This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 133/1/145    most recent pp.103.022632v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Downie, A. B. Articles by Snyder, J. C. Search for Related Content PubMed PubMed Citation Articles by Downie, A. B. Articles by Snyder, J. C. Agricola Articles by Downie, A. B. Articles by Snyder, J. C.

DEVELOPMENT AND HORMONE ACTION

Communication between the Maternal Testa and the Embryo and/or Endosperm Affect Testa Attributes in Tomato^1,[w]

Department of Horticulture (A.B.D., D.Z., L.M.A.D., R.R.T., J.A.P., J.W.B., J.C.S.) and Center of Membrane Sciences, Department of Chemistry (J.L.D., D.A.B.), University of Kentucky, Lexington, Kentucky 40546; and Department of Plant Sciences, Weizmann Institute of Science, Rehovot 76100, Israel (A.A.L.)

	ABSTRACT

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 
Two tomato (Lycopersicon esculentum) mutants with dark testae displaying poor germination rate and percentage on both water and 100 µM gibberellin_{4 + 7} were recovered. The mutants were allelic (black seed1-1; bks1-1 and bks1-2), inherited in Mendelian fashion as a recessive gene residing on chromosome 11. They are not allelic to bs (brown seed) -1, -2, or -4, which impair seed germination and possess dark testae. The bks/bs mutants accumulated dark pigment in the cell layers of the testa above the endothelium, which itself accumulated proanthocyanidins similar to wild type. The poor germination performance of bks mutant seeds was because of impediment of the mutant testae to radicle egress. Imbibition on gibberellin_{4 + 7} did not ameliorate germination percentage or rate. The toughening of the bks testa and associated poor germination were partially overcome when seeds were not dried before germination or were dried under N₂. The seeds of the bks mutant have elevated activity of at least one enzyme responsible for the detoxification of reactive oxygen species. The bks mutant is epistatic to 12 anthocyaninless mutants of tomato. Bio- and physicochemical analysis of the bks testa determined that it accumulated a melanic substance. Inheritance of bks/bs mutations contrasts with that of the anthocyaninless mutants, which are inherited according to the genotype of the maternally derived testa. This suggests that the testa manufactures components before its demise that can maximize testa strength, whereas the endosperm/embryo produces factors that are conveyed to the testa, mitigating this process.

Two other mutants of Arabidopsis in dag1 and dag2 (dna binding with one finger affecting germination1 and 2) either abolish seed dormancy and inhibition of radicle protrusion in total darkness (dag1) or have the opposite effect (dag2). The effect of these mutations is maternal such that the aberrant zinc finger genes, expressed in the vascular tissue of the flower and funiculus, no longer encode functional products that can either impose or alleviate dormancy on the seed (Papi et al., 2000, 2002; Gualberti et al., 2002).

Although deficiencies in phenylpropanoid metabolism can lead to seeds of lighter color; hyperexpression of the regulatory genes encoded by Arabidopsis TT1, TT2, or TT8; overexpression of maize (Zea mays) chal-cone synthase (CHS), isomerase, or dihydroflavonol 4-reductase (DFR; C2, CHI1, and A1) in Arabidopsis; and up-regulation of the endogenous Arabidopsis phenylpropanoid pathway in activation-tagged pap1-D (production of anthocyanin pigment 1-Dominant) mutants, paradoxically, have not resulted in greater than usual testa color (Borevitz et al., 2000; Dong et al., 2001; Nesi et al., 2001; Sagasser et al., 2002). Hence, in Arabidopsis, there are no instances of mutations imparting greater than usual testa pigmentation.

The testae of tomato seeds, like those of Arabidopsis seeds, have been found to regulate the rate of radicle protrusion (Atanassova et al., 1997a), particularly under suboptimal conditions (Atanassova et al., 1997b, 2001). Abscisic acid-deficient mutants of tomato exhibit enhanced germination capabilities under stress (Downie et al., 1999), an attribute correlated with a reduced number of cell layers in the testa (Hilhorst and Downie, 1996). However, unlike Arabidopsis, mutants with darker than usual appearance have been isolated in tomato on five separate occasions. The bs (brown seed) mutants complete germination poorly and occasionally abort the terminal bud (Pilouze, 1974). In fact, bs3 (Yordanov and Stamova, 1971) has been lost due to its poor completion of germination (L. Stamova, personal communication).

In addition to the bs1 and bs4 mutations' negative impacts on seed germination percentage and rate, the BS1 (possibly BS4) genes are on chromosome 1 in locations harboring quantitative trait loci positively influencing salt-tolerant seed germination (Foolad et al., 1998). Both bs1 and bs2 are important morphological markers mapped on chromosomes 1 and 7, respectively (Tanksley et al., 1992; Balint-Kurti et al., 1995), whereas Soressi (1972) tentatively assigned bs4 to chromosome 1. There have been no reported pleiotropic effects of the bs mutations except for bs1, which increases fruit pH and delays flowering time and turning stage (Martiniello et al., 1985). Additionally, the ls (lateral suppressor) mutation, a disruption in a gene encoding a VHIID protein (Schumacher et al., 1999), results in poor seed germination and dark color (Taylor, 1979).

Using a screen for GA-resistant germination, we have isolated two allelic mutants of dark testa color that complete germination slowly and to a low percentage on 100 µM GA_{4 + 7} from a fast neutron-bombarded population of Micro-Tom (MT) tomatoes. The initial characterization of these mutants is reported here.

	RESULTS

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

Mutant Screen, Complementation Analysis, and General Characterization

Between 50 and 100 seeds from each of 616 individual M₃ families derived from a fast neutron-bombarded population of MT tomatoes (see "Materials and Methods") were examined for families that contained at least some seeds that were unable to complete germination on GA_{4 + 7} without surgical removal of the testa and endosperm. Two mutants were recovered that produced seeds that completed germination to a low percentage on water or GA_{4 + 7} and were darker than usual. Complementation tests determined that the mutants were allelic. Analysis of segregating F₂ populations from reciprocal crosses to wild-type (WT) MT indicated that the poor germination phenotype and the black seed color were always linked, probably due to the same recessive nuclear mutation, bks (black seed), inherited in a Mendelian fashion (Table I). When either allele of bks was reciprocally crossed with WT MT, the F₁ seeds were WT in appearance (Fig. 1, A-D). F₂ seeds from selfed F₁ plants segregated three WT to one bks seed (Table I). A literature search revealed five mutants with a similar phenotype. The bs mutants 1 to 4 (Soressi, 1967; Philouze, 1970; Yordanov and Stamova, 1971; Monti, 1972) and the ls (lateral suppressor) mutant (Tucker, 1976; Taylor, 1979) all produce seeds with a darker than usual testa appearance. In addition, bs mutant seeds complete germination poorly (Soressi, 1967; Philouze, 1970; Monti, 1972) and are inherited in a recessive, Mendelian fashion. However, the reduction of tomato seed germination by ls was shown to be an effect expressed late during seed development and was dependent on the genetic composition of the testa (Taylor, 1979). Complementation tests between bks and bs1, 2, and 4 and ls revealed that the new mutant was not allelic with any of the previously identified mutants (data not shown). Due to the dissimilarity of the mode of inheritance of the ls mutation from that of bs and bks, it was not included in this study.

View larger version (84K): [in this window] [in a new window]   Figure 1. The bks and bs mutants result in a darker than usual testa. A to D, F1 seeds from reciprocal crosses of bks1-2 and WT MT are WT in appearance. E, Pigment accumulation in bks1-2 during development. Width of the box containing the insets = 5 mm. F, Hand sections of WT and mutant seeds with dark testae. Embryo and endosperm color is unaffected relative to WT. Pieralbo (PA) is not available (NA); AC, Ailsa Craig; MM, Moneymaker; IG, immature green; MG, mature green; En, endosperm; R, radicle; C, cotyledons; T, testa. Bar in all the micrographs = 1 cm unless stated otherwise.

Pigment commenced accumulating in the bs and bks mutants between the immature green and mature green stages of fruit ripening (Fig. 1E; data not shown). Most seeds were heavily pigmented by the red ripe stage of fruit development, and the dark color was limited to the testa (Fig. 1F).

The seeds of the bks mutant weighed less and were smaller than those of WT MT (Table II). The numbers of seeds per fruit were not influenced, but the pH of the fruit from bks mutants was greater than that of WT MT (Table II).

Analysis of approximately 1,000 F₂ progeny of the cross of bks1-1 with the chromosome 11 markers a (anthocyaninless; green hypocotyls) or hl (hairless; trichomeless stem cells) indicated that bks was linked to hl. This result was confirmed by scoring approximately 1,000 F₂ progeny of the cross of the second allele (bks1-2) with the chromosome 11 marker line (data not shown).

