- Email: [email protected]
DNA transfer from chloroplast to nucleus is much rarer in Chlamydomonas than in tobacco
Gene 316 (2003) 33 – 38 www.elsevier.com/locate/gene
DNA transfer from chloroplast to nucleus is much rarer in Chlamydomonas than in tobacco Diane L. Lister a,1, Joseph M. Bateman b,2, Saul Purton b, Christopher J. Howe a,* a
Centre for Molecular Recognition, Department of Biochemistry, University of Cambridge, Downing Site, Tennis Court Road, Cambridge, CB2 1QW, UK b Department of Biology, University College London, London, UK Received 3 June 2003; accepted 13 June 2003 Received by W. Martin
Abstract By transforming chloroplasts with an antibiotic-resistance gene under the control of a nuclear-specific promoter, we employed a selection scheme to detect the transfer of DNA from the chloroplast to the nucleus in the green alga Chlamydomonas reinhardtii. Among several billion homoplasmic cells tested, we were unable to detect any stable nuclear integration of chloroplast DNA under normal growth conditions or under stress conditions. This contrasts with results reported for the transfer of DNA from chloroplast to nucleus in higher plants and from mitochondrion to nucleus in Saccharomyces cerevisiae. Furthermore, we were unable to detect chloroplast DNA-derived sequences among nuclear genome data for C. reinhardtii, which also contrasts with the situation in higher plants. Taken together, these findings suggest that there is presently little, if any, movement of DNA from chloroplast to nucleus in C. reinhardtii, which may reflect the ultrastructure of the C. reinhardtii cell. D 2003 Elsevier B.V. All rights reserved. Keywords: Endosymbiosis; Genome evolution; Organelle; Gene transfer; Transposition; Translocation
1. Introduction It is generally accepted that chloroplasts originated from the primary endosymbiotic acquisition of oxygenic photosynthetic bacteria by nonphotosynthetic cells. In some lineages, secondary and even tertiary endosymbioses have subsequently occurred (Yoon et al., 2002). Although there is still some controversy over whether all chloroplasts are ultimately derived from a single endosymbiosis (a monophyletic origin) or from several (polyphyletic origins), it is
Abbreviations: bp, base pair; HSM, high salt medium; kb, kilobase pair; mt, mating type; PCR, polymerase chain reaction; TAP, Tris-acetatephosphate; UV, ultraviolet. * Corresponding author. Tel.: +44-1223-333688; fax: +44-1223333345. E-mail address: [email protected] (C.J. Howe). 1 Current address: Department of Pathology, University of Cambridge, UK. 2 Current address: Cancer Research UK Developmental Patterning Laboratory, London, UK. 0378-1119/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0378-1119(03)00754-6
clear that the evolution of the organelle involved the transfer of a large portion of the genetic information of the original endosymbiont(s) to the nucleus of the host organism (Martin et al., 1998, 2002; Howe et al., 2003). Although chloroplasts contain more than a thousand different protein species, their genomes now typically encode of the order of 100 – 200 proteins (Glo¨ckner et al., 2000). The remainder of the chloroplast proteins are encoded in the nucleus and imported posttranslationally through a machinery that is derived at least in part from the bacterial ancestor of the chloroplast (reviewed by Jarvis and Soll, 2001). A number of suggestions have been made as to why transfer of genes to the nucleus might provide a selective advantage (Allen and Raven, 1996; Howe et al., 2000). One suggestion is that transfer to the nucleus isolates genes from damaging species, such as oxygen free radicals, that may be generated during photosynthesis. A second suggestion is that placing a gene in a sexual population (as opposed to the asexual population represented by the uniparentally inherited chloroplast) is advantageous, and a third is that movement of a gene to the nucleus avoids deleterious effects due to the
34
D.L. Lister et al. / Gene 316 (2003) 33–38
biased nucleotide composition associated with many chloroplast genomes. Similarly, several hypotheses have been proposed to explain why a subset of genes has apparently resisted the selective advantages of chloroplast-to-nucleus transfer. These hypotheses include the need to couple the expression of certain genes to a redox-mediated regulatory mechanism within the chloroplast and the difficulty of importing some proteins into the organelle (reviewed by Allen, 2003). Currently, nothing is known of the molecular mechanism of transfer of genes from chloroplast to nucleus although evidence of transfer is seen in the nuclear genomes of higher plants in the form of large tracts (occasionally many kilobases) of chloroplast-derived DNA as well as numerous smaller DNA fragments (Ayliffe and Timmis, 1992; Blanchard and Schmidt, 1995; Arabidopsis Genome Initiative, 2000; Yuan et al., 2002). Studies have been carried out on the analogous transfer of DNA from mitochondrion to nucleus in yeast (Thorsness and Fox, 1990). These authors inserted into the mitochondrial genome a marker (URA3) that would be functional only in the nucleus and then selected for acquisition of URA3 function. They demonstrated an unexpectedly high frequency of transfer from mitochondrion to nucleus (approximately 2 10 5 per cell per generation). In contrast, no transfer of DNA from nucleus to mitochondrion was detectable. The frequency of transfer from mitochondrion to nucleus was increased by treatments expected to disrupt mitochondrial membranes such as growth at elevated temperature or in medium containing 15% glycerol or by freeze thawing. In addition, inactivation of the YME1 gene, encoding a homologue of FtsH postulated as being involved in mitochondrial septum formation, also led to an increased frequency of transfer from mitochondrion to nucleus (Thorsness et al., 1993). This idea that DNA escape from the organelle is a consequence of a breach in the structural integrity of the envelope membrane is supported by studies of DNA movement across the membrane of isolated chloroplasts from pea (Cerutti and Jagendorf, 1995). Recently, Huang et al. (2003) demonstrated directly the transfer of DNA from chloroplast to nucleus in transgenic plants of Nicotiana tabacum. That study used a neomycin phosphotransferase (kanamycin resistance) coding region modified for nuclear expression by attachment of a promoter active in the nucleus and a nuclear intron. The gene was inserted into the chloroplast genome, where it failed to confer kanamycin resistance. Seedling progeny from a cross using a male parent containing this chloroplast genome were selected on kanamycin. The results indicated that the chloroplast kanamycin-resistance gene had been transferred to the nucleus in 16 individuals out of 250,000 seedlings screened. The demonstration of a significant frequency of transfer is important in allowing the process to be studied in the laboratory and also for its implications for plant genetic manipulation. It has been suggested that insertion of transgenes into the chloroplast would reduce the likelihood of
their escaping into the environment because, in most crop species, chloroplasts are not transmitted through the pollen (Maliga, 1993). If gene flow occurs at a significant rate from the chloroplast to the nucleus, this would greatly reduce the strength of this biological ‘containment’. Using the green alga Chlamydomonas reinhardtii, we have carried out similar experiments to those described by Huang et al. (2003) for N. tabacum. We inserted into the chloroplast genome a number of separate constructs based on the ble gene for resistance to the antibiotic zeomycin designed so that the gene would confer resistance only if located in the nucleus (Stevens et al., 1996; Lumbreras et al., 1998). After growth under a range of conditions, cells were selected for resistance to zeomycin, which would be expected if the gene had transferred from the chloroplast to the nucleus. We find that, in contrast to the observations with N. tabacum, DNA transfer from the chloroplast to the nucleus does not occur at a detectable rate in C. reinhardtii and speculate that the presence of only a single chloroplast in the cell prohibits DNA escape to the nucleus.
