Inhibition of gene expression by a short sense fragment

Minicircle DNA vectors achieve sustained expression reflected by active chromatin and transcriptional level

Current efforts in nonviral gene therapy are plagued by a pervasive difficulty in sustaining therapeutic levels of delivered transgenes. Minicircles (plasmid derivatives with the same expression cassette but lacking a bacterial backbone) show sustained expression and hold promise for therapeutic use where persistent transgene expression is required. To characterize the widely-observed silencing process affecting expression of foreign DNA in mammals, we used a system in which mouse liver presented with either plasmid or minicircle consistently silences plasmid but not minicircle expression. We found that preferential silencing of plasmid DNA occurs at a nuclear stage that precedes transport of mRNA to the cytoplasm, evident from a consistent >25-fold minicircle/plasmid transcript difference observed in both nuclear and total RNA. Among possible mechanisms of nuclear silencing, our data favor chromatin-linked transcriptional blockage rather than targeted degradation, aberrant processing, or compromised mRNA transport. In particular, we observe dramatic enrichment of H3K27 trimethylation on plasmid sequences. Also, it appears that Pol II can engage the modified plasmid chromatin, potentially in a manner that is not productive in the synthesis of high levels of new transcript. We outline a scenario in which sustained differences at the chromatin level cooperate to determine the activity of foreign DNA.

Potential roles for short RNAs in lymphocytes

RNA interference (RNAi) is an ancient and evolutionarily conserved process. In some eukaryotes, RNAi silences parasitic genetic elements. In plants, RNAi serves as an immune system against RNA viruses and transgenes and in worms, RNAi silences transposons. In mammals, RNAi has yet unknown functions. However, emerging roles for short RNAs and the factors that interact with them in other eukaryotes include chromatin modification, DNA deletion and DNA methylation, which may provide clues to the roles for short RNA function in mammals. For example, antigen receptor expression in lymphocytes is a highly regulated process and although much is known about chromatin modification and DNA deletion in the immune system, several molecular details of chromatin regulation remain elusive. This review compares emerging roles for short RNA function to processes required for antigen receptor expression in mammalian lymphocytes and predicts that short RNAs direct events required for successful lymphocyte-restricted gene expression.

The evolution of controlled multitasked gene networks: the role of introns and other noncoding RNAs in the development of complex organisms

Eukaryotic phenotypic diversity arises from multitasking of a core proteome of limited size. Multitasking is routine in computers, as well as in other sophisticated information systems, and requires multiple inputs and outputs to control and integrate network activity. Higher eukaryotes have a mosaic gene structure with a dual output, mRNA (protein-coding) sequences and introns, which are released from the pre-mRNA by posttranscriptional processing. Introns have been enormously successful as a class of sequences and comprise up to 95% of the primary transcripts of protein-coding genes in mammals. In addition, many other transcripts (perhaps more than half) do not encode proteins at all, but appear both to be developmentally regulated and to have genetic function. We suggest that these RNAs (eRNAs) have evolved to function as endogenous network control molecules which enable direct gene-gene communication and multitasking of eukaryotic genomes. Analysis of a range of complex genetic phenomena in which RNA is involved or implicated, including co-suppression, transgene silencing, RNA interference, imprinting, methylation, and transvection, suggests that a higher-order regulatory system based on RNA signals operates in the higher eukaryotes and involves chromatin remodeling as well as other RNA-DNA, RNA-RNA, and RNA-protein interactions. The evolution of densely connected gene networks would be expected to result in a relatively stable core proteome due to the multiple reuse of components, implying that cellular differentiation and phenotypic variation in the higher eukaryotes results primarily from variation in the control architecture. Thus, network integration and multitasking using trans-acting RNA molecules produced in parallel with protein-coding sequences may underpin both the evolution of developmentally sophisticated multicellular organisms and the rapid expansion of phenotypic complexity into uncontested environments such as those initiated in the Cambrian radiation and those seen after major extinction events.

Developmentally and transgene regulated nuclear processing of primary transcripts of chalcone synthase A in petunia

The introduction of chalcone synthase A transgenes into petunia plants can result in degradation of chalcone synthase A RNAs and loss of chalcone synthase, a process called cosuppression or post-transcriptional gene silencing. Here we show that the RNA degradation is associated with changes in premRNA processing, i.e. loss of tissue specificity in transcript cleavage patterns, accumulation of unspliced molecules, and use of template-specific secondary poly(A) sites. These changes can also be observed at a lower level in leaves but not flowers of nontransgenic petunias. Based on this, a model is presented of how transgenes may disturb the carefully evolved, developmentally controlled post-transcriptional regulation of chalcone synthase gene expression by influencing the survival rate of the endogenous and their own mRNA.