Aberrations in bks and bs Testa Composition Are Visible Primarily in the Cell Layers Underlying the Epidermis

Tomato mutant seeds were sectioned to determine what tissues accumulated the dark pigment. With the exception of bs4, the preponderance of the dark material was restricted to the cell layers of the integument underlying the epidermis but distal to the endothelium, although the trichomes of the mutants are also consistently darker in appearance than their respective WTs (Fig. 2, A-N). None of the mutants were altered in proanthocyanidin accumulation in the endothelium abutting the endosperm (Figs. 2, B, D, F, H, J, L, and N). In all but the bs4 mutant, the cells containing the dark pigment were peripheral to the proanthocyanidin-containing endothelium as determined by vanillin staining. The vanillin-stained endothelial cells were immediately peripheral to the secondarily thickened cell walls of the first cell files of the endosperm (Fig. 2, A-N). The accumulation of pigment in the bs4 mutant seeds was not limited to the testa but extended into the periclinal wall of the first layer of endosperm cells abutting the testa (Fig. 2, G and I). Vanillin staining of the proanthocyanidins in the endothelium confirmed this observation, the endothelium now present between the pigmented cells rather than interior to them (Fig. 2, H and J).

View larger version (62K): [in this window] [in a new window]   Figure 2. Sections of the testa and endosperm of WT and mutant tomato seeds. All micrographs are 10-µm-thick sections of: A and B, WT Ailsa Craig; C and D, bs1; E and F, bs2; G and H, WT Moneymaker; I and J, bs4; K and L, WT MT; and M and N, bks1-1. The sections of bks1-2 seeds (not shown) were similar in appearance to those of bks1-1. The first column of micrographs is without stain, whereas the second column has been stained with 1% (w/v) vanillin in 4.5 N HCl for 10 min. Bar in the micrographs = 0.5 mm.

Germination Percentage and Rate

The completion of germination of air-dried bks mutant seeds was delayed and, regardless of whether the seeds were imbibed on water or GA_{4 + 7}, did not attain the same final percentage relative to air-dried seeds of WT MT (Fig. 3A). Consistent with the screening regime used to isolate the bks mutants, imbibition on 100 µM GA_{4 + 7} did not increase the percentage germination above that of air-dried mutant seeds imbibed on water (Fig. 3A). The delay in completion of germination of the bks mutant seeds was not due to a deficiency in the rate of imbibition (Fig. 3B).

View larger version (28K): [in this window] [in a new window]   Figure 3. The bks mutant negatively impacts the completion of germination on water or GA4 + 7 but not by impeding imbibition. A, Rate with which the bks mutant commenced radicle protrusion when imbibed on water was retarded relative to that of the WT MT controls. Neither did the bks mutant seeds attain the same final percentage germination as WT MT seeds. Imbibition on GA4 + 7 did not affect the germination rate or the final percentage germination of bks mutant seeds. B, Imbibition rate and moisture content attained by the bks mutant seeds was similar to that of WT MT seeds.

Drying Influences Testa Toughness and Germination Rate

The testa toughness of air-dried WT MT seeds 24 h after imbibition (HAI) was statistically significantly less than that of air-dried bks mutant seeds 24 HAI (Fig. 4). Air-dried WT MT seeds also commenced radicle protrusion faster and/or were more uniform in the rate with which they attained maximum germination percentage relative to mutant seeds (Fig. 4).

View larger version (34K): [in this window] [in a new window]   Figure 4. The rate and final germination percentage of tomato seeds is influenced by how they are dried. The physical toughness of the testa (mean and SE of the puncture force [Newtons] given in each legend by the appropriate treatment symbol) was influenced by if and how the seeds were dehydrated. Fresh or nitrogen-dried seeds consistently completed germination to the greatest percentage and had weak testae. Fresh, Not dried before germination on water; Air, N2, and O2: Dried under an atmosphere of ambient air, pure nitrogen, or oxygen, respectively, in desiccators over activated alumina for 4 d before germination on water. Lowercase letters after the average puncture force values denote significant differences among treatments within a genotype. Uppercase, bold letters denote significant differences among genotypes within a treatment.

Observing an equal incidence of precocious germination in overripe fruit from greenhouse-grown WT MT and bks mutants, it was hypothesized that testa coloration was not itself the cause of poor germination for this mutant but that it could lead to reduced radicle protrusion if the seeds were dried. To test this hypothesis, germination tests and puncture force analyses were conducted on freshly harvested mutant and WT seeds that had been: (a) dried in air, (b) dried in nitrogen, (c) dried in oxygen, or (d) placed directly on water without drying (fresh).

Comparing the force necessary to puncture the testa and endosperm among treatments within a genotype, seeds were toughest or statistically indistinguishable from the toughest when dried in air (Fig. 4). The seeds of the bks mutant completed germination faster and to a greater percentage when drying in air or oxygen was avoided. Air-dried seeds also had statistically significantly tougher testae, such that germination performance was negatively associated with testa strength (Fig. 4). In addition, for the two alleles of the bks mutation, the bks1-1 allele influenced percentage germination the most, and, within a dehydration treatment, this allele consistently had the toughest testa (Fig. 4). However, both bks mutant and WT MT seeds dried under oxygen had weak testae but completed germination poorly; therefore, some aspect of dehydration under an O₂ atmosphere other than testa strengthening negatively impacts seed germination (Fig. 4).

Reactive Oxygen Species (ROS)-Scavenging Enzymes Are Up-Regulated in Dark Testa Mutants

Determining that the testa of the bks mutant seeds was relatively weak when drying in air was avoided, it was hypothesized that oxygen or an ROS was interacting with some testa component, most probably a product of the shikimate pathway, to cross-link and toughen the testa. This presumption was based on precedence of shikimate compounds accumulating in testae, thereby increasing testa strength (Gillikin and Graham, 1991; Todd and Vodkin, 1993; Gijzen, 1997). Such toughening could be enzymatically mediated by either peroxidase (PRX) or polyphenol oxidase (PPO). Catalase (CAT)-insensitive PPO activity was similar among the genotypes (data not shown). However, bks mutant seeds exhibited considerably greater PRX activity relative to WT MT (Fig. 5A). Because PRX is one enzyme in a suite capable of detoxifying ROS, it was possible that CAT and/or superoxide dismutase (SOD) activities might also be up-regulated in the mutant. CAT activity was statistically significantly less in the bks mutants than in WT MT (Fig. 5B, upper), although in-gel staining for CAT activity after native PAGE did not reveal obvious differences (Fig. 5B, lower). Total SOD activity was not significantly different among the genotypes (Fig. 5C, upper). However, using SDS-PAGE, without boiling the sample in SDS-loading buffer, it was determined that a high molecular mass (>113 KD) protein or protein complex with SOD activity was present in the bks mutants but not the WT (Fig. 5C, lower).

View larger version (54K): [in this window] [in a new window]   Figure 5. ROS-scavenging enzymes are up-regulated in the bks mutant. Upper: A, PRX; B, CAT; and C, SOD activities present in WT MT and bks mutant tomato seeds. Within an enzyme assay, different uppercase letters within the bars of the histogram represent statistically significant differences at alpha = 0.05 according to Tukey's mean separation test. Lower: A, Native PAGE (12% [w/v], 20 µg of buffer-soluble protein lane-1) stained for PRX activity from WT MT and bks mutant seeds and a commercial preparation (PRX). B, Native PAGE (9% [w/v], 10 µg of buffer-soluble protein lane-1) stained for CAT activity from WT MT and bks mutant seeds and a commercial preparation. C, SDS-PAGE (12% [w/v]) of crude protein extracts (20 µg lane-1) from WT MT and mutant seeds and a commercial preparation (SOD) dissolved in SDS-loading buffer but not boiled and stained for SOD activity. Numbers to the left of the SDS-PAGE represent the molecular mass of the standards in kilodaltons.

There was no effect of greater than usual ROS-scavenging enzyme activity on the release of ROS from intact, dry, or 24-h-imbibed seeds. Intact, dry seeds of the bks mutant were similar to WT MT in the amount of ROS that was released into the assay media (Fig. 6). Mutant and WT seeds imbibed for 24 h were statistically similar with respect to ROS release (data not shown).

View larger version (21K): [in this window] [in a new window]   Figure 6. Release of ROS from dry seeds upon imbibition. There were no significant differences in ROS release among dry or 24-h-imbibed (data not shown) bks and WT MT.

The bks Mutants Are Epistatic to anthocyaninless Mutants

The anthocyaninless mutants of tomato are disrupted in anthocyanin accumulation, resulting in a green hypocotyl phenotype. This is also a characteristic of at least four of the tt mutants of Arabidopsis (Koornneef, 1990). At least some of the anthocyaninless mutants of tomato may represent lesions in genes orthologous to the Arabidopsis TT genes. In addition, some of the anthocyaninless mutants of tomato exhibit seed associated phenotypes that contrast with the dark-testa mutants. The tomato mutants ah (anthocyaninless of Hoffman), aw (without anthocyanin), and bls (baby lea syndrome) possess: (a) a lighter than WT testa color (Atanassova et al., 1997a, 1997b), (b) a pattern of inheritance dependent on the genotype of the testa (Atanassova et al., 1997a), and (c) more rapid completion of germination than WT (Atanassova et al., 1997a, 1997b). In addition, although Atanassova et al. (1997a) did not include af (anthocyanin free), al (anthocyanin loser), and ag (anthocyanin gainer) in the list of mutants that affected either testa color or germination behavior, these mutants have been determined to possess low CHS activity (af and al) or low activities of both CHS and flavone 3-hydroxylase (ag, ah, and bls; O'Neill et al., 1990). These mutants along with a (anthocyaninless), aa (anthocyanin absent), ae (entirely anthocyaninless), ag-2 (anthocyanin gainer-2), and are (anthocyanin reduced) were all tested for epistasis with the bks mutant. Double-mutant plants (seedlings with green hypocotyls from bks seeds) were produced, and F₃ seed was collected and compared with seeds from WT and bks plants. In every case the bks mutant was epistatic to the anthocyaninless mutants (Fig. 7).

