| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Genetics |
Department of Plant Sciences, University of California Davis, One Shields Ave., Davis, California 95616-8780 USA
Received for publication October 13, 2006. Accepted for publication May 9, 2007.
ABSTRACT
In the processes of plant domestication and variety development, some traits are under direct selection, while others may be introduced by indirect selection or linkage. In the cultivated tomato (Lycopersicon esculentum = Solanum lycopersicum), and all other Solanaceae examined, chloroplasts are normally absent from subepidermal and mesophyll cells surrounding the leaf veins, and thus, veins appear clear upon subillumination. The tomato mutant obscuravenosa (obv), in contrast, contains chloroplasts in cells around the vein, and thus, veins appear as dark as the surrounding leaf tissue. Among tomato cultivars, the obv allele is common in processing varieties bred for mechanical harvest, but is otherwise rare. We traced the source of obv in processing tomatoes to the cultivar Earliana, released in the 1920s. The obv locus was mapped to chromosome 5, bin 5G, using introgression lines containing single chromosome segments from the wild species L. pennellii. This region also contains a quantitative trait locus (QTL) for plant height, pht5.4, which cosegregated with SP5G, a paralog of self-pruning (sp), the gene that controls the switch between determinate and indeterminate growth in tomato. The pht5.4 QTL was partially dominant and associated with a reduced percentage of red fruit at harvest. Our data suggest that the prevalence of obv in nearly all processing varieties may have resulted from its tight linkage to a QTL conferring a more compact, and horticulturally desirable, plant habit.
Key Words: chloroplasts • determinate habit • leaf veins • Lycopersicon esculentum • Lycopersicon pennellii • Solanum lycopersicum • Solanum pennellii
Plant architecture on the macro and micro levels fundamentally influences the way plants grow and respond to the environment. Understanding the molecular basis of these changes in plant development is critical to our ability to make continued progress in plant breeding. For example, improved yield in crop plants is directly affected by the genes and alleles controlling initiation and timing of flowering (Doebley, 2006
). Plant architecture can influence yield by altering the relative biomass in vegetative vs. reproductive organs, the placement of fruit on the plant, and their ease of harvest. Examples include changes in plant habit (determinate vs. indeterminate growth) and the presence or absence of abscission zones in the inflorescence that facilitate separation of fruit from the plant (Pneuli et al., 1998; Szymkowiak and Irish, 2006
). Yield-related traits may also be influenced by leaf development, such as changes in leaf area, photosynthetic efficiency, or stomatal control. Developing leaf veins are of interest because they are morphogenic centers involved in the early differentiation of many leaf cell types (Clay and Nelson, 2002
). Naturally occurring mutations in leaf venation are rare and provide an opportunity to better understand the cell differentiation that defines the architecture of plant leaves (Clay and Nelson, 2002
). While significant research has been conducted on developmental control in monopodial plants such as wheat, maize, arabidopsis, and pea, tomato provides an ideal system for understanding developmental control in sympodial plants.
In leaves of the cultivated tomato, Lycopersicon esculentum Mill. (= Solanum lycopersicum L.), the veins normally transmit more light than the interveinal areas, and thus appear "clear." In contrast, the mutant obscuravenosa (obv) reduces the transmission of light through leaf veins, causing them to appear as dark as or darker than the interveinal areas. This phenotypic difference was first noted by C. M. Rick in segregating backcross derivatives of L. esculentum (obv) x Solanum lycopersicoides (obv+) (Rick et al., 1996
). Further investigations by Rick et al. (1996)
revealed that the clear-veined trait is not unique to the wild species but is also common in cultivated tomatoes. That none of us (C. M. Rick and associates) noted this trait earlier is surprising, given our many years of observing tomato leaves under various conditions and the thousands of varieties and wild populations grown at the University of California at Davis (UC Davis). The prevalence of tomatoes bred for processing (i.e., paste-making) in the program at UC Davis may have been a factor in overlooking such an obvious phenotype for so long. As we report herein, the obv mutant is widespread in processing tomatoes, which also contain the self-pruning gene (sp) conferring a determinate growth habit. This correlation led us to investigate the possibility of a genetic association between obv and sp, or other members of the sp gene family.
Modern tomato breeding began about a century ago, and many traits present in today's cultivars are the product of early selections. Examples are traits such as fruit shape and size, smooth skin, and high yield that are standard features of tomatoes grown today. While large-fruited landraces were developed in the Americas before the arrival of Europeans, domestication and cultivar development has continued since this time. Many of these "primitive" landraces produce highly fasciated fruits on large, indeterminate vines similar to tomatoes first introduced into Europe shortly after Columbus's journeys to the Americas (Smith, 1998
).
Mechanically harvested processing tomatoes were first developed by Jack Hanna and others at UC Davis (Hanna et al., 1964
) in conjunction with parallel efforts to engineer a suitable machine harvester. The rapid adoption of these new cultivars coincided with the end of the guest worker "Bracero" program, which had provided low-wage migrant labor to hand pick tomatoes (Friedland and Barton, 1975
; Krebs, 2005
). Most processing tomatoes, including those bred by Hanna, contain specific alleles at a number of loci that make them better suited for high yield under destructive mechanical harvest. One of the most important of these loci is the gene sp, a recessive mutation that confers a determinate pattern of shoot growth, resulting in a more concentrated fruit set and ripening than is normally obtained on indeterminate vines (Stevens and Rick, 1986
; Yeager, 1927
).
