PlantCell Reports

Plant Cell Reports (1992) 11:132-136

9 Springer-Verlag1992

Transformation of opium poppy (Papaver somniferum L.) with Agrobacterium rhizogenes MAFF 03-01724 Kayo Yoshimatsu and Koichiro Shimomura Tsukuba Medicinal Plant Research Station, National Institute of Hygienic Sciences, 1 Hachimandai, Tsukuba, Ibaraki, 305 Japan Received October 8, 1991/Revisedversion received January 28, 1992- Communicated by E Constabel

Summary.

Transformed cultures of opium poppy (Papaver somniferum L.) were established by infecting hypocotyl segments with Agrobacteriumrhizogenes MAFF 03-01724. Undifferentiated calli formed on the infected site grew satisfactorily on phytohormone-free solid medium in the dark and produced opine, mikimopine, which could not be detected in a normal culture. Numerous adventitious shoots developed from transformed calli during subculture. The transformed shoots separated individually were cultured on phytohormone-free MS solid medium at 22 ~ C under 14 h/day light. They displayed wider leaves and longer internodes than shoots established from seeds or nontransformed root culture. The content of morphinan alkaloids in the cultures and regenerated shoots were quantitatively analyzed by enzyme-linked immunosorbent assay and high performance liquid chromatography. HPLC analysis revealed that non-transformed shoots contained much more codeine (1310 pg/g dry wt.) than morphine (50 ~tg/g dry wt.), while the transformed shoot cultures did not contain morphine, although the level of morphinan alkaloids in the transformed shoots (213 I.tg morphine equivalents/g fr. wt.) was comparable to that in nontransformed shoots (182 ~tg morphine equivalents/g fr. wt.) by ELISA.

Abbreviations MS, Murashige-Skoog (Murashige and Skoog 1962); 1/2 MS, half strength MS; HF, phytohormone-free; NAA, 1naphthaleneacetic acid; ELISA, enzyme-linked immunosorbent assay; HPLC, high performance liquid chromatography. K e y w o r d s : Transformation-Papaver somniferumAgrobacterium rhizogenes-Morphinan alkaloids-ELISAHPLC

Introduction Opium poppy (Papaver somniferum L., Papaveraceae) is

Correspondenceto: K. Shimomura

one of the most important medicinal plants and has been cultivated since early centuries, Opium, the dried cytoplasm of a specialized internal secretory system called the laticifer, is normally collected from the unripe capsule, and opium of P. somniferum is the source for the commercial production of medicinally important alkaloids, including morphine, codeine, thebaine, noscapine and papaverine (Roberts 1988, Nessler 1990). Morphine, which has strong additive properties, is still the most effective analgesic for the treatment of terminal cancer patients in modern medicine, while codeine is commonly used as an antitussive. However, field cultivation of opium poppy has been limited since 1953 by the United Nations Opium Conference Protocol to prevent narcotic crimes. Therefore establishing a tissue culture technique for the production of morphinan alkaloids seems to be desirable for not only medicinal purpose but also decreasing abuse of opiates. Many researchers have so far investigated tissue culture of P. somniferum (Nessler 1990, ref. cited therein), and most cultured P. somniferum cells, either as callus or cell suspensions, readily produced isoquinoline alkaloids such as sanguinarine, but rarely, if even, produce morphinan alkaloids. Kamo et al, (1982), Schuchmann and Wellmann (1983) and Yoshikawa and Furuya (1985) reported the production of morphinan alkaloids in redifferentiated organs, either shoots or somatic embryos, and their results emphasize the importance of the degree of cell differentiation for the biosynthesis of morphinan alkaloids. Recently organ cultures transformed with Agrobacterium have been demonstrated to produce useful secondary metabolites but there have been no reports on the transformation of P. somniferum as far as we know. We succeeded in producing transformed cultures of P. somniferum by infection with A. rhizogenes MAFF 0301724, which was recently isolated in Japan. This strain induces mikimopine production in transformed plant cells (Isogai et al. 1988, 1990). This paper describes the

