Antisense RNA Technology (With Diagram)

This article provides a brief case study on antisense RNA technology.

Introduction:

In cases where a gene has been identified and assigned a particular phenotype, additional approaches are often required to exactly probe the function of gene.

Such hurdles in gene identification and manipulation can be overcome by antisense RNA technology.

Many genomic loci contain transcription units on both strands, therefore two oppositely oriented transcripts can overlap. While one strand codes for a protein, the transcript from the other strand is non-encoding. Such natural antisense transcripts (NATs) can negatively regulate the conjugated sense transcript.

NATs are highly prevalent in a wide range of species—for example, around 15% of human protein-encoding genes have an associated NAT. NATs can be divided into cis-NATs, which are transcribed from opposing DNA strands at the same genomic locus. trans-NATs are transcribed from separate loci. cis-NAT pairs display perfect sequence complementarity, whereas trans- NAT pairs display imperfect complementarity and can target many sense targets to form complex regulation network.

One ingenious and promising approach exploits the specificity of hybridization reactions between two complementary nucleic acid chains. Normally, only one of the two DNA strands in a given portion of double helix is transcribed into RNA and it is always the same strand for a given gene.

If a cloned gene is engineered so that the opposite DNA strand is transcribed instead, it will produce antisense RNA molecules that have a sequence complementary to the normal RNA transcripts. Antisense RNA, when synthesized in large enough amounts, will often hybridize with the “sense” RNA made by the normal genes and thereby inhibit the synthesis of the corresponding protein (Fig. 141).

A related method is to synthesize short antisense nucleic acid molecules by chemical or enzymatic means and then inject (or otherwise deliver) them into cells, again blocking (though only temporarily) production of the corresponding protein.

The first effort that definitely demonstrated the blockage in translation due to the use of antisense RNA in cell-free extracts (CFEs) was carried out by Singer et al. (1963) They showed that synthesis of polyphenylalanine in CFE with polyuridylic acid as template was completely inhibited when polyadenylic acid was added to the translation mixture. This write-up briefly describes the basic and applied aspects of antisense RNA technology, particularly in plant systems.

Natural Antisense RNA Regulation of Gene Expression:

Naturally occurring antisense RNA was involved in gene regulation and this was demonstrated during the study of replication of E. coli ColE1 plasmid. The replication of E. coli plasmid ColE1 involves formation of a RNA primer which is processed by RNase-H while bound to the DNA template. Antisense RNA binds the primer inhibiting the processing of RNA primer and replication of the plasmid, hence the plasmid copy number may be regulated (Tomizawa et al. 1981). Likewise, Staphylococcus aureus plasmid (pT 181) replication and copy number appear to be controlled by antisense RNA (Kumar and Novick, 1985). Translation of E. coli Tn10 transposase mRNA is inhibited by antisense mRNA.

An additional mechanism of translational control, called antisense control, occurs in bacterial cells. This type of regulation is mediated by antisense RNA, which contains sequences complementary to the region of an mRNA containing an initiation codon. Hybridization of the complementary antisense RNA blocks recognition of the initiation codon and binding of the 30S ribosomal subunit to the Shine-Dalgarno sequence (The ribosome binding sites of different bacterial mRNAs display within 10 bases upstream of the AUG a sequence that corresponds to part or all of the hexamer. This polypurine stretch is known as the Shine-Dalgarno sequence), thereby preventing initiation of translation.

Expression of the transposase encoded by the bacterial insertion sequence IS 10 is regulated by the antisense translation-control mechanism. Transposase catalyzes transposition of this mobile DNA element. If too much transposase were expressed, so many mutations would result from IS 10. transposition that the host cell would not survive.

Normally, this does not occur because of antisense control. IS 10 contains two promoters: one called P_IN directs transcription of the strand coding for transposase ; the other called P_OUT lies within the transposase gene and directs transcription of the noncoding strand, producing an antisense RNA complementary to the 5′ end of transposase mRNA (Fig. 14-2). Becuase P_OUT is a much stronger promoter than P_IN, antisense mRNA is produced in greater abundance than transposase mRNA. Hybridization of the antisense RNA to most of the much rarer transposase mRNA prevents translation, thereby assuring that the rate of synthesis of transposase and, in turn, the frequency of transposition, are compatible with survival of the host cells.