Modification of glucose and glutamine metabolism in hybridoma cells through metabolic engineering

The present work describes the genetic modification of a hybridoma cell line with the aim to change its metabolic behaviour, particularly reducing the amounts of ammonia and lactate produced by the cells. The cellular excretion of ammonia was eliminated by transfection of a cloned glutamine synthetase gene. The metabolic characterisation of the transformed cell line includes the analysis of the changes introduced in its intracellular metabolic fluxes by means of a stoichiometric model. Furthermore, the reduction of lactate accumulation was attempted through an antisense mRNA approach, aiming to generate a rate limiting step in the glycolytic pathway, thus lowering the glucose consumption rate. The physiological results obtained with the transformed cells are discussed. A maximum reduction of about 47% in the glucose consumption rate was obtained for one of the transformations. However a main drawback was the lack of stability of the transformed cells.

Specific inhibition of hepatitis B virus replication by sense RNA

We describe effects of sense RNA molecules on hepatitis B virus (HBV) replication and antigen synthesis in transiently transfected cells. When certain subgenomic fragments of HBV were expressed as sense RNA together with a replication-competent genome of HBV, they inhibited HBV replication by up to 75% and HBsAg secretion by up to 60%. The corresponding antisense sequences had a 50% inhibitory effect in one case and no effect in another case. The sense RNA species did not inhibit duck hepatitis B virus (DHBV) replication, suggesting specific inhibitory effects. HBV transcript levels were unaltered in the presence of sense RNA species, consistent with an inhibitory effect mediated at the posttranscriptional level. The inhibition of HBV replication by overexpression of sense RNA derived from the viral genome represents an example of sense cosuppression of an animal virus.

Transcriptional and posttranscriptional silencing of rodent alpha1(I) collagen by a homologous transcriptionally self-silenced transgene

Transient transfection of rodent fibroblasts with plasmids carrying a full-size pro-alpha1(I) collagen gene (pWTC1) results in rapid reduction of the endogenous transcripts by >90%, while the transgene mRNA is undetectable. Using deletion constructs, two adjacent 5' noncoding regions of the gene are identified as causing transcriptional silencing of the endogene in normal and v-fos-transformed cells but not in nontumorigenic revertants, which show partial relief from v-fos transformation-induced alpha1(I) gene suppression. The 3' end of the transgene carries an additional element(s), causing posttranscriptional silencing of the endogene in all cells including the revertant. Data indicate that the transgenes are transcriptionally self-silenced. Genome-integrated transgenes that are transcriptionally active also allow expression of the endogene, suggesting gene activation by chromosomal factors missing in pWTC1. Silencing is not regulated by antisense RNA. Silencing of the endogenous pro-alpha1(I) collagen gene is not linked to the level of transgene expression.

Isolation of an RNA-directed RNA polymerase-specific cDNA clone from tomato

A 3600-bp RNA-directed RNA polymerase (RdRP)-specific cDNA comprising an open reading frame (ORF) of 1114 amino acids was isolated from tomato. The putative protein encoded by this ORF does not share homology with any characterized proteins. Antibodies that were raised against synthetic peptides whose sequences have been deduced from the ORF were shown to specifically detect the 127-kD tomato RdRP protein. The immunoresponse to the antibodies correlated with the enzymatic activity profile of the RdRP after chromatography on Q-, poly(A)-, and poly(U)-Sepharose, hydroxyapatite, and Sephadex G-200 columns. DNA gel blot analysis revealed a single copy of the RdRP gene in tomato. RdRP homologs from petunia, Arabidopsis, tobacco, and wheat were identified by using polymerase chain reaction. A sequence comparison indicated that sequences homologous to RdRP are also present in the yeast Schizosaccharomyces pombe and in the nematode Caenorhabditis elegans. The previously described induction of RdRP activity upon viroid infection is shown to be correlated with an increased steady state level of the corresponding mRNA. The possible involvement of this heretofore functionally elusive plant RNA polymerase in homology-dependent gene silencing is discussed.

