Planta (1996)198:595 603
P l m a t ~
@)Springer-Verlag 1996
The development of phloem anastomoses between vascular bundles
and their role in xylem regeneration after wounding in Cucurbita
and Dahlia
Roni Aloni 1, John R. Barnett 2
i Department of Botany, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel
2 School of Plant Sciences, The University of Reading, Whiteknights, PO Box 221, Reading, RG6 6AS, UK
Received: 11 July 1995/Accepted: 17 August 1995
Abstract. The differentiation of phloem anastomoses
linking the longitudinal vascular bundles has been studied
in stem internodes of Cucurbita maxima Duchesne,
C. pepo L. and Dahlia pinnata Cav. These anastomoses
comprise naturally occurring regenerative sieve tubes
which redifferentiate from interfascicular parenchyma
cells in the young internodes. In all three species, severing
a vascular bundle in a young internode resulted in regeneration of xylem to form a curved by-pass immediately
around the wound. The numerous phloem anastomoses in
these young internodes were not involved in this process,
the regenerated vessels originating from interfascicular
parenchyma alone. Conversely, in mature internodes of
Dahlia, the regenerated vessels originated from initials of
the interfascicular cambia, and their phloem anastomoses
did not influence the pattern of xylogenesis. On the other
hand, in old internodes of Cucurbita, in which an interfascicular cambium was not yet developed, the parenchyma
cells between the bundles had lost the ability to redifferentiate into vessel elements, and instead, regenerated vessels
were produced in the phloem anastomoses. Thus, the
wounded region of the vascular bundle was not bypassed
via the shortest, curved pathway, but by more circuitous
routes further away from the wound. Some of the regenerated vessels produced in the phloem anastomoses were
extremely wide, and presumably efficient conductors of
water. It is proposed that the dense network of phloem
anastomoses developed during evolution as a mechanism
of adaptation to possible damage in mature internodes by
providing flexible alternative pathways for efficient xylem
regeneration in plants with limited or no interfascicular
cambium.
Key words: C u c u r b i t a
Dahlia
Phloem (anastomoses,
development, function) - Xylem (regeneration, efficiency)
This paper is dedicated to the memory of the late Isaac Blachmann
(deceased 19 November 1995),father-in-law of the senior author, for
encouragement and advice throughout the years
Correspondence to: R. Aloni; FAX: 972 (3) 640 9380;
E-mail: [email protected]
Introduction
Phloem anastomoses through the interfascicular
parenchyma connecting the phloems of the longitudinal
vascular strands are common in higher plants. A single
internode grown under favorable summer conditions typically contains several hundred to a few thousand such
anastomoses (Aloni and Sachs 1973; Roberts et al. 1988).
The anastomoses are variable in size, consisting of
bundles of one or more sieve tubes, but are difficult to
visualise in conventional light-microscope sections because of their sinuous nature. They may, however, be
studied in cleared and stained thick preparations in which
whole anastomoses are present (Aloni and Sachs 1973).
Recently, phloem anastomoses were also found to be
common in tumour tissues. In Agrobacterium tumefaciensinduced crown galls grown on stems of Ricinus communis,
numerous phloem anastomoses were observed between
the vascular bundles (Aloni et al. 1995).
A study by Aloni and Peterson (1990) aimed at determining whether the phloem anastomoses are functional
demonstrated that under normal circumstances the anastomoses are not involved in translocation. However, when
longitudinal vascular strands of Dahlia pinnata were severed, the anastomoses in the vicinity of the wound became active transporters, indicating that they serve as an
emergency system capable of by-passing damaged parts
of the phloem system and maintaining translocation by
providing alternative pathways for assimilates around the
stem.