Tomatoes have a sympodial pattern of growth in which shoots are composed of alternating vegetative and reproductive organs. After 6–20 leaf nodes, flowering is initiated from the main shoot meristem, and formation of the inflorescence (normally a terminal structure in tomato) ends growth of that sympodium, with vegetative growth continuing from the uppermost axillary bud. Indeterminate tomatoes (+/+ or +/sp) thereby produce repeating sympodia, each usually consisting of three leaves and an inflorescence, a pattern of growth that results in tall plants, with fruit set occurring throughout development once flowering begins. Determinate plants (sp/sp) form a decreasing number of leaves between inflorescences, which ends when two successive inflorescences are produced without an intervening leaf (hence no axillary meristem) (Pnueli et al., 1998
, 2001
). This pattern of growth produces compact plants with fruit set concentrated toward the later stages of development, yielding a greater proportion of ripe fruit at harvest.
The self-pruning locus in tomato has been cloned and is an ortholog of Centroadialis (CEN) and Terminal Flower 1 (TFL1), genes which regulate flowering in Antirrhinum and Arabidopsis, respectively (Pnueli et al., 1998
). The "CETS" (CEN, TFL1, and SP) family of genes are conserved among many species and are central to flowering regulation (Pnueli et al., 2001
). In pea there are two CETS orthologs: DETERMINATE and LATE FLOWERING. The former gene controls whether inflorescences are determinate or indeterminate, the latter gene regulates the time of flowering (Foucher et al., 2003
). In tomato SP is represented by a gene family with six members (SP, SP2I, SP3D, SP5G, SP6A, and SP9D) distributed on five chromosomes (Carmel-Goren et al., 2003
). SP is located on chromosome 6, bin E ("bin" is explained in the Materials and Methods), and is the gene underlying the sp mutant, which controls the switch between determinate and indeterminate habit. Little is known about the phenotypic effect of the other sp paralogs in tomato. Prior to this study, phenotypes for SP9D and SP5G had been described in L. pennellii (= S. pennellii) introgression lines, which contain single chromosome segments from the wild species in a constant genetic background of "M82", a processing tomato with determinate habit. The chromosome 9 introgression line carrying the wild species allele of SP9D has a semi-determinate phenotype in which growth terminates after an average of nine inflorescences, compared to five for M82 (Carmel-Goren et al., 2003
). Introgression line 5-4, which carries the L. pennellii allele of SP5G, was reported to have several traits that are consistent with a delayed expression of determinacy, including significantly greater plant weight and a higher percentage of green fruit at harvest (Eshed and Zamir, 1995
).
The goal of the present study was to determine the inheritance and map the location of the obv gene in tomato. In addition, we examined the origin of obv during tomato breeding and variety development to understand how this previously undetected mutant allele became nearly fixed in processing tomato cultivars despite its relative rarity in other types of tomatoes. Our results suggest a possible scenario whereby selection at the SP5G locus resulted in fixation of obv.
MATERIALS AND METHODS
Plant material
To determine the origin and genealogy of obscure veins in VF36 and related tomato cultivars, several collections of varieties were screened for their status at the obv locus by examining mature leaves grown under field conditions at Davis and several other locations in California (Table 1). Tomato accessions were obtained from the C. M. Rick Tomato Genetics Resource Center (TGRC) at UC Davis (LA numbered accessions). Other accessions (identified by PI (Plant Introduction) numbers) were provided by the USDA–ARS Plant Genetic Resources Unit, at Geneva, New York, USA. Cultivars without listed accession numbers were observed in plantings of other University of California investigators or of private firms and gardeners. Seeds of the L. pennellii introgression lines (ILs) were obtained through the TGRC. These provide a convenient resource for locating obv because L. pennellii is clear and M82 is obscure-veined. Seed of BCF2 (IL 5-4 x M82) was kindly provided by Dani Zamir at the Hebrew University of Jerusalem, Israel.
|
). These lines, together with 25 sublines, subdivide the tomato genome into 107 mapping "bins", delimited by the overlap or non-overlap of adjacent segments (Pan et al., 2000
). Bin locations are referred to by chromosome number, followed by a letter identifying the bin (e.g., bin 5G = Chr. 5, bin G).
