--Ann E. Stapleton and Virginia Walbot

Solar radiation is both a source of light for photosynthesis and UV that can damage plant DNA. Maintaining the integrity of DNA is of critical importance to all organisms. Plants use both DNA repair and various shielding strategies to minimize DNA damage; relatively little is known about the biochemical basis of irradiation-induced DNA damage in plants (McLennan, AG, DNA Replication in Plants, p. 135, 1987).

UV light is generally classified into 3 subgroups: UV-A (320-390nm), UV-B (280-320nm) and UV-C (less than 280nm). The absorption spectrum of DNA includes wavelengths from 240 to 310nm; the level of solar UV that reaches the surface of the earth is high in the UV-A region of the spectrum, decreases sharply in the UV-B range, and drops to nearly zero by 290nm (McLennan, DNA Replication in Plants, p. 135, 1987). Most studies of the effects of UV light have used wavelengths of 254nm, in the UV-C region, although this wavelength is extremely rare in sunlight.

UV-C and UV-B produce DNA damage primarily via the formation of cyclobutane pyrimidine dimers (PDs). Pyrimidine dimers are 80%-90% of the UV-light induced DNA photoproducts, with most of the rest being pyrimidine(6,4)pyrimidone (Franklin, WA and Haseltine, WA, Mut. Res. 165:1, 1986). The epidermis of plants absorbs 95%-99% of incoming UV light; flavonoid compounds (such as anthocyanins) and cuticular waxes are the agents of UV absorption (McLennan, DNA Replication in Plants, p. 135, 1987). Because anthocyanins and other flavonoids absorb light in the UV-B range, it is commonly suggested that these compounds shield plant DNA from damage. However, this hypothesis has not previously been tested.

We extracted anthocyanins from the purple maize line K8 (an inbred in the W23 background) which contains all the structural genes required to produce anthocyanins and the genes B and Pl that confer anthocyanin expression on nearly every tissue of the plant. For An1 prep 5g husk tissue was frozen in liquid nitrogen, ground to a powder, and mixed with 1% HCl in methanol. After incubation overnight at 4 C and 3 changes of extraction buffer the final volume was 45ml; this was filtered through Whatman #1 to remove particulate matter and stored at 4 C. For An2 prep 5g husk tissue was finely chopped and extracted with 45ml 8:1:1 methanol:acetic acid:water. The resulting solution was passed over a Sephadex LH20 column to remove sugars and stored at 4 C. Quercetin (Sigma) was used from a stock solution of 10mg/ml in methanol.

Genomic DNA was prepared from BMS cell cultures and irradiated. To test the ability of flavonoids to shield DNA, anthocyanin extracts or quercetin solutions were sealed between quartz plates with beeswax for irradiations. For UV-C irradiation we used one GL-15 germicidal lamp positioned 12cm above the sample, with an output of 30 J/m2/sec. For UV-B we used a TR302 transilluminator with filter, output rated as 8 J/m2/sec and a simulator of solar UV-B, output rated as 1 J/m2/sec. We specifically nicked the DNA near the pyrimidine dimers using T4 endonuclease V, which was kindly supplied by Dr. P. C. Hanawalt, PC. DNA was treated for 15 minutes at 37 degrees in 10mM Tris pH 8.0, 100nM NaCl, 10nM EDTA, 1mg/ml BSA, 2 microliters T4 endo V (lot 25). Alkaline loading dye was added to DNA samples to denature them; DNA was size fractionated on 0.6% alkaline gels with running buffer composed of 30mM NaOH, 1mM EDTA. The gels were neutralized, stained with EtBr, photographed, and the amount of unnicked DNA quantitated by densitometry.

The anthocyanins that were isolated from purple maize husk tissue by either of the two extraction methods absorb in the UV range. When these anthocyanins (at a concentration approximately equal to 1/3 the concentration in K8 plants) are interposed between DNA and a UV source they protect the DNA from PD formation (and thus from nicking by T4 endo V) (Table 1). This protection decreases in proportion to increase in dose. Quercetin also protects DNA from damage.

Table 1. Percent of control band densitometer trace of BMS genomic DNA irradiated with UV-B solar simulator (with and without anthocyanin shielding).
 

	0 min		40 min				120 min
T4endoV:	-	+	+	+	+	+	+	+	+	+
Antho:				An1	An2	Q		An1	An2	Q
Expt. 1	79± 1	100	17	86	68	50	0	35	30	28
Expt. 2	104 ±4.3	100	23	83	83	56	11	69	43	21

Control (0 min) and two doses of UV-B were used with the same anthocyanin shielding (An1 and An2) and quercetin shielding (Q). Columns without An1 or An2 or Q headings were unshielded. T4 endo V was omitted in the (-) lane; two or three identical such lanes were run in each experiment to measure loading variation; the ± indicates the variation in percent in these two or three lanes. All data were normalized to the 0 min, plus T4 lane.

When sections of leaf tissue from purple and green plants are irradiated with high doses of UV-C or UV-B and the level of PDs assayed as described above, we find that the purple tissue shows less damage than green tissue. This demonstrates that anthocyanins also protect DNA from UV damage in vivo.

Return to the MNL 66 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage