Planta
Planta (1982)155:473-477
9 Springer-Verlag 1982
Quantitative studies of bud initiation in cultured tobacco tissues Frederick Meins, Jr.*, Rachel Foster, and Joseph Lutz** Department of Botany and Department of Genetics and Development, University of IL, Urbana, Illinois 61801, USA
Abstract. After transferring leaf, pith, and stemcortex tissues of Nicotiana tabacum L. cv. " H a v a n a 425" from a complete medium containing auxin and cytokinin to an inductive medium with auxin deleted, there is lag phase of approx. 14d followed by a linear phase in which the rate of bud initiation is constant. The incidence of buds formed is very low, approx, one bud per 103 or 104 cells. Statistical analysis of the distribution of buds among explants and subcloning experiments provide evidence that the paucity of buds results from neither negative interactions among bud forming centers nor a paucity of cells with the potential for organogenesis. Our results are consistent with the hypothesis that the frequency of bud initiation is determined by the availability of competent cells, by position effects, or by a combination of both mechanisms.
and cytokinin are involved in the initiation process (Thorpe 1980). The present study is concerned with the problem of what factors regulate the incidence of bud initiation in cultured tissues. It is commonly observed that not all cells in tissue explants form buds. For example, tobacco tissues incubated on bud-inducing medium consist of a small number of meristematoid centers, some of which form buds, surrounded by numerous parenchyma cells (Thorpe and Murashige 1968, 1970). Thus, few cells in the explant appear to participate directly in bud formation. We will show that, in the case o f Nicotiana tabacum L. cv. " H a v a n a 425" tissues, the low incidence of buds arising under inductive conditions results from neither interactions among bud-forming centers nor a paucity of cells in the tissues with the potential for bud formation.
Key words: Auxin - Bud initiation - Competence - Cytokinin - Nicotiana (regeneration) - Tissue culture (position effects).
Materials and methods
Introduction Skoog and Miller (1957) established that cultured tobacco tissues could be induced to form roots, shoots, or complete plants by varying the relative concentrations of auxin and cytokinin in the culture medium. Although this method has been applied widely and with considerable success in regenerating plants from cultured tissues (review, Narayanaswamy 1977) it still is not known how organogenesis is initiated, whether shoots arise from single cells or groups of cells (Broertjes and Keen 1980; Henshaw et al. 1982), or how auxin * To whom correspondence should be addressed. Present address: Friedrich Miescher-Institut, P.O. Box 273, CH4002 Basel, Switzerland ** Present address: Crop GeneticsInternational, Dorsey, MD, 21076, USA
The cell lines used in this study were obtained from Nicotiana tabacum L. cv. " H a v a n a 425" plants grown in a greenhouse. The cloned lines of stem-cortex tissue, which are cytokininhabituated, and of leaf tissue, which are cytokinin-requiring, were described by Meins and Lutz (1979). Cloned, cytokininhabituated lines 259 H and 264 H, and cloned, cytokinin-requiring lines 239 N and 266 N were isolated from 35 ~ C-treated pithparenchyma tissue (Meins and Binns 1977). Line 239-75H is a cytokinin-habituated subclone of 239N (Meins et al. 1980).
Methods for culturing tissues are described in detail in Meins and Binns (1977) and Meins and Lutz (1980). Briefly, cytokinin-habituatedtissues were cultured on a standard medium, modifiedfrom Linsmaierand Skoog (1965), supplemented with 2.0 mg 1-1 of the auxin, a-naphthalene-aceticacid. Cytokinin-requiring tissues were cultured on this mediumfurther supplemented with 0.3 mg 1-1 of the cytokinin, kinetin (furfurylaminopurine). To induce bud formation, cube-shaped tissue explants weighingapprox. 20 mg and containingapprox. 5,000 cellswere incubated, 5 pieces per dish, in plastic Petri dishes (100 mm diameter, 20 mm high; Falcon Plastics, Oxnard, Cal., USA) containing 25 ml of the standard medium supplemented with 0.3 mg 1-1 kinetin. The dishes were sealed with Parafilm "M" Laboratory Film (American Canco., Greenwich, Conn., USA)
0032-0935/82/0155/0473/$01.00
474
F. Meins et al. : Bud-initiation in cultured tobacco tissues
5.0
2.O
.~
0 0
Io
20
30
40
50
Doys
Fig. 1. The time course of bud initiation obtained with cloned lines of pith, leaf and cortex tissue of Havana 425 tobacco. Cloned lines 239N ( o - - o ) and 239-75H ( o - - o ) of pith tissue, and cloned lines of cortex (zx--zx) and leaf ( o - - o ) tissue. Error bars: • SE calculated by the method of Lea and Coulson (1949) and incubated at 25 ~ C in light (Meins and Lutz 1980). In measuring the incidence of buds, only green, leafy shoots were scored. The statistical methods used are described by Simpson et al. (1960).
Results 1. Time Course for bud initiation. We estimated the rate of bud-initiation in sets of tissue explants using an indirect method based on the approach developed by Luria and Delbrfick (1943) to study bacterial mutation. In brief, this involved incubating from 10-20 explants of the same shape and size on an inductive medium and then measuring the fraction of explants that do not have at least one bud, P(0), with increasing time. It follows from the Poisson distribution law that the average number of bud-initiations per explant, m, can then be calculated from the relationship: m = - I n P(0). Since m is calculated from the fraction of explants without buds, the estimate is affected by neither interactions among buds in the same explant, nor the formation of multiple buds from the same initiation event. Figure 1 shows the time course for bud-initiation in cloned pith, leaf, and stem-cortex tissues of tobacco. There was an initial lag phase in which few or no buds were formed, followed by a linear phase in which the rate of initiation was roughly constant. In this experiment, the cytokinin habituated tissues (239-75H and cortex), exhibited a sig-
nificantly (P 1. In experiments where it was possible to count five or six buds per explant, 2 and var were calculated and agreement with the expected Poisson distribution was tested by the stringent variance ratio test (Simpson et al. ] 960). Figure 2 shows distributions of P(n) obtained at three times after the transfer of pith clone 266N to an inductive medium
F. Meins et al. : Bud-initiation in cultured tobacco tissues
20 | ~= 1.33 vat = 1.56
'~- -" 35
K
10
105, ~
15
Buds per explant (variance/mean)
% Colonies forming buds ~
Explants
Subclones
Explants
Subclonesd
0.93 • 0.09(2) a 1.06+0.20(4)
1.19+0.08 b 0.91_+0.06
]00 28.9
61.4 77.2
~ ~ , 7--;-'221-- , ,
15-(~)
1:1
Table 1. Comparison of the incidence of buds formed by two cloned lines of H a v a n a 425 tobacco pith tissue and subclones from these lines Cloned line
L__
0
475
Y= 2.49 _
1.92
"(~)
239N 259H
a Mean_+ SE for (n) measurements made after approx. 20 d in different experiments b Mean-t-SE for six measurements made at different times in the same experiment c After approx. 20 d on bud-inducing medium a 145 subclones from each cloned line assayed
R= 3.Bt+ var= 3.4"/
10'
--1
0 0 1 2 3 ~ 5 6~,7
guds/axp[anf { n )
Fig. 2 A - C . The distribution of buds per explant formed on a cloned pith tissue of H a v a n a 425 tobacco. Distributions obtained 15 (A), 22 (B), and 27 (C) d after transferring clone 266 N tissue onto inductive medium. Solid line, observed distribution; broken line, Poisson distribution calculated from the fraction of 49 explants without buds. N o n e of the distributions were significantly different at the 5% level from a Poisson distribution by the variance ratio test
and, for purposes of comparison, the expected Poisson distributions. There was good agreement between the observed and predicted distributions and these results were representative of several cloned lines of pith and leaf tissue tested. The observed distributions were not significantly different at the 5% level from a Poisson distribution, indicating that during the period of bud induction studied there is no interaction among bud-forming centers. Apparently, the low incidence of buds cannot be accounted for by early buds inhibiting the formation of later buds in the same explant.
3. The incidence of organogenetic subclones. Plant tissues often lose their capacity for organogenesis when serially propagated in culture (review, Skirvin 1978). Thus, it may be argued that few buds form in explants of cloned tissues because few cells in the explant are still organogenetic. To test this hypothesis, we isolated subclones from two cloned lines and incubated them on bud-inducing medium. Tissue explants, approximately the same size as the subclones, were excised from the parent tis-
sues and incubated on bud-inducing medium as well, for comparison. If the incidence of buds reflects the incidence of organogenetic cells in the parent tissues, then most subclones should not form buds, and a few - those derived from the organogenetic cells - should form buds in profusion. Therefore, the frequency of bud-forming subclones should be far lower than the frequency of bud-forming explants from the parent tissues, and, because of clustering, the distribution of buds among the subclones should not obey the Poisson law. Table 1 shows the results of bud induction experiments with small explants and subclones from two cloned lines of pith tissue. At least 60% of the subclones formed one or more buds after 20 d on the inductive medium. There was no significant difference between vat/2 values obtained with parent tissues and with their subclones. In both cases, these values were close to one indicating that buds were distributed according to the Poisson law. Therefore, it is unlikely that the paucity of buds can be accounted for by a paucity of cells in the parent tissues with the potential organogenesis.
