close

Вход

Забыли?

вход по аккаунту

?

00221589.1973.11514504

код для вставкиСкачать
Journal of Horticultural Science
ISSN: 0022-1589 (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/thsb19
Flower Bud Dormancy in Coffea Arabica L. I.
Studies of Gibberellin in Flower Buds and Xylem
Sap and of Abscisic Acid in Flower Buds in Relation
to Dormancy Release
G. Browning
To cite this article: G. Browning (1973) Flower Bud Dormancy in Coffea Arabica L. I.
Studies of Gibberellin in Flower Buds and Xylem Sap and of Abscisic Acid in Flower Buds
in Relation to Dormancy Release, Journal of Horticultural Science, 48:1, 29-41, DOI:
10.1080/00221589.1973.11514504
To link to this article: http://dx.doi.org/10.1080/00221589.1973.11514504
Published online: 27 Nov 2015.
Submit your article to this journal
View related articles
Citing articles: 4 View citing articles
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=thsb19
Download by: [UNSW Library]
Date: 28 October 2017, At: 16:34
J. hort. Sci. (1973) 48, 29-41
Flower bud dormancy in Co.lfea arabica L. I. Studies of
gibberellin in flower buds and xylem sap and of abscisic acid in
flower buds in relation to dormancy release
By G. BROWNINGt
Coffee Research Foundation, P.O. Box 4, Ruiru, Kenya
Downloaded by [UNSW Library] at 16:34 28 October 2017
SUMMARY
Changes were studied in gibberellin and abscisic acid in coffee flower buds and
gibberellin in xylem sap when bud dormancy was released by rainfall or irrigation.
Gibberellin levels in the buds increased rapidly, while those in the xylem sap
remained unchanged. Bud gibberellin stopped rising when rapid expansion began
and then decreased as this proceeded. The absolute amount of abscisic acid in
buds remained steady prior to increasing just before anthesis. On a fresh weight
basis levels declined during rapid water uptake but recovered as anthesis
approached. The resumption of active growth leading to blossoming may be
regulated by the liberation of free gibberellin from a bound form in the buds,
but a second, xylem-transported, stimulus could also be involved.
A column chromatographic technique using silicic acid for the separation
of gibberellin and abscisic acid is described.
COFFEE trees blossom a short time after rainfall (Went, 1917) or irrigation (Porteres, 1946),
without which the flower buds stop growing once their microspore mother cells have
matured (Mes, 1957). When stimulated, the flower buds resume active growth even on
laterals which have ~een ring-barked, tipped (Mes, 1957) or defoliated (Browning, unpublished) and probably therefore respond to a stimulus originating either in the roots
or within themselves or in both.
Coster (1926) attributed dormancy release to the rapid fall in temperature which
often accompanies rainfall in the tropics, and evidence supporting his idea has since accumulated. Using controlled environments, both Went (1957) and Mes (1957) were able to
make regularly watered trees blossom by transferring them from 30 °C day/24 oc night
regimes to 23 oc day/17 ac night regimes. Trees growing outdoors in nutrient solutions
have been observed to blossom synchronously with trees in the field after rainfall (Franco,
1962), as have trees in the greenhouse at Ruiru, Kenya (Browning, 1971). At Ruiru blossoming in most cases follows rainfall only when this is accompanied by a temperature
fall of at least 3 ac in 45 minutes or less (M. G. R. Cannell, personal communication).
For irrigation to be effective the trees must first be exposed to water stress (Piringer and
Borthwick, 1955; Alvim, 1960a and 1960b) and water then allowed to permeate the root
zone (Mathew and Chokkana, 1964). Water in the soil must thus reach a critical level for
any buds to respond (Porteres, 1946).
t On secondment from Long Ashton Research Station, University of Bristol.
