Content uploaded by Catalina Iulia Pislariu
Author content
All content in this area was uploaded by Catalina Iulia Pislariu
Content may be subject to copyright.
404 / Molecular Plant-Microbe Interactions
MPM I Vol. 21, No. 4, 2008, pp. 404–410. doi:10.1094/ MPMI -21-4-0404. © 2008 The American Phytopathological Society
Transcription of ENOD8
in Medicago truncatula Nodules Directs ENOD8 Esterase
to Developing and Mature Symbiosomes
Laurent Coque,1 Purnima Neogi,1 Catalina Pislariu,1 Kimberly A. Wilson,1 Christina Catalano,2
Madhavi Avadhani,2 D. Janine Sherrier,2 and Rebecca Dickstein1
1University of North Texas, Department of Biological Sciences, Chestnut and Avenue C, Denton 76203-5220, U.S.A.;
2Department of Plant and Soil Sciences and Delaware Biotechnology Institute, University of Delaware, Newark 19711, U.S.A.
Submitted 25 October 2007. Accepted 5 December 2007.
In Medicago truncatula nodules, the soil bacterium Sino-
rhizobium meliloti reduces atmospheric dinitrogen into
nitrogenous compounds that the legume uses for its own
growth. In nitrogen-fixing nodules, each infected cell con-
tains symbiosomes, which include the rhizobial cell, the
symbiosome membrane surrounding it, and the matrix
between the bacterium and the symbiosome membrane,
termed the symbiosome space. Here, we describe the local-
ization of ENOD8, a nodule-specific esterase. The onset of
ENOD8 expression occurs at 4 to 5 days postinoculation,
before the genes that support the nitrogen fixation capa-
bilities of the nodule. Expression of an ENOD8 promoter–
gusA fusion in nodulated hairy roots of composite trans-
formed M. truncatula plants indicated that ENOD8 is ex-
pressed from the proximal end of interzone II to III to the
proximal end of the nodules. Confocal immunomicroscopy
using an ENOD8-specific antibody showed that the
ENOD8 protein was detected in the same zones. ENOD8
protein was localized in the symbiosome membrane or
symbiosome space around the bacteroids in the infected
nodule cells. Immunoblot analysis of fractionated symbio-
somes strongly suggested that ENOD8 protein was found
in the symbiosome membrane and symbiosome space, but
not in the bacteroid. Determining the localization of
ENOD8 protein in the symbiosome is a first step in under-
standing its role in symbiosome membrane and space dur-
ing nodule formation and function.
Nitrogen-fixing nodules form on the roots of legumes as the
culmination of a successful interaction between the plant and
soil bacteria called rhizobia. Nodule development is a complex
process that involves differentiation of both symbiotic partners
(Brewin 2004; Gage 2004; Hirsch 1992; Jones et al. 2007). After
attaching to root hairs, reorienting root hair cell wall growth,
and forming a bacterial colony within a tightly curled root hair,
rhizobia enter plant roots through plant-derived infection threads
that traverse several cell layers. Ultimately, the rhizobia are de-
posited in nodule primordium cells in a process that resembles
endocytosis or phagocytosis.
A plasma membrane-derived membrane surrounds each bac-
terium and results in a new organelle-like compartment in the
infected nodule cells termed the symbiosome. Within symbio-
somes, coordinated division of rhizobia and the symbiosome
membrane occurs. Eventually, rhizobial division ceases and
rhizobia differentiate into bacteroids, a form recognized as
being capable of nitrogen fixation (Vasse et al. 1990). The ma-
trix that surrounds the bacteroid is called the symbiosome
space. In some literature, the symbiosome membrane and the
symbiosome space are termed the peribacteroid membrane and
peribacteroid space, respectively (Hirsch 1992). Both the sym-
biosome space and the symbiosome membrane become spe-
cialized during nodule development concomitantly with bac-
teroid differentiation, and these changes eventually enable
nitrogen fixation and transport.
In mature indeterminate nodules, such as the type formed by
Medicago truncatula, nodule invasion occurs immediately ad-
jacent to the nodule meristem, in zone II (Vasse et al. 1990).
Rhizobial division and differentiation occur within zone II.
Starch accumulation marks the interzone II to III, proximal to
zone II, where the final stages of bacteroid differentiation
occur. Nitrogen fixation occurs in zone III. In older nodules, a
senescent zone IV also occurs (Hirsch 1992).
The symbiosome membrane serves as a physical barrier and
a mediator of metabolite exchange between the bacteroid and
the plant host cell cytosol (Udvardi and Day 1997). In nitrogen-
fixing nodule cells, the symbiosome membrane resembles both
the plasma membrane and the vacuole membrane with respect
to its substituent proteins (Brewin 1991; Mylona et al. 1995).
Previous studies have focused on the biochemical composition
of the symbiosome membrane and how it differs from other
cellular membranes, how it changes during symbiosome and
bacteroid maturation, and functions during nodule develop-
ment and nitrogen fixation (Catalano et al. 2004, 2007; Panter
et al. 2000; Wienkoop and Saalbach 2003).
Proteomic analysis indicates that approximately 200 proteins
exist in the symbiosome space in M. truncatula (Catalano et al.
2004) and in pea (Saalbach et al. 2002). Immunolocalization
studies have shown that a lectin protein (Dahiya et al. 1997;
Kardailsky et al. 1996) and a cysteine protease are in the sym-
biosome space (Vincent and Brewin 2000) in addition to sev-
Current address of L. Coque: Department of Psychiatry, University o
f
Texas Southwestern Medical Center, Dallas 75390-9070, U.S.A.
Current address of C. Pislariu: The Samuel Roberts Noble Foundation,
Ardmore, OK 73401, U.S.A.