View larger version (25K): [in this window] [in a new window]   Figure 7. The bks mutant is epistatic to the anthocyaninless mutants: Seeds of double mutants between bks and 12 anthocyaninless lines retain a dark testa color. In these mutants, the anthocyaninless testa phenotype could not be scored. a, anthocyaninless; aa, anthocyanin absent; ae, entirely anthocyaninless; af, anthocyanin-free; ag, anthocyanin gainer; ag-2, anthocyanin gainer 2; ah, Hoffmann's anthocyaninless; ai, incomplete anthocyanin; al, anthocyanin loser; are, anthocyanin reduced; aw, without anthocyanin; bls, baby lea syndrome. Bar in the baby lea syndrome square = 1 cm.

The bks Mutant Testa Accumulates a Phytomelanin

Various methods were used in attempts to solubilize and characterize the molecule(s) imparting the dark testa color in bks mutant seed (see Supplemental Table I). Compounds comprised of anthocyanins and lignins were eliminated as candidates. However, both bleach and peroxide were capable of removing the dark testa color of the bks mutant seed. The intransigent nature of the dark compound, particularly its stability under acid hydrolysis (Fogarty and Tobin, 1996) and its susceptibility to the two treatments mentioned above, have been reported as hallmarks of melanin (Sava et al., 2001), a class of chemically resistant phenolic polymers. A technique reported in a patent for purification of melanin (Makordei et al., 1994) was employed. WT seeds, when hydrolyzed with NaOH, produced little or no precipitate when the hydrolysates were subsequently acidified to pH 2. This contrasted with hydrolysates from the bks seed, which produced abundant precipitate upon acidification. Furthermore, the precipitate could be resolubilized in NaOH or in dimethyl sulfoxide (DMSO), consistent with the hypothesis that the black pigment was melanic in nature (Harki et al., 1997).

This conclusion is strengthened by electron paramagnetic resonance (EPR) data indicating that the bks mutant seeds generated an EPR signal at the same frequency (3,475 G) as black sunflower (Helianthus annuus) and niger [Guizotia abyssinica (L.S.) Cass.], known to possess testae comprised of melanic compounds. A white testa sunflower variety and WT MT tomato have appreciably fewer free radicals (Fig. 8; Table III).

View larger version (12K): [in this window] [in a new window]   Figure 8. Physicochemical evidence indicates the presence of un-paired electrons in the bks mutant testa. EPR detected the presence of a single unpaired electron in testae of niger (not shown) and black sunflower (A), both known to contain melanin. A white testa mutant of sunflower (B) did not produce an appreciable signal. Although the bks1-1 mutant testa (C) possessed unpaired electrons, the testa of WT MT did not (D). Black sunflower signal was acquired with one scan, whereas bks1-1, WT MT, and white sunflower were acquired with 100 scans. EPR instrumental parameters are given in "Materials and Methods."

	DISCUSSION

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

bks Mutant Seeds Complete Germination Poorly Due to a Tough Testa

Fresh bks mutant seeds (Fig. 4) approached the final percentage germination of WT MT. These results argue against the poor germination of the mutant being due to the testa altering the light quality impacting the embryo/endosperm invoking far-red-light-inhibition of seed germination. In addition, it is unlikely that the dark pigment in the mutant testa was a germination inhibitor (Walker, 1962) because it did not leach from the seeds during germination tests, and when the testa and endosperm cap opposing the radicle were removed (decapped, a necessary procedure to achieve high percentage germination of bks seeds when establishing linkage with the chromosomal marker lines), the radicle protruded readily and most embryos established (data not shown). Finally, although dark testa coloration has been associated frequently with inhibition of water uptake in seeds during imbibition (Egley et al., 1983), this is not the cause of delayed germination in the bks mutant (Fig. 3B). It was reasoned that the pigment imparting darker color to the mutant testa was able to enhance testa strength, preventing timely emergence of the radicle and decreasing the percentage of embryos capable of doing so. This surmise was proven correct when the force required to penetrate the micropylar endosperm and testa was determined 24 HAI. Air-dried mutant seeds, 24 HAI were always significantly tougher than air-dried, WT MT seeds 24 HAI (Fig. 4). Furthermore, bks seeds dried in air were significantly tougher than fresh, N₂-, or O₂-dried seeds with the same genetic lesion 24 HAI; therefore, the dark pigment must be dried in air to polymerize and increase testa strength (Fig. 4). Why drying in O₂ did not further strengthen the testa is not clear. Nonetheless, the monotegmic tomato testa (Cooper, 1931; Rick, 1946; Mazzucato et al., 1998), similar to the bitegmic Arabidopsis testa, can control if and when seeds complete germination (Atanassova et al., 1997a, 1997b; Koornneef et al., 2002). The association of a darker testa limiting the completion of germination and/or imposing enhanced dormancy on progeny, for multifarious reasons, is a repeated theme in seed biology/weed science (for summary, see Debeaujon et al., 2000).

Pleiotropic Effects of the bks Mutant

It has been reported that the bs mutations do not affect traits other than seed color and germination capacity (Philouze, 1974). However, Martiniello et al. (1985) reported that bs1 negatively impacted fruit pH, flowering time, and turning stage. Investigations of seed size, weight, the number of seeds fruit^-1 and fruit pH revealed significantly smaller seed weight and size for bks mutant seeds relative to WT MT (Table II). The poor germination rate and percentage of bks seeds relative to larger, heavier WT MT seeds is inconsistent with previously published conclusions (Whittington and Fierlinger, 1972; Leviatov et al., 1994) that, for tomato, small seed size leads to faster completion of germination. Neither could Atanassova et al. (1997a) find a consistent relationship between seed weight and rate of germination for anthocyaninless tomato mutants. The fruit from the bks mutant had a greater than usual pH (Table II), a phenotype reported for fruit from bs1 mutants (Martiniello et al., 1985).

ROS-Scavenging Enzymes Are Up-Regulated in the bks Mutant

The bks mutant seeds had greater PRX, significantly less CAT, and an equal amount of SOD activity relative to WT MT seeds (Fig. 5). The alterations in the activity of ROS-scavenging enzymes in the mutant seeds did not affect the quantities of ROS released from intact dry or 24-h-imbibed seeds, the bks mutant being indistinguishable from WT MT (Fig. 6). It is still possible that ROS generation is greater in the bks mutant but that the up-regulation of PRX may decrease the steady-state amounts of ROS to WT levels. Conversely, ROS generation may be normal in mutant seed and the up-regulation of PRX and down-regulation of CAT activity may be a peripheral consequence of the lesion, divorced from ROS generation. Certainly, enzymes associated with ROS scavenging have been implicated previously with developmental processes in the testa. PRX activity has been associated with the polymerization of extensin in the soybean (Glycine max) testa (Cassab and Varner, 1987), polymerization of cinnamyl alcohols during lignification (Gross, 1977), and in suberization (Espelie et al., 1986). Certain PRX enzymes are up-regulated by abscisic acid in diverse organisms experiencing dormancy (Chaloupková and Smart, 1994).

CAT is associated with seed quiescence and germination in a number of ways (McClung, 1997). One report has suggested that inhibition of CAT in dormant seeds is required for dormancy alleviation by redirecting hydrogen peroxide (H₂O₂) to participate in the oxidation of NADPH, thus permitting the pentose phosphate pathway to proceed (Hendricks and Taylorson, 1975). In maize, constitutive CAT activity during seed development can be stimulated, and during germination, inhibited, by the flavonoid-derived phytohormone, salicylic acid (Guan and Scandalios, 1995). However, there have been no reports of CAT affecting the testa.

A large family of germin-like proteins has been identified recently as SDS-resistant, large molecular mass, homo-hexamers of Mn-SOD enzymes (Carter and Thornburg, 2000). These enzymes, as their name implies, are associated with seed germination and dormancy in a plethora of dicot and monocot species (Lane, 1991) and in sporulation of Bacillus subtilis (Inaoka et al., 1999). However, the specific functions they perform during germination remain undefined. Such a germin-like SOD activity, whatever its function, is present in the bks mutant seeds (Fig. 5C, lower).

