The molecular biology of Gaucher disease and the potential for gene therapy

Turnover and distribution of intravenously administered mannose-terminated human acid beta-glucosidase in murine and human tissues

Gaucher disease type 1, the most prevalent lysosomal storage disease, is caused by the defective activity of the lysosomal enzyme, acid beta-glucosidase, or glucocerebrosidase. Infusion of purified acid beta-glucosidase containing alpha-mannosyl-terminated oligosaccharides (alglucerase) is efficacious in reversing hematologic, hepatic, splenic, and bony disease manifestations. The murine tissue distribution and turnover of bolus injections of alglucerase was evaluated by enzymatic activity, quantitative cross-reacting immunologic material analyses, and immunofluorescence studies. Enzyme activity measurements detected distribution to liver, spleen, thymus, kidney, and bone marrow mononuclear cells, but not to lungs and brain. In kidney and thymus, the enzyme was transiently present. In liver and spleen, enzyme activity peaked at about 20 min postinjection followed by a biphasic decrease with t1/2 approximately 40-60 min and approximately 12-14 h. In bone marrow maximal enzyme activity was at 40-60 min with a disappearance t1/2 approximately 60 min. Quantitative cross-reacting immunologic material studies of liver and spleen showed delivery of enzyme with decreased catalytic rate constants whose degradation included denaturation and proteolytic components. By immunofluorescence the human enzyme was distributed primarily to reticuloendothelial cells of the liver and spleen. In autopsy material from a Gaucher disease type 2 patient treated with enzyme, immunohistochemical and activity studies showed distributions similar to those in mice. These studies indicate a complex delivery and intracellular degradation of acid beta-glucosidase with lower intrinsic activity than the administered therapeutic agent.

Drug-selected coexpression of human glucocerebrosidase and P-glycoprotein using a bicistronic vector

Bicistronic cassettes under control of a single promoter have recently been suggested as useful tools for coordinate expression of two different foreign proteins in mammalian cells. Using the long 5' untranslated region of encephalomyocarditis virus as translational enhancer of the second gene, a bicistronic unit composed of cDNA for human P-glycoprotein [the product of the multidrug resistance gene, MDR1 (also called PGY1)] as selectable marker and cDNA for human glucocerebrosidase (GC; EC 3.2.1.45) (a membrane-associated lysosomal hydrolase) was constructed. NIH 3T3 cells transfected with a Harvey murine sarcoma virus retroviral vector carrying this bicistronic cassette (pHaMCG) express active P-glycoprotein and GC and expression of both proteins augments coordinately with selection for increased colchicine resistance. Percoll gradient analysis of homogenates showed that GC was targeted to the lysosomal fraction. The ability to select for expression of GC with natural product drugs after introduction of the pHaMCG retroviral vector may be useful in gene therapy strategies for Gaucher disease.

Gaucher disease as a paradigm of current issues regarding single gene mutations of humans

Gaucher disease is a glycolytic storage disease caused by a deficiency in activity of the catabolic enzyme glucocerebrosidase. Over 35 different mutations have been documented, including missense and nonsense point mutations, splicing mutations, deletions and insertions, a fusion gene, and examples of gene conversion. Gaucher disease is most common in the Ashkenazi Jewish population, in which just five of the mutations in this population account for 98% of the disease-producing alleles. Each of these mutations is found in the context of a single haplotype, a finding consistent with a single origin of each mutation. Although it is clear that these mutations provide a selective advantage in the Jewish population and thus constitute a balanced polymorphism, the nature of the advantage is unknown. Gaucher disease can be treated symptomatically, by administration of the missing enzyme, by allogeneic bone marrow transplantation, and potentially by gene transfer into hematopoietic stem cells. Increasing understanding of this disease has, as in other genetic disorders, created a host of social and ethical dilemmas regarding matters such as the cost of treatment for rare diseases and the advantages and disadvantages of population-targeted genetic screening.

