INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plants, in their need to capture sunlight for photosynthesis, are unavoidably exposed to the damaging effects of UVB radiation. The highly energetic photons in these wavelengths cause damage to DNA and other macromolecules, which can lead to cellular injury, mutagenesis, and death. DNA is one of the major targets of UV damage; however, other components (e.g. proteins, membrane lipids, etc.) also appear to be damaged by UV. For example, UV radiation is known to affect protein synthesis (Andley et al., 1990; Hightower et al., 1994), to directly damage proteins through absorption in the UVB region by aromatic aminoacids (Jagger, 1985; Pigault and Gerard, 1989; Dillon, 1991; Caldwell, 1993), to have effects on enzymatic function (Jordan et al., 1992; Pfundel and Pan, 1992), and to damage lipids through peroxidation (Taira et al., 1992; Hightower et al., 1994). Part of the damage induced by UVB is believed to be caused indirectly through the production of free radicals, such as superoxide radicals, singlet oxygen, and hydroxyl radicals (Pathak and Stratton, 1969; Peak and Peak, 1975; Peak and Peak, 1983; Foyer et al., 1994). A variety of adaptations have evolved that help plants cope with the exposure to UV but we do not yet understand the exact role many of them play in the overall protection against UV damage. The isolation and characterization of mutants has been a powerful tool to learn about various mechanisms that help protect plants against different types of UV radiation damage. Mutants showing hypersensitivity to UV radiation have been instrumental in laying down the groundwork for our understanding of some of these mechanisms. For example, studies of the transparent testa, fah1, and uvs mutants (Li et al., 1993; Lois and Buchanan, 1994; Landry et al., 1995) have increased our understanding of different aspects of the role of phenolics, including flavonoids and sinapates, in the protection of plants against UV radiation (for review, see Bharti and Khurana, 1997). Mutants deficient in DNA repair (Britt et al., 1993; Jiang et al., 1997; Vonarx et al., 1998) have helped elucidate the involvement of mechanisms of photoreactivation and dark repair of DNA in UV protection.

In an attempt to increase our understanding of the various mechanisms of defense against UVB radiation, we set out to search for mutations that would increase, rather than decrease, the level of tolerance of Arabidopsis plants to UV radiation. We describe below the isolation and preliminary characterization of an Arabidopsis mutant that shows a remarkable tolerance of UV radiation levels that are lethal to wild-type plants. Our data indicates that a constitutive increase in accumulation of UV-absorbing pigments due to changes in gene expression is responsible for this elevated resistance to UVB. This mutant may prove to be a powerful tool to further elucidate the molecular mechanisms of protection against UV as well as the potential contribution of UV screens and other mechanisms to the overall tolerance of plants to UV radiation.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Isolation of the uvt1 Mutant

With the goal of increasing our understanding of the different mechanisms of protection against UVB radiation in Arabidopsis, we set out to isolate mutants that show elevated tolerance to UV. To increase our chances of detecting increases in UV tolerance, we used a mutant highly sensitive to UV radiation (uvs), previously isolated in our laboratory, as the genetic background for a new round of ethyl methane sulfonate (EMS)-mediated mutagenesis. As described previously (Lois and Buchanan, 1994), the UV sensitivity of uvs is due to a single recessive lesion in a gene that leads to altered accumulation of UV-absorbing compounds. This mutant does not show the significant increase in absorption in the UV range observed in the wild type upon exposure to UVB radiation. We believe that the increased levels of these UV-absorbing compounds act as a shield that blocks UV before it reaches sensitive targets and that the lack of these derivatives in uvs results in this mutant's extreme sensitivity to UV light.

After EMS mutagenesis, M1 generation uvs seedlings were screened during UV irradiation for plants that showed diminished UV damage as evidenced by greener, healthier appearance than the rest of the uvs population. uvs plants typically become necrotic after a short exposure to low levels of UVB. A screening of approximately 5,000 EMS-treated seeds led to the isolation of three mutants exhibiting a pronounced increase in tolerance to UV, which we have named uvt1, uvt2, and uvt3 (for UV tolerant). The mutant uvt1 showed the most dramatic increase in tolerance to UVB and is described below.