RFLP and CAPS analysis of IL 5-4 sublines
IL 5-4, the only L. pennellii introgression line found to have clear veins (see Results), was backcrossed to M82. DNA was isolated from plants as previously described (Fulton et al., 1995
). Plants were genotyped for molecular markers (Table 2), and those with recombinations between TG351 and TG413 in the BCF2 population were selfed and selected as sublines. Sub-ILs were further genotyped with RFLP and CAPS markers T1541, C2_At4g12590, C2_At3g55360, and SP5G, as well as the phenotypic marker obv. The self-pruning (sp) paralog SP5G was previously mapped to bin 5G within the IL 5-4 introgression (Carmel-Goren et al., 2003
). We developed an RFLP probe by amplifying SP5G and genotyped the sublines and parent lines for this marker to ascertain the relationship between differing plant habits and association with the obv phenotype. RFLP probes were labeled using the random hexamer primers method as described previously (Chetelat and Meglic, 2000
) and hybridized to the blots for 18 h at 65°C. Blots were washed at 65°C to a final stringency of 0.5 x SSC, and exposed to x-ray film (Kodak–BioMax MS and Fuji-Super RX) at –80°C to reveal polymorphisms. SP5G was mapped by Southern analysis using a probe made from primers designed to amplify bp 215-1733 (NCBI accession AY186736) of the tomato self-pruning homolog SP5G. M82 DNA was amplified using Expand Long Template PCR (Roche, Indianapolis, Indiana, USA) according to the manufacturer's instructions. The resulting probe was sequenced to confirm its identity as SP5G. Other RFLP markers were converted to PCR-detectable CAPS using primers designed from published sequences (http://www.sgn.cornell.edu; annealing temperatures were those recommended by the SOL Genomics Network (http://www.sgn.cornell.edu) or a standard of 55°C; PCR reactions were carried out for 35 cycles.
|
M82 x IL 5-4 
, and in 2005 was the reciprocal cross (IL 5-4 x M82). Because of limited seed availability, only four plants per plot of IL 5-4 x M82 could be grown in 2005. Water and fertilizer were applied by drip irrigation. Standard cultural practices, weed, pest, and disease control were applied throughout the season.
Plant phenotypes
The phenotypes of obscure veins (obv) vs. clear veins (obv+) were scored visually by subillumination from the lower side of the leaf over a light box or outdoors by holding a leaflet over a reflective surface (Fig. 1).
|
Statistical analysis
Statistical analyses were performed with STATISTICA software (StatSoft Inc., Tulsa, Oklahoma, USA). Homogeneous groupings for plant heights were determined using ANOVA and the Tukey honest significant difference (HSD) test. We tested the significance of allelic effects of SP5G on plant height and the number of nodes per sympodium using ANOVA, and the Dunnet test (Dunnet, 1955
) for comparisons to the control genotype M82.
Microscopy
To determine the anatomical basis for the obv phenotype, we examined fresh leaf sections using light microscopy. Mid-regions of distal leaflets from freshly collected, fully expanded field-grown leaves were sectioned unfixed at a thickness of approximately 100 microns using a Vibratome (Vibratome Co., St. Louis, Missouri, USA). Unstained sections were mounted in water and examined and photographed using an Olympus BH-2 compound microscope and a MicroFire digital camera with the program Picture Frame v2.2 (Olympus America, Center Valley, Pennsylvania, USA). Distal leaflets from 3–5 leaves/plant, from 3–6 plants of each of the following genotypes, were sectioned: VF36, Moneymaker, L. pennellii, IL 5-3, 5-4, and 5-5.
RESULTS
Phenotype of obv and wild type
Examination of fresh leaflet sections under low magnification revealed that the vascular bundles of obv plants are surrounded by subepidermal and mesophyll cells containing chloroplasts, particularly on the lower side of the leaflets (Fig. 2). Chloroplasts are notably absent from the subepidermal regions above and below the leaflet veins of clear-veined plants. The micrographs clearly show that the pathway of light through veinal areas of wild type leaflets is unobstructed by chloroplasts, in contrast to leaflets of obv. In addition, the veins of wild type are recessed relative to the leaf surface, which produces an overall rugose texture; obscure-veined leaves are smoother and flatter.
|
In a survey of tomato varieties (Table 1, Fig. 3), we found that the obscure-vein trait is common in processing cultivars bred for California, particularly those released by G. C. Hanna and M. A. Stevens at UC Davis (Hanna et al., 1964
; Stevens et al., 1976
; Stevens and Rick, 1986
), including VF36 in which our initial observation was made, as well as M82, a variety related to UC82. The obscure-veined cultivars can be classified into three groups based upon their likely source of obv. Most are found in the two groups that either can be traced directly to Earliana or were bred for California and likely contain the Earliana allele (Fig. 3). However, for several obscure-veined cultivars, the available pedigree information was insufficient to infer the source of obv. One obscure-veined accession, LA0146, is a primitive cultivar collected by J. H. MacGillivray of UC Davis in 1949 at a market in Mexico City and thus can almost certainly be excluded as having been derived from Earliana.
|
|
|
|
|
|
Fine mapping
In a BCF2 population of M82 x IL 5-4, we isolated five plants with recombination events between the flanking markers TG351 and TG413, out of 272 plants examined. This observed recombination frequency of 2.5% represents a nearly four-fold reduction in the expected recombination rate for this region compared to the reference map of tomato, which is based on the F2L. esculentum x L. pennellii (http://www.sgn.cornell.edu/). We genotyped homozygous sublines and the three overlapping ILs (5–3, 5-4, and 5-5) with markers C2_At4g12590, C2_At3g55360, T1541, and SP5G. The resulting map (Fig. 4) refines the location of obv to a region of approximately 1.5 cM between SP5G and C2_At4g12590.
We obtained five additional IL 5-4 sublines produced independently by C. Tauer and associates at Oklahoma State University, Stillwater, Oklahoma, USA. Of these sublines, (identified as recombinants nos. 5, 6, 7, 18, and 20) only no. 5 was clear veined. The recombinant no. 5 introgression extends from TG69 to obv. The phenotype and genotype of these recombinants is consistent with our location of obv.
