Content uploaded by Naima el ghachtouli
Author content
All content in this area was uploaded by Naima el ghachtouli on Mar 23, 2023
Content may be subject to copyright.
Vol. 14, No. 6, 2001 / 737
MPMI Vol. 14, No. 6, 2001, pp. 737–748. Publication no. M-2001-0316-01R. © 2001 The American Phytopathological Society
Medicago truncatula ENOD11: A Novel
RPRP-Encoding Early Nodulin Gene Expressed
During Mycorrhization in Arbuscule-Containing Cells
Etienne-Pascal Journet,1 Naima El-Gachtouli,2 Vanessa Vernoud,1 Françoise de Billy,1
Magalie Pichon,1 Annie Dedieu,1 Christine Arnould,2 Dominique Morandi,2 David G. Barker,1
and Vivienne Gianinazzi-Pearson2
1Laboratoire de Biologie Moléculaire des Relations Plantes-Microorganismes, CNRS-INRA, BP 27, 31326
Castanet-Tolosan Cedex, France; 2UMR INRA–Université de Bourgogne Biochimie, Biologie Cellulaire et
Ecologie des Interactions Plantes-Microorganismes, CMSE-INRA, Dijon, France
Submitted 31 October 2000; Accepted 8 February 2001.
Leguminous plants establish endosymbiotic associations
with both rhizobia (nitrogen fixation) and arbuscular my-
corrhizal fungi (phosphate uptake). These associations
involve controlled entry of the soil microsymbiont into the
root and the coordinated differentiation of the respective
partners to generate the appropriate exchange interfaces.
As part of a study to evaluate analogies at the molecular
level between these two plant–microbe interactions, we
focused on genes from Medicago truncatula encoding puta-
tive cell wall repetitive proline-rich proteins (RPRPs) ex-
pressed during the early stages of root nodulation. Here
we report that a novel RPRP-encoding gene, MtENOD11,
is transcribed during preinfection and infection stages of
nodulation in root and nodule tissues. By means of reverse
transcription-polymerase chain reaction and a promoter-
reporter gene strategy, we demonstrate that this gene is
also expressed during root colonization by endomycorrhi-
zal fungi in inner cortical cells containing recently formed
arbuscules. In contrast, no activation of MtENOD11 is
observed during root colonization by a nonsymbiotic,
biotrophic Rhizoctonia fungal species. Analysis of trans-
genic Medicago spp. plants expressing pMtENOD11-gusA
also revealed that this gene is transcribed in a variety of
nonsymbiotic specialized cell types in the root, shoot, and
developing seed, either sharing high secretion/metabolite
exchange activity or subject to regulated modifications in
cell shape. The potential role of early nodulins with atypi-
cal RPRP structures such as ENOD11 and ENOD12 in
symbiotic and nonsymbiotic cellular contexts is discussed.
Additional keywords: Glomus spp., in situ hybridization, Si-
norhizobium meliloti, transfer cells.
Terrestrial plants have developed the ability to establish mu-
tualistic and reciprocally beneficial symbiotic relationships
with soil microorganisms. In such associations, the micro-
symbionts improve the mineral nutrition of the plant in ex-
change for photosynthates. The most frequent type of root
endosymbiosis, the phosphate-acquiring arbuscular mycorrhi-
zal (AM) association, displays little host specificity and in-
volves more than 80% of extant plant taxa, which interact
with a limited number of fungal species belonging to the Glo-
males order (Smith and Read 1997). When the fungal symbi-
ont contacts the host root surface, it differentiates to form an
appressorium and, subsequently, penetrates the root. Penetrat-
ing hyphae then grow throughout the cortex and differentiate
within inner cortical cells into highly branched structures
known as “arbuscules.” The arbuscule–cortical cell interface
is thought to be the site where phosphate and possibly carbon
are transferred between symbionts (Harrison 1999; Smith and
Smith 1990). Although the process of colonization has been
described clearly, the molecular mechanisms governing rec-
ognition between symbiotic partners and the development of
the AM symbiosis are still largely unknown.
In contrast to the AM association, which dates back to the
Devonian era, the nitrogen-fixing symbiosis with Rhizo-
biaceae bacteria is more recent on the evolutionary time-scale
and restricted almost completely to leguminous plants with a
generally narrow host specificity. This well-studied endo-
phytic association leads to the formation of a novel type of
plant organ known as the “root nodule,” in which bacteria
reduce atmospheric nitrogen to ammonia and thus provide the
host plant with a privileged source of combined nitrogen. In
temperate legumes such as pea and Medicago spp., nodule
development is initiated by the specific attachment of rhizobia
to young emerging root hairs located close to the root tip,
leading to hair curling. Specialized tubular structures known
as “infection threads” are then induced in curled root hairs,
allowing the invading bacteria to progress toward the inner
root cortex, where the initiation of cell division leads to the
formation of a nodule primordium. Subsequent growth and
coordinated differentiation of both symbiotic partners result in
the development of the N2-fixing, indeterminate-type nodule
possessing a persistent apical meristem (Hadri et al 1998;
Mylona et al 1995).
Molecular mechanisms underlying the Rhizobium–legume
symbiotic interaction have been studied in much more detail
than for the AM counterpart. The interaction is initiated by a
molecular dialog between the two organisms and, in particu-
Corresponding author: E.-P. Journet; E-mail: journet@toulouse.inra.fr
Current address of N. El-Gachtouli: Département de Biologie, Faculté
des Sciences et Techniques de Fès Saiss, Fès, Morocco.
738 / Molecular Plant-Microbe Interactions
lar, the synthesis by rhizobia of specific lipo-
chitooligosaccharide (LCO)-signaling molecules known as
“Nod factors,” which are of fundamental importance to suc-
cessful host–bacterial recognition and root nodulation
(Albrecht et al. 1999; Schultze and Kondorosi 1998). The
whole nodulation process involves the specific expression or
up-regulation of a number of plant genes known as “nodulin
genes” (Pawlowski 1997). These genes were initially consid-
ered to be expressed specifically during symbiotic associa-
tions with rhizobia. However, many are now thought to par-
ticipate in other plant developmental processes because the
majority also are expressed in nonsymbiotic plant tissues
(Schultze and Kondorosi 1998). A relatively small number of
early nodulin genes, postulated to play a role in morphoge-
netic and organogenetic plant responses, have been character-
ized. One of the first early nodulin genes to be studied in de-
tail was PsENOD12 (Scheres et al. 1990) encoding a putative
cell wall repetitive proline-rich protein (RPRP). We were able
to isolate a closely related gene, MtENOD12, from the model
legume Medicago truncatula (Pichon et al. 1992), and made
use of a promoter–reporter-gene approach to characterize Nod
factor-dependent MtENOD12 gene activation in the root epi-
dermis (Journet et al. 1994; Pingret et al. 1998).
Despite clear differences between the two root endosymbio-
ses in terms of host specificity and developmental responses
elicited in the host plant, genetic and molecular studies have
shown that several common features characterize various
stages in the establishment of these symbiotic interactions.
The most convincing evidence for this comes from the fact
that a significant proportion of legume mutants that have lost
the ability to form infection threads and nodules (Nod¯) are
also completely resistant to AM fungi (Myc¯), whereas their
interaction with soil pathogens is not affected (Albrecht et al.
1999; Gianinazzi-Pearson 1996; Harrison 1997). Furthermore,
a small number of legume nodulin genes have been identified
that are transcriptionally activated during both AM and rhizo-
bial interactions, e.g., the early nodulin genes MsENOD2,
MsENOD40 (Van Rhijn et al. 1997), PsENOD5, PsENOD12
(Albrecht et al. 1998), and an atypical leghemoglobin gene,
VfLb29 (Frühling et al. 1997). In addition, transcripts coding
for certain polypeptides that are immunologically cross-
reactive with nodule-specific plant proteins (nodulins) have
been detected in mycorrhizal roots, and monoclonal antibodies
raised against components of the infection thread or the peri-
bacteroid membrane in nodules recognize antigenic sites in
the periarbuscular interface (Gianinazzi-Pearson 1996). These
observations, in conjunction with the much earlier evolution
of the AM symbiosis, led to the proposition that some of the
plant processes leading to nodulation may have evolved from
those preexisting for fungal symbioses (Gianinazzi-Pearson
1997; LaRue and Weeden 1994).
