Content uploaded by Ann M. Hirsch
Author content
All content in this area was uploaded by Ann M. Hirsch
Content may be subject to copyright.
Content uploaded by Ann M. Hirsch
Author content
All content in this area was uploaded by Ann M. Hirsch
Content may be subject to copyright.
Content uploaded by Ann M. Hirsch
Author content
All content in this area was uploaded by Ann M. Hirsch
Content may be subject to copyright.
The Plant Cell,
Vol.
2,
1009-1017, October 1990
O
1990 American Society
of
Plant Physiologists
Nodulin Gene Expression and ENOD2 Localization in
Effective, Nitrogen-Fixing and Ineffective, Bacteria-Free
Nodules of Alfalfa
Clemens Van de Wiei," Joanna H. Norris,b Birgit Bochenek," Rebecca Dickstein,d*' Ton Bisseling," and
Ann
M.
Hirsch".'
a
Department of Molecular Biology, Agricultura1 University, Wageningen, The Netherlands
Department of Botany, University of Rhode Island, Kingston, Rhode lsland 02881
Department of Biology, University of California,
Los
Angeles, California 90024
Department
of
Genetics, Harvard Medical School and Department of Molecular Biology, Massachusetts General
Hospital, Boston, Massachusetts 021 14
Alfalfa plants form bacteria-free nodules in response to a number of agents, including
Rhizobium
meliloti exo
mutants, Agrobacterium tumefaciens transconjugants carrying cloned
R.
meliloti nodulation genes, and compounds
that function as auxin transport inhibitors,
N-(
l-naphthy1)phthalamic acid or 2,3,5-triiodobenzoic acid. These bacteria-
free nodules contain transcripts for the nodulins Nms30 and MsENOD2; transcripts for late nodulins like leghemo-
globin are not detected. In situ hybridization studies demonstrated that ENOD2 transcripts were localized in
parenchyma cells at the base and along the periphery of nitrogen-fixing alfalfa root nodules. The ENOD2 gene was
also expressed in
a
tissue-specific manner in nodules elicited by
N-(
l-naphthy1)phthalamic acid and 2,3,5-triiodo-
benzoic acid. In bacteria-free nodules induced by
R.
meliloti exo mutants and
A.
tumefaciens transconjugants carrying
either one or both
R.
meliloti symbiotic plasmids, ENODS transcripts were also detected but were usually localized
to parenchyma cells at the base instead of along the periphery of the nodule. On the basis of the pattern of ENODS
gene expression, we conclude that the developmental pathway
of
bacteria-free nodules, whether bacterially or
chemically induced, is the same as that of nitrogen-fixing nodules, and, furthermore, that the auxin transport
inhibitors in their action mimic some factor(s) that trigger nodule development.
INTRODUCTION
Alfalfa root nodule development represents an excellent
model system for investigating the question of how differ-
entiated root cortical cells escape their developmental
destiny and initiate a new structure, the nodule. Alfalfa
root nodule development begins after specific recognition
of the host plant and Rhizobium meliloti. The inner cortical
cells of the root divide anticlinally and initiate a nodule
primordium, which is soon invaded by an infection thread
that originated in an infected root hair (Dudley et al., 1987).
Rhizobia are released from branches of the infection thread
into cells of the central part of the nodule primor-
dium. Shortly thereafter, the persistent nodule meristem,
consisting of small, densely cytoplasmic, actively dividing
cells, is organized at the apical (distal) end of the nodule
primordium.
The meristem gives rise to cells that differentiate into
the central tissue consisting of infected and uninfected
'
Current address: Laboratoire
de
Biologie Moléculaire
des
Rela-
tions
Plantes-Microorganismes,
CNRS-INRA, Chemin
de
Borde-
rouge,
BP
27,
31 326 Castanet-Tolosan Cedex, France.
To whom correspondence should
be
addressed.
cells, the nodule cortex (formerly called "outer cortex"),
and the nodule parenchyma (formerly called "inner cortex")
(Van de Wiel et al., 1990). The nodule cortex is bounded
at its inner side by the nodule endodermis, whereas the
nodule parenchyma is located between the central tissue
of
infected and uninfected cells and the nodule cortex.
Nodule parenchyma tissue consists of highly vacuolated
parenchyma cells that are more densely packed, with
fewer and smaller intercellular spaces, than the nodule
cortex. Nodule parenchyma also develops at the base of
the nodule. Thus, the central tissue of the nodule is com-
pletely surrounded by, nodule parenchyma, except at the
distal end where the apical meristem is situated. This
arrangement of cells has been related to nodule parenchy-
ma's
property of forming a barrier to the penetration of
free oxygen toward the central tissue (cf Witty et al., 1986).