Downloaded by [UNSW Library] at 16:34 28 October 2017
30
Flower bud dormancy in coffee. I
The internal mechanisms responsible for these responses could well be hormonal,
since the endogenous stimulus can be replaced by exogenous gibberellin (van der Veen,
1968; Browning, 1971) or cytokinin (Browning, 1971) applied directly to the buds, and
in the case of irrigation it may be antagonized by exogenous abscisic acid (van der Veen,
1968). Extracts of coffee flower buds are known to contain abscisic acid (Browning et al.,
1970). Moreover, whereas exogenous gibberellin must be supplied to promote flowering
in frequently irrigated trees growing in the greenhouse (Pagacz, 1959), before buds
respond to cytokinin the trees must first have been exposed to water stress (Browning,
1971). It is possible, therefore, that the responses to rapid temperature drops and water
stress might be mediated by changes in the gibberellin status of the buds, the part played
by xylem transport being to supply cytokinin. The work described below was designed to
test this hypothesis by measuring gibberellin levels in flower buds and xylem sap as
dormancy was released. Gibberellin is synthesized by the roots and exported in the xylem
sap to the shoots in several species (Butcher, 1963; Phillips and Jones, 1964; Carr eta/.,
1964; Jones and Phillips, 1966; Skene, 1967; Jones and Lacey, 1968; Reid and Burrows,
1968; Reid et a/., 1969). Parallel determinations of abscisic acid in the flower buds were
also undertaken in an attempt to clarify the part it plays in the dormancy mechanism.
MATERIALS AND METHODS
Plant material
Flower buds and xylem sap were collected at Ruiru, Kenya, before and after rain
induced bud break from 6-year-old single-stem trees, cvs. SL 34 and 28, and from multiplestem 8-year-old trees, cv. K7, respectively. Flower buds and xylem sap were also collected
from SL 28 trees when flowering was induced by overhead irrigation after three weeks
drought. Flower buds were frozen on dry ice, lyophilized, and ground to a powder. The
fresh and dry weights of 50 buds were recorded after each collection. Sap was collected
from 10 stems at each sampling by water displacement (100 ml) under pressure (10 lb/inch2
nitrogen gas) (Browning 1971). In order to reduce bacterial contamination I ml of 2%
streptomycin was added before displacement to each flask receiving perfusate. Stem fresh
weights were recorded before perfusion. After evaporation in vacuo the sap residue was
dissolved in lO ml 50% aqueous methanol, added to lOg Whatman No. 1 cellulose powder
and then dried for 24 hours at 50 °C. The lyophilized buds and cellulose were air freighted
to Long Ashton Research Station, England, where they were deep frozen until extracted.
Extraction and purification
The lyophilized bud powder (100 g) was extracted, and the extract partitioned according
to the procedure of Browning eta/. (1970) for abscisic acid. After filtration the solid residue
was extracted for cytokinins (Browning, 1971). The acidic ethyl acetate extract was purified
on a 12 x2 em column of the formate form of Dowex I (50-100 mesh) ion exchange resin
and the extract recovered in 400 ml 1 M formic acid. After removing the formic acid by
evaporation in vacuo and then flushing with nitrogen gas, the extract was dissolved in 2 ml
of 10% aqueous methanol and added to 2 g (100-200 mesh) silicic acid (Mallinckrodt,
Silicar CC-4). After drying with nitrogen gas the silicic acid was placed on top of 8 g of
Downloaded by [UNSW Library] at 16:34 28 October 2017
G.
BROWNING
31
partially hydrated silicic acid (0.6 ml water per g) in a 8 x 2 em column. The column was
eluted with 400 ml freshly redistilled di-isopropyl-ether, which removed approximately
90% of the abscisic acid and approximately 5% of the gibberellin. The remaining gibberellin
was recovered by eluting the column with 400 ml ethyl acetate saturated with 0.5 M formic
acid (the eluate had a pH of approximately 3.0). Similar separations of authentic abscisic
acid and gibberellic acid could be obtained by this procedure. After evaporation in vacuo
both fractions were run on thin layer chromatograms.
Xylem sap was recovered from the cellulose powder by eluting with 500 ml of water
(no further biological activity could be obtained by subsequently eluting with methanol).
Volumes equivalent to 750 g fresh weight stem were removed, and after adjusting their
pH to 3.0 with 2N HCl they were partitioned five times with half volumes of ethyl acetate.
After evaporation in vacuo the pooled ethyl acetate extracts were purified on a Dowex I
column as for bud extracts and run on TLC.
Only redistilled solvents were used. The extractions and solvent partitions were carried
out at 3 oc in the dark and all other operations were done under subdued light. Extracts
were evaporated in vacuo using a rotary film evaporator at temperatures not exceeding 40 oc.
