tagged (Meinke, 1992; Castle et al., 1993; Yeung and Meinke,. 1993). ... available the T-DNA-mutagenized populations and Mary Anderson. (Nottingham ...
Plant Physiol. (1994) 106: 1169-1177
arc6, A Fertile Arabidopsis Mutant with Only Two Mesophyll Cell Chloroplasts' Kevin A. Pyke, Stephen M. Rutherford, Elizabeth J. Robertson, and Rache1 M. Leech* Department of Biology, University of York, Heslington, York, YO1 5DD, United Kingdom
size compared to wild type (Pyke and Leech, 1991, 1992, 1994), has required a reassessment of the role of chloroplast division during mesophyll cell development and the attainment of photosynthetic competence. In the stable Mendelian inherited mutants arc3 and arc5 (Pyke and Leech, 1992, 1994), successful chloroplast division is largely prevented, yet the mutant plants grow in a normal manner and are fertile. In those arc mutants with a reduction in the normal chloroplast number the chloroplasts are substantially larger than wild-type chloroplasts, and in arc mutants with more chloroplasts per cell the chloroplasts are smaller than wild type so that the total sizes of the chloroplast compartments per unit cell size in both of these types of mutant and in wild type are similar (Pyke and Leech, 1992, 1994). The high degree of plasticity in the composition of the chloroplast compartment observed in arc mutants suggests that the factors controlling chloroplast biogenesis in the mesophyll cell allow a considerable degree of flexibility in the control of chloroplast division without substantially affecting plant growth. In wild-type Arabidopsis and a11 those arc mutants reported to date (Pyke and Leech, 1992, 1994) the number of young chloroplasts in the smallest, immediately postmitotic mesophyll cells is estimated to be about 14, reflecting the complement of proplastids partitioned into the young postmitotic cells. This suggests that in arc mutants, proplastid division is not radically impaired and the effect of ARC genes is specific to chloroplast division. The arc mutants are a valuable genetic resource that enable the dissection of the nature of the molecular control of chloroplast division. Indeed, our screening of the mutant populations for suitable inherited mutant phenotypes was initiated to enable us to identify suitable mutants for genetic and mechanistic studies of chloroplast division. The isolation of urc genes from the existing arc mutants produced by ethylmethane sulfonate mutagenesis would necessitate a map-based cloning strategy. T-DNA insertional mutagenesis using Agrobacterium-mediated transformation (Feldmann and Marks, 1987; Feldmann, 1991) has an enhanced potential for facilitating gene isolation, since tagged alleles can be used to identify and characterize mutant genes. This paper describes a novel arc mutant isolated in a screen of these T-DNA-mutagenized lines, arc6, in which both chloroplast division and proplastid division are impaired. We report the characterization of this new mutant and consider
A novel mutant of Arabidopsis thaliana, a r d (accumulation and replication of chloroplasts), has been isolated from a transfer DNAmutagenized population of Arabidopsis seedlings. a r 6 has the most extreme arc mutant phenotype we have yet described, with only one to three chloroplasts per leaf mesophyll cell compared to a mean of 83 in cells of the wild-type var Wassilewskija. The chloroplasts of ar& are 20-fold larger than wild-type chloroplasts. Chloroplast division is almost certainly precluded in a r d mesophyll cells, since chloroplast number per cell does not increase during mesophyll cell expansion. arc6 chloroplasts are long and thin in cross-section and only one-half the width of wild-type chloroplasts and the arrangement of thylakoid membranes is largely unaltered. a r d segregates as a monogenic recessive nuclear mutation in a normal Mendelian manner and the arc6 phenotype is stably inherited for at least four generations. a r 6 plants grow normally and are fertile, although the rosette leaves appear curled and twisted. a r d plants accumulate 70 to 75% of the biomass of wild type. The phenotype of this novel mutant is discussed in relation to the nature of the control of chloroplast division in leaf cells.
Chloroplast division has been shown to occur during leaf mesophyll cell development in many species of higher plants (Saurer and Possingham, 1970;Possingham and Smith, 1972; Boffey et al., 1979; Leech et al., 1981) and is considered to be a process vital to the competent development of the chloroplast compartment and of the mesophyll cell (Leech, 1984; Boffey, 1992). The sequence of physical events involved in the division process leading to the production of two daughter chloroplasts has been well characterized by light and electron microscopy (Leech et al., 1981; Hashimoto, 1986). Although several different factors have been shown to affect the number of chloroplast divisions in mesophyll cells, including cell size and nuclear ploidy (Ellis and Leech, 1985; Pyke and Leech, 1987; Leech and Pyke, 1988), genotype (Pyke and Leech, 1987), and several different environmental factors including light quality (Possingham et al., 1975; Hashimoto and Possingham, 1989), the molecular basis of the chloroplast division process and the role of nuclear genes in the control of chloroplast division have not been well established. The recent identification of arc (accumulation and replication of chloroplasts) genes of Arabidopsis thaliana, mutations in which result in radically altered chloroplast number and ~
' Supported by an Agricultura1 and Food Research Council (UK)
Abbreviations: KanR, kanamycin resistant; KanS, kanamycin sensitive; NPTII, neomycin phosphotransferase;T-DNA, transfer DNA; WS, ecotype Wassilewskija.
Plant Molecular Biology I1 grant (LR87/528) to R.M.L. * Corresponding author; fax 44-904-432860. 1169
Pyke et al.
