Endosymbiotic evolution: transfer of plastid DNA to the nucleus and its stability following integration.

Abstract

In eukaryotes, cytoplasmic organellar genomes have contributed to a wide variety of nuclear genes. Therefore DNA transfer from plastids and mitochondria to the nucleus has been an important driving force in eukaryotic evolution. Recent advances have enabled this transfer process to be demonstrated experimentally for the plastid in Nicotiana tabacum. Investigation of the frequencies, mechanisms and consequences of plastid-to-nucleus DNA transfer will shed further light on the complexities of endosymbiotic evolution. To determine the frequencies of plastid-to-nucleus DNA transfer in the male and female germlines of N. tabacum, reciprocal crosses were performed using a transplastomic line (tp7) as either the male or the female parent. In its plastid genome tp7 contained a kanamycin resistance gene (neo) under the control of a nuclear promoter, so any resistant progeny were expected to be the result of neo transfer to the nucleus. In the cross where tp7 was the male parent, 13 resistant plants were obtained from a screen of 146,000 progeny indicating a transfer frequency in the male germline of approximately 1 event for every 11,000 pollen grains. Screening of 273,000 progeny from the cross where tp7 was the female parent revealed only 1 resistant plant, demonstrating a much lower transfer frequency in the female germline. The programmed degradation of plastids during pollen development may release DNA fragments that integrate into the nucleus, thus explaining the high transfer frequency in the male germline. To investigate plastid-to-nucleus DNA transfer in somatic cells, a transplastomic line (tpGUS) was generated containing gus under the control of a nuclear promoter in the plastid genome. Staining of various somatic tissues revealed discrete foci of gus expression. The frequency of gus transfer to the nucleus was estimated to be 1 event for every 200,000 mature leaf cells. Self-fertilised progeny of tpGUS were used to detect stable transfer of gus to the nucleus, revealing a similar germline transfer frequency to that obtained with tp7. To determine whether an RNA intermediate is involved in sequence transfer from the plastid to the nucleus, a transplastomic line (tpneoACG) was generated containing, in its plastid genome, a nuclear promoter-driven neo gene with a start codon that required plastid RNA editing. Screening revealed a number of kanamycin-resistant progeny plants. Surprisingly, neo was unedited in these plants, indicating that neo was active in the absence of RNA editing. Analysis of tpneoACG revealed that only a low proportion of neo transcripts were edited, thus precluding unequivocal conclusions regarding the importance of RNA in plastid-to-nucleus transfer. The stability of newly integrated plastid DNA sequences in the nucleus was analysed using kanamycin-resistant progeny plants of tp7. Around half of the plants showed non-Mendelian segregation, with fewer resistant progeny than expected. Loss of the neo gene was shown to be responsible for this instability. Therefore, plastid DNA integration into the nuclear genome and its subsequent deletion both occur, with the latter process varying from very frequent at some nuclear loci to undetectable at others. These may be important evolutionary processes in the generation of novel nuclear genes.