Inhibition of DNA Polymerase λ , a DNA Repair Enzyme , and Anti-Inflammation : Chemical Knockout Analysis for DNA Polymerase λ Using Curcumin Derivatives

Yoshiyuki Mizushina1,2, Masayuki Nishida3, Takeshi Azuma3 and Masaru Yoshida3,4,5 1Laboratory of Food & Nutritional Sciences Department of Nutritional Science, Kobe-Gakuin University 2Cooperative Research Center of Life Sciences Kobe-Gakuin University 3Division of Gastroenterology, Department of Internal Medicine Kobe University Graduate School of Medicine 4The Integrated Center for Mass Spectrometry Kobe University Graduate School of Medicine 5Division of Metabolomics Research Kobe University Graduate School of Medicine Japan


Effect of curcumin (2) and monoacetyl-curcumin (13) on the activities of pols and other DNA metabolic enzymes
Curcumin (2) and monoacetyl-curcumin (13) were effective at inhibiting human pol λ activity, and the inhibition was dose-dependent with 50% inhibition observed at a concentration of 7.0 and 3.9 µM, respectively (Table 1).These compounds had no influence on the activities of not only DNA replicative pols such as calf pol α, human pol δ and human pol ε, or mitochondrial DNA replicative pols such as human pol γ, but also DNA repair-related pols such as rat pol β, human pols η, ι and κ.It is interesting that these compounds had no affect on the activity of pol β, because pols β and λ both belong to the X-family of pols, and the three-dimensional structure of pol β is thought to be highly similar to pol λ (Garcia- Diaz et al., 2004).Curcumin (2) and monoacetyl-curcumin (13) also had no inhibitory effect on cherry salmon (fish) pols α and δ, cauliflower (higher plant) pol α, prokaryotic pols such as the Klenow fragment of E. coli pol I, Taq pol and T4 pol, and other DNA metabolic enzymes such as calf DNA primase of pol α, human immunodeficiency virus type-1 (HIV-1) reverse transcriptase, T7 RNA polymerase, T4 polynucleotide kinase and bovine deoxyribonuclease I. Therefore, these phenolic compounds were specific inhibitors of human pol λ among the pols and DNA metabolic enzymes tested.Petasiphenol (1) also selectively inhibited the activity of eukaryotic pol λ such as curcumin (2) and monoacetyl-curcumin (13) (Mizushina et al., 2002).
When activated DNA (i.e., bovine deoxyribonuclease I-treated DNA) and dNTP were used as the DNA template-primer and nucleotide substrate instead of synthesized DNA [poly(dA)/oligo(dT) 18 (A/T = 2/1)] and dTTP, respectively, the inhibitory effects of these compounds did not change.

Effect of curcumin derivatives on TPA-induced anti-inflammatory activity
As mentioned in the Introduction, TPA is known to cause inflammation and is commonly used in screens for anti-inflammatory agents (Fujiki & Sugimura, 1987).Curcumin (2) is known as an anti-TPA-induced inflammatory compound (Ammon & Wahl, 1991), but the other agents (compounds 1 and 3-15) had not previously been tested for anti-TPA-induced inflammatory activity.
Fig. 4. Anti-inflammatory activity of curcumin derivatives (compounds 1-15) toward TPAinduced edema on mouse ear.Each compound (250 µg) was applied to one of the mouse ears and, after 30 min, TPA (0.5 µg) was applied to both ears.Edema was evaluated after 7 h.The anti-inflammatory activity (%) is expressed as the percentage reduction in edema as compared with the non-treated ear.Data are shown as the mean ± SE (n=5).
Using an inflammation test in mice, the anti-inflammatory activity of these compounds was examined.The application of TPA (0.5 µg) to a mouse ear induced edema with a 241% increase in the weight of the ear disk at 7 h after application.As expected, curcumin (2) inhibited this inflammation at an applied dose of at least 250 µg (inhibitory effect (IE) = 63%) (Fig. 4).Petasiphenol (1), which was purified from Japanese vegetable (Petasites japonicus), was also an anti-inflammatory agent, although its effect was a third weaker than that of curcumin (2).Thus, both petasiphenol (1) and curcumin (2) could be potent inhibitors of inflammation caused by TPA.Interestingly, other curcumin derivatives also caused a marked reduction in TPA-induced inflammation: notably, the anti-inflammatory effect of monoacetyl-curcumin (13) was stronger than that of curcumin (2) with an IE of 81%, indicating that this compound possesses strong anti-inflammatory activity.

