis: functional cloning of the lysosomal -mannosidase cDNA and identiWcation of a mutation in two aVected siblings, Hum. Mol. Genet. 6 (5) (1997) 717–726. [20] T.C. Doetschman, H. Eistetter, M. Katz, W. Schmidt, R. Kemler, The in vitro development of blastocyst derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium, J. Embryol. Exp. Morphol. 87 (1985) 27–45. [21] J. Rathjen, J.A. Lake, M.D. Bettess, J.M. Washington, G. Chapman, P.D. Rathjen, Formation of a primitive ectoderm like cell population, EPL cells, from ES cells in response to biologically derived factors, J. Cell Sci. 112 (5) (1999) 601–612. [22] B.B. Knowles, C.C. Howe, D.P. Aden, Human hepatocellular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen, Science 209 (1980) 497–499. [23] J.L. Avila, J. Convit, Characterization and properties of alpha-Dmannosidase of human polymorphonuclear leucocytes, Clin. Chim. Acta 47 (3) (1973) 335–345. [24] T. Berg, B. King, P.J. Meikle, O. Nilssen, O.K. Tollersrud, J.J. Hopwood, PuriWcation and characterization of recombinant human lysosomal alpha-mannosidase, Mol. Genet. Metab. 73 (1) (2001) 18–29. [25] L. Zimmerman, B. Parr, U. Lendahl, M. Cunningham, R. McKay, B. Gavin, J. Mann, G. Vassileva, A. McMahon, Independent regulatory elements in the nestin gene direct transgene expression to neural stem cells or muscle precursors, Neuron 12 (1994) 11–24. [26] C. Straznicky, J.C. Vickers, R. Gabriel, M. Costa, A neuroWlament protein antibody selectively labels a large ganglion cell type in the human retina, Brain Res. 582 (1) (1992) 123–128. [27] C.F. Landry, G.O. Ivy, I.R. Brown, Developmental expression of glial Wbrillary acidic protein mRNA in the rat brain analyzed by in situ hybridisation, J. Neurosci. Res. 25 (1990) 194–203. [28] M.H. Rosner, A. Vigano, K. Ozato, P.M. Timmons, F. Poirier, P.W.J. Rigby, L.M. Staudt, A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo, Nature 345 (1990) 686–692. [29] O. Skalli, P. Ropraz, A. Trzeciak, G. Benzonana, D. Gillessen, G. Gabbiani, A monoclonal antibody against alpha-smooth muscle actin: a new probe for smooth muscle diVerentiation, J. Cell Biol. 103 (6 Pt 2) (1986) 2787–2796. [30] S. Chung, T. Andersson, K.C. Sonntag, L. Björklund, O. Isacson, K.S. Kim, Analysis of diVerent promoter systems for eYcient transgene expression in mouse embryonic stem cell lines, Stem Cells 20 (2) (2002) 139–145. [31] H. Sun, M. Yang, M.E. Haskins, D.F. Patterson, J.H. Wolfe, Retrovirus vector-mediated correction and cross-correction of lysosomal -mannosidase deWciency in human and feline Wbroblasts, Hum. Gene Ther. 10 (1999) 1311–1319. [32] S.U. Walkley, M.A. Thrall, K. Dobrenis, M. Huang, P.A. March, D.A. Diegel, S. Wurzelmann, Bone marrow transplantation corrects the enzyme defect in neurons of the central nervous system in a lysosomal storage disease, Proc. Natl. Acad. Sci. USA 91 (1994) 2970–2974.