2. Materials and methods 2.1. Strains and media The C. reinhardtii strains used were the photosystem Ideficient strain C575D (mt+) that has a mutation in the chloroplast gene psaA-3 (Hallahan et al., 1995), the zeomycin-resistant strain M86 that contains a single copy of the ble gene in the nuclear genome and a wild type strain (mt ). Cells were grown in Tris-acetate-phosphate (TAP) medium or high salt minimal (HSM) medium (Harris, 1989). Cell numbers were estimated using a haemocytometer. Media were supplemented as appropriate with 2% w/v agar and zeomycin (Zeocink, Cayla, France) at a range of concentrations. Growth was routinely at 22 jC at an illumination of f 50 AE m 2 s 1 white light. Zeomycin resistance was scored by spotting 5 Al of cells grown to stationary phase (f 1 107 cells ml 1) in TAP medium onto TAP-agar plates with varying zeomycin concentrations. 2.2. Strain construction All recombinant DNA work was performed using Escherichia coli strains DH5a or TG1 and methods as described by Ausubel et al. (2001). The basic plasmid used for construction of C. reinhardtii strains with zeomycinresistance genes in the chloroplast was pBa3.AX.ble, which includes an EcoRI cassette comprising the zeomycin-resistance coding region from Streptoalloteichus hindustanus flanked by 794 bp of the 5V untranslated region and 231 bp of the 3Vuntranslated region of the C. reinhardtii gene RBCS2. This cassette has been shown to function as a dominant nuclear marker in C. reinhardtii (Stevens et al.,
D.L. Lister et al. / Gene 316 (2003) 33–38
1996). The EcoRI cassette was inserted into the EcoRI site of the plasmid pBa3.AX (Hallahan et al., 1995), which contains exon 3 of the psaA gene of C. reinhardtii in a pBluescript vector, to generate pBA3.AX.ble (Fig. 1). Integration of the plasmid into the chloroplast genome can be selected by restoration of photosynthetic growth to the strain C575D which lacks a functional psaA-3 sequence (Hallahan et al., 1995). The plasmid pBA3.AX.ble was modified to form pBa3.AX.117 by replacement of the EcoRI cassette bearing the conventional zeomycin-resistance gene with one containing two copies of the first intron of the C. reinhardtii gene RBCS2. The presence of introns in foreign genes has been shown to enhance expression in C. reinhardtii (Lumbreras et al., 1998). The plasmids pBa3.AX.ble.psbA and pBa3.AX.117.psbA were generated from pBa3.AX.ble and pBa3.AX.117, respectively, by insertion into the MluI site of a 1.8 kb polymerase chain reaction (PCR) product extending from a position 132 upstream from the initiation codon through to the second exon of the C. reinhardtii gene psbA. The four constructs were introduced into C. reinhardtii by particle bombardment of lawns of C575D mutant cells on HSM-agar, essentially as described by Goldschmidt-Clermont (1991). Colonies of transformant cells were visible after 3-week incubation in the light at 22 jC and subjected to another three rounds of propagation on HSM to ensure homoplasmicity. The insertion of the ble constructs into the chloroplast genome downstream of psaA-3 was verified by Southern blotting and PCR analysis.
35
of TAP-agar containing zeomycin at 20 Ag ml 1 in covered bioassay dishes (243 243 18 mm). Colonies obtained were streaked onto TAP-agar containing zeomycin at 20 Ag ml 1 and, if growth occurred, were spot tested on a range of zeomycin concentrations. Mating experiments were carried out as described by Harris (1989). Colchicine treatment was carried out by addition to a stationary phase culture containing 109 –1010 cells in TAP medium to a final concentration of 5 mM. Cells were incubated for 2.5 h, washed, incubated in fresh TAP medium overnight at 22 jC and then plated onto TAP-agar containing zeomycin. Salt stressing was carried out by adding NaCl to a final concentration of 0.2 M to stationary phase cells in TAP medium, followed by overnight incubation at 22 jC, washing and selection on TAP-agar containing zeomycin. Ultraviolet (UV) stressing was carried out by incubation of 5 ml aliquots of cells, containing 2.5 107 cells, adjacent to a Stratagene crosslinker providing a total irradiation of 2 104 – 5 104 AJ.cm 2. Cells were then added to 250 ml aliquots of TAP, incubated for 3 days at 22 jC and then plated in the presence of zeomycin. Counting of viable cells showed that the UV conditions used killed up to 70% of cells. Larger quantities (1010) of cells were also subjected to UV treatment by stirring under UV irradiation provided by a bactericidal lamp in a microbiological flow hood.
3. Results 3.1. Selection for zeomycin resistance
2.3. Selection for zeomycin resistance Selection for acquisition of zeomycin resistance used cultures grown in 1 l of TAP medium at 15, 22 or 35 jC until stationary phase (f 1 107 cells ml 1). Cells were harvested by centrifugation and resuspended in soft TAPagar (containing 0.5% w/v agar) and spread on the surface
Fig. 1. Constructs used in chloroplast transformation. The figure shows the location of sequences inserted into the C. reinhardtii chloroplast transformation vector pBa3.AX. The constructs containing particular insertions are indicated, as are the coding regions of psaA-3, psbA [partial], and ble.