.I 170
the relevance of its novel phenotype to our understanding of the control of chloroplast division in the leaf mesophyll cell. MATERIALS AND METHODS Plant Material and Mutant Screening
Forty-nine sets of mutants of Arabidopsis thaliana cv WS, derived from plants mutagenized by T-DNA insertional mutagenesis (Feldmann and Marks, 1987; Feldmann, 1991), were obtained from the Nottingham Arabidopsis Stock Centre (University of Nottingham, Nottingham, UK; stock No. N3115). Each set contained approximately 12 seeds from each of 100 transformed lines, constituting a total population of 4900 T-DNA transformations. Seeds were sown on compost in a grid arrangement (Pyke and Leech, 1991) and the first leaves from individual plants were harvested 22 d after sowing in a controlled-environment cabinet (Pyke and Leech, 1991). Suspensions of fixed, isolated mesophyll cells from putative mutants were prepared from first leaves (Pyke and Leech, 1991) and screened microscopically for obvious abnormalities in the number and size of chloroplasts per mesophyll cell, using Nomarski differential interference contrast optics. Those plants selected by this screen were allowed to self, seed was collected, and the mutant cellular phenotype was confirmed in the next generation. Analysis of Chloroplast Number and Mesophyll Cell Size
The chloroplast complements of individual mesophyll cells were determined in intact, separated cells using Nomarski differential interference contrast optics and plan areas of individual mesophyll cells and of the chloroplasts within them were determined by image analysis (Pyke and Leech, 1991). Guard-cell chloroplasts were counted after mounting epidermal peels in 1%(w/v) AgN03. Ultrastructural Analysis
First leaves from wild type (WS) and arc6 were harvested 33 d after sowing. Slices of tissue 2 to 3 mm wide were cut from near the distal end of the intact leaf. Tissue was fixed and embedded into Spurr's epoxy resin (TAAB Laboratories Equipment Ltd, Reading, Berkshire, UK) (Spurr, 1969) as described previously (Robertson et al., 1993). Ultrathin sections (90 nm) were cut using a diamond knife and mounted on 200-square mesh copper grids coated with O.S%'formvar. Double staining in uranyl acetate and lead citrate preceded examination on a Jeol JEM-1200EX electron microscope. Thick (1 pm) sections were also cut, stained with toluidine blue, and observed by light microscopy to determine leaf anatomy and cell shape. Measurements of chloroplast dimensions and thylakoid membrane arrangement were made from electron micrographs by image analysis (Seescan Imaging Ltd, Cambridge, UK). Assay for Kanamycin Resistance
The T-DNA used in the mutagenesis procedure contains a single NPTII gene conveying resistance to kanamycin (Feldmann and Marks, 1987; Feldmann, 1991). To analyze the
Plant Physiol. Vol. 106, 1994
segregation pattems of kanamycin resistance, seeds in sachets of folded Whatman 3M filter paper were sterilized by immersion first in 70% ethanol for 2 min then in 5% sodium hypochlorite containing 0.5% SDS for 15 min (Balc,ells, 1991). After two 5-min washes in sterile, double-distilled water, the seeds were plated onto germination medium containing 50 mg/L kanamycin (Valvekens et al., 1988). Seeds were arranged on a 10 X 10 array on each plate to facilitate scoring. A pretreatment of 4OC in the dark for 48 h prior to germination in the light at 22OC resulted in >98% germination. Each plate contained wild-type seeds as a KanS control. RESULTS
The relationship between chloroplast number and mesophyll cell plan area was first established for WS (F'ig. 1).This is necessary because the T-DNA insertional mutants were constructed in the WS, i.e. not the Landsberg erecta, background of existing arc mutants. WS contains fewer chloroplasts for a given mesophyll cell plan area than Landsberg erecta: the mean chloroplast number per mesophyll cell for WS is 83 (Table I) compared to a mean of 120 chloroplasts per cell for Landsberg erecta (Pyke and Leech, 1992). Characterization of the arc6 Mutant of A. thaliar~avar WS
Cell suspensions from the first leaves of ll,OC10putative mutant seedlings were screened microscopically by eye and several mutants with altered chloroplast number and/or chloroplast size were isolated. One of these mutants with a novel chloroplast phenotype was termed arc6. Analysis of F2 populations resulting from reciprocal backcrossing of arc6 to wild type shows that arc6 segregates as a monoger.ic nuclear recessive trait in a normal Mendelian manner (300 wild type:115 arc6; x2 = 1.7, P > 0.05). The arc6 mutant phenotype is stably inherited through at least four generations. Allelic crosses with existing arc mutants with similar phenotypes indicated that the arc6 mutant represents a new arc locus.
type/
Mesophyll cell plan area (ym2)
Figure 1. The relationship between chloroplast number per mesophyll cell and mesophyll cell plan area for wild-type var WS (O)and the arc 6 mutant (O) of A. thaliana. Each data point rep'resents the measurement from one cell whose area was measured and the chloroplast complement counted. Values for r z are wild type, 0.86; arc6, 0.3.
1171
The arc6 Chloroplast Division Mutant of Arabidopsis
Table I. Mean mesophyll cell size, mean chloroplast number, and chloroplast size for populations of mesophyll cells from fully expanded first leaves of wild-type and mutant arc6 A. thaliana varWS Mesophyll cell plan area is a mean of 150 cells and mean chloroplast plan area is a mean of at least 100 chloroplasts from 50 different mesophyll cells. Mean chloroplast number for each line was determined from a regression of chloroplast number per cell on mesophyll cell plan area using the value for mean mesophyll plan area. The mean cell plan area of cells in which chloroplasts were measured was not significantly different from the mean cell plan area of 150 cells and was a representative sample.
Wild type 403020100- •
I
1
0
100 200 300 400 500 600
b
Ol
s
arc6
ob
Mesophyll Cells Arabidopsis Genotype
Mean cell plan area
Mean chloroplast number
Mean chloroplast plan area
Chloroplasts per unit area 3
n