The identification of additional nodulin genes expressed
during the establishment of the arbuscular mycorrhizal sym-
biosis should facilitate comparative studies between both root
endosymbioses and, at the same time, provide new molecular
markers to better characterize the early stages of AM forma-
tion. In this article, we report the characterization of
MtENOD11, an early nodulin gene from the model legume M.
truncatula encoding a novel RPRP. We show that MtENOD11
is a molecular marker for both early preinfection responses
and later infection-related processes occurring within root and
symbiotic nodule tissues. Furthermore, we demonstrate that
this gene is expressed strongly during the arbuscular my-
corrhizal interaction in inner cortical cells containing recently
formed arbuscules. We discuss the significance of the similari-
ties and differences between the expression patterns of the two
early nodulin genes MtENOD11 and MtENOD12 in symbiotic
and nonsymbiotic contexts and, in particular, the possible
functions of the corresponding putative cell wall RPR-
proteins.
RESULTS
Identification and cloning
of the single-copy MtENOD11 gene.
In a hybridization experiment performed a number of years
ago, several bands of variable intensity were observed when a
pea ENOD12 cDNA (Scheres et al. 1990) was used to probe
restriction digests of M. truncatula genomic DNA under low-
stringency conditions (Pichon et al. 1992). The cloning and
characterization of the single-copy gene which we named
MtENOD12, corresponding to the strongest hybridization
signal and clearly homologous at the sequence level with
PsENOD12, has been reported (Pichon et al. 1992). Subse-
quently, we isolated a genomic clone corresponding to one of
the weaker cross-hybridizing bands (the 3.0 kb EcoRI ge-
nomic DNA fragment in Figure 1 in Pichon et al. 1992). This
clone contained a gene encoding a second early nodulin,
which we named MtENOD11. Southern analysis revealed that
the MtENOD11 gene also is single-copy in the M. truncatula
genome (data not shown).
Sequence analysis of the MtENOD11 gene
and the predicted encoded RPRP.
The nucleic acid sequence of the MtENOD11 gene, includ-
ing immediate flanking regions, has been submitted to the
EMBL databank as accession no. AJ297721. The 174 amino
acid-long sequence deduced from the coding strand of this
gene is presented in Figure 1. No alternative open reading
frame of significant length could be identified on either strand
of the genomic sequence hybridizing to the pea ENOD12
probe, and there is no evidence from sequence analysis that
MtENOD11 contains introns. In line with this, reverse tran-
scription-polymerase chain reaction (RT-PCR) amplification
of MtENOD11 cDNA generates a fragment of identical size to
that predicted from the genomic sequence (see below).
In MtENOD11, the ATG initiation codon is followed by 24
amino acids that, according to the rules of Von Heijne (1983),
are likely to serve as a hydrophobic transit signal peptide. This
putative N-terminal signal sequence is very similar (60 to 75%
identity) to that found in pea, vetch, alfalfa, and M. truncatula
ENOD12 putative precursor proteins (Allison et al. 1993;
Pichon et al. 1992; Scheres et al. 1990; Vijn et al. 1995) as
well as that of the MsENOD10 nodulin precursor (Löbler and
Hirsch 1993). The remaining coding sequence is composed
mainly of the repeating pentapeptide element PPXXX, a struc-
tural feature characteristic of a family of (hydroxy)proline-rich
cell wall proteins known as RPRPs, which were originally
characterized in soybean (Hong et al. 1990). The mature
MtENOD11 sequence (as is the case for MtENOD12) con-
tains relatively few (5) tyrosines compared with classical
RPRPs (Kieliszewski and Lamport 1994) due to the fact that
Vol. 14, No. 6, 2001 / 739
this amino acid virtually is absent from the repeating pen-
tapeptide motif. Although MtENOD11 and MtENOD12 pos-
sess similar overall structures, MtENOD11 is approximately
twice as long as MtENOD12. Furthermore, the two PPXXX
repeat regions cannot be aligned, and the sequences flanking
the two repeat regions share no homology. Moreover, the re-
peated pentapeptide motifs are extremely variable in
MtENOD11 and cannot be grouped into decapeptide repeats
as is the case for MtENOD12 (Pichon et al. 1992).
MtENOD11 transcripts localize to the prefixation zone
of the nitrogen-fixing nodule.
Northern hybridization experiments with a MtENOD11 spe-
cific probe had initially shown that this gene is expressed
throughout most of nodule development and that the
MtENOD11 transcript level is at its highest prior to the onset
of N2 fixation (data not shown). This also is the case for
MtENOD12 (Pichon et al 1992), although MtENOD11 tran-
script levels are approximately fivefold higher. The tissue-
specific location of MtENOD11 transcripts within the nodule
was examined using in situ hybridization. Sections of M. trun-
catula nodules, harvested either 4 days or 2 to 3 weeks
postinoculation, were hybridized with a 35S-labeled antisense
RNA probe. Figure 2A shows that MtENOD11 mRNA is pre-
sent throughout most of the central tissue of the young emerg-
ing nodule at 4 days post-inoculation (dpi). Control experi-
ments with sense probes gave low-level, uniform backgrounds
(data not shown). At this early stage of development, the nod-
ule is approximately spherical in shape with a central differen-
tiating tissue in which ramifying intercellular infection threads
penetrate the plant tissue located below the recently formed
nodule meristem. Equivalent experiments carried out on sec-
tions of 2- to 3-week-old mature nitrogen-fixing nodules
showed that MtENOD11 transcripts are now restricted to a
narrow band at the distal end of prefixation zone II, the region
of the nodule where bacteria are released from the infection
threads (Fig. 2B) (definition of nodule zones in Vasse et al.
1990). A closer examination of this region (Fig. 2C) revealed
that transcripts are predominantly present in the submeris-
tematic preinfection zone, i.e., in the two or three cell layers
in which infection threads are not yet visible (Pichon et al
1992). The hybridization signal drops to background levels
toward the middle of zone II, where bacteroids and plant cells
are rapidly codifferentiating. Therefore, MtENOD11 expres-
sion is initiated in an identical region of the nodule, as de-
scribed for MtENOD12 (Pichon et al 1992). MtENOD12 tran-
scripts, however, are still present in more proximal tissues of
the prefixation zone, suggesting either a wider transcription
window for MtENOD12 or differences in the transcript stabil-
ity of the two genes.
MtENOD11 also is expressed during early stages
of the interaction with Sinorhizobium meliloti.
To complement in situ hybridization studies and facilitate
analysis of MtENOD11 gene expression during both the ear-
lier stages of nodulation and the association with endosymbi-
otic fungi (see below), we made use of a transgenic M. trun-
catula line expressing a fusion between the MtENOD11
promoter (2.3 kb) and gusA reporter gene. The construction of
the transcriptional fusion and the subsequent generation of
this representative line (L416) containing a single-copy
pMtENOD11–gusA homozygous transgene, are described
elsewhere (D. G. Barker, J. L. Pingret, M. Chabaud, and E.-P.
Journet, unpublished).
Following Sinorhizobium meliloti inoculation, activation of
the MtENOD11 promoter was first detected after 3 to 6 h in
epidermal tissues lying behind the root tip and corresponding
to the region of root hair emergence and development (Fig.
2D). This epidermal response occurs both in the primary and
lateral roots, precisely in the region where successful infec-
tions will occur 1 to 2 days later (Bhuvaneswari et al. 1981).
In line with these results, accumulation of MtENOD11 tran-
scripts as early as 3 h postinoculation had been previously
detected in polyA+ mRNA extracts (Gamas et al. 1996). Note
that GUS staining also is present at the apex of L416 roots
(Fig. 2D). This corresponds to nonsymbiotic MtENOD11 gene
expression in root cap cells and will be discussed in more
detail later. At 3 dpi, developing nodules first become visible
as root outgrowths (Fig. 2E) and GUS staining is now present
within the central tissue, which agrees with the in situ hy-
bridization results presented earlier for 4-day-old nodules. In
addition, intense GUS staining is clearly visible on the root
surface at infection sites (Fig. 2E), either above developing
nodules or where infections in root hairs have arrested (Pen-
metsa and Cook 1997; Vernoud et al 1999). At later stages of
nodule development, GUS activity becomes restricted to the
distal end of the central tissue (Fig. 2F), again in line with the
in situ hybridization data (Fig. 2B and C). Thus, our results
show that, as for MtENOD12 (Pichon et al. 1992),
MtENOD11 gene expression can be correlated with preinfec-
tion and infection events throughout nodulation.
Expression of MtENOD11
during the arbuscular endomycorrhizal interaction.
Because the AM symbiotic interaction shares common fea-
tures at the molecular and genetic levels with its rhizobial
counterpart (see above), we examined whether the early
nodulin gene MtENOD11 also is transcribed during AM
formation. As transgenic plants carrying promoter–GUS gene
Fig. 1. Predicted amino-
acid sequence of the MtENOD11 polypeptide. The
putative signal peptide is in italics, and the peptide cleavage site is marked
by an arrow. The 24 proline-rich pentapeptide motifs are double under-
lined, and the five tyrosine residues in the mature polypeptide are boxed.