Four constructs were successfully introduced into the C. reinhardtii chloroplast genome to produce homoplasmic transformants. To check the stability of the constructs within the chloroplast genome, transformant lines were examined by PCR using primers flanking the integration site downstream of psaA-3. For all lines, a PCR product of the predicted size was obtained (data not shown). The first construct included a cassette comprising the ble coding region from S. hindustanus fused to the 5Vand 3Vuntranslated regions of the C. reinhardtii gene RBCS2 for the small subunit of ribulose bis-phosphate carboxylase (in plasmid pBa3.AX.ble). This cassette has been shown to function as a nuclear dominant selectable marker (Stevens et al., 1996). The resulting strain was designated CRble for simplicity. The second construct contained the ble gene modified by the insertion of two copies of the first intron of RBCS2 of C. reinhardtii (in plasmid pBa3.AX.117). The insertion of C. reinhardtii introns into the ble gene has previously been shown to enhance its expression (Lumbreras et al., 1998). The resulting strain was designated CRble.int. The third and fourth constructs (using plasmids pBa3.AX.ble.psbA and pBa3.AX.117.psbA) were as the first and the second except that a region, including a region upstream of the psbA gene through to part of the second exon, was inserted upstream of the ble coding
36
D.L. Lister et al. / Gene 316 (2003) 33–38
regions, and the resulting strains were designated as CRble.psbA and CRble.int.psbA, respectively. We screened the four different transformant lines for chloroplast-to-nucleus DNA transfer events that resulted in the introduction of the ble marker into the nuclear genome by plating large numbers (109 – 1010) of cells on medium containing zeomycin and scoring for antibiotic-resistant colonies. A range of different treatments was also employed prior to plating on the selective medium with the aim of affecting the integrity of the chloroplast envelope membrane. These treatments included growth under low and high temperatures, growth in high salt to produce osmotic stress, growth under high UV irradiance and growth in the presence of colchicine, a drug that interferes with cytokinesis in Chlamydomonas (Harris, 1989). The numbers of zeomycin-resistant colonies observed with prior treatment at different growth temperature are shown in Table 1. No colonies were seen when similar numbers of cells that lacked the ble constructs were plated on medium containing zeomycin. No zeomycin-resistant colonies were observed when strains CRble and CRble.int were treated with high salinity, UV or colchicine. The zeomycin resistance of the individual resistant colonies obtained after growth at 15 and 22 jC and several of the resistant colonies obtained after growth at 35 jC were assessed by spotting onto medium containing a range of zeomycin concentrations (Fig. 2). All showed some level of zeomycin resistance above that seen with the original chloroplast transformants although the level varied between different lines. 3.2. Characterization of zeomycin-resistant cells Examination of each of the zeomycin-resistant lines by light microscopy showed that most had an unusual morphology, being generally nonmotile in liquid culture and forming clumpy (palmelloid) colonies rather than single cells. This phenomenon has been reported in laboratory strains of C. reinhardtii and appears to be the consequence of mutation resulting in the failure of daughter cells to hatch from the mother cell wall during vegetative cell division (Harris, 1989). This palmelloid phenotype represented a severe hindrance to genetic studies aimed at investigating the inheritance of the zeomycin-resistant phenotype. Nonetheless, strains showing the best motility were used in crosses to a wild-type strain (mt ) as described by Harris Table 1 Numbers of zeomycin-resistant colonies observed when 109 – 1010 cells carrying ble constructs were plated on medium containing zeomycin after growth at a range of temperatures
15 jC 22 jC 35 jC
CRble
CRble.int
CRble.psbA
CRble.int.psbA
N.D. 0 0
0 0 0
1 (7 109) 1 (4 109) 52 (2.4 109)
N.D. 0 0
Where zeomycin-resistant colonies were observed, the number in parentheses indicates the number of cells plated. N.D., not determined.
Fig. 2. Assay of zeomycin resistance after initial selection. Colonies showing zeomycin resistance were grown to stationary phase, and aliquots spotted onto agar plates containing a range of concentrations of zeomycin. The figure shows the results with wild type (sensitive) cells, a zeomycinresistant strain (M86) that has a single resistance gene in the nucleus, an example of one of the strains used for selection (CRble.psbA), the resistant strain recovered after growth at 15 jC (CR15), the resistant strain recovered after growth at 22 jC (CR22), and examples of four of the resistant strains recovered after growth at 35 jC (CR35/1 – 4). The numbers along the top show the zeomycin concentration in Ag ml 1.
(1989). If resistance were conferred by a nuclear gene, such a mating would be expected to produce progeny with anything from half showing zeomycin resistance (if the parental resistant cell contained a single resistance gene) to all showing zeomycin resistance (if there were multiple copies of the resistance gene throughout the genome of the parental resistant cell). Although the crosses resulted in the formation of diploid zygospores, the resulting haploid progeny showed poor viability and no clear ratio of zeomycin resistant to sensitive cells could be obtained. Furthermore, all viable progeny were found to be sensitive to much lower concentrations of zeomycin than the parental strains. All progeny cells were killed by zeomycin at 10 Ag ml 1 and some were killed by zeomycin at 5 Ag ml 1, in contrast to the parental lines used, which were resistant to zeomycin at 20 Ag ml 1. DNA was prepared from six of the original zeomycinresistant lines and subjected to restriction digestion, electrophoresis and Southern blotting. The blots were hybridized with a probe specific for the ble gene, but no bands in addition to those due to the chloroplast copies of ble were
D.L. Lister et al. / Gene 316 (2003) 33–38
seen (data not shown). A positive control (strain M86) known to have a single copy of the ble gene in the nucleus gave a clear signal in the Southern blot.
4. Discussion Despite repeated attempts using considerable numbers of cells (>109 per experiment) the majority of the chloroplast transformants carrying the ble constructs did not give rise to any zeomycin-resistant colonies, whether grown under normal or stressed conditions. Because the ble constructs used are known to be able to confer resistance to zeomycin when expressed in the nucleus, we infer that they failed to transfer to the nucleus under the conditions used. For the transformant strain that did give rise to zeomycin-resistant colonies, these were observed at a maximum frequency of 1 in 4 107. Because this colony count was made at the end of the 35 jC incubation and therefore includes any clonal multiplication of these cells during that period, the actual frequency is likely to be lower. In any case, it seems unlikely that the zeomycin-resistant phenotype observed was due to stable integration of the marker into the nucleus of the kind that was observed for the neomycin phosphotransferase gene in N. tabacum (Huang et al., 2003). Rather, it is more likely that the rare examples of increased resistance were a consequence of mutation resulting in the formation of palmelloid colonies such that aggregation of large numbers of cells during growth limited drug uptake and cell death. This is further supported by the genetic analysis because viable progeny would tend to be those without the palmelloid phenotype, and these would be expected to show lower resistance to zeomycin. It was also not possible to demonstrate a nuclear copy of the marker in a Southern blot (although, if a large piece of chloroplast DNA had been inserted into the nucleus, one would not be able to differentiate hybridization to the ble restriction fragment within this from hybridization to the bona fide chloroplast-located sequence). Given the low numbers of zeomycin-resistant cells seen, it is difficult to assess whether the presence of psbA sequence or the growth at elevated temperature had any significant effect on the frequency with which resistant cells are generated. The study of Huang et al. (2003) showed stable transfer of the neomycin phosphotransferase gene from the chloroplast to the nucleus in a minimum of approximately 1 pollen grain in 16,000 in N. tabacum. We therefore infer that the frequency of transfer in C. reinhardtii, if it occurs at all, is much lower than in N. tabacum. Consistent with this, a BLAST search of the C. reinhardtii nuclear genome (http:// genome.jgi-psf.org/chlre1/) with the entire chloroplast genome (Maul et al., 2002) with a filter of 100 bp failed to identify any chloroplast sequences in the nucleus (S.P., unpublished observations). In comparison, 17 such insertions totaling in excess of 11 kbp were observed for Arabidopsis thaliana (Arabidopsis Genome Initiative,
37
2000) and very large insertions of 18 and 33 kb of chloroplast DNA have been reported in the nuclear genomes of tobacco and rice, respectively (Ayliffe and Timmis, 1992; Yuan et al., 2002). Although a number of genes that are located in the chloroplast in higher plants are located in the nucleus in C. reinhardtii, many of these genes are also located in the nucleus in Chlorella, indicating that the transfer was not recent (Maul et al., 2002). Furthermore, the C. reinhardtii nuclear genes are generally GC-rich and contain introns, also suggesting that the transfer was not recent (allowing the genes time to acquire typical nuclear features). Taken together, our experiments and these observations suggest that although C. reinhardtii and other green algae may in the past have lost more genes to the nucleus than green plants (Maul et al., 2002), the rate of gene flow has subsequently slowed dramatically and the movement of DNA from chloroplast to nucleus is now very rare indeed, if it occurs at all. It is interesting to speculate why this should be. C. reinhardtii has a single chloroplast. If the mechanism of transfer of DNA to the nucleus involved chloroplast rupture, it would be very unlikely that a organism with a single chloroplast could undergo this while retaining viability. The plastid compartment is the site of several important biosynthetic pathways in addition to photosynthesis. Consequently, loss of a functional plastid would be lethal even in organisms that grow heterotrophically such as the plastidbearing protists (Wilson et al., 2003). Conversely, chloroplast rupture need not result in cell death in higher plants, where there are many chloroplasts per cell. It would be interesting to see if green algal species with multiple chloroplasts have higher frequencies of DNA transfer to the nucleus. It has been argued (Allen, 1993) that one reason for the retention of an organelle genome is to allow for rapid redox regulation of gene expression. In addition, it seems possible that in some species, cell ultrastructure and the requirement to retain an intact chloroplast prohibit further gene loss from taking place.
