5. EFFECT OF VITAMIN B6 COMPOUNDS ON HUVEC

PROLIFERATION AND TUBE FORMATION

Many endogenous inhibitors of angiogenesis inhibit endothelial cell proliferation in

vitro. Vitamin B6 compounds were applied to human umbilical vein endothelial cell

(HUVEC) proliferation-stimulated human basic fibroblast growth factor (bFGF) in a 72 h

proliferation assay. Among the vitamin B6 compounds, PL and PLP inhibited HUVEC

proliferation in a dose-dependent manner, and LD50 values were 53.9 and 112 μM,

respectively (Table 4). On the other hand, PN and PM, which did not inhibit the activities of

any mammalian pols (Table 1), had no influence on HUVEC proliferation. These results

suggested that PL and PLP must be able to penetrate the cell membrane of HUVEC, and the

inhibitory activity of mammalian replicative pols such as pol α by PL and PLP might be

important for HUVEC proliferation. These results were consistent with the previous report in

which PLP and PL inhibited angiogenesis in a rat aortic ring assay [9]. The effect on HUVEC

proliferation-stimulated vascular endothelial growth factor (VEGF) was also examined, and a

similar inhibitory effect of PLP was ascertained in the assay (data not shown).

14 Yoshiyuki Mizushina, Norihisa Kato, Hiromi Yoshida et al.

HUVEC on reconstituted basement membrane migrated, attached to each other and

formed tube structures. PLP and PL did not affect HUVEC tube formation on the

reconstituted basement membrane at the concentration at which they strongly inhibited

HUVEC proliferation (Figure 4); thus, vitamin B6 would have no effect on such HUVEC

functions.

Table 4. LD50 values of vitamin B6 compounds on HUVEC proliferation

Vitamin B6 compound LD50 values (μM)

PL 53.9

PN >1000

PM >1000

PLP 112

HUVEC was purchased from Kurabo Industries (Osaka, Japan). HUVEC was dispersed with trypsin

and suspended in HuMedia EG2 medium. A cell suspension (15,000 cell/ml) was plated onto 6-well

culture plates (2 ml/well), and incubated at 37 °C in a humidified 5 % CO2 for 24 h. The medium was

replaced with fresh HuMedia EG2 containing vitamin B6 compounds. After 72 h, cells were dispersed

with trypsin, suspended in the medium, and counted.

Figure 4. Effect of PL and PLP on HUVEC tube formation on reconstituted basement membrane gel.

(A) control, (B) 250 μM PLP, and (C) 250 μM PL. Tube formation assay was performed using an In

Vitro Angiogenesis Assay Kit (Chemicon International, Inc., Temecula, CA, U.S.A.). Cells were plated

on reconstituted gel and observed 12 h later. Tube formation was observed under an inverted light

microscope at 40 X magnification.

6. EFFECTS OF PL AND PLP ON

CULTURED HUMAN CANCER CELLS

To investigate the anti-cancer effects of vitamin B6 compounds, human epitheloid

carcinoma of the cervix cell line HeLa was tested. Cell viability was determined by the MTT

(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide) assay [20]. Cell growth

inhibition dose-curves are shown in Figure 5. These results indicated that PL had potent cell

proliferation inhibitory effects on this cancer cell line with an LD50 value of 224 μM.

Surprisingly, none of the PLP tested showed such an inhibitory effect. PLP was more

Inhibition of DNA Polymerase and Topoisomerase by Vitamin B6 15

effective than PL for pol and topo inhibition (Table 1), but PLP showed no effect on the

human cancer cell growth, suggesting that it could not penetrate the cell membrane of HeLa.

Vitamin B6 compound (μM)

Cell viability (%)

0

20

40

60

80

100

0 50 100 150 200 250

Figure 5. Effect of vitamin B6 compounds on the proliferation of human epitheloid carcinoma of cervix

(HeLa) cells. Dose-responsive curves of the growth inhibition of HeLa cells incubated with PL (circle),

PN (diamond), PM (triangle) and PLP (square) for 48 h. HeLa cell line was obtained from the Health

Science Research Bank (Osaka, Japan). Cell proliferation was determined using the MTT (3-(4,5-

dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide) assay [20]. Data are shown as the means ±

SEM of four independent experiments.

7. CONVERSION FROM PL TO ITS 5'-PHOSPHATE FORM IN

HUMAN CANCER CELLS

PLP, which is the 5'-phosphate form of PL, must be the active form of PL in human

cancer cells. The question arises as to whether PL is converted to its 5'-monophosphate form

in vivo. To determine whether the conversion occurs in vivo, we examined the 5’-phosphate

production of PL in a HeLa cell culture with 224 μM. After 48 h of incubation, the cell

extract was isolated, and applied to thin layer chromatography (TLC, 75 % methanol). As

shown in Figure 6, PL was found in the cell extract after 1 h of incubation (lane 3), and PL

and PLP were found in the cell extract after 48 h of incubation (lane 4). These results indicate

that PL rapidly penetrated cells and phosphorylated into its 5'-phosphate form (i.e., PLP).

