Biochemical Characterization of Two Tobacco Cell Lines With

tapping button above. Home Screen Newspaper. No matches found. No publications added. Journals. Books. Book Series. Working Papers. If it takes too lo...
0 downloads 0 Views 425KB Size
B I 0CH E 11I CA4L CHX RX C T E R I ZA4TI 0S 0F T W 0 T 0B X CC0 CELL LISES WITH DIFFEREST LEYELS OF CINS,41IOYL PUTRESCISES .JOCHES

BERLIS, KARL-HEISZ KSOBLOCH, GERHARD HOFLEand LL-DGER ~VITTE

Geselisihizft f. Biotechnologischc FurschiiT!g w.O.H., S f n s c h e r o d e r !l.eg 1 , D 5300 - Bri2it~rscii;ceig, Federol R e p u b l i c of Gertnciny ABSTRACT.--.~ p-fluorophenylalanine-resistanttobacco cell line fTSi I and a wildtype culture i T S l , were compared biochemically. T h e former yielded 6 times more cinnamoyl putrescines than t h e l a t t e r , and t h e main phenolic compounds of both cell iines were identified as caffeoyl-, feruloyl-, and p-coumaroyl putrescine. T h e increased accumulation b?- T I 1 cells was evidently due to increased enzyme activities of related enzymes, phenylalanine ammonia-lyase. trans-cinnamate 1-hydroxylase, p-coumarate-Co.4 ligase, ornithine and arginine decarboxylases,

In rr.ost plant cell cultures, secondary patlin-ays are not well expressed. However. it has been shon-n that high yielding strains can be obtained from lon- producing parent cultures bj- analytical or biochemical selection (1-6). ; ibiochemical characterization of variant strains with different capacities in the synthesis and accumulation of secondary metabolites ma>- help to identify biochemical requirenients of high producing strains and to elucidate the enzyrno1og:- of secondar:. pathn-ap-. The value of plant cell cultures for ,studying biochemical aspects of secondary pathways has been impressively demonstrated by the progress in the enzymologj- of flavonoids and indole alkaloids ( i - 8 ) . Palmer and Kidholm selected a p-fluorophenylalanine (PFP) resistant celi line (TX4j from a parent culture of S i c o f i a i i a tabaczim L.cv.Xanthi ( T X l ) (9). The resistant line accumulated 6 to 10 times more phenolic compounds, and its phenylalanine ammonia-lyase (P-iL) activity was coneiderably increased (10). From feeding of labeled precursors and from their uv-.spectra. the phenolic compounds n-ere tentatively regarded a.5 cinnamoyl putrescine derivatives (11). Prelimirmy results indicated that not only P-\L but also activities of other enzymes of the proposed pathway n-ere increased i l l ) . Here we report on a detailed identification of the cinnamoj-l putrescines of the tobacco cultures and on the activities of five enzymes involved in the biosyntlmis of these compounds in lox- and high producing cells. EXPERIIIESTa\L PL.*sr \ 1 ~ ~ ~ ~ 1 ~ ~ . - 1 l a i n t e and n a nsome c e characteristics of nild-type cells TS1 and t h e PFP-resistant cell line T S i iSicotiiinn tnbncitni L. cv. S a n t hi I have been described previousl319,10,12,13). Inoculum size was 1 g fr.wt. per 70 nil 11s-medium (11) containing 2 p 1 l 2,1-D. Cells were harvested b3- vacuum filtration for fresh wt and d r y wt determinations and for biochemical analysis. I S O L A T I O S OF

