Photosynthesis Research 44: 75-79, 1995. © 1995 KluwerAcademic Publishers. Printed in the Netherlands.

Regular paper

Isolation and initial characterization of virescent mutants of Arabidopsis

thaliana J u d y A. B r u s s l a n 1 & E l a i n e M. T o b i n Department of Biology, 405 Hilgard Ave, University of California, Los Angeles, Los Angeles, CA 90024-1606, USA; 1Present address: Department of Biological Sciences, 1250 Bellflower Blvd., California State University, Long Beach, Long Beach, CA 90840-3702, USA Received 27 September 1994; accepted in revised form 21 December 1994

Key words: chloroplast development, fast-neutron mutagenesis, fluorescence emission spectroscopy, greening, light-harvesting

Abstract In higher plants, development of the chloroplasts must be coordinated with development of the leaf. In order to study the signals that synchronize these two developmental processes, we have isolated virescent (delayed in greening) mutants ofArabidopsis thaliana. Two such mutants that have pale-green young leaves which gradually green more fully during leaf maturation have been partially characterized. The two, virl and vir2, are due to separate nuclear recessive mutations. The pale leaves of virl and vir2 both had reduced 77 °K fluorescence emission at 730-734 nm relative to that at 686-687 nm, indicating a reduction in the relative amount of LHC I compared to WT. As leaves greened, the amount of LHC I increased to near wildtype levels. The shift in the fluorescence emission peak from 730 nm to 734 nm, characteristic of maturing LHC I, was seen for virl, but not vir2, suggesting that virl is a regulatory mutant while vir2 may be defective in a specific aspect(s) of LHC I function.

Abbreviations: D - dark; EMS - ethyl methanesulfonate; er - erecta; gll - glabrous1; L - light; LHC I - light harvesting complex of Photosystem I; LHC I I - light harvesting complex of Photosystem II; M 2 - second generation of mutagenized seed; M3 - third generation of mutagenized seed; vir - virescent; WT - wildtype Introduction Development of the chloroplast in higher plants has been studied extensively, and much is known about the changes in ultrastructure, photochemical activities, and accumulation of proteins and mRNAs that accompany plastid maturation (Baker 1984; Leech 1984; Mullet 1993; Dreyfuss and Thornber 1994a,b). However, little is known about the regulatory signals that control development of the chloroplast. Many nuclear mutants that affect chloroplast development have been isolated in angiosperms, but with one exception, the genes that are defective in these mutants have not been cloned, and the molecular events underlying these phenotypes are not known.

The iojap mutation of maize has been cloned and sequenced, lojap is a nuclear mutation that results in green and white striped leaves. The plastids located in the white stripes have undeveloped lamellae and lack photosynthetic activity, ribosomes and proteins (Walbot and Coe 1979). The gene responsible for the iojap mutation was cloned via transposon tagging (Han et al. 1992). It encodes a 25 kD protein that has no homology to proteins in databases, thus its function is not yet understood. Another class of mutants that affects chloroplast development displays a delay in greening, and such mutants have been termed virescent. These were first defined as recessive nuclear mutations of maize that have white or yellow seedlings that green after a lapse of time (Emerson 1912; Demerec 1924). These

76 lines are indistinguishable from wildtype (WT) when mature, and only display the virescent phenotype when grown at cooler temperatures (Phinney and Kay 1954). Other virescent lines resulting from recessive nuclear mutations have been isolated from barley (Maclachlan and Zalik 1963), Phaseolus vulgaris (Dale and Heyes 1970), cotton (Benedict et al. 1972), peanut (Benedict and Ketring 1972) and maize (Langdale et al. 1987). The Vir-C mutant of tobacco also displays a virescent phenotype, but it is maternally inherited (Archer and Bonnet 1987). Our understanding of the molecular basis of virescent mutants is limited to the fact that nuclear and plastid genes can control plastid development, but the identity and function of these genes is not known. Genes responsible for virescent mutations are important because they are likely to be involved in coordinating leaf and chloroplast development. Understanding their function could help in elucidating the communication that occurs between the plastid and the nucleus. We describe in this report two newly isolated virescent mutants of Arabidopsis thaliana, the model plant system that is amenable to molecular analysis (Meyerowitz 1987; Hauge et al. 1993). These two mutants differ from other virescent mutants in that they have pale-green and not white or yellow young leaves, but like the other mutants, they accumulate WT levels of chl as they mature, albeit more slowly.