Thin layer chromatography (TLC)
,..m
This was conducted using solvent-washed 400
layers of Kieselgel GF.254 (Merck)
or Kieselgur (Merck) spread on 20 X 20 em plates. Extracts and marker spots of abscisic
acid and gibberellic acid were applied to the plates in the manner described by Milborrow
(1967). After development, the GF.254 plates were examined under a UV lamp and the
UV absorbing bands marked. Authentic standards run on Kieselgur were made visible
by examining the plates under UV light after spraying with 5% sulphuric acid in ethanol
and heating for 5 minutes at 100 °C. The adsorbent was scraped from the plates in bands
corresponding to 0. I RF values and eluted with water-saturated ethyl acetate. Small volumes
of the extract were later dried onto 4.25 em Whatman No. I filter paper discs in 4.5 em
diameter petri dishes for elution and bioassay.
Bioassays
Coleoptile assay
This was based on the assay developed by Nitsch and Nitsch (1956). Extracts were
eluted overnight at 3 oc in 1 ml 0.2% sucrose buffered at pH 5.0 with a phosphate/citrate
buffer and assayed for abscisic acid (ABA) using ten 10 mm long sections cut 3 mm below
the tip from 72 hours old wheat coleoptiles, cv. Atle. The sections were incubated with
the test solutions in the dark for 20 hours at 25 oc and then measured to the nearest millimeter.
Lettuce hypocotyl assay (Frankland and Wareing, 1960)
Extracts were eluted overnight at 3 oc with I ml distilled water from the filter paper
discs, which had previously been impregnated with 0.5 ,..g zeatin. Ten selected pregerminated seeds, cv. Grand Rapids, were placed in each petri dish. Standard solutions
from 0.001-10 ,_,g GA 3 /ml were also assayed each time. Hypocotyls were measured to the
32
Flower bud dormancy in coffee. I
nearest millimeter after the dishes had been left for 72 hours in a growth cabinet illuminated
by fluorescent tubes and maintained at 25 oc.
Downloaded by [UNSW Library] at 16:34 28 October 2017
Rumex senescence retardation assay
This was based OI). the assay developed by Whyte and Luckwill (1966). Extracts were
eluted overnight at 3 °C in 1 ml distilled water, and ten 7 mm diameter leaf discs, cut from
leaves detached and pre-aged in the dark at 25 oc for 24 hours, were added to each petri
dish. The dishes were run in triplicate. When a full range response to the standards (0.00011.0 p.g GA 3/ml) was obtained, each lot of ten discs was extracted in 10 ml methanol for
24 hours and the optical density of the solution measured at 665 p.m.
Extracts within each series from bud break to anthesis were bioassayed at the time.
Statistical analysis
Responses at the zones corresponding to each RF value were tested for statistical
significance at P<O.Ol.
Quantitative bioassays were based on the parallel line technique (Finney, 1952),
abscisic acid thus being estimated as described by Browning et al. (1970).
Gibberellin extracts were assayed at three ten-fold dilutions. The responses (y) were
transformed to Logits (Z) by calculating Z = i- loge
(A~y)' where A= the empirically
estimated saturation response. Because the transformed data could no longer be assumed
to have uniform variance, the weighting W = y2(A-y)2 was applied. The dose response
curves obtained were tested for linearity and parallelism using regression analysis. The
difference in mean response for each sample from the grand mean response, calculated
from the pooled slopes, was then used to estimate the equivalent log concentration and
hence the relative concentration. The least significant difference (L.S.D.) for each comparison was also calculated. Concentrations of abscisic acid, relative concentrations of
gibberellin and L.S.D.s were determined on a dry weight basis. Transformation to bud and
fresh weight units was obtained by adjusting the values for dry weights appropriately.
RESULTS
Whilst gibberellin activity could be detected by bioassay after TLC of bud extracts
from preliminary extractions, it proved difficult to separate all the three components found
from inhibitory activity associated with abscisic acid. The resolution obtained with seven
different solvent systems is shown in Table I, which also gives the RF values for authentic
GA 3 and ABA.
The problem was overcome using the silicic acid column procedure described previously. Goren and Goldschmidt (1970) have reported that abscisic acid-like inhibitors in
·citrus fruit can be partitioned at pH 6.0 into di-isopropyl ether, leaving most of the gibberellin in the aqueous phase. Further steps to obtain unambiguous measurements of activity
were taken by including 0.5 ppm zeatin in the test media assayed with lettuce hypocotyls.
The abscisic acid-inhibited promotion of hypocotyl growth by gibberellin is overcome by
·cytokinin (Sankhla and Sankhla, 1968). The lettuce system was thus used for quantitative
G. BROWNING
33
TABLE I
RF values in seven solvent systems on TLC using Kieselgel
or Kieselgur for the sap and flower bud gibberellins
Solvent
systemt
RF values
GA activity
RF values
GA3
RF values
ABA
1
2
0.0-0.2
0.0-0.2
0.3-0.5
0.8-\.0
0.0-0.3
0.4-0.6
0.8-1.0
0.3-0.7
0.2-0.5
0.7-0.9
0.2-0.4
0.5-0.85
0.12
0.17
0.10
0.88
0.05
0.89
0.15
0.86
0.26
0.08
0.49
0.29
0.70
0.65
3
4
Downloaded by [UNSW Library] at 16:34 28 October 2017
5
6
7
t The solvents used were as follows: (1) Di-isopropyl
ether:acetic acid (95:5); (2) Carbon tetrachloride: acetic
acid: water (50:19:31) on Kieselgur, lower phase+20%
ethyl acetate ; (3) Benzene : ethyl acetate : acetic acid
(50:5:2); (4) Water; (5) N-butanol:n-propanol:ammonia:water(6:2:1 :2); (6) Ethyl acetate; (7) Isopropanol:
water :ammonia (80: 19.95:0. 05).
determinations of gibberellin by assaying bud or sap extracts directly without prior chromatography on TLC. Corroborative evidence was obtained by running aliquots from each
extract on Kieselgur plates equibrilated in carbon tetrachloride:water: acetic acid (50:19:31)
upper phase and developed with lower phase+20% ethyl acetate, and assaying with the
Rumex test. This solvent separated abscisic acid from two of the three bands of gibberellin
activity (see Table 1.)
Buds collected 2-4 days after 'blossom' showers consistently yielded extracts which, vwhen subjected to these procedures, were found to contain much higher levels of gibberellin
than those detected in extracts of dormant buds. The pattern taken by these increases
emerged clearly in bioassays of extracts from buds collected at intervals from induction
until an thesis and can be seen in Figure I, which shows the relative concentrations calculated
after extracts from buds in the irrigation experiment were assayed at 5, 0.5 and 0.05 g/ml
with the lettuce test.
Comparing these changes with those recorded for bud fresh and dry weights, the
increase in gibberellin appeared to precede the rapid rise in water content, and presumably
therefore water uptake from the xylem, which started as the buds began to expand and
elongate 4-5 days after dormancy release was induced. This can be seen from Figure 2,
which shows the bud weights recorded in the irrigation experiment.
For one bud series, the fresh and dry weights of three replicate, as opposed to one,
subsamples of 50 buds were recorded at each time to provide an estimate of error. From
these data it was calculated that a significant (P<0.05) increase in water content had not
occurred until the fourth sampling, in this case the sixth day after the 'blossom' shower.
The pattern of bud growth and water uptake described above and depicted in Figure 2
corresponds closely with that reported by Mathew and Chokkana (1964't for coffee growing
3
34
Flower bud dormancy in coffee. I
10·0
..c
·..,"
...."
c
10·0 :;;;
.,.
...;,.
.c
1·0
1·0
~
g
~
.:"'
0·1
0·1
.
...
;"
0
~
0 01
0·01
"'
.....
0
1!"
.,.
0
.....
...
<
C)
Downloaded by [UNSW Library] at 16:34 28 October 2017
~
c
ll
"
~
.,.
c
.2
c
g
C>
0·001
0
6
c
9
10
11
12
Ocys after irrigation
FIG. 1
Relative gibberellin concentrations in flower buds collected just before and at two-day intervals after
overhead irrigation, assayed with the lettuce test.
The closed circles show concentrations calculated in bud units and the open ones fresh weight units.
The L.S.D.s at P<O. 05 for some comparisons are given below:
2-0
4-2
8-4
8-0
10--2
10-0
Difference in gibberellin levels between days compared
L.S.D. for bud unit basis
4.0
3.4
3.1
6.2
2.8
8.3
L.S.D.forfreshweightbasis
4.1
2.8
3.4
4.7
2.3
5.9
in India. Once rapid growth got under way, absolute levels of gibberellin began to decline
and were diluted as the water content increased, until at anthesis the initial increase was
reversed, with the absolute levels being even lower (P <0.05) than those in dormant buds.
There appeared to be little difference between the magnitude of the rise in bud gibberellin
associated with 'blossom' showers and that in the irrigation experiment. That the increase
detected in acidic gibberellins was not accompanied by changes in their ratios can be seen
from Figure 3, which shows the activity detected with the Rumex assay following TLC
of extracts in a bud series collected after rainfall. The rainfall which induced the flowering
to which the data referred was accompanied by a rapid temperature drop.