Isolation of an RNA-directed RNA polymerase-specific cDNA clone from tomato

A 3600-bp RNA-directed RNA polymerase (RdRP)-specific cDNA comprising an open reading frame (ORF) of 1114 amino acids was isolated from tomato. The putative protein encoded by this ORF does not share homology with any characterized proteins. Antibodies that were raised against synthetic peptides whose sequences have been deduced from the ORF were shown to specifically detect the 127-kD tomato RdRP protein. The immunoresponse to the antibodies correlated with the enzymatic activity profile of the RdRP after chromatography on Q-, poly(A)-, and poly(U)-Sepharose, hydroxyapatite, and Sephadex G-200 columns. DNA gel blot analysis revealed a single copy of the RdRP gene in tomato. RdRP homologs from petunia, Arabidopsis, tobacco, and wheat were identified by using polymerase chain reaction. A sequence comparison indicated that sequences homologous to RdRP are also present in the yeast Schizosaccharomyces pombe and in the nematode Caenorhabditis elegans. The previously described induction of RdRP activity upon viroid infection is shown to be correlated with an increased steady state level of the corresponding mRNA. The possible involvement of this heretofore functionally elusive plant RNA polymerase in homology-dependent gene silencing is discussed.

Uncoordinated inhibition of gene expression for muscle proteins by a troponin T antisense oligodeoxynucleotide

The effect of antisense oligodeoxynucleotide to rat troponin T (TnT) mRNA on its expression in differentiated rat L6 myotubes in culture was examined. The target sequence following the initiation codon was between nucleotides 83 and 97 and is found in all mRNAs produced from the f-TNT gene. Our studies showed that chimeric oligomer with one phosphorothioate linkage at the 3'-end was considerably more resistant to nucleases than was a phosphodiester oligomer. The chimeric oligomer produced >50% inhibition of TnT polypeptide synthesis. Synthesis of myosin heavy chain (MHC), troponin I (TnI), and alpha and beta tropomyosins (Tm) was not inhibited by the anti-TnT oligomer. However, synthesis of alpha-actin and troponin C (TnC) was somewhat affected by this treatment. Furthermore, compared with the untreated control myotubes, the steady-state level of TnT mRNA was reduced by approximately 40%-50% in anti-TnT oligomer-treated myotubes. Cellular levels of three other muscle mRNAs, alpha-Tm, s-TnI, and alpha-actin were also reduced by approximately 30%-40%. In contrast, fast TnI, beta-Tm, and TnC mRNA levels were not significantly affected by this treatment. Therefore, inhibition of TnT synthesis in differentiated myotubes uncoupled the coordinated expression of muscle proteins.

The Frequency and Degree of Cosuppression by Sense Chalcone Synthase Transgenes Are Dependent on Transgene Promoter Strength and Are Reduced by Premature Nonsense Codons in the Transgene Coding Sequence

By comparing the effects of strong and weak promoters that drive sense chalcone synthase (Chs) transgenes in large populations of independently transformed plants, we show here that a strong transgene promoter is required for high-frequency cosuppression of Chs genes and for production of the full range of cosuppression phenotypes. In addition, sense Chs transgenes driven by a cauliflower mosaic virus 35S promoter possessing a single copy of the upstream activator region (UAR) were found to produce a significantly lower degree of cosuppression than they did when the transgene promoter possessed two or four copies of the UAR. It has been shown elsewhere that 35S promoter strength increases with increasing UAR copy number. Frameshift mutations producing early nonsense codons in the Chs transgene were found to reduce the frequency and the degree of cosuppression. These results suggest that promoter strength and transcript stability determine the degree of cosuppression, supporting the hypothesis that sense cosuppression is a response to the accumulation of transcripts at high concentrations. This conclusion was shown to apply to single-copy transgenes but not necessarily to inversely repeated transgenes. The results presented here also have significance for efficient engineering of cosuppression phenotypes for use in research and agriculture.