133 establishment of transformed cultures of P. somniferum and their capability to produce morphinan alkaloids. Material and method Induction of transformed calli. Seeds of P. somniferum L. var. Ikkanshu were surface-sterilized according to the method of Ikeda et at. (1988), placed on agar solidified medium (0.5 % sucrose and 0.5 % agar) and kept at 18 ~ C under 14 h/day light (80 I.tEm'2S-l). After 1 week, hypocotyl segments (ca. 0.5 cm) excised from the seedlings were transferred to I/2 MS liquid medium (20 ml/100 ml Erlenmeyer flask) and co-cultured with A. rhizogenes MAFF 03-01724 (ca. 105 - 106 bacteria in the liquid medium) at 25 ~ C in the dark on a rotary shaker (100 rpm). After 1 day, the segments were placed on 1/2 MS solid medium containing 500 mg/l Claforan| (Hoechst Japan Ltd.). Ten days later, white calli had formed on the segments. The bacteria-free calli were transferred to tlF MS solid medium (25 ml/90 i.d. x 20 mm petri dish) and maintained at 22 ~ C in the dark with subculturing at 3-4 week intervals. Numerous shoots developed spontaneously on the caUi during subculture. The medium used for the experiments was MS medium supplemented with 30 g/1 sucrose and adjusted to pH 5.7 before autoclaving at 121 ~ for 15 min. Solid media were solidified with 0.2 % Gelrite (Merck & Co., U. S. A.).

Induction of non-transformed roots and culture of shoot-forming calli. Hypocotyl segments were cultured on MS solid medium containing 0.5 mg/l NAA at 22 ~ C in the dark. After 1-2 months, newly developed roots were excised and subcultured under the same conditions. Shootforming calli appeared on the roots sporadically, and were maintained on MS solid medium containing 0.5 mg,tl NAA at 22 ~ C in the dark with subcuhuring at 4 week intervals.

Culture of non-transformed and transformed shoots. The shoots formed on transformed and non-transformed calli were separated and cultured individually on HF MS solid medium (30 ml/30 i.d. x 150 mm test tube) at 22 ~

under 14 h/day light (80l.t Em-2S - 1).

Detection of mikimopine.

F r e s h t i s s u e (100-200 mg) was homogenized with a plastic rod in a microtube, and the crude extract (50 p.1) obtained after centrifugation at 6,000 g for 1 min was subjected to high-voltage paper electrophoresis according to the method of Petit et al. (1983). The detection of mikimopine was performed with the Pauly reagent (Isogai et al. 1988, 1990).

ELISA of morphinan alkaloids.

The crude extract was immediately subjected to ELISA analysis as described previously (Yoshimatsu et al. 1990). The assay conditions were as follows: coating antigen, morphine-conjugated ovalbumin (2 gg/ml) in 50 mM NaHCO 3 buffer (pH 9.6) (50 lal/well); morphine-specific monoclonal antibody, MOR 131.5.13 (Sawada et al. 1988) diluted xl0,000 with 10 mM phosphatebuffered saline containing 1 g/1 casein (C-PBS, pH 7.2) (50 g.l/well); second antibody-enzyme conjugate, peroxidase-conjugated sheep antimouse IgG (BIOSYS, S.A., France, BI 3413/8249) diluted x 10,000 with C-PBS (50 p.l/well); substrate, TMB (Kirkegaard & Perry Laboratories Inc.). Morphine equivalents in samples were calculated from a standard curve of morphine.

ttPLC conditions. Alkaloid extracts were injected onto a TSK gel ODS120A column (TOSOH, 4.6 i.d. x 250 mm), mobile phase acetonitrile10 mM sodium l-heptanesulphonate (pH 3.5) gradient (25 - 80 % acetonitrile, see Fig. 4), flow rate 1.0 ml/min, column temperature 35 ~ C, UV detection at 284 nm. Each peak corresponding to morphine, codeine, papaverine, noscapine and sanguinarine was confirmed by evaluating its UV spectrum using a photodiode array detector (Waters 990J). The data from these extracts are shown as the mean of three replicates.