Phloem development is induced by low auxin levels,
while for xylem differentiation there is need for higher
auxin concentrations (Aloni 1980, 1987b). It has been
suggested that phloem anastomoses are induced by lowlevel streams of auxin (Aloni 1987a, 1995). Normally, the
phloem anastomoses are not associated with xylem elements, and equivalent xylem anastomoses are not found
in the stem. However, when high auxin concentrations
were applied to decapitated Luffa cylindrica stems, regenerated xylem elements were formed within some of the
phloem anastomoses (Aloni 1987a, 1995). This suggests
that the phloem anastomoses can provide preferred pathways for auxin flow between the longitudinal vascular
596
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
strands. If this suggestion is true, one might expect that
when a vascular b u n d l e is severed in an i n t e r n o d e which
has n u m e r o u s p h l o e m anastomoses, the polar auxin flow
descending from a b o v e the w o u n d would move laterally
t h r o u g h the p h l o e m anastomoses. O u r w o r k i n g hypothesis was that if the c o n c e n t r a t i o n of this lateral a u x i n flow
is high e n o u g h it w o u l d induce xylem regeneration in
some of the p h l o e m anastomoses.
We have studied the n o r m a l d e v e l o p m e n t of p h l o e m
a n a s t o m o s e s in intact stems of p l a n t species which typically have t h o u s a n d s of these a n a s t o m o s e s per internode.
I n order to elucidate the possible role of p h l o e m anastomoses in xylem r e g e n e r a t i o n a r o u n d a w o u n d we selected two types of plant: Cucurbita species which usually
do n o t p r o d u c e interfascicular c a m b i u m , a n d Dahlia species which do. I n addition, to u n c o v e r the possible role of
the age of interfascicular p a r e n c h y m a cells in xylem regeneration, the effect of severing l o n g i t u d i n a l vascular
b u n d l e s was studied in i n t e r n o d e s r a n g i n g in d e v e l o p m e n t
from y o u n g to mature.
Materials and methods
Plants. Plants of Dahlia pinnata Cav., Cucurbita pepo L. and
C. maxima Duchesne (Suttons Seeds, Torquay, Devon, UK) were
grown in the experimental gardens of the School of Plant Science at
the University of Reading. The stems of both Cucurbita species were
about 5 m long and the Dahlia stems about 1 m tall at the time of
the experiments. The stems of both Cucurbita species developed
five internodes (addition of 1 m) per week, while the Dahlia stems
produced two internodes (addition of 0.15 m) weekly. The stems
were selected for uniformity of the length of the leaf blade of their
# 2 leaves (the second unfolded leaves below the apical bud) which
were defined as those whose blades were 50-90 mm long in C.
maxima, 41~60 mm long in C. pepo, and 4(~70 mm in D. pinnata.
The length of internode number 2 (defined as the internode below
leaves #2) ranged between 30 and 60 mm in the three species
studied. The growth of internode # 2 was extremely rapid in both
Cucurbita species, reaching about 150 mm in length within one
week. The length of internode #6, ranging from 130 mm to
170 mm, remained almost constant during the experiments, which
were carried out in July and August 1994.
Development of phloem anastomoses and wounding experiments. To
observe the development of phloem anastomoses in intact stems,
internodes of a range of ages from newly forming (2 mm long) to
mature (up to 170 mm) were collected. They were cleared in lactic
acid and stained with lacmoid according to the method developed
by Aloni and Sachs (1973) and slightly modified by Aloni et al.
(1995). When possible, after clearing with lactic acid, the phloem
tissue was separated from the xylem in the mature internodes.
In order to study the effect of internode age on xylem regeneration around a wound, internodes numbers 2, 3, 5 and 6 were pierced
at their midpoint using a narrow sharp blade (1 mm wide) to sever
a vascular strand. These internodes were sampled 5 and 7 d after
wounding. The wounding experiment was repeated three times on
C. maxima and twice on both C. pepo and D. pinnata, with 10
repetitions per experiment. The wound region in each internode was
prepared for examination as previously described (Aloni and Sachs
1973; Aloni et al. 1995).
Microscopy. Very thick (2-5 mm) pieces of stem were examined and
photographed using a Reichert (Vienna, Anstria) Polyvar 2 photomicroscope and a Wild (Heerbrugg, Switzerland) M400 photomacroscope.
Results
Development o f phloem anastomoses. The specimens prepared for this study were very thick in order to permit
visualisation of the whole of each phloem anastomosis.