As is the case for many legumes, alfalfa nodule devel-
opment can be arrested at severa1 different stages. How-
ever, alfalfa roots are more susceptible than many other
legumes to agents that elicit the formation of ineffective,
bacteria-free nodules.
R.
meliloti
mutants defective in ex-
MW
69-
46-
30-
14-
OD
($>
»^-
.
Ill
69-
46-
30-
14-
MW
69-
46-
30-
Q)
(5)
Q)
10
PH
3 10
pH
14-
Q)
(P
O
I
10
PH
ENOD2 Localization in Alfalfa Nodules 1 O1 1
opolysaccharide synthesis (exo) (Finan et al., 1985; Leigh
et al., 1987) and Agrobacterium tumefaciens transconju-
gants carrying
R.
meliloti nodulation genes (Wong et al.,
1983; Truchet et al., 1984; Hirsch et al., 1984,1985) induce
nodules that superficially resemble normal, nitrogen-fixing
nodules. They have a central tissue surrounded by nodule
parenchyma and peripheral vascular bundles, but cells of
the central tissue are devoid of bacteria. Also, the nodules
are not elongate like effective, nitrogen-fixing nodules but
are small and broad, often with the margins of one nodule
overlapping the borders of adjacent ones, like beads on a
string. Meristematic activity is spread along the dista1 end
of the nodule.
In alfalfa, two nodulins, Nms30 and MsENOD2, are
associated with nodule morphogenesis. The Nms30 no-
dulin is known as an in vitro translation product only.
Neither its amino acid sequence nor its localization in the
nodule has been studied
so
far. The ENOD2 nodulin was
originally identified as a cDNA clone from a soybean nodule
library. Its deduced amino acid sequence shows a repeti-
tive structure in which two different pentapeptides alter-
nate, each pentapeptide starting with
2
proline residues.
Because many proline-rich proteins are hydroxylated and
localized in the cell wall, it is likely that ENOD2 represents
a (hydroxy)proline-rich cell wall protein (Franssen et al.,
1987). Recently, in situ hybridization studies have shown
that the ENOD2 gene is exclusively expressed in the
nodule parenchyma
of
soybean as well as of pea nodules
(Van de Wiel et al., 1990).
Alfalfa plants can also form bacteria-free root nodule-
like structures in response to the auxin transport inhibitors
(ATls) such as N-(1 -naphthyl)phthalamic acid (NPA) and
2,3,5-triiodobenzoic acid (TIBA) (Hirsch et al., 1989). In
severa1 respects, the histology of the ATI-induced nodules
resembles that of the bacteria-free (“empty”) nodules elic-
ited on alfalfa by
R.
meliloti exo mutants or A. tumefaciens
transconjugants carrying cloned
R.
meliloti nod genes.
However, their histology differs from that of bacteria-
induced empty nodules in that their vascular tissue is
confined to the proximal end of the nodule (Hirsch et al.,
1989).
Like the exo mutant-induced nodules (Dickstein et al.,
1988; Norris et al., 1988), the ATI-elicited nodules contain
transcripts for the nodulins Nms30 and MsENOD2. Tran-
scripts for late nodulins, such as leghemoglobin, are not
found in the ATI-induced nodules (Hirsch et al., 1989). The
presence of the early nodulin MsENOD2 and the nodulin
Nms3O has been used as a diagnostic criterion to deter-
mine whether the ATI-induced structures were comparable
with
R.
meliloti-induced root nodules. Because these no-
dulin genes are expressed in ATI-induced structures as
well as in empty nodules formed in response
to
R.
meliloti
exo mutants or
A.
tumefaciens transconjugants carrying
R.
meliloti nodulation sequences, we concluded that they
represent nodules fully comparable with those induced by
R.
meliloti (Hirsch et al., 1989).
In this report, we will demonstrate that the ENOD2 gene
is expressed in empty nodules, both bacteria-induced and
ATI-induced, in a tissue-specific manner that is the same
as that
of
wild-type
R.
meliloti-induced nodules even
though the growth pattern of the ATI-induced structures
deviates from the ”normal” growth pattern of alfalfa root
nodules. However, even among the bacteria-induced
empty nodules, there is variation in growth pattern (cf
Wong et al., 1983; Hirsch et al., 1984, 1985; Finan et al.,
1985). Therefore, we have included empty nodules elicited
by different bacterial strains in this study. Such a compar-
ison is significant because it demonstrates that structures
that
may
exhibit some morphological divergence follow
the same developmental pathway and are, therefore,
equivalem.
RESULTS
Nodule Development and Nodulin Gene Expression
The morphology of the nodules elicited by
R.
meliloti exoA
and exoF mutants conforms to that
of
other
R.
meliloti exo
mutant-induced nodules, as previously described by Finan
et al. (1985) and Leigh et al. (1987). Nodules induced by
Figure
1.
