Control of Rhizobia Endosymbiosis by Coupling ER Expansion with Enhanced UPR

Abstract

Legumes establish symbiosis with rhizobia by forming a symbiotic interface that enables cross‐kingdom exchanges of signaling molecules and nutrients. However, how host organelles interact with symbiosomes at the symbiotic interface remains elusive during rhizobia endosymbiosis. Here, symbiotic cells are reconstructed using 3D scanning electron microscopy (SEM) and uncover that the host endoplasmic reticulum (ER) undergoes dynamic expansion to gradually enwrap symbiosomes, facilitating their compartmentalization and endosymbiosis. Consistently, altering ER lamellar expansion by overexpressing MtRTNLBs, the reticulons responsible for ER tubulation, impairs rhizobia accommodation and symbiosome development. Intriguingly, unfolded protein response (UPR)‐marker genes, bZIP60 and IRE1A/B, show continuously activated expression during nodule development, and the two UPR‐deficient mutants, ire1b, and bzip60, exhibit compromised ER biogenesis and defective symbiosome development. Collectively, the findings underpin ER expansion and UPR activation as two key events in rhizobia accommodation and reveal an intrinsic coupling of ER morphology with proper UPR during root nodule symbiosis.

Full text

RESEARCH ARTICLE
www.advancedscience.com
Control of Rhizobia Endosymbiosis by Coupling ER
Expansion with Enhanced UPR
Jing Ren, Qi Wang, Xiaxia Zhang, Yongheng Cao, JingXia Wu, Juan Tian, Yanjun Yu,
Qingqiu Gong, and Zhaosheng Kong*
Legumes establish symbiosis with rhizobia by forming a symbiotic interface
that enables cross-kingdom exchanges of signaling molecules and nutrients.
However, how host organelles interact with symbiosomes at the symbiotic
interface remains elusive during rhizobia endosymbiosis. Here, symbiotic
cells are reconstructed using 3D scanning electron microscopy (SEM) and
uncover that the host endoplasmic reticulum (ER) undergoes dynamic
expansion to gradually enwrap symbiosomes, facilitating their
compartmentalization and endosymbiosis. Consistently, altering ER lamellar
expansion by overexpressing MtRTNLBs, the reticulons responsible for ER
tubulation, impairs rhizobia accommodation and symbiosome development.
Intriguingly, unfolded protein response (UPR)-marker genes, bZIP60 and
IRE1A/B, show continuously activated expression during nodule
development, and the two UPR-deficient mutants, ire1b,andbzip60, exhibit
compromised ER biogenesis and defective symbiosome development.
Collectively, the findings underpin ER expansion and UPR activation as two
key events in rhizobia accommodation and reveal an intrinsic coupling of ER
morphology with proper UPR during root nodule symbiosis.
1. Introduction
Legumes (Fabaceae or Leguminosae) represent the third-largest
family of flowering plants. It is divided into six subfamilies.
Among these, Papilionoideae and the Caesalpinioideae, contain
species that enter into a root-nodulating N2-fixing symbiotic re-
lationship with endosymbiotic rhizobia. The vast majority of the
J. Ren, Q. Wang, X. Zhang, Y. Cao, J. Wu, J. Tian, Y. Yu, Z. Kong
State Key Laboratory of Plant Genomics
Institute of Microbiology
Chinese Academy of Sciences
Beijing 100101, China
E-mail: zskong@im.ac.cn
J. Ren, Y. Cao, J. Wu, Z. Kong
University of Chinese Academy of Sciences
Beijing 100049, China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/advs.202414519
© 2025 The Author(s). Advanced Science published by Wiley-VCH
GmbH. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
DOI: 10.1002/advs.202414519
Papilionoideae subfamily is known to be
nodulated, but the Caesalpinioideae is
largely non-nodulated with the exception of
the tribe Mimosae (formerly the subfamily
Mimosoideae or the Mimosoid clade) which
is mostly nodulated.[1–4]Within legume
root nodules, rhizobia differentiate into bac-
teroids that fix atmospheric nitrogen (N2)
into ammonia (NH3), which is then trans-
ferred to the host plant. Bacteroids ex-
hibit distinct differentiation patterns: in
some Papilionoid legumes within the In-
verted Repeat–Lacking Clade (IRLC), such
as Medicago spp., they become swollen
and terminally differentiated, whereas in
legumes of the Phaseolid clade, such as