The sinuous n a t u r e of the anastomoses, coupled with the
limited depth of field of the light microscope m e a n s that
while the whole of their course t h r o u g h the p a r e n c h y m a
between the l o n g i t u d i n a l b u n d l e s could be followed by
optical sectioning, it was difficult to o b t a i n views of whole
a n a s t o m o s e s in a single micrograph. A n a s t o m o s e s which
a p p e a r to end in the p a r e n c h y m a in some of the illustrations in this paper in fact do so because they have m o v e d
out of the plane of focus. I n practice, all phloem anastomoses illustrated were observed t h r o u g h o u t their
length, a n d all were c o n t i n u o u s at each end with a vascular bundle, or formed a b r a n c h with a n o t h e r anastomosis.
The p a t t e r n of phloem a n a s t o m o s i s d e v e l o p m e n t was
similar in Dahlia a n d Cucurbita, a n d micrographs have
been selected which best illustrate the p h e n o m e n a
described.
I n very y o u n g internodes, shorter than 4 m m long,
there were no phloem a n a s t o m o s e s between the longitudinal vascular bundles in all three species studied. The first
visible signs of the differentiation of p h l o e m anastomoses
were detected in very y o u n g internodes a b o u t 5 m m long
(Fig. 1A,B). Such internodes, which had been cleared,
stained, split in half l o n g i t u d i n a l l y a n d m o u n t e d whole,
already c o n t a i n e d m a t u r e sieve elements in the longitudinal vascular b u n d l e s (Fig. 1A). These elements were easily
detectable by virtue of the presence of sieve plates. The
earliest stage of a n a s t o m o s i s formation was manifested as
the presence of strands of cells clearly derived by recent
cell divisions of interfascicular p a r e n c h y m a lying between
the m a t u r e phloem of adjacent l o n g i t u d i n a l b u n d l e s
(Fig. IA, B). A r e m a r k a b l e feature of these cells was the
Fig. 1A-E. Typical patterns of phloem anastomosis differentiation
in young (A, B) and mature (C-E) internodes of Dahlia pinnata (A-C)
and Cucurbita maxima (D,E). The micrographs show cleared
and lacmoid-stained, thick (2 5 mm) longitudinal sections.
Bars = 100 gm. A Whole mount of half of a developing D. pinnata
internode, 5 mm long at the time of sampling. Mature sieve elements
in the axial vascular strands may be identified by the presence of
sieve plates (white arrows) with low callose levels. Elsewhere, transverse divisions of axial parenchyma cells have occurred to produce
a strand of small cells with prominent nuclei (black arrows). This
strand of cells is the precursor of a phloem anastomosis, x 240.
B Specimen as A. In this view, the cells of the developing phloem
anastomosis (arrows) may be seen to merge into the phloem of an
axial vascular strand (right side), x 240. C Mature internode of D.
pinnata, 70 mm long. The mature phloem anastomosis (black arrows) is clearly visible and identifiable, as are the phloem elements
of the axial strands which it links, by the sieve plates (white arrows)
with relatively high callose levels between individual sieve elements, x 190. D The central region of a mature internode, 150 mm
long of C. maxima. Several anastomoses (arrows) in various orientations are linked to the phloem of a prominent axial strand (S). The
thickness of the specimen means that some parts of the anastomoses
are out of the plane of focus, x 60. E Detail of a wide phloem
anastomosis in a mature internode of C. maxima. The anastomosis
consists of numerous sieve tubes (arrows), only a fraction of which lie
in the plane of focus of this micrograph. Many more could be
observed in the same anastomosis by use of optical sectioning, x 100
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
597
598
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
prominence of their nuclei even though the nuclei had not
been stained specifically (Figs. 1A, B). As the internodes
increased in length with age, these cells elongated and
differentiated to form sieve elements, again identifiable by
the presence of sieve plates (Fig. 1C).