Treatment of inherited neurometabolic diseases: the future

The past 10 years' experience with bone marrow transplantation from normal, immunologically compatible donors indicates its possible use in various neurometabolic diseases, particularly in a patient who has not suffered irreparable brain damage. This experience may be a prelude to treatment by somatic gene therapy. This can be applied as an autologous bone marrow transplant, grafting the patient's own stem cells inserted with the normal gene. Although somatic gene therapy will be relatively easy for tissues with dividing cells, its application to target tissues with little or no cell division may pose difficulties. Meanwhile, techniques for the preservation, culture, and grafting of fetal neurons in humans have been developed and have been used in the treatment of Parkinson's disease. These procedures could readily be transferred to the treatment of other neurodegenerative diseases that cause significant morbidity, but ethical, legal, and religious considerations must be taken into account. All these efforts promise novel and improved management of inborn neurometabolic errors.

Restoration of arylsulphatase A activity in human-metachromatic-leucodystrophy fibroblasts via retroviral-vector-mediated gene transfer

Metachromatic leukodystrophy is a lysosomal storage disease caused by the deficiency of arylsulphatase A (ASA). A human ASA cDNA was subcloned into the retroviral vector pXT1. Replication-defective retrovirus was generated by transfection of the vector into the amphotropic packaging cell line PA317. Human fibroblasts from a patient suffering from metachromatic leucodystrophy was infected with the recombinant retrovirus. Infected fibroblasts expressed ten times more ASA compared with control fibroblasts from a normal individual. The ASA encoded by the integrated provirus was shown to be correctly transported into the lysosomes and to normalize the impaired degradation of cerebroside sulphate.

Toward gene therapy for Gaucher disease

We are studying the transfer and expression by retroviral vectors of the human glucocerebrosidase (GC) gene into bone marrow cells as a model of gene therapy for genetic diseases of hematopoietic cells. A simple retroviral vector (G2) was developed that contains a normal human GC cDNA under the control of the Moloney murine leukemia virus long-terminal repeat (LTR) enhancer/promoter. Murine bone marrow was transduced with the G2 vector and maintained in long-term bone marrow culture (LTBMC). Expression of the human GC gene in the transduced murine LTBMC cells exceeded the level of endogenous murine GC mRNA. Murine bone marrow cells were also transduced with G2 and transplanted into irradiated syngeneic recipients. High levels of GC gene transfer and expression were seen in day-12 CFU-S foci, and to a lesser extent in the hematopoietic organs 4 months after gene transfer/bone marrow transplant (BMT). Human bone marrow, from a patient with Gaucher disease, was also used in studies of GC gene transduction. Gene transfer into 35-40% of the Gaucher hematopoietic progenitor cells was achieved, following prestimulation of the marrow with recombinant hematopoietic growth factors. Equal rates of gene transfer were obtained using either total marrow mononuclear cells or progenitor cells enriched 100-fold by immunomagnetic bead separation. GC gene transduction corrected the enzymatic deficiency of the Gaucher marrow. Our results demonstrate the potential utility of retroviral vector-mediated gene transfer for gene therapy of Gaucher disease. Current efforts are aimed at achieving more consistent in vivo GC expression in the murine BMT model and demonstrating transduction of pluripotent human hematopoietic stem cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Restoration of arylsulphatase B activity in human mucopolysaccharidosis-type-VI fibroblasts by retroviral-vector-mediated gene transfer

The Maroteaux-Lamy syndrome (mucopolysaccharidosis type VI; MPS VI) is a lysosomal storage disease caused by deficiency of the enzyme arylsulphatase B (ASB). A human ASB cDNA has been subcloned into the retroviral vector pXT1 containing the bacterial neomycin-resistance gene and an internal thymidine kinase promoter for transcription of the inserted gene. Replication defective retrovirus was generated by transfecting the construct into the amphotropic packaging cell line PA317. Human MPS VI fibroblasts infected with recombinant retrovirus integrated the provirus into their genome and expressed retrovirus-encoded ASB mRNAs. In infected fibroblasts the level of ASB was up to 36-fold higher than in normal fibroblasts. Biosynthesis and processing of ASB in infected MPS VI fibroblasts was accomplished as in normal fibroblasts, and mature, enzymically active, ASB accumulated in dense lysosomes, indicating that the ASB deficiency in MPS VI fibroblasts was corrected by the retroviral gene transfer.