soybean (Glycine max L.), bacteroids re-
main non-swollen and retain the ability to
regenerate outside the nodule.[5]Among
a variety of plant-associated microorgan-
isms, symbiotic rhizobia are striking ex-
amples of bacteria that are successfully
taken up into host cells and subsequently
accommodated in the membrane-enclosed compartments within
symbiotic cells.[2–4]In the past twenty years, over 200 genes have
been reported to be involved in symbiotic nitrogen fixation us-
ing either forward or reverse genetics approaches,[6–8]which have
elucidated the main signaling pathways in regulating symbi-
otic nitrogen fixation. However, the mechanisms underlying the
uptake of rhizobia into living host plant cells and subsequent
Q. Wang, Z. Kong
Houji Laboratory in Shanxi Province
Academy of Agronomy
Shanxi Agricultural University
Taiyuan 030031, China
Q. Wang
Department of Plant Microbe Interactions
Max Planck Institute for Plant Breeding Research
50829 Cologne, Germany
Q. Gong
State Key Laboratory of Microbial Metabolism & Joint International
Research Laboratory of Metabolic and Developmental Sciences
School of Life Sciences and Biotechnology
Shanghai Jiao Tong University
Shanghai 200240, China
Adv. Sci. 2025,12, 2414519 2414519 (1 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
accommodation within the plant-derived membrane system in
symbiotic cells remain largely unknown, even though this pro-
cess is a prerequisite for root nodule symbiosis.[7]
In symbiotic root nodules of the IRLC clade (and in many
other studied legumes, but not all), rhizobia are entrapped into
host cells via the specialized transcellular apoplastic compart-
ments known as infection threads (ITs).[9,10]Subsequently, the
rhizobia are taken up into symbiotic cells by an endocytosis-like
process from ITs and grow in the organelle-like vesicles called
symbiosomes.[4]Along with rhizobia differentiation, symbiotic
cells also undergo dramatic endoreduplication, ultimately accom-
modating thousands of symbiosomes to support robust nitrogen
fixation.[11]During symbiosome development, the symbiosome
membrane undergoes massive expansion and dynamic organi-
zation, forming the symbiotic interface to achieve the accommo-
dation of symbiotic rhizobia that efficiently fix nitrogen.[11,12]As
early as the 1970s, electron microscopy studies have observed
the presence of intracellular organelles, including the endoplas-
mic reticulum (ER), in symbiotic cells. Previous studies have
shown that the ER participates in the formation of cytoplas-
mic bridges surrounding the IT tip to prompt progressive IT
development.[2,9,13]Among numerous organelles, the ER rep-
resents a multifunctional organelle harboring a wide range of
structural, biosynthetic, and metabolic functions and exhibits an
elaborate and flexible membrane web-like structure spreading
throughout the cytoplasm.[14]However, how the host ER system
contributes to symbiotic interface formation during rhizobia en-
dosymbiosis remains to be uncovered.[15]
Moreover, another important function of the ER is to control
the quality of proteins: only properly folded proteins are packaged
into ER-exit vesicles and allowed to move onward to their destina-
tion sites.[16]In contrast, misfolded proteins are degraded by the
ER-associated degradation (ERAD) system.[17]When a mismatch
occurs between the load of unfolded or misfolded proteins in the
ER and the capacity of the cellular machinery, it causes ER stress
and further triggers the unfolded protein response (UPR).[18,19]
The UPR is a cytoprotective response that senses an insufficiency
in the ER’s protein-folding capacity and restores cellular home-
ostasis following physiological stress exerted on the ER. The out-
comes of UPR include inhibition of mRNA translation, an in-
crease in ER protein folding capacity, and enhanced ERAD, which
are coordinated with ER membrane biogenesis to reduce the load
of misfolding-prone proteins.[20,21]However, although UPR plays
a critical role in responding to cellular biosynthetic demands, its
link to nodule symbiosis has yet to be explored. In particular, the
interplay between ER morphology and UPR during nodule sym-
biosis is poorly understood.