Acknowledgements We would like to thank Bill Martin for helpful discussions, the Wellcome Trust, the Gatsby Charitable Foundation and the Leverhulme Trust for their financial support. JMB was supported by a studentship from the Biotechnology and Biological Sciences Research Council, U.K.
References Allen, J.F., 1993. Control of gene expression by redox potential and the requirement for chloroplast and mitochondrial genomes. J. Theor. Biol. 165, 609 – 631. Allen, J.F., 2003. The function of genomes in bioenergetic organelles. Philos. Trans. R. Soc. B 358, 19 – 38. Allen, J.F., Raven J.A., 1996. Free-radical-induced mutation vs. redox
38
D.L. Lister et al. / Gene 316 (2003) 33–38
regulation: costs and benefits of genes in organelles. J. Mol. Evol. 42, 482 – 492. Arabidopsis Genome Initiative, 2000. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796 – 815. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K. (Eds.), 2001. Current Protocols in Molecular Biology. Wiley, New York. Ayliffe, M.F., Timmis, J.N., 1992. Plastid DNA sequence homologies in the tobacco nuclear genome. Mol. Gen. Genet. 236, 105 – 112. Blanchard, J.L., Schmidt, G.W., 1995. Pervasive migration of organellar DNA to the nucleus in plants. J. Mol. Evol. 41, 397 – 406. Cerutti, H., Jagendorf, A., 1995. Movement of DNA across the chloroplast envelope: implications for the transfer of promiscuous DNA. Photosynth. Res. 46, 329 – 337. Glo¨ckner, G., Rosenthal, A., Valentin, K., 2000. The structure and gene repertoire of an ancient red algal plastid genome. J. Mol. Evol. 51, 382 – 390. Goldschmidt-Clermont, M., 1991. Transgenic expression of aminoglycoside adenine transferase in the chloroplast: a selectable marker for sitedirected transformation of Chlamydomonas. Nucleic Acids Res. 19, 4083 – 4089. Hallahan, B.J., Purton, S., Ivison, A., Wright, D., Evans, M.C.W., 1995. Analysis of the proposed Fe – S – X binding region of photosystem I by site directed mutation of PsaA in Chlamydomonas reinhardtii. Photosynth. Res. 46, 257 – 264. Harris, E.H., 1989. The Chlamydomonas Sourcebook: A Comprehensive Guide to Biology and Laboratory Use. Academic Press, San Diego. Howe, C.J., Barbrook, A.C., Lockhart, P.J., 2000. Organelle genes: do they jump or are they pushed? Trends Genet. 16, 65 – 66. Howe, C.J., Barbrook, A.C., Koumandou, V.L., Nisbet, R.E.R., Symington, H.A., Wightman, T.F., 2003. Evolution of the chloroplast genome. Philos. Trans. R. Soc. B 358, 99 – 107. Huang, C.Y., Ayliffe, M.A., Timmis, J.N., 2003. Direct measurement of the transfer rate of chloroplast DNA into the nucleus. Nature 422, 72 – 76. Jarvis, P., Soll, J., 2001. Toc, tic and chloroplast import. Biochim. Biophys. Acta 1541, 64 – 79. Lumbreras, V., Stevens, D.R., Purton, S., 1998. Efficient foreign gene
expression in Chlamydomonas reinhardtii mediated by an endogenous intron. Plant J. 14, 441 – 447. Maliga, P., 1993. Towards plastid transformation in flowering plants. Trends Biotech. 11, 101 – 107. Martin, W., Stoebe, B., Goremykin, V., Hansmann, S., Hasegawa, M., Kowallik, K.V., 1998. Gene transfer to the nucleus and the evolution of chloroplasts. Nature 393, 162 – 165. Martin, W., Rujan, T., Richly, E., Hansen, A., Cornelsen, S., Lins, T., Leister, D., Stoebe, B., Hasegawa, M., Penny, D., 2002. Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proc. Natl Acad. Sci. U. S. A. 99, 12246 – 12251. Maul, J.E., Lilly, J.W., Cui, L., de Pamphilis, C.W., Miller, W., Harris, E.H., Stern, D.B., 2002. The Chlamydomonas reinhardtii plastid chromosome: islands of genes in a sea of repeats. Plant Cell 14, 2659 – 2679. Stevens, D.R., Rochaix, J.-D., Purton, S., 1996. The bacterial phleomycin resistance gene ble as a dominant selectable marker in Chlamydomonas. Mol. Gen. Genet. 251, 23 – 30. Thorsness, P.E., Fox, T.D., 1990. Escape of DNA from the mitochondria to the nucleus in Saccharomyces cerevisiae. Nature 346, 376 – 379. Thorsness, P.E., White, K.H., Fox, T.D., 1993. Inactivation of YME1, a member of the ftsH – sec18 – pas1 – cdc48 family of putative ATPase encoding genes, causes increased escape of DNA from mitochondria in Saccharomyces cerevisiae. Mol. Cell. Biol. 13, 5418 – 5426. Wilson, R.J.M., Rangachari, K., Saldanha, J.W., Rickman, L., Buxton, R.S., Eccleston, J.F., 2003. Parasite plastids: maintenance and functions. Philos. Trans. R. Soc. B 358, 155 – 162. Yoon, H.S., Hackett, J.D., Bhattacharya, D., 2002. A single origin of the peridinin- and fucoxanthin-containing plastids in dinoflagellates through tertiary endosymbiosis. Proc. Natl Acad. Sci. U. S. A. 99, 11724 – 11729. Yuan, Q., Hill, J., Hsiao, J., Moffat, K., Ouyang, S., Jiang, J., Cuell, C.R., 2002. Genome sequencing of a 239-kb region of rice chromosome 10 L reveals a high frequency of gene duplication and a large chloroplast DNA insertion. Mol. Gen. Genomics 267, 713 – 720.
