A TRANSCRIPTIONALLY ACTIVE STATE IS REQUIRED FOR POSTTRANSCRIPTIONAL SILENCING (COSUPPRESSION) OF NITRATE REDUCTASE HOST GENES AND TRANSGENES

Using tobacco nitrate reductase cosuppression as a model system of pos t-transcriptional gene silencing, we analyzed the influence of DNA and RNA dosages both together and independently, For this purpose, zero, one, two, or four active or transcriptionally silenced copies of a cau liflower mosaic virus 35S-Nia2 transgene were combined by transformati on and subsequent crosses with zero, one, two, three, or four active, disrupted, or transcriptionally repressed copies of the wild-type host Nia genes. The analysis of the corresponding transgenic lines reveale d that (1) the percentage of isogenic plants that are affected by cosu ppression depends directly upon the relative dosage of both host gene and transgene; (2) transcriptional silencing of the 35S-Nia transgene impedes cosuppression; and (3) the absence of host gene transcription reduces the frequency of cosuppression or delays its triggering. Taken together, these results indicate that transgene DNA per se is not suf ficient to trigger post-transcriptional cosuppression of nitrate reduc tase host genes and transgenes. The requirement for a transcriptionall y active state is discussed with respect to both the RNA dosage and th e DNA-DNA pairing hypotheses.