Two-Dimensional Polyacrylamide
Gel
Analysis
of
in Vitro Translations of Total RNA lsolated from Alfalfa Tissue.
(A)
From root tissue.
The
open arrow
(in
all panels) points to glutamine synthetase (see Norris
et
al., 1988, for identification of this
translation product as
GS).
The circles indicate prominent translation products that are evident after in vitro translations of
RNA
isolated
from infected root nodules but that are absent in root
RNA.
(B)
From nitrogen-fixing nodules induced
by
wild-type
R.
meliloti
strain RmlO21.
The
boxed-in translation product
is
a nodule-specific/
enhanced glutamine synthetase. The arrowhead points to Nms30, and the arrows indicate leghemoglobin.
(C)
From Fix-, bacteria-free nodules induced
by
Rm7055 (exoF::Tn5). Nms30
is
present (arrowhead)
as
well
as Nms25 (tandem arrows),
but
leghemoglobin translation products (circles) are absent.
(D)
From Fix-, bacteria-free nodules induced
by
A128
(A.
tumefaciens carrying Rm1021 pSyma). Nms30 is present (arrowhead), but
leghemoglobin translation products (circles) are not.
(E)
From Fix-, bacteria-free alfalfa nodules induced
by
Rm7061 (exoA::Tn5).
The
open arrow points to glutamine synthetase.
The
circles
indicate leghemoglobin translation products observed after in vitro translation of RNA isolated from infected nodules but absent in RNA
isolated from
these
nodules.
The
Nms30 translation
product
is
present (arrowhead), as
is
Nms25 (tandem arrows).
1012
The
Plant Cell
Figure 2. Detection of ENOD2 Transcript in Alfalfa Nodules by in Situ Hybridization.
ENOD2 Localization
in
Alfalfa Nodules
1 O1
3
A. tumefaciens strain A128, which carries the
R.
meliloti
megaplasmid with the nod and niflfix genes (pSyma), or
strain A135, which contains both pSyma as well as pSymb
(the megaplasmid with the exoABDF genes) (Finan et
al.,
1986) are also devoid of intracellular bacteria like the
A.
tumefaciens transconjugant-induced empty nodules re-
ported earlier by Hirsch et al. (1984, 1985) and Truchet et
al. (1 984). These nodules will be described in more detail
in the next section.
Previous analyses of in vitro translated RNA on two-
dimensional gels have shown that about 20 different no-
dulin translation products can be identified in wild-type
R.
meliloti-induced alfalfa nodules, as seen in Figures 1 A and
1B. In nodules formed in response to
R.
meliloti exo
mutants or A. tumefaciens transconjugants carrying
R.
meliloti nodulation genes, only the nodulin translation prod-
uct Nms30 can be detected (Dickstein et al., 1988; Norris
et al., 1988). In addition, a nodulin of variable isoforms,
Nms25, that appears to be unique to exo mutant-induced
nodules, has been described by Leigh et al. (1987). Figures
1C and 1E illustrate that the empty nodules used in the
present study conformed to this pattern of gene expres-
sion. Nms30 as well as Nms25 translation products were
detectable in nodules elicited by exoF and exoA mutants.
On the other hand, nodules formed in response to A128
inoculation appeared to contain only the Nms30 transcript
(Figure 1 D).
The early nodulin ENOD2 mRNA
has
been shown by
RNA transfer blot analysis to be present in exo mutant-
induced nodules as well as in
A.
tumefaciens transconju-
gant-elicited nodules (Dickstein et al., 1988).
A
similarly
sized 1.4-kb ENOD2 transcript was detected in nodules
induced by the A. tumefaciens transconjugants A1 28 and
A1 35 used in the present study (data not shown).
Thus, the pattern of nodulin gene expression
is
remark-
ably similar in all empty nodules, except that the nodulin
Nms25 transcript apparently occurs only in
R.
meliloti exo
mutant-induced nodules.
Localization of
ENODS
Transcripts in
Nodules
Sections of the various alfalfa root nodules were hybridized
to antisense RNA transcribed from pA2ENOD2, which
contains an approximately 0.3-kb insert of alfalfa ENOD2
cDNA. The 4-week-old, wild-type
R.
meliloti-induced
nod-
ule showed a hybridization pattern like the one described
for pea (cf Van de Wiel et al., 1990). The ENOD2 mRNA
was present in parenchyma cells (nodule parenchyma)
surrounding the central tissue of infected and uninfected
cells and also at the base
of
the nodule, as shown in
Figures 2A and 28. There was no ENOD2 transcript
detectable at the top of the nodule, the site of the apical
meristem (Figure 2A). Three-dimensionally, ENOD2 local-
ization was in the shape of a cup, which was open at the
nodule meristem. No ENOD2 transcripts were found within
the vascular strands along the periphery of the nodule
(Figure 2A).