Current address of K. A. Wilson: Departments of Pediatrics and Bio-
chemistry, University of Texas Southwestern Medical School, Dallas, TX
75390-9063, U.S.A.
Corresponding author: Rebecca Dickstein;
E-mail: beccad@unt.edu; Fax +1.940.565.3821.
*The e-Xtra logo stands for “electronic extra” and indicates that a sup-
plemental figure is published online.
e-Xt
r
a*
Vol. 21, No. 4, 2008 / 405
eral members of a calcium-binding protein family (Liu et al.
2006). However, the functional significance of many of the
proteins in the symbiosome space is largely unknown.
In this report, we describe the localization of ENOD8. In
M. truncatula, ENOD8 is a member of a duplicated gene
family of which only ENOD8 is highly active in root nodules
(Dickstein et al. 2002). ENOD8 belongs to the GDSL family
encoding hydrolytic enzymes (Akoh et al. 2004; Upton and
Buckley 1995). This gene family, consisting of plant and bac-
terial genes, encodes hydrolytic proteins, some of which have
esterase, lipase, or carbohydrate hydrolase activity. Of the
GDSL plant enzymes that have been biochemically character-
ized, lanatoside 15′-O-acetylesterase (LAE) from Digitalis
lanata is a close homolog to ENOD8, with 44% sequence
identity at the amino acid level (Kandzia et al. 1998). LAE
deacetylates the substrate lanatoside A, an acetylated glycosy-
lated steroid, to purpureaglycoside A in cardenolide metabo-
lism. Another close homolog of ENOD8 (47% identical) en-
codes an exopolygalacturonase from Daucus carota, carrot
(Tanaka et al. 2002). The exopolygalacturonase is 97% identi-
cal to iEP4 (Bertinetti and Ugalde 1996) and EP4, found
extracellularly (van Engelen et al. 1995). Purified M. sativa
ENOD8 proteins have esterase activity that is more active on
shorter-chain esters than on longer-chain esters (Pringle and
Dickstein 2004). ENOD8 protein was shown to be completely
soluble in aqueous solutions (Pringle and Dickstein 2004).
Curiously, ENOD8 protein also was identified in purified sym-
biosome membrane fractions (Catalano et al. 2004). To date,
however, there has been no comprehensive study of the distri-
bution of ENOD8 protein localization.
Here, we report on the spatial localization of a pENOD8-
gusA transgene in M. truncatula nodules. We also report on the
localization of the ENOD8 protein product to infected cells of
this same region of nitrogen-fixing nodules. Within the nodules,
ENOD8 protein localizes to the symbiosome space and sym-
biosome membrane. These results provide intriguing insights
into the role of potential esterase activity of ENOD8 protein in
nodule formation and function.
RESULTS
Localization of ENOD8 gene expression.
To determine the spatial localization of ENOD8-regulated
transcription, a translational fusion of the 2.98-kb sequence
upstream of the highly active ENOD8.1 gene and the reporter
gene gusA was constructed (pMtENOD8-gusA). This was used
to transform M. truncatula A17 roots in a hairy root system
(Boisson-Dernier et al. 2001). Control plants were transformed
with Agrobacterium rhizogenes carrying the pMtENOD11-gusA
fusion (Boisson-Dernier et al. 2001). Composite plants with
transformed roots were placed in aeroponics chambers and
inoculated with Sinorhizobium meliloti to induce nodulation.
Three weeks after inoculation, nodulated root systems were
harvested and stained with 5-bromo-4-chloro-3-indolyl-β-D-
glucuronide, cyclohexylammonium salt (X-Gluc) (Fig 1).
pMtENOD8-gusA expression was found primarily in nodule
zone III. Occasionally, expression also was detected in the
proximal end of interzone II to III (IZ II-III). The zones were
identified by staining nodules with iodine to identify the starch
that accumulates in IZ II-III (not shown). No expression was
seen in the meristem or in the uninfected nodule periphery. Ex-
pression of pMtENOD11-gusA in control nodules showed X-
Gluc staining at the distal end of the nodule, similar to published
results (not shown) (Boisson-Dernier et al. 2001; Journet et al.
2001). The nodules formed on transgenic hairy roots with both
the pMtENOD8-gusA transgene construct and control constructs
were smaller than those found on untransformed tissue. There
was some variability of staining from nodule to nodule; most of
the nodules examined came from independent transformation
events. The nodule shown in Figure 1 represents the staining
pattern found in most pMtENOD8-gusA-transformed nodules.
Localization of ENOD8 protein at the tissue level.
An ENOD8 antiserum was used to localize ENOD8 protein
in fixed nodule sections. This antibody that was raised against
the oligopeptide CKNPSTRITWDGTHYTEAA was reported
previously (Dickstein et al. 2002; Pringle and Dickstein 2004).
This sequence corresponds to amino acids 336 to 354 of the
M. sativa ENOD8 putative polypeptide sequence (GenBank
accession AAB41547) and an 89% identical sequence in M.
truncatula’s predicted ENOD8 protein (GenBank accession
AAC26810). This antiserum has high specificity for ENOD8
protein from both M. sativa and M. truncatula (Dickstein et al.
2002). In conditions tested previously, the antiserum was not
an effective ENOD8 protein localization tool (Pringle and
Dickstein 2004). After trial and error, a protocol was found
that recovers antigen signal loss during fixation (see Methods).
Additionally, and in confirmation of staining patterns obtained,
we raised another ENOD8 antiserum to a recombinant mal-
tose-binding protein (MBP)-ENOD8 fusion protein and used it
instead of the ENOD8 oligopeptide antiserum in additional
immunolocalization experiments not reported here. In every
tested case, identical localization patterns were observed when
the MBP-ENOD8 fusion protein antiserum were used (data
not shown), instead of the ENOD8 oligopeptide antiserum.