Detailed analysis of antisense control in bacteria has revealed that the secondary structures of the complementary RNAs greatly influence the rate at which they hybridize, and thus the efficiency of antisense control. The RNAs transcribed from P_OUT and P_IN IS 10 have been optimized by natural selection to hybridize at extremely high rates at physiological temperature and salt concentrations, so transposase expression is inhibited very effectively.

Besides prokaryotes, in eukaryotes antisense RNA is involved in splicing of hnRNA as it involves small nuclear ribonucleoproteins which have a RNA component complementary to the splice site (Ragers and Wall, 1980). Antisense small nucleolar RNAs encoded by introns have also been reported to play a role in rRNA methylation (Nicoloso et al. 1996).

Naturally occurring antisense RNAs are known to regulate gene expression in plants too. These include antisense RNA transcripts to barley K-amylase mRNA (Rogers, 1988), antisense mRNA complementary to niv gene encoding for enzyme of flavonoid pathway, chalcone synthase (CHS) (Coen and Carpenter, 1988). Rogers identified two antisense transcripts in barley, both were imperfectly complementary to K-amylase gene whereas in case of niv gene, antisense transcripts arose due to an inverted duplication of un-translated leader sequences.

Therefore a tentative mechanism has been proposed for the generation of antisense transcripts. Antisense RNA arises when transcription of a gene proceeds in the strand opposite to template in absence of a strong transcription termination site in the short intergenic region. Antisense transcripts have also been identified in Brassica for the S locus receptor kinase gene which controls self-incompatibility in Brassica (Cock et al. 1997).

The regulation of antisense RNA involves certain basic mechanisms, on the basis of which they have been classified into three classes (Takamaya and Inouye, 1990):

Class I – antisense RNAs are directly complementary to coding region or the SD sequence, resulting in direct inhibition of translation or mRNA destabilization.

Class II – RNAs include those that bind to non-coding regions of the target RNA, resulting in indirect effects produced by, e.g. alternative secondary structure formation that sequesters the ribosome-binding site.

Class III – antisense RNAs regulate transcription of the target mRNA by a mechanism similar to transcriptional attenuation.

Artificial Antisense RNA Regulation of Gene Expression:

Antisense RNA has also been used to artificially modulate gene expression in plants and animals. An example of the use of the binary system to introduce functional genes in plants comes from experiments using antisense RNA to control plant gene expression. Polygalacturonase (PG) is an enzyme that solubilizes the walls of plant cells by digesting pectin. Reducing expression of polygalacturonase could lead to fruit that bruise less easily.

A segment from the 5′ end of the PG gene was inserted in the reverse orientation, together with the promoter from the cauliflower mosaic virus (CaMV), into a binary vector. This vector was transferred to Agrobacterium and subsequently to tomato plants. Ripe fruit from transformed plants expressing the antisense construct had significantly reduced levels of PG enzyme activity.

Disappointingly, these fruits were as soft as wild-type tomatoes, presumably because PG activity is just one of several factors that contribute to fruit softening. Another factor is ethylene, and successful inhibition of tomato ripening has now been achieved by expressing an antisense RNA for an enzyme in the metabolic pathway of ethylene.

The grower can ship such tomatoes without bruising them and then use ethylene to ripen them. To determine to optimum requirements for an efficient antisense RNA regulation varied lengths as well as regions of the sense gene 2-galactosidase were targeted (Pestka et al. 1998)

Such experiments led to certain general conclusions:

1. The antisense RNA must be complementary to 5′ end to sense mRNA and a functional ribosome-binding site on the 5′ end (Coleman et al. 1984).

2. A significant correlation exists between the concentration of antisense RNA and sense mRNA. To produce maximum inhibition the molar ratios between antisense RNA and the sense mRNA vary from 50 : 1 to 600 : 1. It was 150 : 1 in case of inhibition of 2-galactosidase synthesis. Although there exist examples where significant inhibition is produced with molar ratio of 1 : 1 between sense RNA and antisense RNA, there is competition between ribosome and antisense RNA for binding to the ribosome-binding site of sense mRNA. Hence the factors that contribute to the increase in concentration of antisense RNA help in producing stronger inhibition (Melton, 1985).

3. The factors that increase the rate of synthesis of antisence RNA as well as increase the half-life contribute towards the effectiveness of antisense RNA. Thus the promoter for antisense gene should be a strong one or antisense gene should be present in high copy number plasmid for high concentration of the antisense RNA. The length of antisense RNA and its configuration influence the stability and hence the effectiveness of the RNA (Matsuyama and Mizuslima, 1985).