An anastomosis in Cucurbita starts with either one or
a few sieve tubes in a young internode. During internode
elongation m a n y of the anastomoses become thick and
might comprise numerous sieve tubes (Fig. 1D, E). The
exact number of sieve tubes in a bundle of a mature
internode is difficult to determine because so little of it
appears in focus in any optical section (Fig. 1E), but it is
estimated that in the widest anastomoses of Cucurbita
between 20 to 50 sieve tubes are present. In Dahlia
pinnata the widest anastomoses contain 7 sieve tubes. The
number of phloem anastomoses per mature internode
(130 170 m m long) ranged between 5000 and 10000 in the
two Cucurbita species and between 3000 and 5000 per
internode (70-100 m m long) in the Dahlia plants.
Xylem regeneration along Cucurbita stems. Wounding the
young second internode of either C. maxima or C. pepo,
which already possessed phloem anastomoses, resulted in
a curved pattern of xylem regeneration immediately
around the wound (Fig. 2A) by 7 d after wounding. The
arcs of xylem had regenerated within the interfascicular
parenchyma, restoring continuity of the severed longitudinal bundle by the shortest route. The numerous regenerated vessel elements involved in the repair were very
compact. Interestingly, the phloem anastomoses in the
young internodes do not influence xylem regeneration.
In the third internode, the regenerated vessels were
more widely separated after 7 d, consistent with extension
growth of the internode during the xylem regeneration
process (Fig. 2B). There was also a tendency for some
regenerated vessels to form anastomoses with adjacent
intact longitudinal bundles. These xylem anastomoses differentiated in association with pre-existing phloem anastomoses. By the fifth internode, this tendency had become
very pronounced, with direct links formed by xylem anastomoses between the xylem of the severed bundles, and
that of adjacent intact bundles (Fig. 2C). Such xylem anastomoses were formed in preference to arcs of xylem
by-passing the severed region to directly reconnect the
damaged longitudinal bundle.
In the sixth internode, the wounding resulted in regeneration of xylem anastomoses which were associated exclusively with phloem anastomoses. N o arcs of xylem
regeneration were formed around the wound (Fig. 2D) by
7 d after wounding.
In a few mature internodes the regeneration of xylem
extended laterally to form connections with the intact
bundle second-removed from the severed bundle. This
pattern was not observed in young internodes and occurred along phloem anastomoses only in the mature sixth
internodes.
Polarity of xylem regeneration in mature internodes of
Cucurbita. The regeneration of xylem was very polar in
mature internodes studied 7 d after wounding. There was
an increasing tendency with internode age for more xylem
anastomoses to form below the wound than above it
(Fig. 2C, D). This was true for both C. maxima (Fig. 2D)
and C. pepo (Fig. 3A). The anastomoses were always close
to the wound when they formed above it, but were more
dispersed, and present for a much greater distance below
the wound (Figs. 2D, 3A). When xylem regeneration was
studied 5 d after wounding, the formation of vessels often
appeared complete above the wound, but incomplete
below it. This polar pattern was evidenced by faint outlines of developing, but as yet not fully differentiated
regenerated vessels below the wound, while well-developed regenerated vessels were present above it (Fig. 3D).
Giant regenerated vessels. Extremely wide regenerated
vessels were frequently formed in the older Cucurbita
internodes, both above and below the wound (arrows in
Figs. 2C, D and 3A-D). Whereas the majority of the
regenerated vessel elements were narrow and appeared to
have developed from interfascicular parenchyma without
significant cell enlargement, the wide vessel members of
the giant vessels, which differentiated within the phloem
anastomoses, had undergone considerable expansion in
diameter during differentiation. The diameter of the giant
regenerated vessels in C. maxima ranged from 250 to
400 gm, while the diameter of normal regenerated vessels
was less than 70 ~am.