DNA transfer from chloroplast to nucleus is much rarer in Chlamydomonas than in tobacco Diane L. Lister a,1, Joseph M. Bateman b,2, Saul Purton b, Christopher J. Howe a,* a
Centre for Molecular Recognition, Department of Biochemistry, University of Cambridge, Downing Site, Tennis Court Road, Cambridge, CB2 1QW, UK b Department of Biology, University College London, London, UK Received 3 June 2003; accepted 13 June 2003 Received by W. Martin
Abstract By transforming chloroplasts with an antibiotic-resistance gene under the control of a nuclear-specific promoter, we employed a selection scheme to detect the transfer of DNA from the chloroplast to the nucleus in the green alga Chlamydomonas reinhardtii. Among several billion homoplasmic cells tested, we were unable to detect any stable nuclear integration of chloroplast DNA under normal growth conditions or under stress conditions. This contrasts with results reported for the transfer of DNA from chloroplast to nucleus in higher plants and from mitochondrion to nucleus in Saccharomyces cerevisiae. Furthermore, we were unable to detect chloroplast DNA-derived sequences among nuclear genome data for C. reinhardtii, which also contrasts with the situation in higher plants. Taken together, these findings suggest that there is presently little, if any, movement of DNA from chloroplast to nucleus in C. reinhardtii, which may reflect the ultrastructure of the C. reinhardtii cell. D 2003 Elsevier B.V. All rights reserved. Keywords: Endosymbiosis; Genome evolution; Organelle; Gene transfer; Transposition; Translocation
1. Introduction It is generally accepted that chloroplasts originated from the primary endosymbiotic acquisition of oxygenic photosynthetic bacteria by nonphotosynthetic cells. In some lineages, secondary and even tertiary endosymbioses have subsequently occurred (Yoon et al., 2002). Although there is still some controversy over whether all chloroplasts are ultimately derived from a single endosymbiosis (a monophyletic origin) or from several (polyphyletic origins), it is
Abbreviations: bp, base pair; HSM, high salt medium; kb, kilobase pair; mt, mating type; PCR, polymerase chain reaction; TAP, Tris-acetatephosphate; UV, ultraviolet. * Corresponding author. Tel.: +44-1223-333688; fax: +44-1223333345. E-mail address: [email protected] (C.J. Howe). 1 Current address: Department of Pathology, University of Cambridge, UK. 2 Current address: Cancer Research UK Developmental Patterning Laboratory, London, UK. 0378-1119/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0378-1119(03)00754-6
clear that the evolution of the organelle involved the transfer of a large portion of the genetic information of the original endosymbiont(s) to the nucleus of the host organism (Martin et al., 1998, 2002; Howe et al., 2003). Although chloroplasts contain more than a thousand different protein species, their genomes now typically encode of the order of 100 – 200 proteins (Glo¨ckner et al., 2000). The remainder of the chloroplast proteins are encoded in the nucleus and imported posttranslationally through a machinery that is derived at least in part from the bacterial ancestor of the chloroplast (reviewed by Jarvis and Soll, 2001). A number of suggestions have been made as to why transfer of genes to the nucleus might provide a selective advantage (Allen and Raven, 1996; Howe et al., 2000). One suggestion is that transfer to the nucleus isolates genes from damaging species, such as oxygen free radicals, that may be generated during photosynthesis. A second suggestion is that placing a gene in a sexual population (as opposed to the asexual population represented by the uniparentally inherited chloroplast) is advantageous, and a third is that movement of a gene to the nucleus avoids deleterious effects due to the
34
D.L. Lister et al. / Gene 316 (2003) 33–38
biased nucleotide composition associated with many chloroplast genomes. Similarly, several hypotheses have been proposed to explain why a subset of genes has apparently resisted the selective advantages of chloroplast-to-nucleus transfer. These hypotheses include the need to couple the expression of certain genes to a redox-mediated regulatory mechanism within the chloroplast and the difficulty of importing some proteins into the organelle (reviewed by Allen, 2003). Currently, nothing is known of the molecular mechanism of transfer of genes from chloroplast to nucleus although evidence of transfer is seen in the nuclear genomes of higher plants in the form of large tracts (occasionally many kilobases) of chloroplast-derived DNA as well as numerous smaller DNA fragments (Ayliffe and Timmis, 1992; Blanchard and Schmidt, 1995; Arabidopsis Genome Initiative, 2000; Yuan et al., 2002). Studies have been carried out on the analogous transfer of DNA from mitochondrion to nucleus in yeast (Thorsness and Fox, 1990). These authors inserted into the mitochondrial genome a marker (URA3) that would be functional only in the nucleus and then selected for acquisition of URA3 function. They demonstrated an unexpectedly high frequency of transfer from mitochondrion to nucleus (approximately 2 10 5 per cell per generation). In contrast, no transfer of DNA from nucleus to mitochondrion was detectable. The frequency of transfer from mitochondrion to nucleus was increased by treatments expected to disrupt mitochondrial membranes such as growth at elevated temperature or in medium containing 15% glycerol or by freeze thawing. In addition, inactivation of the YME1 gene, encoding a homologue of FtsH postulated as being involved in mitochondrial septum formation, also led to an increased frequency of transfer from mitochondrion to nucleus (Thorsness et al., 1993). This idea that DNA escape from the organelle is a consequence of a breach in the structural integrity of the envelope membrane is supported by studies of DNA movement across the membrane of isolated chloroplasts from pea (Cerutti and Jagendorf, 1995). Recently, Huang et al. (2003) demonstrated directly the transfer of DNA from chloroplast to nucleus in transgenic plants of Nicotiana tabacum. That study used a neomycin phosphotransferase (kanamycin resistance) coding region modified for nuclear expression by attachment of a promoter active in the nucleus and a nuclear intron. The gene was inserted into the chloroplast genome, where it failed to confer kanamycin resistance. Seedling progeny from a cross using a male parent containing this chloroplast genome were selected on kanamycin. The results indicated that the chloroplast kanamycin-resistance gene had been transferred to the nucleus in 16 individuals out of 250,000 seedlings screened. The demonstration of a significant frequency of transfer is important in allowing the process to be studied in the laboratory and also for its implications for plant genetic manipulation. It has been suggested that insertion of transgenes into the chloroplast would reduce the likelihood of
their escaping into the environment because, in most crop species, chloroplasts are not transmitted through the pollen (Maliga, 1993). If gene flow occurs at a significant rate from the chloroplast to the nucleus, this would greatly reduce the strength of this biological ‘containment’. Using the green alga Chlamydomonas reinhardtii, we have carried out similar experiments to those described by Huang et al. (2003) for N. tabacum. We inserted into the chloroplast genome a number of separate constructs based on the ble gene for resistance to the antibiotic zeomycin designed so that the gene would confer resistance only if located in the nucleus (Stevens et al., 1996; Lumbreras et al., 1998). After growth under a range of conditions, cells were selected for resistance to zeomycin, which would be expected if the gene had transferred from the chloroplast to the nucleus. We find that, in contrast to the observations with N. tabacum, DNA transfer from the chloroplast to the nucleus does not occur at a detectable rate in C. reinhardtii and speculate that the presence of only a single chloroplast in the cell prohibits DNA escape to the nucleus.