One recombinant (IL 5-4-1) had a crossover between obv and SP5G (Fig. 4). SP5G was thereby mapped to the region between TG351, which delineates the end of IL 5-3 and the obv locus. We estimate the approximate distance between SP5G and obv to be 1.5 cM on the tomato map by multiplying the observed genetic distance (0.37 cM) times the reduction in recombination (4x) observed in this BCF2 population.
Phenotypes of the chromosome 5 introgression lines and sublines
Mature plants of the chromosome 5 introgression line IL 5-4 were significantly taller in both years than the M82 control (Fig. 5). We refer herein to this plant height QTL as pht5.4. The sublines IL 5-4-1, 5-4-2, and 5-4-3 were significantly taller than M82, while IL 5-4-4 and IL 5-5 were not significantly different from M82 (Fig. 5). These ILs and sublines were grouped according to their genotype at SP5G and tested for significant differences. In 2004 the lines homozygous (p/p) for the L. pennellii allele of SP5G were 32% taller than lines homozygous for the M82 allele (+/+, P < 0.001). The heterozygote (+/p) was 17% taller (P = < 0.001). In 2005 the SP5G homozygotes (p/p) were 56% taller than +/+ lines (P < 0.001), and the heterozygote was 17% taller (P < 0.05).
|
All plants displayed the determinate habit typical of M82, yet the initiation of flowering appeared delayed (i.e., there were more leaves before the first inflorescence) in plants with the L. pennellii allele of SP5G (Fig. 6). We tested this allelic effect by pooling the data for plants based on their genotype at SP5G. In 2004 the first inflorescence in plants homozygous for the L. pennellii allele (p/p) occurred at a mean value of 11.8 leaf nodes compared to 8.0 nodes for lines homozygous for the M82 allele. The first inflorescence in heterozygous plants occurred at 10.9 nodes and was not significantly different (at P < 0.05) from the p/p homozygotes. The number of leaf nodes between inflorescences 2 through 8 was not significantly different between genotypes in 2004. In 2005 the first inflorescence in p/p genotypes occurred at a mean value of 12.0 leaf nodes, compared to 9.2 nodes for +/+ lines. However, unlike the previous year, in 2005 the number of nodes per sympodium was significantly greater for the p/p homozygotes for every inflorescence up to the eighth (Fig. 6, panels A and B, P < 0.05).
|
DISCUSSION
Anatomy and morphology of obv
The phenotype of the obv mutant appears to have a relatively simple anatomical basis: the presence of chloroplasts in the subepidermal and mesophyll cell layers around leaf veins. In contrast, leaves of the wild type lack chloroplasts in these cells. Interestingly, this chloroplast-free zone is confined to the cells above and below the vascular bundles. The boundary between cells with and without chloroplasts is relatively abrupt, suggesting that the cells surrounding the vein have a specified identity and/or fate that differs from subepidermal and mesophyll cells in the leaflet lamina. One possible explanation for the obv phenotype, then, is that subepidermal cells in the veinal region have lost some feature of their normal identity, at least with respect to chloroplast development. It may be significant that the obv leaf is flatter and smoother than the wild type, suggesting that certain cell types surrounding the vascular tissues do not develop and/or expand in the normal fashion.
Origin, distribution, and genealogy of obv
The presence of obv in a primitive Mexican landrace suggests that the obscure vein trait may have been present in tomatoes that were originally brought from Mexico to Europe and then to the Americas. Under this hypothesis, all obv cultivars may ultimately be derivations from the same source. The obv phenotype appears to be rare; for example, it has not been reported in any of the large-scale, tomato mutagenesis projects. However, this mutant phenotype could easily be missed. After all, the trait went unnoticed, despite its presence in widely grown cultivars, until Rick first observed it in the 1990s (Rick et al., 1996
), after 50+ yr of working with tomato varieties and mutants of many types.
An ancient origin of obv would be consistent with our observation that nearly all obv cultivars trace back to one variety (Earliana) and possibly represent a single mutation event. This hypothesis would, however, conflict with our observation that the variety Stone is clear veined. Earliana, which was the original source for obv in modern processing types, is reported to have been a single plant selection in a field of tomatoes purchased as ‘Stone' by George Sparks of Salem, New Jersey, USA (Morrison, 1938
). If a mutation in Stone was the source of obscure veins in Earliana, then a single, ancient origin of obv would be excluded. It is not likely that Earliana could have been the progenitor of the primitive Mexican land race with obscure veins. However, the introduction of modern cultivars into Mexico cannot be ruled out as the source of obv in LA0146. Earliana was released 20 yr prior to the collection of LA0146, and it is not uncommon to find modern cultivars from North America replacing indigenous landraces in Central and South America. The morphology of fruits and flowers of LA0146, however, are similar to those of other landraces from this region and distinct from Earliana. The flowers and fruit of LA0146 are highly fasciated and often deformed with strongly exerted stigmas, all of which are typical of many Latin American landraces. The significance of obv in LA0146 is difficult to assess, but the existence of this accession raises the possibility that obv might have been present in the area of domestication during pre-Columbian times. The mutation in Earliana could conceivably trace to one of these landraces, or it could represent a novel, spontaneous allele.