Two-week-old to 3-week-old NPA-induced and TIBA-
induced nodules were more or less similar
to
each other
(data not shown). Meristematic activity was frequently
spread over a relatively large part of the dista1 end of the
nodule instead of being concentrated to the apex, as in
wild-type
R.
meliloti-induced nodules. Vascular traces were
confined to the proximal part of the nodule and did not
separate into distinct strands extending distally into the
peripheral tissue of the nodule.
A
peripheral tissue and
central tissue, the latter consisting of cells rich in plastids
with prominent starch grains, can be distinguished (Figure
2C). Around this central tissue, a relatively narrow zone
consisting of highly vacuolated, densely packed cells con-
tained transcripts that hybridized to the ENOD2 antisense
_____~
Figure
2.
(continued).
(A)
Dark-field photograph of a nitrogen-fixing alfalfa nodule. ENOD2 transcripts as visualized by the brightly reflecting silver grains (see
Methods) are localized to parenchyma cells along the periphery of the nodule surrounding the vascular bundle (vb) and also along the
base of the nodule. The meristem
(m)
is devoid of MsENOD2 transcript. Median longitudinal section. Magnification
~85.
Bar
=
100
pm.
(6)
Bright-field photograph of a nitrogen-fixing nodule cut obliquely. The nodule meristem is out of the plane of section. ENOD2 transcripts
were localized on paraffin sections using the Boehringer Mannheim RNA labeling kit (Genius) (see Methods).
A
blue color is present in
parenchyma cells along the edge of the infected cells and at the base of the nodule but not in the nodule cortex (nc). The color that
appears to be in the root (r) is actually at the base of a second lobe of a two-lobed nodule. The second lobe is not observed in this
section. Magnification
x85.
Bar
=
100
pm.
(C)
Dark-field photograph of a cluster of nodules elicited
2
weeks after addition of
NPA.
ENOD2 transcripts (brightly reflecting silver
grains) are detected along the periphery of the nodule, but not in the root (r). Slightly oblique section. Magnification
x40.
Bar
=
100
pm.
(D)
Bright-field photograph of
(C).
Magnification
~40.
(E)
Dark-field photograph of a 5-week-old, bacteria-free nodule elicited by
Rm5078
(exoB::Tn5). The nodule meristem
(m)
is devoid of
ENOD2 mRNA, as is the nodule cortex (nc). ENOD2 transcripts extend along the periphery of the nodule and are also localized to
parenchyma cells at the base of the nodule. The vascular bundle (vb; observed in transverse section) is also devoid of ENOD2 transcript.
Nonmedian longitudinal section. Magnification
~85.
Bar
=
1
O0
pm.
(F)
Bright-field photograph of a bacteria-free nodule induced by Rm7061 (exoA::Tn5). The nodule has been sectioned transversely (see
Figure 3A for approximate location of the section). Severa1 vascular bundles (vb) are evident; these are surrounded by parenchyma cells
that express MsENOD2. The nodule cortex (nc) is devoid of ENOD2 mRNA. Magnification
x85.
Bar
=
100 pm.
1014
The
Plant Cell
Figure
3.
Detection
of
ENOD2 Transcript
in
Alfalfa Nodules
by in
Situ Hybridization.
(A)
Bright-field photograph
of a
3-week-old, bacteria-free nodule induced
by
Rm5078 (exoB::Tn5).
The
presumed meristematic region
(m)
is
very broad
and
lacks ENOD2 mRNA. ENOD2 transcripts
are
detected
at the
base
of the
nodule
in
parenchyma tissue that surrounds
ENOD2 Localization in Alfalfa Nodules 1015
RNA (Figures 2C and 2D). Outside this zone, a less densely
packed nodule cortex, where no ENOD2 transcript could
be detected, was present.
Figure 2E illustrates an example of a 5-week-old exoB
mutant-induced nodule. As in wild-type
R.
melilofi-induced
nodules, it is possible to distinguish an apical meristem, a
nodule cortex, a peripheral zone, and a central tissue,
which, however, was free of intracellular bacteria. ENOD2
transcript was found in the densely packed, highly vacuo-
lated parenchymatous tissue that constitutes the inner
part of the peripheral tissue. This parenchymatous tissue
was traversed by a vascular strand that is free of ENOD2
transcript (data not shown). The outer part of the peripheral
tissue, which is more loosely packed, was also free of
ENOD2 transcript.
The majority of exoB, exoA, or exoF mutant-induced
nodules are not elongate like effective, nitrogen-fixing al-
falfa nodules. They are small and broad, often with the
margins of one nodule overlapping the borders of adjacent
ones, thus giving the appearance of beads on a string.