Because uvt1 was isolated in a screening of the M1 population, it was unlikely that we would isolate any recessive mutations. Thus, from a selfing cross of the uvt1 mutant we expected to find a segregating population of resistant and sensitive plants. Furthermore, because the M1 population consists of chimeric plants harboring the mutation only in certain clonal sectors, it was clear that the uvt1 mutation would be passed on to the next generation only if the mutated sector included the gamete-forming cells. The progeny from a uvt1 self-cross segregated into two distinct populations of plants of UV-sensitive and UV-resistant phenotypes in a ratio consistent with that of a dominant heritable characteristic caused by a lesion in a single Mendelian locus (data not shown). Figure 1A shows the effects of irradiating uvt1 and uvs seedlings with UVB. Under these conditions, we observed no signs of damage in uvt1 plants, whereas the typical severe necrosis was clearly evident in uvs. We subsequently crossed the uvt1 mutant to wild-type plants and allowed the resulting progeny to self. The F₂ population segregated with respect to UV resistance (Fig. 1B). We observed individuals displaying three distinct phenotypes, a uvt1-like phenotype (top), a uvs-like phenotype (bottom), and an intermediate UV tolerance phenotype (center) that later proved to have wild type-like UV tolerance. Because the original uvt1 mutant also carried a gl1 genotype, after the backcross the population is segregating with respect to both uvt1 and gl1 phenotypes. Thus, it is important to note that the UV-tolerant phenotype is observed in plants with or without trichomes (Fig. 1C). The same is true of the UV-sensitive and wild-type phenotypes (not shown). This suggests that the intrinsic differences in UV tolerance between these plants are not significantly affected by the presence or absence of trichomes in their leaves.

View larger version (72K): [in this window] [in a new window]   Figure 1.   Effect of UVB-radiation on UV-sensitive and UV-tolerant seedlings of Arabidopsis (A) and on a segregating population of uvt1/uvs seedlings (B and C). A, Twelve-day old uvs and uvt1 plants were irradiated with 0.17 W m2 UVB for 48 h and allowed to recover for 48 h under fluorescent lighting before being photographed. Representative individuals of UV-tolerant (uvt1) and UV-sensitive (uvs) plants are shown. B, Seventeen-day old seedlings from the F2 generation of a cross between wild-type and uvt1 plants were exposed to 0.17 W m2 UVB for 37 h and allowed to recover for 27 h under fluorescent lighting before being photographed. Representative individuals of each of three distinct UV tolerance phenotypes observed, highly tolerant (uvt1), intermediate tolerance (WT) ,and low tolerance (uvs), are shown. C, Additional seedlings of the same experiment are shown to depict the segregation of the glabrous mutation (gl1) among uvt1 plants. All images were digitized before printing.

From these populations segregating for uvt1, gl1, and uvs, we obtained uvt1 plants in a wild-type genetic background for both the gl1 and uvs loci. The level of UV tolerance of the uvt1/uvs double mutant, however, appears not to be significantly different from that of uvt1 plants in a wild-type background (data not shown). Figure 2 shows an assessment of the UV tolerance of mature plants from uvs, wild-type controls, and uvt1 mutants (wild-type background). As previously described (Lois and Buchanan, 1994), under identical UVB irradiation conditions, the uvs mutant showed more pronounced necrotic lesions than the wild type. The uvt1 mutant, however, showed significantly higher UVB tolerance than both uvs and the wild type. This is especially true under high intensity UVB irradiation conditions (high UV) that are well tolerated by uvt1 plants but that are lethal to both uvs and the wild type. From its extreme UVB tolerance, we can infer that uvt1 cannot be simply a revertant from uvs to wild-type genotype, but rather a new mutation that leads to hyper-resistance to UV. Furthermore, it appears that the elevated tolerance of uvt1 plants is evident at both early (Fig. 1) and late stages of development (Fig. 2) and in both young and older leaves. In UV-irradiated wild-type Arabidopsis, one normally observes a gradient of damage whereby older leaves are more damaged than younger leaves (Lois, 1994). This appears not to be the case in uvt1 where all leaves, including the oldest leaves, are highly tolerant to UV radiation.

View larger version (93K): [in this window] [in a new window]   Figure 2.   Effect of different regimes of UVB irradiation on mature wild-type and mutant Arabidopsis plants. One-month-old uvt1, wild-type (WT), and uvs plants were exposed to 0.10 W m2 UVB (mid UV), 0.18 W m2 UVB (high UV), or kept under fluorescent lighting (no UV) for 3 d and allowed to recover for 3 d under fluorescent lighting before being photographed. This image was digitized before printing.

Increased UV Tolerance and Dark-Green Phenotypes in uvt1 Are Likely Due to Constitutively Elevated Flavonoid Accumulation

The uvt1 mutation leads to different phenotypes that may or may not be related to the effects it has on UV tolerance. These include significantly shortened hypocotyls, a slower rate of growth, and delayed flowering with respect to wild-type plants. Some of these changes suggest that the defect in uvt1 may involve components of the light signal transduction. On the other hand, uvt1 seedlings show a phototropic response indistinguishable from the wild type (data not shown). As can be observed in Figure 2, leaves from uvt1 plants also appear to be significantly darker than those in wild-type or uvs plants, especially in the petioles and midribs of UV-exposed plants, which show a purple coloration. These observations indicated that the increased UV tolerance of uvt1 might be related to increased pigmentation. Therefore, we investigated the possibility that the darker coloration and UV-tolerant phenotypes may be due to the same genetic lesion and would therefore segregate together. After scoring and marking 586 dark plants, according to their light and dark coloration, these plants were exposed to UV radiation and analyzed for their sensitivity to UV. As expected, almost all plants showed tolerance to UV. Only three among the 586 plants marked as dark were found to show a UV-sensitive phenotype. Because we cannot rule out that these three plants may have been erroneously marked as dark, from this data we conclude that the two phenotypes are very tightly linked and may even be caused by the same genetic lesion.