ENOD8 antibodies were purified from the ENOD8 oligo-
peptide antiserum by immunoaffinity chromatography and were
used in immunolocalization experiments. To assess the utility
of the ENOD8 antibodies in localization, they were immuno-
depleted with ENOD8 oligopeptide. The ENOD8 antibodies
specifically label infected cells in the interior of the nodule,
but also apparently have nonspecific binding in the nodule
periphery (Supplemental Figure 1). The specific staining disap-
pears in ENOD8 antibodies preincubated with ENOD8 oligo-
peptide, whereas the nonspecific staining persists. Preimmune
antisera did not label nodules.
To determine which tissues within M. truncatula nodules
ENOD8 protein localizes, nodule sections were treated with
immunopurified ENOD8 antibodies followed by Cy5-labeled
secondary antibodies and then observed by confocal micros-
copy (Fig. 2A). Sections serial to the immunolabeled section
were stained with iodine to localize IZ II-III and observed in
the light microscope (Fig. 2B). Comparison of Figure 2A and
B show that ENOD8 protein is found in IZ II-III and in zone
III, the nitrogen-fixing zone. The protein distribution is similar
to the pattern of expression found for pMtENOD8-gusA trans-
gene expression. Except for the nonspecific nodule periphery
staining noted above, ENOD8 protein staining appears to be
confined to the infected cells, because the cells staining for
expression are large and the staining pattern is ring shaped,
consistent with infected cells. ENOD8 protein staining is more
intense at the proximal end of the nodule (Fig. 2A); this pat-
tern is consistent in all nodule sections observed.
Intracellular localization of ENOD8 protein.
Higher magnification confocal microscopy was used to lo-
calize ENOD8 protein at the cellular level. To control for anti-
body diffusion into the fixed nodule tissue and accessibility of
the antiserum to the nodule cells, a commercially available
control antibody to histone H1 was used. The assumption is
that the diffusion of the histone H1 antibodies and the ENOD8
antibodies should be similar. TRITC-conjugated secondary anti-
serum was used to detect the histone H1 antibody (Fig. 3, blue)
and Cy5-conjugated secondary antibody was used to detect the
406 / Molecular Plant-Microbe Interactions
immunopurified ENOD8 antibody (Fig. 3, red). SYTO13 was
used as a counterstain to show the distribution of the rhizobia
(Fig. 3, green) (Haynes et al. 2004). No ENOD8 label was
detected in uninfected cells (data not shown). In infected cells
of IZ II-III and zone III, ENOD8 label was correlated with
symbiosomes (Fig 3). No ENOD8 label was detected in infected
cells from the infection zone after bacteroid replication and
elongation (Fig. 3B). Only cells in IZ II-III and zone III label
with ENOD8 antibodies (Fig. 3C). From the merged images in
Figure 4C, the left-hand side of the infected cell has symbio-
somes that have their long axis almost perpendicular to the
focal plane (arrow). In these symbiosomes, the ENOD8 label
clearly can be seen to surround the bacteroid label. At higher
magnification, ENOD8 antibodies detected ENOD8 protein in
areas immediately surrounding the bacteroids, in the symbio-
some space or symbiosome membrane (Fig. 4).
We attempted to localize ENOD8 protein using immuno-
electron microscopy. However, under the harsh fixation condi-
tions required for electron microscopy (Sherrier et al. 2005),
ENOD8 could not be detected with either ENOD8 antiserum.
Immunodetection of ENOD8 protein
in fractionated nodules.
Confocal microscopy does not permit the immunolocalization
of ENOD8 protein within the symbiosome. Mass spectrometry
has indicated that ENOD8 protein may be a component of the
symbiosome membrane (Catalano et al. 2004). In order to fur-
ther provide the biochemical evidence for localization of
ENOD8 in the symbiosome, blots containing total protein from
the symbiosome membrane, symbiosome space, and bacteroid
fractions were probed with ENOD8 antiserum. Significantly
higher levels of ENOD8 protein were detected in the symbio-
some space relative to the symbiosome membrane (Fig. 5B).
Control blots were stained with Ponceau (Fig. 5A) and stained
for markers of the symbiosome membrane (Fig. 5C) and bacter-
oid fractions (Fig. 5D) using nodulin-26 and dinitrogenase re-
ductase antisera, respectively. Ponceau was found to stain the
hydrophobic proteins in the symbiosome membrane less effec-
tively than the more soluble proteins in the other fractions (Fig.
5A). Nodulin-26 was detected as an approximately 32-kDa pro-
tein in the symbiosome membrane fraction, consistent with the
predicted size of M. truncatula nodulin-26. Although nodulin-26
antisera showed significant cross reactions with the proteins in
the symbiosome space fractions, no protein corresponding to the
size of nodulin-26 was detected in that fraction suggesting the
purity of the symbiosome membrane fraction. A 31-kDa S.
meliloti-specific dinitrogenase reductase was detected in the
bacteroid fraction (Fig. 5D). However, a control immunoblot for
the symbiosome space fraction could not be performed due to
the nonavailability of specific antiserum. Thus, these experi-
ments suggest the localization of ENOD8 protein to the symbio-
some space, with a smaller amount localizing to the symbio-
some membrane. It also is possible that the ENOD8 protein in
the membrane fraction represents a small amount trapped inside
sealed vesicles of the symbiosome membrane preparation.
DISCUSSION
In this study, we report the distribution of pMtENOD8-gusA
transgene expression and the ENOD8 protein in symbiotic
nodules of M. truncatula, providing insight into the role of
ENOD8 in nodule development. ENOD8 proteins have esterase
activity, more active on methyl esters than on longer-chain
esters of model ester substrates (Pringle and Dickstein 2004).