Meristematic activity, which is spread along a relatively
long region of the subtending root, leads to the deposition
of a limited amount of tissue in a plane perpendicular to
the root axis. Consequently, little peripheral and central
tissue is formed at the apical end of the nodule, as shown
in Figures 3A and 38. ENOD2 transcripts were localized
to nodule parenchyma cells at the base of the nodule. Also
at the base of the nodule, several vascular traces con-
nected to the root stele were present that did not contain
ENOD2 transcript (Figures 2F, 3A, and 3B). A more distally
located tissue consisting of cytoplasmically rich cells con-
taining amyloplasts was also free of ENOD2 transcript
(Figures 3A and 38).
The nodules elicited by the
A.
fumefaciens transconju-
gants A1 28 and A1 35 were quite similar to the small, bead-
like, exo mutant-induced nodules described above. There
was a limited amount of meristematic activity over a broad
zone of the subtending root (Figures 3C to 3F). Proximally,
several vascular traces were connected to the root stele
(Figures 3E and 3F). ENOD2 transcripts were clearly de-
tectable in a zone of densely packed, vacuolated paren-
chymatous tissue, containing amyloplasts, at the base
of these nodules and around the vascular traces. The dis-
tal tissue of these nodules, which is highly vacuolated
and free of amyloplasts, contained no ENOD2 transcript
(Figure 3D).
DISCUSSION
Our in situ hybridization studies clearly showed that the
ENOD2 gene was expressed in the nodule parenchyma of
nitrogen-fixing alfalfa nodules, following a pattern in agree-
ment with the recent observations of pea nodules (Van de
Wiel et al., 1990). Moreover, we found that the ENOD2
gene was expressed in a tissue-specific manner in nodules
elicited by the ATls, NPA, and TIBA: ENOD2 transcripts
were present in a zone of densely packed, highly vacuo-
lated parenchyma cells between a central tissue of cells,
rich in amyloplasts, and an outer layer of more loosely
packed cortical tissue. No ENODP mRNA was detected in
the dista1 end of the nodule. The position and morphology
of
the ENOD2 gene-expressing zone corresponded to
those of the nodule parenchyma of wild-type
R.
melilofi-
induced nodules. Hence, by using the criterion of tissue-
specific ENOD2 gene expression, one may conclude that
the empty ATI-elicited nodules are indeed comparable with
nodules induced by wild-type
R.
meliloti.
The
R.
meliloti exo mutants and
A.
tumefaciens tran-
sconjugants induce a heterogeneous population of nod-
ules (cf Wong et al., 1983; Hirsch et al., 1984, 1985; Finan
et al., 1985). Relatively few are of the elongate type that
most resemble wild-type
R.
meliloti-induced nodules. In
such nodules, the presence of ENOD2 transcripts was
restricted to a tissue that, both positionally and morpho-
logically, was comparable with the nodule parenchyma of
wild-type
R.
meliloti-induced nodules (cf Figure 2E and
Figures 2A and 2B). The vast majority of nodules, however,
were small and broad, in which little peripheral and central
tissue was differentiated perpendicularly to the long axis
of the root. Nevertheless, ENOD2 gene expression was
restricted to a zone of densely packed parenchymatous
tissue extending along the base of the nodule and around
the vascular traces (Figures 3A and
38).
The location
of
this zone was the same as the ENOD2 gene-expressing
Figure
3.
(continued).
the vascular bundles (vb). The nodule cortex (nc) forms the outer boundary of the nodule. The arrows denote the approximate location of
the transverse section illustrated in Figure 2F. Near median longitudinal section. Magnification ~82.5. Bar
=
100
fim.
(6)
Dark-field photograph of
(A).
Magnification ~82.5.
(C)
Dark-field photograph
of
a bacteria-free nodule induced by A135. ENOD2 transcripts are localized to the parenchyma cells at the base
of the nodule. Oblique section. Magnification x82.5.
(D)
Bright-field photograph of
(C).
ENOD2 mRNA is localized to the base of the nodule. A nodule cortex (nc) is observed, but cells typical
of a meristematic region are not identifiable in this nodule. Magnification x82.5. Bar
=
100 pm.
(E)
Bright-field photograph
of
a bacteria-free nodule formed in response to A1 28 infection. The nodule is sectioned transversely to
obliquely. ENOD? mRNA is detected at the base of the nodule surrounding the vascular bundles (vb). Isodiametric, dividing cells typical
of
a meristematic region are not observed in this nodule, but the loosely packed nodule cortex (nc) is present. Magnification x82.5. Bar
=
100Gm.
(F)
Dark-field
photograph of
(D).