It is conceivable that the Stone we observed in our study is not the same as that purchased by Sparks many years ago, particularly given the propensity of early seed companies to use the same name for different varieties. Thus, it is virtually impossible to know with certainty the obv genotype of the original Stone from which Sparks selected Earliana. Furthermore, the plant selected by Sparks and named Earliana could have been the result of an outcross or seed mixture. Ultimately, cloning of the obv gene may provide the means to resolve questions about its origin.
The selection of Earliana by Sparks was part of a rivalry among growers to produce the earliest tomatoes (i.e., those requiring the fewest days until harvest). It is tempting to speculate that the selection for earliness may have resulted in selection for an allele of SP5G in Earliana. If the Earliana allele of SP5G positively influenced plant habit (i.e., made a more compact plant) and/or earliness, then the obscure vein trait could have been swept into some cultivars by selection at the tightly linked SP5G locus. Indeed, many of the obscure-veined cultivars (Fig. 3) are from geographic regions where earliness was important. For instance, Fargo, Bison, Red River, and Farthest North—in addition to Earliana—were all tomatoes from North Dakota, and Chatham came from the tomato breeding program in New Hampshire Yeager (1933)
. Both of these tomato breeding programs sought tomatoes that could grow in the short growing seasons of their respective northern temperate climates.
Plant height, sympodial index, and SP5G
Our results demonstrate that IL 5-4 contains a QTL that affects plant height and sympodial index, consistent with earlier reports (Eshed and Zamir, 1994
) of an increased plant biomass (weight) QTL in this region. In our trials, increased biomass was associated with increased plant height and branching in all of our recombinant sublines homozygous for the L. pennellii allele of SP5G. The increase in plant height of p/p genotypes appears to be due to both a delay in flowering and a greater number of leaf nodes between successive inflorescences.
This delay in flowering associated with the L. pennellii allele of SP5G is consistent with the role of sp in controlling the switch from vegetative to reproductive growth. In our 2004 trial the effect seemed to be primarily a delay in the initiation of the first inflorescence. However, in 2005 this delay was accompanied by a greater number of nodes between each inflorescence in plants homozygous for the L. pennellii SP5G allele. This difference is illustrated in Fig. 6, panels A and B, where in 2004 the allelic effect increased total nodes in the plants with the L. pennellii allele of SP5G, but the shape of the curve remained the same; plants were taller because of the delay in initiation. In contrast, the L. pennellii allele in 2005 resulted in continued shoot growth well beyond the point at which the plants with the L. esculentum allele had reached maximum height; hence, plants were taller because of the initiation delay and the increase in nodes per sympodium. The difference is likely due to year to year environmental variation.
Little is known about the function of SP paralogs during growth and development of tomato. However, breeders have long recognized that plant habit in tomato is controlled by loci other than sp (Elkind et al., 1991
; G. C. Hanna, unpublished manuscript). Carmel-Goren (2003) reported that the sp paralog SP9D was associated with a "semi-determinate" (i.e., less compact) habit. It is important to recognize that all the L. pennellii introgression lines are determinate (sp), with the exception of IL6-2 and 6-3, which carry the sp+ allele from L. pennellii. Thus, the effects of QTL that might influence the expression of determinate habit are more readily detected in these ILs than if they were in an indeterminate background.
Processing tomatoes and obscuravenosa
A striking aspect of the obv mutation is its persistence in processing cultivars, particularly those bred for California. We traced obv in these cultivars back to Pearson and Earliana, which were in the pedigrees of many varieties released by G. C. Hanna at UC Davis in the 1950s and 1960s. The obscure-vein trait is otherwise rare in other market classes (e.g., fresh market, greenhouse). Surprisingly, the obv mutant was not eliminated following crosses to other stocks used in the development of modern processing cultivars—as sources of disease resistances or other traits (Stevens and Rick, 1986
)—many of which were clear-veined.
One can infer from these observations that either obv has a desirable pleiotropic affect, or it is tightly linked in coupling to a locus that has been under consistent positive selection in breeding of processing tomatoes. Here we present evidence that in a very small genomic region on chromosome 5, obv is linked to a locus that can dramatically impact plant height and red fruit yield, at least in the L. pennellii introgression lines. Such traits would clearly be beneficial to performance of processing tomatoes. When crosses are made to clear-veined varieties or wild species, heavy selection to return to the desired horticultural types is applied and resulting varieties are almost exclusively obscure-veined. Compact plant size is under strong selection in California processing tomatoes because plants which set fruit in the furrows (i.e., further out on the branches), as opposed to on the bed tops, are undesirable. In fact, breeding of the first processing tomatoes designed for machine harvest required significant effort to achieve small plants with greater determinacy than that provided by sp alone (G. C. Hanna, unpublished manuscript).