Differential microtubule requirements for transcytosis in MDCK cells

Given the role of microtubules in directing the transport of many intracellular organelles, we investigated whether intact microtubules were also required for transcytosis across epithelia. Using polarized MDCK cells expressing receptors for the Fc domain of IgG (FcRII-B2) or polymeric immunoglobulin (pIg-R), we examined the involvement of microtubules in apical to basolateral and basolateral to apical transcytosis, respectively. While depolymerization of microtubules with nocodozole had no effect on apical to basolateral transcytosis via FcR, basolateral to apical transcytosis of dimeric IgA via pIg-R was almost completely blocked. Inhibition due to nocodozole was selective for basolateral to apical transcytosis, since neither endocytosis nor receptor recycling was significantly affected at either plasma membrane domain. As shown by confocal microscopy, the block in transcytosis was due to the inability of MDCK cells to translocate IgA-containing vesicles from the basolateral to the apical cytoplasm in the absence of an intact microtubule network. The nocodazole sensitive step could be partially by-passed, however, by allowing cells to internalize IgA at 17 degrees C prior to nocodazole treatment. Although incubation at 17 degrees C blocked release of IgA into the apical medium, it did not prevent translocation of IgA-containing vesicles to the apical cytoplasm. Thus, receptor-mediated transcytosis in opposite directions exhibits distinct requirements for microtubules, a feature which reflects the spatial organization of MDCK cells.

Reversal of creatine kinase translational repression by 3' untranslated sequences

A subline of U937 cells (U937D) was obtained in which creatine kinase B (CK-B) messenger RNA was present and bound to ribosomes, but CK activity was undetectable. Transformation of U937D cells with retrovirus vectors that contain the 3' untranslated region (3' UTR) of CK-B messenger RNA exhibited CK activity with no change in abundance of CK-B mRNA. The 3' UTR formed a complex in vitro with a component of S100 extracts from wild-type cells. This binding activity was not detectable in S100 extracts from cells that expressed CK activity after transformation with the 3' UTR-containing vector. These results suggest that translation of CK-B is repressed by binding of a soluble factor or factors to the 3' UTR.

Antisense RNA complementary to 3' coding and noncoding sequences of creatine kinase is a potent inhibitor of translation in vivo

Antisense RNA is a potentially powerful tool for creating dominant negative mutations, but one of the limitations of this strategy has been the relative inefficiency of antisense transcripts in blocking target gene expression. To identify more effective target sequences, helper-free retrovirus-mediated gene transfer was used to introduce antisense RNAs complementary to multiple functional regions of the human creatine kinase B (CK-B) mRNA into U937 cells. Antisense RNA complementary to the last third of the coding and all of the noncoding regio of this mRNA is highly effective; one or two antisense transcripts is sufficient to block the expression of one CK-B mRNA. In contrast, antisense RNA from which sequences complementary to the last 17 codons and all the 3' noncoding region have been deleted has no effect on CK-B expression. Neither antisense RNA alters the abundance of the target message, processing of the primary transcript, egress of the CK-B message from the nucleus, or the polysome profile of CK-B mRNA in sucrose gradients. These results point to a direct effect of the antisense transcript on translation and suggest that this effect may be explained at least in part by an inhibition of elongation or termination as a consequence of the duplex formed in the distal coding and/or 3' noncoding region.

Expression of macrophage-lymphocyte Fc receptors in Madin-Darby canine kidney cells: polarity and transcytosis differ for isoforms with or without coated pit localization domains

