The putative tumour suppressor EXT1 alters the expression of cell-surface heparan sulfate.

Hereditary multiple exostoses (HME) is an autosomal dominant disorder characterized by the formation of cartilage-capped tumours (exostoses) that develop from the growth plate of endochondral bone. This condition can lead to skeletal abnormalities, short stature and malignant transformation of exostoses to chondrosarcomas or osteosarcomas. Linkage analyses have identified three different genes for HME, EXT1 on 8q24.1, EXT2 on 11p11-13 and EXT3 on 19p (refs 6-9). Most HME cases have been attributed to missense or frameshift mutations in these tumour-supressor genes, whose functions have remained obscure. Here, we show that EXT1 is an ER-resident type II transmembrane glycoprotein whose expression in cells results in the alteration of the synthesis and display of cell surface heparan sulfate glycosaminoglycans (GAGs). Two EXT1 variants containing aetiologic missense mutations failed to alter cell-surface glycosaminoglycans, despite retaining their ER-localization.

Length of huntingtin and its polyglutamine tract influences localization and frequency of intracellular aggregates.

It is unclear how polyglutamine expansion is associated with the pathogenesis of Huntington disease (HD). Here, we provide evidence that polyglutamine expansion leads to the formation of large intracellular aggregates in vitro and in vivo. In vitro these huntingtin-containing aggregates disrupt normal cellular architecture and increase in frequency with polyglutamine length. Huntingtin truncated at nucleotide 1955, close to the caspase-3 cleavage site, forms perinuclear aggregates more readily than full-length huntingtin and increases the susceptibility of cells to death following apoptotic stimuli. Further truncation of huntingtin to nucleotide 436 results in both intranuclear and perinuclear aggregates. For a given protein size, increasing polyglutamine length is associated with increased cellular toxicity. Asymptomatic transgenic mice expressing full-length huntingtin with 138 polyglutamines form exclusively perinuclear aggregates in neurons. These data support the hypothesis that proteolytic cleavage of mutant huntingtin leads to the development of aggregates which compromise cell viability, and that their localization is influenced by protein length.

Identification and characterization of a mouse cell mutant defective in the intracellular transport of glycoproteins.

We have isolated a mutant line of mouse L cells, termed gro29, in which the growth of herpes simplex virus (HSV) and vesicular stomatitis virus (VSV) is defective. The block occurs late in the infectious cycle of both viruses. We demonstrate that HSV and VSV enter gro29 cells normally, negotiate the early stages of infection, yet are impaired at a late stage of virus maturation. During VSV infection of the mutant cell line, intracellular transport of its glycoprotein (G protein) is slowed. Pulse-chase experiments showed that oligosaccharide processing is impeded, and immunofluorescence localization revealed an accumulation of G protein in a juxtanuclear region that contains the Golgi complex. We conclude that export of newly made glycoproteins is defective in gro29 cells, and speculate that this defect may reflect a lesion in the glycoprotein transport apparatus.

Selective destruction of gliomas in immunocompetent rats by thymidine kinase-defective herpes simplex virus type 1.

BACKGROUND: Thymidine kinase-deficient herpes simplex virus type 1 [tk(-) HSV-1] replicates well in dividing cells but not in nondividing cells such as neurons, suggesting a potential use in the treatment of brain tumors. PURPOSE: We attempted to examine the efficacy of using tk(-) HSV-1 for treating brain tumors in immunocompetent animals. METHODS: 9L glioma cells were cultured and subsequently implanted intracerebrally in immunocompetent, adult male Long-Evans rats. A thymidine kinase-defective HSV-1 virus, KOS-SB, was used to infect 9L cells in culture, and the viability of the infected cells was compared with that of mock-infected (i.e., uninfected) cells. We also injected the virus intratumorally and determined the mortality of the tumor-bearing animals. Tumor regression and viral spread following virus injection were examined by histologic and immunocytochemical assays. RESULTS: In vitro, the tk(-) virus destroyed cultured 9L cell monolayers at multiplicities of infection of 0.1 and 1.0 within 48 hours. With the same quantity of virus, no remarkable difference in survival of neural cells was found. Foscarnet, an antiviral drug that acts independently of tk activity, blocked viral replication by greater than 99% at a concentration of 100 micrograms/mL. The mortality of animals bearing tumors declined with an increase in the amount of virus injected. Histologic examination showed that the HSV-1 treatment caused severe tumor regression. Immunocytochemistry using an anti-HSV-1 antibody revealed only a weak staining within the regressing tumors, and few immunopositive neurons were evident in the surrounding brain tissue. CONCLUSIONS: The results indicate that tk(-) HSV-1 mutants can selectively and effectively destroy glioma cells both in vitro and in vivo in normal, immunocompetent animals. IMPLICATIONS: Our failure to detect viral spread associated with regressing tumors suggests that some other cytopathic factors might be involved in the tumor regression. Regardless of the precise mode of tumor cell killing, HSV-1 may be useful for treating brain tumors.