Abscisic acid was estimated quantitatively after running 50 g equivalents of the
abscisic acid extract on Kieselgel GF.254 plates in benzene :ethyl acetate :acetic acid (50 :5 :2),
which separates the relatively inactive trans, trans isomer from the natural cis, trans
(Milborrow, 1968). The solvent was allowed to develop half way up the plate which wa:s
removed, dried, and then run again allowing the solvent to develop the whole distance.
After assaying zones corresponding to each RF at 5 g equivalentsfml, the zone with inhibitory
activity and UV absorption corresponding to the RF of the authentic abscisic acid marker
was assayed at 10 g equivalents/ml and tenfold dilutions down to 0.01 g equivalents/mi.
Only very slight traces of a UV absorbing band corresponding to the trans, trans isomer
(RF 0.25) were seen.
Absolute levels of abscisic acid in buds were not significantly different until 4 days
before anthesis, when an increase was found. On a fresh weight basis, however, there was
G.
35
BROWNING
75
75
70
70
60
60
...E so
50
...
::1
J:J
i
40""
e
40
.... ...,..,.
30
i
~
~
Downloaded by [UNSW Library] at 16:34 28 October 2017
....
::>
-
30
~
.I::
c
i
<C(
~
oo
20
10
0
~10
-------0
2
3
7
6
5
4
Days a rter irrigation
8
9
10
11
12
FIG. 2
Fresh (open circles) and dry weights (closed circles) of flower buds collected just before and at two-day
intervals after irrigation.
a significant dilution once the buds started to take up water rapidly, although the levels later
began to recover. The relationship between abscisic acid concentration changes on a fresh
weight and bud unit basis can be seen in Figure 4. The 95% confidence interval on a dry
weight basis for each estimate is given below the caption.
Representative bioassay results on a dry weight basis for thin layer chromatograms
of abscisic acid extracts from buds collected before and after rain-induced bud break are
shown in Figure 5. The responses shown were approximately one order of magnitude less
than the saturation response.
After chromatography on GF.254 plates developed in isopropanol:water:ammonia
(80: 19.95 :0.05), xylem sap extracts equivalent to approximately four stems (1 kg fresh
weight) yielded only very slight inhibitory activity and UV absorbance corresponding to
authentic abscisic acid. Correlative changes in xylem sap abscisic acid were therefore not
examined. Gibberellin extracts from xylem sap were run on Kieselgur plates as for bud
extracts, and zones corresponding to each RF assayed at 100 g stem fresh weight equivalents/
ml with the Rumex test. As shown in Figure 6 no differences in activity could be detected,
nor were changes in the relative activity at each zone apparent. Corroborative evidence
was obtained by determining relative concentrations directly by assaying at 200, 20 and 2 g
stem fresh weight equivalents/ml with the lettuce assay. The gibberellins found in xylem sap
extracts ran at the same RF values in the solvent systems shown in Table I as those resolved
from bud extracts.
36
Flower bud dormancy in coffee. I
c
500
A
300
· -· • ·---- ·-· • •• ---- •• • ---·0·01 ppm
GA3
200
·0·001 ppm
GA3
50
Downloaded by [UNSW Library] at 16:34 28 October 2017
500
8
D
- --O·Ippm
GA3
0·0
0·2
0·4
Rf
0·6
0·8
0z
0·0
1·0
0·4
0·6
0·8
1·0
Rf
500
E
._
"'.X ;;..
400
..
0
300
..
-------------------·0·01 ppm
0
0
..
~
0
...,,.._
GA3
... 0
50
0·0
O·Z
0·4
Rf
FIG.
0·6
0·8
1·0
3
Rumex assay of buds collected (A) three days before, and (B) three days, (C) five days, (D) seven days,
(E) ten days after rainfall.
Extracts assayed at 5 g dry weight equivalents/ml after chromatography on Kieselgur with carbon tetrachloride:acetic acid:water (50:19:31) lower phase+20% ethyl acetate. Darkened areas indicate activity
significant at P<O.Ol.
DISCUSSION
The data clearly are consistent with the hypothesis identifying gibberellin as the
stimulus for dormancy release generated in coffee flower buds. This is particularly evident
from the phasing in gibberellin changes, which implied a causal rather than coincidental
relationship between gibberellin supply and the resumption of active growth. The signifi-
G.