740 / Molecular Plant-Microbe Interactions
Fig. 2. Analysis of MtENOD11 expression during early and late stages of nodulation. A and B, Dark-field images of MtENOD11 mRNA localization in
7-µm nodule longitudinal sections by in situ hybridization with 35S-labeled antisense riboprobes (radioactive labeling appears as bright spots). A, Young
nodule harvested 4 days postinoculation showing a uniform pattern of hybridization throughout most of the central tissue. White star indicates the meris-
tematic zone; white arrow indicates the root vasculature. Bar = 80 µm. B, Mature nodule harvested 3 weeks after inoculation. The hybridization signal is
localized at the distal end of prefixation zone II. Bar = 150 µm. C, Bright-field image of the framed region in B, showing that MtENOD11 transcripts
(radioactive labeling now appears as dark spots) are present predominantly in the 2–3 cell layers of the submeristematic preinfection zone where infec-
tion threads are not yet visible. White star indicates the nodule meristematic zone, and arrows point to distally located infection threads. Bar = 50 µm.
D–F, Histochemical localization of GUS activity in the transgenic line L416 expressing pMtENOD11–gusA. D,
Young lateral root 18 h postinoculation,
showing epidermal MtENOD11
promoter activation in the region of developing root hairs. Note the constitutive GUS staining in root cap cells at the
root apex (Fig. 4E). Bar = 900 µm. E, Lateral root 3 days postinoculation. GUS staining is visible in the outer tissues of an emerging nodule correspond-
ing to the initial infection site and in the deeper tissues corresponding to the invaded nodule primordium. Several sites of arrested infections also are
visible on the lower root surface (arrows). Bar = 250 µm. F, Longitudinal section of a 3-week-old nitrogen-fixing nodule. GUS activity is restricted to
the distal end of the central tissue. Bar = 200 µm.
Vol. 14, No. 6, 2001 / 741
fusions are particularly well suited for analyzing gene expres-
sion at the cellular level, we initially made use of the M. trun-
catula L416 transgenic line described above to evaluate
pMtENOD11-gusA expression in root systems colonized by
Glomus species. L416 plants were therefore grown in pots and
inoculated with Glomus fasciculatum (see below), and roots
were harvested at a stage corresponding to approximately 20
to 30% root colonization. Histological GUS staining revealed
strong reporter gene expression in longitudinal stretches lo-
cated in inner root tissues (Fig. 3A). A closer examination of
root segments making use of both transverse root sectioning
(Fig. 3B and C) and counterstaining for fungal structures (Fig.
3D) allowed us to determine that the intense GUS staining
correlated with the presence of arbuscule-containing cells
located in the innermost cortical cell layers. Adjacent cortical
cells also showed some GUS activity, although at a signifi-
cantly lower level (Fig. 3B and C). In contrast, GUS activity
in the root inner cortex was totally absent both in control
uninoculated transgenic plants and in nontransgenic mycorrhi-
zal roots (data not shown). This latter control allowed us to
conclude that the staining in mycorrhizal L416 transgenic
roots cannot be attributed to fungal GUS-like enzyme activity,
which means that the blue coloration occasionally found
within internal hyphae in the vicinity of arbuscule-containing
cells in L416 (Fig. 2C) is probably an artifact resulting from
the local diffusion of the soluble monomeric GUS-reaction
product (Gallagher 1992). MtENOD11 expression in arbus-
cule-containing cells was systematically observed whether
plants were grown in soil or in the axenic in vitro system de-
scribed later and with all Glomus species tested (see below).
Finally, a transgenic alfalfa (Medicago varia) line harboring
the pMtENOD11–gusA construct gave very similar results in
terms of GUS localization following fungal colonization (data
not shown).
Because Northern analysis with total RNA (P. Gamas,
personal communication) and in situ hybridization proved
insufficiently sensitive to detect MtENOD11 mRNA accu-
mulation in mycorrhizal roots, we used RT-PCR to evaluate
endogenous MtENOD11 gene activation during symbiotic
fungal colonization. Figure 4 shows that MtENOD11 tran-
scripts could be detected clearly by RT-PCR in RNA ex-
tracted from mycorrhizal roots of M. truncatula plants
grown in smaller pots to speed up colonization (see below)
and harvested 2 weeks after inoculation with Glomus
mosseae (30% root colonization). MtENOD11 mRNA, how-
ever, was undetectable in control uninoculated roots (Fig. 4).
Furthermore, no DNA was amplified when the reverse tran-
scriptase step was omitted, thus showing that the PCR signal
is not the result of amplification of contaminating genomic
DNA (data not shown). The quality and the relative quanti-
ties of the cDNAs used were confirmed by performing
parallel amplifications with primers specific for the
constitutively expressed MtPR10-1 gene (see below) (Ga-
mas et al. 1998). Comparable results were obtained in three
independent experiments with mRNA extracted from my-
corrhizal tissues either 2 or 3 weeks after inoculation.
The time course of pMtENOD11-gusA expression over a
lengthy mycorrhization period was then determined with
plants grown in larger pots (see below). Transgenic M. varia
seedlings were transplanted to a clay substrate containing a G.
fasciculatum inoculum and whole root systems were harvested
at regular intervals 1 to 10 weeks later. The degree of my-
corrhization and the functional activity of internal fungal
structures were assessed by quantifying arbuscule abundance
in representative root samples (Trouvelot et al. 1986) that
were stained for total fungal structures with trypan blue (TB)
or for alkaline phosphatase activity (ALP), a vacuolar enzyme
marker for active fungal structures (see below) (Smith and
Gianinazzi-Pearson 1988). Note that root colonization was
slower under these culture conditions than for the RT-PCR
experiments described above. Over the 10-week period, there
was a progressive increase in the proportion of the root con-
taining visible arbuscules (TB staining), reaching a final value
of 70% (Fig. 5). On the other hand, and despite some varia-
tion, both the proportion of the root system containing func-
tional arbuscules (ALP staining) and the proportion with posi-
tive GUS staining reached a plateau value of approximately
10% after 2 to 4 weeks. Similar results were obtained with M.
truncatula L416 transgenic plants (data not shown). We pre-
sume that this correlation reflects the association of ALP and
GUS activities in the same arbuscule-containing cells and is
consistent with the fact that GUS-negative, arbuscule-
containing cells were generally found in older parts of the root
system (data not shown) where arbuscules are mostly inactive
or senescent (Tisserant et al. 1996). Our interpretation of these
data is that MtENOD11 expression correlates with the func-
tional arbuscule and that transcription is switched off prior to
or during arbuscule senescence.
In order to obtain information concerning reporter gene
expression during the earlier stages of fungal colonization,
experiments were performed making use of an axenic in
vitro mycorrhization system developed to facilitate spatio-
temporal studies. Germinated seedlings of transgenic M.
truncatula or M. varia expressing pMtENOD11–gusA were
transferred to an agar medium containing an actively grow-
ing inoculum composed of Glomus intraradices-infected
carrot hairy roots (see below) (Bécard and Piché 1992).
GUS-positive sites were first detected approximately 5 days
after planting, and microscopical analysis revealed that all
of these sites contained at least a single arbuscule (not
shown). Most sites staining for GUS were located on lateral
roots that had formed after planting and therefore corre-
sponded to recent (i.e., ≤ 3-day-old) colonization events.
These observations again show that MtENOD11 gene ex-
pression within the root inner cortex during the AM interac-
tion is linked to fungal arbuscule formation and further sug-
gest that the gene is activated during the early stages of
arbuscule development.
To determine whether ENOD expression can be triggered
by other biotrophic root fungal associations, we challenged M.
truncatula transgenic plants with a nonpathogenic root-
infecting fungus, a binucleate Rhizoctonia species (see be-
low). The mycelium of this fungus develops on the root sur-
face and forms typical monilioid clusters in epidermal cells
but does not penetrate deeper into the root. Whole-root sys-
tems were stained for GUS activity and the presence of fungal
structures at 2 and 3 weeks postinoculation. Although the ex-
tent of root infection by Rhizoctonia reached 15 and 35%,
respectively, pMtENOD11 reporter gene activity could not be
detected in the root periphery or cortex (data not shown). This
suggests that fungal induction of MtENOD11 is specific to the
symbiotic mycorrhizal association.
742 / Molecular Plant-Microbe Interactions
Expression of pMtENOD11-gusA
in nonsymbiotic cellular contexts.
During our studies of MtENOD11 expression in root tissues
related to the establishment of both bacterial and fungal endo-
symbioses, we also observed reporter gene expression in non-
symbiotic tissues. As mentioned earlier, GUS activity is pre-
sent in root apices of actively growing roots of L416 (Fig.