cISs.~~foYL Pt-TRESCISES.-FreeZe

dried cells were extracted t x i c e with

11CW (methanol-chloroform-water, 12:5:31. T h e combined extracts were separated in two phases by t h e addition of 3 ml chloroform and 3 ml water to 10 ml 1ICR-extract. The chloroform-phase was discarded, and the m e t h a n o l - m t e r phase was concentrated for chromatography in t h e systems L l (n-butanol-acetic acid-water, i : l : l jWhatman 3 1111,cellulose or silca gel tlc), L? iisobutylmethylketone-formicacid-water, 11:3:2, silica gel tlc I and L3 in-propanolx+-ater-Tris-HC10.1 31 p H 8.0, 70:30:1, cellulose,. Paper chromatography with L1 separated caffeoyl putrescine (pale fluorescent) from feruloyl putrescine (blue fluorescent I and p coumaroyl putrescine (dark absorbing). T h e hands were cut out and eluted with 505 methanol. Feruloyl putrescine and p-coumaroyl putrescine were further purified from each other b y tlc in t h e systems L3 and L1 on silica gel. L2 was mainly used t o check for other impurities.

IDESXFICATIOS OF CISSAMOYL P L - T R E S C I S E S . - . ~ ~ ~ isolated compounds were compared with chemically synthesized cinnamoyl putrescines (l5j as h>-drochlorides. a; C a f f e o y l putrescine.-The isolated product had an identical uv-spectrum with the synthesized compound (Xmas 320, 295, Xmin 260 in joyc I I e O I I i , showed a bathochromic shift with X a O H (Xmas 365j, and a similar shift q-ith AICIP (Xmax 328, 352 s h ) , and was ninhydrin positive. Mass spectroscopy of the synthesized, a s well as the isolated compound, did not provide a molecular ion nor ani- t?-pical fragmentation pattern. Therefore 83

s4

Journal of Natural Product's

[Vol. 45, No. 1

both compounds were refluxed in 2 ml acetic acid anhydride for 10 min. After evaporat,ion to dryness, a mixture of tri-, tetra- and pentaacetyl derivatives was detected b y chromatography in ethyl acetate-methanol, 9:l (silica gel). T h e mixture was boiled in ethanol for 2 min t o give only t h e triacetyl derivative (Rf=0.2). h i s , m/z (%) 376 (hi+, 10) 334,(8), 317(4), 292(92), 233(16), 163(100). T h e nmr-spectra of t h e two triacetyl derivatives were also identical (15). b) Feruloyl putrescine.-The isolated product and t h e synthetic compound showed identical uv-spectra (Xmax 320, 295, Xmin 260). T h e bathochromic shift b y alkali (Xmax 358) was smaller than for caffeoyl putrescine and t h e uv-spectra were not altered b y AlC13. T h e compound was ninhydrin positive. T h e purified compound (L1, cellulose, silica gel) was directly analyzed by mass spectroscopy. T h e fragmentation pattern was identical with t h e synthesized compound and d a t a published b y Wheaton and Stewart (16): MS, m/z(yc) 264 (hfL, l l ) , 177(83), 145(55), iO(100). T h e isolated compound and t h e synthetic compound were also analyzed by gcims a s trifluoroacetyl derivatives (TFA.4, 60°C for 30 m i n ) : MS, m / z ( % ) 552 (Mi, l o ) , 455(4), 273(100), 245(3), 176(15). Gc-conditions: 15, x 0.25 mm i.d. fused silica capillary column coated with SE 30, carrier gas helium 0,65 bar, split ratio 1:30, temperature program 200-300"C 6"C/min, detection FID. I=2435 according t o t h e K o v a t s retention index (17). c ) P-Coumnroyl putrescine.-The uv-spectra of t h e isolated compound (L1 Whatman 3 MM; L3 cellulose; L1 silica gel) and t h e synthesized compound were identical (Xmax 302, 289 Xmin 245). T h e bathochromic shift t o Xmax 340 b y K a O H indicated t h e free phenolic hydroxyl group and t h e positive ninhydrin reaction t h e free amino group. Since t h e m s pattern of t h e synthetic compound m / z ( % ) 234 (hi-, 3 ) , 147(100), 119(31), 91(27), 70(87) was observed in t h e mass spectrum of t h e isolated product only in addition t o signals from impurities, the isolated and synthetic compounds were also compared b y gc/ms as their trifluoroacetyl derivatives (conditions see above) I=2275. M S , m / ~ ( 522 7 ~(M+, ) 3), 425(1), 243(100), 215(5). QUASTIFICA~ION OY OIL PUTREScISES.-The absorption coefficients for t h e hydrore for caffeoyl putrescine € a z o = 18600, feruloyl putrescine t3*0= chlorides in 50% meth 19630 and p-coumaroyl putrescine e302=18750. T h e crude MCW-extracts of T X 1 and T X 4 cells showed uv-spectra typical for caffeoyl/feruloyl putrescine. Since t h e ratio of isolated caffeoyl-feruloyl-coumaroyl putrescines was roughly 8 : l : l in both cell lines, t h e quantification from MCW-extracts was based on t h e ~ ~ ~ o - v a lofu ecaffeoyl putrescine. Equivalent levels were found b y quantifying cinnamoyl putrescines by t h e Folin-method (10).