Discontinuous patterns of xylem regeneration. Samples
taken 5 d after wounding revealed early stages of xylem
regeneration with various patterns of incomplete xylem
anastomoses (Fig. 3C, D). It was clear from optical sectioning that the vessels seen in these internodes were
incomplete, and that they did not simply move out of the
plane of section. From Fig. 3C it is clear that the vessels
develop piecemeal, and not sequentially from one end or
other of the anastomosis. Thus, while some regenerated
vessel elements form strands connected at one end to
the xylem of a longitudinal vascular bundle, others form
Fig. 2A-D. Micrographs of typical regenerated vessels (arrowheads)
and giant regenerated vessels(arrows)around a wound (W) in young
(A, B) and mature (C,D) internodes of C. maxima sampled seven
days after wounding, showing the gradual transition from the typical
curved by-pass of xylem regeneration immediately around the
wound (A) to xylem regeneration limited to phloem anastomoses
further away from the wound (D). Bars = 1 ram. A In the second
internode, numerous parenchyma cells have redifferentiated to produce vessels by-passing the wound and rejoining the xylem of the
damaged bundle. The repair is very compact in internodes of this
age. x 10. B In the third internode the regenerated vessels are more
extended in form around the wound, x 15. C In the fifth internode
a few vessel elements have formed a bridge around the wound
(arrowhead on its left side), but most vessel formation has followed
the course of the phloem anastomoses leading to the wound being
by-passed indirectly via adjacent, undamaged vascular bundles.
Many more regenerated xylem anastomoses have formed below,
than above the wound. Note the presence of some giant regenerated
vessels (arrows).x 10. D In the sixth internode, there are no direct
bridges of xylem formed around the wound. The differentiation
of the regenerated xylem elements is exclusively associated
with phloem anastomoses. Note that most of the xylem anastomoses consist of giant regenerated vessels (arrows)although some
of the smaller regenerated vessels (arrowheads) are present (lower
left), x 10
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
599
600
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
strands within the interfascicular parenchyma without
being attached at either end to the xylem of a longitudinal
vascular bundle (Fig. 3C).
Xylem regeneration along Dahlia stems. The major difference in vascular development between mature internodes
of Dahlia pinnata and those of the two Cucurbita species is
that the former possesses an interfascicular cambium
which actively produces the secondary xylem. A wound
in mature internodes (both fifth and sixth internodes) of
Dahlia induced regenerated vessels which originated from
the cambial initials in the familiar oval pattern immediately around the wound. Phloem anastomoses in the mature internodes of Dahlia did not influence the pattern of
xylem regeneration.
In young internodes of Dahlia the regenerated vessels,
which originated from interfascicular parenchyma cells,
differentiated in the typical curved pattern around the
wound in a similar way to that found in Cucurbita
(Fig. 2A). The numerous phloem anastomoses which occurred in the young Dahlia internodes remained inactive
in xylem regeneration.
Discussion
Vascular differentiation, which continues as long as the
plant grows, may be broadly classified into three types:
(i) primary differentiation, which occurs by direct changes
in the cells derived from the primary meristem - the
procambium; (ii) secondary differentiation, which occurs in
cells formed from the lateral meristem - the cambium; and
(iii) regenerative differentiation, which occurs by a re-differentiation of parenchyma cells or changes in the cambium (Sachs 1981). Regenerative differentiation occurs
Fig. 3A-D. Micrographs of regenerated vessels (arrows and arrowheads) around a wound (W) in mature internodes of Cucurbita pepo
(A) and C. maxima (B-D) sampled either 7 d (A, B) or 5 d (C, D) after
wounding, showing polar patterns of xylem regeneration (A,D),
discontinuous patterns (C) and giant regenerated vessels (A-D).
Bars = 1 mm. A The sixth internode in C. pepo showing a typical
pattern of xylem regeneration within phloem anastomoses similar to
that of C. maxima (see Fig. 2D). Note that most of the xylem
anastomoses differentiated mainly below the wound, x 10. B Detail
of the fifth internode in C. maxima, showing a giant regenerated
vessel (arrows) compared with regular regenerated vessels (arrowheads). While the majority of regenerated vessel elements may be
seen to have differentiated directly from phloem-anastomosisassociated parenchyma cells whose dimensions, orientation and
position have been retained, the elements of the giant vessel may be
seen to have undergone considerable enlargement during differentiation, x 20. C. Part of the fifth internode in C maxima 5 d after
wounding showing something of the time course of regenerative
xylem development. While many of the vessels develop outwards by
differentiation of parenchyma cells in contact with the first-occurring axial bundles, it is apparent that some isolated regenerated
vessel elements appear in the parenchyma out of contact with the
axial bundles (asterisks).x 20. D Part of the sixth internode
in C. maxima 5 d after wounding. Here the vessels above the wound
are mature, while they are still developing below it (asterisks),
demonstrating the role of polarity of auxin flow in their formation. x 10
601
most often following wounding, or associated with internal developmental changes such as the growth of lateral
roots (Sachs 1981; Aloni and Plotkin 1985). The present
study has revealed another type of internal regenerative
differentiation - the development of phloem anastomoses.