4. Restoration of bud-forming capacity. To find out whether or not cells in regions of explants that do not form buds retain their potential for organogenesis, we incubated explants of a pith line, 266 N, on the inductive medium and measured the time course of bud initiation. After 35d, when most of the explants had formed buds, we excised pieces of tissues without visible buds from the bud-forming colonies. One set of bud-free explants was transferred onto the inductive medium; another set of these explants was pre-incubated for one week on a complete medium containing auxin and
476
F, Meins et al. : Bud-initiation in cultured tobacco tissues 2.4
1.6~ 1.2
t
0
~10
- "-'~-- - - 0 " * - - ~ - - ~-~ 20 50 40
"~
J 50
Days
Fig, 3. The effect of incubating a cloned line of Havana 425 tobacco pith tissue on inductive and non-inductive media. Tissue of cloned line 266 N on inductive medium ( o - - o ) . Explants from bud-free regions of explants forming buds incubated on inductive medium ( u - - u ) . Explants from bud-free regions of explants shown in ( o - - o ) preincubated for 7 d on complete medium and then transferred onto inductive medium (zx--zx). Error bars: _+SE for 55 explants calculated by the method of Lea and Coulson (1949)
cytokinin, and then transferred onto the inductive medium. Both sets of explants were then scored for bud initiation. The original tissue formed buds during the linear phase at a rate of 0.102+0.009 initiation per day per explant (_+standard error, N = 3) (Fig. 3). Explants from the bud-free regions of tissues did not form buds when incubated an additional 40 d on the inductive medium. In contrast, after pre-incubation on the complete medium, these explants formed buds at a rate of 0.064 + 0.006 initiations per day per explant, about 60% the rate of the original tissues. These results show that cells in the bud-free regions of induced explants retain their potential for organogenesis, which is expressed after incubation on the complete medium.
Discussion In agreement with earlier histological studies (Thorpe and Murashige 1968, 1970), we found that the incidence of buds initiated by cultured tobacco tissues is very low; approximately one bud is formed per 103 or 104 cells over a 20 d period. Statistical analysis of the distribution of buds showed that, at least during early stages of shoot regeneration, there is no significant interaction among bud-forming centers. This was then confirmed by the observation that bud-free regions of bud-forming explants, when incubated a second
time on the inductive medium in the absence of buds, still do not form buds. Early buds do not inhibit formation of later buds in the same tissue explant. Subcloning experiments provide evidence that the paucity of buds does not result from a paucity of organogenetic cells in the tissues. Most subclones isolated from cloned lines formed buds under inductive conditions. If there were a dramatic selection for organogenetic cells during the cloning process, then the incidence of buds obtained with subclones would be far higher than the incidence of buds obtained with explants of the parent cloned line. This was not observed. Thus, it is likely that the incidence of bud-forming clones provides a reliable estimate of the incidence of organogenetic cells in the parent cloned lines. The value obtained in this way, which was at least 10 -t, is 102-103 times the incidence of buds formed by the same tissue. There are far more cells present with the potential to form buds than the number of buds that form. This conclusion is also consistent with the observation that tissues that have not given rise to buds during the induction process form buds under inductive conditions after brief incubation on complete medium. Cells which did not form buds under inductive conditions retain the potential to do so. In the course of development, cells often change their competence to respond to inductive signals (Waddington 1966). Thus, it may be argued that although most or all cells in the tissue explants have the potential to form buds, only a few are competent to express this potentiality under the particular experimental conditions we used. According to this hypothesis, bud-forming tissues without buds do not form buds when excised and placed on the inductive medium a second time because they are devoid of competent cells. Thus, the discrepancy between the high frequency or organogenetic cells estimated by subcloning and the low frequency of bud formation expressed on a per cell basis reflects the fact that non-competent cells give rise to competent cells on the complete medium but not on the inductive medium. This is also consistent with the observation that preincubation of regions from bud-forming tissues without buds on complete medium restores the capacity to form buds. In principle, the competent state could be either transient or stable, i.e., transmitted to daughter cells when the competent cell divides. If one assumes that competent cells are stable and arise at low rates from non-competent cells in tissues grown on complete medium, then when subclones
F. Meins et al. : Bud-initiation in cultured tobacco tissues