Results and discussion

Establishment of non-transformed and transformed calli Three clones of transformed calli (MAFF clone 1, 2 and 3) were established by infection of P. somniferum with A. rhizogenes MAFF 03-01724, although the infection rate was low (below 20 %). Numerous adventitious shoots differentiated spontaneously on the callus, with clone 1 showing the strongest shoot-forming capability (Fig. 1 B). A. rhizogenes MAFF 03-01724 causes hairy root formation at the infection site on tobacco plants (Isogai et al. 1988), but in the case of opium poppy no roots appeared. Shootforming callus was also obtained from non-transformed roots cultured with 0.5 mg/l NAA (Fig. 1 A). Mikimopine, however, was only detected in the MAFF clone (Fig. 2). Infection of hypocotyl segments with A. rhizogenes A4 was also attempted, but no transformed cultures were obtained.

Morphinan alkaloid contents in non-transformed and transformed calli The morphine-specific antibody used in this study has a high affinity for morphine, codeine, ethylmorphine, dihydromorphine and dihydrocodeine, and less reactivity with dihydromorphinone, dihydrocodeinone and norcodeine (Sawada et al. 1988). The morphinan alkaloid content of non-transformed and transformed calli was first analyzed by ELISA to evaluate their morphinan alkaloid productivity from small amount of crude extracts (Table 1). Based on the ELISA results, MAFF clone 1 which had the strongest shoot-forming capability contained the highest level of morphinan alkaloids (1.76 Ixg morphine equivalents/g fr. T a b l e 1. M o r p h i n a n

cultures

in t h e c a l l i o f P.

morphine equivalents (~tg / g fr. wt. + s.d.)

n o n - t r a n s f o r m e d callus (NAA 0.5 mg/l)

0.17

_-4- 0 . 0 6

MAFF clone 1

1.76

+_ 0 . 8 7

Extraction of alkaloids.

Alkaloids were extracted using a modified method of Staba et al. (1982). Each lyophilized sample (ca. 50 rag) was ground to a fine powder and extracted with 1 ml of 5% acetic acid two times by ultrasonication. The aqueous solution was then washed with 2 ml of chloroform. The aqueous phase was made alkaline with conc. ammonium hydroxide, and extracted with chloroform three times (2, 2, and 1 ml). The combined chloroform extracts were evaporated in vacuo, dissolved in an appropriate volume of methanol and analyzed by HPLC.

alkaloid contents

somniferum

12.00

_+ 4 . 0 0 *

MAFFclone2

0.22

+

0.01

MAFF clone 3

0.08

_+ 0 . 0 3

Calli were maintained on MS solid medium at 22 ~ C in the dark or under 14 h/day light (*) for 4 weeks. Morphinan alkaloids were analyzed by ELISA. The data were shown as the mean of two replicates.

134

Fig.2. Detection L a n e s 1, 2 and M A F F clone 1, callus (0.5 mg/I

Fig.1.

Non-transformed ( A : 0.5 m g / I N A A ) a n d t r a n s f o r m e d (B; MAFF clone 1, without phytohormone)

calli of P. somniferum cultured on MS solid medium at 22 ~ in the dark.

wt.) among MAFF clones cultured in the dark. Kamo et at. (1982) and Yoshikawa et al. (1985) suggested that the biosynthesis of morphinan alkaloids might be closely

of mikimopine. 5; m i k i m o p i n e , lane 3; lane 4; n o n - t r a n s f o r m e d NAA)

related to the differentiation of shoots, and this result supports their suggestion. However, the nontransformed callus contained ten-fold less morphinan alkaloid (0.17 I.tg morphine equivalents/g fr. wt.) although it showed almost the same shoot-forming capability as MAFF clone 1. Therefore, the difference in morphinan alkaloid production observed among MAFF clones is related more to clonal line differences (Mano et al. 1986). The transformed shoots on MAFF clone 1 turned green when cultured under 14 h/day light. Those shoots contained approximately 7-fold higher levels of morphinan alkaloids and codeine was clearly detected by TLC analysis (data not shown). Then transformed shoots on MAFF clone 1 were used for further study.