2D). Microscopical examination of whole root tips and thin
longitudinal sections revealed that pMtENOD11–gusA is, in
fact, expressed in the root cap cells (Fig. 3E). Isolated periph-
eral GUS-staining cells also were found in loose contact with
the epidermis behind the root apex. These correspond to the
“border cells” (Hawes et al. 1998) that have sloughed off from
the root cap. GUS staining was also observed in lateral root
primordia (Charlton 1996) in clusters of dividing pericycle
cells adjacent to the primary root vasculature and was maxi-
mal at early stages of lateral root formation (Fig. 3F). Occa-
sionally, nonlocalized GUS activity was also detected in the
pericycle tissue of older root parts.
In aerial organs of L416, relatively strong GUS activity was
detected in cotyledon (Fig. 3G) and leaf vascular strands (illus-
trated for a recently opened leaf in Fig. 3H) and in pulvini, leaf-
moving organs located at the base of the stalk of each leaflet and
leaf (data not shown). Stomatal guard cells (except on young
stalks before leaf expansion and leaf blades) also stain for GUS
activity (Fig. 3I). Reporter gene expression was observed for
glandular trichomes, although in this case staining was most
intense on young stalks, where stomata do not stain (Fig. 3J).
The precise relationship between MtENOD11 gene expression
and the developmental stage of the plant tissue for these two
specialized epidermal cell types remains to be determined.
During seed development, the pMtENOD11-gusA reporter
was strongly expressed in i) the embryo suspensor as early as 5
to 7 days after pollination (Fig. 3K); ii) the epidermis and under-
lying tissue of the cotyledons, hypocotyl, and radicle in matur-
ing embryos (approximately 18 days after pollination onward
(Fig. 3L); and iii) the outer cell layer of the endosperm (ap-
proximately 10 days after pollination onward) (Fig. 3M), which
appears as a thin envelope enclosing the embryo. This cell layer
displays a characteristic cell morphology that may be related to
transfer cell function (Fig. 3N). The identity of this tissue was
established by crossing wild-type M. truncatula plants with the
transgenic L416 line. In the resulting developing hybrid F1
seeds, GUS activity was always in the embryo and the surround-
ing endosperm, both of which derive from the same double-
fertilization event. The GUS-staining pattern described for the
L416 line in nonsymbiotic tissues was also observed in all other
independent transgenic M. truncatula lines analyzed and was
unaltered in the presence of rhizobial or fungal symbionts. Simi-
lar patterns were observed in the case of M. varia lines, except
for expression in trichomes. In the control wild-type Medicago
spp. plants, no evidence was found for endogenous GUS-like
activity in any organ, except for a low-level background blue
coloration in the developing seed endosperm, mature pollen, and
secondary vascular tissue.
DISCUSSION
MtENOD11, a new molecular marker for early stages
of root nodulation.
In this paper we report the identification of a novel early
nodulin gene from M. truncatula, MtENOD11, and the charac-
terization of its expression with complementary approaches
involving both in situ hybridization and the histochemical
localization of reporter gene activity (GUS) driven by the
MtENOD11 promoter in transgenic Medicago spp. plants.
Within hours of inoculation by S. meliloti, the pMtENOD11–
gusA fusion is activated in differentiating epidermal tissues
located close to the root tip, corresponding to the region of
root hair emergence and development. In legumes such as
Medicago spp., these types of emerging root hairs are the
main targets for S. meliloti attachment and penetration leading
to successful infection and nodulation (Bhuvaneswari et al
1981; Caetano-Anollès and Gresshoff 1991). Parallel experi-
ments revealed that transcription of pMtENOD11-gusA can
also be triggered in differentiating epidermal cells following
the addition of subnanomolar concentrations of purified Nod
factors of S. meliloti (D. G. Barker, J. L. Pingret, M. Chabaud,
and E.-P. Journet, unpublished). At later stages of nodulation,
MtENOD11 promoter activity in both root and nodule tissues
correlates with the presence of infection threads. This gene is
therefore a molecular marker in Medicago spp. for both prein-
fection and infection events in root and nodule cells through-
out the rhizobial symbiotic interaction and, in this respect,
closely resembles the MtENOD12 gene, also encoding a
RPRP (Pichon et al. 1992). The significantly higher steady-
state transcript levels of MtENOD11 (approximately fivefold),
also reflected in the respective transgenic lines, make this a
superior molecular marker in terms of sensitivity. Interest-
ingly, there are differences between MtENOD11 and
MtENOD12 gene expression patterns and those reported for
ENOD12 genes in other legumes. For example, PsENOD12A
(Scheres et al. 1990), VsENOD12 (Vijn et al. 1995), and
MsENOD12A (Bauer et al. 1996) are all expressed in nodule
primordia prior to bacterial invasion, whereas this is not the
case for either MtENOD12 (Pichon et al. 1994) or MtENOD11
(not shown). Finally, whilst genomic hybridization studies
indicate that ENOD11 probably also exists in the closely re-
lated alfalfa (not shown), we do not yet know if an ENOD11
ortholog exists in other legumes.
MtENOD11 is expressed in root inner cortical cells
during endomycorrhizal arbuscule development.
When transgenic Medicago spp. roots are inoculated with
Glomus spp. of AM fungi, the pMtENOD11-gusA fusion is
activated strongly and specifically in inner cortical cells con-
taining arbuscules. We confirmed by RT-PCR that the en-
dogenous MtENOD11 gene is, indeed, activated transcription-
ally during AM colonization. The rapidity of pMtENOD11–
gusA transcriptional activation during mycorrhizal coloniza-
tion of M. truncatula seedlings, together with results from
time-course studies with mature plants, are consistent with the
induction of the MtENOD11 promoter during arbuscule devel-
opment. Because MtENOD11–gusA expression is probably
switched off prior to or during arbuscule senescence (Fig. 5)
and arbuscules are ephemeral fungal structures that start to
senesce several days after reaching full development (Alexan-
der et al. 1989), this reporter gene provides a useful molecular
marker to follow the asynchronous formation of AM infection
units in whole root systems and, in particular, the longitudinal
progression of root colonization in the inner cortex.
Preliminary observations suggest that, during such longitudi-
nal colonization of the root by internal hyphae, pMtENOD11–
Vol. 14, No. 6, 2001 / 743
gusA expression precedes visible arbuscular differentiation in
inner cortical cells. This suggests that the gene may be
switched on very early during arbuscule development. Ex-
periments with transgenic Medicago spp. lines expressing a
pMtENOD12–gusA fusion (Chabaud et al. 1996; Pichon et al.
1992) revealed that the MtENOD12 promoter is also activated
in mycorrhizae with the same cell specificity as for
MtENOD11 (not shown). MtENOD12 expression levels, how-
ever, appear to be significantly lower than for MtENOD11,
reflecting the quantitative difference already observed during
root nodulation.
In a recent study, Albrecht et al. (1998) examined the ex-
pression of the two early nodulin genes PsENOD5 and
PsENOD12A during the interaction of pea roots with the AM
fungus Gigaspora margarita. It was shown by RT-PCR that
PsENOD12A transcripts accumulate at an early stage of the
endomycorrhizal interaction and that steady-state levels de-
crease as colonization proceeds and the first arbuscules de-
velop. This pattern of expression is clearly different from that
observed for MtENOD11 and MtENOD12. However, we are
currently developing an in vitro-targeted inoculation system to
investigate whether the two MtENOD genes might also be
switched on transiently during the earliest stages of mycorrhi-
zal colonization. Other nodulin genes are also expressed dur-
ing the AM symbiosis, including a Vicia faba leghemoglobin
gene (Frühling et al. 1997) and alfalfa ENOD2 and ENOD40
genes (Van Rhijn et al. 1997). In situ hybridization analyses
have shown that MsENOD40 is expressed in the root pericycle
and inner cortical cells containing immature arbuscules,
whereas MsENOD2 transcripts are located in cortical cells
containing mature arbuscules (Van Rhijn et al. 1997). Taken
together, these results show that several types of nodulin
genes are also activated transcriptionally during AM symbio-
sis. It is important to underline, however, that little is known
so far about the extent to which the regulatory pathways lead-
ing to the activation of these various genes are common to
both nodulation and endomycorrhization.
Possible roles for the ENOD11
and ENOD12 repetitive proline-rich proteins.
The nucleotide sequences of the MtENOD11 and
MtENOD12 genes predict that they both encode RPRPs
(Kieliszewski and Lamport 1994; Showalter 1993). However,
the respective PPXXX repeat regions cannot be aligned, and
the sequences flanking the repeat regions share no homology,
making it unlikely that they have evolved from a common
ancestral gene. The presence of an N-terminal signal peptide
that is well conserved among members of the legume PRP
family, e.g., MtPRP1 and MtPRP2 (Wilson et al. 1994),
PsENOD12 (Scheres et al. 1990), and SbPRP3 (Hong et al.