Methods

Mutagenesis EMS mutagenesis was performed as described in Karlin-Neumann (1991) using 20 mM EMS on Arabidopsis thaliana ecotype Columbia. Fast-neutron mutagenesis (60 Gy) was also performed on Arabidopsis thaliana ecotype Columbia by Dr Hans Brunner at the International Atomic Energy Agency in Vienna, Austria. M 1 plants were grown to maturity in the greenhouse, and selfed M2 progeny were collected from families of 1000 M1 plants.

Screening for virescent mutants The screening for virescent mutants was done on GP medium [1 x Murashige and Skoog salts (Sigma), 0.5 g/l MES, pH 5.7, 0.7% phytagar (GibcoBRL)] that contained 25 #M gibberellic acid (GA4+7, Abbott Laboratories). Seeds were sterilized in 10% commer-

cial bleach containing 0.02% Ivory soap (Proctor & Gamble) for 10 min, rinsed twice with sterile Milli-Qpurified H20 (Millipore), resuspended in 500 pl 0.12% sterile agarose (GibcoBRL), and spread onto plates. Seeds were imbibed for 2 d at 4 °C, and then exposed to white light (20/~E/m2/s) for 1 h to stimulate germination before being transferred to darkness for 3 d. Plates containing seedlings were transferred to white light and allowed to green for 16 h. Seedlings that had pale cotyledons were then transplanted to the GP medium described above but containing 2% sucrose and no GA4+7. After one week in a growth chamber (60 #E/m2/s, 16 h L: 8 h D), seedlings were transferred to soil, and grown to maturity. Selfed M3 seeds were collected from individual plants, and retested for the virescent phenotype.

Genetic analysis M3 plants that retained the virescent phenotype were used in reciprocal crosses to Arabidopsis ecotype Columbia that had a glabrous 1 (gll, lacks trichomes) marker. Virescent mutants were also crossed to Arabidopsis ecotype Landsberg that had an erecta (er, shorter stature, shorter siliques)marker. F1 were grown to maturity, and the F2 progeny were analyzed for segregation of the virescent phenotype. F2 seeds were cold-imbibed in 0.1% agarose for 2 days at 4 °C, sown to soil and grown in a low light growth chamber (60 #E/ma/s) for 2 weeks, then put into a high light growth chamber (300/iE/m2/s). This regimen enhanced the virescent phenotype, and plants were scored after 1-2 weeks of growth in high light.

Fluorescence emission spectroscopy Fluorescence emission spectra were taken of whole leaves of Arabidopsis plants that had been grown as described above. New leaves in the center of the rosette were pale while the older, but not senescent, leaves were green. WT leaves of similar age as the pale leaves were used as a control. Leaves were cooled to 77 ° K in liquid N2, and fluorescence emission was measured in an Aminco SPF-500 fluorometer. Excitation was at 436 nm, emission between 600 and 800 nm was recorded, and the bandpass was 2 nm. The peak of emission was determined by visual inspection.