Many cells of the immune system and certain epithelia express receptors for the Fc domain of IgG (FcR). On mouse macrophages and lymphocytes, two distinct receptor isoforms have been identified, designated FcRII-B1 and FcRII-B2. The isoforms are identical except for an in-frame insertion of 47 amino acids in the cytoplasmic tail of FcRII-B1 that blocks its ability to be internalized by clathrin-coated pits. We have recently found that at least one IgG-transporting epithelium, namely placental syncytial trophoblasts, expresses transcripts encoding a receptor similar or identical to macrophage-lymphocyte FcRII. To determine whether FcRII of hematopoietic cells might also function as a transcytotic receptor if expressed in epithelial cells, FcRII-B1 and -B2 were transfected into Madin-Darby canine kidney (MDCK) cells and grown on permeable filter units. The two FcRII isoforms exhibited different patterns of polarized expression: FcRII-B1 was localized mainly to the apical plasma membrane domain, whereas FcRII-B2 was found predominantly on the basolateral surface. As expected for FcR in placenta, FcRII-B2 and to a lesser extent FcRII-B1 mediated transcellular transport of IgG-complexes from the apical to the basolateral plasma membrane. Neither receptor mediated transcytosis in the opposite direction, although FcRII-B2 also delivered ligand to lysosomes when internalized from either the basolateral or apical domains. Furthermore, FcRII-B2 was capable of transporting monovalent antireceptor antibody Fab fragments across the cell, suggesting that transcytosis was not dependent on receptor cross-linking. These findings suggest the possibility that FcRII can mediate transepithelial IgG transport when expressed in placental syncytial trophoblasts in addition to its "classical" endocytic and signaling activities when expressed in macrophages. Because FcRII-B1 and -B2 are expressed with distinct polarities, the results also suggest that interactions with clathrin-coated pits may play a role in generating the polarized distribution of at least some plasma membrane proteins in MDCK cells.

Production of human glucocerebrosidase in mice after retroviral gene transfer into multipotential hematopoietic progenitor cells

The human glucocerebrosidase (GC) gene has been transferred efficiently into spleen colony-forming unit (CFU-S) multipotential hematopoietic progenitor cells, and production of human GC RNA and protein has been achieved in transduced CFU-S colonies. High-titer retroviral vectors containing the human GC cDNA were constructed. Mouse bone marrow cells were stimulated with hematopoietic growth factors, infected by coculture with producer cells, and injected into lethally irradiated animals. Four vectors were compared with respect to gene-transfer efficiency into CFU-S progenitors. One vector (G vector) required high concentrations of interleukins 3 and 6 during stimulation and coculture for efficient transduction of CFU-S progenitors. The remaining three vectors (NTG, GTN, and GI vectors) transduced these progenitors at infection frequencies approaching 100% using low concentrations of hematopoietic growth factors to stimulate cell division prior to and during the infection. Vectors using the viral long terminal repeat enhancer/promoter to drive the human GC cDNA produced high levels of human GC RNA in the progeny of CFU-S progenitors after gene transfer. When an internal herpes simplex thymidine kinase promoter assisted by a mutant polyoma enhancer was used to drive the human GC cDNA (NTG vector), little or no human GC RNA was detected in transduced CFU-S colonies. All three vectors producing human GC RNA in CFU-S colonies can generate human GC as detected by immunochemical analysis of CFU-S colonies. NTG vector-infected bone marrow cells were transplanted into W/Wv recipients to generate long-term reconstituted mice. The capacity of the viral long terminal repeat and the internal thymidine kinase promoter to direct synthesis of RNA in transduced bone marrow and spleen cells 5 months after bone marrow transplantation reflected the performance of these promoters in NTG-transduced CFU-S colonies.

Characterization and expression of the human rhoH12 gene product

The rho genes constitute an evolutionarily conserved family having significant homology to the ras oncogene family. These genes have been found in Saccharomyces cerevisiae, Drosophila melanogaster, rat, and human; their 21,000-dalton products show strong conservation of structure. In humans, three classes of rho cDNA clones have been identified which differ by virtue of the presence of variable C-terminal domains: rhoH12, rhoH6, and rhoH9. The predicted 193 amino acids of human rhoH12 protein show 88% similarity with those of the human rhoH6 clone, 96.8% similarity with those of the Aplysia rho product, and 81.8% similarity with those of the yeast RHO1 protein. Rat-1 and NIH 3T3 mouse fibroblasts were transfected with clones containing the normal human rhoH12 allele as well as the variants encoding valine in place of the glycine and leucine in place of the glutamine normally found at residues 14 and 64, respectively. These replacements mirror the changes responsible for oncogenic activation of the related ras-encoded p21 proteins. These mutant rhoH12 clone alleles did not cause focus formation in monolayers or growth in soft agar. However, amplification of normal rhoH12 via cotransfection with a dihydrofolate reductase gene resulted in colonies that displayed reduced dependence on serum for growth, grew to higher saturation densities, and were tumorigenic when inoculated into nude mice. Normal p21rho protein was detected in the transfected cell lines as well as in normal cell lines by Western immunoblot and immunoprecipitation analysis with rabbit antibodies raised against the peptide corresponding to amino acids 122 to 135.