Mode of Action of Antisense RNA:

The study of natural antisense regulation as well as artificial antisense inhibition does not point towards the existence of any single mechanism of gene activation. Several modes of action were suggested by the evidences accumulated. The first stage at which a target gene can be inhibited is transcription but as yet no evidence has been brought to light which supports inhibition of transcription as the rate of transcription of both sense and antisense genes is unaffected by the expression of antisense gene (Sheehy et al. 1988).

The second stage where antisense transcript can interfere is the RNA-processing stage, it was shown by Tieman et al. (1992) that when the two introns in pectin methyl esterase gene were placed in antisense orientation, they were not spliced out. The formation of a duplex between antisense and sense transcripts is a factor contributing to inhibition, it has been hypothesized that a RNA-RNA duplex is unstable and susceptible to nucleases but no direct evidence for duplex formation has been found perhaps due to degradation of the duplex (Bourque and Flok, 1992).

Therefore the formation of such a duplex would hinder the processing or transport of the sense mRNA across the nuclear membrane or would lead to degradation of antisense-sense duplex by nucleases making the sense transcript unavailable for translation (Bonque and Folk, 1992).

Inhibition may also occur at the translational stage. The antisense transcript would compete with the ribosomes to bind 5′ end of the sense RNA, hence inhibiting the translation. It has been observed by Pestka et al. (1984), that if the antisense RNA is not complementary to the 5′ end of mRNA, then the extent of inhibition is significantly reduced through a lower but significant level of inhibition remains; this shows that antisense RNA-mRNA duplex formation occurs but the ribosome which binds the 5′ end is capable of stripping the antisense RNA. Thus once the ribosome binds to mRNA, mRNA-antisense RNA duplex formation is greatly reduced and so is the inhibition (Pestka et al. 1984).

Factors Influencing Antisense RNA Regulation of Gene Expression:

While making antisense constructs for specifically regulating the expression of a particular gene, certain factors have to be taken into account. The presence of antisense transcript much in excess of target mRNA is a prerequisite for effective inhibition, therefore the choice of promoter is important. Cauliflower mosaic virus (CaMV) 35S RNA promoter is a constitutive one and is the most widely used one.

Other commonly used promoters include nopaline synthase promoter, chlorophyll a/b-binding protein gene promoter, and CHS gene promoter. Identical promoters and terminators may be employed for both sense and antisense constructs but usually an excess of antisense transcripts is required, thus antisense gene is cloned along with constitutive promoters (Robert et al. 1989). Some workers have utilized tissue-specific promoters for antisense regulation of a particular function in a specific tissue.

Since the formation of duplex of sense and antisense transcripts is the critical step for inhibition, the degree of homology and homology in certain specific regions is important. Though some heterogeneity is tolerated, e.g. starch synthase antisense gene from cassava could suppress the starch synthase in potato (Salehuzzaman et al. 1993), antisense apple ACC oxidase has been reported to inhibit ethylene production in tomato (Bolitho et al. 1997) yet low inhibition is seen in cases where degree of homology is low.

Transgenic tobacco plants carrying antisense gene for tomato ACC oxidase showed variable inhibition in different parts and physiological states; this may be due to varying degree of homology or fluctuations in different tissues (Einset, 1996).

Antisense Oligonucleotides:

Transient inhibition of a specific gene expression can be achieved by using antisense oligonucleotides, which are short (14-18 bases) DNA molecules complementary to the 5′ leader sequence or 3′ end of mRNA. In cases where oligonucleotide is complementary to the coding sequence, the 5′-fragment can be translated to generate a truncated polypeptide.

The binding of antisense deoxy oligonucleotide and target mRNA leads to formation of a DNA-RNA duplex which is unstable and is recognized by RNase- H which selectively degrades the RNA strand in a DNA-RNA duplex, thus inhibiting translation. In case of blockage of RNase-H activity, oligonucleotide directed to cap sites can only inhibit translation as it inhibits binding of 40S subunit of ribosome (Walder and Walder, 1988).

Another class of oligonucleotides exists which are called the code blockers or triplex forming oligonucleotides (Riordam and Martin, 1991), these bind in the major groove of DNA target sequence, thus inhibiting transcription. They can either inhibit the binding of transcription factors by binding upstream to the coding region or can inhibit movement of RNA polymerase. Triplex formation requires the target site to have a homopurine or homopyrimidine sequences and third strand binds by hoogsten base pairing, triplex strategy though is quite effective, especially in case of actively transcribing genes, yet it has its disadvantages like the requirement of a homopurine/homopyrimidine target sequence and further the problem of triplex instability under physiological conditions (Maher et al. 1989).