The Epistatic Relationship between bks and anthocyaninless Mutants

Precedence exists for the color of seed coats (testa and pericarp) to be imparted by flavonoid metabolites (Mol et al., 1998). Reports in the literature of tomato mutants affected in anthocyanin production producing seeds with lighter than usual testa color (ah, aw, and bls; Atanassova et al., 1997a) led to the speculation that the flavonoid biosynthetic pathway was functioning to produce condensed tannins in the tomato testa in much the same way it does in Arabidopsis. This conviction was strengthened by the fact that the only anthocyaninless defect currently identified in tomato (aw) is DFR (Goldsbrough et al., 1994), an enzyme in anthocyanin biosynthesis before the pathway branch leading to condensed tannin formation (Nesi et al., 2000) and producing seeds of a discernibly lighter testa color (Atanassova et al., 1997a). In addition, because CHS plays an essential role as the first committed enzyme channeling general phenylpropanoid metabolites into the myriad of products of the flavonoid biosynthetic pathway (Ferrer et al., 1999; Winkel-Shirley, 2001), the down-regulation of CHS activity in the tomato mutants bls and ah (O'Neill et al., 1990) would be expected to lead to seeds with a testa color lighter than normal (Atanassova et al., 1997a) and to a phenotype epistatic to the bks mutant if the latter's main effect was in the flavonoid pathway. The tt4 mutation in Arabidopsis is a null mutation in the single CHS gene in Arabidopsis (Shirley et al., 1995) and results in plants totally devoid of all flavonoids (Burbulis et al., 1996). These were the reasons to conduct the epistasis analysis with the anthocyaninless mutants in tomato. The dark testa color of the double mutants was in no case even mitigated, let alone eliminated, when combined with the anthocyaninless mutants. Hence, the anthocyaninless mutants are not epistatic to bks. However, is bks truly epistatic to the anthocyaninless mutants in that it is a lesion in a gene whose product is involved in a step higher in the metabolic pathway leading to the flavonoids, or is it epistatic only due to masking the anthocyaninless phenotype in the testa? The bks mutation is not epistatic to any of the anthocyaninless mutations with regards to hypocotyl pigments. This suggests that the dark testa color of the double-mutant seeds might be due to a masking of the anthocyaninless testa phenotype in aw, ah, and bls (i.e. the testa of the double mutants is deficient in anthocyanin-derived pigments, but it is not possible to determine this because of the dark color imparted by bks). In this case, bks would not be a lesion in a gene producing a perturbation in the flavonoid bio-synthetic pathway. Alternatively, bks may be diverting shikimate metabolites down an alternative pathway before the flavonoid branch at CHS. Such a branch point exists at p-coumaric acid (Goodwin and Mercer, 1983). This would mean that the bks phenotype must be restricted to the testa and would not be expected to alter the anthocyaninless phenotype in the hypocotyl. If this is the case and for reasons stated above, it is only possible to definitively state that the bks mutant must be at least higher in the shikimate biosynthetic pathway than DFR. Only a single gene is thought to encode DFR in tomato (Goldsbrough et al., 1994). Furthermore, the dark pigment present in the bks mutant testa accumulated predominately in the outer cell layers of the testa, whereas the endothelial layers, based on vanillin staining, present a more WT appearance relative to the accumulation of proanthocyanidins. The tt mutants of Arabidopsis differ in this respect, usually reducing condensed tannin accumulation in the endothelium of the inner integument as a consequence of any one of a variety of lesions (Devic et al., 1999; Nesi et al., 2000, 2001, 2002; Debeaujon et al., 2001; Johnson et al., 2002; Sagasser et al., 2002). Nonetheless, the Arabidopsis testa accumulates pigments reminiscent of those in the tomato testa in the cell layers exterior to the endothelium of the inner integument commencing at approximately the "walking stick" cotyledonary stage of development (Debeaujon et al., 2001).

A Bio- and Physicochemical Evaluation of the Pigment in the bks Mutant Testa

The determination that the bks mutant was epistatic to the anthocyaninless mutants, particularly ah, aw, and bls known to exhibit light testa coloration (Atanassova et al., 1997a) and the al and af mutants in which both CHS genes were down-regulated (O'Neill et al., 1990), was strong evidence that the lesions producing the bks mutant were not in the flavonoid biosynthetic pathway. The biochemical analysis of the bks testa pigment confirmed that the dark pigment was not extractable by any means known to be effective on anthocyanins (Supplemental Table I). In addition, the black pigment was highly resistant to acid hydrolysis, indicative of a melanic compound (Fogarty and Tobin, 1996). Furthermore, protocols used to solubilize lignin (Lu and Ralph, 1997) were not effective in solubilizing pigments from the bks mutant, suggesting that this mutant is not produced by increased flux of metabolite down the biosynthetic pathway leading to S, G, or H lignin (Sewalt et al., 1997). The bio- and physicochemical evidence leading to the assignment of the dark pigment present in the bks testa as a melanin is supported by the epistatic relationship of the bks mutant to all known anthocyaninless mutants. Melanin production resulting in black pigmentation proceeds by one of two pathways in plants. The first leads to compounds of the allomelanin variety deviating in the shikimic acid pathway at p-coumarate before the flavonoids (Goodwin and Mercer, 1983). The second produces eumelanin from Tyr via oxidation of DL-dioxy-Phe and is divorced from the shikimate pathway altogether. Either production scenario would be epistatic to the anthocyaninless lines. The dark appearance of the testa and its greater than usual toughness, indicative of a greater structural strength imparted by polymerization (Fogarty and Tobin, 1996), are also in support of the production of a melanic compound. In addition, polymerization of melanic compounds produces peroxide (Blois, 1978) and trapped free radicals that should lead to an up-regulation of PRX activity and an EPR signal, respectively. EPR is highly sensitive and the only direct means of detecting free radicals (Butterfield, 1982). One advantage of EPR over optical methods is the ability to examine opaque paramagnetic samples and paramagnetic solids. The resulting EPR spectrum of nonoriented solids (as the seeds examined in this study are) is normally a one-line spectrum, due to anisotropies of the g and hyperfine tensors not being averaged out by the motion found in dilute solutions of paramagnetic species. The amplitude and resonance position of the single line gave insight into the amount of unpaired spins present and the nature of the radical species. In the studies performed here, niger and sunflower seeds are rich in EPR-active components, most likely melanin-like substances (phytomelanin). That the EPR spectra of the black seeds was located at the same resonance position is suggestive that phytomelanin is present here as well. The amplitudes of the EPR spectra presented were indicative of the number of spins present, and the demarcation of results indicates that not all seeds have paramagnetic species present. These data are consistent with and corroborate the biochemical and genetic findings presented.

Contrasting Patterns of Inheritance between anthocyaninless and the bks Mutants

The manifestation in the seeds of all reported tt, ats, mum (Arabidopsis), and anthocyaninless (tomato) mutants affecting the biochemical and physical properties of the testa depends solely on the genotype of the maternal parent (i.e. the genetic composition of the maternally derived testa; Atanassova et al., 1997a). Hence, there is a lag of one generation in the elimination (mutant female) or manifestation (mutant male) of these phenotypes in crosses between nonallelic parents. However, the bs1, 2, 4, and bks mutants are all inherited as recessive, monogenic, Mendelian traits (i.e. their phenotype is not determined strictly by the genotype of the testa). This necessitates a hitherto unsuspected and certainly undescribed communication between the testa and the underlying endosperm/embryo. There appears, therefore, to be an antagonistic interplay between the maternal testa genome and that of the underlying endosperm/embryo with their paternal contribution that influences testa color and toughness. This antagonism provides a molecular basis for the parent-offspring conflict in seed germination (Ellner, 1986) without providing direct information on its mechanism(s) of action (i.e. although it is clear that the endosperm/embryo are influencing the color and toughness of the testa, it is not at all clear how this is accomplished).

Regardless of the mechanism that produces the bks mutant, the paternal contribution to testa attributes assisting eventual embryo egress represented by the bs mutants has been established and the paradigm reconfirmed by the recovery of the bks mutant. Our understanding of how the maternally derived testa is prepared to carry out its final task of protecting the tissue within has increased in sophistication with the realization that this preparation is influenced by underlying tissues with a paternal genetic component.

	MATERIALS AND METHODS

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 

Plant Material

Tomato (Lycopersicon esculentum Mill. cv MT; Scott and Harbaugh, 1989) seeds derived from fast neutron-mutagenized plants were used in this study. The fast neutron mutant collection was produced by irradiating MT seeds in the International Atomic Energy Agency accelerator (Vienna) with a dose of 15 Gy. M₁ plants were grown, and M₂ seeds were extracted from each plant separately. M₂ families (1,500) were grown, and M₃ seeds were collected in bulk from the eight to 12 M₂ plants grown in each family. The mutant screen was performed on M₃ seeds of 616 families as described below.

Mutant Screens

Up to 100 M₃ seeds were placed on 4 mL of distilled, deionized water on two 4-cm-diameter blotting paper discs (Stults Scientific Eng. Corp., Springfield, IL) in a petri dish. Dishes were placed inside plastic containers lined with water-saturated paper towels and incubated at 25°C in constant light (135 µmol m^-2 s^-1 photosynthetically active radiation). Tomato seeds that did not complete germination after 1 week on water were transferred to two 4-cm-diameter blotter discs (Stults Scientific Eng. Corp.) saturated with an aqueous solution of 100 µM GA_{4 + 7}. Seeds completing germination were discarded. Seeds that did not complete germination on GA_{4 + 7} within 5 d were removed from the dishes, the micropylar end of the testa and endosperm cap was surgically removed with a razor blade, and the seeds placed back on GA_{4 + 7} for an additional week. Seeds that had not completed germination after 1 week were discarded. All seeds that completed germination after surgical manipulation were transferred to soil. The pots were placed on an automatic controlled water table fertigation system (Buxton and Jia, 1999) in the greenhouse, and the germinants were grown to maturity. Seeds were harvested from the putative mutants and dried for at least 4 d at room temperature before being retested for their inability to complete germination under the same conditions used in the primary screen. Those mutants exhibiting the same phenotype as the M₄ parent were planted in soil along with WT MT tomato seeds (Totally Tomatoes, Augusta, GA), grown to maturity, and reciprocally crossed. Seeds from each parent of the reciprocal cross were examined to determine heritability. F₁ seeds from the crosses were planted, and the mutant seeds (identified by the dark color of the testa in 25% of the F₁ seeds) were recovered from the fruit and replanted to increase mutant seed numbers for subsequent testing. Seeds were harvested, cleaned (0.1 M HCl for 1 h), washed in tap water, dried (5% moisture content fresh weight basis), and stored at -20°C.