EXTRACTIOZ A K D A S S A Y S OF EszriiEs.-Extraction and measurement of phenylalanine ammonia-lyase (EC 4.3.1.5: P A L ) , L-ornithine decarboxylase ( E C 4.1.1.17; ODC) and Larginine decarboxylase ( E C 4.1.1.19; ADC) have been described previously (10,13). F o r t h e determination of 4-coumarate:CoA ligase ( E C 6.2.1.12) and trans-cinnamate 4-hydroxylase (EC 1.14.13.14) 2.5 g (fr.wt.) cells were homogenized with quartz sand and 5 mlO.1 h i Tris-HC1, p H 7.5, containing 7 m l 1 2-mercaptoethanol and 15% glycerol in t h e presence of 0.4 g buffer saturated Polyclar AT. After centrifugation (20,ooO g, 20 min) t h e supernatant was chromatographed on Se hadex G 25 equilibrated with t h e same buffer. From t h e eluate t h e activity of 4-cournarate:8oL4 ligase was measured according t o (18) with an altered CoASH concentration of 0.2 mhi. Cinnamic acid 4-hydroxylase was measured according t o (19) from a micros3mal fraction prepared from t h e enzyme extract after gel filtration b y centrifugation a t 100,ooO g for 70 min. T h e pellet was resuspended in 0.5 ml 0.1 hi Tris-HC1, p H 7.5, containing 3 m h l 2-mercaptoethanol. Protein was determined by t h e L o m y method (20).

RESULTS A S D DISCUSSIOK IDENTIFICATION A N D QUASTIFICATIOS OF CINSAMOTL PUTRESC1SES.-The

purpose of our work was to investigate biochemical differences between tobacco cell lines with low (TX1) and high (TX4) accumulation of phenolics. After purification by paper and thin-layer chromatography, the phenolic compounds have now been identified as caffeoyl-, feruloyl- and p-coumaroyl putrescines by their uvspectra, their chemical reactions (e.g. to ninhydrin and Folin-reagent'), and their mass fragmentation in comparison with authentic material (see Experimental), I n particular, caffeoyl putrescine caused some problems during identification since it was not possible to obtain mass spectra from the isolated product. After chemical synthesis of the compound (15), it was noted that only derivatives of caffeoyl putrescine resulted in typical mass fragmentation patterns. Caffeoyl putrescine is a rather sensitive compound. Thus, the compound was destroyed by chromatography in systems above p H 9 or on silica gel plates. Light caused t'rans-cisisomerization (21) and, therefore, often tx-o bands were found in second or third chromatography systems. Due to the lack of an ortho-hydroxyl group, feruloyland p-coumaroyl putrescine were even stable in basic chromatographic systems like n-propanol-ammonium hydroxide solution (7 :3) on cellulose and were also eluted from silica gel plates. M7hilewe were able to separate the three synt'hesized compounds on tlc-plates (L3) for direct quant'ification by a Shimadzu-TLC-scanner,

