the seed protective coats, was not amenable to experimental manipulation. ... ylalanine (dopa) was followed according to Romeo and Preston. (13). Crude cell ..... Montague, M. J., Armstrong, T. A. & Jaworski, E. G. (1979). Plant Physiol.
Proc. Nadl. Acad. Sci. USA Vol. 82, pp. 2799-2803, May 1985
Developmental Biology
a-Amanitin resistance is developmentally regulated in carrot (amatoxins/somatic embryogenesis/cefl differentiation/oxidative detoxification/tyrosinase)
L. PITTO, F. Lo SCHIAVO, AND M. TERZI Istituto di Mutagenesi e Differenziamento, Consiglio Nazionale delle Richerche, Via Svezia, 10, 1-56100 Pisa, Italy
Communicated by G. Pontecorvo, December 26, 1984
on standard medium supplemented with 0.8% Oxoid agar for all lines. Cultivation was done at 250C in Erlenmeyer flasks on a gyratory shaker (80 rpm) under a light intensity of 500 lx. Cells were subcultured regularly every 15 days, inoculating 2 ml (packed cell volume after centrifugation for 10 min at 200 x g) in 50 ml of fresh medium. Plants were regenerated from 7-day-old cultures as described by Giuliano et al. (8) with minor modifications. The culture fraction utilized was sieved through 120-,um filters and retained by 50-tkm filters; the cell densities used were 10,000 clumps per ml for var. S. Valery cells and 1,000 clumps per ml for WOO1C and derivatives. Embryonal stages, when necessary, were purified with a routine procedure developed in this laboratory (9). ama' was determined spectrophotometrically: culture aliquots (at different times after addition of a-amanitin) were taken and spun at 10,000 x g for 5 min. The concentration of a-amanitin in the supernatant was determined from the extinction coefficient e305 = 15,400 M-l cm-l (10). Cycloheximide inactivation (CH') was estimated by the agar-gel diffusion assay as described by Sung et al. (2). Native, two-dimensional NaDodSO4/polyacrylamide gel electrophoresis was performed as described by Sung and Okimoto (3). Protoplast fusions were performed by the method of Kao and Michayluk (11). a3 protoplasts (106)-inactivated for 20 min with 0.3 M iodoacetate (12) (lithium salt; Calbiochem) at 25°C-were fused with equal amounts of C15. Four days afterwards the protoplast medium was diluted 1:1 with b5+, and after 4 more days, the cell aggregates detached from the dish were washed in b5+ and plated on top of agarized HAT medium. The colonies that developed in each dish were harvested by washing before they had reached full-grown size (20 days after plating). They were then resuspended and replated. After 1 week the cells were harvested again and set to grow in suspension. Tyrosinase (EC 1.14.18.1) activity was determined spectrophotometrically: the oxidation of L-3,4-dihydroxyphenylalanine (dopa) was followed according to Romeo and Preston (13). Crude cell extracts were made by grinding the frozen cell pellet in a mortar with liquid nitrogen, as described (14).
Carrot cells are capable of inactivating aABSTRACT amanitin only in embryogenic conditions (regenerating cells and embryoids). Instead, the mutant line a3 is capable of inactivating the drug also in nonembryogenic conditions (vegetative growth). The mutation is dominant in somatic hybrids and is pleiotropic, allowing expression during vegetative growth of other embryonal functions. The inactivation of a-amanitin is due to the oxidative activity of tyrosinase.
Research on plant development is much less advanced a field than its animal counterpart. For a long time this was due to the difficulty of access to the embryo that, being covered by the seed protective coats, was not amenable to experimental manipulation. Much of this difficulty could be circumvented after the definition of efficient and reproducible conditions for embryogenesis from somatic cell cultures (for a review on the developmental biology of somatic embryogenesis, see ref. 1). However, the lack of biochemical markers is still a major block, and finding a number of such markers has become a priority task. So far only two specific functions have been described during carrot somatic embryogenesis: cycloheximide resistance (2) and production of embryonal proteins El and E2 (3). In the present work we report that the capacity to inactivate a-amanitin (amai) is a developmentally regulated function: it is possessed by carrot cultured cells during organized growth (differentiating) and not during unorganized growth (in nonembryogenic medium). Moreover, the characterization of a mutant constitutively exhibiting amai will also be presented.