1
O1
6 The Plant Cell
nodule parenchyma region in wild-type
R.
meliloti-induced
nodules. Thus, in spite of the morphological anomalies
caused by the diffuse and limited meristematic activity in
these empty nodules, the pattern of ENOD2 expression
did not differ fundamentally from that observed in wild-
type
R.
meliloti-induced nodules.
Localization studies of mRNAs that mark specific nodule
tissues may provide insight into the nature of aberrant root
nodules. Although distinct tissues, like nodule paren-
chyma, should be recognizable by the combination of
positional and morphological criteria, such distinctions in
tissue types are relatively difficult to assess by microscop-
ical studies alone, especially in small and broad empty
nodules of the type described above. In the cases where
a clearly differentiated nodule endodermis is detectable (cf
Truchet et al., 1984; Finan et al., 1985), it is relatively easy
to delimit nodule parenchyma cells from the more loosely
packed (“outer”) cortex. However, delimitation from a bac-
teria-free central tissue is frequently more difficult because
possible differences in the extent of intercellular space
between the two tissues are not easy to discern. Further-
more, other morphological criteria are difficult to evaluate;
the nodule parenchyma cells are often rich in amyloplasts,
like the central tissue itself. Also, the extent of vacuolation
in the central tissue is variable in nodules of the small and
broad type. In these cases, localization studies of ENOD2
transcripts are helpful. Such analysis may also be impor-
tant for understanding even more peculiar types of nod-
ules, such as those elicited on clover by
A.
tumefaciens or
R.
leguminosarum bv tfifolii transconjugants carrying
R.
meliloti nodulation sequences (Hirsch et al., 1985; Truchet
et al., 1985). or the ones elicited on alfalfa by
R.
meliloti
with a Tn5 insertion
3
kb downstream from nodC (Dudley
et al., 1987). Unlike other empty nodules, these pseudo-
nodules more closely resemble lateral roots in that they
have a centrally located vascular bundle. In contrast to
normal lateral roots, however, one such type of nodule
has a nodule endodermis-like layer in its cortex (Dudley et
al., 1987). Localization of ENOD2 transcripts may indicate
whether a nodule parenchyma-like tissue is present in
these pseudonodules and,
if
so,
where it is located. It may
further indicate whether these structures are develop-
mentally equivalent to authentic root nodules.
Thus, based on the observation that the ATls induce a
normal pattern of ENOD2 gene expression, we conclude
that in their action the ATls mimic some factor(s) respon-
sible for nodule development. The ATls are known to cause
alterations in the endogenous hormone balance of the
plant. Cell divisions and ENOD2 gene expression, both of
which are induced by ATI treatment, thus may be related
to the hormonal status of the root. Recently, Lerouge et
al. (1 990) have identified a Rhizobium-produced molecule,
NodRm-1, which elicits alfalfa root hair deformation, as a
P-sulfated tetraglucosamine with acetyl and acyl substitu-
tions. The ATls, which are often substituted benzoic acids,
are structurally different from NodRm-1, an oligosacchar-
ide. Thus, it is unlikely that both act at the same site. This
implies that
Rhizobium
may produce other molecules, sep-
arate from NodRm-1, that trigger the cascade of events
leading to nodule development. Alternatively, NodRm-1
may stimulate the production of an endogenous plant
factor that interacts with the same site as the ATls.
METHODS
Piant Material
Alfalfa (Medicago
safiva
L cv Iroquois) plants were grown
as
described by Norris et al. (1988). At the time of sowing, individual
bins containing alfalfa seeds were left uninoculated or inoculated
with either Rhizobium melilofi wild-type strains Rm2011 or
Rm1021,
R.
meliloti mutant in exoB
(Rm5078),
exoA (Rm7061),
or exoF (Rm7055), or Agrobacferium tumefaciens transconju-
gants, either strain A128, which harbors the
nodlnif
symbiotic
megaplasmid (pSyma), or strain A135, which carries the
nodlnif
rnegaplasmid (pSyma) and the megaplasmid bearing exo genes
(pSymb) (Leigh et al., 1985; Finan et al., 1986). Plants were
treated with the ATls as described previously (Hirsch et al., 1989).
RNA Analysis
Tissue was harvested, frozen
in
liquid nitrogen, and stored at
-8OOC until use. Total RNA was isolated from nodule or root
tissue, in vitro translations were performed, and the translation
products were separated by two-dimensional electrophoresis as
described by Norris et al. (1988). The RNA for transfer blots was
transferred to GeneScreen (Du Pont-New England Nuclear) ac-
cording to the manufacturer’s directions. A 292-bp insert of
pA2ENOD2 (Dickstein et al., 1988) was prepared and nick trans-
lated. Hybridization and washing conditions followed the Gene-
Screen manufacturer’s protocol, washing at 65OC.
