Coexpression after electroporation of plasmid mixtures into muscle in vivo.

AIM: Muscle is perhaps the most frequently considered tissue for non-viral gene therapy, in particular after gene transfer by electroporation. Expression in muscle is stable, but since the cell turnover is so slow incorporation in the host genome is not required. This raises interesting practical and theoretical questions related to the behaviour of the transgenic DNA under such conditions. METHODS: We have investigated expression of reporter genes from plasmid mixtures electroporated into the extensor digitorum longus (EDL) muscle in mice in order to assess the degree of coexpression. RESULTS: Under conditions where the reporter is easily identified the coexpression rate was 100%, as none of 287 fibres from five different muscles expressing blue fluorescent protein (BFP) failed to express green fluorescent protein (GFP). With other reporter combinations the rate was lower, but this we attribute to marginal sensitivity for fluorescent proteins, or from reporter protein degradation for beta-galactosidase. CONCLUSIONS: The high degree of coexpression suggests that a large copy number takes part in the final transcription with this system. The finding also enhances the usefulness of muscle and electroporation for gene therapy and experimental biology.

Number and spatial distribution of nuclei in the muscle fibres of normal mice studied in vivo.

We present here a new technique with which to visualize nuclei in living muscle fibres in the intact animal, involving injection of labelled DNA into single cells. This approach allowed us to determine the position of all of nuclei within a sarcolemma without labelling satellite cells. In contrast to what has been reported in tissue culture, we found that the nuclei were immobile, even when observed over several days. Nucleic density was uniform along the fibre except for the endplate and some myotendinous junctions, where the density was higher. The perijunctional region had the same number of nuclei as the rest of the fibre. In the extensor digitorum longus (EDL) muscle, the extrajunctional nuclei were elongated and precisely aligned to the long axis of the fibre. In the soleus, the nuclei were rounder and not well aligned. When comparing small and large fibres in the soleus, the number of nuclei varied approximately in proportion to cytoplasmic volume, while in the EDL the number was proportional to surface area. Statistical analysis revealed that the nuclei were not randomly distributed in either the EDL or the soleus. For each fibre, actual distributions were compared with computer simulations in which nuclei were assumed to repel each other, which optimizes the distribution of nuclei with respect to minimizing transport distances. The simulated patterns were regular, with clear row-like structures when the density of nuclei was low. The non-random and often row-like distribution of nuclei observed in muscle fibres may thus reflect regulatory mechanisms whereby nuclei repel each other in order to minimize transport distances.