These are naturally occurring regenerative sieve tubes
which redifferentiate from interfascicular parenchyma
cells between the longitudinal vascular bundles.
Phloem anastomoses do not exist in very young internodes (less than 3 m m long) of Dahlia and Cucurbita, but
during internode development, thousands of phloem
anastomoses redifferentiate naturally in each growing internode. This normal type of regenerative phloem differentiation which characterizes intact shoots is relatively
difficult to visualise in conventional light-microscope sections because of the sinuous nature of the phloem anastomoses. The lacmoid clearing technique (Aloni and Sachs
1973) used in this study permits the observation of the
development of phloem anastomoses between the longitudinal vascular bundles in very thick stem pieces (2-5 mm).
The technical difficulty of observing phloem anastomoses
in thin sections means that they have been ignored or have
received very little attention in plant a n a t o m y texts
(Esau 1965, 1969; Mauseth 1988; Fahn 1990; Raven et al.
1992; Moore et al. 1995).
Our study emphasizes that the normal course of vascular differentiation in a dicotyledon stem is as follows: (i)
primary differentiation of longitudinal strands of phloem
and xylem occurs in young growing internodes; (ii) regenerative phloem differentiation (phloem anastomoses development between the vascular bundles) occurs during
internode elongation, and at later developmental stages,
usually when elongation has ceased; then (iii) secondary
differentiation of both phloem and xylem occurs.
This study has shown that phloem anastomoses start
to differentiate after the first mature primary phloem
elements are present in the axial strands. The first evidence
of their development being in the form of patterns of cell
divisions within the interfascicular parenchyma of very
young internodes. The differentiation of phloem anastomoses is probably induced by low-level streams of auxin
which leak from the longitudinal vascular bundles during
internode elongation (Aloni 1987a, 1995). According to
the 'canalization hypothesis' (Sachs 1969, 1981, 1986)the
flow of auxin induces the formation of a vascular network.
Before a vascular system has been established, auxin will
necessarily follow an irregular pathway along which vascular differentiation will take place. This in turn will
provide a preferred pathway for auxin flow, reinforcing
the differentiation along that pathway. The sinuous patterns of phloem anastomoses reflect the r a n d o m pathways
of auxin movement between the longitudinal bundles. The
gradual development of very wide phloem anastomoses
with numerous sieve tubes in the two Cucurbita species
indicates that during internode elongation and the development of phloem anastomoses they serve as the preferred
pathways for low-level streams of auxin between the
longitudinal bundles.
The regeneration of xylem within the phloem anastomoses in mature internodes of the two Cucurbita species
supports our working hypothesis that the high-level
streams of polar auxin flow descending from above the
602
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
wound move preferably through the phloem anastomoses.
Where the concentration of these auxin streams was high
enough, xylem regeneration was induced in some of the
phloem anastomoses. Conversely, phloem anastomoses
were not involved in xylem regeneration around a wound
in mature internodes of Dahlia. This result demonstrates
that when a cambium is present it becomes the preferred
pathway for auxin and therefore the anastomoses in Dahlia remain inactive. However, although the phloem anastomoses in Dahlia do not function in xylem regeneration
they do function in assimilate translocation following
wounding (Aloni and Peterson 1990).
Although many phloem anastomoses were already
present before wounding in the young internodes of the
three species studied, the regenerated vessels redifferentiated from interfascicular parenchyma cells and the
anastomoses remained inactive. This result indicates that
when parenchyma cells are young and can redifferentiate
into vascular elements they will serve as the preferred
pathway for auxin. However, when the parenchyma cells
become old, they lose the ability to respond, and if no
cambium is present, the anastomoses become the main
pathway for the polar auxin descending from above the
wound, resulting in xylem regeneration within the anastomoses.