are isolated from these tissues, those arising from competent cells will form many more buds than those arising from noncompetent cells. This will distort the uniform distribution of buds among the subclones and the value of var/2 should be greater than 1.0. Because this distortion was not observed in experiments with 145 subclones, the frequency of stable, competent cells can be estimated as being less than approx. 10 -3 (upper limit, 95% confidence level), which is roughly the same range as the frequency of bud formation expressed on a per cell basis. Thus, our cloning experiments neither support nor rule out the hypothesis that buds arise from a subpopulation of stable, competent cells present in the tissues. A second, less cumbersome, explanation for the incidence of buds focuses on the effects of position rather than on the properties of the cells. Thorpe and his associates (Ross et al. 1973; Maeda and Thorpe 1979) found that buds arise from specific regions of cultured tobacco tissues and proposed that physiological gradients established by contact of the tissues with the agar substratum determine where buds are formed. According to this interpretation, few buds form because few positions within the explant favor organogenesis. If it is position rather than the incidence of competent cells that determines the frequency of buds, then the incidence of buds in explants from subclones and the parent tissues should be roughly equal, and this was the case. Superficially, the observation t h a t explants from bud-free regions do not form buds when incubated on inductive medium is not consistent with a position hypothesis. Because the position of cells relative to the boundries of the explant and agar substratum is changed new bud-forming sites should be generated. The difficulty in interpreting these results is that tissue explants are transferred from the complete medium containing auxin and cytokinin, to an inductive medium with auxin deleted. It may be argued that the putative hormone gradients specifying bud sites are only present early in the induction period. Hence, as was observed, explants from regions without buds would only form buds after pre-incubation on the complete medium. In conclusion, the present experiments provide strong evidence that the low incidence of bud formation results from neither negative interactions among bud-forming centers nor a paucity of organogenetic cells in the tissues. Our results are consistent with two hypotheses, which are not necessarily mutually exclusive: I) buds are derived from relatively rare competent cells, which arise at low rates from a more abundant population of non-
477
competent cells in the tissue; and 2) buds arise at special positions in the tissue explant; and these positions are rare and randomly distributed among the tissue explants. We wish to thank Debra Mohnen and Pat King for their useful comments and criticism. This work was supported by grants No. CA 20053 from the National Cancer Institute, U.S. Public Health Service, and No. 78-10203 from the National Science Foundation.
References Broertjes, C., Keen, A. (1980) Adventitious shoots: Do they develop from one cell? Euphytica 29, 73 87 Henshaw, G.G., O'Hara, J.F., Webb, K.J. (1982) Morphogeneric studies in plant tissue cultures. In: IV. Symposium of the British Society for Cell Biology: Differentiation in vitro, pp. 231-251, Yeoman, M.M., Truman, D.E.S., eds. Cambridge University Press, Cambridge, U.K. Lea, D.E., Coulson, C.A. (1949) The distribution of the numbers of mutants in bacterial populations. J. Genet. 49, 264~285 Linsmaier, E.M., Skoog, F. (1965) Organic growth factor requirements of tobacco tissue cultures. Physiol. Plant. 18, 100-127 Luria, S.E., Delbriick, M. (1943) Mutation of bacteria from virus sensitivity to virus resistance. Genetics 28, 491-51I Maeda, E., Thorpe, T.A. (1979) Shoot histogenesis in tobacco callus cultures. In Vitro 15, 415-424 Meins, F., Jr., Binns, A.N. (1977) Epigenetic variation of cultured somatic cells: Evidence for gradual changes in the requirement for factors promoting cell division. Proc. Natl. Acad. Sci. USA 74, 2928 2932 Meins, F., Jr., Lutz, J. (1980) The induction of cytokinin habituation in primary pith explants of tobacco. Planta 149, 402-407 Narayanaswamy, S. (1977) Regeneration of plants from tissue cultures. In: Plant cell, tissue, and organ culture, pp. 179206, Reinert, J., Bajaj, Y.P.S., eds. Springer, Berlin Heidelberg New York Ross, M.K., Thorpe, T.A., Costerton, J.W. (1973) Ultrastructural aspects of shoot initiation in tobacco callus cultures. Am. J. Bot. 60, 788-795 Simpson, G.G., Roe, A., Lewontin, R.C. (1960) Quantitative zoology. Harcourt, Brace and World, New York Chicago San Francisco Atlanta Skirvin, R.M. (1978) Natural and induced variation in tissue culture. Euphytica 27, 241-266 Skoog, F., Miller, C.O. (1957) Chemical regulation of growth and organ formation in plant tissue cultivated in vitro. Symp. Soc. Exp. Biol. 11, 118-131 Thorpe, T.A. (1980) Organogenesis in vitro : structural, physiological and biochemical aspects. Int. Rev. Cytol. (Suppl.) l l A , 71-111 Thorpe, T.A., Murashige, T. (1968) Starch accumulation in shoot-forming tobacco callus cultures. Science 160, 421-422 Thorpe, T.A., Murashige, T. (1970) Some histochemical changes underlying shoot initiation in tobacco callus cultures. Can. J. Bot. 48, 277-285 Waddington, C.H. (1966) New patterns in genetics and development. Columbia University Press, New York London Whittaker, R.H. (1975) Communities and ecosystems, pp. 6769, 2nd edn. Macmillan, New York Received 7 December 1981 ; accepted 12 June 1982
Planta (1982)155:473-477
9 Springer-Verlag 1982
Quantitative studies of bud initiation in cultured tobacco tissues Frederick Meins, Jr.*, Rachel Foster, and Joseph Lutz** Department of Botany and Department of Genetics and Development, University of IL, Urbana, Illinois 61801, USA
Abstract. After transferring leaf, pith, and stemcortex tissues of Nicotiana tabacum L. cv. " H a v a n a 425" from a complete medium containing auxin and cytokinin to an inductive medium with auxin deleted, there is lag phase of approx. 14d followed by a linear phase in which the rate of bud initiation is constant. The incidence of buds formed is very low, approx, one bud per 103 or 104 cells. Statistical analysis of the distribution of buds among explants and subcloning experiments provide evidence that the paucity of buds results from neither negative interactions among bud forming centers nor a paucity of cells with the potential for organogenesis. Our results are consistent with the hypothesis that the frequency of bud initiation is determined by the availability of competent cells, by position effects, or by a combination of both mechanisms.
