Repair of CFTR mRNA by spliceosome-mediated RNA trans-splicing.

Most messenger RNA precursors (pre-mRNA) undergo cis-splicing in which introns are excised and the adjoining exons from a single pre-mRNA are ligated together to form mature messenger RNA. This reaction is driven by a complex known as the spliceosome. Spliceosomes can also combine sequences from two independently transcribed pre-mRNAs in a process known as trans-splicing. Spliceosome-mediated RNA trans-splicing (SMaRT) is an emerging technology in which RNA pre-therapeutic molecules (PTMs) are designed to recode a specific pre-mRNA by suppressing cis-splicing while enhancing trans-splicing between the PTM and its pre-mRNA target. This study examined the feasibility of SMaRT as a potential therapy for genetic diseases to correct mutations using cystic fibrosis (CF) as an example. We used several versions of a cystic fibrosis transmembrane conductance regulator (CFTR) mini-gene expressing mutant (deltaF508) pre-mRNA targets and tested this against a number of PTMs capable of binding to the CFTR target intron 9 and trans-splicing in the normal coding sequences for exons 10-24 (containing F508). When 293T cells were cotransfected with both constructs, they produced a trans-spliced mRNA in which normal exon 10-24 replaced mutant exon 10. To test whether SMaRT produced mature CFTR protein, proteins were immunoprecipitated from lysates of cotransfected cells and detected by Western blotting and PKA-phosphorylation. Tryptic phosphopeptide mapping confirmed the identity of CFTR. This proof-of-concept study demonstrates that exon replacement by SMaRT can repair an abnormal pre-mRNA associated with a genetic disease.

Spliceosome-mediated RNA trans-splicing as a tool for gene therapy.

We have developed RNA molecules capable of effecting spliceosome-mediated RNA trans-splicing reactions with a target messenger RNA precursor (pre-mRNA). Targeted trans-splicing was demonstrated in a HeLa nuclear extract, cultured human cells, and H1299 human lung cancer tumors in athymic mice. Trans-splicing between a cancer-associated pre-mRNA encoding the beta-subunit of human chorionic gonadotropin gene 6 and pre-trans-splicing molecule (PTM) RNA was accurate both in vitro and in vivo. Comparison of targeted versus nontargeted trans-splicing revealed a moderate level of specificity, which was improved by the addition of an internal inverted repeat encompassing the PTM splice site. Competition between cis- and trans-splicing demonstrated that cis-splicing can be inhibited by trans-splicing. RNA repair in a splicing model of a nonfunctional lacZ transcript was effected in cells by a PTM, which restored significant beta-galactosidase activity. These observations suggest that spliceosome-mediated RNA trans-splicing may represent a general approach for reprogramming the sequence of targeted transcripts, providing a novel approach to gene therapy.

Molecular cloning and characterization of an invertebrate cellular retinoic acid binding protein.

We have cloned a cDNA and gene from the tobacco hornworm, Manduca sexta, which is related to the vertebrate cellular retinoic acid binding proteins (CRABPs). CRABPs are members of the superfamily of lipid binding proteins (LBPs) and are thought to mediate the effects of retinoic acid (RA) on morphogenesis, differentiation, and homeostasis. This discovery of a Manduca sexta CRABP (msCRABP) demonstrates the presence of a CRABP in invertebrates. Compared with bovine/murine CRABP I, the deduced amino acid sequence of msCRABP is 71% homologous overall and 88% homologous for the ligand binding pocket. The genomic organization of msCRABP is conserved with other CRABP family members and the larger LBP superfamily. Importantly, the promoter region contains a motif that resembles an RA response element characteristic of the promoter region of most CRABPs analyzed. Three-dimensional molecular modeling based on postulated structural homology with bovine/murine CRABP I shows msCRABP has a ligand binding pocket that can accommodate RA. The existence of an invertebrate CRABP has significant evolutionary implications, suggesting CRABPs appeared during the evolution of the LBP superfamily well before vertebrate/invertebrate divergence, instead of much later in evolution in selected vertebrates.

Molecular organization and alternative splicing in zipper, the gene that encodes the Drosophila non-muscle myosin II heavy chain.