Use of a pMtENOD8-gusA transgene in A. rhizogenes-trans-
formed roots of composite Medicago plants shows expression
in the parts of the nodule where cells infected with rhizobia
inside symbiosomes are found. However, not all developmental
stages of symbiosome-containing cells express the pMtENOD8-
gusA transgene. Expression was not found in zone II, which
contains dividing and differentiating rhizobia inside symbio-
some membranes (Vasse et al. 1990). Use of an immunopurified
antibody that was raised against an ENOD8-specific peptide
allowed visualization of the ENOD8 protein inside rhizobia-
infected cells at the tissue level and inside symbiosomes at the
Fig. 2. ENOD8 protein localizes to cells in interzone (IZ) II to III and zone
III. A, Nodule section immunostained with immunopurified ENOD8 anti-
bodies imaged by confocal microscopy. Cells that label specifically fo
r
ENOD8 have ring structures, indicative of infected cells. B, Section serial
to the section in A, stained with iodine to localize the IZ II-III. The double
arrow marks IZ II-III. Bar = 1 mm.
Fig. 1. Expression of pMtENOD8-gusA transgene in nodulated hairy root o
f
composite Medicago A17 plants transformed via Agrobacterium rhizogenes.
Roots were transformed with pMtENOD8-gusA and subsequently inoculate
d
with Sinorhizobium meliloti in an aeroponic system. At 25 days postinocu-
lation, the roots were fixed, sectioned, and stained with 5-bromo-4-chloro-
3-indolyl-β-D-glucuronide, cyclohexylammonium salt (blue), visualizing
the pMtENOD8-gusA expression in the proximal part of the nodule. The
double arrows mark the locations of zones II and III. Bar = 0.5 mm.
Vol. 21, No. 4, 2008 / 407
subcellular level. Previous proteomic experiments to identify
proteins found in the symbiosome membrane showed that
ENOD8 is found in this membrane (Catalano et al. 2004). In
studies reported here, differentially separated nodule fractions
probed on Western blots suggest that ENOD8 protein is found
in the symbiosome space as well.
This is the first report of a biochemical change in a plant
component of symbiosomes as they mature from having newly
deposited type 1 bacteroids inside the symbiosome membrane
to nitrogen-fixing forms with elongated type 4 bacteroids
(Vasse et al. 1990). ENOD8 protein is observed in developing
and mature symbiosomes. In symbiosomes at the distal end of
zone II, bacteroids already have elongated (Vasse et al. 1990)
and ENOD8 is not observed (Figs. 2 and 3B). At the distal end
IZ II-III, ENOD8 expression begins, persisting to the nitrogen-
fixing zone (Fig. 3C).
We previously reported that ENOD8 encoded a protein with
a signal sequence and no predicted transmembrane domains
(Dickstein et al. 1993, 2002). Based on similarities with ho-
mologous proteins that are secreted and associated with the
apoplast (Kandzia et al. 1998; van Engelen et al. 1995) and its
biochemical property of complete solubility in aqueous solu-
tions, we anticipated that ENOD8 protein also would be se-
creted and localize to the exterior of the plasma membrane
(Pringle and Dickstein 2004). Topologically, the symbiosome
is external to the cytosol; thus, the localization of ENOD8 pro-
tein to symbiosomes is consistent with previous predictions.
Other proteins that localize to symbiosomes also have a signal
sequence (Kardailsky et al. 1996; Liu et al. 2006), but no one
has yet determined whether consensus amino acid motifs exist
which might be recognized by host-targeting machinery to tar-
get plant host proteins there. Thus, there is as yet no way to
predict that ENOD8 encodes a symbiosome-localized protein.
Most ENOD8 protein was found to localize to the symbio-
some space, with a smaller amount localizing to the symbio-
some membrane, although it was impossible to rule out a small
amount of symbiosome space inside sealed vesicles contami-
nating the symbiosome membrane fractions. Because of its solu-
bility in aqueous solutions and lack of predicted transmembrane
domains, the fraction of ENOD8 protein that is apparently in
the symbiosome membrane is likely a peripheral membrane
protein, bound to another symbiosome membrane molecule
that is intrinsically attached to the membrane.
Because of its localization, ENOD8 protein may be integral
to symbiosome function. ENOD8 protein is a member of the
GDSL family of hydrolytic enzymes. In plants, GDSL member
proteins are involved in secondary metabolism. Although we
previously have shown ENOD8 protein to have esterase activity
Fig. 3. Distribution of ENOD8 in infected nodule cells. Staining with ENOD8 antisera (red) with histone antisera (blue) as an antibody accessibility control
and SYTO13 (green), a soluble nucleic-acid binding dye, shows positions of bacteroids. A, Stained nodule shows positions of individual cells investigated at
high magnification in B and C. The double arrow marks interzone (IZ) II to III. Bar = 50 μm. B, Infected cell from middle of IZ II-III shows elongated
bacteroids (green, upper left) and histone staining in nuclei (blue, lower left), but no ENOD8 staining (red, upper right) in merged image (lower right). Bar =
10 μm. C, Infected cell from nitrogen-fixing zone III. Elongated bacteroids (arrow) and an infection thread remnant (double arrow), and the nucleus
(arrowhead) stain with SYTO13 (green, upper left), histone staining in nuclei (blue, lower left), prominent ENOD8 staining (red, upper right) in the merged
images (lower left) show that ENOD8 labeling surrounds the bacteroids’ staining. At points in the cell where the symbiosomes are almost perpendicular to
the confocal plane, the ENOD8 staining (red) can be seen to surround the bacteroid staining (green) (arrow). Bar = 10 μm.
Fig. 4. Immunolocalization of ENOD8 in symbiosomes. High magnifica-
tion of symbiosome region in an infected cell. A, Staining with SYTO13
visualizes bacteroids. B, Staining with anti-ENOD8 antibodies. C, Merged
image shows ENOD8 protein surrounding the bacteroids. Bar = 10 μm.