77 Table 1. Genetic analysis of virl and vir2

Cross female × male

F2 phenotype vir % vir

WT

X2

p

> 0.5 >0.1 > 0.5

gll x virl

118

39

25

0.03

virl x er

69 103 102

15 35 16

18 25 14

2.3 0.04 8.7

er x vir2 vir2× gll

Results Screening for virescent mutants was performed on 5,000 EMS mutagenized and 18,000 fast-neutron mutagenized M2 seeds. Sucrose was not included in the growth medium because it can affect chloroplast development (Sheen 1994). After three days of growth in darkness, seedlings were allowed to green in white light, and those with pale or pigmentless cotyledons were selected. Approximately 0.05% of the M2 seedlings were initially selected, but most of these did not survive because they could not accumulate carotenoids and/or chlorophyll. Many other selected lines turned out to be false positives, greening normally when the M3 were retested. From the EMS mutagenized lines, virl, and from the fast-neutron mutagenized lines, vir2 were selected for further study. The visible phenotypes of virl and vir2 are similar. All newly emerging leaves of the rosette are pale green, and full greening occurs within 4 to 5 days. This is significantly slower than WT where leaves emerge fully green. Vir2 has a weaker phenotype than virl, greening a day earlier. The pale phenotype in virl is seen most clearly along the outer leaf margin while in vir2 the pale phenotype is seen throughout the entire leaf. Both mutants grow more slowly than WT, and bolting is delayed by one week, but they are both fertile. The virescent phenotype of both lines is clearly evident in high light (300 #E/m2/s), but is barely visible in plants grown under low light (40 #E/m2/s). Increased temperature (28 °C versus the normal 24 °C) does not rescue the virescent phenotype in either mutant line. Genetic analysis was performed on virl and vir2, and the results are summarized in Table 1. When virl was crossed to gll, one-quarter of the F2 progeny displayed the virescent phenotype demonstrating that virl is a single recessive nuclear mutation. A reciprocal cross in which er was used as the female confirmed this observation. Crosses involving vir2 also showed

> 0.005

Table 2. Analysis of fluorescence emission spectra of leaves

Shorter A

Longer A

Ratio of longer/

maxima (nm) maxima (nm) shorter A emission WT

687 687

734 730

6.3 3.1

687 686

734 732

4.6 2.9

vir2-green 686

732

5.1

virl-pale virl-green vir2-pale

r34

7~4

A

B

c

>

"6

687 )

% ...'.-.;T....." t.

650

800 650

L

.2.-

800

Wavelength (nm) Fig. 1. 77 o K fluorescence emission spectra of whole leaves of WT and the virescent mutants. Fluorescence emission from 650 to 800 nm is shown for leaves of (A) WT and virl and (B) WT and vir2. WT is shown as black lines, fully greened leaves of vir mutants are shown as dashed lines, and pale leaves of vir mutants are show,a as dotted lines.

that vir2 is a single recessive nuclear mutation. In some cases, the percentage of vir in the F2 generation was below 25%, and X2 values were high. This was probably due to slow growth and premature senescence of

78 some vir plants. Scoring of gll and er markers suggested that virl is not linked to either marker, but that vir2 is located approximately 10 map units from gll on chromosome 3. Analysis of additional F2 progeny will be necessary to confirm this linkage. virl and vir2 were also subjected to complementation analysis to determine if the mutations were allelic. virl, er or vir2, er were used as females in crosses to vir2 or virl, respectively. The lack of the er phenotype in the F 1 generation demonstrated that the crosses were successful. F1 plants from reciprocal crosses did not show a delay in greening, demonstrating that virl and vir2 are separate mutations. Fluorescence emission spectra of pale and green leaves of virl and vir2 and for young green leaves of WT are shown in Fig. 1. The emission at 686-687 nm is thought to arise from LHC II, while the emission at 730-734 nm arises from LHC I (Hipkins and Baker 1986). The locations of the emission maxima are listed in Table 2. The peak of emission at 730-734 nm was sharply defined while the peak near 686-687 nm was broader, and thus more difficult to assign a precise wavelength. WT and virl had similar 687 nm maxima, while a 1-nm shift to a shorter wavelength was observed in vir2. The emission maximum arising from LHC I showed a larger shift. In WT, it occurred at 734 nm, while in virl it shifted from 730 to 734 nm during the greening process. In vir2 the LHC I emission maximum is observed at an intermediate wavelength, 732 nm, but it does not shift during greening. Table 2 also shows the ratio of emission at 730-734 nm to emission at 686-687 nm. In WT, the ratio is 6.3, while in both vir mutants the ratio shifts from approximately 3 to 5 during greening.