Genomic sequence of the entire zipper gene, that encodes non-muscle myosin II heavy chain (MHC) in Drosophila melanogaster, reveals a new, differentially spliced exon in this essential locus and identifies a molecular lesion that is responsible for a severe embryonic lethal zipper allele. There are two alternative splices in the head domain. The first is present in the 5' untranslated sequence which, when employed, produces an N-terminal extension of 45 amino acids (aa). This splicoform produces a protein that is stable in flies but less prevalent than the isoform that lacks the extension. The second alternative exon (40 aa) is close to the nucleotide binding pocket. The position, size and sequence of this exon is conserved in D. simulans and putative alternative exons of different size (7 to 16 aa) but identical position have been reported for other myosins throughout phylogeny. The functional significance of neither alternative splice is clear. Sequence analysis of genomic DNA identifies the lesion responsible for zipIIF107, one of the original severe embryonic lethal zipper alleles. Our primary structural data confirm and correct our previous sequence of the cDNA, establish the spatial relationship between zipper and unzipped (the gene originally thought to have been disrupted in zipper mutations), and provide a high resolution template for the precise mapping of mutations.

A phosphorylation epitope on MAP 1B that is transiently expressed in growing axons in the developing rat nervous system.

We have isolated a monoclonal antibody (150) that recognizes a phosphorylation epitope on the microtubule-associated protein (MAP) 1B. Immunoblot analysis of the developing rat central nervous system shows that monoclonal antibody 150 is directed against a protein of approximately 325 kDa (MAP 1B) that copolymerizes with microtubules through successive cycles of temperature-dependent assembly and disassembly. Furthermore, immunoprecipitated MAP 1B contains the epitope recognized by monoclonal antibody 150. Removal of phosphate from blotted proteins using alkaline phosphatase abolishes the binding of monoclonal antibody 150 to MAP 1B, indicating that the epitope is phosphorylated. In the developing rat nervous system, immunohistochemistry with monoclonal antibody 150 shows that the phosphorylation epitope on MAP 1B is transiently expressed in growing axons but not in dendrites. For instance, in the neonatal rat cerebellum, the parallel fibres of granule cells are stained only during elongation and not after synaptogenesis. The monoclonal antibody 150 epitope is also transiently expressed in radial glial fibres and in certain cell nuclei. All immunostaining of sections with monoclonal antibody 150 was completely abolished by alkaline phosphatase treatment. These observations and previous ones made by us in cell culture (Mansfield et al., J. Neurocytol., 20, 654-666, 1991) suggest that the phosphorylation epitope on MAP 1B recognized by monoclonal antibody 150, which has not been previously detected in vivo, may be important in axonogenesis.

The organization of F-actin and microtubules in growth cones exposed to a brain-derived collapsing factor.

In previous work we characterized a brain derived collapsing factor that induces the collapse of dorsal root ganglion growth cones in culture (Raper and Kapfhammer, 1990). To determine how the growth cone cytoskeleton is rearranged during collapse, we have compared the distributions of F-actin and microtubules in normal and partially collapsed growth cones. The relative concentration of F-actin as compared to all proteins can be measured in growth cones by rationing the intensity of rhodamine-phalloidin staining of F-actin to the intensity of a general protein stain. The relative concentration of F-actin is decreased by about one half in growth cones exposed to collapsing factor for five minutes, a time at which they are just beginning to collapse. During this period the relative concentration of F-actin in the leading edges of growth cones decreases dramatically while the concentration of F-actin in the centers decreases little. These results suggest that collapse is associated with a net loss of F-actin at the leading edge. The distributions of microtubules in normal and collapsing factor treated growth cones were examined with antibodies to tyrosinated and detyrosinated isoforms of alpha-tubulin. The tyrosinated form is found in newly polymerized microtubules while the detyrosinated form is not. The relative proximal-distal distributions of these isoforms are not altered during collapse, suggesting that rates of microtubule polymerization and depolymerization are not greatly affected by the presence of collapsing factor. An analysis of the distributions of microtubules before and after collapse suggests that microtubules are rearranged, but their polymerization state is unaffected during collapse. These results are consistent with the hypothesis that the brain derived collapsing factor has little effect on microtubule polymerization or depolymerization. Instead it appears to induce a net loss of F-actin at the leading edge of the growth cone.