Jan-Feb 19S2]

85

Berlin et al. : Cinnamoyl Putrescines

this n-as only possible from plant extracts after several purification steps. Separation and quantification of the three compounds as their trifluoroacetyl derivatives by capillary gc proved to be unsuccessful since caffeoyl putrescine gave an unstable product for gc and up to 50% decomposition products were found for feruloyl- and p-coumaroyl putrescines. Therefore our quantification was based on the fact that caffeoyl putrescine was the main compound (fig. 1). From the quantities of isolated compounds, a ratio of S:1:1 (caffeoyl-feruloyl-p-coumaroyl putrescine) was calculated. Therefore direct measurement of MCW-extracts a t 320 nm was sufficient for our purposes. We have observed the productivity of TX1 and T X 4 cells for more than 6 years. It is now evident that the p-fluorophenylalanine resistant cell line T X 4 is a very stable line with regard to its resistance and its productivity in the absence of the selective agent. A typical kinetic curve for grorvth and production over a grou-th cycle is given in fig. 1; but. as in all cell culture systems, variations were occasionally observed. Thus, TX4 cells had levels between 5-10yc and TX1 betueen 0,3 to 1,5yccinnamoyl putrescines on a dry weight basis over the years. However, the difference betv een T X 1 and T X 4 cells always remained 6 to IO-fold. Conqidering the good growth of TX-l cells (fig. 1) [30O g fresh wt liter in 20 liters g d r y weight n 0

d

h)

0

0

A

w

0

f’

o

m 0

0-l

0

mg caffeoyl putresc1ne.g dry weight-’ FIG.1. Growth kinetics of T X l ( 0 7 0 ) and T X 4 cells and accumulation of caffeoyl putresclne by TX1 (3+) and T X I (E--3) cells.

)+.(

air-lift fermenter (unpublished)] this line classifies in the list of high producing cell cultures ( 2 2 ) with a yield of caffeoyl putrescine of 0,6-1,0 g,liter. The occurrence of caffeoyl-, feruloyl- and p-coumaroyl putrescine in callus cultures of a different Sicotiaita tabacum variety has already been described ( 2 3 ) . More recently such hydroxycinnamic acid amides have also been found in intact plants of tobacco (24). I t has been wggested that these amides may play an important role in flowering processes ( 2 5 ) and male sterility (26). The presence of bound putrescine may also be related to the biosynthesis of nicotine in tobacco ( 2 3 ) . This alkaloid n a s not found in our cell lines. E s z n i o L o G i OF C I S S A ~ I O Y L PUTREscI?iEs.-our main interest if-as directed to the question of what is the biochemical basis of the increased productivity of

86

[Vol. 45, s o . 1

Journal of Satural Products

T X 4 cells. Therefore the activities of enzymes involved in the phenylpropanoid pathway and those involved in the formation of putrescine were compared in TX1 and TX4 cells (fig. 2). The activities of the first three enzymes of the general phenylpropanoid pathway, according to Hahlbrock the “group I” enzymes (27), were increased manifold in T X 4 cells. The changes in activities during the growth cycle indicated a coordinated regulation of these three enzymes. A coordinated behavior of these enzymes has also been noted in several other systems (27). We have recently shown that the conjugate putrescine may be formed via arginine or

I

B

A

A ‘!