A purple coloration in leaf tissues is commonly the result of an accumulation of the pigmented flavonoids, anthocyanins (Holton and Cornish, 1995). It seemed likely then that an increased level of flavonoids in the uvt1 mutant might be responsible for its dark coloration. By the same token, since flavonoids and other UV-absorbing phenolics have been postulated as important UV defense mechanisms (Li et al., 1993; Lois, 1994; Lois and Buchanan, 1994; Landry et al., 1995; Mazza et al., 2000), it appeared possible that the increased UV resistance of uvt1 may be due to the increase in UV-absorbing compounds. To address these questions, we investigated if the levels of these pigments were higher in the uvt1 mutant than in uvs and wild-type controls. Normally, wild-type plants accumulate low levels of UV-absorbing pigments, and these levels increase significantly upon UVB exposure. The absorption spectra in Figure 3 are an extension of data presented previously by Lois and Buchanan (1994) and show the normal UV-mediated increase in UV absorption in the region between 260 and 350 nm in wild-type controls, as well as the minimal increase in UV absorption in this region in uvs when plants are exposed to UVB. In contrast, the absorption spectrum of uvt1 shows elevated levels of UV-absorbing pigments prior to UV irradiation. A similar result was observed when data for extract absorption at 330 nm (the peak of absorption in Fig. 3) from three independent experiments were pooled (Fig. 4). Again, we observe the normal UV-mediated increase in absorption at 330 nm in the wild type, and the minimal increase in uvs. We also see significantly elevated levels of UV-absorbing compounds in nonexposed uvt1 when compared with uvs and wild-type controls. We conclude from these data that before UV exposure the levels of UV-absorbing pigments in uvt1 tissue are significantly higher than those in uvs and wild type, and that they are comparable to the levels seen in wild type after UV exposure. This in turn indicates that the cause of the increased UV tolerance of uvt1 might be mediated by the increased accumulation of UV-absorbing pigments. This is supported by absorption studies in the progeny of a selfing cross of a uvt1 heterozygote. When analyzing extracts from over 50 individual UV-tolerant and UV-sensitive plants from the F₁ generation we obtained in all cases, low absorption value at 330 nm (A₃₃₀) for individual UV-sensitive plants and high levels for uvt1 plants (data not shown). Therefore, the simplest explanation for the data presented is that the high level of pigment absorption at 330 nm is the cause of the elevated resistance of uvt1 to UV radiation. In this light, the fact that uvt1 plants show constitutive accumulation of UV-absorbing pigments before exposure to UV might explain why this mutant is even more UV resistant than the wild type. The high level of UV-absorbing pigments in uvt1 at the time of initial exposure to UV may protect uvt1 plants from early UV-mediated damage. On wild-type Arabidopsis plants, detectable increases in flavonoid accumulation begin only after about 7 h of UV exposure (Lois, 1994). Thus, damaging levels of UV radiation may be able to penetrate (and damage) the leaf for some time until the accumulation of UV-screening compounds is above a threshold of effective protection.

View larger version (18K): [in this window] [in a new window]   Figure 3.   Changes in the absorption spectra of pigment extracts from uvs, uvt1, and wild type after exposure to UVB. Thirteen-day-old wild type (WT), uvs, and uvt1 plants were exposed to fluorescent lighting with (+) or without () additional 0.15 W m2 UVB-radiation for 21 h. Extracts were prepared from equal amounts of tissue in 80% (v/v) ethanol. This figure represents an extension of previously published work (Lois and Buchanan, 1994)

View larger version (20K): [in this window] [in a new window]   Figure 4.   Level of UV-absorbing pigments in mutant and wild-type Arabidopsis plants. Pigments were extracted in 80% (v/v) ethanol from aerial tissues of 13-d-old plants before () and after (+) 21 h exposure to 0.15 W m2 UVB -radiation. The height of the bars represents the mean value of A330 per milligrams tissue of three independent experiments. Error bars correspond to one SD. Lines below bars depict relevant cases of pairs of bars for which there is a statistically significant difference between their means (non-paired Student's t test, P < 0.05).