The varicella zoster virus glycoprotein B (gB) plays a role in virus binding to cell surface heparan sulfate proteoglycans.

Varicella-zoster virus (VZV) interacts with cell surface heparan sulfate proteoglycans during virus attachment. In the present study, we investigated the potential involvement of two VZV glycoproteins, gB and gE, in the virus adsorption process. We showed that gB, but not gE, binds specifically to cellular heparan sulfate proteoglycans (HSPGs). Indeed, soluble recombinant gB protein (recgB) was found to bind to immobilized heparin and to MRC5 and L cells, a binding which was inhibited by heparin. Furthermore, recgB binding to two heparan sulfate-minus mutant L cell lines, gro2C and sog9 cells, was markedly reduced as compared to the parental L strain. Under the same experimental conditions, soluble recombinant VZV gE protein did not interact with heparin or with cell surfaces. We also demonstrated that the gB-HSPGs interactions were relevant to the VZV attachment to cells. Indeed, although polyclonal antibodies directed to gB did not impair the VZV binding, recgB could delay the virus adsorption. Our results thus strongly suggest that the interactions between gB and heparan sulfate proteoglycans take part in the initial VZV attachment to cell surfaces.

Serological distance coefficients.

We define experimentally measurable similarity coefficients S[A,B/C] that specify the extent to which a pair of substances, A and B, are similar in the context of a diverse reagent, C, which can be a non-immune serum. We describe how this definition can be applied to mixtures of antibodies, for example the antibodies of two serum samples, A and B. A "distance coefficient" D[A,B/C] between two sera, A and B, in the context of C, is defined as one minus the similarity coefficient of the two sera in the same context. We discuss the problem of the experimental measurement of these coefficients, and the possible use of the coefficients in the diagnosis and prognosis of disease conditions.

A bcl-2 expressing viral vector protects cortical neurons from excitotoxicity even when administered several hours after the toxic insult.

The product of the bcl-2 oncogene has been shown to play an important role in apoptosis and programmed cell death. In this study, a herpes simplex virus type-1 vector was constructed to carry the human bcl-2 gene. The possible role of bcl-2 in protecting neurons from excitoxicity was investigated by using the viral vector to deliver the gene into neuronal cultures before or after the cells were exposed to glutamate under conditions in which 50-80% of neurons died. Infection with the bcl-2 expressing vector 24 h prior to glutamate treatment effectively prevented the cell death that normally follows this treatment. Moreover, infection with the vector as late as 8 h after the glutamate insult still resulted in substantial neuroprotective effects. These results have potential implications for new therapies in stroke or ischemic neuropathies.

A perspective of gene therapy in the glaucomas.

Gene therapy in the anterior and posterior segment tissues may have the potential to favorably influence aqueous hydrodynamics and retinal ganglion cell biology, thereby preventing, delaying, or minimizing glaucomatous damage to the optic nerve. We demonstrated the feasibility of using a herpes viral vector (ribonucleotide reductase defective HSV-1, hrR3) to deliver the lacZ reporter gene to living cat and rat eyes. Cats received injections into the anterior chamber and rats into the vitreous cavity. In cats, lacZ expression was detectable at 1 to 2 days in the anterior outer portion of the ciliary muscle and the lining of the intertrabecular spaces of the corneoscleral and uveal meshwork. Rat eyes showed lacZ expression in the retinal pigment epithelium and photoreceptor outer segments 2 days after injection.

Examination of conditions affecting the efficiency of HVS-1 amplicon packaging.

Defective herpes simplex virus type 1 vectors (HSV amplicons) have been used as vehicles for efficiently delivering foreign genes into non-dividing cells such as neutrons in vitro and in vivo. This system is useful for studying neuronal physiology and may have potential for human gene therapy of neuronal disorders. The preparation of infectious amplicon particles is normally achieved by transfecting amplicon plasmid DNA, which contains the HSV replication origin and packaging signal, into mammalian cell lines followed by infection of the cells with HSV helper virus. This allows for replication and packaging of both viral and amplicon plasmid DNA. To improve the packaging efficiency of amplicons, several parameters involved in the packaging process were investigated. By introducing the SV40 DNA replication origin into an amplicon plasmid and prereplicating it before HSV infection, it was demonstrated that the existing amount of amplicon DNA prior to infection in the cells is not a rate-limiting step during HSV packaging. In addition, it was shown that the yield of the packaged amplicon particles can be improved by: (1) using a relatively small amount of HSV helper virus up to multiplicity of infection (m.o.i.) equal to 0.1 at infection; (2) infecting with HSV helper virus at 2 or 3 days post-transfection; and (3) passaging the initial packaged amplicon stocks 1-2 times on fresh host cells.

