A link between DNA methylation and epigenetic silencing in transgenic Volvox carteri

Abstract

Epigenetic silencing of foreign genes introduced into plants poses an unsolved problem for transgenic technology. Here we have used the simple multicellular green alga VOLVOX: carteri as a model to analyse the relation of DNA methylation to transgenic silencing. VOLVOX: DNA contains on average 1.1% 5-methylcytosine and 0.3% N6-methyladenine, as revealed by electrospray mass spectrometry and phosphoimaging of chromatographically separated (32)P-labelled nucleotides. In two nuclear transformants of V.carteri, produced in 1993 by biolistic bombardment with a foreign arylsulphatase gene (C-ars), the transgene is still expressed in one (Hill 181), but not in the other (Hill 183), after an estimated 500-1000 generations. Each transformant clone contains multiple intact copies of C-ars, most of them integrated into the genome as tandem repeats. When the bisulphite genomic sequencing protocol was applied to examine two select regions of transgenic C-ars, we found that the inactivated copies (Hill 183) exhibited a high-level methylation (40%) of CpG dinucleotides, whereas the active copies (Hill 181) displayed low-level (7%) CpG methylation. These are average values from 40 PCR clones sequenced from each DNA strand in the two portions of C-ars. The observed correlation of CpG methylation and transgene inactivation in a green alga will be discussed in the light of transcriptional silencing.

Figure 1

HPLC elution profile of dNs and mass spectra of fractions B and F. (A) dNs from 50 µg of hydrolysed and dephosphorylated V.carteri HK10 DNA were separated by HPLC and monitored by absorbance at 254 nm. Fractions A–F were collected and analysed by mass spectrometry. (B) Mass spectra of HPLC fractions B and F from (A). Signals assigned by their m/z values identify protonated molecules (MH⁺) of two methylated derivatives, ^5mC and ^6mA. The complete assignments of fractions A–F are compiled in Table 1.

Figure 2

Repeat structure of integrated C-ars copies and Southern analysis of transformants. (A) Proposed tandem arrangement of transgenic DNA (containing C-ars) in transformant clones Hill 181 and Hill 183, and extent of two restriction fragments (I and II) used for probing Southern blots. P-Vβ2, Volvox β2-tubulin promoter; C-ars, arylsulphatase gene from C.reinhardtii. (B) Southern blots of plasmids (pIK2, pJD27) and genomic DNAs (identified by clone numbers), probed as indicated. DNAs were restriction-digested as follows: lanes 2 and 9, BamHI; lanes 4, 5, 7 and 11, SalI; lanes 1, 3, 6, 8 and 10, BamHI–SalI. Genomic DNAs were applied at 2 µg per lane, plasmid DNAs were applied at 1950 (pIK2) or 950 pg (pJD27) per lane (corresponding to 10 copies per genome; lanes 2–4) and at 195 (pIK2) or 95 pg (pJD27) per lane (corresponding to one copy per genome; lanes 9–11). The resulting differences in signal intensities served as a reference for densitometric estimates of transgene copy numbers. Bands are identified by their sizes (indicated to the left or right of the gels).

Figure 3

Analysis of tandemly integrated transgenes by plasmid rescue. As illustrated, the transgenes (abbreviated as in Fig. 2; nitA, nitrate reductase gene) were recovered by SalI digestion of transformant DNA, ligation and plasmid replication (ori, origin of replication) and selection in an E.coli host on ampicillin (bla, β-lactamase gene). The plasmid structures were confirmed by restriction mapping and sequence analysis. The pIK2/pVcNR1 hybrid structure is suggestive of homologous recombination that joins transforming plasmids within the identical pUC-derived vector portions (black/red transitions) before they are integrated (as concatemers) into the genome.

Figure 4

(Opposite and above) Physical map of the P-Vβ2/C-ars transgene (A) and CpG methylation patterns in regions I and II compared between Hill 181 (B) and Hill 183 (C). (A) Sequence-derived map of the P-Vβ2/C-ars construct introduced into Hill 181 and Hill 183. The numbering of transcription initiation (+1), translation start (ATG) and stop (TAG) refers to the submitted Volvox β2tub (GenBank accession no. L24547) and Chlamydomonas ars (GenBank accession no. AF333184) genomic sequences. Exons (blue boxes), introns (black lines), promoter (grey arrow) and the P-Vβ2/C-ars junction (red) are delineated. Bars marked ‘Region I’ and ‘Region II’ delimit the portions analysed by bisulphite genomic sequencing (B and C). (B) Hill 181- and (C) Hill 183-derived transgene nucleotide sequences of region I (B1 and C1) and region II (B2 and C2), respectively. Numbering and colour codes as in (A). The methylatable CpG dinucleotide positions are marked by columns of 10 circles in each strand representing 10 clones isolated and sequenced for each set of primers. Each horizontal set of circles represents the methylation pattern of a single cloned PCR product. Filled circles indicate methylated cytosines, open circles unmethylated cytosines.

Figure 4

(Opposite and above) Physical map of the P-Vβ2/C-ars transgene (A) and CpG methylation patterns in regions I and II compared between Hill 181 (B) and Hill 183 (C). (A) Sequence-derived map of the P-Vβ2/C-ars construct introduced into Hill 181 and Hill 183. The numbering of transcription initiation (+1), translation start (ATG) and stop (TAG) refers to the submitted Volvox β2tub (GenBank accession no. L24547) and Chlamydomonas ars (GenBank accession no. AF333184) genomic sequences. Exons (blue boxes), introns (black lines), promoter (grey arrow) and the P-Vβ2/C-ars junction (red) are delineated. Bars marked ‘Region I’ and ‘Region II’ delimit the portions analysed by bisulphite genomic sequencing (B and C). (B) Hill 181- and (C) Hill 183-derived transgene nucleotide sequences of region I (B1 and C1) and region II (B2 and C2), respectively. Numbering and colour codes as in (A). The methylatable CpG dinucleotide positions are marked by columns of 10 circles in each strand representing 10 clones isolated and sequenced for each set of primers. Each horizontal set of circles represents the methylation pattern of a single cloned PCR product. Filled circles indicate methylated cytosines, open circles unmethylated cytosines.