In the absorption spectra in Figure 3, one can also observe higher levels of absorption for the peaks in the 400- to 500-nm and the 650-nm regions. Because these peaks correspond to the absorption by carotenoids and chlorophyll, it would appear that uvt1 might accumulate higher levels of these pigments in addition to the pigments absorbing in the 330-nm range. Pooling the data from three independent experiments using a larger number of plants, however, we have been unable to show that the elevated levels of carotenoids and chlorophyll between uvt1 and wild-type plants are statistically significant (data not shown).

We have hypothesized above that the dark-green phenotype characteristic of uvt1 (Fig. 2) may be due to the accumulation of anthocyanins. To test this, we analyzed the absorption at the characteristic maxima at 530 nm of anthocyanins in acidified methanol extracts. This method allows specific quantification of anthocyanins as opposed to A_330, which reflects absorbance by a variety of phenolics including anthocyanins. In Figure 5, we can see a significantly higher level of 530 nm absorption in uvt1 than in wild-type controls and uvs. This leads us to conclude that uvt1 accumulates higher levels of anthocyanins in the absence of UV exposure and that this may be the cause, at least in part, for the darker coloration in the leaves and petioles of these plants (see Fig. 2).

View larger version (33K): [in this window] [in a new window]   Figure 5.   Anthocyanin levels in mutant and wild-type Arabidopsis plants. Anthocyanins were extracted in acidified methanol from leaves of 36-d-old wild-type control (WT) and uvs plants and from UV-tolerant individuals (uvt1). The height of the bars represents the mean value of A530 per milligrams tissue of three independent experiments. Error bars correspond to one SD. There is a statistically significant difference between the uvt1 value and each of the other two values (non-paired Student's t test, P < 0.05).

Because both flavonoids and sinapates absorb in the UVB range and have been implicated in UV protection (Landry et al., 1995), we set out to determine if the increased UVB absorption in uvt1 could be mediated by an increased accumulation of sinapates as well as flavonoids. HPLC analysis of pigment extracts (Fig. 6) showed that several compounds (the most prominent peaks in the A₃₃₀ chromatogram) display significantly elevated levels in uvt1 when compared with wild-type plants. Fluorescence monitoring of the eluate showed that one of the major peaks of absorbance (arrow) corresponds to a highly fluorescent compound that accumulates at a higher level in uvt1 plants. Chapple et al. (1992) showed that these highly fluorescent compounds in Arabidopsis correspond to sinapates. Based on this and on our observations that extracts from fah1, an Arabidopsis mutant lacking sinapates, do not display that prominent peak of fluorescence (data not shown), we conclude that uvt1 has elevated levels of one of the sinapates. This is interesting because it indicates that the effects of the uvt1 lesion involve two biosynthetic pathways, flavonoid and sinapate, and suggests that it may affect a common regulator of these pathways.

View larger version (25K): [in this window] [in a new window]   Figure 6.   HPLC separation of UV-absorbing compounds from leaf tissues of uvt1 and wild-type plants. Pigments extracted from 16-d-old plants in 70% (v/v) methanol were separated by HPLC on a C18 column. The spectra represent the elution profiles between 9 and 19 min (the region where the most prominent peaks eluted) as monitored by spectrophotometry (A330, thick lines) and fluorescence (thin lines). The arrow indicates a peak corresponding to a highly fluorescent sinapate.

Because we detected the accumulation of UVB-absorbing compounds in uvt1 using leaf extracts, the possibility existed that these pigments may not be distributed homogeneously in the leaf and thus not really function as a protective filter for the whole leaf surface. This is especially true in the case of anthocyanins because of their localized accumulation in petioles and leaf midrib. If all UV-absorbing compounds were non-homogeneously distributed, their accumulation would not result in protection to most of the leaf surface. To address this question, we measured the capacity to absorb UV at different places on the lamina of different leaves, excluding the midrib in our mutants and in control plants. More specifically, we determined the amount of UV radiation transmitted through uvt1 leaves and compared it with that transmitted by uvs and control wild-type leaves. Figure 7 shows the spectra of light transmitted through the different types of leaves and Figure 8 shows the level of transmittance of UVB radiation through these leaves. In both figures, it is clear that highest and lowest transmission of UVB occurred in the cases of uvs and uvt1 leaves, respectively, whereas intermediate transmission was seen in wild-type leaves. Therefore, a one-to-one correlation exists between high tolerance to UV and leaf absorption of UVB radiation. Furthermore, despite some variation in the percent transmission between different places on the lamina or between different leaves, in all cases uvt1 leaves transmitted less UVB than either uvs or wild-type controls. Thus, the distribution of UV-absorbing compounds in these leaves is homogeneous enough for these pigments to filter UV differentially and thus protect the whole leaf area from UV damage. As a consequence, the tight correlation between in vivo leaf absorption in the UVB range and the overall resistance of the plant to UVB radiation suggest an important role for UV-absorbing compounds in the tolerance of uvt1 to UVB. In summary, uvt1 plants display an extreme tolerance to UV radiation levels that would kill the wild type, which may be explained by the increased levels of UV-absorbing compounds deposited in their leaves.