In Situ Hybridization
Tissues were fixed either as described by Van de Wiel et al.
(1990) or in
4%
glutaraldehyde, 1.5% paraformaldehyde in 0.1
M
phosphate buffer, pH 7.2. After fixation, the tissues were rinsed
twice in buffer and dehydrated in a graded alcohol series until
they were in
50%
alcohol. The tissues were then transferred
to
the first tertiary butyl alcohol step, dehydrated in the tertiary butyl
alcohol series (Sass, 1958), and embedded in Paraplast. Sections
either 7
pm
or
8
pm thick were affixed to poly-L-lysine-coated
slides. The in situ hybridizations using 35S-UTP antisense RNA
probes were based on a protocol published by Cox and Goldberg
(1988) and were performed essentially as described by Van de
Wiel et al. (1990). After developrnent
of
the emulsion, sections
were stained either with 1% aqueous safranin or with
0.05%
toluidine blue.
Alternatively, the RNA labeling kit (Genius) using digoxigenin-
labeled UTP (Boehringer Mannheim) was utilized. Sense and
antisense probes were made according to the manufacturer’s
directions and added to deparaffinized sections that had been
ENOD2 Localization in Alfalfa Nodules 1 O1 7
pretreated as described for radioactive probes. Some nodules
were fixed overnight in formaldehyde-acetic acid-alcohol (Sass,
1953), rinsed overnight in
50%
alcohol, slowly rehydrated, and
then frozen on a block in liquid nitrogen. The nodules were
sectioned at 16 pm using a cryostat, and the sections were placed
on poly-L-lysine-coated slides. The nodule sections were gradually
warmed to room temperature and further treated as for radioactive
probes. Antisense probe was added to the cryostat-sectioned
material at 10 ng/mL to 100 ng/mL of hybridization buffer per
slide. The manufacturer’s protocol for pre-hybridization, hybridi-
zation, and post-hybridization was followed for both sense and
antisense probes; sections were hybridized at 37OC overnight.
MsENOD2 transcripts were detected immunologically according
to the manufacturer’s directions. For paraffin-embedded material,
12 hr to 72 hr were required for maximal color development,
whereas for cryostat-sectioned material, color development
generally occurred within 60 min. The sections were not
counterstained.
After dehydration through an alcohol series, the sections were
mounted with DPX or Eukitt, and were photographed either with
a Nikon or Zeiss Axiophot microscope equipped with bright-field
and dark-field optics.
ACKNOWLEDGMENTS
We acknowledge the generosity of Agway, Inc., Syracuse, NY,
for providing us with alfalfa seeds, and we also thank Ethan
R.
Signer, John Leigh, and Turlough Finan for bacterial strains. We
are especially grateful to Judith Lengyel, who introduced us to
the “Genius Kit,” and to Stefan Kirchanski for critically reading the
manuscript. We also thank the members of our respective labo-
ratories for helpful comments and useful discussions. A.M.H. was
supported at the beginning of this project by a fellowship from the
Organization for Economic Cooperation and Development on
Food Production and Preservation (summer 1988). We also ac-
knowledge support from National Science Foundation Grant DCB-
8703297.
Received June 20, 1990; accepted August 13, 1990
REFERENCES
Cox, K.H., and Goldberg, R.B.
(1988). Analysis of plant gene
expression. In Plant Molecular Biology:
A
Practical Approach,
C.H. Shaw, ed (Oxford:
IRL
Press), pp. 1-34.
Dickstein,
R.,
Bisseling, T., Reinhold, V.N., and Ausubel, F.M.
(1988). Expression of nodule-specific genes in alfalfa root nod-
ules blocked at an early stage
of
development. Genes Dev. 2,
Dudley, M.E., Jacobs, T.W., and Long,
S.R.
(1987). Microscopic
studies of cell divisions induced in alfalfa roots by Rhizobium
melilofi. Planta 171, 289-301.
Finan, T.M., Hirsch, A.M., Leigh, J.A., Johansen, E., Kuldau,
G.A., Deegan,
S.,
Walker, G.C., and Signer, E.R.
(1985).
Symbiotic mutants of Rhizobium meliloti that uncouple plant
from bacterial differentiation. Cell 40, 869-877.
Finan, T.M., Kunkel, B., De Vos, G.F., and Signer, E.R.
(1986).
Second symbiotic megaplasmid in Rhizobium meliloti carrying
exopolysaccharide and thiamine synthesis genes. J. Bacteriol.
677-687.
167,66-72.
Franssen, H.J., Nap, J.P., Gloudemans, T., Stikema, W., Van
Dam, H., Govers, F., Louwerse, J., Van Kammen, A., and
Bisseling, T.
(1 987). Characterization of cDNA for nodulin-
75 of soybean:
A
gene product involved in early stages of
root nodule development. Proc. Natl. Acad. Sci. USA 84,
4495-4499.