The Frequency and Degree of Cosuppression by Sense Chalcone Synthase Transgenes Are Dependent on Transgene Promoter Strength and Are Reduced by Premature Nonsense Codons in the Transgene Coding Sequence

By comparing the effects of strong and weak promoters that drive sense chalcone synthase (Chs) transgenes in large populations of independently transformed plants, we show here that a strong transgene promoter is required for high-frequency cosuppression of Chs genes and for production of the full range of cosuppression phenotypes. In addition, sense Chs transgenes driven by a cauliflower mosaic virus 35S promoter possessing a single copy of the upstream activator region (UAR) were found to produce a significantly lower degree of cosuppression than they did when the transgene promoter possessed two or four copies of the UAR. It has been shown elsewhere that 35S promoter strength increases with increasing UAR copy number. Frameshift mutations producing early nonsense codons in the Chs transgene were found to reduce the frequency and the degree of cosuppression. These results suggest that promoter strength and transcript stability determine the degree of cosuppression, supporting the hypothesis that sense cosuppression is a response to the accumulation of transcripts at high concentrations. This conclusion was shown to apply to single-copy transgenes but not necessarily to inversely repeated transgenes. The results presented here also have significance for efficient engineering of cosuppression phenotypes for use in research and agriculture.

Suppression of c-fos induction in rat brain impairs retention of a brightness discrimination reaction

Recently, the induction of transcription factor-encoding immediate-early genes such as c-fos was observed in distinct brain regions of rats trained to acquire a footshock-motivated brightness discrimination in a Y-maze. The functional relevance of inducible transcription factors for learning and memory formation is, however, not clear. To address this question in the present study, we have used a synthetic antisense phosphorothioate oligodeoxynucleotide to suppress in vivo the expression of c-fos in rat brain. Intrahippocampal application of the oligodeoxynucleotide 10 hr and 2 hr before starting a brightness discrimination training drastically reduced the induction of c-Fos immunoreactivity normally observed in limbic and cortical areas after the training session. Acquisition of the discrimination reaction was not affected by this treatment. In a relearning test 24 hr after the first training, retention of the discrimination reaction was specifically impaired compared with rats pretreated with control oligodeoxynucleotide or saline. Our findings are consistent with the hypothesis that the inducible transcription factor c-Fos is involved in processes underlying the formation of long-term memory.

HOMOLOGY-DEPENDENT GENE SILENCING IN PLANTS

Homology-dependent gene silencing phenomena in plants have received considerable attention, especially when it was discovered that the presence of homologous sequences not only affected the stability of transgene expression, but that the activity of endogenous genes could be altered after insertion of homologous transgenes into the genome. Homology-mediated inactivation most likely comprises at least two different molecular mechanisms that induce gene silencing at the transcriptional or posttranscriptional level, respectively. In this review we discuss different mechanistic models for plant-specific inactivation mechanisms and their relationship with repeat-specific silencing phenomena in other species.

Stably expressed antisense RNA to cytomegalovirus UL83 inhibits viral replication

The human cytomegalovirus (HCMV) open reading frame UL83 encodes a phosphoprotein of 64 to 68kDa (pp65) which is a major constituent of this virion and dense bodies. To determine the importance of the HCMV gene in the virus cycle, we studied HCMV replication in astrocytoma cells stably transfected with a retroviral vector carrying an antisense UL83 cDNA. Reverse transcription-PCR detected antisense RNA in the cytoplasm. The steady-state level of a 4-kb RNA containing coding sequences for pp65 was significantly reduced after infection of antisense cells. Concomitant with this, levels of expression of pp65 and pp71 (UL82) were severely reduced. Extracellular HCMV production was almost completely blocked, irrespective of the multiplicity of infection or the time after infection studied. The block occurred at an early phase, since immediate-early protein synthesis occurred normally, while several late proteins (e.g., pp150 [ppUL32] and assembly protein [UL80]) were absent or strongly inhibited. Normal replication of herpes simplex virus and of a pp65 deletion mutant of HCMV (RVAd65), lacking target sequences of antisense RNA, demonstrated the specificity of the block for wild-type HCMV in the antisense-stabilized cells and indicated that the block was not due to indirect interference with cellular genes. Our results appear to contradict those of Schmolke et al (S. Schmolke, H.F. Kern, P. Drescher, G. Jahn, and B. Plachter, J. Virol. 69:5959-5968, 1995), which show that UL83 is a nonessential gene for HCMV replication in vitro. This contradiction is discussed in light of the fact that the 4-kb mRNA, which codes for pp65 and was targeted in UL83-antisense cell lines, may be a bicistronic mRNA which also codes for pp71 (UL82). Thus, interference of expression from the genes encoding pp65 and pp71 by blocking of this putative bicistronic message leads to severe impairment of viral replication.