Structure-activity relationship of curcumin derivatives
Pol λ inhibition had a significant correlation (correlation coefficient = 0.9608) with antiinflammatory activity, as shown by Fig. 3 and Fig. 4, which led us to speculate that TPAinduced inflammation may involve a process requiring pol λ, which is a DNA repair-related pol.Thus, to confirm whether there is a relationship between pol λ inhibition and antiinflammation, the inhibitory effects of the curcumin derivatives (compounds 1-15) on the two bio-activities were compared.
Considering the structure of monoacetyl-curcumin (13) (Fig. 5), the essential moieties of the structure for these activities might be: <1> two enone moieties, <2> one hydroxyl group at position C4', and <3> one acetoxy group at position C4".These moieties are specific to monoacetyl-curcumin (13); therefore, these moieties are likely to be involved in the activities of both pol λ inhibition and anti-inflammation. 3"

Inhibitory activity of curcumin (2) and monoacetyl-curcumin (13) against inflammatory responses in cultured cells
Next, because curcumin (2) and monoacetyl-curcumin (13) might be chemical knockout agents for DNA repair-related pol λ activity (Table 1), we used these compounds to investigate the anti-inflammatory mechanism of pol λ specific inhibitors in the murine macrophage cell line RAW264.7 treated with lipopolysaccharide (LPS or endotoxin), which stimulates macrophages to release inflammatory cytokines, interleukins (ILs) and tumor necrosis factor (TNF) (Hsu & Wen, 2002).RAW264.7 cells were seeded on a 12-well plate at 1x10 5 cells/well and incubated for 24 h.The cells were pre-treated with 10 or 50 µM curcumin (2) or monoacetyl-curcumin (13) for 30 min and then stimulated with 100 ng/mL of LPS.After 30 min or 24 h, the cell culture medium was collected to measure the levels of inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), nuclear factor-κB (NF-κB) and IκB.In RAW264.7 cells, cytotoxicity of these compounds at 50 µM was not observed (data not shown).
Next, the effect of monoacetyl-curcumin (13) on the expression level of pol λ protein in LPStreated RAW264.7 cells was investigated.Fig. 6B shows that these macrophages underwent a more than 3-fold increase in the expression of pol λ after LPS stimulation, but this increase was suppressed by 50 µM monoacetyl-curcumin (13).These results suggest that there is the positive correlation between inflammatory induction by LPS and pol λ expression; thus, not only the DNA polymerization activity but also the protein expression of DNA repair-related pol λ is likely to be important in inflammation.
Anti-oxidative activity has been reported to be linked to anti-inflammatory activity (Rahman et al., 2006).We therefore investigated the anti-oxidative activity of curcumin (2) and monoacetyl-curcumin (13) against the production of reactive oxygen species (ROS) induced by TNF-α.Measurement of intracellular ROS was performed according to the method of a previous report (Corda et al., 2001).In RAW264.7 cells, at 50 µM, the two compounds decreased the production of ROS by 50 ng/mL of TNF-α to 59.5% and 32.1%, respectively (Fig. 6E).These results suggest that both compounds possess anti-oxidative activity, but that monoacetyl-curcumin (13) has stronger activity than curcumin (2).