The distribution and phosphorylation of the microtubule-associated protein MAP 1B in growth cones.

Primary cultures of dissociated embryonic day 18 rat cerebral cortices were labelled by immunofluorescence with antibodies directed either against phosphorylated and non-phosphorylated MAP 1B (antibody 81) or against phosphorylated MAP 1B (antibody 150). Both antibodies stain cortical neurons, including their neurites and growth cones, in early (18 h) cultures, whereas only antibody 81 stained glial cells. By 4 days in culture, phosphorylated MAP 1B is largely restricted to axonal processes and growth cones, where it is often distributed in a gradient that is highest distally. In axonal processes and growth cones after 18 h and 4 days in culture, the phosphorylated form of MAP 1B is present both in a soluble form and bound to microtubules. Growth cones isolated from postnatal day 5 rat forebrain were labelled in vitro with 32P-orthophosphate and detergent soluble and insoluble (cytoskeleton) fractions prepared. SDS-PAGE analysis revealed several major phosphoproteins in isolated growth cone cytoskeletons, including MAP 1B. Phosphorylated MAP 1B was also present in the detergent soluble fraction of growth cones. Immunoblotting and immunoprecipitation with MAP 1B antibodies confirmed the identification of MAP 1B and that the protein is phosphorylated in growth cones. These data show that MAP 1B, in particular the phosphorylated isoform, is present in growth cones and suggest that phosphorylation of MAP 1B may play an important role in neurite elongation.

Dynamic post-translational modification of tubulin in rat cerebral cortical neurons extending neurites in culture: effects of taxol.

Dissociated embryonic (E18-E20) rat cortical neurons were grown in culture and double-labelled by immunofluorescence with antibodies directed against tyrosinated (YL 1/2), detyrosinated (SUP GLU), and acetylated (6-11B-1) alpha-tubulin. Within 90 min of plating, neurons extended growth cones that were YL 1/2+ but SUP GLU- and 6-11B-1-. The neurite that forms behind the advancing growth cone is also, initially, YL 1/2+ and SUP GLU-/6-11B-1-. However, when it has attained a length of about half the cell body diameter, it becomes SUP GLU+ and 6-11B-1+. The effects of the microtubule polymerizing agent taxol (15 microM) on growth cone and neurite alpha-tubulins was investigated. Taxol, as reported previously, caused the formation of microtubule loops in the central domain of the growth cone, a loss of filopodia, and the collapse of the growth cone onto the loops. The taxol effects peaked at 60 min, when over 85% of neurites showed microtubule loops, and declined thereafter, so that at 420 min in taxol, only about 23% of neurites had microtubule loops. Over this period there was an inverse correlation between the presence of microtubule loops and growth cones. Taxol had striking effects on the intensity of SUP GLU and 6-11B-1 staining in neurons. In 48 h cultures, a 30 min exposure to taxol enhanced the SUP GLU and 6-11B-1 staining of dendrites and axons and produced a loss of YL 1/2 staining in axons. Immunoblotting experiments confirmed that there was an overall reduction in YL 1/2 immunoreactivity and an increase in SUP GLU immunoreactivity. These observations support previous suggestions that the neurite microtubules are assembled in the growth cone and post-translationally modified in the neurite and, in addition, imply that growth cones can overcome the effects of taxol in the continued presence of the compound.

Direct visualisation of the soluble pool of tubulin in the neuronal growth cone: immunofluorescence studies following taxol polymerisation.

Previously we have shown that the predominant form of soluble alpha-tubulin in the growth cone is C-terminally tyrosinated (Gordon-Weeks and Lang. Dev. Brain Res., 42 (1988) 156-160). Here we show that when growth cones are incubated in taxol, this soluble pool of alpha-tubulin is polymerised onto the ends of the neurite microtubules that enter the proximal part of the growth cone indicating that it is assembly-competent.