Allelism Test of bks with bs Mutants

Based on published reports concerning altered seed coloration in tomato, five mutations were identified that might be allelic to the black seeded mutants. Seeds for the ls mutation and three of four known brown seed mutants were obtained from the C.M. Rick Tomato Genetic Resource Center (University of California, Davis; bs3 [Yordanov and Stamova, 1971] has been lost due to its poor germination percentage [L. Stamova, personal communication]). The bks, bs, and ls mutant plants were crossed reciprocally, and F₁ seeds were examined for dark or normal testa color.

Ontogeny of Pigment Accumulation in bks/bs

Flowers at anthesis on bks/bs mutant plants were tagged daily. As fruit development progressed past the red ripe stage, one fruit from each of five different plants (all of the same age) was collected each day, and the seeds were examined for the first appearance of black coloration.

To localize the site of coloration, mature seeds were imbibed in water at 4°C for 4 h to facilitate dissection. These seeds were dissected into testa, endosperm, and embryo and observed under a dissecting microscope. Some seeds were sliced longitudinally with a razor blade to reveal the testa, endosperm, and embryo cell layers (Fig. 1F).

To determine the site in the testa in which the dark pigment was accumulating, seeds harvested from breaker stage fruit were glued to a block and sectioned fresh on a Vibratome Series 1500 tissue sectioning system (The Vibratome Company, St. Louis), and the sections (10 µm) were placed in water under a coverslip on a microscope slide. The sections were viewed on an Olympus BX40 (Olympus, Tokyo), and images were captured using a Canon EOSD30 and downloaded for archiving to ZoomBrowser EX software (Canon USA Inc., Lake Success, NY). Micrographs of sections were obtained before and after staining for proanthocyanidin accumulation using 1% (w/v) vanillin in 1 M HCl (Kristensen and Aastrup, 1986) and for phenolic compounds by staining for 5 s in 1 N ammonium hydroxide followed by mounting in 1 N NaOH (Bensley and Bensley, 1938).

Mutant Characterization

Seed weight was calculated from eight replications of 100 seeds each (International Seed Testing Association, 1993) from red ripe fruit. The planar area of one face of an individual seed's silhouette was calculated from four replications of 25 seeds each. Seeds were isolated from each other on a clear plastic 96-well plate lid and placed on a flat bed scanner with a transparency adaptor. Scans were made with the transparency adaptor engaged to eliminate shadowing. Images were scanned at 300 dpi and saved as gray scale tif files. Images were opened in SigmaScan Pro 5.0 for Windows (SPPC Science, Chicago), and the threshold function was used to select the pixels comprising each seed's image. The pixels comprising each seed image were then enumerated using the measure function of the same software, sorted in descending order, and exported to a spreadsheet (Excel, Microsoft, Redmond, WA). Excel values were imported into SAS (version 8, Statistical Analysis Systems, 1999) and statistically analyzed by two-way ANOVA (see below). Seeds were harvested from individual fruit, counted, and seed number fruit^-1 analyzed for significantly deviating means. Fruit pH at the red ripe stage was analyzed from 10 individual fruit replications by immersing a pH electrode into a test tube in which the fruit had been crushed.

Seed Germination

Seeds were plated onto three layers of germination blotter paper in square germination trays. The seeds were scanned every 12 h for 14 d using a Paradigm Seed Imaging System (Paradigm Research, Inc., South Haven, MN). After the completion of the test, the mean percentage germination for WT and mutant seed were compared using the Statistical Analysis System.

Four replicates of 25 seeds each were placed on GA_{4 + 7}, and radicle protrusion was recorded every 12 h for 14 d. In addition, seeds were harvested from fruit and fermented in the juice for 24 h to remove the sheath. The seeds were then apportioned into three equal fractions, one of which was dried over activated alumina in an air-, one in a nitrogen-, and one in an oxygen-flushed desiccator (Grabe, 1989). One day before the commencement of the germination test, seeds were extracted from additional fruit of each genotype. After 24 h of fermentation, they were cleaned. At this point, four replicates of 25 seeds each that had been dried for 4 d in each of the atmospheres or had not been dried were plated on germination paper (see above). The undried control in this experiment was possible because tomato seeds do not require dehydration to complete germination (Berry and Bewley, 1991).

Analysis of Puncture Force

Three replications of 10 seeds each 24 HAI on water were bisected transversely, the radicle tip was removed from the micropylar half seed, and a blunt probe of the same diameter as the radicle was inserted into the embryo cavity. The force in Newtons required to push through the endosperm and testa of the micropyle was recorded with a Chatillon force analyzer (Chatillon, Greensboro, NC).

Enzyme Assays

PRX
PRX activity was measured in extracts of total protein from three replications of 100 whole dry seeds. Seed samples were weighed, transferred to a mortar, and pulverized in liquid nitrogen. One milliliter of extraction buffer (25 mM potassium phosphate [pH 7.5]) was added, the mixture was ground into a slurry, transferred to a microfuge tube, and centrifuged at 13,000g for 10 min at 4°C, and the supernatant was assayed directly.

Activity from 10 µL of homogenate was monitored for 10 min at A₄₉₀ with in a UV-2101PC spectrophotometer (Shimadzu, Kyoto) using 8 mM Guaiacol and 0.5 mM H₂O₂ in 50 mM potassium acetate (pH 5.5) at 25°C (Gillikin and Graham, 1991).

PPO
Three replications of 100 mature dry seeds were extracted in 0.1 M cacodylate buffer (pH 7.2) and 0.01% (w/v) SDS and assayed in the same buffer made 25 mM with respect to DL-dioxy-Phe (Sigma, St. Louis; Vaughn and Duke, 1981). Air was bubbled though the assay buffer for 5 min just before the commencement of each assay. Assays were monitored at A₄₇₀ using 100 µL of enzyme extract in 2 mL of assay buffer at 25°C for 5 min. Duplicate assays were run on each extract. To one assay, 100 µL of extraction buffer was added to the cuvette, whereas in the second assay, 420 units of CAT (Sigma) in extraction buffer was added to prevent peroxidation of the substrate (Vaughn and Duke, 1981), thereby distinguishing between PPO and PRX (Egley et al., 1983).

CAT
The activity of CAT in seed extracts was measured at a constant 30°C using an Oxylab oxygen electrode (version 1.09, Hansatech Instruments Ltd, Norfolk, UK). Three replications of 100 dry seeds per genotype were pulverized in liquid nitrogen and ground in 1 mL of a 50 mM sodium phosphate buffer (pH 7.8). The slurry was centrifuged, and aliquots of the supernatant were diluted 10- or 20-fold depending on the activity of the extract. Assays were conducted for 1 min commencing immediately upon the addition of 2 µL of diluted supernatant delivered with a Hamilton syringe into 1 mL of 17.6 mM H₂O₂ (Fath et al., 2001) in the reaction chamber. Activity in the seed extracts was quantified using a standard curve of commercial bovine liver CAT (Sigma). An in-gel determination of the isoforms of CAT present in the seeds of the various genotypes was performed (Michán et al., 2002).

SOD
SOD was assayed by following the quenching of the reduction of acetylated cytochrome c (Sigma) by xanthine/xanthine oxidase-generated superoxide (Flohé and ötting, 1984). Three replications of 100 mature, dry seeds per genotype were pulverized in liquid N₂, and 1 mL of 50 mM sodium phosphate buffer (pH 7.8) and 0.1 mM EDTA were added. After centrifugation, the supernatant was dialyzed overnight at 4°C against 50 mM sodium phosphate buffer (pH 7.8). Dialyzed extract was diluted 10-fold, and 50 µL was used in a final assay volume of 3 mL. The rate of cytochrome c reduction was followed for 5 min at A₅₅₀, and the quenching of this reduction was compared with a standard curve generated from commercial SOD from human erythrocytes (Sigma). Investigation of the mass of the holoenzyme(s) in the bks mutants that exhibited SOD activity used an in-gel assay (Carter and Thornburg, 2000).

For each enzyme assay and activity gel, protein determinations were made for each replication with Bradford's reagent using bovine serum albumin fraction V (Sigma) at various dilutions as a standard.