The effectiveness of oligonucleotides depends on several factors like position of target site against which they are directed, length of the oligonucleotide and further the presence or absence of secondary structure at the binding site. Stability of oligonucleotide is very essential for effective inhibition, and, thus, in case of unmodified oligonucleotides, poor uptake and nuclease degradation were the limiting factors.

Therefore, various types of modifications have been developed which render the oligonucleotide resistance to nucleases. The modifications are either of the phosphate backbone (e.g. methylphospho- nates, phosphorothioates, and phosphoroselonates) (Stein and Cohen, 1988). or the oligomer ends may be modified (K-oligo decarboxynucleotide).

The oligomers can be coupled with intercalating agents and reactive metal agents like EDTA-Fe, phenanthroline-Cu, etc. Synthetic polymers like PAMAM den-drimers have also been investigated to function as delivery system for targeted gene modulation. Novel methyl phosphonate oligonucleotides have been designed which have an internal non- nucleotide based linker moiety due to which the complementary unpaired base of RNA becomes sensitive to cleavage, hence site-specific cleavage of target RNA is possible (Reynolds et al. 1996).

Applications of Antisense RNA Technology:

Antisense strategies have been applied to plant systems as well as animal systems not only for production of novel mutants but also for studying the steps involved in particular metabolic pathways, identifying gene function, plant development, crop improvement and other novel uses.

Antisense RNA provides an opening in the study of regulation of viral genes, as an antisense inhibition can be taken to be a leaky mutation which would be useful in studying genes, mutations in which are lethal and their partial inhibition also leads to a significant change in phenotype.

Such partial inhibition was used to create a tobacco mutant deficient in NADH-hydroxypyruvate reductase to study the role of photo-respiration in stress protection (Oliver et al. 1993). Besides unravelling the vital gene functions, antisense RNA inhibition has been used to observe various steps in metabolic pathways. Majeau et al. (1994) modified the activity of carbonic anhydrase which had no significant impact on CO₂ assimilation but it brought forward the effect of decline in carbonic anhydrase activity on stomatal conductance and susceptibility to water stress.

Antisense mutants of tobacco with drastic decrease in Rubisco content resulted in low photosynthetic rate; however, the leaf development was normal and independent of Rubisco content though leaf development was delayed (Jiang and Rodermel 1995). The biochemical target of various herbicides is acetolactate synthase, and this was confirmed by raising transgenic potato plants expressing antisense acetolactate synthase which were inviable without amino acid supplementation, thus an in vivo model for herbicide action was put forward (Hofgen et al. 1995).

Similarly, the effect of ethylene on shoot morphogenesis was studied via the production of transgenic mustard plants expressing antisense 1-aminocyclopropane -1-carboxylic acid (ACC) oxidase gene, and such plants showed marked increase in regeneration potential and corresponding decrease in ethylene production (Pua and Lee, 1995). Cotton fibre protein genes have also been characterized using antisense RNA inhibition of a particular gene (Hohn, 1996). Antisense inhibition has been utilized to work out the role of lipoxygenase (LOX) in lentil protoplast by the introduction of antisense LOX gene (Maccarone et al. 1995).

The antisense RNA technology has formed the basis for elucidating the flavonoid biosynthetic pathway, and, as a matter of fact, CHS gene was the first endogenous gene targeted by antisense RNA in plants. Antisense CHS petunia plants produced flowers with pale corolla pigmentation but the steady state levels of mRNA of other flavonoid-specific genes were not affected (Van der Krol et al. 1990).

S-adenosyl-methionine (SAM) is a common precursor for both ethylene biosynthesis as well as biosynthesis of polyamines (spermine and spermidine), and the antisense inhibition of SAM decarboxylase gene expression in potato transgenic plants provided a molecular approach to study the effect of manipulation of polyamine levels on growth and development (Kumar et al. 1996).

The potato plants expressing the antisense SAM decarboxylase gene constitutively showed aberrant phenotypes due to the depletion of cellular polyamines and elevated levels of ethylene (Kumar et al. 1996). Further, potato transgenic were also produced with SAM decarboxylase antisense gene driven by an tetracycline-inducible promoter, and these transgenic provide an opportunity for further investigations into the inter-relationship between polyamine and ethylene metabolic pathways.