Subsequently, since converted PLP must inhibit pols α and ε activities in vivo, cell

proliferation is thought to be suppressed. In lane 4, the ratio of PL: PLP was 87 : 13. Since

PL was not effectively but only slightly converted to PLP in the cells, the LD50 value of PL

16 Yoshiyuki Mizushina, Norihisa Kato, Hiromi Yoshida et al.

for cell growth inhibition (i.e., 224 μM) was considered to be approximately 6.8-fold higher

than the IC50 values of PLP for pols α and ε inhibitions (33.8 and 32.6 μM, respectively, in

Figure 3).

PLP

PL

Front

Start

Lane 1 2 3 4

Figure 6. Thin layer chromatograms of the phosphorylation of PL into human epitheloid carcinoma of

cervix (HeLa) cells. Lanes 1 to 4 are PLP (control), PL (control), HeLa cell extract cultured with PL for

1 h, and HeLa cell extract cultured with PL for 48 h, respectively. A photograph of thin layer

chromatography (TLC, methanol / water (75 : 25, v / v)) detected by iodine is shown.

8. CONCLUSION

It has been shown that supraphysiological doses of vitamin B6 suppress tumor growth

and metastasis in mice [4] and that dietary supplemental vitamin B6 suppresses colon

tumorigenesis in mice [8,9]. Epidemiological studies also indicated that vitamin B6 lowers

the risk of colon and lung cancer [5,6]; therefore, interest is increasing in the anti-tumor

effect of vitamin B6 [18,19].

Inhibition of DNA Polymerase and Topoisomerase by Vitamin B6 17

We reported previously that vitamin B6 had anti-angiogenic activity as a potent

mechanism of its anti-tumor effect and demonstrated the activity in an ex vivo angiogenesis

assay using a rat aortic ring [10]. PL and PLP inhibited HUVEC proliferation stimulated by

bFGF in a dose-dependent manner within a range of 25 - 200 μM (Table 4). This result was

consistent with our previous experiment using a rat aortic ring [9]. PLP and PL also inhibited

HUVEC proliferation stimulated by VEGF in a similar manner. On the other hand, PLP and

PL did not affect HUVEC tube formation on a reconstituted basement membrane, implying

that it has no effect on the mobility and attachment functions of HUVEC. Thus, the antiangiogenic effect of vitamin B6 appears to be mediated through the suppression of

endothelial cell proliferation. Moreover, vitamin B6 reportedly suppresses cancer cell

proliferation in vitro [1-3], and supplemental vitamin B6 suppresses the expression of cell

proliferation-related genes, c-myc and c-fos, in the colon epithelium of mice receiving

azoxymethane [8].

These results arouse interest in the identify of the molecular target of PL and PLP. PL

and PLP seem to be very similar to the base of bredinin, an analog of pyrimidine base,

inosine [21]. As described previously [21], the base of bredinin is an inhibitor of cancer cell

proliferation and pols. Since the inhibition of cell proliferation is mostly a result of the

inhibition of DNA replication directly or indirectly, based on the experience of bredinin

studies [21], we tested here the effects of PLP and PL on DNA metabolic enzymes such as

pols. In particular, PLP was a potent inhibitor of eukaryotic pols and human topos, especially

pols α and ε, and interestingly, had hardly any effect on repair-related pols. The cellular

effects described above by vitamin B6 must be caused by the inhibition of DNA replication.

The problem with this speculation is that PL only weakly influenced pols, and PLP is thought

to hardly penetrate living human cancer cells such as HeLa. It is possible that PL penetrates

cells and converts to PLP, which selectively inhibits pols α and εactivities, and subsequently

DNA replication and cell proliferation.

These indicated results imply that PLP has a physiological role in maintaining the proper

structure of the chromosome by controlling the activities of replicative pols and topos. It has

been demonstrated that vitamin B6 deficiency enhanced gene expression in rat liver [22,23]

and that vitamin B6 supplementation suppressed some gene expression in cancer cells [24].

These observations could be explained by the effect of vitamin B6 on pols and topos

elucidated in this review.

In conclusion, this review revealed evidence for the inhibitory effects of vitamin B6 on

replicative pol and topo activities, and the proliferation of endothelial cells and cancer cells.

These effects may relate to the anti-angiogenesis and anti-cancer effect of vitamin B6.

ACKNOWLEDGMENTS

We are grateful for the donations of calf pol αby Dr. M. Takemura of Tokyo University

of Science (Tokyo, Japan), rat pol β by Dr. A. Matsukage of Japan Women's University

(Tokyo, Japan), human pol γ by Dr. M. Suzuki of Nagoya University School of Medicine

(Nagoya, Japan), human pols δ and ε by Dr. K. Sakaguchi of Tokyo University of Science

(Chiba, Japan), human pols η and ι by Dr. F. Hanaoka and Dr. C. Masutani of Osaka

18 Yoshiyuki Mizushina, Norihisa Kato, Hiromi Yoshida et al.