ROS Quantification

Quantification of total ROS used a fluorometric assay (Yatin et al., 2000) based on the oxidation of dichlorofluorescin to dichlorofluorescein and was performed on whole seeds that had or had not been imbibed 24 h on water (Schopfer et al., 2001). Five single, dry, or 24-h-imbibed seed replications of bks mutant or WT MT were placed in 200 µL of 20 mM potassium phosphate buffer (pH 6.0 containing 50 µM esterase-treated, 2',7'-dichlorofluorescin [Sigma] just before use according to the instructions of the authors [Schopfer et al., 2001]) in black microtitre plate wells, and the increase in fluorescence caused by the conversion of dichlorofluorescin to dichlorofluorescein by seed-released ROS was determined every 5 min for 30 min using a Spectra-Max Gemini model XS plate reader (Molecular Devices, Sunnyvale, CA) at 485-nm excitation and 530-nm emission wavelengths.

Genetic Analysis

Twelve chromosomal marker lines were obtained from the C.M. Rick Tomato Genetic Resource Center and crossed with bks1-1 and bks1-2. Seeds from F₁ plants were separated into groups of dark and normal testa color, surface sterilized for 20 min in 3.5% (v/v) commercial bleach, rinsed well, and placed on separate, moist germination blotter papers at 25°C. After 1 week, bks seeds that had not yet completed germination had the testa removed with forceps and were placed back on a moist germination blotter. Upon the completion of germination, seedlings emerging from black or normal colored seeds were transferred to soil in appropriately designated trays. When all seedlings were of sufficient size to manifest the mutation of interest, seedlings were scored and the results compared using ² for deviations from the expected 9:3:3:1 segregation ratio. Evidence of linkage in repulsion phase was verified by scoring F₂ seeds from the appropriate cross of the allelic bks mutant and the chromosomal marker line under scrutiny.

The bks/bs mutants were crossed with 12 anthocyaninless mutants (obtained from the C.M. Rick Tomato Genetic Resource Center), some of which are known to produce seeds with lighter than usual testa color, to test the epistatic relationship between anthocyaninless mutants and bks. Dark F₂ seeds were planted, and F₂ seedlings that were devoid of anthocyanin in the hypocotyl were grown to maturity. The F₃ seeds from the double mutant, possessing a testa homozygous for the anthocyaninless mutation under examination, were examined for testa color. F₃ seeds that were darker than normal were allowed to complete germination, and the green hypocotyl color of the seedlings was verified.

Biochemical Characterization of the Accumulating Pigment in bks Testae

An array of solvents was used in attempts to solubilize the black pigment present in bks seeds, some of which were effective in solubilizing anthocyanins (Bate-Smith, 1975) and lignans (Lu and Ralph, 1997). Positive controls for anthocyanin extraction were from black turtle bean (Phaseolus vulgaris) testa, and monomer standards for derivatization followed by reductive cleavage products were 4-acetoxycinnamyl acetate, coniferyl diacetate, sinapyl diacetate, and an internal standard of 4,4'-ethylidenebisphenol. One hundred milligrams of WT or bks testae was pulverized in liquid nitrogen, and the powder was placed in 1 mL of the designated solvent. These mixtures were allowed to steep at room temperature in organic solvents overnight. The resulting solutions were filtered through fiberglass filters and compared both visually and by scanning from A₂₀₀ to A₈₀₀ nm to determine absorption maxima, if any. Acid hydrolysis was carried out for 2 to 4 h at either 95°C or 100°C in the presence and absence of air. The latter was accomplished by placing 100 mg of ground testa in an ignition tube, adding 1 mL of the appropriate concentration of acid (Supplemental Table I), freezing the contents in liquid nitrogen, evacuating, sealing, and then heating the tube and its contents. Evaluation of differences between mutant and WT testae extracts were carried out as outlined above.

For base hydrolysis, 1.0 mL of 0.5 N NaOH was added to ground mutant or WT testae (100 mg). The mixture was heated (1 h, 60°C) cooled, centrifuged, and decanted. The residue was washed twice (0.5 N NaOH) and combined with the original hydrolysate. The pH of the solution was adjusted to 2 (6 N HCl), and the resulting suspension was centrifuged and the pellet resuspended in NaOH and reprecipitated as above (Makordei et al., 1994). The pellet was lyophilized, and its solubility in DMSO assessed. Solubility in DMSO is a property indicative of a melanic compound (Li et al., 2001).

EPR Analysis of bks Testae

Aliquots of testae ground in liquid nitrogen were packed in quartz capillary tubes and placed in EPR quartz sample tubes. EPR spectra were acquired according to established procedure (Butterfield, 1982) on a Bruker (Billerica, MA) EMX EPR spectrometer. Instrumental parameters were as follows: microwave power, 20 mW; modulation amplitude, 0.3 G; receiver gain, 1 x 10⁵; conversion time, 40.96 ms; center field, 3,495 G; and sweep width, 100 G. The empty cavity background (100 scans) was subtracted from testae spectra (100 scans) using Bruker WINEPR software. The height of the peak centered around 3475 G was measured using the WINEPR software. To compare the mutant testae spectra that were acquired with 100 scans to the positive control spectra, niger [Guizotia abyssinica (L.S.) Cass.] and sunflower (Helianthus annuus), which were obtained with only one scan, the intensity of the signals obtained were divided by the number of scans.

Statistical Analysis

Seed weight, number of seed per fruit, fruit pH, puncture force, enzyme activity, release of ROS from seeds, and final germination percentage were subjected to analysis of variance among genotypes using ANOVA (SAS, Cary, NC). If the ANOVA indicated that there were significant differences among means, the analysis was rerun using Tukey's mean separation test to distinguish among significantly deviating means.

For the analysis of linkage of bks with chromosomal marker lines, ² analysis of goodness-of-fit to a 9:3:3:1 segregation ratio for two independent (unlinked) loci were calculated using Excel.

	ACKNOWLEDGMENTS

 
The GA_{4 + 7} used in these experiments was the kind gift of Abbott Biochemicals (Chicago). The ls and bs mutant lines, chromosomal marker lines, and anthocyaninless lines were provided by Roger Chetelat from stock maintained at the C.M. Rick Tomato Genetic Resource Center (University of California, Davis). 4-acetoxycinnamyl acetate, coniferyl diacetate, sinapyl diacetate, and 4,4'-ethylidenebisphenol, used as positive controls in the DFRC assays, were the kind gift of Dr. John Ralph (University of Wisconsin, Madison). We wish to acknowledge the intellectual contribution of Prof. Jan St. Pyrek, who initially suggested that the black substance in the mutant tomato testa might be melanin. Mr. Manjul Dutt and Prof. Robert Geneve developed the technique for acquiring and measuring seed silhouettes. Mr. David McNertney allowed us the use of the Paradigm Co. seed imaging and analysis system. Darrell Slone, Kay Oakley, and David Lowry maintained the plants at the University of Kentucky Horticulture Experimental Farm (Lexington).

Received February 25, 2003; returned for revision April 28, 2003; accepted June 7, 2003.

	FOOTNOTES

 
¹ This work was supported by the Department of Horticulture (University of Kentucky, Lexington) and by Hatch funds.

^[w] The online version of this article contains Web-only data.

² These authors contributed equally to the paper.

^* Corresponding author; e-mail adownie{at}uky.edu; fax 859-257-7874.

	LITERATURE CITED

TOP ABSTRACT RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

 
Association of Official Seed Analysts (1983) Seed Vigor Testing Handbook, No. 32. Association of Official Seed Analysts, XXXX, XX

Atanassova B, Daskalov S, Shtereva L, Balatcheva E (2001) Anthocyanin mutations improving tomato and pepper tolerance to adverse climatic conditions. Euphytica 120: 357-365[CrossRef]

Atanassova B, Shtereva L, Molle E (1997a) Effect of three anthocyaninless genes on germination in tomato (Lycopersicon esculentum Mill.): I. Seed germination under optimal conditions. Euphytica 95: 89-98[CrossRef]

Atanassova B, Shtereva L, Molle E (1997b) Effect of three anthocyaninless genes on germination in tomato (Lycopersicon esculentum Mill.): II. Seed germination under stress conditions. Euphytica 97: 31-38[CrossRef]

Balint-Kurti PJ, Jones DA, Jones JDG (1995) Integration of the classical and RFLP linkage maps of the short arm of tomato chromosome 1. Theor Appl Genet 90: 17-26

Bate-Smith EC (1975) Phytochemistry of proanthocyanindins. Phytochemistry 14: 1107-1113[CrossRef][Web of Science]

Bensley RR, Bensley SH (1938) Handbook of Histological and Cytological Technique. The University of Chicago Press, Chicago

Berry T, Bewley JD (1991) Seeds of tomato (Lycopersicon esculentum Mill.) which develop in a fully hydrated environment in the fruit switch from a developmental to a germinative mode without a requirement of desiccation. Planta 186: 27-34

Bewley JD (1997) Seed germination and dormancy. Plant Cell 9: 1055-1066[CrossRef][Web of Science][Medline]

Blois MS (1978) The melanins: their synthesis and structure. Photochem Photobiol Rev 3: 115-134

Borevitz JO, Xia Y, Blount J, Dixon RA, Lamb C (2000) Activation tagging identifies a conserved MYB regulator of phenylpropanoid biosynthesis. Plant Cell 12: 2383-2393[Abstract/Free Full Text]