In this study, we revealed that the ER forms a niche to com-
partmentalize and nourish symbiosomes in rhizobia-infected
cells in Medicago truncatula nodules. Further genetic valida-
tion confirmed that MtRTNL-overexpression lines show reduced
ER lamellar expansion and impaired symbiosome development.
Strikingly, genes controlling UPR are continuously induced dur-
ing nodule development. As expected, symbiosome development
was found to be impaired in UPR-deficient mutants, which ex-
hibit altered ER morphology. Our findings, for the first time, un-
cover that ER morphological changes coordinate with enhanced
UPR to achieve rhizobia accommodation and symbiosome devel-
opment during root nodule symbiosis.
2. Results
2.1. ER Dynamic Expansion and Compartmentalization Regulate
Symbiosome Accommodation and Development
To thoroughly explore the engagement of the host ER system in
root nodule symbiosis, we performed large-scale live-cell imag-
ing and observed that the ER (pseudo-colored in red) network
forms reticulate structures surrounding symbiosomes (pseudo-
colored in green) in rhizobia-infected cells in M. truncatula nod-
ules (Figure S1a, Supporting Information). We next employed
scanning electron microscopy (SEM) for detailed examination.
In these symbiotic cells, ER spreads extensively into the cytosol
as a network of sheets and tubules accompanied by other mem-
brane organelles (Figure S1b, Supporting Information). We then
took advantage of the AutoCUTS-SEM (Automatic collector of ul-
trathin sections-SEM) technique and generated 3D-SEM recon-
struction images to show detailed ER information in a spatial
manner.[22–24]Interestingly, central flat ER sheets were observed
to transform into cortical tubular structures to capture rhizobia
(Figure S2a and Movie S1, Supporting Information). Moreover,
ER tubules distributed in the periphery of symbiosomes show
lumen width broadening compared to initial ER lamellar sheets
(Figure S2b–e, Supporting Information).
Next, we examined the morphological changes of ER along the
nodule zones at different stages of symbiosome development.
[25]As shown in Figure 1a, the developing symbiosomes emerge
randomly and entangle with ER tubules from the initial infec-
tion zone (zone II) to the transition zone (zone II-III). In con-
trast, in the nitrogen-fixing zone (zone III), ER tubules are fully
expanded throughout the host cell and orient nearly parallel to
mature symbiosomes that radiate regularly around a large cen-
tral vacuole. Moreover, symbiosomes interact frequently with the
ER through direct contact with each other throughout the sym-
biosome developmental stages (Figure 1a–c). In addition to the
distribution, we also examined the morphological changes of ER
by measuring the lumen width, and the width of ER surround-
ing the symbiosomes narrows gradually from zone II to zone III
(Figure 1d). We also selected rhizobia-infected cells with ultra-
thin high-quality sections in each nodule zone (Movies S3–S5,
Supporting Information) using the AutoCUTS-SEM technique
and generated 3D-SEM images to show the reconstructions of
symbiosomes and ER. It turns out that ER tubules with high
curvature form cage-like structures that arrest symbiosomes in
zone II (Figure 1E;MovieS6, Supporting Information). In con-
trast, from zone II-III (Figure 1f;MovieS7, Supporting Informa-
tion) to zone III (Figure 1g;MovieS8, Supporting Information),
symbiosomes are gradually enclosed by lamellar ER structures,
characterized by connected flattening structures with lower cur-
vature continuity on the Z-axis of continuous slices. In addition,
the volume and surface area of the ER surrounding symbiosomes
increase with symbiosome enlargement from zone II to zone III
(Figure 1h,i). Interestingly, the contact frequency between the ER
and symbiosomes obviously increases alone with ER expansion
(Figure 1j–m).