1990), implies that both polypeptides follow a common secre-
tory pathway and are probably targeted to the extracellular
matrix by analogy with the known cellular location of SbPRPs
(Ye et al. 1991) and pea RPRPs (Sherrier and VandenBosch
1994).
One striking feature of ENOD11 and ENOD12 proteins is
their low overall tyrosine (Tyr) content and the virtual absence
of this particular amino acid within the repeated pentapeptide
motif. This contrasts with the situation for classical RPRPs,
where Tyr is frequently part of the repeat motif (Kieliszewski
and Lamport 1994). It has been proposed that Tyr residues in
proline-rich structural proteins can be involved in in-
tramolecular and intermolecular cross linking, and that this
could be responsible for the observed insolubilization of
HRGPs and PRPs in cell walls during developmental or de-
fense responses. Such posttranslational cross-linking is pre-
sumed to result in the mechanical strengthening of the cell
wall and the consequent fixation of cell shape. Cross-linking
also might contribute to the restriction of wall porosity that
appears to be determined primarily by the pectin matrix
(Carpita and Gibeaut 1993, Showalter 1993). If this is the
case, then the presence in the cell wall of significant quantities
of ENOD11 and ENOD12-type proteins with a low potential
for the formation of Tyr-dependent intermolecular cross-links
might result in a reduced capacity for wall strengthening and a
higher matrix porosity.
During the interaction of roots with Rhizobium spp., symbi-
otic MtENOD11 and MtENOD12 expression is linked to
(pre)infection events in epidermal, cortical, and nodule cells.
Thus, both gene products may be involved either in the modi-
fication of the root hair cell wall for subsequent penetration of
bacteria or in the synthesis of new components required for
the infection thread matrix, as proposed previously for
PsENOD12 (Scheres et al. 1990). It is probable that lower
levels of cross-linking and greater plasticity are important
requirements for such walls. Our transgenic plant studies have
revealed that MtENOD11 also is expressed in root cap cells,
cotyledon and leaf vascular strands, leaf pulvini, stomatal
guard cells, glandular trichomes, the suspensor and outer tis-
sues of the developing embryo, and the seed endosperm (Fig.
3). All of these specialized cell-types and/or tissues share the
capacity for high secretion/metabolite exchange, i.e., root cap
and border cells (Hawes et al. 1998; Rougier 1981), phloem
and associated cells in leaves (Esau 1977), glandular
trichomes (Wagner 1991), endosperm (Lopes and Larkins
1993), embryo suspensor (Yeung and Meinke 1993), the outer
epidermis of embryonic cotyledons (Weber et al. 1998),
and/or regulated and reversible variations of cell
shape/volume, i.e., pulvinus (Coté 1995) and stomatal guard
cells (Assmann 1993), for which high cell wall porosity or
plasticity are probably essential characteristics.
In the case of MtENOD12 and in contrast with MtENOD11,
nonsymbiotic expression has been detected only in differenti-
ating xylem cells of lateral root primordia, outer tissues of
developing embryos, and the surrounding endosperm (Journet
et al. 1994; data not shown). Whereas the nonsymbiotic ex-
pression patterns of the two early nodulin genes are therefore
significantly different, it is striking that their transcriptional
regulation is very similar in symbiotic contexts. It is thus pos-
sible that MtENOD11 and MtENOD12 were recruited for
symbiotic functions through the acquisition of additional pro-
moter elements to ensure coordinated expression. It has been
reported that the absence of a functional ENOD12 allele in
particular genotypes of diploid alfalfa does not affect their
capacity to establish an efficient N2-fixing symbiosis with S.
meliloti (Csanadi et al. 1994). Bearing in mind the similarities
between the symbiotic expression patterns of MtENOD11 and
MtENOD12, this finding could be explained most simply by
functional redundancy between the two RPRPs.
Finally, MtENOD11 and MtENOD12 are also expressed
during the endomycorrhizal symbiosis with AM fungi in root
cortical cells containing recently formed arbuscules. Several
744 / Molecular Plant-Microbe Interactions
plant genes and proteins are expressed in cortical cells where
arbuscules develop, including cell wall proteins that are pre-
sumed to be involved in the major extracellular matrix modi-
fications characteristic of this highly differentiated structure
(Bonfante 2001). Arbuscules result from intracellular penetra-
tion by the microsymbiont and correspond to the mature func-
tional structure assumed to be involved in metabolite ex-
change between the two symbiotic partners (Gianinazzi-
Pearson 1996; Harrison 1999). The expression of MtENOD11
and MtENOD12 in these cells might therefore be related either
to hyphal infection, leading to modifications in the properties
of the existing cell wall, or to elaboration of the new arbuscu-
lar matrix interface. To help answer these questions, studies
are now under way to determine the precise cellular location
Vol. 14, No. 6, 2001 / 745
of MtENOD11 and MtENOD12 proteins during symbiotic
interactions and nonsymbiotic developmental processes.
MATERIALS AND METHODS
Isolation of the MtENOD11 genomic clone.
The construction of the genomic library of M. truncatula
leaf DNA in a phasmid vector and the screening of this library
with a pea ENOD12 cDNA probe (Scheres et al 1990) have
been described (Gallusci et al 1991; Pichon et al. 1992). The
clone containing MtENOD11 had a total insert size of ap-
proximately 13.5 kb, and the MtENOD11 coding sequence
was localized within the insert by hybridization. The nucleo-
tide sequence of MtENOD11 has been submitted to the EMBL
databank as accession no. AJ297721.
Plant material and microbial strains.
The wild-type M. truncatula Gaertn. cv. Jemalong lines
used in this study are either A17 or J5 (provided by T. Huguet,
Toulouse, France, and G. Duc, Dijon, France, respectively).
The construction of the pMtENOD11–gusA transcriptional
fusion and the subsequent generation of the M. truncatula line
L416 containing a single-copy homozygous transgene are
reported elsewhere (D. G. Barker, J. L. Pingret, M. Chabaud,
and E.-P. Journet, unpublished). The M. varia lines expressing
pMtENOD11–gusA were obtained using the transformation
protocol described in Pichon et al. (1992). The M. varia and
M. truncatula transgenic lines expressing the pMtENOD12–
gusA construct are described in Pichon et al. (1992) and
Chabaud et al. (1996), respectively.
The wild-type S. meliloti strain RCR2011 was grown at
28°C in standard liquid TY medium supplemented with 6 mM
CaCl2. G. mosseae (Nicol. & Gerd.) Gerd. and Trappe (isolate
BEG12) and G. fasciculatum (Thaxter sensu Gerd.) Gerd. &
Trappe (isolate BEG 53) were cultured on roots of the host
plant Allium porrum L. in soil-based gnobiotic pot cultures in
a growth cabinet (200 µE cm−2 s
−1, 25°C, 16-h photoperiod,
70% relative humidity). A G. intraradices inoculum, provided
by G. Bécard (Université P. Sabatier, Toulouse, France), was
propagated by axenic co-culture of the AM fungus with Dau-
cus carota hairy roots on solid medium according to Bécard
and Piché (1992). The nonpathogenic binucleate Rhizoctonia
species used in this study was isolated from a soil sample of
the INRA Experimental Domaine (Dijon, France) and charac-
terized as belonging to the anastomosis group AG A. This
Rhizoctonia species was cultivated in vitro according to Cam-
porota (1989) to produce an inoculum that was incorporated in
the plant growth substrate at 1g of mycelium per liter.
Plant growth and inoculation conditions.
For nodulation experiments, germinated M. truncatula or
M. varia seedlings were grown in aeroponic conditions at
22°C with a relative humidity of 75%, a 16-h photoperiod, and
light intensity (photosynthetically active radiation) of 200 µE
m−2 s
−1. After 2 to 3 weeks growth on medium containing
5 mM NH4NO3 (Gallusci et al. 1991), combined nitrogen was
removed from the medium and plants were inoculated 3 days
later with the wild-type S. meliloti strain (5 × 105 bacteria per
ml).
For mycorrhization experiments with transgenic lines,
seedlings were transplanted into large pots (400 ml) contain-
ing 2/3 vol of a calcined clay substrate (Terragreen) mixed
with 1/3 volume of a 10-week-old G. fasciculatum inoculum
(roots, spores, mycelium, and soil) and grown in a constant
environment chamber (300 µE cm−2 s−1, 18/22°C, 16-h photo-
period, and 70% relative humidity). The plants were provided
with a modified nutrient Long Ashton solution (Hewitt 1966)
without phosphate but with 15 mM nitrate to ensure nodula-
tion-free conditions. Alternatively, we used an axenic in vitro
mycorrhization system, which was developed in our labora-
tory in collaboration with G. Bécard (Vernoud 1999). Magenta
boxes containing 60 ml of sterile solid M medium (Bécard and
Fortin 1988) with 0.2% (wt/vol) Phytagel (Sigma, St. Louis,
MO, U.S.A.) were initially inoculated with the G. intraradi-
ces–carrot hairy root dual culture (see above) and grown at
25°C in the dark. After approximately 6 weeks, fungal hyphae
had colonized most of the medium and the first newly formed
spores were visible. The medium was then washed five times
(90-min immersion per wash) with approximately 100 ml of
fresh M liquid medium lacking sucrose. Surface-sterilized,
germinated seedlings of transgenic M. truncatula or M. varia
were then planted in the Magenta box medium (up to 15 seed-
lings per box) so that the developing roots grew vertically
through the highly infectious hyphal network. Plants could be
grown for 2 to 3 weeks in a growth chamber under these con-
ditions (25°C, 16-h photoperiod, 70 µE m−2 s−1).