Burbulis IA, Iacobucci M, Shirley BW (1996) A null mutation in the first enzyme of flavonoid biosynthesis does not affect male fertility in Arabidopsis. Plant Cell 8: 1013-1025[Abstract]

Butterfield DA (1982) Spin labeling in disease. In LJ Berliner, J Reuben, eds, Biological Magnetic Resonance, Vol. IV. Plenum Press, New York, pp 1-78

Buxton JW, Jia W (1999) A controlled water table irrigation system for hydroponic lettuce production. In AP Papdopoulos, ed, Proceedings of the International Symposium on Growing Media and Hydroponics: Acta Horticulturae No. 481. International Society for Horticultural Science, Leuven, Belgium, pp 281-287

Carter C, Thornburg RW (2000) Tobacco nectarin I: purification and characterization as a germin-like manganese superoxide dismutase implicated in the defense of floral reproductive tissues. J Biol Chem 275: 36726-36733[Abstract/Free Full Text]

Cassab GI, Varner JE (1987) Immunocytolocalization of extensin in developing soybean seed coats by immunogold-sliver staining and by tissue printing on nitrocellulose paper. J Cell Biol 105: 2581-2588[Abstract/Free Full Text]

Chaloupková K, Smart CC (1994) The abscisic acid induction of a novel peroxidase is antagonized by cytokinin in Spirodela polyrrhiza L. Plant Physiol 105: 497-507[Abstract]

Cooper DC (1931) Macrosporogenesis and the development of the macrogametophyte of Lycopersicon esculentum. Am J Bot 18: 739-748[CrossRef]

Debeaujon I, Léon-Kloosterziel KM, Koornneef M (2000) Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis. Plant Physiol 122: 403-414[Abstract/Free Full Text]

Debeaujon I, Koornneef M (2000) Gibberellin requirement for Arabidopsis seed germination is determined both by testa characteristics and embryonic abscisic acid. Plant Physiol 122: 415-424[Abstract/Free Full Text]

Debeaujon I, Peeters AJM, Léon-Kloosterziel KM, Koornneef M (2001) The TRANSPARENT TESTA12 gene of Arabidopsis encodes a multidrug secondary transporter-like protein required for flavonoid sequestration in vacuoles of the seed coat endothelium. Plant Cell 13: 853-871[Abstract/Free Full Text]

Devic M, Guilleminot J, Debeaujon I, Bechtold N, Bensaude E, Koornneef M, Pelletier G, Delseny M (1999) The BANYULS gene encodes a DFR-like protein and is a marker of early seed coat development. Plant J 19: 387-398[CrossRef][Web of Science][Medline]

Dong X, Braun EL, Grotewold E (2001) Functional conservation of plant secondary metabolic enzymes revealed by complementation of Arabidopsis flavonoid mutants with maize genes. Plant Physiol 127: 46-57[Abstract/Free Full Text]

Downie B, Gurusinghe S, Bradford KJ (1999) Internal anatomy of individual tomato seeds: relationship to abscisic acid and germination physiology. Seed Sci Res 9: 117-128

Egley GH, Paul RN Jr, Vaughn KC, Duke SO (1983) Role of peroxidase in the development of water-impermeable seed coats in Sida spinosa L. Planta 157: 224-232[CrossRef]

Ellner S (1986) Germination dimorphisms and parent-offspring conflict in seed germination. J Theor Biol 123: 173-185[CrossRef]

Espelie KE, Franceschi VR, Kolattukudy PE (1986) Immunocytochemical localization and time course of appearance of an anionic peroxidase associated with suberization in wound-healing potato tuber tissue. Plant Physiol 81: 487-492[Abstract/Free Full Text]

Fath A, Bethke PC, Jones RL (2001) Enzymes that scavenge reactive oxygen species are down-regulated prior to gibberellic acid-induced programmed cell death in barley aleurone. Plant Physiol 126: 156-166[Abstract/Free Full Text]

Ferrer J-L, Jez JM, Bowman ME, Dixon RA, Noel JP (1999) Structure of CHALCONE SYNTHASE and the molecular basis of plant polyketide biosynthesis. Nat Struct Biol 6: 775-784[CrossRef][Web of Science][Medline]

Flohé L, ötting F (1984) Superoxide dismutase assays. In Lester Packer, ed, Methods in Enzymology. 105: 93-104[Web of Science][Medline]

Fogarty RV, Tobin JM (1996) Fungal melanins and their interactions with metals. Enz Microbiol Technol 19: 311-317

Foolad MR, Chen FQ, Lin GY (1998) RFLP mapping of QTLs conferring salt tolerance during germination in an interspecific cross of tomato. Theor Appl Genet 97: 1133-1144[CrossRef]

Gijzen M (1997) A deletion mutation at the ep locus causes low seed coat peroxidase activity in soybean. Plant J 12: 991-998[CrossRef][Web of Science][Medline]

Gillikin JW, Graham JS (1991) Purification and developmental analysis of the major anionic peroxidase from the seed coat of Glycine max. Plant Physiol 96: 214-220[Abstract/Free Full Text]

Goldsbrough A, Belzile F, Yoder JI (1994) Complementation of the tomato anthocyanin without (aw) mutant using the DIHYDROFLAVONOL 4-REDUCTASE gene. Plant Physiol 105: 491-496[Abstract]

Goodwin TW, Mercer EI (1983) Introduction to Plant Biochemistry, Ed 2. Pergamon Press, Oxford

Grabe DF (1989) Measurement of seed moisture. In PC Stanwood and MB McDonald, Jr., eds, Seed Moisture: Special Publication 14. Crop Science Society of America, Madison, WI, pp 69-92

Gross GG (1977) Biosynthesis of lignin and related monomers. Rec Adv Phytochem 11: 141-184

Gualberti G, Papi M, Bellucci L, Ricci I, Bouchez D, Camilleri C, Costantino P, Vittorioso P (2002) Mutations in the Dof Zinc finger genes DAG2 and DAG1 influence with opposite effects the germination of Arabidopsis seeds. Plant Cell 14: 1253-1263[Abstract/Free Full Text]

Guan L, Scandalios JG (1995) Developmentally related responses of maize catalase genes to salicylic acid. Proc Natl Acad Sci USA 92: 5930-5934[Abstract/Free Full Text]

Harki E, Talou T, Dargent R (1997) Purification, characterization and analysis of melanin extracted from Tuber melanosporum Vitt. Food Chem 58: 69-73[CrossRef]

Hendricks SB, Taylorson RB (1975) Breaking of seed dormancy by catalase inhibition. Proc Natl Acad Sci USA 72: 306-309[Abstract/Free Full Text]

Hilhorst WMH, Downie B (1996) Primary dormancy in tomato (Lycopersicon esculentum cv. Moneymaker): studies with the sitiens mutant. J Exp Bot 47: 89-97

Inaoka T, Matsumura Y, Tsuchido T (1999) SodA and manganese are essential for resistance to oxidative stress in growing and sporulating cells of Bacillus subtilis. J Bacteriol 181: 1939-1943[Abstract/Free Full Text]

International Seed Testing Association (1993) International rules for seed testing. Seed Sci Technol Suppl 21: xxx-xxx

Johnson CS, Kolevski B, Smyth DR (2002) TRANSPARENT TESTA GLA BRA2, a trichome and seed coat development gene of Arabidopsis, encodes a WRKY transcription factor. Plant Cell 14: 1359-1375[Abstract/Free Full Text]

Koornneef M (1990) Mutations affecting the testa colour in Arabidopsis, Arabidopsis Inf Serv 27: 1-4

Koornneef M, Bentsink L, Hilhorst H (2002) Seed dormancy and germination. Curr Opin Plant Biol 5: 33-36[CrossRef][Web of Science][Medline]

Koornneef M, Karssen CM (1994) Seed dormancy and germination. In EM Meyerowitz, CR Somerville, eds, Arabidopsis. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp 313-334

Kristensen KN, Aastrup S (1986) A non-destructive screening method for proanthocyanidin-free barley mutants. Carlsberg Res Commun 51: 509-513

Lane BG (1991) Cellular desiccation and hydration developmentally regulated proteins and the maturation and germination of seed embryos. FASEB J 5: 2893-2901[Abstract]

Léon-Kloosterziel KM, Keijzer CJ, Koornneef M (1994) A seed shape mutant in Arabidopsis thaliana that is affected in integument development. Plant Cell 6: 385-392[Abstract]

Leviatov S, Shoseyov O, Wolf S (1994) Roles of different seed components in controlling tomato seed germination at low temperature. Sci Hortic 56: 197-206

Li X, Wang Z, Jiang X (2001) Purification of melanin extracted from squid ink. Shanghai Shuichan Daxue Xuebao 10: 252-256

Lu F, Ralph J (1997) DFRC method for lignin analysis: I. New method for -aryl ether cleavage: lignin model studies. J Agric Food Chem 45: 4655-4660[CrossRef]

Makordei FV, Venger LA, Slyusarenko LI, Barba IN (1994) Allomelanins. Isolation methods, physicochemical properties, and possibilities of practical use. Izvestiya Vysshikh Uchebnykh Zavedenii Khimiya i Khimicheskaya Tekhnologiya 37: 4-6

Martiniello P, Falabigna A, Soressi GP (1985) Influence of the tomato Lycopersicon esculentum BROWN SEED gene on plant and fruit characteristics. Genet Agraria 39: 417-422