408 / Molecular Plant-Microbe Interactions
against model substrates, especially shorter ester chains (Pringle
and Dickstein 2004), we still do not understand the nature of
its true substrates. Because of the previous observation that
ENOD8 expression persists after nitrate addition (Dickstein et
al. 2002), it is unlikely that ENOD8 protein is required for nitro-
gen fixation per se. ENOD8 protein has high similarity to LAE
and to the exopolygalacturonase EP4 (Bertinetti and Ugalde
1996; Tanaka et al. 2002; van Engelen et al. 1995), both of
which act on carbohydrate substrates; thus, we predict that
ENOD8 protein’s substrate is also an oligo- or polysaccharide.
iEP4 was found to be inducible by elicitors (Bertinetti and
Ugalde 1996), which raises the possibility that it and its
homologs, including ENOD8, have defensive roles. The find-
ing that ENOD8 protein is in the symbiosome space and, pos-
sibly, membrane opens the intriguing possibility that ENOD8’s
substrate could be a surface molecule of bacteroid origin.
Could ENOD8’s substrate be rhizobially produced Nod factor,
exopolysaccharide, or lipopolysaccharide? The latter is known
to change during bacteroid development (Kannenberg and
Carlson 2001). It is equally possible that ENOD8 acts on a
plant symbiosome-targeted substrate that controls bacteroid
differentiation, function, or proliferation or an aspect of sym-
biosome function.
MATERIALS AND METHODS
Plant growth.
Wild-type A17 and composite A17 plants with transformed
roots were grown in aeroponic chambers as previously described
(Veereshlingam et al. 2004), misted in a nitrogen-free nutrient
media (Lullien et al. 1987), and inoculated with S. meliloti 5
days after germination. Plants were maintained on a regime of
16 h of light and 8 h of dark at 22°C.
ENOD8.1 promoter-gusA fusion construct.
pRD022 was constructed by cloning the 3.2-kb SphI piece
of pCAMBIA2301 into pUC18. This 3.2-kb piece of pCAM-
BIA2301 contains lacZα, the Cauliflower mosaic virus 35S
promoter, and β-glucuronidase containing the catalase intron.
Six ENOD8 genes have been found to be on the fully se-
quenced BAC mth1-64n13 (GenBank accession AC139354),
of which one, ENOD8.1, has been found to be highly ex-
pressed in nodules (Dickstein et al. 2002). ENOD8.1 also is
found on the partially sequenced BAC mth1-19n23 that over-
laps BAC mth1-64n13. The ENOD8.1 promoter was obtained
with flanking SacI and NcoI restriction sites by polymerase
chain reaction (PCR) from BAC mth1-19n23 by PCR with the
following primers: SacI-ENOD8p: 5′-GATCGAGCTCCACC
GGACCTATTGACTAGC and ENOD8p-NcoI: 5′-GCATGCC
ATGGATTTCATGAAGCAATAAAGGAACC. These primers
amplify DNA from nucleotides 7,570 to 10,551 from GenBank
accession AF463407 (BAC mth1-19n23) that contains 2.98 kb
of DNA immediately upstream of the translational start codon
of ENOD8.1. The 2.98-kb PCR fragment was digested with
SacI and NcoI and ligated to similarly digested pRD022, creat-
ing a translational fusion between the ENOD8.1 upstream se-
quence and gusA. The PCR manipulations changed the sequence
at the translational start of ENOD8.1 from 5′-TTCATGG to 5′-
TCCATGG. The resulting plasmid then was digested with SacI
and BstEII. The 4.23-kb SacI-BstEII fragment containing the
ENOD8.1 promoter fused to gusA subsequently was ligated
into pCAMBIA2301, creating pRD027. pRD027 was trans-
formed into A. rhizogenes ARqua1 and used for hairy root
transformations (Boisson-Dernier et al. 2001). In addition to
nodule staining in the nodule as shown in the results, X-Gluc
staining in the root stele also was observed occasionally in
roots transformed with pMtENOD8-gusA transgene and the
control transgene, which may be a stress response to hairy root
transformation (D. Barker, personal communication).
Nodules containing the pMtENOD8-gusA transgene or
pMtENOD11-gusA control transgene were fixed in 4%
(vol/vol) formaldehyde in 0.1 M PIPES buffer, and stained
with X-Gluc (Gold Biotechnology, St. Louis) as described
(Journet et al. 1994).
MBP-ENOD8 protein fusion construct.
The nucleotides 100 to 1,158 sequence of the M. sativa
ENOD8 cDNA (GenBank L18899) was engineered by PCR
Fig. 5. Immunoblot analysis of ENOD8 protein in Medicago root nodule protein fractions. A, Ponceau-stained blot showing total protein (50 μg/lane) in the
symbiosome membrane (SM), symbiosome space (SS), and bacteroid (B) fractions. Blot probed with B, anti-ENOD8; C, anti-nodulin 26, a marker for the
symbiosome membrane; and D, anti-dinitrogenase reductase, a marker for the bacteroid fraction. Arrows indicate the relevant chemiluminescent signal in
panels B, C, and D.
Vol. 21, No. 4, 2008 / 409
with the ForpMAL primer (GCACCCCGGGACACATTG TG
ATTTTCCTGCC) to add a SmaI site at the 5′ end of the PCR
product and with the RevpMAL primer (GCACGTCGACTCA
CTTTCTATAACATGCCATATCTAG) to add a SalI site to the
3′ end of the PCR product. The PCR product was sequentially
digested with SmaI, then SalI, ligated into an Xmn1, SalI-
digested pMAL-p2X vector (New England Biolabs, Beverly,
MA, U.S.A.), and transformed in TOP10 competent cells
(Invitrogen, Carlsbad, CA, U.S.A.). Positive colonies were se-
quenced (Lone Star Labs, Inc., Houston) with custom primers
to identify the correct clone.