View larger version (26K): [in this window] [in a new window]   Figure 7.   Spectra of UV light transmitted through leaves from wild-type and mutant Arabidopsis plants. Irradiance spectra were obtained from incident light (Lamp) and light transmitted through leaves from 7-week-old wild-type (WT), uvs, and uvt1 plants exposed to UVB radiation. The left ordinate corresponds to the spectra of UV transmitted through the leaves and the right ordinate to the incident light spectrum.

View larger version (22K): [in this window] [in a new window]   Figure 8.   Relative transmission of UVB light through wild-type and mutant Arabidopsis plants. Irradiance spectra were obtained from incident light (Lamp) and light transmitted through leaves from 7-week-old wild-type (WT), uvs, and uvt1 plants exposed to UVB radiation. The height of the bars indicate the averages of percent transmission measurements for wavelengths between 290 and 320 nm taken in three or four places in each leaf lamina from at least three different leaves from uvs, wild-type, and uvt11 plants. Error bars correspond to one SD. The numbers above each bar represent the full range of percent transmittance values for each plant type.

Chalcone Synthase (CHS) mRNA Levels in uvt1 Are Constitutively Elevated

Thus far, our results show that uvt1 accumulates higher levels of UV-absorbing compounds in plants that have not been exposed to the stimulus of UV radiation. We have shown that some of these compounds are pigmented flavonoids and other phenolics, and that their excessive accumulation is responsible, at least in part, for the darker coloration of uvt1. As a first step in defining the mechanism underlying the changes in pigment accumulation, we wanted to determine if this elevated constitutive accumulation of phenolics in uvt1 was due to changes in gene expression. To this end, using RNA-blot analysis, we measured the mRNA levels of CHS, the gene encoding the enzyme that catalyzes the first committed step to flavonoid biosynthesis. The results in Figure 9 show a significantly higher level of CHS mRNA accumulation, before exposure to UV, in uvt1 than in wild type (wild type/UV and uvt1/UV). In fact, we were unable to detect any CHS transcripts in wild-type plants not exposed to UV even after a much longer autoradiographic exposure of the blot (data not shown). Thus, uvt1 displays elevated levels of expression of CHS mRNA and accumulation of flavonoids before exposure to UV. Therefore, it is likely that the increased level of pigment accumulation observed in these plants is the result of altered gene expression. The fact that uvt1 plants display increased accumulation of both flavonoids and sinapates argues against the possibility that the CHS gene may be the mutated locus in uvt1 and suggests that the defect may be at a gene involved in the regulation of more than one biosynthetic pathway.

View larger version (105K): [in this window] [in a new window]   Figure 9.   Accumulation of CHS mRNA in mutant and wild-type Arabidopsis. Northern blot from total RNA samples isolated from aerial tissues of 13-d-old wild-type, uvt11, and uvs plants not exposed (UV) and exposed (+UV) to 0.15 W m2 UVB radiation for 21 h, hybridized with a CHS gene probe. The upper image shows the rRNA stained with methylene blue. This image was digitized before printing.

Because CHS gene expression is regulated by various stimuli, including UV light (Tsukaya et al., 1991; Fuglevand et al., 1996; Ni et al., 1996; van Eldik et al., 1997; Hartmann et al., 1998; Kubasek et al., 1998), it was important to investigate if the constitutive expression of CHS in uvt1 was due to a defect in the UV-mediated regulation of the CHS gene. The data in Figure 9 (lane 4) show that CHS gene mRNA levels are significantly elevated after UVB exposure and, therefore, that the UVB-mediated regulation of CHS gene expression is still operating in uvt1. The fact that the UVB signal transduction pathway can still function argues that a full deregulation of the UV signal transduction pathway leading to increased CHS expression may not be the problem in uvt1. This indicates that uvt1 may have a partially deregulated UV signal transduction pathway, or alternatively, that the genetic lesion in uvt1 is affecting one of the other signaling pathways that control CHS gene expression. The fact that uvt1 plants show shortened hypocotyls and delayed flowering however suggests that the uvt1 defect might be in one of the components of the white light regulation of CHS, perhaps the induction of CHS mediated by high irradiance (Feinbaum et al., 1991).

As was the case with UV-absorbing pigment accumulation, we also found a tight correlation between high CHS expression and increased resistance to UV. Analysis of CHS expression in pools of over 500 plants from the UV-sensitive or the UV-tolerant populations from a selfing cross of a uvt1 heterozygote resulted in high level of CHS mRNA both before and after UV irradiation in the UV-resistant population, whereas undetectable levels were observed in the UV-sensitive pools (data not shown). Thus, the UVB tolerance of uvt1 correlates with deregulated gene expression and may even be a consequence of it.