Hirsch, A.M., Wilson, K.J., Jones, J.D.G., Bang,
M.,
Walker,
V.V., and Ausubel, F.M.
(1 984). Rhizobium melilofi nodulation
genes allow Agrobacferium tumefaciens and Escherichia coli to
form pseudonodules on alfalfa. J. Bacteriol. 158, 11 33-1 143.
Hirsch, A.M., Drake, D., Jacobs, T.W., and Long,
S.R.
(1985).
Nodules are induced on alfalfa roots by Agrobacferium fume-
faciens and Rhizobium
frifolii
containing small segments of
the Rhizobium melilofi nodulation region.
J.
Bacteriol. 161,
Hirsch, A.M., Bhuvaneswari, T.V., Torrey, J.G., and Bisseling,
T.
(1989). Early nodulin genes are induced in alfalfa root out-
growths elicited by auxin transport inhibitors. Proc. Natl. Acad.
Sci. USA 86,1244-1248.
Leigh, J.A., Signer,
E.R.,
and Walker, G.C.
(1985). Exopolysac-
charide-deficient mutants of Rhizobium melilofi that form inef-
fective nodules. Proc. Natl. Acad. Sci. USA 82, 6231 -6235.
Leigh, J.A., Reed, J.W., Hanks, J.F., Hirsch, A.M., and Walker,
G.C.
(1 987). Rhizobium melilofi mutants that fail to succinylate
their calcofluor-binding exopolysaccharide are defective in nod-
ule invasion. Cell 51, 579-587.
Lerouge, P., Roche, P., Faucher, C., Maillet,
F.,
Truchet, G.,
Promé, J.C., and Denarié, J.
(1 990). Symbiotic host-specificity
of
Rhizobium
meliloti is determined by a sulphated and acylated
glucosamine oligosaccharide signal. Nature 344, 781 -784.
Norris, J.H., Macol, L.A., and Hirsch, A.M.
(1988). Nodulin gene
expression in effective alfalfa nodules and in nodules arrested
at three different stages of development. Plant Physiol. 88,
Sass, J.E.
(1953). Botanical Microtechnique, 3rd ed (Ames,
IA:
lowa State University Press), pp. 26-28.
Truchet, G., Rosenberg, C., Vasse, J., Julliot, J.-S., Camut,
S.,
and Dénarié, J.
(1984). Transfer of Rhizobium melilofi Sym
genes into Agrobacferium tumefaciens: Host-specific nodulation
by atypical infection. J. Bacteriol. 157, 134-142.
Truchet, G., Debelle,
F.,
Vasse, J., Terzaghi, B., Garnerone,
A.-M., Rosenberg, C., Batut, J., Maillet, F., and Denarié, J.
(1985). ldentification of a Rhizobium melilofi
pSym2011
region
controlling the host specificity of root hair curling and nodulation.
J. Bacteriol. 164, 1200-121
O.
Van de Wiel, C., Scheres, B., Franssen, H., van Lierop, M.-J.,
Van Lammeren, A., Van Kammen, A., and Bisseling, T.
(1990). The early nodulin transcript ENOD2 is located in the
nodule parenchyma (inner cortex) of pea and soybean root
nodules. EMBO
J.
9,
1-7.
Witty, J.F., Minchin, F.R., Skst, L., and Sheehy, J.E.
(1986).
Nitrogen fixation and oxygen in legume root nodules. Oxford
Surv. Plant
MOI.
Cell Biol. ,3, 275-314.
Wong, C.H., Pankhurst, C.E., Kondorosi, A., and Broughton,
W.J.
(1983). Morphology of root nodules and nodule-like
structures formed by
Rhizobium
and Agrobacterium strains
containing a Rhizobium melilofi megaplasmid. J. Cell Biol.
97,
223-230.
321 -328.
787-794.
DOI 10.1105/tpc.2.10.1009
1990;2;1009-1017Plant Cell
C. Van de Wiel, J. H. Norris, B. Bochenek, R. Dickstein, T. Bisseling and A. M. Hirsch
Bacteria-Free Nodules of Alfalfa
Nodulin Gene Expression and ENOD2 Localization in Effective, Nitrogen-Fixing and Ineffective,
This information is current as of October 14, 2011
Permissions 8X
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw153229
eTOCs http://www.plantcell.org/cgi/alerts/ctmain
Sign up for eTOCs at:
CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain
Sign up for CiteTrack Alerts at:
Subscription Information http://www.aspb.org/publications/subscriptions.cfm is available at:Plant Physiology and The Plant CellSubscription Information for
ADVANCING THE SCIENCE OF PLANT BIOLOGY
© American Society of Plant Biologists