Replicating herpes simplex virus vectors for cancer gene therapy.

Attenuated viral vectors based on herpes simplex virus (HSV) are capable of killing cancer cells directly while sparing normal tissue in animal models of disease. This selective ability is likely due to the evolutionary constraints on the virus to establish lifelong infection in its host without causing destruction of normal tissues. However, extensive experimental animal data show that cancer cells are able to sustain a productive viral infection, which ultimately leads to cell death and tumour regression. Moreover, preliminary results generated in two Phase I clinical studies of modified replicating HSV for the treatment of brain tumours (e.g., glioblastoma multiforme) have been encouraging and suggest that the safety data generated in animals are predictive of human safety. Although much progress has been made in developing oncolytic HSV vectors for clinical use, there is still a long way to go to determine which combinations of virus, surgery, radiation and chemotherapy will provide improved therapy for the control and eradication of a variety of human cancers.

Herpes simplex virus particles are unable to traverse the secretory pathway in the mouse L-cell mutant gro29.

The mouse L-cell mutant gro29 was selected for its ability to survive infection by herpes simplex virus type 1 (HSV-1) and is defective in the propagation of HSV-1 and vesicular stomatitis virus (F. Tufaro, M. D. Snider, and S. L. McKnight, J. Cell Biol. 105:647-657, 1987). In this report, we show that gro29 cells harbor a lesion that inhibits the egress of HSV-1 virions during infection. We also found that HSV-1 glycoprotein D was slow to traverse the secretory pathway en route to the plasma membrane of infected gro29 cells. The movement of glycoproteins was not blocked entirely, however, and immunofluorescence experiments revealed that infected gro29 cells contained roughly 10% of the expected amount of glycoprotein D on their cell surface at 12 h postinfection. Furthermore, nucleocapsids and virions assembled inside the cells during infection, suggesting that the lesion in gro29 cells impinged on a late step in virion maturation. Electron micrographs of infected cells revealed that many of the intracellular virions were contained in irregular cytoplasmic vacuoles, similar to those that accumulate in HSV-1-infected cells treated with the ionophore monensin. We conclude from these results that gro29 harbors a defect that blocks the egress of HSV-1 virions from the infected cell without seriously impeding the flux of individual glycoproteins to the cell surface. We infer that HSV-1 maturation and egress require a host cell component that is either reduced or absent in gro29 cells and that this lesion, although not lethal to the host cell, cannot be tolerated by HSV-1 during its life cycle.

The influence of N-linked glycosylation on the antigenicity and immunogenicity of rubella virus E1 glycoprotein.

Rubella virus E1 glycoprotein contains three functional N-linked glycosylation sites. The role of N-linked glycosylation on the antigenicity and immunogenicity of E1 glycoprotein was studied using vaccinia recombinants expressing E1 glycosylation mutants. Expressed E1 glycosylation mutant proteins were recognized by a panel of E1-specific monoclonal antibodies in radioimmunoprecipitation, immunofluorescence, and immunoblotting, indicating that carbohydrate side chains on E1 are not involved in the constitution of epitopes recognized by these monoclonal antibodies. This observation was further supported by the fact that removal of oligosaccharides on E1 by glycosidase digestion did not significantly change the antigenicity of E1. All the glycosylation mutants were capable of eliciting anti-RV E1 antibodies. The single glycosylation mutants (G1, G2, and G3), but not the double mutant (G23) or the triple mutant (G123), were found to be capable of inducing virus neutralizing antibodies. Among the single glycosylation mutants, only G2 and G3 were active in producing hemagglutination inhibition antibodies in mice. Our findings suggest that although carbohydrate on E1 is not directly involved in the antigenic structures of E1, it is important in maintaining proper protein folding and stable conformation for expression of immunological epitopes on E1.

Structural and functional analysis of an alternatively spliced chicken TK messenger RNA.

The nucleotide sequence of independent cDNA clones revealed that TK transcripts can undergo alternative splicing within the 3' nontranslated portion of the message. The most abundant mRNA is 2.1 kb in length and is colinear with the underlying genomic sequence for at least the first 1052 bases of its 3' nontranslated region. A less abundant mRNA, 1.3 kb in length, differs from the larger mRNA in that an 863 base intron is spliced from the 3' nontranslated region. The possibility that removal of the alternatively spliced intron allows the small mRNA to persist in postreplicative cells was investigated in two ways. First, the levels of large and small TK mRNA in tissues expressing differential growth rates was determined. Second, the pattern of TK enzyme expression during differentiation was analyzed in muscle cells transformed with recombinant TK genes lacking the 3' nontranslated intron. Both lines of experimentation indicated that the small TK mRNA was as dependent as the large TK mRNA on the replicative state of the cell.