MBP-ENOD8 protein fusion purification.
An Escherichia coli strain carrying the MBP-ENOD8 con-
struct was grown at 37°C on a shaker in Luria-Bertani media
supplemented with streptomycin at 10 μg/ml and ampicillin at
100 μg/ml, until absorbance at 600 nm was 0.5. Isopropyl-β-D-
thiogalactoside was added to 0.03 mM to induce recombinant
expression overnight at room temperature (RT) with shaking.
The induced bacterial pellet was recovered by centrifugation
and resuspended in column buffer (20 mM Tris-HCl, pH 7.4;
200 mM NaCl; and 1 mM EDTA, pH 8.0), stored overnight at
–80°C, then thawed on ice water. Cells were lysed with a
French press. After centrifugation to collect cell debris, the su-
pernatant was incubated with an amylose bead slurry (New
England Biolabs) and washed several times with column
buffer. The MBP-ENOD8 protein was eluted from the amylose
beads with column buffer supplemented with 10 mM maltose.
After centrifugation, the supernatant was collected and con-
centrated by ultrafiltration with Ultrafree centrifugal filter
units (Millipore, Billerica, MA, U.S.A.). Concentration of the
purified recombinant MBP-ENOD8 protein was evaluated by
Bradford assay with Coomassie protein assay reagent (Pierce,
Rockford, IL, U.S.A.) and sodium dodecyl sulfate (SDS) poly-
acrylamide gel electrophoresis.
MBP-ENOD8 polyclonal antiserum.
Antisera were generated by immunizing rabbits with the
purified MBP-ENOD8 fusion protein (Bio-synthesis, Inc.,
Lewisville, TX, U.S.A.). Preimmune serum was collected from
each rabbit prior to immunization. Immunoblot analysis
showed that the MBP-ENOD8 antiserum had the same speci-
ficity toward the ENOD8 protein from nodule-soluble extract,
as the anti-ENOD8 oligopeptide antiserum previously tested
(Dickstein et al. 2002).
Affinity-purified anti-ENOD8 oligopeptide antibody.
The anti-ENOD8 oligopeptide antisera (Dickstein et al.
2002) was incubated in a SulfoLink affinity chromatography
column (Pierce), whose beads had been conjugated with the
ENOD8 peptide, CKNPSTRITWDGTHYTEAA (SynPep,
Dublin, CA, U.S.A.). The ENOD8 antibodies were eluted from
the column by 0.2 M glycine, pH 2.0. ENOD8 antibody frac-
tions were collected directly in microcentrifuge tubes contain-
ing 1 M Tris-HCl, pH 9.2. The activity of the affinity-purified
ENOD8 antibody fraction was validated by Western analysis.
Immunomicroscopy.
Mature pink nodules were collected from M. truncatula
grown in aeroponics chambers. Nodules were vacuum infiltrated
seven times, 30 s each with 0.5% (vol/vol) formaldehyde (Elec-
tron Microscopy Sciences, Hatfield, PA, U.S.A.) in phosphate-
buffered saline (PBS; 137 mM NaCl, 10 mM sodium phosphate,
and 2.7 mM KCl), pH 7.0. After 20 min at RT, nodules were
rinsed six times for 10 min in PBS, pH 7.0. Nodules were trans-
ferred to ice-cold 100% methanol and agitated overnight at 4°C.
Nodules were rinsed in PBS, pH 7.0, six times for 10 min each
and were embedded in 6% (wt/vol) low-melting-point agarose in
PBS, pH 7.0. Tissues were sectioned with a Vibratome 1000
classic (Vibratome, St. Louis) mounted with a single-edge razor
blade. Then, 50-μm sections were collected affixed on slides,
which had been spotted with a drop of poly-L-lysine (Sigma-
Aldrich, St. Louis) surrounded by a thin water-repellent film. All
incubations for immunostaining were carried in 50-μl volumes
at RT in humidity chambers. Sections were incubated for 1 h in
blocking buffer (3% [wt/vol] bovine serum albumin [BSA], 5%
[vol/vol] normal goat serum [NGS], 0.1% [vol/vol] Tween 20,
PBS pH 7.0), then incubated for 1 h with a 1:100 dilution of the
affinity-purified anti-ENOD8 oligopeptide antibody. In some
experiments, the anti-ENOD8 oligopeptide antibody was co-
incubated with a 1:100 dilution of anti-histone H1 antibody (Up-
state, Lake Placid, NY, U.S.A.). Sections were rinsed twice with
blocking buffer and subsequently incubated with 1:500 dilution
of secondary antibodies (Jackson ImmunoResearch Laborato-
ries, West Grove, PA, U.S.A.) for 1 h: tetramethyl rhodamine
isothiocyanate (TRITC)-conjugated goat anti-mouse (H+L),
Cy5-conjugated goat anti-rabbit (H+L). Sections were rinsed
twice with blocking buffer and stained for 15 min with 5 μM
SYTO13 (Invitrogen) in blocking buffer. Sections were washed
twice for 5 min in blocking buffer. Sections were covered by a
drop of anti-photobleaching media, ProLong Gold antifade re-
agent (Invitrogen) before laying a coverslip on them. Confocal
images were acquired with a Perkin-Elmer UltraVIEW ERS
spinning disk confocal microscope connected to a Zeiss Axio-
vert 200M (Carl Zeiss, Oberkochen, Germany), with 488 nm
(SYTO13), 568 nm (TRITC), and 640 nm (Cy5) laser lines. Ob-
servations were made using a ×20 (NA 0.75) or ×100 (NA 1.4)
objective lenses. Images were acquired as z-stacks using the
software developed by Perkin-Elmer provided with the Ultra-
view ERS. Data then was processed with ImageJ from NIH.