A 5

’. 10

days

FIG.2. Comparison of enzyme activities involved in the biosynthesis of cinnamoyl putrescines in TX1 (filled symbols) and TX4 (open symbols). A) Enzymes of general phenylpropanoid metabolism: phenylalanine ammonia-lyase (O-C), cinnamate 4hydroxylase (A-A), p-coumarate: CoA ligase (5-51. B) Enzj-mes of putrescine biosynthesis: ornithine decarboxylase (A-A) and arginine decarboxylase (3-3).

ornithine decarboxylation (13). The activities of both enzymes n ere significantly increased in TX4 cells. However, the pattern of ornithine and arginine decarboxylase activities differed from that of the enzymes of the phenylpropanoid pathway. I n cell lines accumulating no cinnamoyl putrescine similar changes in the activities of arginine decarboxylase were found (28). The formation of cinnamoyl putrescines by TX1 cells was greatly increased when the cells were transferred to a medium devoid of phosphate and 2.4-D (29). I n these medium-induced TX1 cells, the activity of phenylalanine ammonia-lyase, but not of ornithine decarboxylase, shou-ed a close correlation with increased product accumulation compared with non-induced cells (29). Therefore, it is concluded that the activities of the decarboxylases and of the enzymes of the general phenylpropanoid pathway are regulated independently, the former having less influence on the regulation of product formation. We also tried to find a hydroxycinnamoyl-CoX :putrescine hydroxycinnamoyltransferase in TX1 and T X 4 cells. Under conditions allowing conjugation with quinate or shikimate (30,31), we failed to detect such an enzyme in our cell lines. However, feeding experiments with putrescine clearly indicated that T X 4 cells should have a higher activity of a conjugating enzyme (13). Thus, one may conclude that the increased accumulation of cinnamoyl putrescines by T X 4 cells is due

Jan-Feb 1%2]

Berlin e / (11. : Cinnanioyl Putre,cines

87

to the increased activities of many enzymes related to this secondary pathway. From results n-it h the P-IL-inhibitor a-aminooxy-fi-phenylpropionic acid or t'he 5-enolpyruvylshikiniate-3-phosphatesynthase inhibitor glyphosate (32), it can be concluced that no regulatory alterations in phenylalanine supply had occurred in TS4 cells ( 3 3 ) . The productivity of cell cultures evidently depends very much on the activity of related enzymes. I n T S 4 cells all of the tested enzymes were increased considerably. Thi.. however. is not a l x i p s necessary. K e have recently ;hewn that tlie bios\-nthesis of indole alkaloids by Catharaizfhus cultures w i s induced n-lien lo-day-old cells \yere transferred into an 8% sucrose solution (34). The actkit!- of strictosidine synthase was not enhanced in the induction medium. whereas the activit?. of tr\-ptophan decarboxylase n-as increased manifold during the period of rapid alkaloid sccuniulation (34). Thus, the competence of cell cultures for production may already be achieved if the activity of the first enzynieia) diyerting priniarJ- metabolites are increased. If we know that the increased activitiei: of the enzyme;: leading into the secondary pathways are a prerequisite for high yielding cell line>. one may contemplate biochemical selection sy.-tems for high producing .strains which ma!- give rise to quite stable high yielding cell lines. ;ICKSOWLEDGE;\IEST T h e excellent technical assistance of 11s. -1.Zimnier? 11s. H . Gittermann and Mr. D . Pape is gratefully acknowledged. Recei:,ed 2 Jioie 1951

LITERA\TURE CITED 1. 11.T a b a t a , T. Ogino, K . l*oshioka, S . Toshikawa, and S . Hiraoka, "Frontiers of Plant

Tissue Culture 1978" [ E d . T. A . Thorpe,, Univ. of Calgary, Canada, p. 213. 2 . 11.H . Zenk, H . El-Shagi, H . hrens, J . Stockigt, E . W.Weiler, and B. Deus, "Plant Tissue Culture and Its Bio-technological .ipplication" ( E d . W.Barz, E. Reinhard, 11. H . Zenk), Springer-\.erlag, Berlin-Heidelberg-Sea- Tork 1977, p . 27. 3. A . W . .ilfermann, D . l l e r z , and E. Reinhard, Plant0 X e d . S u p p i . , 19i5, p. 70. 4 . T. N a t s u m o t o , T. Ikeda, S . Kanno, T . Kisaki, and 11.Soguchi, .4gric. Biol. Cheni., 44, 967 (1980 I . 5.