and cytokinin are involved in the initiation process (Thorpe 1980). The present study is concerned with the problem of what factors regulate the incidence of bud initiation in cultured tissues. It is commonly observed that not all cells in tissue explants form buds. For example, tobacco tissues incubated on bud-inducing medium consist of a small number of meristematoid centers, some of which form buds, surrounded by numerous parenchyma cells (Thorpe and Murashige 1968, 1970). Thus, few cells in the explant appear to participate directly in bud formation. We will show that, in the case o f Nicotiana tabacum L. cv. " H a v a n a 425" tissues, the low incidence of buds arising under inductive conditions results from neither interactions among bud-forming centers nor a paucity of cells in the tissues with the potential for bud formation.
Key words: Auxin - Bud initiation - Competence - Cytokinin - Nicotiana (regeneration) - Tissue culture (position effects).
Materials and methods
Introduction Skoog and Miller (1957) established that cultured tobacco tissues could be induced to form roots, shoots, or complete plants by varying the relative concentrations of auxin and cytokinin in the culture medium. Although this method has been applied widely and with considerable success in regenerating plants from cultured tissues (review, Narayanaswamy 1977) it still is not known how organogenesis is initiated, whether shoots arise from single cells or groups of cells (Broertjes and Keen 1980; Henshaw et al. 1982), or how auxin * To whom correspondence should be addressed. Present address: Friedrich Miescher-Institut, P.O. Box 273, CH4002 Basel, Switzerland ** Present address: Crop GeneticsInternational, Dorsey, MD, 21076, USA
The cell lines used in this study were obtained from Nicotiana tabacum L. cv. " H a v a n a 425" plants grown in a greenhouse. The cloned lines of stem-cortex tissue, which are cytokininhabituated, and of leaf tissue, which are cytokinin-requiring, were described by Meins and Lutz (1979). Cloned, cytokininhabituated lines 259 H and 264 H, and cloned, cytokinin-requiring lines 239 N and 266 N were isolated from 35 ~ C-treated pithparenchyma tissue (Meins and Binns 1977). Line 239-75H is a cytokinin-habituated subclone of 239N (Meins et al. 1980).
Methods for culturing tissues are described in detail in Meins and Binns (1977) and Meins and Lutz (1980). Briefly, cytokinin-habituatedtissues were cultured on a standard medium, modifiedfrom Linsmaierand Skoog (1965), supplemented with 2.0 mg 1-1 of the auxin, a-naphthalene-aceticacid. Cytokinin-requiring tissues were cultured on this mediumfurther supplemented with 0.3 mg 1-1 of the cytokinin, kinetin (furfurylaminopurine). To induce bud formation, cube-shaped tissue explants weighingapprox. 20 mg and containingapprox. 5,000 cellswere incubated, 5 pieces per dish, in plastic Petri dishes (100 mm diameter, 20 mm high; Falcon Plastics, Oxnard, Cal., USA) containing 25 ml of the standard medium supplemented with 0.3 mg 1-1 kinetin. The dishes were sealed with Parafilm "M" Laboratory Film (American Canco., Greenwich, Conn., USA)
0032-0935/82/0155/0473/$01.00
474
F. Meins et al. : Bud-initiation in cultured tobacco tissues
5.0
2.O
.~
0 0
Io
20
30
40
50
Doys
Fig. 1. The time course of bud initiation obtained with cloned lines of pith, leaf and cortex tissue of Havana 425 tobacco. Cloned lines 239N ( o - - o ) and 239-75H ( o - - o ) of pith tissue, and cloned lines of cortex (zx--zx) and leaf ( o - - o ) tissue. Error bars: • SE calculated by the method of Lea and Coulson (1949) and incubated at 25 ~ C in light (Meins and Lutz 1980). In measuring the incidence of buds, only green, leafy shoots were scored. The statistical methods used are described by Simpson et al. (1960).