37
BROWNING
0·1
10·0
0·075 ..;
:.
~
"0
...,
:::0
0
0
~ 1·0
.
0·050 ;
...
...
0.
a>
...
""
::t..
.;.
-
0
Downloaded by [UNSW Library] at 16:34 28 October 2017
.....
...a.
..."'
...
0·025 ::t.
...
.....0
.c
c:
... 0·01
0·1
0
2
3
4
6
Days after irrigation
8
9
10
11
12
FIG. 4
Wheat coleoptile assay of ABA in flower buds collected just before and at two-day intervals after irrigation.
The closed circles show concentrations calculated in bud units and the open ones fresh weight units.
The 95% confidence interval for each determination is given below in dry weight units:
12
10
8
6
4
2
0
Day
0.22
0.13
0.13
0.11
0.12
0.13
0.15
Confidenceinterval
0.46
0.28
0.30
0.18
0.22
0 29
0.29
(l'g/g dry weight)
cance of the data describing changes in ABA levels is less clear, and raises the question of
whether in this case steady-state measurements will provide insight into the part it plays
in the dormancy mechanism. For example, the failure to detect a decline in absolute levels
during dormancy release is difficult to reconcile with the action as a primary block for
transcription or translocation in dormancy systems which has been suggested for ABA
(van Overbeek eta/., 1967; Wareing eta/., 1968). On the other hand, without information
about the localization of ABA and gibberellin in relation to their site(s) of action, such
judgements are probably premature. The same can be said for the decline detected in ABA
concentrations calculated on a fresh weight basis, which appeared to indicate a dilution
following from the rapid water uptake when active growth began again.
If gibberellin is the stimulus operating in the buds, the dormancy mechanism would
probably hinge on those processes regulating its supply. Attempts to prevent dormancy
being released by spraying buds with B9 or CCC (Browning, 1971), which in many plants
inhibit gibberellin biosynthesis, have failed, thus seeming to rule out a mechanism involving
de novo gibberellin biosynthesis at the time release occurs. Of the two alternatives, the data
obtained here can be argued to preclude supply from the xylem sap, and hence the gibberellin
may therefore have been converted or released in the buds from a bound or other inactive
form.
That the factors regulating gibberellin supply may indeed provide the key to the
dormancy mechanism has emerged from recent work suggesting that, while it can prevent
natural flowering (van der Veen, 1968), exogenous ABA does not antagonize the dormancy-
38
Flower bud dormancy in coffee. I
24
2~
22
21
A
20
E
E
-...go
.t:
_.
19
·18
17
16
15.
------------- ------- -1-0ppm
Downloaded by [UNSW Library] at 16:34 28 October 2017
14
A8A
14
24
23
22
21
21
E
E
.r.:
20
19
20
c:
""
18
18
17
16
17-
""
--'
19
16f
15-
15
14
14
00
0·2
04
Rf
0·6
1·0
1·0
FIG. 5
Wheat coleoptile assay of buds collected (A) three days before and (B) three days, (C) five days, (D) ten
days after rainfalL
Extracts assayed at 5g dry weight equivalents/ml after chromatography on Kieselgel in benzene:ethyl
acetate:acetic acid (50:5:2). Darkened areas indicate activity significant at P<O.Ol.
releasing action of exogenous gibberellin (Browning, 1972). Thus, not only is it unlikely
that the gibberellin responsible for dormancy release is imported from the xylem sap,
at least in a free form, but the role of ABA might be to regulate gibberellin supply in the
buds, conceivably by determining the equilibrium between free and bound forms.
Although the interpretation offered here relies heavily on the weight given to the data
showing the sequential difference in gibberellin changes and water uptake, it does find
support in the reports ofPagacz (1959) and van der Veen (1968), who were unable to obtain
flowering on frequently watered trees unless the flower buds were treated with exogenous
gibberellin. It is also consistent with evidence that for natural flowering in the field when
water in the soil is plentiful, rapid temperature drops are required (Browning, 1972). In
fact, there clearly is some justification for connecting rapid temperature drops directly
with the release of bud gibberellin. Similarly, as exposure to water stress is reported to
increase the response of dormant buds to exogenous gibberellin (Alvim, 1960a; Browning.
G.
...c -...
u
300
A
0
.f ~
~
300
200
u
.a 0
<Ct'"- 100
. ._ - - - . - - - - - . -
u 0
c !:;
~ :;
::; u
..