-6. I .

8. 9. 10. 11. 12. 13. 14. 15. 16. 1i. 18. 19

20. 21,

22. 23. 24.

25.

T. 11.Murphy. C . 11.Hamilton, and H . E . Street, Plniit Physiol., 64, 936 (1979). J . Berlin, L.P j n n r c n p h y s i o l . , 97, 317 (1980). K. Hahlbrock and H . Grisehach, A n n . Re;,.Plotit Physiol., 30, 105 11979). AI. H . Zenk, J . S n f . Prod., 43. 43s (1950). ,J. E. Palmer and .J. 11.Widholm, Plnitt Piiysiol.. 36, 233 (1975). ,J. Berlin and J . 11.Widholm, Ploiii Piixsiol., 59, 550 ( 1 9 i i ) . J. Berlin and K . - H . Knobloch, H o p p e S e y i e r ' s 2.Pliysiol. Chew:., 361, 219 (1980). J. Berlin and .J. 11.Kidholm? Z.-\-otilcforsch.q 34c, 631 (1978). *J. Berlin, Phy:ociiem., 20, 53 11981 I . T. Nurashige and F . Skoog, Physiol. Plaii!., 15, 473 (1962). G. Hofle. in preparation. R . -1.Wheaton and I . Stewart, -\-0/i~re! 206, 620 11965,. .\. Wehrli and E . Kovats, H e l l . Chitit. d c t i l , 42, 2709 (1959). K.-H. Knobloch and K . Hahlbrock, .Arch. Biocheni. B i o p h y s . , 184, 237 (1977). T . Buche and H . Sandermann?.Arch. Biochem. B i o p h y s . , 158, 445 (1973). 0 . H. Lowry, S . .J. Rosenbrough, .\. L. Farr and R.,J. Randall, 1.Bioi. Chew:. 193, 265 11951 1 .

J. C. Challice and A . H. Willianis, J . C h r o m a t o g . , 21, 357 (196G). 11. H . Zenk in "Frontiers of Plant Tissue Culture 1978" ( E d . T . -1.Thorpe), Univ. of Calgary, Calgar>-, Canada, p. 1. 8 . lfizusaki, I-.Tanabe, 11.Soguchi and E. Tamaki, Phyfochewiisiry, 10, 1347 (1971). F . Cabanne, *J. Martin-Tanguy and C . Martin, Piiysiol. I-eg., 15, 129 (1977). .I. Martin-Tanguy, F. Cabanne, E. Perdrizet and C. Martin, Phytochemistry, 17, 1927 (1978).

J . Martin-Tanguy, A . Deshayes, E . Perdrizet and C . Martin, FEBS Lett., 108, l i 6 (19i9). K . Hahlbrock and H . Grisehach, .-1tin. Re?. Plaint Physiol., 30, 105 (19i9). 11..J. l l o n t a g u e , T. .i. Armstrong R- E. G. Jaworski, Plnnt Physiol., 63, 341 (19i9). K.-H.Knohloch and .J. Berlin, Plant0 X e d . , 42: 167 (1981). B. L-lbrich, J. Stockigt and M.H . Zenk, Sat:irl;.issenschiiTten, 63, 184 (19iG). B . L-lbrich and 11. H . Zenk, Phyfocheniistry, 19, 1625 (1980). T.Amrhein, J . Schah and H . C . SteinMcken, .\ntitr;~,.isse,zscia~te~i,67, 356 11980). J . Berlin and L . Witte, Z . .~-\htii$orsch., 36c, 210 11981). 34. K.-H. Knobloch, B . Hansen, and J . Berlin: Z . a\-onturforscii., 36c, 40 (1981). 26. 27, 28. 29. 30. 31. 32. 33.