Results 1. Time Course for bud initiation. We estimated the rate of bud-initiation in sets of tissue explants using an indirect method based on the approach developed by Luria and Delbrfick (1943) to study bacterial mutation. In brief, this involved incubating from 10-20 explants of the same shape and size on an inductive medium and then measuring the fraction of explants that do not have at least one bud, P(0), with increasing time. It follows from the Poisson distribution law that the average number of bud-initiations per explant, m, can then be calculated from the relationship: m = - I n P(0). Since m is calculated from the fraction of explants without buds, the estimate is affected by neither interactions among buds in the same explant, nor the formation of multiple buds from the same initiation event. Figure 1 shows the time course for bud-initiation in cloned pith, leaf, and stem-cortex tissues of tobacco. There was an initial lag phase in which few or no buds were formed, followed by a linear phase in which the rate of initiation was roughly constant. In this experiment, the cytokinin habituated tissues (239-75H and cortex), exhibited a sig-
nificantly (P 1. In experiments where it was possible to count five or six buds per explant, 2 and var were calculated and agreement with the expected Poisson distribution was tested by the stringent variance ratio test (Simpson et al. ] 960). Figure 2 shows distributions of P(n) obtained at three times after the transfer of pith clone 266N to an inductive medium
F. Meins et al. : Bud-initiation in cultured tobacco tissues
20 | ~= 1.33 vat = 1.56
'~- -" 35
K
10
105, ~
15
Buds per explant (variance/mean)
% Colonies forming buds ~
Explants
Subclones
Explants
Subclonesd
0.93 • 0.09(2) a 1.06+0.20(4)
1.19+0.08 b 0.91_+0.06
]00 28.9
61.4 77.2
~ ~ , 7--;-'221-- , ,
15-(~)
1:1
Table 1. Comparison of the incidence of buds formed by two cloned lines of H a v a n a 425 tobacco pith tissue and subclones from these lines Cloned line
L__
0
475
Y= 2.49 _
1.92
"(~)
239N 259H
a Mean_+ SE for (n) measurements made after approx. 20 d in different experiments b Mean-t-SE for six measurements made at different times in the same experiment c After approx. 20 d on bud-inducing medium a 145 subclones from each cloned line assayed
R= 3.Bt+ var= 3.4"/
10'
--1
0 0 1 2 3 ~ 5 6~,7
guds/axp[anf { n )
Fig. 2 A - C . The distribution of buds per explant formed on a cloned pith tissue of H a v a n a 425 tobacco. Distributions obtained 15 (A), 22 (B), and 27 (C) d after transferring clone 266 N tissue onto inductive medium. Solid line, observed distribution; broken line, Poisson distribution calculated from the fraction of 49 explants without buds. N o n e of the distributions were significantly different at the 5% level from a Poisson distribution by the variance ratio test
and, for purposes of comparison, the expected Poisson distributions. There was good agreement between the observed and predicted distributions and these results were representative of several cloned lines of pith and leaf tissue tested. The observed distributions were not significantly different at the 5% level from a Poisson distribution, indicating that during the period of bud induction studied there is no interaction among bud-forming centers. Apparently, the low incidence of buds cannot be accounted for by early buds inhibiting the formation of later buds in the same explant.
3. The incidence of organogenetic subclones. Plant tissues often lose their capacity for organogenesis when serially propagated in culture (review, Skirvin 1978). Thus, it may be argued that few buds form in explants of cloned tissues because few cells in the explant are still organogenetic. To test this hypothesis, we isolated subclones from two cloned lines and incubated them on bud-inducing medium. Tissue explants, approximately the same size as the subclones, were excised from the parent tis-
sues and incubated on bud-inducing medium as well, for comparison. If the incidence of buds reflects the incidence of organogenetic cells in the parent tissues, then most subclones should not form buds, and a few - those derived from the organogenetic cells - should form buds in profusion. Therefore, the frequency of bud-forming subclones should be far lower than the frequency of bud-forming explants from the parent tissues, and, because of clustering, the distribution of buds among the subclones should not obey the Poisson law. Table 1 shows the results of bud induction experiments with small explants and subclones from two cloned lines of pith tissue. At least 60% of the subclones formed one or more buds after 20 d on the inductive medium. There was no significant difference between vat/2 values obtained with parent tissues and with their subclones. In both cases, these values were close to one indicating that buds were distributed according to the Poisson law. Therefore, it is unlikely that the paucity of buds can be accounted for by a paucity of cells in the parent tissues with the potential organogenesis.