MATERIALS AND METHODS Our wild-type line (Daucus carota var. S. Valery) and its a-amanitin-resistant mutants a3 and aS have been described (4). C1S is a double mutant showing alterations in RNA polymerase II and in hypoxanthine phosphoribosyl-transferase (5). Line WOO1C (from wild carrot) and its derivatives WOO1, (5-methyltryptophan resistant) and WCH105 (cycloheximide resistant) are described by Sung (6) and Sung et al. (2). Line HA, (var. Juwarot) and its growing conditions also have been described (7). Basal media, obtained as powder from Flow, were Gamborg's b5 for var. S. Valery and derivatives and Murashige and Skoog for WOO1C and derivatives. The hormonal addition for WOO1C was 0.1 mg of dichlorophenoxyacetic acid (2,4-D) per liter. The standard medium (b5, nonembryogenic) for our var. S. Valery lines was b5 supplemented with 0.5 mg of 2,4-dichlorophenoxyacetic acid and 0.25 mg of 6-benzylaminopurine per liter. HAT medium was bS+ containing hypoxanthine (100 ,uM), aminopterin (100 ,uM), and glycine (100 ,M). Cell plating was done
RESULTS Carrot cells can reproduce vegetatively (unorganized growth) in the presence of auxin (nonembryogenic medium) or, in its absence, regenerate through several differentiative steps in which growth is organized and oriented to the productionin succession-of globular, heart- and torpedo-shaped embryos (1). a-Amanitin, a cyclic octapeptide showing an absorption peak at 305 nm, at 50 ,ug/ml completely blocks both orga-
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviations: amai and CHi, a-amanitin and cycloheximide in-
activating ability. 2799
2800
Developmental Biology: Pitto et al.
U
Proc. Natl. Acad. Sci. USA 82 (1985)
0.6
t
0.4
t9
0.2
260
280
300
320
340
260
300
280
320
340
Wavelength, nm FIG. 1. Spectrophotometric'reading of a-amanitin-containing b5' nonembryogenic medium (Left) and b5 embryogenic medium (Right) at 3, 6, and 9 days after the addition of the drug (time zero). Seven-day-old wild-type cells growing in b5+ medium were transferred at time zero (Left) to fresh b5 containing a-amanitin (50 pg/ml) at the standard cell density (-2.5 x 101 cell aggregates per ml) or, after washing (Right), to b5 basal medium at 104 cell aggregates per ml.
nized and unorganized growth (4). However, regenerating cells and embryos of all stages, after a lag lasting a few days, could resume differentiation in a normal way. In parallel, spectrophotometric reading of the regeneration medium showed a decrease of the a-amanitin absorption peak. Fig. 1 presents the time course of ama' in the case of differentiating cells. Curves of similar shape were obtained with purified embryonal forms of all stages (data not shown). ama' was present in carrot cells of different origin only during differentiation (Fig. 2). The different slopes of the inactivation curves were related to the embryogenic potential of the various lines. Fig. 3 shows that a cell line with null embryogenic potential (WOOl, 5methyltryptophan resistant) also lost ama'; aS, an a-amanitinresistant mutant whose resistance is due to an altered RNA polymerase II (3), after 1 year in culture lost much embryogenic potential and also some ama'. When one of its rare regenerants was made into callus, the regeneration potential was restored, and in the same way ama' was increased (Fig. 3). This a5 line, frequently rejuvenated to keep high embryogenic potential, was subsequently used as control instead of wild type for its capacity of vegetatively growing in the presence of a-amanitin without degrading it. In an attempt to determine the factors that influence the expression of this developmentally regulated inactivating function, we looked at cell density and fresh medium, which are known to play a fundamental role in somatic embryogenesis (3, 15). It appears that both high cell density and presence of auxin are necessary to prevent expression of ama'; fresh medium does not seem to play an important role. The factor responsible for ama' is not secreted into the medium, and the time course of inactivation is the same in fresh and old medium; "conditioned" media-i.e., media in which amanitin has been degraded-have no ama' left (after
the cells have been filtered out) nor inducing capacity for it (data not shown).