The role of phloem anastomoses in xylem regeneration
was not detected earlier (Sachs 1981; Aloni 1987b, 1995;
Roberts et al. 1988) because regeneration of xylem was
studied either in species possessing active cambium in
their mature internodes, or in young internodes in which
the young parenchyma cells were capable of redifferentiating into vascular elements.
Vessel regeneration in both young and mature internodes was clearly polar in several respects. In the young
internodes the vessel elements above the wound regenerated closer to the wound than those below it, confirming
earlier reports (see Baum et al. 1991). This was also true
for xylem regeneration in the phloem anastomoses, where
the xylem anastomoses above the wound were always
close to the cut, but were more dispersed and present for
a much greater distance below it. Furthermore, the differentiation of regenerated vessel elements above the wound
was faster than below it, so that mature regenerated vessel
elements were first observed above the wound. These
polar patterns of xylem regeneration around the wound
have been interpreted as being a consequence of interruption to the basipetal polar flow of auxin by the wound,
resulting in a high local auxin level immediately above the
cut (Aloni 1987a, 1995).
This study is the first to describe giant regenerated
vessels around a wound. In old internodes, especially near
growing fruit, where anastomoses are the only pathway
for xylem regeneration, and no other direct connections
around the wound are formed, the giant vessels so frequently observed may be particularly important. They
must compensate for the loss of the normal pathway as
a result of wounding and must be capable of efficient
conduction of large quantities of water to the large leaves
and the fast growing fruits. These extremely wide regenerated vessels, which were frequently formed in the mature
internodes of both Cucurbita species, are potentially very
eff•
conductors of water since the rate at which water
can be conducted through a vessel is proportional to the
fourth power of its radius (Poiseuille's law) (Zimmermann
1983).
The present study demonstrates the effect of internode
age, or distance from the young leaves, on the diameter of
regenerated vessels. According to the 'six-point hypothesis' (Aloni and Zimmermann 1983), high auxin levels
occurring near the young leaves induce numerous narrow
vessels, while low concentrations occurring away from the
young leaves result in a few wide vessels. In the young
internodes of Cucurbita all the numerous regenerated
vessels were narrow, while in the old internodes some
giant regenerated vessels differentiated. The extremely
wide vessel elements of the giant vessels in the phloem
anastomoses had undergone considerable diameter expansion growth during differentiation. It has been shown
that slow vessel differentiation, which permits vessel expansion, is induced by relatively low auxin concentrations
(Aloni and Zimmermann 1983). Therefore, the differentiation of the giant regenerated vessels could only occur
under relatively low auxin levels typically found in old
internodes located further away from the young leaves
which are the main sources of auxin.
Of particular interest is the observation that the xylem
cells associated with phloem anastomoses do not necessarily differentiate sequentially and that isolated groups of
cells at an advanced stage of differentiation may be present within the parenchyma, apparently as yet unconnected
to the main vascular strands. This contrasts with the
statement of Jacobs (1952), based on the work of Kaan
Albert (1934), that new xylem cells are differentiated in
a strictly basipetal direction, by the transformation of
already-existing parenchyma cells, the new cells always
being in continuity with the previously differentiated ones
(our italics). While this situation applies in the case of cells
differentiating from parenchyma, it seems that cells differentiating within or among the phloem anastomoses can
do so in a more-random order, as local conditions permit.
These discontinuous patterns of xylem differentiation indicate different levels of responsiveness of the differentiating cells along the pathway of auxin.
Aloni and Peterson (1990) demonstrated that the
phloem anastomoses in Dahlia stems serve as an emergency system for assimilate translocation. They are capable
of functioning but usually do not do so under normal
conditions. Only when a stem is wounded do the anastomoses begin to function, thus providing alternative
pathways for assimilates around a wound. The present
study further shows that in mature internodes without
an interfascicular cambium the phloem anastomoses also
serve as an emergency system which provides numerous
lateral routes for xylem regeneration around the stem. The
ability of Cucurbita plants to carry out this type of repair
is significant from an evolutionary point of view, since
they thus have retained the potential for xylem regeneration in the absence of a cambium between the vascular
bundles. The repair mechanism uses preformed pathways,
already established between mature vascular strands and
ensures rapid restoration of water and nutrient flow
around a damaged region. We therefore propose that the
dense network of thousands of phloem anastomoses per
internode has developed during evolution as a mechanism
R. Aloni and J.R. Barnett: Role of phloem anastomoses in xylem regeneration
of a d a p t a t i o n to possible d a m a g e in m a t u r e i n t e r n o d e s by
p r o v i d i n g flexible alternative p a t h w a y s for efficient xylem
r e g e n e r a t i o n in plants with limited or n o interfascicular
cambium.