Fig.3. R e g e n e r a t e d s h o o t s of P. somniferum A; control (from seed) B; non-transformant (0.5 mg/I NAA) C; transformant (MAFF clone 1)

135

Culture of non-transformed and transformed shoots The non-transformed and transformed shoots (MAFF clone 1) induced on callus in the dark were transferred on HF MS solid medium (30 ml/30 i.d. x 150 mm test tube) and their growth and features compared with intact shoots established from seeds (Fig. 3). The non-transformed shoots showed almost the same growth and appearance as shoots germinated from seed, while transformed shoots, which also produced mikimopine (data not shown), demonstrated aberrant structural features such as wider leaves and longer internodes.

Morphinan alkaloid content of non-transformed and transformed shoots To analyze morphine and codeine more specifically and quantitatively, HPLC fractionation of the extracts was employed. An HPLC method to detect sanguinarine simultaneously with morphine and codeine was desirable, because in vitro cultures readily produce sanguinarine (Roberts 1988, Furuya et al. 1972, Ikuta et al. 1974, Schuchmann and Wellmann 1983, Tyler et al. 1988, Songstad et al. 1989). Therefore we established .

.

.

.

i

Sangulnarine 100' 0.5

......

80.

50" 430.4

2S-

10 o~

Time

0.3

15

22

26

(min)

an HPLC method capable of analyzing morphine, codeine, papaverine, noscapine and sanguinarine simultaneously (Fig. 4). The content of morphinan alkaloids in regenerated shoots was analyzed by ELISA and HPLC (Table 2). The T a b l e 2. M o r p h i n a n a l k a l o i d shoots of P. somniferum

content

in r e g e n e r a t e d

ELISA (IXg/gft. wt. _+s.d.) HPLC(gg/g dry wt. +.+s.d.) . material

morphine equivalents

morphine

codeine

control(from ~e,eds)

472 _+ 155

70_+ 40

2030_+ 330

non-transformant~AA0,sm~)

182 -+ 62

50-+ 10

1310-+650

MAFF clone 1

213 +_ 4

n.d.

750 _+ 230

n. d.: not detected Shoots were cultured on M S sofid m e d i u m under 14 h/day light at 22 o C for 6 weeks. The data w e r e shown as the m e a n of three replicates.

regenerated shoots apparently could accumulate much more morphinan alkaloids than their original calli (approximately 1000-fold higher level of morphine equivalents for nontransformant, 100-fold higher level of morphine equivalents for transformant). The transformed shoots contained almost the same level of morphinan alkaloids as non-transformed shoots when being analyzed by ELISA, but no morphine could be detected by HPLC. In contrast both of nontransformed shoots contained much more codeine (13102030 gg/g dry wt.) than morphine (50-70 I-tg/g dry wt.). Other alkaloids ( papaverine, noscapine and sanguinarine) were detected in both non-transformed and transformed shoots. In general concept, morphine is biosynthesized from codeine by demethylation (Trease and Evans 1989). Yoshikawa and Furuya (1985) observed that codeine was the main morphinan alkaloid at the early stage shoot (0.5-3.0 cm in height) and the morphine content increased during shoot development. These results imply that the biosynthesis of morphine from codeine is, to some extent, correlated with plant cell maturation.

=

Conclusion O

0.2

< . . . . . . . . .

0.1

[l oph'e o/i ................I

.........................

0 ....

5 ....

! .........................

10 ....

15 . . . .

!r .................

20 ....

25

"

Time (min) F i g . 4 . H P L C c h r o m a t o g r a m of o p i o u m a l k a l o i d s . The f i g u r e i n s i d e i n d i c a t e s g r a d i e n t p r o g r a m of m o b i l e phase.