Other than elucidating the steps of metabolic pathways, antisense RNA inhibition has found its use in identification of gene function as in case of a ripening gene (pTOM5), which was found to be a part of carotenoid pathway (Bird et al. 1991).

Another ripening-related gene of tomato (pTOM13) was found to be involved in ethylene synthesis and thus may be a part of ACC-oxidase system involved in conversion of ACC to ethylene (Thang et al. 1992) Arabidopsis expressing antisense RNA against ankyrin repeat (AKR) containing gene indicated the involvement of AKR gene in regulation of chloroplast differentiation (Bouzayer et al. 1992). Antisense repression of nuclear encoded NADH-binding subunit of mitochondrial respiratory chain complex-I in potato plants led to normal vegetative growth but reduced male fertility which might be due to insufficient mitochondrial respiratory chain (Heiser et al. 1997).

Transgenic Flaveria bidentis plants with antisense Rubisco gene were used to study the relationship between CO₂ assimilation and Rubisco content in C₄ plants and it was observed that the inhibition of Rubisco led to increase in CO₂ concentration and its leakage in bundlesheath (Caemmerer et al. 1997).

The importance of peptide transport gene AtPTR2-B from Arabidopsis was evaluated by producing transgenic with antisense AtPTR2-B gene, the transgenics had altered phenotype, delayed flowering and no seed set, suggesting a major role of the gene in growth and development (Song et al. 1997).

Other examples of the antisense RNA technology being utilized for elucidating gene functions include the transgenic tobacco-expressing antisense ascorbate peroxidase (APX) gene leading to increased susceptibility of transgenics to ozone injury, suggesting the major role of APX in oxidative stress tolerance (Orvar and Ellis, 1997). Antisense inhibition of biotin carboxylase gene in tobacco led to severe retardation of growth, reinforcing the importance of biotin for plant growth (Shintami et al. 1997).

The antisense RNA technology has also been used for crop improvement, besides being used to gain knowledge in the basics of plant development. The technology has been used in modifying seed oil composition of Brassica seed oil (Knutzon et al. 1992). The desaturase enzyme gene was inhibited, leading to production of seeds with high-stearate oil content without there being any decrease in the seed lipid content.

Antisense RNA inhibition of chitinase gene expression has resulted in increased fungal disease susceptibility in Arabidopsis plants, elucidating some role of chitinases in plant protection (Samac and Shah, 1994). This technique has also been used to confer resistance to viral plant infections. Transgenic potato plants expressing antisense RNA to potato leaf roll luteovirus (PLRV) coat protein were resistant to the infection (Kawchuk et al. 1991). Stanley et al. (Stanley et al. 1990) produced transgenic cassava resistant to cassava latent virus (CLV) by introducing a tandem repeat of subgenomic DNA B of CLV.

Transgenic Nicotiana benthaniana plants expressing antisense C1 gene (which encodes for Rep protein) of tomato yellow leaf curl virus (TYLCV) were found to be resistant to TYLCV infection (Bendalmane and Gronenborn, 1997), thus tomato transgenics may be produced which would be resistant to TYLCV, a major disease in tomato.

The antisense RNA technology has recently been reported to generate Hsp70 mutant in A. thaliana and this brought forward the protective role of HSp70 in thermo tolerance and a regulatory effect on heat shock transcription factors leading to autoregulation of the heat shock response (Lee and Schoftl 1996).

The antisense gene strategy has been applied to inhibit the expression of an allergen gene during seed maturation in rice, and inhibition persisted in the progeny of the transformed plants (Tada et al. 1996).

The antisense RNA technology has certain basic lacunae which need to be overcome before its full potential is exploited. Firstly, there seems to be no single universal mechanism of antisense action, each system has to be studied independently for its mechanistic aspects of inhibition.

The degree of inhibition is quite variable even when single copy of antisense gene is present, this aspect needs to be studied further in correlation with copy number and positional effect on the antisense gene expression. Besides these, the long-term effects of expression of antisense gene or dosage of antisense oligonucleotides, their degradation and effects of nonspecific binding need to be investigated.

In conclusion, the antisense RNA technology holds promise both for the plant systems as well as the animal systems but before its extensive use, the basics of the technology have to be elucidated and the technology accordingly modulated so that it may be exploited to its full potential.