In situ hybridization.
For the preparation of the single-stranded RNA probes, the
430-bp internal MaeI–BstEII restriction fragment from the
←
Fig. 3. Histochemical analysis of MtENOD11–gusA expression in transgenic M. truncatula A–D, during the arbuscular mycorrhizal symbiosis and
E–N, in nonsymbiotic tissues. A, Whole root segment showing GUS activity in arbuscular mycorrhizal-
colonized inner root cortical cells. External
hyphae (arrowheads) and spores of Glomus intraradices are stained red with acid fuchsin. Bar = 250 µm. B, 80-
µm transverse root section of a colonized
root. Bar = 150 µm. C, Semi-
thin transverse section (4 µm; counterstained with toluidine blue) of a colonized root. Arrows indicate sections through
internal hyphae and arrowheads indicate arbuscule-containing cells. Bar = 80 µm. D, Detail of arbuscule-
containing cells after double staining with
Magenta–glcUA for GUS activity (in red) and ink and vinegar for fungal structures (in dark-blue). Arrow points to an internal hypha. Bar = 60 µm. E,
Longitudinal thin section (4 µm) of root tip viewed under dark-field optics. GUS activity (blue staining appears reddish in dark-field) is associated with
cells of the root cap that surrounds the root meristem. Bar = 125 µm. F, Late
ral root primordium within a whole root segment. GUS staining marks the
first dividing pericycle cells next to the root vascular strands (star). Bar = 75 µm. G,
Abaxial part of a fully expanded seedling cotyledon showing GUS
activity associated with the whole vascular network. Bar = 600 µm. H, Portion of a recently opened leaf showing GUS activity associated predominantly with
the minor vein network. Bar = 750 µm. I, Stomatal guard cells on the leaf stalk epidermis with strong GUS activity. Bar = 30 µm. J, Glandular trichome lo-
cated on a young stalk and showing intense GUS activity. Note that GUS staining is absent for the two stomata visible on this image (see text). Bar = 50 µm.
K, GUS activity in the embryo suspensor (11 days after pollination), which remains attached to the embryonic root tip. Bar = 500 µm. L, GUS staining o
f
whole maturing embryo 3 weeks after pollination. Staining is associated with the outer surface of cotyledons, correlating with the distribution of epidermal
transfer cells in these organs (Weber et al. 1998). Bar = 750 µm. M, Whole endosperm (15 days after pollination) stained after removal of embryo. Bar = 750
µm. N, Detail of the GUS-positive outer cell layer of the endosperm exhibiting a characteristic cell morphology. Bar = 50 µm.
746 / Molecular Plant-Microbe Interactions
MtENOD11 gene was subcloned into pBlueScript KS+ and
SK+ vectors (Stratagene, La Jolla, CA, U.S.A.). RNA synthe-
sis with T7 DNA polymerase and partial hydrolysis of radio-
labeled sense and antisense RNA was carried out as described
in De Billy et al. (1991). In situ hybridizations on 7-µm nod-
ule sections with 35S-labeled riboprobes were performed as
described previously (De Billy et al. 1991), except for the
addition of a 24-h prehybridization step with the standard
hybridization buffer minus dextran sulfate.
Histochemical localization
of reporter enzyme activities and fungal structures.
Histochemical staining (blue) for β-glucuronidase (GUS)
activity was performed on whole-root segments for up to 24 h
with the substrate X-glcUA (5-bromo-4-chloro-3-indolyl glu-
curonide, cyclohexylammonium salt) (Biosynth, Staad, Swit-
zerland) as described in Journet et al. (1994). Alternatively,
the Magenta–glcUA substrate (Biosynth) was used to obtain
red-colored GUS staining. Stem, leaf, and flower pieces were
vacuum infiltrated and stained in the presence of 0.02% Triton
X-100. Intact nodules were prefixed with 0.5% paraformalde-
hyde prior to slicing into 80-µm sections (Microcut H1200;
Bio-Rad, Hercules, CA, U.S.A.). GUS staining times for sec-
tions were generally below 2 h. When necessary, stained intact
samples or slices were cleared briefly with sodium hypochlo-
rite, as described in Pichon et al. (1992), or with 100% ethanol
for chlorophyll-containing explants and observed with a
Wild/Leitz stereomicroscope (Leica Microsystems, Wetzler,
Germany) or a Zeiss Axiophot light microscope (Carl Zeiss,
Le Pecq, France). For the localization of GUS activity at the
cellular level, stained roots were postfixed for 1 h in 2.5%
glutaraldehyde buffered in 0.1 M potassium phosphate (pH
7.0), dehydrated in an alcohol series, and embedded in Tech-
novit 7100 resin (Heraeus Kulzer, Wehrheim, Germany). Sec-
tions (4 µm) were observed with bright- and dark-field mi-
croscopy after counterstaining using basic fuchsin (0.5% in
distilled water).
After digestion of mycorrhizal samples with 10% KOH
(90ºC, 10 to 45 min), total fungal structures were stained with
either 0.05% (wt/vol) Trypan blue, 0.5% (wt/vol) acid fuchsin
(Bundrett et al. 1984), or ink and vinegar (Vierheilig et al.
1998). The double-staining procedure used for mychorrhizal
roots to detect both GUS (Magenta-glcUA) and fungal struc-
tures (ink and vinegar) was developed by M. Chabaud. Fungal
alkaline phosphatase activity was localized after staining with
I-napthyl phosphate (Tisserant et al. 1993). Parameters of
mycorrhizal colonization (mycorrhizal frequency, F%; arbus-
cule abundance, A%) were determined after staining according
to Trouvelot et al. (1986).
RT-PCR assay to detect MtENOD11 transcripts.
Jemalong J5 seedlings were transferred to small, 100-ml
multipots, which were inoculated with the fungal isolate G.
mosseae (BEG 12) and watered with a modified nutrient Long
Ashton solution without phosphate, as described above.
Whole root systems of 7 to 16 plants per sample were har-
vested 2 and 3 weeks after transplantation and ground in liq-
uid nitrogen. Total RNA was isolated by a standard phenol–
sodium dodecyl sulfate protocol and quantified using
spectrophotometry and agarose gel electrophoresis. Prior to
RT-PCR, genomic DNA was eliminated and the first cDNA
strand was synthesized, as described in Pingret et al. (1998).
cDNA aliquots (2 µl), equivalent to 60 ng of initial total RNA,
were then used in PCR reactions. A 437-bp MtENOD11 cDNA
fragment was amplified (30 cycles at 94°C for 30 s, 60°C for
30 s, and 72°C for 30 s) at nucleotide positions 83 to 519
(relative to the ATG codon) of the genomic sequence (EMBL
accession no. AJ297721) with forward primer 5′-CTCCATC-
CCACAATATGCCTCC-3′ and reverse primer 5′-ATGGATG-
CTAGGTGGAGGCT-3′. Parallel amplification of the gene
MtPR10-1 expressed constitutively in roots (Gamas et al.
1998) (GenBank accession no. Y08726) was performed to
control for equivalent cDNA levels in samples. MtPR10-1
cDNA was amplified during 22 cycles according to Pingret et
al. (1998). PCR products were analyzed by Southern blotting
and hybridization with a 430-bp internal MaeI–BstEII DNA
Fig. 4. Transcriptional activation of MtENOD11 in Medicago truncatula
arbuscular mycorrhiza. Total RNA was extracted from entire root sys-
tems of M. truncatula plants 2 weeks postinoculation with Glomus
mosseae and from noninoculated control plants. MtENOD11 transcript
levels were evaluated by reverse transcription-polymerase chain reaction
analysis (see text). Constitutive expression of the MtPR10-1 gene was
used as an internal control for the quality and quantity of cDNA sam-
ples. Identical results were obtained in three independent experiments.