The most recent, modern commercial hybrids do not appear to be as heavily biased toward obscure veins as are the processing cultivars developed in the 1950s through the 1970s. Several facts could explain this observation. Obscure veins are recessive, so both inbred parents would need to be obv for their F1 hybrid to show the phenotype. In our model system, the plants heterozygous for the L. pennellii allele of SP5G were shorter than homozygous plants, indicating that SP5G may act in a partially dominant manner. This would require only one parent of the F1 hybrids to carry the linkage block with obv and SP5G, resulting in a compact plant with clear veins. It is also possible that in some inbred lines linkage between obv and SP5G has been broken. Because each inbred line may be used to create more than one hybrid variety, as well as being used to develop superior inbreds, a single crossover event between obv and SP5G could show up in several varieties.
While we present evidence that fixation of obv could have resulted from indirect selection for SP5G, our data are based on study of the L. pennellii introgression lines. We have not yet demonstrated that variation for this gene, or for the pht5.4 QTL, exists within tomato varieties. An alternative hypothesis is that the obscure-vein trait confers a desirable characteristic that is under direct selection. For example, the presence of chloroplasts in the subepidermal cells surrounding the vascular bundles might increase photosynthetic capacity and ultimately yield. With their strongly determinate habit, finite number of leaves, and concentrated fruit set, processing tomatoes bred for mechanical harvest are probably more source-limited than indeterminate types. This is consistent with data indicating that determinate genotypes produce fruit with lower soluble solids than indeterminate isolines (Emery and Munger, 1970
; Rousseaux et al., 2005
). Interestingly, the leaves of many processing cultivars tend to be rolled rather than flat, which exposes the abaxial surfaces to direct sunlight. In this orientation, chloroplasts in the cells beneath the vascular bundles might contribute significantly to photosynthetic light capture. However, the prevalence of clear veins in all wild species examined to date seems to suggest that if obv confers a photosynthetic advantage, it is a relatively minor one. Or, species with indeterminate growth might optimize light capture more efficiently by growing new leaves and/or altering the plant canopy.
There might also be other loci (besides pht5.4 or obv) under selection in this region of chromosome 5. To our knowledge, this region does not contain any known major disease resistance (Zhang et al., 2002
), fruit shape, or fruit size loci currently mapped in tomato (Tanksley, 2004
) that could be candidates for this selection. A minor disease resistance locus (rx-3), one of three recessive loci contributing to resistance to Xanthomonas campestris pv. vesicatoria race 1, maps to bin 5G. However, it is unlikely that rx-3 has been under selection in processing tomatoes, because this disease is limited to tropical/subtropical areas, and resistance has not been a focus of the processing tomato breeding programs. Causse et al. (2004)
and Eshed and Zamir (1995)
both reported a QTL for higher Brix (soluble solids, mostly sugars) levels in bin 5G of the L. pennellii introgression lines. We measured Brix in the sublines used in this study and found that higher levels were associated with taller plants containing the L. pennellii allele of SP5G (data not shown). Hence, the allele for high Brix is in repulsion to the alleles for obscure veins and compact plant habit and would therefore not drive fixation at obv. Furthermore, the Brix QTL in IL 5-4 could be the result of the less determinate habit of these plants, as is the case with the sp locus on chromosome 6 noted earlier.
The molecular basis of crop domestication is a subject of both fundamental and applied interest. To date, several important "domestication" loci in tomato have been cloned and characterized, including loci responsible for fruit size (Alpert et al., 1995
; Frary et al., 2000
; Lippman and Tanksley, 2001
) and color (Ronen et al., 2000
). Additional loci confer desirable fruit or horticultural characteristics in selected varieties. Examples include the SUN locus affecting fruit shape (van der Knaap et al., 2004
), the JOINTLESS gene controlling fruit abscission (Mao et al., 2000
, 2001
), and the sp locus described previously.
Conclusions and outlook
A reduction in variation at the nucleotide level may indicate the presence of loci under natural or artificial selection. In the study of crop evolution, regions of depleted sequence variation suggest the presence of domestication loci or other genes of agricultural importance (Yamasaki et al., 2005
). However, these regions are less variable and informative within cultivated gene pools. Hence, while domestication history may be revealed, many loci currently under selection for crop improvement may remain hidden.
Breeding and selection for distinct horticultural types within cultivated germplasm has a long tradition in the development of food crops. Little is known, however, about the fixation or extent of linkage disequilibrium around loci that determine the differences between horticultural types in these crops. Presumably most major loci related to domestication are fixed in all cultivated types, while minor loci (or "modifiers"), such as the sp homologs in tomato, may vary. It is these secondary loci that may often be the unrecognized targets of phenotypic selection in breeding programs. The linkage between obv and SP5G thus provides new insights into how selection in breeding programs drives fixation at linked loci. Understanding such relationships can help researchers to identify regions lacking useful variation and suggest new opportunities for targeted introgressions.
In this study we focused on the effects of L. pennellii introgressions on chromosome 5. In these derivatives, plant height differences cosegregated with allelic differences at the SP5G locus. These data suggest SP5G is a candidate for the pht5.4 QTL. In addition, we demonstrated variation at the tightly linked obv locus among cultivated tomatoes, yet near uniformity in processing tomatoes for the mutant allele. Further work is required to determine whether sequence variation at SP5G exists within cultivated germplasm, and if so, whether it underlies the pht5.4 QTL and/or the fixation of obv.