Viral RNA as a potential target for two independent mechanisms of replicase-mediated resistance against cucumber mosaic virus

Transgenic tobacco showing replicase-mediated resistance against cucumber mosaic virus (CMV) can be infected by the strain K-CMV. By use of chimeric constructs between full-length cDNA clones of RNA2 of strains Fny-CMV and K-CMV, the existence of two independent mechanisms of replicase-mediated resistance against viral replication and movement of Fny-CMV was demonstrated in these plants. The data indicate that viral RNA may serve as the target for both mechanisms of resistance. A positive correlation was observed between the amount of K-CMV RNA2 sequence present in the chimeric constructs and the ability to overcome the inhibition of replication, whereas a sequence domain was delimited in K-CMV RNA2 responsible for the ability of this strain to break resistance against virus movement.

Transgenes and gene suppression: telling us something new?

Transgenes provide unique opportunities to assess the relationship between genotype and phenotype in an organism. In most cases, introduction and subsequent expression of a transgene will increase (with a sense RNA) or decrease (with an antisense RNA) the steady-state level of a specific gene product. However, a number of surprising observations have been made in the course of many transgenic studies. We develop a hypothesis that suggests that many examples of endogenous gene suppression by either antisense or sense transcripts are mediated by the same cellular mechanism.

Virus-mediated or transgenic suppression of a G-protein alpha subunit and attenuation of fungal virulence

Strains of the chestnut blight fungus Cryphonectria parasitica harboring RNA viruses of the genus Hypovirus exhibit significantly reduced levels of virulence (called hypovirulence). The accumulation of a heterotrimeric GTP-binding protein (G protein) alpha subunit of the Gi class was found to be reduced in hypovirus-containing C. parasitica strains. Transgenic cosuppression, a phenomenon frequently observed in transgenic plants, reduced the accumulation of this alpha subunit in virus-free fungal strains. Significantly, the resulting transgenic fungal strains were also hypovirulent. These results indicate a crucial role for G-protein-linked signal transduction in fungal pathogenesis and suggest a molecular basis for virus-mediated attenuation of fungal virulence.

Specific inhibition of aphthovirus infection by RNAs transcribed from both the 5' and the 3' noncoding regions

RNA molecules containing the 3' terminal region of foot-and-mouth disease virus (FMDV) RNA in both antisense and sense orientations were able to inhibit viral FMDV translation and infective particle formation in BHK-21 cells following comicroinjection or cotransfection with infectious viral RNA. Antisense, but not sense, transcripts from the 5' noncoding region including the proximal element of the internal ribosome entry site and the two functional initiation AUGs were also inhibitory, both in in vitro translation and in vivo in comicroinjected or cotransfected BHK-21 cells. This effect was not observed with nonrelated RNA transcripts from lambda phage. The inhibitions found were permanent, sequence specific, and dose dependent; an inverse correlation between the length of the transcript and the extent of the antiviral effect was seen. In all cases, the extent of inhibition increased when viral RNAs and transcripts were allowed to reanneal before transfection, concomitant with a decrease in the doses required. The antiviral effect was specific for FMDV, since transcripts failed to inhibit infective particle formation by other picornavirus, such as encephalomyocarditis virus. These results indicate that the ability of RNA transcripts to inhibit viral multiplication depends on their efficient hybridization with target regions on the viral genome. Furthermore, cells transfected with the 5'1as transcript, which is complementary to the 5' noncoding region, showed a significant reduction of plaque-forming ability during the course of a natural infection. RNA 5'1as was able to inhibit FMDV RNA translation in vitro, suggesting that the inhibitions observed are mediated by a blockage of the viral translation initiation. Conversely, hybridization of short sequences of both sense and antisense transcripts from the 3' end induces distortion of predicted highly ordered structural motifs, which could be required for the synthesis of negative-stranded viral RNA, and correlates with inhibition of viral propagation.

Multiple domains in a ribozyme construct confer increased suppressive activity in monkey cells

An expression vector designed to express a long (approximately 1 kb) RNA containing multiple ribozyme domains was cotransfected into mammalian cells with a plasmid encoding, as target, messenger RNA for chloramphenicol acetyltransferase (CAT). In comparative studies the multimeric ribozyme construct proved to be significantly more effective at suppressing CAT expression than either the corresponding antisense RNA, or a transcript carrying a single ribozyme domain. Suppression of gene activity was apparently specific because expression of an independently expressed gene for human growth hormone was unaffected. The profile of CAT RNA extracted from transfected cells was consistent with RNA cleavage.