Herpes simplex virus: discovering the link between heparan sulphate and hereditary bone tumours.

To gain entry into the host, viruses use host cell surface molecules that normally serve as receptors for other ligands. Herpes simplex virus type 1 (HSV-1) uses heparan sulphate (HS) glycosaminoglycans (GAGs) as receptors for initial attachment to the host cell surface. HS GAGs are both ubiquitous and structurally diverse, and normally serve as critical mediators of interactions between the cell and the extracellular environment. We have used the HS binding ability of HSV-1 to identify the function of a cellular gene, EXT1, which is involved in HS polymerisation. Cellular factors that affect virus growth and replication are often key regulators of the cell cycle and EXT1 is no different-humans with inherited mutations in EXT1 have developmental defects that lead to bone tumours (hereditary multiple exostoses, HME) and sometimes chondrosarcomas. Thus, as a result of using HSV-1 as a molecular probe, a functionally orphaned disease gene now has a defined function. These findings highlight the utility of viruses for investigating important cellular processes.

Brefeldin A and monensin arrest cell surface expression of membrane glycoproteins and release of rubella virus.

The maturation of rubella virus (RV) glycoproteins E2 and E1 was examined by using brefeldin A (BFA) and monensin. BFA, which induces the rapid redistribution of Golgi enzymes residing in the Golgi complex into the endoplasmic reticulum (ER), was used to locate the intracellular site for the modification of carbohydrate side-chains on RV E1 and E2 proteins. The monovalent ionophore monensin, which inhibits intracellular transport of proteins through the ER-Golgi complex, was used to block the transport of E1 and E2 glycoproteins through the Golgi complex. BFA and monensin effectively blocked the cell surface expression of RV E2 and E1 proteins, secretion of an anchor-free form of E2 and budding of RV from the plasma membrane. For O-linked glycosylation, addition of N-acetylgalactosamine and galactose to E2 protein was found to take place in the medial to the trans Golgi. A dramatic change in the intracellular distribution of RV structural proteins was observed when transfected COS cells were treated with BFA or monensin, although the proteolytic processing of RV structural protein precursor was not affected. In the presence of BFA or monensin, virus release from infected Vero cells was only 0.1% of the intracellular virus, and the intracellular virus titre decreased as well. Our results suggest that O-linked glycosylation on the E2 protein occurred in the post-ER region and the transport of RV structural proteins to the Golgi complex and post-Golgi compartment may be a rate-limiting step in RV assembly and budding.

New perspectives on the molecular basis of hereditary bone tumours.

Bone development is a highly regulated process sensitive to a wide variety of hormones, inflammatory mediators and growth factors. One of the most common hereditary skeletal dysplasias, hereditary multiple exostoses (HME), is an autosomal dominant disorder characterized by skeletal malformations that manifest as bony, benign tumours near the end of long bones. HME is usually caused by defects in either one of two genes, EXT1 and EXT2, which encode enzymes that catalyse the biosynthesis of heparan sulphate, an important component of the extracellular matrix. Thus, HME-linked bone tumours, like many other skeletal dysplasias, probably result from disruptions in cell surface architecture. However, despite the recent success in unravelling functions for several members of the EXT gene family, significant challenges remain before this knowledge can be used to develop new approaches for the diagnosis and treatment of disease.

Post-transcriptional restriction of gene expression in sea urchin interspecies hybrid embryos.

The synthesis of many paternal species-specific proteins is reduced in all stages of sea urchin interspecies hybrid embryos, due to the reduced amounts of some paternal mRNA species in hybrid embryos compared with the embryos of the paternal species (Tufaro and Brandhorst 1982). Possible explanations for this restriction were tested. Cloned cDNAs were selected that were specific for paternal RNA sequences having reduced amounts (to 2-20% of normal) in hybrid embryos derived from a cross of Stronglyocentrotus purpuratus eggs with Lytechinus pictus sperm. Several of these RNA species are barely detectable in the eggs, but they accumulate extensively (5- to 40-fold) during L. pictus embryogenesis. Thus, the restricted expression of these paternal genes in hybrid embryos is not the result of the persistence of stable maternal mRNA species stored in eggs and not replaced by zygotic transcription. The accumulation of some of these L. pictus transcripts is also reduced in the reciprocal cross (L. pictus eggs X S. purpuratus sperm); therefore, the full expression of these L. pictus genes in hybrid embryos is not dependent on species-specific maternal factors stored in the egg. The transcriptional activity of one such gene was estimated using a run-on assay in isolated nuclei; it is as actively transcribed in hybrid as it is in homospecific embryos, but in hybrid embryos the cytoplasmic transcript accumulates to only 2-15% of the normal level. Sequence analysis indicates that this gene encodes a metallothionein. Mechanisms are discussed that might account for the post-transcriptional restriction of expression of some genes in hybrid embryos.