Preparation of symbiosome fractions and
immunoblot analysis.
Symbiosome membrane, symbiosome space, and bacteroid
protein fractions were purified as described previously (Catalano
et al. 2004). Protein concentrations were determined using the
Bio-Rad DC protein assay according to the manufacturer’s
instructions (Bio-Rad Laboratories, Hercules, CA, U.S.A.).
Symbiosome membrane, symbiosome space, and bacteroid
protein fractions were resolved on SDS gels for immunoblot
analyses as described (Catalano et al. 2004, 2007). Proteins were
transferred to 0.45-μm nitrocellulose membrane using a semidry
electrophoretic transfer cell from Bio-Rad according to manu-
facturer’s instructions (Bio-Rad Laboratories). The blot to be
probed with ENOD8 antisera was blocked in 10 ml of 1× Tris-
buffered saline (TBS), pH 7.4, 2% (wt/vol) BSA, 1% (wt/vol)
NGS, and 0.05% (vol/vol) Tween 20 at RT for 1 h. The blot then
was incubated in 10 ml of block containing 1:5000 ENOD8 an-
tisera for 2 h at RT followed by rinsing the blot five times in 10
ml of TBS-Tween (TBST) for 5 min. The blot was incubated in
10 ml of block containing 1:10,000 goat anti-rabbit immu-
noglobulin-G secondary antibody conjugated to horseradish per-
oxidase (Sigma-Aldrich) for 1 h at RT. Finally, the blot was
rinsed four times in 10 ml of TBST for 5 min and once in 10 ml
of TBS for 5 min. Protein signal was detected using enhanced
chemiluminescence. Nodulin-26 and dinitrogenase reductase
Western blots were performed as described previously (Catalano
et al. 2004; Guenther et al. 2003).
ACKNOWLEDGMENTS
We thank E. Blancaflor for suggestions on tissue fixation and antigen
recovery; K. Luby-Phelps for help with confocal microscopy and use of
the Live Cell Imaging Core at University of Texas Southwestern; K.
410 / Molecular Plant-Microbe Interactions
VandenBosch for help with attempts at immunoelectron microscopy; A.
Wang and J. Park for help with the pMtENOD8-gusA construct; and S.
Banker, B. Blackmon, and B. Huskinson for help characterizing the anti-
ENOD8 and anti-MBP-ENOD8 antisera. This project was partially sup-
ported by United States Department of Agriculture Cooperative State Re-
search, Education and Extension Service National Research Initiative Ni-
trogen Fixation Program grants 99-35305-8574, 99-35305-8693 to R.
Dickstein and Plant Biochemistry Program grant 2005-35318-16215 to D.
J. Sherrier.
LITERATURE CITED
Akoh, C. C., Lee, G. C., Liaw, Y. C., Huang, T. H., and Shaw, J. F. 2004.
GDSL family of serine esterases/lipases. Prog. Lipid Res. 43:534-
552.
Bertinetti, C., and Ugalde, R. A. 1996. Studies on the response of carrot
cells to a Sclerotinia sclerotiorum elicitor: induction of the expression
of an extracellular glycoprotein mRNA. Mol. Plant-Microbe Interact.
9:658-663.
Boisson-Dernier, A., Chabaud, M., Garcia, F., Becard, G., Rosenberg, C.,
and Barker, D. G. 2001. Agrobacterium rhizogenes-transformed roots of
Medicago truncatula for the study of nitrogen-fixing and endomycorrhi-
zal symbiotic associations. Mol. Plant-Microbe Interact. 14:695-700.
Brewin, N. J. 1991. Development of the legume root nodule. Annu. Rev.
Cell Biol. 7:191-226.
Brewin, N. J. 2004. Plant cell wall remodelling in the Rhizobium–legume
symbiosis. Crit. Rev. Plant Sci. 23:293-316.
Catalano, C., Lane, W. S., and Sherrier, D. J. 2004. Biochemical charac-
terization of symbiosome membrane proteins from Medicago trunca-
tula root nodules. Electrophoresis 25:519-531.
Catalano, C. M., Czymmek, K. J., Gann, J. G., and Sherrier, D. J. 2007.
Medicago truncatula syntaxin SYP132 defines the symbiosome mem-
brane and infection droplet membrane in root nodules. Planta. 225:541-
550.
Dahiya, P., Kardailsky, I. V., and Brewin, N. J. 1997. lmmunolocalization
of PsNLEC-1 I a lectin-like Glycoprotein expressed in developing pea
nodules. Physiol. Plant 115:1431-1442.
Dickstein, R., Prusty, R., Peng, T., Ngo, W., and Smith, M. E. 1993.
ENOD8, a novel early nodule-specific gene, is expressed in empty al-
falfa nodules. Mol. Plant-Microbe Interact. 6:715-721.
Dickstein, R., Hu, X., Yang, J., Ba, L., Coque, L., Kim, D.-J., Cook, D. R.,
Yeung, A. T. 2002. Differential expression of tandemly duplicated
Enod8 genes in Medicago. Plant Sci. 163:333-343.
Gage, D. J. 2004. Infection and invasion of roots by symbiotic, nitrogen-
fixing rhizobia during nodulation of temperate legumes. Microbiol.
Mol. Biol. Rev. 68:280-300.
Guenther, J. F., Chanmanivone, N., Galetovic, M. P., Wallace, I. S., Cobb,
J. A., and Roberts, D. M. 2003. Phosphorylation of soybean nodulin 26
on serine 262 enhances water permeability and is regulated develop-
mentally and by osmotic signals. Plant Cell 15:981-991.
Haynes, J. G., Cyzmmek, K. J., Carlson, C. A., Veereshlingam, H.,
Dickstein, R., and Sherrier, D. J. 2004. A novel method for rapid analy-
sis of legume root nodule development using confocal microscopy. New
Phytol. 163:661-668.