Fibroblasts transfected with Torpedo acetylcholine receptor beta-, gamma-, and delta-subunit cDNAs express functional receptors when infected with a retroviral alpha recombinant

Torpedo californica acetylcholine receptor (AChR) alpha-, beta-, gamma-, and delta-subunit cDNAs were each stably introduced into muscle and/or fibroblast cell lines using recombinant retroviral vectors and viral infection, or using SV-40 vectors and DNA-mediated cotransfection. The expressed proteins were characterized in terms of their molecular mass, antigenicity, posttranslational processing, cell surface expression, stability in fibroblasts, stability in differentiated and undifferentiated muscle cells, and ability (of alpha) to bind alpha-bungarotoxin (BuTx). We demonstrated that the alpha, beta, gamma, and delta polypeptides acquired one, one, two, and three units of oligosaccharide, respectively. If all four subunits were expressed in the same cell, fully functional cell surface AChRs were produced which had a Kd for BuTx of 7.8 X 10(-11) M. In contrast, subunits expressed individually were not detected on the surface of fibroblasts and the Kd for BuTx binding to individual alpha polypeptides was only approximately 4 X 10(-7) M. The half-lives of the alpha, gamma, and delta subunits at 37 degrees C were all found to be quite short (approximately 43 min), while the half-life of the beta subunit was found to be even shorter (approximately 12 min). The unique half-life of the beta subunit suggests that it might perform a key regulatory role in the process of AChR subunit assembly. One stable fibroblast cell line was established by transfection that expressed beta, gamma, and delta subunits simultaneously. When this cell line was infected with a retroviral alpha recombinant, fully functional cell surface AChRs were produced. The successful expression of this pentameric protein complex combining transfection and infection techniques demonstrates one strategy for stably introducing the genes of a heterologous multisubunit protein complex into cells.

Characterization and expression of the human rhoH12 gene product

The rho genes constitute an evolutionarily conserved family having significant homology to the ras oncogene family. These genes have been found in Saccharomyces cerevisiae, Drosophila melanogaster, rat, and human; their 21,000-dalton products show strong conservation of structure. In humans, three classes of rho cDNA clones have been identified which differ by virtue of the presence of variable C-terminal domains: rhoH12, rhoH6, and rhoH9. The predicted 193 amino acids of human rhoH12 protein show 88% similarity with those of the human rhoH6 clone, 96.8% similarity with those of the Aplysia rho product, and 81.8% similarity with those of the yeast RHO1 protein. Rat-1 and NIH 3T3 mouse fibroblasts were transfected with clones containing the normal human rhoH12 allele as well as the variants encoding valine in place of the glycine and leucine in place of the glutamine normally found at residues 14 and 64, respectively. These replacements mirror the changes responsible for oncogenic activation of the related ras-encoded p21 proteins. These mutant rhoH12 clone alleles did not cause focus formation in monolayers or growth in soft agar. However, amplification of normal rhoH12 via cotransfection with a dihydrofolate reductase gene resulted in colonies that displayed reduced dependence on serum for growth, grew to higher saturation densities, and were tumorigenic when inoculated into nude mice. Normal p21rho protein was detected in the transfected cell lines as well as in normal cell lines by Western immunoblot and immunoprecipitation analysis with rabbit antibodies raised against the peptide corresponding to amino acids 122 to 135.