Epigenetic repeat-induced gene silencing (RIGS) in Arabidopsis

In several plant systems expression of structurally intact genes may be silenced epigenetically when a transgenic construct increases the copy number of DNA sequences. Here we report epigenetic silencing in Arabidopsis lines containing transgenic inserts of defined genetic structure, all at the same genomic locus. These comprise an allelic series that includes a single copy of the primary insert, which carries repeated drug resistance transgenes, and a set of its derivatives, which as a result of recombination within the insert carry different numbers and alleles of resistance genes. Although the drug resistance genes remained intact, both the primary and some recombinant lines nevertheless segregated many progeny that were partly or fully drug-sensitive because of silencing. As in other systems silencing was reversible, and correlated with decreased steady-state mRNA and increased DNA methylation. Each different number and combination of genes, on the same or different (i.e., homologous) chromosomes, conditioned its own idiosyncratic segregation pattern. Strikingly, lines with a single gene segregated only a few slightly drug-sensitive progeny whereas multi-gene lines segregated many highly sensitive progeny, indicating dependence of silencing at this locus on repeated sequences. This argues strongly against explanations based on antisense RNA, but is consistent with explanations based on ectopic DNA pairing. One possibility is that silencing reflects the interaction of paired homologous DNA with flanking heterologous DNA, which induces condensation of chromatin into a non-transcribable state.

An Arabidopsis mutant with a reduced level of cab140 RNA is a result of cosuppression

We analyzed a mutant of Arabidopsis with a severely reduced level of cab140 RNA. This mutant, named lct for low level of cab140 transcript, was obtained during a selection for phytochrome signal transduction mutants. The selection was based on reduced expression of the tumor morphology shoots gene (tms2), an introduced counter-selectable marker under the control of the cab140 promoter. Expression of the introduced cab140::tms2 gene was also greatly reduced in lct, but surprisingly, expression of other phytochrome-regulated genes was not comparably affected. Furthermore, the lct phenotype could not be separated genetically from the T-DNA insert; thus, we suggest that this phenotype was caused by cosuppression of the introduced construct and the endogenous cab140 gene, and that the mutation causing the cosuppression was located on the T-DNA insert. In vitro nuclear transcription experiments demonstrated that the suppression was occurring at the level of transcription. We also found that the suppressed cab140 genes were not significantly more methylated than the nonsuppressed cab140 genes.

Co-suppression of Dictyostelium discoideum myosin II heavy-chain gene expression by a sense orientation transcript

We have previously demonstrated effective inhibition of Dictyostelium discoideum myosin heavy chain (MIIHC) gene expression after introduction of an antisense gene on a transformation vector into cells. We now report that transcription of a sense orientation MIIHC gene has a similar, though not as dramatic, effect on expression from the endogenous gene. Cells transformed with a sense fragment of the MIIHC gene are defective in cytokinesis and become giant and multinucleated when growing on a surface. This morphological phenotype is correlated with decreased expression of the endogenous MIIHC gene. Sense transformants accumulate less MIIHC protein and mRNA than control transformed cells, concomitant with accumulation of a vector-derived sense transcript. Expression of the introduced sequences seems to be required for the sense effect. The possible causes of this surprising phenomenon are discussed.

Variables affecting antisense RNA inhibition of gene expression

We have sought to determine what variables affect the extent of inhibition of gene expression by stable nuclear-derived antisense RNAs. Myosin heavy chain gene II (MHCII) expression in Dictyostelium was used as a model system, because previous results have shown that nearly complete inhibition of expression can be achieved under appropriate conditions. Various fragments of the myosin gene were inserted into several transformation vectors in both sense and antisense orientation, and the effects on expression of protein and RNA from the endogenous MHC II gene were assayed. The results indicate that the critical factor was the particular fragment of the gene used to produce the antisense RNA. Some fragments produced complete inhibition of expression, whereas others gave only slight inhibition. The fragments that produced the greatest inhibition were from the tail region of the gene. In addition, cells were capable of overcoming the inhibition while still expressing the antisense RNA. We hypothesize that the three-dimensional topology of the antisense and sense RNAs determines their accessibility for interstrand hybridization. Transformation with sense fragments of the myosin gene also caused inhibition of the endogenous myosin gene. The inhibition seen with sense expression is not as dramatic as that for antisense, but it had the same phenotypic consequences for the cells. This phenomenon, which has now been documented in other systems, may be mechanistically similar to inhibition by antisense RNA in our system.