Hirsch, A. M. 1992. Developmental biology of legume nodulation. New
Phytol. 122:211-237.
Jones, K. M., Kobayashi, H., Davies, B. W., Taga, M. E., and Walker, G.
C. 2007. How rhizobial symbionts invade plants: the Sinorhizobium–
Medicago model. Nat. Rev. Microbiol. 5:619-633.
Journet, E.-P., Pichon, M., Dedieu, A., de Billy, F., Truchet, G., and Barker,
D. G. 1994. Rhizobium meliloti Nod factors elicit cell-specific transcrip-
tion of the ENOD12 gene in transgenic alfalfa. Plant J. 6:241-249.
Journet, E. P., El-Gachtouli, N., Vernoud, B., de Billy, F., Pichon, M.,
Dedieu, A., Arnould, C., Morandi, D., Barker, D. G., and Gianinazzi-
Pearson, V. 2001. Medicago truncatula ENOD11: a novel RPRP-encod-
ing early nodulin gene expressed during mycorrhization in arbuscule-
containing cells. Mo. Plant-Microbe Interact. 14:737-748.
Kandzia, R., Grimm, R., Eckerskorn, C., Lindemann, P., and Luckner, M.
1998. Purification and characterization of lanatoside 15′-O-acetyles-
terase from Digitalis lanata Ehrh. Planta 204:383-389.
Kannenberg, E. L., and Carlson, R. W. 2001. Lipid A and O-chain modifi-
cations cause Rhizobium lipopolysaccharides to become hydrophobic
during bacteroid development. Mol. Microbiol. 39:379-391.
Kardailsky, I. V., Sherrier, D. J., and Brewin, N. J. 1996. Identification of a
new pea gene, PsNlec1, encoding a lectin-like glycoprotein isolated
from the symbiosomes of root nodules. Plant Physiol. 111:49-60.
Liu, J., Miller, S. S., Graham, M., Bucciarelli, B., Catalano, C., Sherrier,
D. J., Samac, D. A., Ivashuta, S., Fedorova, M., Matsumoto, P., Gantt, J.
S., and Vance, C. P. 2006. Recruitment of novel calcium binding pro-
teins for root nodule symbiosis in Medicago truncatula. Plant Physiol.
141:167-177.
Lullien, V., Barker, D. G., de Lajudie, P., and Huguet, T. 1987. Plant gene
expression in effective and ineffective root nodules of alfalfa (Medicago
sativa). Plant Mol. Biol. 9:469-478.
Mylona, P., Pawlowski, K., and Bisseling, T. 1995. Symbiotic nitrogen
fixation. Plant Cell 7:869-885.
Panter, S., Thomson, R., de Bruxelles, G., Laver, D., Trevaskis, B., and
Udvardi, M. 2000. Identification with proteomics of novel proteins as-
sociated with the peribacteroid membrane of soybean root nodules.
Mol. Plant-Microbe Interact. 13:325-333.
Pringle, D., and Dickstein, R. 2004. Purification of ENOD8 proteins from
Medicago sativa root nodules and their characterization as esterases.
Plant Physiol. Biochem. 24:73-79.
Saalbach, G., Erik, P., and Wienkoop, S. 2002. Characterisation by pro-
teomics of peribacteroid space and peribacteroid membrane prepara-
tions from pea (Pisum sativum) symbiosomes. Proteomics 2:325-337.
Sherrier, D. J., Taylor, G. S., Silverstein, K. A. T., Gonzales, M. B., and
VandenBosch, K. 2005. Accumulation of extracellular proteins bearing
unique proline-rich motifs in intercellular spaces of the legume nodule
parenchyma. Protoplasma 225:43-55.
Tanaka, R., Ikeda, M., Funatsuki, K., Yukioka, H., Hashimoto, Y., Fujimoto,
S., Takata, M., Katoh, K., and Konno, H. 2002. Molecular cloning and
cytochemical analysis of exopolygalacturonase from carrot. Planta
215:735-744.
Udvardi, M. K., and Day, D. A. 1997. Metabolite transport across symbi-
otic membranes of legume nodules. Annu. Rev. Plant Physiol. Plant
Mol. Biol. 48:493-523.
Upton, C., and Buckley, J. T. 1995. A new family of lipolytic enzymes?
Trends Biochem. Sci. 20:178-179.
van Engelen, F. A., de Jong, A. J., Meijer, E. A., Kuil, C. W., Meyboom, J.
K., Dirkse, W. G., Booij, H., Hartog, M. V., Vandekerckhove, J., de
Vries, S. C., and van Kammen, A. 1995. Purification, immunological
characterization and cDNA cloning of a 47 kDa glycoprotein secreted
by carrot suspension cells. Plant Mol. Biol. 27:901-910.
Vasse, J., de Billy, F., Camut, S., and Truchet, G. 1990. Correlation be-
tween ultrastructural differentiation of bacteroids and nitrogen fixation
in alfalfa nodules. J. Bacteriol. 172:4295-4306.
Veereshlingam, H., Haynes, J. G., Sherrier, D. J., Penmetsa, R. V., Cook,
D. R., and Dickstein, R. 2004. nip, A symbiotic Medicago truncatula
mutant that forms root nodules with aberrant infection threads and plant
defense-like response. Plant Physiol. 136:3692-3702.
Vincent, J. L., and Brewin, N. J. 2000. Immunolocalization of a cysteine
protease in vacuoles, vesicles, and symbiosomes of pea nodule cells.
Plant Physiol. 123:521-530.
Wienkoop, S., and Saalbach, G. 2003. Proteome analysis. Novel proteins
identified at the peribacteroid membrane from Lotus japonicus root
nodules. Plant Physiol. 131:1080-1090.