ce et 6
ct
._0 so
-U
0
3
6
9
12
15
Time, days FIG. 2. Time course of a-amanitin inactivation in nonembryogenic medium (o, 0, A) or embryogenic medium (e, m, A). o, *, WOO1C; A, *, aS; o, m, wt. The procedure was as in Fig. 1.
Developmental Biology:
Proc. Natl. Acad. Sci. USA 82 (1985)
Pitto et A
4. Therefore, a3 can be considered a constitutive mutant producing the embryonal function amai during vegetative growth. The next question we asked is whether this "misregulation" concerns only the expression of amai or if a more general switch was involved. Therefore, we measured the synthesis of El embryonal protein (Fig. 5) and CHi. Table 1 presents the results obtained with a3 and a5; other lines are presented for comparison. As can be seen, ama' and CHi were always expressed simultaneously. El might (as in lines originated from wild carrot) or might not (in domesticated carrot lines) follow the same regulation. The next question we tried to answer concerns the dominance of the mutation in a3. For this purpose we hybridized a3, inactivated with iodoacetate, with the double mutant line C15 (8-azaguanine resistant and a-amanitin resistant) and selected the hybrid in HAT medium (5); 50 to 100 colonies per dish were found. No colonies were found in the uniparental control dishes. The surviving colonies, in principle all hybrids, were harvested and replated. One week later they were harvested again (-106 cells per dish) and set to grow in suspension. One week later they were tested for production of ama'. As they did produce ama', this character can be considered dominant. Biochemical Characterization of ama1. a-Amanitin is likely to be degraded by an enzymatic process, but the specificity of the enzymatic reaction is not known. As there is a recent report in the literature of amai present in nontoxic fungi, which depends on a tyrosinase (13) (that, by oxidizing the hydroxytryptophan moiety and opening the indole ring, would also eliminate its specific absorbance at 305 nm), we tested the presence of tyrosinase activity in a3 and aS in embryogenic and nonembryogenic conditions. Table 2 shows that tyrosinase activity paralleled ama' in being constitutive in a3 and being induced during organized
100
C80 M
8 60
g 40 t \ 40~~~~
20
0
8
4
12
16
20
24
Time, days FIG. 3. a-Amanitin inactivation by cell lines with different embryogenic potential in regenerating conditions (embryogenic medium). *, W001; o, aS (1 year old) line; *, a5 (1 month old) line; A, a5 F, aS line freshly isolated from embryo-generated callus.
Characterization of the Constitutive Mutant. a-Amanitin blocks cell growth through a specific binding to RNA polymerase II (10). As we had isolated some a-amanitin-resistant mutants and had shown that only some of them had an altered RNA polymerase II (4), we thought it worthwhile to investigate whether the resistance of the remaining mutants could be due to ama'. That this was indeed the case is shown in Fig. 100
80
C~~~
S
c60
40
0
2
2801
3
4
Time, days FIG. 4. a-Amanitin inactivation by a3 mutant cell line. *, Nonembryogenic medium; A, embryogenic medium.
2802
_
Developmental Biology: Pitto et al.
4ft
-dw
Proc. Natl. Acad Sci. USA 82 (1985) go. w
_
Abla
--&'
AAAAW ..,-
'.
qb ,.
-
W
4m
a
141,
GP
.. .
.Imob..