Comparison of uvt1 with Other Mutants That Show Altered CHS Expression

Jackson et al. (1995) described a mutant, icx1, which like uvt1 shows constitutively elevated CHS gene expression. The icx1 mutation differs from uvt1, however, in that it segregates as a single recessive lesion and that it leads to other phenotypic characteristics. For example, uvt1 does not show the narrower young leaves or the decreased number of trichomes observed in icx1 leaf. Therefore, it is unlikely for these two mutations to represent the same genetic locus.

The cop/det/fus family of Arabidopsis mutants has been described as failing to repress light-activated genes (Wei and Deng, 1996). One of the characteristics of the cop, det, and fus mutants is that they display deregulated expression of CHS and an excessive accumulation of anthocyanins in young seedlings and in mature seeds. Because the expression of CHS and anthocyanins levels are also altered in uvt1, we wanted to know if uvt1 displayed the short hypocotyls of dark grown fusca, det, and cop mutants. Therefore, we grew uvt1, wild-type, and uvs Arabidopsis seeds in darkness for 12 d and found that uvt1 seedlings displayed an etiolated phenotype, indistinguishable from that of wild-type controls or uvs seedlings (data not shown). We conclude that uvt1 does not belong to the same class of mutants as the cop/det/fus mutants and, therefore, that the alteration of CHS regulation and excessive accumulation of anthocyanins in uvt1 define a new class of mutant with marked tolerance to UV radiation.

A recently described mutant of Arabidopsis (pap1-D) generated by activation tagging (Borevitz et al., 2000) may shed some light on the molecular mechanisms underlying the uvt1 phenotype. This mutant overexpresses a MYB transcription factor and displays elevated pigmentation similar to that in uvt1. The phenotype of this mutant further resembles that of uvt1 in that it is caused by a single dominant lesion that leads to elevated CHS expression as well as enhanced flavonoid and sinapate accumulation. These similarities in the phenotypes of both mutants suggest that both mutations might affect similar if not identical targets. However, these two mutants differ from each other in their pigmentation pattern. The enhanced pigmentation of uvt1 is restricted to leaves and stems, whereas in pap1-D it affects the whole plant including roots, sepals, anthers, and carpels in addition to stems and leaves. These differences do not appear to be merely a quantitative difference in pigment accumulation between the two mutants, but a differential effect of each mutation in different organs. For example, the level of pigmentation in roots and stems are comparable in pap1-D plants, whereas no pigmentation is detected in the roots of uvt1 plants that show deep stem pigmentation. This organ specificity in uvt1 pigmentation and the lack thereof in pap1-D then argues that these two mutations might be affecting different regulatory targets. One alternative is that these different targets may be part of the same gene. Another alternative is that the uvt1 mutation affects a different regulator of phenylpropanoid biosynthesis and that this leads to similar pigmentation effects but with different organ specificity. Preliminary mapping data of the uvt1 mutation has ruled out a close linkage to pap1-D or to the related PAP2 gene (L. Procter and R. Lois, unpublished data) lending support to the latter alternative.

In conclusion, we have presented evidence that a dominant mutation in Arabidopsis, leading to altered expression of the CHS gene and to increases in accumulation of UV-absorbing compounds, confers a high level of tolerance to UV radiation in these plants. The uvt1 mutant may prove to be an important tool not just to help define in more exact terms the role of UV-absorbing pigments in the overall tolerance to UV radiation but also to help elucidate the relative contribution of different mechanisms of protection against UV. For example, we expect that studies of the combined effects of the uvt1 mutation in the background of some of the already known mutations leading to increased UV sensitivity (like fah1 and the different DNA repair mutants) may help us understand the way in which different mechanisms of protection interact and how they contribute to the overall tolerance of a plant to UV radiation. Finally, we expect that uvt1 in combination with other pigmentation mutants may help understand the molecular mechanisms regulating phenylpropanoid biosynthesis in different organs.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Growth and UVB Treatment

Arabidopsis (Columbia/glabrous-1 mutant) were grown in potting soil (Home Pro, Chico, CA, or Sunshine Special Blend, McConkey Co., Sumner, WA) under white fluorescent light (between 90-150 µEi m² s¹ photosynthetically active radiation) in 12-h-light/-dark cycles at 23°C. For UV treatments, plants were transferred to continuous lighting of white fluorescent light (between 30-70 µEi m² s¹ photosynthetically active radiation), supplemented with 0.10 to 0.18 W m² UVB light (from one fluorescent F40 UVB bulb, Phillips, Holland; or an FS40T12 UVB bulb, Light Sources, Milford CT). Nonexposed control plants were either covered with a 0.13-mm mylar filter that blocks wavelengths of light below 320 nm, or maintained in a different incubator in the absence of UV under similar white light conditions.