2. Materials and methods 2.1. Strains and media The C. reinhardtii strains used were the photosystem Ideficient strain C575D (mt+) that has a mutation in the chloroplast gene psaA-3 (Hallahan et al., 1995), the zeomycin-resistant strain M86 that contains a single copy of the ble gene in the nuclear genome and a wild type strain (mt ). Cells were grown in Tris-acetate-phosphate (TAP) medium or high salt minimal (HSM) medium (Harris, 1989). Cell numbers were estimated using a haemocytometer. Media were supplemented as appropriate with 2% w/v agar and zeomycin (Zeocink, Cayla, France) at a range of concentrations. Growth was routinely at 22 jC at an illumination of f 50 AE m 2 s 1 white light. Zeomycin resistance was scored by spotting 5 Al of cells grown to stationary phase (f 1 107 cells ml 1) in TAP medium onto TAP-agar plates with varying zeomycin concentrations. 2.2. Strain construction All recombinant DNA work was performed using Escherichia coli strains DH5a or TG1 and methods as described by Ausubel et al. (2001). The basic plasmid used for construction of C. reinhardtii strains with zeomycinresistance genes in the chloroplast was pBa3.AX.ble, which includes an EcoRI cassette comprising the zeomycin-resistance coding region from Streptoalloteichus hindustanus flanked by 794 bp of the 5V untranslated region and 231 bp of the 3Vuntranslated region of the C. reinhardtii gene RBCS2. This cassette has been shown to function as a dominant nuclear marker in C. reinhardtii (Stevens et al.,
D.L. Lister et al. / Gene 316 (2003) 33–38
1996). The EcoRI cassette was inserted into the EcoRI site of the plasmid pBa3.AX (Hallahan et al., 1995), which contains exon 3 of the psaA gene of C. reinhardtii in a pBluescript vector, to generate pBA3.AX.ble (Fig. 1). Integration of the plasmid into the chloroplast genome can be selected by restoration of photosynthetic growth to the strain C575D which lacks a functional psaA-3 sequence (Hallahan et al., 1995). The plasmid pBA3.AX.ble was modified to form pBa3.AX.117 by replacement of the EcoRI cassette bearing the conventional zeomycin-resistance gene with one containing two copies of the first intron of the C. reinhardtii gene RBCS2. The presence of introns in foreign genes has been shown to enhance expression in C. reinhardtii (Lumbreras et al., 1998). The plasmids pBa3.AX.ble.psbA and pBa3.AX.117.psbA were generated from pBa3.AX.ble and pBa3.AX.117, respectively, by insertion into the MluI site of a 1.8 kb polymerase chain reaction (PCR) product extending from a position 132 upstream from the initiation codon through to the second exon of the C. reinhardtii gene psbA. The four constructs were introduced into C. reinhardtii by particle bombardment of lawns of C575D mutant cells on HSM-agar, essentially as described by Goldschmidt-Clermont (1991). Colonies of transformant cells were visible after 3-week incubation in the light at 22 jC and subjected to another three rounds of propagation on HSM to ensure homoplasmicity. The insertion of the ble constructs into the chloroplast genome downstream of psaA-3 was verified by Southern blotting and PCR analysis.
35
of TAP-agar containing zeomycin at 20 Ag ml 1 in covered bioassay dishes (243 243 18 mm). Colonies obtained were streaked onto TAP-agar containing zeomycin at 20 Ag ml 1 and, if growth occurred, were spot tested on a range of zeomycin concentrations. Mating experiments were carried out as described by Harris (1989). Colchicine treatment was carried out by addition to a stationary phase culture containing 109 –1010 cells in TAP medium to a final concentration of 5 mM. Cells were incubated for 2.5 h, washed, incubated in fresh TAP medium overnight at 22 jC and then plated onto TAP-agar containing zeomycin. Salt stressing was carried out by adding NaCl to a final concentration of 0.2 M to stationary phase cells in TAP medium, followed by overnight incubation at 22 jC, washing and selection on TAP-agar containing zeomycin. Ultraviolet (UV) stressing was carried out by incubation of 5 ml aliquots of cells, containing 2.5 107 cells, adjacent to a Stratagene crosslinker providing a total irradiation of 2 104 – 5 104 AJ.cm 2. Cells were then added to 250 ml aliquots of TAP, incubated for 3 days at 22 jC and then plated in the presence of zeomycin. Counting of viable cells showed that the UV conditions used killed up to 70% of cells. Larger quantities (1010) of cells were also subjected to UV treatment by stirring under UV irradiation provided by a bactericidal lamp in a microbiological flow hood.
3. Results 3.1. Selection for zeomycin resistance
2.3. Selection for zeomycin resistance Selection for acquisition of zeomycin resistance used cultures grown in 1 l of TAP medium at 15, 22 or 35 jC until stationary phase (f 1 107 cells ml 1). Cells were harvested by centrifugation and resuspended in soft TAPagar (containing 0.5% w/v agar) and spread on the surface
Fig. 1. Constructs used in chloroplast transformation. The figure shows the location of sequences inserted into the C. reinhardtii chloroplast transformation vector pBa3.AX. The constructs containing particular insertions are indicated, as are the coding regions of psaA-3, psbA [partial], and ble.