0
m
v
_
_ -
If
40~~~
A_MM
.q0
_. _
.N .V
FIG. 5. Autoradiography showing the presence in two-dimensional electropherograms of El spot (arrow). The extracts were all from cells in nonembryogenic conditions. (Left) The entire picture is shown for a3. (Inset) Region around a3. (Right) This same region is shown for a3 (Top), a5 (Middle), and WOiC (Bottom).
growth in aS. That tyrosinase and ama' are indeed the same activity is indicated by the fact that ama' was inhibited in vivo by dopa (the degradation of a-amanitin in a standard ama' assay went from 84% to 12% in the presence of 5 mM dopa). In addition, the determinations of tyrosinase activity at various concentrations of substrate (dopa) showed that the Km is apparently increased from 1.42 mM to 2.08 mM in the presence of 0.167 mM a-amanitin. The two curves (with and without a-amanitin) intersect at the y axis in a double-reciprocal plot, indicating competitive inhibition (16). Table 1. Expression of embryonal functions in carrot cell lines Plant Embryogenic Embryo conditions amai CHi El production Cell line origin + + + +(2) +(3) WOOlC Wild aS
Domestic
+
+
+
+
+
a3
Domestic
+
+
+
+
+
_
+
+
+
_
Wild
+
-
-
-(3)
-
WCHlO5 Wild
+
+
+(2) +(3)
+
_
+
+
+
_
+
+
+
+
+ +
+
_
WOOl
HA
Domestic
+
Numbers in parentheses refer to references from which the data were obtained; the data have all been confirmed by us. See also ref. 1.
DISCUSSION Carrot somatic embryos and differentiating cells are capable of inactivating a-amanitin. In the same system, a-amanitin blocks transcription by RNA polymerase II in a matter of 15 min or less (14). As amaW is seen a few hours after vegetatively growing cells are transferred into embryogenic a-amanitincontaining medium, it is not clear how this amal can be expressed when RNA polymerase II is inactive. As wild type var. S. Valery cells' and aS (a mutant showing a RNA polymerase enzyme resistant to a-amanitin) do not differ in the time course of ama', models invoking escape or residual transcription carried out by RNA polymerase II seem unlikely. Of course, it may be that mRNA synthesis is not necessary for amai to be expressed, (post-transcriptional activation), or it is possible that the class of RNA to be synthesized can utilize RNA polymerase I or III (both amanitin insensitive) as suggested for other RNA species by Carlson and Ross (17). ama' is not present in vegetatively growing cells. It resides within differentiating cells, but it is not liberated into the Table 2. Tyrosinase specific activity (units/mg of protein) in carrot cells Cell line aS Ratio a3/a5 a3 Conditions 12 3 36 Embryogenic 5 15 0.3 Nonembryogenic
Developmental Biology: Pitto et al. medium. Cells kept at low density, even in the presence of hormones, show this activity, which is not present when auxin is added to high cell density-i.e., in growing conditions. Habituated cells, unable to differentiate, such as 5-methyltryptophan-resistant lines, do not show amai, no matter how cultivated, and there is a correlation between amai and regenerating potential. A constitutive mutant a3 has been isolated. It is capable of degrading amanitin also in nonembryogenic conditions and is an overproducer during regeneration. This mutation is pleiotropic: it also affects CH' expression. Conversely, WCH105, a mutant constitutive for CHi is also constitutive for ama'. The two constitutive mutants can be differentiated because a3 behaves as dominant and WCH105 as recessive (2) in somatic hybrids. Another embryonal function, El, also is expressed coordinately with CHi and ama' but only in carrot lines of wild origin. This pattern of expression suggests the following interpretation: the natural tendency of cultured carrot cells is toward embryogeny; the