Glycosylation and processing of high levels of active human glucocerebrosidase in invertebrate cells using a baculovirus expression vector

A human cDNA containing the complete coding region for the lysosomal glycoprotein glucocerebrosidase (EC 3.2.1.45) was introduced into the genome of Autographa californica nuclear polyhedrosis virus downstream from the polyhedrin promoter. Infection of Spodoptera frugiperda cells (SF9) with recombinant virus produced high levels of glucocerebrosidase, 40% of which was in the culture medium. The amino-terminal amino acid sequence of the recombinantly produced enzyme was identical to that of mature, human placental glucocerebrosidase, demonstrating that the signal sequence of the human preenzyme was recognized and appropriately removed in the SF9 invertebrate cells. The glucocerebrosidase in both the culture supernatant and SF9 cell pellet was glycosylated and contained, in part, high mannose oligosaccharide. These results demonstrate that insect cells can be used to produce abundant quantities of active mature human glucocerebrosidase that contains high mannose oligosaccharide as a consequence of post-translational processing.

Structural analysis of the human glucocerebrosidase genes

Two different genomic clones containing the entire coding sequence of human glucocerebrosidase were isolated from a fetal liver library using a cDNA probe previously cloned by us. These clones correspond to two human glucocerebrosidase genes, designated 6-1 and 10-2. Clone 6-1 contains sequences homologous to the cDNA we cloned previously. The promoter regions of the genes were identified by S1 analysis and sequenced. They contain TATA- and CAAT-like boxes, but lack a GGCGGG motif. When coupled to the bacterial gene coding for chloramphenicol acetyl transferase (CAT) and transfected to Gaucher skin fibroblast lines, both promoter fragments enhanced CAT activity. The promoter of gene 6-1 was eight times more efficient than the promoter of gene 10-2. Northern blot analysis revealed three human glucocerebrosidase RNA species of 6, 2.6, and 2.2 kb in size. The 6-kb transcript is probably a nuclear transcript whereas the 2.6-kb and 2.2-kb transcripts are cytoplasmic species which emerge from polyadenylation at different sites.

Iron regulates ferritin mRNA translation through a segment of its 5' untranslated region

In previous studies, we showed that acute administration of iron to intact rats or to rat hepatoma cells in culture induces synthesis of the iron-storage protein ferritin by activating translation of inactive cytoplasmic ferritin mRNAs for both the heavy (H) and the light (L) subunits. In the course of activation, these ferritin mRNAs are recruited onto polysomes. To elucidate the structural features of these mRNAs involved in the translational response to iron, a chimera was constructed from the 5' and 3' untranslated regions (UTRs) of ferritin L subunit mRNA fused to the reading frame of the mRNA of bacterial chloramphenicol acetyltransferase (CAT). This chimera and deletion constructs derived from it were introduced into a rat hepatoma cell line by retrovirus-mediated gene transfer. The complete chimera showed increased CAT activity in response to iron enrichment of the medium, whereas deletion of the first 67 nucleotides of the 5' UTR, which contain a highly conserved sequence, caused loss of regulation by iron. Whereas cis-acting sequences located in the 5' flanking regions of many genes have been repeatedly implicated in modulating their transcriptional expression, we report here a specific regulatory translational sequence found within the 5' UTR of a eukaryotic mRNA.

Signal sequence and DNA-mediated expression of human lysosomal alpha-galactosidase A

Twelve complementary DNA clones for human lysosomal alpha-galactosidase A were isolated from an Okayama-Berg library constructed from SV40-transformed human fibroblasts. The identity of these clones was confirmed by complete colinearity of the nucleotide-deduced amino acid sequence with that determined by direct chemical sequencing of human placental alpha-galactosidase A. Hybridization of the alpha-galactosidase A cDNA to genomic DNA from individuals with varying numbers of X chromosomes as well as from interspecies somatic-cell hybrids showed only a single locus in the genome at Xq 13.1-Xq 22. One cDNA clone (pcD-AG210) contained the complete coding sequence for both the signal peptide and mature alpha-galactosidase A. The signal peptide of 31 amino acids contains the expected hydrophobic domains consisting of Leu-Gly-Cys-Ala-Leu-Ala-Leu and Phe-Leu-Ala-Leu-Val and has Ala at the signal peptidase cleavage site. Twelve out of fifteen G residues flanking the 5' end of the cDNA in pcD-AG210 were removed and the truncated fragment was ligated into the original vector. This construct, pcD-AG502, encoded enzymatically active human alpha-galactosidase A in monkey COS cells.