Fig. 5. Time-
course experiment illustrating the relationship between the
extent of root colonization, arbuscule functionality, and pMtENOD11–
gusA expression. Whole root systems of transgenic Medicago varia
inoculated with Glomus fasciculatum were harvested at the indicated
time points, and samples (30 randomly selected 1-cm fragments pe
r
treatment) were stained for fungal structures (trypan blue), fungal alka-
line phosphatase (ALP), or reporter GUS activity. Four plants per time
point were used as replicates. Percentage values of arbuscule abundance
as revealed by the various histochemical staining procedures were esti-
mated according to Trouvelot et al. (1986). Small bars indicate standard
deviations.
Vol. 14, No. 6, 2001 / 747
fragment from MtENOD11 and the complete cDNA from
MtPR10-1. Identical results were obtained whether plants
were grown in the absence or presence of 15 mM nitrate (to
inhibit nodulation by possible contaminating rhizobia), con-
firming that the MtENOD11 signal was not the result of rhizo-
bial contamination of the fungal inoculum.
ACKNOWLEDGMENTS
We are grateful to T. Bisseling (Wageningen, The Netherlands) for his
initial gift of the PsENOD12 cDNA clone, and to G. Bécard (Université
P. Sabatier, Toulouse, France) for providing us with an axenic Glomus
intraradices–hairy root dual culture and his expert advice during the
setting up of the axenic in vitro mycorrhization system. Our thanks also
to M. Chabaud for providing us with the photograph illustrating the
double staining (for GUS activity and fungal tissue) of the arbuscule-
containing cells, M. Chabaud and Pascal Gamas for critical reading of
the manuscript, and other colleagues for helpful comments. Financial
support was provided by INRA (AIP Biologie du Développement 1995-
97), the EU in the framework of TMR network funding (ref.
ERBFMRXCT980243), the Human Frontier Research Program (ref. RG-
327/98), and INRA–CNRS for the Medicago truncatula genome project
(1998-2000). V. Vernoud received a Ph.D. grant from the French Min-
istère de l’Enseignement Supérieur et de la Recherche.
LITERATURE CITED
Albrecht, C., Geurts, R., Lapeyrie, F., and Bisseling, T. 1998. Endomy-
corrhizae and rhizobial Nod factors both require SYM8 to induce the
expression of the early nodulin genes PsENOD5 and PsENOD12A.
Plant J. 15:605-614.
Albrecht, C., Geurts, R. and Bisseling, T. 1999. Legume nodulation and
mycorrhizae formation: Two extremes in host specificity meet. EMBO
J. 18:281-288.
Alexander, T., Toth, R., Meier, R., and Weber, H. C. 1989. Dynamics of
arbuscule development and degeneration in onion, bean and tomato
with reference to vesicular-arbuscular mycorrhizae in grasses. Can. J.
Bot. 67:2505-2513.
Allison, L. A., Kiss, G. B., Bauer, P., Poiret, M., Pierre, M., Savouré, A.,
Kondorosi, E., and Kondorosi, A. 1993. Identification of two alfalfa
early nodulin genes with homology to members of the pea ENOD12
gene family. Plant Mol. Biol. 21:375-380.
Assmann, S. M. 1993. Signal transduction in guard cells. Annu. Rev.
Cell Biol. 9:345-375.
Bauer, P., Ratet, P., Crespi, M. D., Schultze, M., and Kondorosi, A.
1996. Nod factors and cytokinins induce similar cortical cell division,
amyloplast deposition and MsEnod12A expression patterns in alfalfa
roots. Plant J. 10:91-105.
Bécard, G., and Fortin, J. A. 1988. Early events of vesicular-arbuscular
mycorrhiza formation on Ri T-DNA transformed roots. New Phytol.
108:211-218.
Bécard, G., and Piché, Y. 1992. Establishment of vesicular-arbuscular
mycorrhiza in root organ culture: Review and proposed methodology.
Pages 89-108 in: Methods in Microbiology, Vol. 24. J. R. Norris, D. J.
Read, and A. K. Varma, eds. Academic Press, London.
Bhuvaneswari, T. V., Bhagwat, A. A., and Bauer, W. D. 1981. Transient
susceptibility of root cells in four common legumes to nodulation by
Rhizobia. Plant Physiol. 68:1144-1149.
Bonfante, P. 2001. At the interface between mycorrhizal fungi and
plants: The structural organization of cell wall, plasma membrane and
cytoskeleton. Pages 45-61 in: The Mycota IX, Fungal Associations. B.
Hock, ed. Springer-Verlag, Berlin.
Bundrett, M. C., Piché, Y., and Peterson, R. L. 1984. A new method for
observing the morphology of vesicular-arbuscular mycorrhizae. Can.
J. Bot. 62:2128-2134.
Caetano-Anollés, G., and Gresshoff, P. M. 1991. Alfalfa controls nodula-
tion during the onset of Rhizobium-induced cortical cell division.
Plant Physiol. 95:366-373.
Camporota, P. 1989. Maladies des plantes dues à Rhizoctonia solani
(Khün): Stratégie et techniques d’études. Agronomie 9:327-334.
Carpita, N. C., and Gibeaut, D. M. 1993. Structural models of primary
cell walls in flowering plants: Consistency of molecular structure with
the physical properties of the walls during growth. Plant J. 3:1-30.
Chabaud, M., Larsonneau, C., Marmouget, C., and Huguet, T. 1996.
Transformation of barrel medic (Medicago truncatula Gaertn) by
Agrobacterium tumefaciens and regeneration via somatic embryo-
genesis of transgenic plants with the MtENOD12 nodulin promoter
fused to the gus reporter gene. Plant Cell Rep. 15:305-310.
Charlton, W. A. 1996. Lateral root initiation. Pages 149-173 in: Plant
Roots: The Hidden Half, 2nd ed. Y. Waisel, A. Eskel, and U. Kafkafi,
eds. Marcel Dekker, New York.
Coté, G. G. 1995. Signal transduction in leaf movement. Plant Physiol.
109:729-734.
Csanadi, G., Szecsi, J., Kalo, P., Kiss, P., Endre, G., Kondorosi, A.,
Kondorosi, E., and Kiss, G.B. 1994. ENOD12, an early nodulin gene,
is not required for nodule formation and efficient nitrogen fixation in
alfalfa. Plant Cell 6:201-213.
De Billy, F., Barker, D. G., Gallusci, P., and Truchet, G. 1991.
Leghaemoglobin gene transcription is triggered in a single cell layer
of the indeterminate nitrogen-fixing root nodule of alfalfa. Plant J.
1:27-35.
Esau, K. 1977. Anatomy of Seed Plants, 2nd ed. Wiley & Sons, New
Yo rk .
Frühling, M., Roussel, H., Gianinazzi-Pearson, V., Puhler, A., and
Perlick, A. M. 1997. The Vicia faba leghemoglobin gene VfLb29 is
induced in root nodules and in roots colonized by the arbuscular my-
corrhizal fungus Glomus fasciculatum. Mol. Plant-Microbe Interact.
10:124-131.
Gallagher, S. R. 1992. GUS protocols: Using the GUS gene as a reporter
of gene expression . S. R. Gallagher, ed. Academic Press, London.
Gallusci, P., Dedieu, A., Journet, E.-P., Huguet, T., and Barker, D. G.
1991. Synchronous expression of leghaemoglobin genes in Medicago
truncatula during nitrogen-fixing root nodule development and re-
sponse to exogenously supplied nitrate. Plant Mol. Biol. 17:335-349.
Gamas, P., Niebel, F. D. C., Lescure, N., and Cullimore, J. V. 1996. Use
of a subtractive hybridization approach to identify new Medicago
truncatula genes induced during root nodule development. Mol.
Plant-Microbe Interact. 9:233-242.
Gamas, P., De Billy, F., and Truchet, G. 1998. Symbiosis-specific ex-
pression of two Medicago truncatula nodulin genes, MtN1 and
MtN13, encoding products homologous to plant defense proteins.
Mol. Plant-Microbe Interact. 11:393-403.
Gianinazzi-Pearson, V. 1996. Plant cell responses to arbuscular my-
corrhizal fungi: Getting to the roots of the symbiosis. Plant Cell
8:1871-1883.
Gianinazzi-Pearson, V. 1997. Have common plant systems co-evolved in
fungal and bacterial root symbioses? Pages 321-324 in: Biological
Fixation of Nitrogen for Ecology and Sustainable Agriculture: NATO
ASI Series, Vol G39. A. Legocki, H. Bothe, and A. Puhler, eds.
Springer-Verlag, Berlin.
Hadri, A. E., Spaink, H. P., Bisseling, T., and Brewin, N. J. 1998. Diver-
sity of root nodulation and rhizobial infection process. Pages 347-360
in: The Rhizobiaceae—Molecular biology of model plant-associated
bacteria. H. P. Spaink, A. Kondorosi, and P. J. J. Hooykaas, eds. Klu-
wer Academic Publishers, Dordrecht, The Netherlands.