presence of auxin and high cell density prevent the process (3), but the initial expression of embryonal functions may occur even during vegetative growth to an extent that may be different in different lines. Therefore, the study of embryonal functions could help us to draw a temporal (or hierarchical) sequence of events. In this light WO0C would be the least embryogenic in showing no embryonal functions. WOO1, a habituated line unable to regenerate, would not express any of the embryonal functions; a5 would have acquired the capacity of expressing El; and a3-by mutation-would have acquired the capacity to express simultaneously amal and CH'. Another line, derived from domesticated carrot, the haploid line HA (7), would be the most embryogenic, expressing constitutively El, CHi, and ama', thus being similar to mutants such as a3 and WCH105. a3 is also an overproducer of tyrosinase. This enzyme is responsible for the degradation of a-amanitin in nontoxic fungi (13). Substrate competition experiments indicate that tyrosinase is not another coordinately expressed activity but rather the enzyme actually responsible for carrying out the ama' function. Another interesting question concerns the biological meaning of this ama'. It is possible that inactivating a-amanitin (or for that matter, cycloheximide) can be useful to the species that encounter those toxins in the soil. And, of course, this
Proc. Natl. Acad. Sci. USA 82 (1985)
2803
is particularly true for the seeds-i.e., during embryonal stages. It should be mentioned also that a good correlation was found between resistance to a-amanitin and eating habits of six Drosophila species (18). What happens is that Drosophila species feeding on mushrooms are resistant to a-amanitin for a mechanism other than resistant RNA polymerase, whereas the nonmycophagous species do not need this resistance mechanism. It would be interesting to know if the alternative mechanism of resistance depends on tyrosinase, the enzyme whose activity correlated well with the capacity for in vivo inactivation of a-amanitin in amanitin-accumulating and nonaccumulating Amanita species (13). This work was supported by Consiglio Nazionale delle Richerche, P.F. IPRA (publication 260). 1. Sung, Z. R., Fienberg, A., Chorneau, R., Borkird, C., Furner, I., Smith, J., Terzi, M., Lo Schiavo, F., Giuliano, G., Pitto, L.
& Nuti Ronchi, V. (1984) Plant Mol. Biol. Rep. 2, 3-14. 2. Sung, Z. R., Lazar, G. B. & Dudits, D. (1981) Plant Physiol. 68, 261-264. 3. Sung, Z. R. & Okimoto, R. (1981) Proc. Natl. Acad. Sci. USA 78, 3683-3687. 4. Vergara, M. R., Biasini, G., Lo Schiavo, F. & Terzi, M. (1982) Z. Pflanzenphysiol. 107, 313-319. 5. Lo Schiavo, F., Giovinazzo, G. & Terzi, M. (1983) Mol. Gen. Genet. 192, 326-329. 6. Sung, Z. R. (1979) Planta 145, 339-345. 7. Smith, J., Furner, I. & Sung, Z. R. (1981) In Vitro 17, 315-321. 8. Giuliano, G., Lo Schiavo, F. & Terzi, M. (1984) Theor. Appl. Genet. 67, 179-183. 9. Giuliano, G., Rosellini, D. & Terzi, M. (1983) Plant Cell Rep. 2, 216-218. 10. Cochet-Meilhac, M. & Chambon, P. (1974) Biochim. Biophys. Acta 353, 160-184. 11. Kao, K. N. & Michayluk, N. R. (1974) Planta 115, 355-367. 12. Cella, R., Carbonera, D. & lodarola, P. (1983) Z. Pflanzenphysiol. 112, 449-457. 13. Romeo, T. & Preston, J. F. (1984) Exp. Mycol. 8, 25-36. 14. Pitto, L., Lo Schiavo, F., Giuliano, G. & Terzi, M. (1983) Plant Mol. Biol. 2, 231-237. 15. Montague, M. J., Armstrong, T. A. & Jaworski, E. G. (1979) Plant Physiol. 63, 341-345. 16. Dixon, M. & Webb, E. C. (1979) Enzymes (Longman, New York), 3rd Ed., pp. 82-92. 17. Carlson, D. P. & Ross, J. (1983) Cell 34, 857-864. 18. Janicke, G., Grimaldi, D. A., Sluder, A. E. & Greenleaf, A. L. (1983) Science 221, 165-166.