Four constructs were successfully introduced into the C. reinhardtii chloroplast genome to produce homoplasmic transformants. To check the stability of the constructs within the chloroplast genome, transformant lines were examined by PCR using primers flanking the integration site downstream of psaA-3. For all lines, a PCR product of the predicted size was obtained (data not shown). The first construct included a cassette comprising the ble coding region from S. hindustanus fused to the 5Vand 3Vuntranslated regions of the C. reinhardtii gene RBCS2 for the small subunit of ribulose bis-phosphate carboxylase (in plasmid pBa3.AX.ble). This cassette has been shown to function as a nuclear dominant selectable marker (Stevens et al., 1996). The resulting strain was designated CRble for simplicity. The second construct contained the ble gene modified by the insertion of two copies of the first intron of RBCS2 of C. reinhardtii (in plasmid pBa3.AX.117). The insertion of C. reinhardtii introns into the ble gene has previously been shown to enhance its expression (Lumbreras et al., 1998). The resulting strain was designated CRble.int. The third and fourth constructs (using plasmids pBa3.AX.ble.psbA and pBa3.AX.117.psbA) were as the first and the second except that a region, including a region upstream of the psbA gene through to part of the second exon, was inserted upstream of the ble coding
36
D.L. Lister et al. / Gene 316 (2003) 33–38
regions, and the resulting strains were designated as CRble.psbA and CRble.int.psbA, respectively. We screened the four different transformant lines for chloroplast-to-nucleus DNA transfer events that resulted in the introduction of the ble marker into the nuclear genome by plating large numbers (109 – 1010) of cells on medium containing zeomycin and scoring for antibiotic-resistant colonies. A range of different treatments was also employed prior to plating on the selective medium with the aim of affecting the integrity of the chloroplast envelope membrane. These treatments included growth under low and high temperatures, growth in high salt to produce osmotic stress, growth under high UV irradiance and growth in the presence of colchicine, a drug that interferes with cytokinesis in Chlamydomonas (Harris, 1989). The numbers of zeomycin-resistant colonies observed with prior treatment at different growth temperature are shown in Table 1. No colonies were seen when similar numbers of cells that lacked the ble constructs were plated on medium containing zeomycin. No zeomycin-resistant colonies were observed when strains CRble and CRble.int were treated with high salinity, UV or colchicine. The zeomycin resistance of the individual resistant colonies obtained after growth at 15 and 22 jC and several of the resistant colonies obtained after growth at 35 jC were assessed by spotting onto medium containing a range of zeomycin concentrations (Fig. 2). All showed some level of zeomycin resistance above that seen with the original chloroplast transformants although the level varied between different lines. 3.2. Characterization of zeomycin-resistant cells Examination of each of the zeomycin-resistant lines by light microscopy showed that most had an unusual morphology, being generally nonmotile in liquid culture and forming clumpy (palmelloid) colonies rather than single cells. This phenomenon has been reported in laboratory strains of C. reinhardtii and appears to be the consequence of mutation resulting in the failure of daughter cells to hatch from the mother cell wall during vegetative cell division (Harris, 1989). This palmelloid phenotype represented a severe hindrance to genetic studies aimed at investigating the inheritance of the zeomycin-resistant phenotype. Nonetheless, strains showing the best motility were used in crosses to a wild-type strain (mt ) as described by Harris Table 1 Numbers of zeomycin-resistant colonies observed when 109 – 1010 cells carrying ble constructs were plated on medium containing zeomycin after growth at a range of temperatures
15 jC 22 jC 35 jC
CRble
CRble.int
CRble.psbA
CRble.int.psbA
N.D. 0 0
0 0 0
1 (7 109) 1 (4 109) 52 (2.4 109)
N.D. 0 0
Where zeomycin-resistant colonies were observed, the number in parentheses indicates the number of cells plated. N.D., not determined.
Fig. 2. Assay of zeomycin resistance after initial selection. Colonies showing zeomycin resistance were grown to stationary phase, and aliquots spotted onto agar plates containing a range of concentrations of zeomycin. The figure shows the results with wild type (sensitive) cells, a zeomycinresistant strain (M86) that has a single resistance gene in the nucleus, an example of one of the strains used for selection (CRble.psbA), the resistant strain recovered after growth at 15 jC (CR15), the resistant strain recovered after growth at 22 jC (CR22), and examples of four of the resistant strains recovered after growth at 35 jC (CR35/1 – 4). The numbers along the top show the zeomycin concentration in Ag ml 1.
(1989). If resistance were conferred by a nuclear gene, such a mating would be expected to produce progeny with anything from half showing zeomycin resistance (if the parental resistant cell contained a single resistance gene) to all showing zeomycin resistance (if there were multiple copies of the resistance gene throughout the genome of the parental resistant cell). Although the crosses resulted in the formation of diploid zygospores, the resulting haploid progeny showed poor viability and no clear ratio of zeomycin resistant to sensitive cells could be obtained. Furthermore, all viable progeny were found to be sensitive to much lower concentrations of zeomycin than the parental strains. All progeny cells were killed by zeomycin at 10 Ag ml 1 and some were killed by zeomycin at 5 Ag ml 1, in contrast to the parental lines used, which were resistant to zeomycin at 20 Ag ml 1. DNA was prepared from six of the original zeomycinresistant lines and subjected to restriction digestion, electrophoresis and Southern blotting. The blots were hybridized with a probe specific for the ble gene, but no bands in addition to those due to the chloroplast copies of ble were
D.L. Lister et al. / Gene 316 (2003) 33–38
seen (data not shown). A positive control (strain M86) known to have a single copy of the ble gene in the nucleus gave a clear signal in the Southern blot.
4. Discussion Despite repeated attempts using considerable numbers of cells (>109 per experiment) the majority of the chloroplast transformants carrying the ble constructs did not give rise to any zeomycin-resistant colonies, whether grown under normal or stressed conditions. Because the ble constructs used are known to be able to confer resistance to zeomycin when expressed in the nucleus, we infer that they failed to transfer to the nucleus under the conditions used. For the transformant strain that did give rise to zeomycin-resistant colonies, these were observed at a maximum frequency of 1 in 4 107. Because this colony count was made at the end of the 35 jC incubation and therefore includes any clonal multiplication of these cells during that period, the actual frequency is likely to be lower. In any case, it seems unlikely that the zeomycin-resistant phenotype observed was due to stable integration of the marker into the nucleus of the kind that was observed for the neomycin phosphotransferase gene in N. tabacum (Huang et al., 2003). Rather, it is more likely that the rare examples of increased resistance were a consequence of mutation resulting in the formation of palmelloid colonies such that aggregation of large numbers of cells during growth limited drug uptake and cell death. This is further supported by the genetic analysis because viable progeny would tend to be those without the palmelloid phenotype, and these would be expected to show lower resistance to zeomycin. It was also not possible to demonstrate a nuclear copy of the marker in a Southern blot (although, if a large piece of chloroplast DNA had been inserted into the nucleus, one would not be able to differentiate hybridization to the ble restriction fragment within this from hybridization to the bona fide chloroplast-located sequence). Given the low numbers of zeomycin-resistant cells seen, it is difficult to assess whether the presence of psbA sequence or the growth at elevated temperature had any significant effect on the frequency with which resistant cells are generated. The study of Huang et al. (2003) showed stable transfer of the neomycin phosphotransferase gene from the chloroplast to the nucleus in a minimum of approximately 1 pollen grain in 16,000 in N. tabacum. We therefore infer that the frequency of transfer in C. reinhardtii, if it occurs at all, is much lower than in N. tabacum. Consistent with this, a BLAST search of the C. reinhardtii nuclear genome (http:// genome.jgi-psf.org/chlre1/) with the entire chloroplast genome (Maul et al., 2002) with a filter of 100 bp failed to identify any chloroplast sequences in the nucleus (S.P., unpublished observations). In comparison, 17 such insertions totaling in excess of 11 kbp were observed for Arabidopsis thaliana (Arabidopsis Genome Initiative,
37
2000) and very large insertions of 18 and 33 kb of chloroplast DNA have been reported in the nuclear genomes of tobacco and rice, respectively (Ayliffe and Timmis, 1992; Yuan et al., 2002). Although a number of genes that are located in the chloroplast in higher plants are located in the nucleus in C. reinhardtii, many of these genes are also located in the nucleus in Chlorella, indicating that the transfer was not recent (Maul et al., 2002). Furthermore, the C. reinhardtii nuclear genes are generally GC-rich and contain introns, also suggesting that the transfer was not recent (allowing the genes time to acquire typical nuclear features). Taken together, our experiments and these observations suggest that although C. reinhardtii and other green algae may in the past have lost more genes to the nucleus than green plants (Maul et al., 2002), the rate of gene flow has subsequently slowed dramatically and the movement of DNA from chloroplast to nucleus is now very rare indeed, if it occurs at all. It is interesting to speculate why this should be. C. reinhardtii has a single chloroplast. If the mechanism of transfer of DNA to the nucleus involved chloroplast rupture, it would be very unlikely that a organism with a single chloroplast could undergo this while retaining viability. The plastid compartment is the site of several important biosynthetic pathways in addition to photosynthesis. Consequently, loss of a functional plastid would be lethal even in organisms that grow heterotrophically such as the plastidbearing protists (Wilson et al., 2003). Conversely, chloroplast rupture need not result in cell death in higher plants, where there are many chloroplasts per cell. It would be interesting to see if green algal species with multiple chloroplasts have higher frequencies of DNA transfer to the nucleus. It has been argued (Allen, 1993) that one reason for the retention of an organelle genome is to allow for rapid redox regulation of gene expression. In addition, it seems possible that in some species, cell ultrastructure and the requirement to retain an intact chloroplast prohibit further gene loss from taking place.