4. Restoration of bud-forming capacity. To find out whether or not cells in regions of explants that do not form buds retain their potential for organogenesis, we incubated explants of a pith line, 266 N, on the inductive medium and measured the time course of bud initiation. After 35d, when most of the explants had formed buds, we excised pieces of tissues without visible buds from the bud-forming colonies. One set of bud-free explants was transferred onto the inductive medium; another set of these explants was pre-incubated for one week on a complete medium containing auxin and
476
F, Meins et al. : Bud-initiation in cultured tobacco tissues 2.4
1.6~ 1.2
t
0
~10
- "-'~-- - - 0 " * - - ~ - - ~-~ 20 50 40
"~
J 50
Days
Fig, 3. The effect of incubating a cloned line of Havana 425 tobacco pith tissue on inductive and non-inductive media. Tissue of cloned line 266 N on inductive medium ( o - - o ) . Explants from bud-free regions of explants forming buds incubated on inductive medium ( u - - u ) . Explants from bud-free regions of explants shown in ( o - - o ) preincubated for 7 d on complete medium and then transferred onto inductive medium (zx--zx). Error bars: _+SE for 55 explants calculated by the method of Lea and Coulson (1949)
cytokinin, and then transferred onto the inductive medium. Both sets of explants were then scored for bud initiation. The original tissue formed buds during the linear phase at a rate of 0.102+0.009 initiation per day per explant (_+standard error, N = 3) (Fig. 3). Explants from the bud-free regions of tissues did not form buds when incubated an additional 40 d on the inductive medium. In contrast, after pre-incubation on the complete medium, these explants formed buds at a rate of 0.064 + 0.006 initiations per day per explant, about 60% the rate of the original tissues. These results show that cells in the bud-free regions of induced explants retain their potential for organogenesis, which is expressed after incubation on the complete medium.
Discussion In agreement with earlier histological studies (Thorpe and Murashige 1968, 1970), we found that the incidence of buds initiated by cultured tobacco tissues is very low; approximately one bud is formed per 103 or 104 cells over a 20 d period. Statistical analysis of the distribution of buds showed that, at least during early stages of shoot regeneration, there is no significant interaction among bud-forming centers. This was then confirmed by the observation that bud-free regions of bud-forming explants, when incubated a second
time on the inductive medium in the absence of buds, still do not form buds. Early buds do not inhibit formation of later buds in the same tissue explant. Subcloning experiments provide evidence that the paucity of buds does not result from a paucity of organogenetic cells in the tissues. Most subclones isolated from cloned lines formed buds under inductive conditions. If there were a dramatic selection for organogenetic cells during the cloning process, then the incidence of buds obtained with subclones would be far higher than the incidence of buds obtained with explants of the parent cloned line. This was not observed. Thus, it is likely that the incidence of bud-forming clones provides a reliable estimate of the incidence of organogenetic cells in the parent cloned lines. The value obtained in this way, which was at least 10 -t, is 102-103 times the incidence of buds formed by the same tissue. There are far more cells present with the potential to form buds than the number of buds that form. This conclusion is also consistent with the observation that tissues that have not given rise to buds during the induction process form buds under inductive conditions after brief incubation on complete medium. Cells which did not form buds under inductive conditions retain the potential to do so. In the course of development, cells often change their competence to respond to inductive signals (Waddington 1966). Thus, it may be argued that although most or all cells in the tissue explants have the potential to form buds, only a few are competent to express this potentiality under the particular experimental conditions we used. According to this hypothesis, bud-forming tissues without buds do not form buds when excised and placed on the inductive medium a second time because they are devoid of competent cells. Thus, the discrepancy between the high frequency or organogenetic cells estimated by subcloning and the low frequency of bud formation expressed on a per cell basis reflects the fact that non-competent cells give rise to competent cells on the complete medium but not on the inductive medium. This is also consistent with the observation that preincubation of regions from bud-forming tissues without buds on complete medium restores the capacity to form buds. In principle, the competent state could be either transient or stable, i.e., transmitted to daughter cells when the competent cell divides. If one assumes that competent cells are stable and arise at low rates from non-competent cells in tissues grown on complete medium, then when subclones
F. Meins et al. : Bud-initiation in cultured tobacco tissues