Mazzucato A, Taddei AR, Soressi GP (1998) The parthenocarpic fruit (pat) mutant of tomato (Lycopersicon esculentum Mill.) sets seedless fruits and has aberrant anther and ovule development. Development 125: 107-114[Abstract]

McClung CR (1997) Regulation of catalases in Arabidopsis. Free Radic Biol Med 23: 489-496[CrossRef][Web of Science][Medline]

Michán S, Lledías F, Baldwin JD, Natvig DO, Hansberg W (2002) Regulation and oxidation of two large monofunctional catalases. Free Radic Biol Med 33: 521-532[CrossRef][Medline]

Mol J, Grotewold E, Koes R (1998) How genes paint flowers and seeds. Trends Plant Sci 3: 212-217[CrossRef][Web of Science]

Monti LM (1972) Three "brown seed" mutations induced in tomato by neutron treatments. Report Tom Genet Coop 22: 18

Nesi N, Debeaujon I, Jond C, Pelletier G, Caboche M, Lepiniec L (2000) The TT8 gene encodes a basic helix-loop-helix domain protein required for expression of DRF and BAN genes in Arabidopsis siliques. Plant Cell 12: 1863-1878[Abstract/Free Full Text]

Nesi N, Debeaujon I, Jond C, Stewart AJ, Jenkins GI, Caboche M, Lepiniec L (2002) The TRANSPARENT TESTA14 locus encodes the ARABIDOPSIS BSISTER MADS domain protein and is required for proper development and pigmentation of the seed coat. Plant Cell 14: 2463-2479[Abstract/Free Full Text]

Nesi N, Jond C, Debeaujon I, Caboche M, Lepiniec L (2001) The Arabidopsis TT2 gene encodes an R2R3 MYB domain protein that acts as a key determinant for proanthocyanidin accumulation in developing seed. Plant Cell 13: 2099-2114[Abstract/Free Full Text]

O'Neill SD, Tong Y, Spörlein B, Gorkmann G, Yoder J (1990) Molecular genetic analysis of CHALCONE SYNTHASE in Lycopersicon esculentum and an anthocyanin-deficient mutant. Mol Gen Genet 224: 279-288[CrossRef][Web of Science][Medline]

Papi M, Sabatini S, Altamura MM, Hennig L, Schafer E, Costantino P, Vittorioso P (2002) Inactivation of the phloem-specific Dof zinc finger gene DAG1 affects response to light and integrity of the testa of Arabidopsis seeds. Plant Physiol 128: 411-417[Abstract/Free Full Text]

Papi M, Sabatini S, Bouchez D, Camilleri C, Costantino P, Vittorioso P (2000) Identification and disruption of an Arabidopsis zinc finger gene controlling seed germination. Genes Dev 14: 28-33[Abstract/Free Full Text]

Philouze J (1970) A second gene for brown color of tomato seed. Report Tom Genet Coop 20: 44

Philouze J (1974) Étude de deux genes de coloration brune de la graine de tomate. Ann Amélior Plantes 24: 317-333

Rick CM (1946) The development of sterile ovules in Lycopersicon esculentum Mill. Am J Bot 33: 250-256[CrossRef]

Sagasser M, Lu G-H, Hahlbrock K, Weisshaar B (2002) A. thaliana TRANS PARENT TESTA1 is involved in seed coat development and defines the WIP subfamily of plant zinc finger proteins. Genes Dev 16: 138-149[Abstract/Free Full Text]

Sava VM, Yang S-M, Hong M-Y, Yang P-C, Huang GS (2001) Isolation and characterization of melanic pigments derived from tea and tea pholyphenols. Food Chem 73: 177-184[CrossRef]

Schumacher K, Schmitt T, Rossberg M, Schmitz G, Theres K (1999) The Lateral suppressor (Ls) gene of tomato encodes a new member of the VHIID protein family. Proc Natl Acad Sci USA 96: 290-295[Abstract/Free Full Text]

Schopfer P, Plachy C, Frahry G (2001) Release of reactive oxygen intermediates (superoxide radicles, hydrogen peroxide, and hydroxyl radicals) and peroxidase in germinating radish seeds controlled by light, gibberellin, and abscisic acid. Plant Physiol 125: 1591-1602[Abstract/Free Full Text]

Scott JW, Harbaugh BK (1989) Micro-Tom: a miniature dwarf tomato. Florida Agric Exp Stn Circ 370: 1-6

Sewalt VJH, Ni W, Blount JW, Jung HG, Masoud SA, Howles PA, Lamb C, Dixon RA (1997) Reduced lignin content and altered lignin composition in transgenic tobacco down-regulated in expression of L-phenylalanine ammonia-lyase or cinnamate 4-hydroxylase. Plant Physiol 115: 41-50[Abstract]

Shirley BW, Kubasek WL, Storz G, Bruggemann E, Koornneef M, Ausubel FM, Goodman HM (1995) Analysis of Arabidopsis mutants deficient in flavonoid biosynthesis. Plant J 8: 659-671[CrossRef][Web of Science][Medline]

Soressi GP (1967) Brown seed (bs), a tomato seed character which behaves as an endosperm trait. Report Tom Gen Coop 17: 50

Soressi GP (1972) Allelism test for different brown seed (bs) genes. Rep Tom Gen Coop 22: 25-26 Statistical Analysis Systems (1999) Version 8. SAS Institute Inc., Cary NC, USA.

Statistical Analysis Systems (1999) Version 8. SAS Institute Inc., Cary, NC.

Tanksley SD, Ganal MW, Prince JP, de Vincente MC, Bonierbale MW, Broun P, Fulton TM, Giovannoni JJ, Grandillo S, Martin GB et al. (1992) High density molecular linkage maps of the tomato and potato genomes. Genetics 132: 1141-1160[Abstract]

Taylor IB (1979) The effect of the lateral suppressor gene on seed germination in tomato. Euphytica 28: 93-97[CrossRef]

Todd JJ, Vodkin LO (1993) Pigmented soybean (Glycine max) seed coats accumulate proanthocyanidins during development. Plant Physiol 102: 663-670[Abstract]

Tucker DJ (1976) Endogenous growth regulators in relation to sideshoot development in the tomato. New Phytol 77: 561-568

Vaughn KC, Duke SO (1981) Tentoxin-induced loss of plastidic polyphenol oxidase. Physiol Plant 53: 421-428[CrossRef]

Walker JRL (1962) Phenolic acids in "cloud" and normal tomato fruit wall tissue. J Sci Food Agric 13: 363-367[CrossRef]

Western TL, Burn J, Tan WL, Skinner DJ, Martin-McCaffrey L, Moffatt BA, Haughn GW (2001) Isolation and characterization of mutants defective in seed coat mucilage secretory cell development in Arabidopsis. Plant Physiol 127: 998-1011[Abstract/Free Full Text]

Whittington WJ, Fierlinger P (1972) The genetic control of time to germination in tomato. Ann Bot 36: 873-880[Abstract/Free Full Text]

Winkel-Shirley B (2001) Flavonoid biosynthesis: a colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol 126: 485-493[Free Full Text]

Xie D-Y, Sharma SB, Paiva NL, Ferreira D, Dixon RA (2003) Role of ANTHOCYANIDIN REDUCTASE, encoded by BANYULS in plant flavonoid biosynthesis. Science 299: 396-399[Abstract/Free Full Text]

Yatin SM, Varadarajan S, Butterfield DA (2000) Vitamin E prevents Alzheimer's amyloid beta-peptide (1-42)-induced protein oxidation and reactive oxygen species formation. J Alzheimer's Dis 2: 123-131[Medline]

Yordanov M, Stamova L (1971) A new brown seed mutant. Report Tom Genet Coop 21: 41

This article has been cited by other articles:

				A. Kaga, T. Isemura, N. Tomooka, and D. A. Vaughan The Genetics of Domestication of the Azuki Bean (Vigna angularis) Genetics, February 1, 2008; 178(2): 1013 - 1036. [Abstract] [Full Text] [PDF]

				S. Penfield, E. L. Rylott, A. D. Gilday, S. Graham, T. R. Larson, and I. A. Graham Reserve Mobilization in the Arabidopsis Endosperm Fuels Hypocotyl Elongation in the Dark, Is Independent of Abscisic Acid, and Requires PHOSPHOENOLPYRUVATE CARBOXYKINASE1 PLANT CELL, October 1, 2004; 16(10): 2705 - 2718. [Abstract] [Full Text] [PDF]

				A. B. Downie, L. M. A. Dirk, Q. Xu, J. Drake, D. Zhang, M. Dutt, A. Butterfield, R. R. Geneve, J. W. Corum III, K. G. Lindstrom, et al. A physical, enzymatic, and genetic characterization of perturbations in the seeds of the brownseed tomato mutants J. Exp. Bot., May 1, 2004; 55(399): 961 - 973. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 133/1/145    most recent pp.103.022632v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Downie, A. B. Articles by Snyder, J. C. Search for Related Content PubMed PubMed Citation Articles by Downie, A. B. Articles by Snyder, J. C. Agricola Articles by Downie, A. B. Articles by Snyder, J. C.