These subcellular features clearly demonstrate that the ER
forms a cradle to support and coordinate the spatial distribu-
tion of symbiosomes in infected cells. Additionally, the ER ex-
hibits a significant increase in surface area, volume, and contact
Adv. Sci. 2025,12, 2414519 2414519 (2 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Figure 1. Dynamic expansion of ER surrounding the symbiosomes in M.truncatula nodules. a–c) Representative SEM images of 3-week-old nodules in
zone II (a), zone II-III (b), and zone III (c). The red arrowheads indicate ER; s, symbiosome; m, mitochondrion; g, Golgi apparatus. Scale bars =1μm.
d) Comparison of the ER width with 2D images in different rhizobial zones (II, n=150; II-III, n=118; III, n=182). e–g) 3D reconstruction derived from
a 50-μm-thick section showing that ER (blue) forms a cage-like structure to enclose symbiosomes (magenta) in zone II (e), zone II-III (f), and zone III
(g). The vertically arranged ones formed a group, showing the interaction between ER and symbiosomes from two different perspectives. Scale bars =1
μm. At least 10 cells were examined, and 2 cells were reconstructed in each zone with the same results. h), Quantification and comparison of the surface
area and the volume of 3D symbiosomes in different nodule zones (II, n=10; II-III, n=6; III, n=10). i, Quantification and comparison of the surface
area and the volume of the 3D ER system surrounding the individual symbiosome in different nodule zones (symbiosomes: II, n=3; II-III, n=3; III, n
=3). j–l) Representative partial enlarged display of the e-g. Scale bars =0.3 μm. m), Quantification and comparison of the ER contact frequency and
area of each symbiosome in different nodule zones (symbiosomes: II, n=3; II-III, n=3; III, n=3). All data are shown as mean ±standard deviation
(SD). The statistical analysis was performed using the two-way analysis of variance (ANOVA) and post-hoc comparisons. The letters indicate values with
statistically significant (p<0.05) and non-significant (p>0.05) differences, respectively.
Adv. Sci. 2025,12, 2414519 2414519 (3 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
sites between the ER and symbiosomes, with sustained morpho-
logical stretching and expansion of the lamella-tubule-lamella
shift along with symbiosome accommodation and development.
These results indicate that dynamic ER expansion plays a key
role in symbiotic interface formation during symbiosome devel-
opment.
2.2. Disturbance of ER Morphology Affects Symbiosome
Development and Accommodation
To further validate that ER expansion is a prerequisite to sym-
biosome development, we took a genetic approach and ec-
topically expressed reticulons, which are called reticulon-like
proteins (RTNLs) in M. truncatula.[26]RTNLs are membrane-
bending proteins that shape the ER membrane into tubules,
and overexpression of RTNLs has been shown to prevent
ER membrane expansion and sheet formation.[27–29]We con -
structed constitutive expression vectors ProLjUb::MtRTNLB4-1
and ProLjUb::MtRTNLB4-2, which were then introduced into
M.truncatula ecotype A17 through hairy root transformation
(Figure S3a,b, Supporting Information).[30]We observed that
both MtRTNLB4-1-andMtRTNLB4-2-overexpressing nodules
exhibit developmental decay with smaller sizes (Figure 2a,b;
Figure S3c–f, Supporting Information). Semi-thin sections op-
tical observation further revealed abnormally enlarged ITs and
deficient nodule development, and staining of infected cells de-
tect a low level of accommodation and developmental retarda-
tion of symbiosomes (Figure 2c,d). Subsequent SEM observation
shows that the ER in infected cells generates longer unbranched
tubules with local bundle aggregation (Figure 2e–h). Meanwhile,
the symbiosomes around the unbranched ER tubules appear dis-
ordered, and their development is arrested (Figure 2e–i). Further
3D reconstruction results demonstrated that highly curved ER
undergoes local aggregation without lamella expansion [20,31–33]
aggregation attenuates the interaction between the ER and sym-
biosomes (Figure 2j; Movies S9 and S10, Supporting Informa-
tion). Taken together, these findings indicated that excessive ER
tubulation impairs rhizobia accommodation and symbiosome
development, further supporting the essential role of ER mor-
phology transformation in symbiosis.