are isolated from these tissues, those arising from competent cells will form many more buds than those arising from noncompetent cells. This will distort the uniform distribution of buds among the subclones and the value of var/2 should be greater than 1.0. Because this distortion was not observed in experiments with 145 subclones, the frequency of stable, competent cells can be estimated as being less than approx. 10 -3 (upper limit, 95% confidence level), which is roughly the same range as the frequency of bud formation expressed on a per cell basis. Thus, our cloning experiments neither support nor rule out the hypothesis that buds arise from a subpopulation of stable, competent cells present in the tissues. A second, less cumbersome, explanation for the incidence of buds focuses on the effects of position rather than on the properties of the cells. Thorpe and his associates (Ross et al. 1973; Maeda and Thorpe 1979) found that buds arise from specific regions of cultured tobacco tissues and proposed that physiological gradients established by contact of the tissues with the agar substratum determine where buds are formed. According to this interpretation, few buds form because few positions within the explant favor organogenesis. If it is position rather than the incidence of competent cells that determines the frequency of buds, then the incidence of buds in explants from subclones and the parent tissues should be roughly equal, and this was the case. Superficially, the observation t h a t explants from bud-free regions do not form buds when incubated on inductive medium is not consistent with a position hypothesis. Because the position of cells relative to the boundries of the explant and agar substratum is changed new bud-forming sites should be generated. The difficulty in interpreting these results is that tissue explants are transferred from the complete medium containing auxin and cytokinin, to an inductive medium with auxin deleted. It may be argued that the putative hormone gradients specifying bud sites are only present early in the induction period. Hence, as was observed, explants from regions without buds would only form buds after pre-incubation on the complete medium. In conclusion, the present experiments provide strong evidence that the low incidence of bud formation results from neither negative interactions among bud-forming centers nor a paucity of organogenetic cells in the tissues. Our results are consistent with two hypotheses, which are not necessarily mutually exclusive: I) buds are derived from relatively rare competent cells, which arise at low rates from a more abundant population of non-
477
competent cells in the tissue; and 2) buds arise at special positions in the tissue explant; and these positions are rare and randomly distributed among the tissue explants. We wish to thank Debra Mohnen and Pat King for their useful comments and criticism. This work was supported by grants No. CA 20053 from the National Cancer Institute, U.S. Public Health Service, and No. 78-10203 from the National Science Foundation.
References Broertjes, C., Keen, A. (1980) Adventitious shoots: Do they develop from one cell? Euphytica 29, 73 87 Henshaw, G.G., O'Hara, J.F., Webb, K.J. (1982) Morphogeneric studies in plant tissue cultures. In: IV. Symposium of the British Society for Cell Biology: Differentiation in vitro, pp. 231-251, Yeoman, M.M., Truman, D.E.S., eds. Cambridge University Press, Cambridge, U.K. Lea, D.E., Coulson, C.A. (1949) The distribution of the numbers of mutants in bacterial populations. J. Genet. 49, 264~285 Linsmaier, E.M., Skoog, F. (1965) Organic growth factor requirements of tobacco tissue cultures. Physiol. Plant. 18, 100-127 Luria, S.E., Delbriick, M. (1943) Mutation of bacteria from virus sensitivity to virus resistance. Genetics 28, 491-51I Maeda, E., Thorpe, T.A. (1979) Shoot histogenesis in tobacco callus cultures. In Vitro 15, 415-424 Meins, F., Jr., Binns, A.N. (1977) Epigenetic variation of cultured somatic cells: Evidence for gradual changes in the requirement for factors promoting cell division. Proc. Natl. Acad. Sci. USA 74, 2928 2932 Meins, F., Jr., Lutz, J. (1980) The induction of cytokinin habituation in primary pith explants of tobacco. Planta 149, 402-407 Narayanaswamy, S. (1977) Regeneration of plants from tissue cultures. In: Plant cell, tissue, and organ culture, pp. 179206, Reinert, J., Bajaj, Y.P.S., eds. Springer, Berlin Heidelberg New York Ross, M.K., Thorpe, T.A., Costerton, J.W. (1973) Ultrastructural aspects of shoot initiation in tobacco callus cultures. Am. J. Bot. 60, 788-795 Simpson, G.G., Roe, A., Lewontin, R.C. (1960) Quantitative zoology. Harcourt, Brace and World, New York Chicago San Francisco Atlanta Skirvin, R.M. (1978) Natural and induced variation in tissue culture. Euphytica 27, 241-266 Skoog, F., Miller, C.O. (1957) Chemical regulation of growth and organ formation in plant tissue cultivated in vitro. Symp. Soc. Exp. Biol. 11, 118-131 Thorpe, T.A. (1980) Organogenesis in vitro : structural, physiological and biochemical aspects. Int. Rev. Cytol. (Suppl.) l l A , 71-111 Thorpe, T.A., Murashige, T. (1968) Starch accumulation in shoot-forming tobacco callus cultures. Science 160, 421-422 Thorpe, T.A., Murashige, T. (1970) Some histochemical changes underlying shoot initiation in tobacco callus cultures. Can. J. Bot. 48, 277-285 Waddington, C.H. (1966) New patterns in genetics and development. Columbia University Press, New York London Whittaker, R.H. (1975) Communities and ecosystems, pp. 6769, 2nd edn. Macmillan, New York Received 7 December 1981 ; accepted 12 June 1982