This research was supported by an International Scientific Exchange
Award to R.A. from the Israel Academy of Sciences and The Royal
Society.
References
Aloni R (1980) Role of auxin and sucrose in the differentiation of
sieve and tracheary elements in plant tissue cultures. Planta 150:
255-263
Aloni R (1987a) The induction of vascular tissues by auxin. In:
Davis PJ (ed) Plant hormones and their role in plant growth
and development. Martinus Nijhoff, Dordrecht Boston, pp
363-374
Aloni R (1987b) Differentiation of vascular tissues. Annu Rev Plant
Physiol 38:179-204
Aloni R (1995) The induction of vascular tissues by auxin and
cytokinin. In: Davis PJ (ed) Plant hormones: physiology, biochemistry and molecular biology. 2nd edn. Kluwer Academic
Publishers, Dordrecht, London, pp 531-546
Aloni R, Peterson CA (1990) The functional significance of phloem
anastomoses in stems of Dahlia pinnata Cav. Planta 182:
583-590
Aloni R, Plotkin T (1985) Wound-induced and naturally occurring
regenerative differentiation of xylem in Zea mays L. Planta 163:
126 132
Aloni R, Sachs T (1973) The three-dimensionalstructure of primary
phloem systems. Planta 113:345-353
Aloni R, Zimmermann MH (1983) The control of vessel size and
density along the plant axis a new hypothesis. Differentiation
24:203 208
603
Aloni R, Pradel KS, Ullrich CI (1995) The three-dimensional structure of vascular tissues in Agrobacterium tumefaciens-induced
crown galls and in the host stems of Ricinus communis L. Planta
196:597 605
Baum SF, Aloni R, Peterson CA (1991) The role of cytokinin in
vessel regeneration in wounded Coleus internodes. Ann Bot 67:
543-548
Esau K (1965) Vascular differentiation in plants. Holt Rinehart
Winston, New York
Esau K (1969) The phloem. In: Zimmermann W, Ozenda P, Wulff
HD (eds) Encyclopedia of plant anatomy, vol 5, pt 2. Gebruder
Borntraeger, Berlin
Fahn A (1990) Plant anatomy. 4th edn. Pergamon Press, Oxford
New York
Jacobs WP (1952) The role of auxin in differentiation of xylem
around a wound. Am J Bot 39:301-308
Kaan Albert A yon (1934) Anatomische und physiologische Untersuchungen fiber die Entstehung yon Siebr6hrenverbindungen.
Z Bot 27:1-92
Mauseth JD (1988) Plant anatomy. Benjamin/Cummings, Menlo
Park California
Moore R, Clark WD, Stern KR, Vodopich D (1995) Botany. Brown
Publishers, Dubuque Iowa
Raven PH, Evert RF, Eichhorn SE (1992) Biology of plants. 5th edn.
Worth Publishers, New York
Roberts LW, Gahan PB, Aloni R (1988) Vascular differentiation and
plant growth regulators. Springer-Verlag, Berlin Heidelberg
New York
Sachs T (1969) Polarity and the induction of organised vascular
tissues. Ann Bot 33:263-275
Sachs T (1981) The control of the patterned differentiation of vascular tissues. Adv Bot Res 9:151-262
Sachs T (1986) Cellular patterns determined by polar transport. In:
Bopp M (ed) Plant growth substances. Springer-Verlag, Berlin
Heidelberg, pp 231-235
Zimmermann MH (1983) Xylem structure and the ascent of sap.
Springer-Verlag, Berlin Heidelberg New York