We succeeded in the induction of transformed cultures of P. somniferum by infection of hypocotyl segments with A. rhizogenes MAFF 03-01724. The transformed cells of P. somniferum easily formed adventitious shoots instead of roots, and shoot production was correlated with production of morphinan alkaloids. MAFF clone 1 accumulated about 10-fold higher levels of morphinan alkaloids than did non-transformed shootforming callus, and light clearly influenced morphinan alkaloid production (Table 1). The regenerated shoots from MAFF clone 1 did not accumulate any detectable amount of morphine. Roberts (1988) points out that codeine production by stable cell cultures has been an obvious target for exploitation because of world requirements and the limited availability of codeine from the opium poppy plant. The increasing abuse of opiates has also stimulated the search for raw materials

136 from sources other than P. somniferum which might contain non-addictive morphinan alkaloids (Trease and Evans 1989). Our results here present the possibility of producing morphine-free P. somniferum plant by transformation with A. rhizogenes, though further study will be required to clarify the biosynthetic capability of transformed plants at the latter stage. Acknowledgments. Authors thank Dr. Jun-ichi Sawada (Division of Biochemistry and Immunochemistry, National Institute of Hygienic Sciences) for providing the morphine-specific monoclonal antibody. This work was supported in part by Special Coordination Funds of the Science and Technology Agency of the Japanese Government.

References Furuya T, Ikuta A and Sy6no K (1972) Phytochem. 11: 3041-3044 Ikeda K, Teshima D, Aoyama T, Satake M and Shimomura K (1988) Plant cell Rep. 7:288-291 Ikuta A, Sy6no K and Furuya T (1974) Phytochem. 13: 2175-2179 Isogai A, Fukuchi N, Hayashi M, Kamada H, Harada H and Suzuki A (1988) Agric. Biol. Chem. 52:3235-3237 Isogai A, Fukuchi N, Hayashi M, Kamada H, Harada H and Suzuki A (1990) Phytochem. 29:3131-3134 Kamo KK, Kimoto W, Hsu A-F, Mahlberg PG and Bills DD (1982) Phytochem. 21:219-222 Mano Y, Nabeshima S., Matsui C and Ohkawa H (1986) Agric. Biol. Chem. 50:2715-2722 Murashige T and Skoog F (1962) Physiol. Plant. 15: 473-

497 Nessler CL (1990) In: Handbook of Plant Cell Culture vol. 5, Poppy, Ammirato PV, Evans DR, Sharp WR and Bajaj YPS (ed), pp 693-715, McGraw-Hill Publishing Co., New York Petit A, David C Dahl GA, Ellis JG, Guyon P, CasseDelbart F and Temp6 J (1983) Mol. Gen. Gent. 190: 204214 Roberts MF (1988) In: Cell Culture and Somatic Cell Genetics of Plants vol. 5, Isoquinolines (Papaver Alkaloids), Constabel F and Vasil IK (ed), pp 315-334, Academic Press Inc., San Diego Sawada J, Janejai N, Nagamatsu K and Terao T (1988) Molecular Immunology 9:937-943 Schuchmann R and Wellmann E (1983) Plant Cell Rep. 2: 88-91 Songstad DD, Giles KL, Park J, Novakovski D, Epp D, Friesen L and Roewer I (1989) Plant Cell Rep. 8: 463466 Staba EJ, Zito S and Amin M (1982) J. Nat. Prod. 45: 256-262 Trease GE and Evans WC (1989) In: Pharmacognosy 13th edition, p 582-591, Baillibre Tindall, London Tyler RT, Eilert U, Rijnders COM, Roewer IA and Kur~ WGW (1988) Plant Cell Rep. 7:410-413 Yoshikawa T. and Furuya T. (1985) Planta Med. 110-113 Yoshimatsu K, Satake M, Shimomura K, Sawada J and Terao T (1990) J. Nat. Prod. 53:1498-1502