2.3. Stimulation of UPR Upon Rhizobia Infection and
Colonization
The intensive increase in surface area and active morphology
changes of the ER during symbiosis development reflect an effec-
tive regulation strategy behind it.[18]Given the large protein syn-
thesis requirements in the legume-rhizobia symbiosis, we spec-
ulated that UPR plays a key role and active ER expansion meets
the demand of robust UPR for successful symbiosis.[20,31–33]
To investigate the link between UPR and symbiosis, we first
analyzed the expression profiles of two UPR marker genes.[34–36]
One is bZIP60, which is spliced by Inositol-requiring enzyme
type 1 (IRE1) into a spliced version of bZIP60 (sbZIP60) at the
onset of ER stress, and the other is BIP3, which is one of the
target genes of the sbZIP60 transcription factor and encodes a
chaperone.[37,38]Quantitative real-time PCR (qRT-PCR) analyses
showed that the abundance of sbZIP60 mRNA and BIP3 mRNA
is significantly elevated compared to the control, indicating that
rhizobia infection stimulates UPR in root cells (Figure 3a,b). To
investigate whether the enhanced expression of these two genes
is relevant to the subsequent nodule development, we compared
the expression levels of the two genes in different organs. As
shown in Figure 3c, the expression of the two UPR marker genes
is significantly higher in nodules than in the other organs, con-
firming the association of UPR with the legume-rhizobia sym-
biosis.
We further examined the tissue-specific expression pattern
of three essential genes involved in UPR, including MtIRE1A,
MtIRE1B,andMtbZIP60 in various zones of the nodules
(Figure 3d–f) using the dataset of a laser-capture microdissection
study together with RNA sequencing.[39]We found that the tran-
scripts of three genes tend to increase from zone II (IId, p) to zone
III (ZIII) during the development and maturity of bacteroids for
nitrogen fixation.
To further explore the expression patterns of the three genes
in living cells, we detected the expression of the GUS reporter
in nodule cells driven by the promoters of MtIRE1A,MtIRE1B,
and MtbZIP60 genes, respectively. The results show that ex-
pression of these three genes is specifically induced in infected
cells (Figure 3g–i). Furthermore, we constructed Pro35S::RFP-
MtbZIP60 to visualize its subcellular localization by fluorescence
imaging of living cells, fluorescent signals were observed both at
the ER and in the nucleus of infected cells (Figure S4d, Support-
ing Information), indicating that MtbZIP60 mRNAs are spliced
in infected cells and that the resulting truncated proteins are
translocated into the nucleus.[40]Collectively, these results verify
the inherent UPR stimulation in the nodule cells from an intra-
cellular perspective.
2.4. Symbiosome Development is Retarded in UPR-Deficient
Mutants
To assess the effect of UPR on the regulation of ER biogenesis
in rhizobia-infected cells and further symbiotic development, us-
ing CRISPR–Cas9-mediated gene-editing technology, we gener-
ated stable transgenic Medicago plants of ire1a-, ire1b-andbzip60-
knockout mutants, respectively (Figure S5a–c, Supporting Infor-
mation). Further phenotypic analysis revealed that, under non-
symbiotic conditions, those mutants do not show obvious growth
defects compared to the wild-type plants (Figure S5d–i, Support-
ing Information), consistent with the observations in previous
Arabidopsis studies.[41]We next evaluated their sensitivity to ER
stress by treating the seedlings with tunicamycin (TM), which has
been widely used to induce ER stress.[42]Notably, upon TM treat-
ment, the growth and development of the mutants are affected
compared to the wild-type plants, confirming that the knockout
of these key players in UPR decreases the ability of mutants to
deal with ER stress (Figure S6a–d, Supporting Information).
Importantly, these mutants provide valuable materials to in-
vestigate whether the UPR pathway regulates nodule symbio-
sis. We first analyzed early infection events, including the for-
mation of infection foci, infection threads (ITs), and nodule pri-
mordia. Quantitative analysis showed that these UPR-deficient
mutants do not exhibit obvious defects during the early stages of
Adv. Sci. 2025,12, 2414519 2414519 (4 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Figure 2. Disturbance of ER dynamics influences nodule development. Plasmids of ProLjUb::MtRTNLB4-1 and empty vector as control were introduced
into Medicago A17 plants by hairy-root transformation, respectively. GFP expressed from the construct was used as a selection marker for transformants
and positive nodules were selected at 21days post-inoculation (dpi). a–h) the nodule phenotype observation in the empty vector group (a, c, e, g) and the
MtRTNLB4-1 overexpression group (b, d, f, h), including nodule growth (a, b) nodule resin section (c, d) SEM image of the mature infected cell (e, f ), the
symbiosome and ER distribution by local enlargement of E, F (g, h). The red arrowheads indicate ER. i) Quantification and comparison of symbiosomes
width between the MtRTNLB4-1 overexpression group and the control group (MtRTNLB4-1 overexpression, n=40; Empty vector, n=40). All data are
shown as mean ±SD. The statistical analysis was performed using the unpaired two-sided Student’s t-test, ****p<0.0001. j) 3D reconstruction of the
interaction and distribution of symbiosomes (magenta) and ER (blue) in nodule-infected cells of MtRTNLB4-1 overexpression plants. Scale bars =1mm
(a, b), 0.1 mm (c, d), 10 μm(e,f),0.5μm(g,h),1μm (left) and 0.8 μm (right) (j). At least two biological replicates were performed for each genotype.