Discussion Two new virescent mutants, virl and vir2, have been isolated from the model plant A. thaliana. Genetic analysis showed that they are non-allelic, recessive nuclear mutations. The visible phenotypes of virl and vir2 were similar, showing a delay of approximately four days in the greening of all leaves. In contrast, greening in WT occurred prior to leaf emergence. The virescent phenotype of the Arabidopsis lines differs from other virescent mutants isolated in maize, barley, Phaseolus, cotton and peanut. These mutants initially had white or yellow leaves, but then became fully green as other leaves emerged. Our mutants also differ from the original virescent lines isolated in maize

in that they were not rescued when grown at a higher temperature. Interestingly, the phenotypes of virl and vir2 were more obvious when grown under high light (300/~E/me/s). This is probably due to an increase in the rate of leaf growth that cannot be matched by the delayed rate of chloroplast development in the virescent lines. Consistent with this explanation is the observation that the slow rate of greening is not apparent in plants grown in dim light (40 ~E/me/s). Fluorescence emission spectroscopy was used as a diagnostic tool for studying the biogenesis of LHCs in the virescent mutants. In WT Arabidopsis grown under white light, a high 730-734 nm to 686-687 nm ratio was achieved prior to leaf emergence, but in the virescent mutants grown under the same light conditions, this ratio was low, and as leaves greened, it increased to near WT levels. As emission at 730-734 nm arises from LHC I while emission at 686-687 nm is thought to arise from LHC II (Hipkins and Baker 1986), it appears that the ratio of LHC I to LHC II was low in the pale leaves of virl and vir2 and subsequent LHC I biogenesis was greatly slowed. Differences in the wavelength of the emission maximum arising from LHC I were observed in the three lines. In virl, the maximum shifted from 730 to 734 nm during greening. A similar shift to a longer wavelength along with an increase in the ratio of 730-734 nm to 686-687 nm emissions has been observed during greening of intermittent-light grown barley that was exposed to white light (Dreyfuss and Thornber 1994b). Thus, the virl mutant displayed a normal, albeit delayed, pattern of LHC I accumulation which is what would be expected of a regulatory mutant. We have tested phytochrome regulation of Lhcbl*3 gene expression in virl, and have found that it is identical to WT (data not shown), thus ruling out mutations in the phytochrome regulatory network as the cause of the virescent phenotype. In contrast, the vir2 mutant showed a slightly abnormal pattern of LHC I accumulation. Although changes in the ratio of emissions at 730-734 nm to 686-687 nm appeared to follow a normal, but slower progression than WT, the emission maximum was at an intermediate wavelength (732 nm) in pale leaves, and did not shift during development to the 734 nm peak characteristic of the mature complex. It is possible that a component of the PS I antenna is lacking in this mutant which results in the accumulation of a slightly altered LHC I. LHC II content appeared to be normal, having an emission maximum at 687 nm in WT and in virl. There

79 was a small apparent shift in the location of the peak to 686 n m in vir2 suggesting that L H C II, as well as L H C I, m i g h t be altered in this mutant. The use of a non-denaturing ' g r e e n ' gel system for Arabidopsis will enable us to determine whether all L H C protein c o m p l e x e s are present in the vir2 mutant (Peter et al. 1991). The slow greening o f the vir mutants could result f r o m leaky blocks in m a n y aspects of chloroplast function, thus a k n o w l e d g e o f the m o l e c u l a r basis o f these mutations will be required. In the future, the expanding availability o f m o l e c u l a r markers (Hauge et al. 1993; K o n i e c z n y and A u s u b e l 1993; Bell and Ecker 1994) should enable us to c l o n e the genes responsible for the virl and vir2 mutations of Arabidopsis.

Acknowledgements We thank Susanne Preiss for help with fluorescence e m i s s i o n spectroscopy and Philip T h o r n b e r for insightful discussions. This w o r k was supported by grant G M 23167 f r o m the National Institutes o f Health.