FOOTNOTES
1 The authors acknowledge the late Prof. Ernest M. Gifford at UC Davis, who provided micrographs during the early development of this project. D. Zamir provided seed of the IL 5-4 F2, and seed of selected cultivars were provided by the USDA Plant Genetic Resources Unit, D. St. Clair, B. Farnsworth, G. Miyao (UCD), B. Sterling, and T. Beck-Bunn (Seminis) provided access to their field trials for evaluation. C. Tauer and B. Martin at Oklahoma State University generously provided seed of their sub-ILs. The authors are also grateful for the valuable assistance of R. Benitez along with the staff and students at the C.M. Rick Tomato Genetics Resource Center. They dedicate this paper to the late Prof. Charles M. Rick, who noted "seeing is not the same as observing" in reference to his belated detection of obscuravenosa after 50+ years of tomato research. 
2 Author for correspondence (e-mail: trchetelat{at}ucdavis.edu
), phone: +1–530-752-6726
3 Present address: Seminis Vegetable Seeds, 37437 State Highway 16, Woodland, CA 95695 USA
4 Present address: Campbell Research and Development, 28605 County Road 104, Davis, CA 95618 USA
LITERATURE CITED
Alpert K. B. Grandillo S. Tanksley S. D.. 1995. Fw 2.2: a major QTL controlling fruit weight is common to both red- and green-fruited tomato species. Theoretical and Applied Genetics 91: 994-1000..[Web of Science]
Carmel-Goren L. Liu Y. S. Lifschitz E. Zamir D.. 2003. The SELF-PRUNING gene family in tomato. Plant Molecular Biology 52: 1215-1222..[CrossRef][Web of Science][Medline]
Causse M. Duffe P. Gomez M. C. Buret M. Damidaux R. Zamir D. Gur A. Chevalier C. Lemaire-Chamley M. Rothan C.. 2004. A genetic map of candidate genes and QTLs involved in tomato fruit size and composition. Journal of Experimental Botany 55: 1671-1685..
Chetelat R. T. Meglic V.. 2000. Molecular mapping of chromosome segments introgressed from Solanum lycopersicoides into cultivated tomato (Lycopersicon esculentum). Theoretical and Applied Genetics 100: 232-241..[CrossRef][Web of Science]
Clay N. K. Nelson T.. 2002. VH1, a provascular cell specific receptor kinase that influences leaf cell patterns in Arabidopsis. Plant Cell 14: 2707-2722..
Doebley J.. 2006. Unfallen grains: how ancient farmers turned weeds into crops. Science 312: 1318-1319..
Dunnet C. W.. 1955. A multiple comparison procedure for comparing several treatments with a single control. Journal of the American Statistical Association 50: 1096-1121..[CrossRef][Web of Science]
Elkind Y. Gurnick A. Kedar N.. 1991. Genetics of semideterminate growth habit in tomato. HortScience 26: 1074-1075..
Emery G. C. Munger H. M.. 1970. Effects of inherited differences in growth habit on fruit size and soluble solids in tomato. Journal of the American Society for Horticultural Science 95: 410-412..
Eshed Y. Zamir D.. 1994. A genomic library of Lycopersicon pennellii in L. esculentum: a tool for fine mapping of genes. Euphytica 79: 175-179..[CrossRef][Web of Science]
Eshed Y. Zamir D.. 1995. An introgression line population of Lycopersicon pennellii in the cultivated tomato enables the identification and fine mapping of yield-associated QTL. Genetics 141: 1147-1162..[Abstract]
Foucher F. Morin J. Courtiade J. Cadioux S. Ellis N. Banfield M. J. Rameau C.. 2003. DETERMINATE and LATE FLOWERING are two TERMINAL FLOWER1/CENTRORADIALIS homologs that control two distinct phases of flowering initiation and development in pea. Plant Cell 15: 2742-2754..
Frary A. Nesbitt T. C. Grandillo S. Knaap E. Cong B. Liu J. Meller J. Elber R. Alpert K. B. Tanksley S. D.. 2000. fw 2.2: a quantitative locus key to evolution of tomato fruit size. Science 289: 85-88..
Friedland W. H. Barton A.. 1975. Destalking the wily tomato: a case study in social consequences in California agricultural research. Research Monograph No. 15. Dept. of Applied Behavioral Sciences, University of California, Davis, California, USA..
Fulton T. M. Chunwongse J. Tanksley S. D.. 1995. Microprep protocol for extraction of DNA from tomato and other herbaceous plants. Plant Molecular Biology Reporter 13: 207-209..[Web of Science]
Graham T. O.. 1959. Impact of Mendelian factors on the tomato 1929–1950. Report of the Tomato Genetics Cooperative 9: 36-37..
Hanna G. C. Gentle A. Smith P. G. Lippe L. F. Davis G. N. McCoy D. N.. 1964. Recently developed vegetables aid mechanical cultivation. California Agriculture 18: 8-10..
Krebs A. V.. 2005. La Causa: the word was made flesh. Website http://www.farmworkermovement.org/essays/essays/KREBS%20MANUSCRIPT%20LA%20CAUSA.pdf [accessed 12 July 2006]..
Lippman Z. Tanksley S. D.. 2001. Dissecting the genetic pathway to extreme fruit size in tomato using a cross between the small-fruited wild species L. pimpinellifolium and L. esculentum, var. Giant Heirloom. Genetics 158: 413-422..