infection (Figure S7a–f, Supporting Information). Next, we as-
sessed nodule development in these UPR mutants and found
that they formed smaller nodules compared to the wild type
(Figure 4a–d). The nodules of UPR mutants had a significantly re-
duced area proportion of the mature nitrogen fixation zone (zone
III) relative to the wild type (Figure 4e–f). Consistently, although
the nodule number in mutants is similar to that of the wild type
(Figure 4g), the nitrogenase activity in the mutant nodules is
dramatically lower than in the wild type with an equal number
(Figure 4h). Collectively, these results indicate that the develop-
ment of symbiosomes is retarded in the UPR-deficient mutants.
It is worth mentioning that the ire1a mutant does not ex-
hibit clear symbiotic development defects comparable to ire1b
and bzip60 (Figure S8a–g, Supporting Information). Com-
pared with the results of the nodule single-cell sequencing
database, the expression level of the IRE1A gene is signifi-
cantly lower than that of the IRE1B gene (Figure S8h, Support-
ing Information).[43]This is further confirmed by the expres-
sion analysis by using different tissues of M. truncatula (Figure
S8i, Supporting Information). Thus, IRE1B may play a more
critical role in UPR regulation of the symbiotic process than
IRE1A.
Adv. Sci. 2025,12, 2414519 2414519 (5 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Figure 3. UPR induction during legume-rhizobia symbiosis. a,b) Fluorescent quantitative PCR detection of spliced MtbZIP60 (sMtbZIP60) (a) and
MtBIP3 mRNA (b) in early symbiosis (0, 24, 48h post-inoculation, pi). c) Relative expression levels of sMtbZIP60 and MtBIP3 genes in multiple organs
of the same plant. d–f) Relative expression levels of MtIRE1A (d), MtIRE1B (e), and MtbZIP60 (f ) along the symbiotic process are represented by distinct
nodule sections. Zone I (meristematic), zone IId (distal) and zone IIp (proximal) constitute zone II (infection and differentiation zone). The data were
obtained from the Symbimics website, curated from a previous publication,[38], and represent the means of three technical replicates. Individual data
points are not available. g-i) Nodules at 14 dpi were dissected from M. truncatula transgenic lines expressing the 𝛽-glucuronidase (GUS) reporter under
the control of promoters of MtIRE1A (g), MtIRE1B (h) and MtbZIP60 (i) respectively and stained with GUS solution for semi-section. Ruthenium red
staining was performed for imaging. Scale bars =0.1 mm. For (a–c), all data are shown as mean ±SD. n=3, ****p<0.0001 versus mock sample
(a, b) and other organs (c). The statistical analysis was performed using the unpaired two-sided Student’s t-test in (a, b) and ANOVA with post-hoc
comparisons in (c-f). The letters indicate values with statistically significant (p<0.05) and non-significant (p>0.05) differences, respectively.