0
.a.._
Downloaded by [UNSW Library] at 16:34 28 October 2017
<0
200
100
50
c
-0·001~~~
~~GA3
~
50
. - 300
39
BROWNING
B
100
50
300
~~
200
100
50
D
~~
L:=J
0·0
FIG.
6
Rumex assay of xylem sap collected (A) just before, and (B) four days, (C) eight days, (D) twelve days
after irrigation.
Extracts assayed at 100 g fresh weight equivalents/ml after chromatography on Kieselgur with carbon
tetrachloride:acetic acid:water (50:19:31) lower phase+20% ethyl acetate. Darkened areas indicate activity
significant at P<O.Ol.
1971), temperature drop might therefore affect the buds in the same way, explaining how
these climatic factors are able to replace each other in their effect on flowering. Given the
evidence now available, it thus appears possible that two endogenous stimuli operate, one
supplied from the xylem sap and entrained to the soil moisture status, and the other being
gibberellin generated within the flower buds themselves. From the phasing of changes in
bud gibberellin and water uptake, these might have to act sequentially, and, if the xylemtransported factor is cytokinin, evidence for which will be presented in a forthcoming
paper, then the inability of exogenous cytokinin to induce flowering when the trees have
not been exposed to water stress (Browning, 1971) could thereby be explained. One
difficulty with this hypothesis, however, is the ability of exogenous gibberellin on its own
to break dormancy; this can nevertheless be overcome by postulating that the level of
gibberellin in the buds determines their sink strength for xylem sap cytokinin, with external
applications achieving this to such an extent that the soil moisture status no longer is a
limiting factor.
Another difficulty with the hypothesis is the similarity detected in the phasing of
changes accompanying rainfall- and irrigation-induced blossoming. However, recent work
in the field indicates that when the overhead irrigation was applied in July conditions
may not have favoured the fulfilment of the stress requirement (Browning, 1972), and so
the irrigation itself may have stimulated a rapid temperature drop. In view of this, studies
of hormone changes associated with exposure to water stress and of flowerings subsequently
induced by ground irrigation will now be undertaken.
The author wishes to thank Mr C. R. Baines for help with the statistical analysis,
Dr L. C. Luckwill for his interest in the project and for useful comments on the manuscript,
40
Flower bud dormancy in coffee. 1
and F. H. Hoffman-La Roche & Co. for a sample of abscisic acid. During this work the
author was supported by an Overseas Development Administration Junior Fellowship.
Downloaded by [UNSW Library] at 16:34 28 October 2017
REFERENCES
ALVIM, P. de T. (1958). Estimulo de Ia floraci6n y fructifici6n del cafeto por aspersiones
con acido giberelico. Turrialba, 8, 64-72.
ALVIM, P. de T. (1960a). Moisture stress as a requirement for flowering of coffee. Science,
N.Y., 132, 354.
ALVIM, P. de T. (l960b). Physiology of growth and flowering in coffee. Turrialba, 2, 57-62.
BROWNING, G., HOAD, G. V., and GASKIN, P. (1970). Identification of abscisic acid in·
flower buds of Coffea arabica L. Planta, 94, 213-19.
BROWNING, G. (1971). The hormonal regulation of flowering and cropping in Coffea
arabica L. Ph. D. Thesis, Univ. Bristol.
BROWNING, G. (1972). Physiological studies of Coffea arabica. Rep. Long Ashton Res. Stn
for 1971, 41-44.
BUTCHER, D. M. (1963). The presence of gibberellin in excised tomato roots. J. exp. Bot.,
14, 272-80.
CARR, D. S., REID, D. M., and SKENE, K. G. M. (1964). The supply of gibberellin from the
root to the shoot. Planta, 63, 382-92.
CoSTER, C. (1926). Periodische Bltiteerscheinungen in den Tropen. Ann. lard. bot. Buitenz.,
35, 125-62.
FINNEY, D. J. (1952). Statistical method in biological assay. Griffin, London.
FRANCO, C. M. (1962). Cur so internacional de fisiologia vegetal. Piracicaba, Brazil.
FRANKLAND, B., and WAREING, P. F. (1960). Effect of gibberellic acid on hypocotyl growth
of lettuce seedlings. Nature, Lond., 185, 255-6.
GOREN, R., and GOLDSCHMIDT, E. E. (1970). Regulative systems in the developing citrus
fruit I. The hormonal balance in orange fruit tissues. Physiologia Pl., 17, 697-709.