Harrison, M. J. 1997. The arbuscular mycorrhizal symbiosis: An under-
ground association. Trends Plant Sci. 2:54-56.
Harrison, M. J. 1999. Molecular and cellular aspects of the arbuscular
mycorrhizal symbiosis. Annu. Rev. Plant Physiol. Plant Mol. Biol.
50:361-389.
Hawes, M. C., Brigham, L. A., Wen, F., Woo, H., and Zhu, Y. 1998.
Function of root border cells in plant health: Pioneers in the
rhizosphere. Annu. Rev. Phytopathol. 36:311-327.
Hewitt, E. J. 1966. Sand and water culture methods used in the study of
plant nutrition. Pages 430-434 in: Technical Communication, 22nd
ed. Commonwealth Agricultural Bureau, London.
Hong, J. C., Nagao, R. T., and Key, J. L. 1990. Characterization of a
proline-rich cell wall protein gene family of soybean. A comparative
analysis. J. Biol. Chem. 265:2470-2475.
Journet, E.-P., Pichon, M., Dedieu, A., De Billy, F., Truchet, G., and
Barker, D. G. 1994. Rhizobium meliloti Nod factors elicit cell-specific
transcription of the ENOD12 gene in transgenic alfalfa. Plant J. 6:241-
249.
Kieliszewski, M. J., and Lamport, D. T. 1994. Extensin: Repetitive mo-
tifs, functional sites, post-translational codes, and phylogeny. Plant J.
748 / Molecular Plant-Microbe Interactions
5:157-72.
LaRue, T. A., and Weeden, N. F. 1994. The symbiosis genes of the host.
Pages 147-151 in: Proceedings of the 1st European Nitrogen Fixation
Conference. G. B. Kiss and G. Endre, eds. Officina Press, Szeged,
Hungary.
Löbler, M., and Hirsch, A. M. 1993. A gene that encodes a proline-rich
nodulin with limited homology to PsENOD12 is expressed in the in-
vasion zone of Rhizobium meliloti-induced alfalfa root nodules. Plant
Physiol. 103:21-30.
Lopes, M. A., and Larkins, B. A. 1993. Endosperm origin, development,
and function. Plant Cell 5:1383-1399.
Mylona, P., Pawlowski, K., and Bisseling, T. 1995. Symbiotic nitrogen
fixation. Plant Cell 7:869-885.
Pawlowski, K. 1997. Nodule-specific gene expression. Physiol. Planta-
rum 99:617-631.
Penmetsa, R. V., and Cook, D. R. 1997. A legume ethylene-insensitive
mutant hyperinfected by its rhizobial symbiont. Science 275:527-530.
Pichon, M., Journet, E.-P., Dedieu, A., De Billy, F., Truchet, G., and
Barker, D. G. 1992. Rhizobium meliloti elicits transient expression of
the early nodulin gene ENOD12 in the differentiating root epidermis
of transgenic alfalfa. Plant Cell 4:1199-1211.
Pichon, M., Journet, E.-P., De Billy, F., Dedieu, A., Huguet, T., Truchet,
G., and Barker, D. G. 1994. ENOD12 gene expression as a molecular
marker for comparing Rhizobium-dependent and -independent nodula-
tion in alfalfa. Mol. Plant-Microbe Interact. 7:740-747.
Pingret, J. L., Journet, E.-P., and Barker, D. G. 1998. Rhizobium nod
factor signaling. Evidence for a G protein-mediated transduction
mechanism. Plant Cell 10:659-672.
Rougier, M. 1981. Secretory activity of the root cap. Pages 542-574 in:
Encyclopedia of Plant Physiology, Vol. 13B. W. Tanner and F. A.
Loewus, eds. Springer-Verlag, Berlin.
Scheres, B., Van de Wiel, C., Zalensky, A., Horvath, B., Spaink, H., Van
Eck, H., Zwartkruis, F., Wolters, A.-M., Gloudemans, T., Van
Kammen, A., and Bisseling, T. 1990. The ENOD12 gene product is
involved in the infection process during the pea-Rhizobium interac-
tion. Cell 60:281-294.
Schultze, M., and Kondorosi, A. 1998. Regulation of symbiotic root
nodule development. Annu. Rev. Genet. 32:33-57.
Sherrier, D. J., and VandenBosch, K. 1994. Localization of repetitive
proline-rich proteins in the extracellular matrix of pea root nodules.
Protoplasma 183:148-161.
Showalter, A. M. 1993. Structure and function of plant cell wall pro-
teins. Plant Cell 5:9-23.
Smith, S. E., and Gianinazzi-Pearson, V. 1988. Physiological interac-
tions between symbionts in vesicular-arbuscular mycorrhizal plants.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 39:221-244.
Smith, S. E., and Smith, F. A. 1990. Structure and function of the inter-
faces in biotrophic symbioses as they relate to nutrient transport. New
Phytol. 114:1-38.
Smith, S. E., and Read D. J. 1997. Mycorrhizal Symbiosis, 2nd ed. Aca-
demic Press, London.
Tisserant, B., Gianinazzi-Pearson, V., Gianinazzi, S., and Gollotte, A.
1993. In planta histochemical staining of fungal alkaline phosphatase
activity for analysis of efficient arbuscular mycorrhizal infections.
Mycol. Res. 97:245-250.
Tisserant, B., Gianinazzi, S., and Gianinazzi-Pearson, V. 1996. Relation-
ships between lateral root order, arbuscular mycorrhiza development,
and the physiological state of the symbiotic fungus in Platanus aceri-
folia. Can. J. Bot. 74:1947-1955.
Trouvelot, A., Kough, J. L., and Gianinazzi-Pearson, V. 1986. Mesure du
taux de mycorhization VA d’un système radiculaire. Recherche de
méthodes d’estimation ayant une signification fonctionnelle. Pages
217-221 in: Physiological and Genetic Aspects of Mycorrhizae. V.
Gianinazzi-Pearson and S. Gianinazzi, eds. INRA Press, Paris.
Van Rhijn, P., Fang, Y., Galili, S., Shaul, O., Atzmon, N., Wininger, S.,
Eshed, Y., Lum, M., Li, Y., To, V., Fujishige, N., Kapulnik, Y., and
Hirsch, A. M. 1997. Expression of early nodulin genes in alfalfa my-
corrhizae indicates that signal transduction pathways used in forming
arbuscular mycorrhizae and Rhizobium-induced nodules may be con-
served. Proc. Natl. Acad. Sci. USA 94:5467-5472.
Vasse, J., De Billy, F., and Truchet, G. 1990. Correlation between ultra-
structural differentiation of bacteroids and nitrogen fixation in alfalfa
nodules. J. Bacteriol. 172:4295-4306.
Vernoud, V. 1999. Le gene MtENOD20: Un marqueur moléculaire pré-
coce de l’organogenèse nodulaire lors de l’interaction symbiotique
Sinorhizobium meliloti/Medicago. PhD thesis. Université P. Sabatier,
Toulouse, France.
Vernoud, V., Journet, E.-P., and Barker, D. G. 1999. MtENOD20, a Nod
factor-inducible molecular marker for root cortical cell activation.
Mol. Plant-Microbe Interact. 12:604-614.
Vierheilig, H., Coughlan, A. P., Wyss, U., and Piché, Y. 1998. Ink and
vinegar, a simple staining technique for arbuscular-mycorrhizal fungi.
Appl. Environ. Microbiol. 64:5004-5007.
Vijn, I., Yang, W. C., Pallisgard, N., Jensen, E. O., Van Kammen, A., and
Bisseling. T. 1995. VsENOD5, VsENOD12 and VsENOD40 expres-
sion during Rhizobium-induced nodule formation on Vicia sativa
roots. Plant Mol. Biol. 28:1111-1119.
Von Heijne, G. 1983. Patterns of amino-acids near signal sequence
cleavage sites. Eur. J. Biochem. 133:17-21.
Wagner, G. J. 1991. Secreting glandular trichomes: More than just hairs.
Plant Physiol. 96:675-679.
Weber, H., Heim, U., Golombek, S., Borisjuk, L., and Wobus, U. 1998.
Assimilate uptake and the regulation of seed development. Seed Sci.
Res. 8:331-345.
Wilson, R. C., Long, F., Maruoka, E. M., and Cooper, J. B. 1994. A new
proline-rich early nodulin from Medicago truncatula is highly ex-
pressed in nodule meristematic cells. Plant Cell 6:1265-1275.
Ye, Z. H., Song, Y. R., Marcus, A., and Varner, J. E. 1991. Compara-
tive Localization of three classes of cell wall proteins. Plant J.
1:175-183.
Yeung, E. C., and Meinke, D. W. 1993. Embryogenesis in angiosperms:
Development of the suspensor. Plant Cell 5:1371-1381.