Mao L. Begum D. Chuang H. W. Budiman M. A. Szymkowiak E. J. Irish E. E. Wing R. A.. 2000. JOINTLESS is a MADS-box gene controlling tomato flower abscission zone development. Nature 406: 910-913..[CrossRef][Medline]
Mao L. Begum D. Goff S. A. Wing R. A.. 2001. Sequence and analysis of the tomato JOINTLESS locus. Plant Physiology 126: 1331-1340..
Morrison G.. 1938. Tomato varieties. Michigan State College Agricultural Experiment Station, Special Bulletin 290: 1-68..
Pan Q. Liu Y. S. Budai-Hadrian O. Sela M. Carmel-Goren L. Zamir D. Fluhr R.. 2000. Comparative genetics of nucleotide binding site-leucine rich repeat resistance gene homologues in the genomes of two dicotyledons: tomato and arabidopsis. Genetics 155: 309-322..
Pnueli L. Carmel-Goren L. Hareven D. Gutfinger T. Alvarez J. Ganal M. Zamir D. Lifschitz E.. 1998. The SELF-PRUNING gene of tomato regulates vegetative to reproductive switching of sympodial meristems and is the ortholog of CEN and TFL1. Development 125: 1979-1989..[Abstract]
Pnueli L. Gutfinger T. Hareven D. Ben-Naim O. Ron N. Adir N. Lifschitz E.. 2001. Tomato SP-interacting proteins define a conserved signaling system that regulates shoot architecture and flowering. Plant Cell 13: 2687-2702..
Porter D. R. Jones H. A. Davis G. N.. 1941. Newly developed vegetable varieties for use in California. University of California College of Agriculture Experiment Station Bulletin 646: 1-26..
Rick C. M. Gifford E. M. Chetelat R. T.. 1996. Genetics, anatomy, and genealogy of obscuravenosa, a monogenic tomato leaf trait. Proceedings of the Molecular Biology of Tomato Symposium UC Berkeley, Berkeley, California, USA,1–4 August 1996, abstract O15..
Ronen G. Carmel-Goren L. Zamir D. Hirschberg J.. 2000. An alternative pathway to beta-carotene formation in plant chromoplasts discovered by map-based cloning of Beta and old gold color mutations in tomato. Proceedings of the National Academy of Sciences, USA 97: 11102-11107..
Rousseaux M. C. Jones C. M. Adams D. Chetelat R. Bennett A. Powell A.. 2005. QTL analysis of fruit antioxidants in tomato using Lycopersicon pennellii introgression lines. Theoretical and Applied Genetics 111: 1396-1408..[CrossRef][Web of Science][Medline]
Smith B.. 1998. Foreword, Livingston and the tomato. Ohio State University Press, Columbus, Ohio, USA..
Stevens M. A. Dickinson G. L. Aguirre M. S.. 1976. UC82, a high yielding processing tomato. Bulletin, University of California, Davis, Vegetable Crops Series 183, Davis, California, USA..
Stevens M. A. Rick C. M.. 1986. Genetics and breeding. In J. G. Atherton, J. Rudich [eds.] The tomato crop: a scientific basis for improvement, 35-105. Chapman and Hall, New York, New York, USA..
Szymkowiak E. J. Irish E. E.. 2006. JOINTLESS suppresses sympodial identity in inflorescence meristems of tomato. Planta 223: 646-658..[CrossRef][Web of Science][Medline]
Tanksley S. D.. 2004. The genetic, developmental, and molecular bases of fruit size and shape variation in tomato. Plant Cell 16: S181-S189..
Tanksley S. D. Ganal M. W. Prince J. P. De Vicente M. C. Bonierbale M. W. Broun P. Fulton T. M. Giovannoni J. J. Grandillo S.. 1992. High density molecular linkage maps of the tomato and potato genomes. Genetics 132: 1141-1160..[Abstract]
van der Knaap E. Sanyal A. Jackson S. A. Tanksley S. D.. 2004. High-resolution fine mapping and fluorescence in situ hybridization analysis of sun, a locus controlling tomato fruit shape, reveals a region of the tomato genome prone to DNA rearrangements. Genetics 168: 2127-2140..
Yamasaki M. Tenaillon M. I. Bi I. V. Schroeder S. G. Sanchez-Villeda H. Doebley J. F. Gaut B. S. McMullen M. D.. 2005. A large-scale screen for artificial selection in maize identifies candidate agronomic loci for domestication and crop improvement. Plant Cell 17: 2859-2872..
Yeager A. F.. 1927. Determinate growth in the tomato. Journal of Heredity 28: 263-265..
Yeager A. F.. 1933. Tomato breeding. North Dakota Agricultural Experiment Station Bulletin 1-20..
Zhang L. P. Khan A. Nino-Liu D. Foolad M. R.. 2002. A molecular linkage map of tomato displaying chromosomal locations of resistance gene analogs based on a Lycopersicon esculentum x Lycopersicon hirsutum cross. Genome 45: 133-146..[Medline]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) represent the mean values for lines homozygous for the M82 allele and open boxes (
) the L. pennellii allele of SP5G. Gray boxes represent heterozygotes at SP5G. Leaf pictographs symbolize the obscuravenosa (obv) phenotype of each line.