JoNES, 0. P., and LACEY, H. J. (1968). Gibberellin-like substances in the xylem stream of
apple and pear trees. J. exp. Bot., 19, 526-31.
JONES, R. L., and PHILLIPS, I. D. J. (1966). Organs of gibberellin synthesis in light grown
sunflower plants. Pl. Physiol., Lancaster, 41, 1381-6.
LANG, A. (1970). Gibberellins: structure and metabolism. A. Rev. Pl. Physiol., 21, 537-70.
MATHEW, P. K., and CHOKKANA, H. G. (1964). Studies on the intake of water and nutrients
during development of flower buds to blossoms in coffee. Indian Coffee, 25, 264-88.
MES, M. G. (1957). Studies on the flowering of Coffea arabica L. I. The influence of temperature on the initiation and growth of coffee flower buds. II. Breaking the dormancy
of coffee flower buds. Port. Acta bioi., 4, 328-56.
MILBORROW, B. V. (I 967). The identification of ( )-abscisin II [( )-dormin)] in plants
and the measurement of its concentrations. Planta, 16, 93-113.
MILBORROW, B. V. (1968). Identification and measurement of ( )-abscisic acid in plants.
In Biochemistry and physiology of plant growth substances, 1531-45. Ed. Wightman,
F., and Setterfield, G. Runge Press, Ottawa.
+
+
+
Downloaded by [UNSW Library] at 16:34 28 October 2017
G. BROWNING
41
NITSCH, J. P., and NITSCH, C. (1956). Studies on the growth of coleoptile and first internode
sections. A new, sensitive, straight growth test for auxins. Pl. Physiol.,- Lancaster,
41, 94-111.
VAN OvERBEEK, J., LoEFFLER, J. E., and MASON, M. I. R. (1967). Dormin (abscisin II)
inhibitor of plant DNA synthesis? Science, N.Y., 156, 1497-9.
PAGACZ, E. A. (1959). Quelques considerations sur la fioraison du cafetier. Bull. agric.
Congo. beige., 50, 1531-8.
PHILLIPS, I. D. J., and JONES, R. L. (1964). Gibberellin-like activity in bleeding sap of root
systems of Helianthus unnuus detected by a new dwarf pea epicotyl assay and other
methods. Planta, 63, 269-78.
PIRINGER, A. A., and BORTHWICK, H. A. (1955). Photoperiodic responses of coffee.
Turrialba, 5, 72-7.
PORTERES, R. (1946). Action de l'eau, apres une periode seche, sur le declenchement de la
floraison chez Coffea arabica L. Agron. trop., Nogent, 1, 148-58.
REID, D. M., and BuRROWS, W. J. (1968). Cytokinin-like and gibberellin-like activity in
the spring sap of trees. Experimenta, 24, 189-90.
REID, D. M., CROZIER, A., and HARVEY, B. M. R. (1969). The effects of flooding on the
export of gibberellins from the root to the shoot. Planta, 89, 376--9.
SANKHLA, N., and SANKHLA, D. (1968). Reversal of ( +) abscisin II induced inhibition of
lettuce seed germination and seedling growth by kinetin. Physio/ogiaPl., 21, 190-5.
SKENE, K. G. M. (1967). Gibberellin-like substances in root exudate of Vilis vinifera.
Planta, 74, 250-62.
VAN DER YEEN, R. (1968). Plant hormones and flowering in coffee. Acta bot. neerl., 17,
373-6.
WAREING, P. F., Goon, J., and MANUEL, J. (1968). Some possible physiological roles of
abscisic acid. In Biochemistry and physiology of plant growth substances, 1561-79.
Ed. Wightman, F., and Setterfield, G. Runge Press, Ottawa.
WENT, F. A. F. C. (1917). Periodische Erscheinungen beim Bliihen tropischer Gewaschse.
Naturwissenschaften, 5, 72-6.
WENT, F. W. (1957). The experimental control of plant growth. Chronica Botanica Co.,
Waltham, Massachusetts.
WHYTE, P., and LucKWILL, L. C. (1966). A sensitive bioassay for gibberellins based on
retardation of leaf senescence in Rumex obtusifolius. Nature, Lond., 210, 3160.
(Received 20th March, 1972.)
Документ
Категория
Без категории
Просмотров
0
Размер файла
870 Кб
Теги
1973, 11514504, 00221589
1/--страниц
Пожаловаться на содержимое документа