2.5. ER Expansion Couples with Enhanced UPR to Achieve
Rhizobia Accommodation and Symbiosome Development
Effective UPR alleviates ER stress by enhancing ER membrane
expansion to provide more ER surface and luminal space. To in-
vestigate whether this holds true in our case, we examined the
ER alterations in the two UPR-deficient mutants to determine if
the ER membrane expansion of infected cells is also affected. No-
tably, the ER lumen width of infected cells narrows significantly
within the entire three zones (as mentioned earlier) in UPR mu-
tant nodules, such as ire1b-1 and bzip60-1, compared to the WT
(Figure 5a–f). These results indicate that functional deficiency of
UPR leads to reduced ER membrane expansion and defective
symbiosome development, and UPR is indeed involved in the
Adv. Sci. 2025,12, 2414519 2414519 (6 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Figure 4. Deficiency in UPR function affects nodule symbiosis. a) Representative phenotypes in plant growth, from left to right: WT, ire1b-1,ire1b-2,
bzip60-1,bzip60-2. b) Statistical analysis of the fresh weight of the above plants (WT n=20, ire1b-1 n =20, ire1b-2 n =20, bzip60-1 n =19, bzip60-2
n=19). c) Representative phenotypes in nodules growth at 14 dpi, from left to right: WT, ire1b-1,ire1b-2,bzip60-1,bzip60-2.Scalebars=2 mm. d)
Quantitative statistical analysis on nodule length of the above plants (WT n=36, ire1b-1 n =45, ire1b-2 n =40, bzip60-1 n =30, bzip60-2 n =33). e)
Adv. Sci. 2025,12, 2414519 2414519 (7 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
regulation of ER morphology and symbiosome accommodation
during symbiotic development.
To further validate the importance of ER morphology changes
resulting from enhanced host UPR for symbiosomes develop-
ment, we first examined the differentiation of bacteroid and
found no severe developmental defects, such as early bacteroid
senescence and degradation (Figures 4e and 5a–c). We next an-
alyzed the expression levels of nifD,nifK,andnifH,whichen-
code nitrogenases of N2fixation in rhizobia.[44]The qPCR re-
sults show that the expression levels of nifD,nifK,andnifH de-
crease significantly in the above mutants compared to the wild
type (Figure 5g–i). Together, these results confirm that UPR func-
tional deficiency impairs nodule development and nitrogen fixa-
tion.
Since RTNLB4-1 overexpression results in abnormal expan-
sion of the ER membrane (see Figure 2j), we asked whether such
a phenotype is associated with UPR deficiency. Indeed, the ex-
pression of both the two UPR markers (MtbZIP60 and MtBIP3)
and the downstream CALNEXIN genes (MtCNX1 and MtCNX2)
and PROTEIN DISULFIDE ISOMERASE gene (MtPDI)[45]ap-
peared to be down-regulated in the transgenic plants (Figure S9,
Supporting Information). Collectively, these results indicate an
immediately reciprocal feedback regulation between the abnor-
mal ER membrane expansion and the weakened UPR signaling.
Taken together, we propose that the host ER surrounding the
symbiosomes undergoes dynamic expansion, and the compart-
mentalization of the ER controls the rhizobia intracellular colo-
nization. Importantly, ER expansion couples with enhanced UPR
to achieve rhizobia accommodation and symbiosome develop-
ment during the legume-rhizobia symbiosis. Thus, function de-
ficiency of UPR in ire1b and bzip60 mutants causes defective ER
structures and impaired symbiosome development (Figure 6).
3. Discussion
Symbiosis is a reciprocally beneficial process for the plant and
bacterial partners. Understanding the cellular and molecular
mechanisms involved in the accommodation of symbiosomes
is a huge challenge to study the endosymbiosis process during
legume symbiosis. Our study unveiled two important features of
host ER-related functions in the process of symbiosis. One aspect
is the dynamic expansion and remodeling of the host ER along
with the symbiosome accommodation, and the other is the stim-
ulation of UPR in the infected cells, which is required to shape
nodule symbiosis throughout the stages of symbiosome develop-
ment.
Cells constantly adjust the sizes and shapes of their organelles
according to physiological requirements.[46]It has been well doc-
umented that the ER can connect directly with other organelles
and undergo size and shape changes in the complicated mem-
brane network to regulate different cellular activities in eukary-
otic cells.[14,47,48]In our study, we observed that the ER changes its
size and shape spatiotemporally along with the development of
symbiosis. At different stages of symbiosome development, ER
increases its surface area and volume coordinately with the ex-
panding of symbiosomes; in multiple zones of nodule, the dy-
namic arrangement of the ER is well matched with the size en-
largement of symbiosomes; and ER morphology is transformed
from tubules to lamellae with the membrane expansion to adapt
the growth and development of symbiosomes in the infected
cells. Consistently, the disturbance of ER morphology through
RTNLBs overexpression leads to disorganization and defects in
symbiosome development. Thus, our results provide a solid line
of evidence showing the dynamic changes of ER surface area and
the morphology patterns in coupled with the process of symbio-
some accommodation and development. Thus, our results pro-
vide a solid line of evidence showing that the dynamic changes
of ER surface area and the changes in patterns of morphology are
coupled with the process of symbiosome accommodation and de-
velopment.