Inhibitory activity of curcumin (2) and monoacetyl-curcumin (13) against LPS-induced inflammation in vivo
To assess their anti-inflammatory effects in vivo, the inhibitory activity of curcumin (2) and monoacetyl-curcumin (13) against LPS-induced acute inflammation was investigated in mice (Fig. 7).As shown in Fig. 7A, treatment with 250 µg/kg (body weight, BW) of LPS increased the serum TNF-α level, and an oral injection of 100 mg/kg (BW) of monoacetylcurcumin (13) significantly decreased the LPS-induced production of TNF-α to 36%.By contrast, curcumin (2) had no effect.Next, the inhibitory effects of these compounds on nuclear translocation of NF-κB in the liver were examined.Fig. 7B shows that LPS caused translocation of NF-κB into the nucleus, and monoacetyl-curcumin (13) blocked this nuclear translocation.Notably, curcumin (2) also inhibited nuclear translocation of NF-κB even though it did not block TNF-α production.
The serum levels of curcumin (2) and monoacetyl-curcumin (13) 2 h after oral administration were measured in the mice by liquid-chromatography mass spectrometry.The serum concentrations were below the detection limit and, thus, were less than 0.3 nM for both curcumin (2) and monoacetyl-curcumin (13) (data not shown).It has been reported that curcumin (2) is poorly absorbed in the body (Anand et al., 2007).Thus, a lower concentration of monoacetyl-curcumin (13) than of curcumin (2) might be able to decrease the serum TNF-α level in mice treated with LPS.

Discussion
Inflammatory mediators, such as TPA and LPS, quickly stimulate ROS (Hsu & Wen, 2002), and ROS are known to mediate oxidative DNA damage.As shown in Fig. 8, DNA repair pols such as pol λ induce protein expression and increase DNA polymerization activity to repair the damaged DNA.Furthermore, we consider that pol λ might have a great effect on inflammatory responses, such as TNF-α production, NF-κB activation, secretion of cytokines [e.g.(IFNs) and interleukins (ILs) etc], tissue damage and cell death.The results summarized in this review suggest that inhibition of DNA repair by pol λ is related to anti-inflammatory pathways, and that pol λ-specific inhibitors such as monoacetyl-curcumin (13) might be chemotherapeutic drugs for inflammatory diseases.The detailed molecular mechanism underlying the correlation between DNA repair inhibition by pol λ and anti-inflammatory responses is not yet known; therefore, experiments with small interfering RNA (siRNA) targeting pol λ would help in further analyses.As mentioned above, eukaryotic cells reportedly contain 14 pol species belonging to four families (Friedberg et al., 2000;Takata et al., 2006).Among the X family of pols, pol λ has an unclear biochemical function, although it seems to work in a similar way to pol β (Garcia- Diaz et al., 2002).Pol β is involved in the short-patch base excision repair (BER) pathway (Matsumoto & Kim, 1995;Singhal & Wilson, 1993;Sobol et al., 1996), as well as playing an essential role in neural development (Sugo et al., 2000).Recently, pol λ was found to possess 5'-deoxyribose-5-phosphate (dRP) lyase activity, but not apurinic/apyrimidinic (AP) lyase activity (Garcia-Diaz et al., 2001).Pol λ is able to substitute for pol β during in vitro BER, suggesting that pol λ also participates in BER.Northern blot analysis indicated that transcripts of pol β are abundantly expressed in the testis, thymus and brain in rats (Hirose et al., 1989), whereas pol λ is efficiently transcribed mostly in the testis (Garcia-Diaz et al., 2000).Bertocci et al. reported that mice in which pol λ expression is knocked down are not only viable and fertile, but also display a normal hyper-mutation pattern (Bertocci et al., 2002).

Macrophages (RAW264.7 cells)
DNA damage As well as causing inflammation, TPA influences cell proliferation and has physiological effects on cells because it has tumor promoter activity (Nakamura et al., 1995).Therefore, anti-inflammatory agents are expected to suppress DNA replication/repair/recombination in nuclei in relation to the action of TPA.Because pol λ is a DNA repair-related pol (Garcia-Diaz et al., 2002), our finding -that the molecular target of curcumin derivatives as monoacetyl-curcumin ( 13) is pol λ -is in good agreement with this expected mechanism of anti-inflammatory agents.As a result, any inhibitor of DNA repair-related pol λ might also be an inflammatory suppressor.