For dark-growth experiments, seed samples were surface sterilized in a solution of 0.1% (v/v) Triton X-100 and 10% (v/v) bleach in water for 10 min and rinsed three times in vast excess of sterile double distilled water for 20 min. Sterilized seeds were sown on 4 g L¹ agar containing 2% (w/v) Suc and Murashige and Skoog basal salts (Gibco Life Technologies, Inc., Grand Island, NY).

Mutagenesis

Seeds from the Arabidopsis uvs mutant (Lois and Buchanan, 1994; Columbia/glabrous-1 background), were exposed to 0.20% (v/v) EMS (Sigma, St. Louis) for 12 to 16 h. Seeds were rinsed twice in vast excess double distilled water, soaked in double-distilled water for 24 h, and dried. Approximately 10,000 seeds were sown, grown for 2 weeks, and then exposed to 0.15 W m² UV radiation for 2 d or until plants showed signs of visible damage. Prospective UV-tolerant mutants (uvt) were selected based on a lack of visible damage and a healthier overall appearance compared with the rest of the plants. Selected plants were transplanted, allowed to recover under white fluorescent light, and self-crossed.

It should be noted that because of the uvs genetic background, we chose Columbia/gl-1 plants (wild type for the uvs and uvt1 loci) as the most appropriate controls for all experiments. We refer to these plants as wild-type controls throughout the paper unless it is specifically stated otherwise.

Plants from a self-cross of uvt1 showing the UV-tolerant phenotype (in a uvs and gl1 background) were used as female recipient for a cross to Columbia plants wild type for both uvs and gl1. From these populations segregating for uvt1, gl1,and uvs, we were able to obtain uvt1 plants in a wild-type genetic background for both the gl1 and uvs loci.

Spectrophotometric Analysis

Pigment extracts were prepared by incubating 0.1 g of leaf tissue in 500 µL 80% (v/v) ethanol at 65°C for 10 to 20 min. Absorption spectra of undiluted extracts were obtained using a diode array spectrophotometer (Beckman DU 7400, Fullerton, CA). In those cases where the total aerial tissues were used for pigment extraction, absorbance values are presented as a ratio between A₃₃₀ to milligrams of tissue.

Anthocyanin Analysis

Leaf samples were ground in liquid nitrogen and extracted overnight in 1.0 mL 1% (v/v) HCl, in methanol. After extraction, 1.0 mL of water and 2.0 mL of chloroform were added, and the sample was vortexed and centrifuged at 3,000g for 2 min. Relative anthocyanin levels were determined by measuring A₅₃₀ of the upper aqueous phase (Mancinelli et al., 1988) using a Beckman DU 7400 spectrophotometer.

HPLC Analysis

Leaf samples (0.4g) were ground in liquid nitrogen and extracted in a solution of 1.6 mL 70% (v/v) methanol and 1% (v/v) HCl. The extract was clarified by centrifugation at 5000g for 3 min and filtered through a 0.22-µm membrane before injection into a 4.6- × 250-mm Partisphere RTF C18 column (Whatman, Clifton, NJ). HPLC chromatography was carried out using a Shimadzu HPLC system with a UV spectrophotometric detector monitoring A₃₃₀ and a Kratos FS970 LC-Fluorometer (Kratos, Chestnut Ridge, NY) using 350-nm excitation and a 370-nm emission cutoff filter. Chromatography was based on a modification of procedure described by Graham (1991) and Li et al. (1993). Chromatography essentially was at 1.0 mL min¹ using increasing concentrations of HPLC grade acetonitrile mixed with 0.01% (v/v) glacial acetic acid in water. After injection, acetonitrile was increased linearly from 1% to 32% (v/v) in 5 min, maintained at 32% (v/v) for 5 min, increased linearly to 34% (v/v) in 5 min, to 37% (v/v) in 2 min, to 40% (v/v) in 3 min, and to 99% (v/v) and maintained at 99% (v/v) for 5 min.

Radiometric Analysis

Irradiance and transmission measurements were performed with a CCD diode array detector (model Instaspec IV, Oriel, Stratford, CT) connected to a Spectrograph/Monochromator (Oriel model Multispec 1/8M) with 0.2-nm resolution using a 1,200 lines mm¹ grating and a 120-µm slit. The input to the Monochromator was either a 15.3-cm integrating sphere (Oriel model 70451), for most irradiance determinations, or a 2-m, small slit, high-grade fused silica fiber optic cable (Oriel model 77532) for the measurements of the transmission spectra through intact leaves. Comparative measurements of UV irradiance were routinely performed with a hand-held UV light meter with a UVB probe (Oriel Goldilux model 70215/70221). All three instruments were calibrated with a standard of spectral irradiance lamp (Oriel model 63361) following the manufacturer's instructions. In the transmission experiments, to avoid differential reflection due to trichomes, all plants used had a glabrous phenotype.