References Archer EK and Bonnett HT (1987) Characterization of a virescent chloroplast mutant of tobacco. Plant Physiol 83:920-925 Baker NR (1984) Development of chloroplast photochemical functions. In: Baker NR and Barber J (eds) Chloroplast Biogenesis, pp 207-251. Elsevier Science Publishers, Amsterdam Bell CJ and Ecker JR (1994) Assignment of 30 microsatellite loci to the linkage map ofArabidopsis. Genomics 19:137-144 Benedict CR and Ketring DL (1972) Nuclear gene affecting greening in virescent peanut leaves. Plant Physio149:972-976 Benedict CR, McCree KJ and Kohel RJ (1972) High photosynthetic rate of a chlorophyll mutant of cotton. Plant Physio149:968-971 Dale JE and Heyes JK (1970) A virescens mutant of Phaseolus vulgaris; growth, pigment and plastid characters. New Phyto169: 733-742 Demerec M (1924) Genetic relations of five factor pairs for virescent seedlings in maize. Comell Univ Agr Expt Sta Memoir 84:1-25 Dreyfuss BW and Thornber JP (1994a) Assembly of the lightharvesting complexes (LHCs) of Photosystem II: Monomeric LHC lib complexes are intermediates in the formation of oligomeric LHC lib complexes. Plant Physiol 106:829-839

Dreyfuss BW and Thornber JP (1994b) Organization of the lightharvesting complex of Photosystem I and its assembly during plastid development. Plant Physiol 106:841-848 Emerson RA ( 1912) The inheritance of certain forms of chlorophyll reduction in corn leaves. Nebr Agr Expt Sta Ann Rept 25:89-105 Han C-D, Coe EH Jr and Martienssen R (1992) Molecular cloning and characterization of iojap (ij), a pattern striping gene of maize. EMBO J 11: 4037-4046 Hauge BM, Hanley SM, Cartinhour S, Cherry JM, Goodman HM, Koorneef M, Stam P, Chang C, Kempin S, Medrano L and Meyerowitz EM (1993) An integrated genetic/RFLP map of the Arabidopsis thaliana genome. Plant J 3:745-754 Hipkins MF and Baker NR (1986) Spectroscopy. In: Hipkins MF and Baker NR (eds) Photosynthesis Energy Transduction: A Practical Approach, pp 5 l- 101. IRL Press, Oxford Karlin-Neumann GA (1991) A genetic approach to the dissection of phytochrome-regulated gene transcription in Arabidopsis thaliana. PhD Dissertation, Los Angeles, University of California Konieczny A and Ausubel FM (1993) A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCRbased markers. Plant J 4:403-4 10 Langdale JA, Metzler MC and Nelson T (1987) The Argentia mutation delays normal development of photosynthetic cell-types in Zea mays. Dev Biol 122:243-255 Leech RM (1984) Chloroplast development in angiosperms: current knowledge and future prospects. In: Baker NR and Barber J (eds) Chloroplast Biogenesis, pp 1-21. Elsevier Science Publishers, Amsterdam Maclachlan S and Zalik S (1963) Plastid structure, chlorophyll concentration, and free amino acid composition of a chlorophyll mutant of harley. Can J Bot 41:1053-1060 Meyerowitz EM (1987) Arabidopsis thaliana. Ann Rev Genet 21: 93-111 Mullet JE (1993) Dynamic regulation of chloroplast transcription. Plant Physiol 103:309-313 Peter GF, Takeuchi T and Thomber JP ( 1991) Solubilization and twodimensional electrophoretic procedures for studying the organization and composition of photosynthetic membrane polypeptides. Methods: A Companion to Methods in Enzymology 3: 115-124 Phinney BO and Kay RE (1954) Interaction of environment and genotype in the expression of a virescent gene, pale-yellow- 1, of maize. Hilgardia 23:185-196 Sheen J (1994) Feedback control of gene expression. Photosynth Res 39:427438 Walbot V and Coe EH Jr (1979) Nuclear gene iojap conditions a programmed change to ribosome-less plastids in Zea mays. Proc Natl Acad Sci, USA 76:2760-2764

Isolation and initial characterization of virescent mutants of Arabidopsis thaliana.

In higher plants, development of the chloroplasts must be coordinated with development of the leaf. In order to study the signals that synchronize the...
413KB Sizes 0 Downloads 0 Views