Acknowledgements We would like to thank Bill Martin for helpful discussions, the Wellcome Trust, the Gatsby Charitable Foundation and the Leverhulme Trust for their financial support. JMB was supported by a studentship from the Biotechnology and Biological Sciences Research Council, U.K.
References Allen, J.F., 1993. Control of gene expression by redox potential and the requirement for chloroplast and mitochondrial genomes. J. Theor. Biol. 165, 609 – 631. Allen, J.F., 2003. The function of genomes in bioenergetic organelles. Philos. Trans. R. Soc. B 358, 19 – 38. Allen, J.F., Raven J.A., 1996. Free-radical-induced mutation vs. redox
38
D.L. Lister et al. / Gene 316 (2003) 33–38
regulation: costs and benefits of genes in organelles. J. Mol. Evol. 42, 482 – 492. Arabidopsis Genome Initiative, 2000. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796 – 815. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K. (Eds.), 2001. Current Protocols in Molecular Biology. Wiley, New York. Ayliffe, M.F., Timmis, J.N., 1992. Plastid DNA sequence homologies in the tobacco nuclear genome. Mol. Gen. Genet. 236, 105 – 112. Blanchard, J.L., Schmidt, G.W., 1995. Pervasive migration of organellar DNA to the nucleus in plants. J. Mol. Evol. 41, 397 – 406. Cerutti, H., Jagendorf, A., 1995. Movement of DNA across the chloroplast envelope: implications for the transfer of promiscuous DNA. Photosynth. Res. 46, 329 – 337. Glo¨ckner, G., Rosenthal, A., Valentin, K., 2000. The structure and gene repertoire of an ancient red algal plastid genome. J. Mol. Evol. 51, 382 – 390. Goldschmidt-Clermont, M., 1991. Transgenic expression of aminoglycoside adenine transferase in the chloroplast: a selectable marker for sitedirected transformation of Chlamydomonas. Nucleic Acids Res. 19, 4083 – 4089. Hallahan, B.J., Purton, S., Ivison, A., Wright, D., Evans, M.C.W., 1995. Analysis of the proposed Fe – S – X binding region of photosystem I by site directed mutation of PsaA in Chlamydomonas reinhardtii. Photosynth. Res. 46, 257 – 264. Harris, E.H., 1989. The Chlamydomonas Sourcebook: A Comprehensive Guide to Biology and Laboratory Use. Academic Press, San Diego. Howe, C.J., Barbrook, A.C., Lockhart, P.J., 2000. Organelle genes: do they jump or are they pushed? Trends Genet. 16, 65 – 66. Howe, C.J., Barbrook, A.C., Koumandou, V.L., Nisbet, R.E.R., Symington, H.A., Wightman, T.F., 2003. Evolution of the chloroplast genome. Philos. Trans. R. Soc. B 358, 99 – 107. Huang, C.Y., Ayliffe, M.A., Timmis, J.N., 2003. Direct measurement of the transfer rate of chloroplast DNA into the nucleus. Nature 422, 72 – 76. Jarvis, P., Soll, J., 2001. Toc, tic and chloroplast import. Biochim. Biophys. Acta 1541, 64 – 79. Lumbreras, V., Stevens, D.R., Purton, S., 1998. Efficient foreign gene
expression in Chlamydomonas reinhardtii mediated by an endogenous intron. Plant J. 14, 441 – 447. Maliga, P., 1993. Towards plastid transformation in flowering plants. Trends Biotech. 11, 101 – 107. Martin, W., Stoebe, B., Goremykin, V., Hansmann, S., Hasegawa, M., Kowallik, K.V., 1998. Gene transfer to the nucleus and the evolution of chloroplasts. Nature 393, 162 – 165. Martin, W., Rujan, T., Richly, E., Hansen, A., Cornelsen, S., Lins, T., Leister, D., Stoebe, B., Hasegawa, M., Penny, D., 2002. Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proc. Natl Acad. Sci. U. S. A. 99, 12246 – 12251. Maul, J.E., Lilly, J.W., Cui, L., de Pamphilis, C.W., Miller, W., Harris, E.H., Stern, D.B., 2002. The Chlamydomonas reinhardtii plastid chromosome: islands of genes in a sea of repeats. Plant Cell 14, 2659 – 2679. Stevens, D.R., Rochaix, J.-D., Purton, S., 1996. The bacterial phleomycin resistance gene ble as a dominant selectable marker in Chlamydomonas. Mol. Gen. Genet. 251, 23 – 30. Thorsness, P.E., Fox, T.D., 1990. Escape of DNA from the mitochondria to the nucleus in Saccharomyces cerevisiae. Nature 346, 376 – 379. Thorsness, P.E., White, K.H., Fox, T.D., 1993. Inactivation of YME1, a member of the ftsH – sec18 – pas1 – cdc48 family of putative ATPase encoding genes, causes increased escape of DNA from mitochondria in Saccharomyces cerevisiae. Mol. Cell. Biol. 13, 5418 – 5426. Wilson, R.J.M., Rangachari, K., Saldanha, J.W., Rickman, L., Buxton, R.S., Eccleston, J.F., 2003. Parasite plastids: maintenance and functions. Philos. Trans. R. Soc. B 358, 155 – 162. Yoon, H.S., Hackett, J.D., Bhattacharya, D., 2002. A single origin of the peridinin- and fucoxanthin-containing plastids in dinoflagellates through tertiary endosymbiosis. Proc. Natl Acad. Sci. U. S. A. 99, 11724 – 11729. Yuan, Q., Hill, J., Hsiao, J., Moffat, K., Ouyang, S., Jiang, J., Cuell, C.R., 2002. Genome sequencing of a 239-kb region of rice chromosome 10 L reveals a high frequency of gene duplication and a large chloroplast DNA insertion. Mol. Gen. Genomics 267, 713 – 720.