Conclusion
This review summarizes data showing that a major anti-inflammatory food compound, curcumin (2), selectively inhibits the activity of pol λ among 9 species of mammalian pols tested.Monoacetyl-curcumin (13) was the strongest inhibitor of pol λ among the 13 chemically synthesized derivatives of curcumin (2), suggesting that monoacetyl-curcumin (13) is a potent candidate for a functional compound.In addition, the inhibitory effects of monoacetyl-curcumin (13) on inflammatory responses in comparison to those of curcumin (2) in vitro and in vivo were investigated.Monoacetyl-curcumin (13) suppressed NF-κB activation induced by LPS and TNF-α in RAW264.7 murine macrophages.Moreover,

Fig. 3 .
Fig.3.Inhibitory effect of curcumin derivatives (compounds 1-15) on the activities of mammalian pols.Each compound (10 µM) was incubated with calf pol α (B-family pol), human pol γ (A-family pol), human pol κ (Y-family pol) and human pol λ (X-family pol) (0.05 units each).Pol activity in the absence of the compound was taken as 100%, and the relative activity is shown.Data are shown as the mean ± SE (n=3).

Fig. 6 .
Fig.6.Inhibitory activities of curcumin (2) and monoacetyl-curcumin (13) against inflammatory responses in the cultured murine macrophage cell line RAW264.7.(a) RAW264.7 cells were pre-treated with 10 or 50 µM curcumin (2) or monoacetyl-curcumin (13) for 30 min and then incubated with 100 ng/mL of LPS for 24 h.The TNF-α level in the culture medium was measured by ELISA.Data are shown as the mean ± SE (n=4).(b) RAW264.7 cells were pre-treated with 50 µM monoacetyl-curcumin (13), and then incubated with 100 ng/mL of LPS for 30 min.The expression level of pol λ was evaluated by Western blot analysis.The intensity of each band was analyzed, and the values relative to nontreatment with LPS are represented at the lower edge of the image.(c and d) RAW264.7 cells were pre-treated with 50 µM curcumin (2) or monoacetyl-curcumin (13), and then incubated with 100 ng/mL of LPS for 30 min.Nuclear translocation of NF-κB p65 (c) and the phosphorylation of IκB (d) were evaluated by Western blot analysis.The intensity of each band was analyzed, and the values relative to treatment with LPS alone are represented at the lower edge of the image.(e) RAW264.7 cells were pre-treated with 50 µM curcumin (2) or monoacetyl-curcumin (13) for 30 min, and then treated with 50 ng/mL of TNF-α and 4 µM DCFH-DA for 30 min.The fluorescent intensity of DCF, which indicates ROS production, was measured as described in a previous report(Corda et al., 2001).Data are shown as the mean ± SE (n=4).

Fig. 7 .
Fig. 7.The inhibitory activity of curcumin (2) and monoacetyl-curcumin (13) against LPSinduced inflammation in vivo.Male 8-week-old C57BL/6 mice were given an oral dose of 100 mg/kg (BW) of curcumin (2) or monoacetyl-curcumin (13) dissolved in corn oil or 200 µL of corn oil as a vehicle control.After 2 h, the mice were intraperitoneally injected with 250 µg/kg (BW) of LPS dissolved in phosphate-buffered saline (PBS) or 200 µL of PBS as a vehicle control.After 1 h, the mice were killed.(a) The TNF-α level in serum was measured by ELISA.Data are shown as the mean ± SE (n=4).The treatment with corn oil and LPS (positive control) was taken as 100% (TNF-α level, 728 pg/mL) and that with corn oil and saline (negative control) as taken as 0% (TNF-α level, 32 pg/mL).(b) NF-κB p65 in the nuclei of mouse liver cells was detected by Western blotting.The intensity of each band was analyzed, and the values relative to treatment with LPS alone are represented at the lower edge of the image.

Fig. 8 .
Fig. 8.The relationship between DNA repair by pol λ and inflammation