CHS mRNA Analysis

For each treatment, approximately 0.1 g leaf tissue was harvested, frozen in liquid nitrogen, and stored at 80°C until extraction. Frozen tissue was ground using a mortar and pestle under liquid nitrogen and total RNA was extracted using an RNeasy kit (Qiagen, Santa Clarita, CA). RNA concentration was determined by A₂₆₀. Analysis by RNA blotting was carried out as a modification of the methods described by Sambrook et al. (1989). For electrophoresis, RNA samples were prepared by mixing 3 µL of RNA with 17 µL RNA loading buffer (57% [v/v] formamide, 20% [v/v] formaldehyde, 0.025% [w/v] bromphenol blue, 0.025% [w/v] xylene cyanole FF, 3% [v/v] glycerol, 5 mM NaH₂PO₄, 5 mM Na₂HPO₄, 5 mM sodium acetate, and 1 mM EDTA), incubated at 65°C for 10 min, then cooled on ice before loading. Five micrograms total RNA was separated on a 1.2% (w/v) agarose 6% (v/v) formaldehyde denaturing gel in 1× phosphate running buffer (5 mM NaH₂PO₄, 5 mM Na₂HPO₄, 5 mM sodium acetate, and 1 mM EDTA). Samples were electrophoresed at 70 V for 3 to 4 h, transferred by capillary elution to a Gene-Screen nylon membrane (NEN Life Science, Boston) in 25 mM NaH₂PO₄ (pH 6.5) for 16 h, cross-linked at 1,200 µW cm² with a CL-1000 UV cross-linker (Ultra-Violet Products, Inc., San Gabriel, CA), and stained with 0.04% (w/v) methylene blue in 0.5 M sodium acetate for 5 min to visualize rRNA.

Probe Synthesis

Primers 5'-CTGACTACTACTTCCGCATC-3' and 5'-GTGATCTCAGAGCAGACAAC-3' (Universal DNA, Inc., Tigard, OR) were used to amplify a 453-bp fragment from the CHS gene from Arabidopsis cDNA to use as a probe. For cDNA synthesis, 1.5 µg total leaf RNA, pre-incubated at 65°C for 10 min and cooled on ice, was incubated for 1 h at 37°C with 13.3 units Moloney Murine Leukemia Virus (M-MLV) reverse transcriptase (Promega, Madison, WI); 1.5 µM Oligo dT 15 mer primer; 0.5 µM each of dATP, dCTP, dGTP, and dTTP (Boehringer Mannheim, Indianapolis); 30 units of RNase inhibitor (Stratagene, LaJolla, CA); and 3 µL 10× M-MLV reverse transcriptase buffer (Promega) in a total volume of 30 µL. Synthesis of cDNA was terminated by incubating at 70°C for 10 min and samples were kept at 20°C until used for probe amplification. Amplification reactions were carried out using approximately 1 µg of cDNA, with 0.5 unit of Taq polymerase (Promega); 20 mM each of dATP, dCTP, dGTP, and dTTP (Boehringer Mannheim); 50 µM each of CHS primer; 10 µL of 10× Taq polymerase buffer (Promega); and 25 mM magnesium chloride in a 100-µL reaction. Amplification reactions consisted of 1 cycle at 94°C for 3 min, 30 cycles of 55°C for 1.5 min, 72°C for 2 min, and 94°C for 2 min, and a final 72°C incubation for 2 min.

Amplified fragments were purified after electrophoresis on a 1.5% (w/v) agarose gel containing 0.5 µg mL¹ ethidium bromide (Sigma), and band excision using QIAquick Gel Extraction Kit (Qiagen). A 25-ng sample of amplified CHS probe was labeled using ³²P-labeled dCTP by random priming (Boehringer Mannheim) following the manufacturer's recommendations. Unincorporated nucleotides were removed from the radiolabled probe by spinning through a 1-mL syringe filled with Sepharose CL-6B (Sigma). Prehybridization was carried out in a hybridization oven at 42°C in 15 mL 0.75 M NaCl, 50 mM NaH₂PO_4, 5 mM EDTA, 50% (v/v) formamide (Fluka Chemical Corp., Ronkonkoma, NY), 0.1% (w/v) SDS, 0.5% (w/v) nonfat dry milk, and 1.7 µg mL¹ denatured calf thymus DNA for 20 to 30 min. Hybridization was overnight and included 2 × 10⁷ cpm of denatured probe in 15-mL prehybridization buffer. Membranes were washed three times in approximately 300 mL of 0.3 M NaCl, 20 mM NaH₂PO_4, 2 mM EDTA, and 0.1% (w/v) SDS at 60°C for 25 min. The membrane was wrapped with plastic wrap, and autoradiographed with Kodak Imaging film (Blue XB-1) with Kodak X-Omatic regular intensifying screens from overnight to 3 weeks at 80°C depending on the signal.