It has been reported that perturbation of ER homeostasis
causes ER stress, which, in turn, activates UPR and enhances pro-
tein and lipid synthesis to expand ER membranes and volume in
yeast and mammalian cells.[49,31,50]Based on the observation of
enhanced biosynthesis of ER surface and volume along with the
process of symbiosis, we predicted that the intense production
of membrane components could result in unfolded protein accu-
mulation, and the ER processing capacity might become insuffi-
cient for the requirements of protein folding under this circum-
stance, thus leading to the occurrence of ER stress. As a result,
UPR and related reprogramming events were triggered to alle-
viate ER stress and maintain ER homeostasis. Indeed, through
cellular, molecular, and genetic analyses, we obtained convinc-
ing results showing that high-intensity execution of UPR exists
in nodules, where rhizobia undergo intracellular accommoda-
tion and colonization. Consistently, in mutants with knock out
of UPR-related genes such as ire1b and bzip60, ER membrane
biogenesis capacity decreased significantly compared to the wild
type, associated with the formation of smaller mature zone III,
which established a communication hub between the necessity
of the ER membrane biogenesis mediated by UPR and nodule
development. In addition, we tested the effect of UPR at the late
stage of symbiosis by examining the decreased expression level
of nifD, nifK,andnifH genes in the nodules of relevant mu-
tants, which further indicated that UPR could mediate balance
beneficial for the survival and function persistence of bacteroids.
Taken together, these data demonstrated that maintaining dy-
namic morphological transformation of the ER is a requirement
for nodule development. By means of ER membrane expansion,
larger space could be supplied to hold more misfolded proteins.
Semi-thin sections of nodules at 14 dpi were stained with 0.4% toluidine blue. Scale bars =0.1 mm. f) Quantitative statistical results of area proportion
in zone III. (WT, n=24; ire1b-1,n=17; ire1b-2,n=24; bzip60-1,n=20; bzip60-2,n=20). g) Statistical result of pink nodules number at 14 dpi. n=
12. h) Acetylene reduction assay (ARA) revealed significantly decreased nitrogen-fixing activities of ire1b-1,ire1b-2,bzip60-1,andbzip60-2 nodules at 20
dpi. Each dot represents the mean value from 55 nodules (WT, n=3; ire1b-1, n =3; ire1b-2, n =3; bzip60-1, n =3; bzip60-2, n =3). All data are shown
as means ±SD. The statistical significance of the differences was tested using one-way ANOVA and post-hoc comparisons. Different and same letters
indicate values with statistically significant (p<0.05) and nonsignificant (p>0.05) differences, respectively; The above nodules for semithin sections
were collected from at least 60 plants with three biological replicates.
Adv. Sci. 2025,12, 2414519 2414519 (8 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Figure 5. Alteration of ER biosynthesis in UPR mutants results in defects in nitrogen fixation. a–c) Representative SEM images of 2-week-old nodules in
zone II (a), zone II-III (b), and zone III (c), from left to right: WT, ire1b-1,bzip60-1. The ER lumen was indicated by the red arrow. Scale bars =0.5 μm. d)
Quantitative analysis of the ER lumen width in zone II. (WT, n=300; ire1b-1,n=293; bzip60-1,n=297). e) Quantitative analysis of the ER lumen width
in zone II-III. (WT, n=118; ire1b1,n=183; bzip60-1,n=170). f) Quantitative analysis of the ER lumen width in zone III. (WT, n=183; ire1b-1,n=185;
Adv. Sci. 2025,12, 2414519 2414519 (9 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
bzip60-1,n=183). All data are shown as mean ±SD. Statistical analysis was performed with the one-way ANOVA method with post-hoc comparisons.
The above SEM statistical results were randomly selected from 3 cells in each zone. g-i) Transcript level of nifD (g), nifK (h), nifH (i) was determined
by qRT-PCR, and the rhizobia housekeeping gene rpoA was used for normalization, n=6. All data are shown as mean ±SD. The statistical analysis
was performed by the two-way ANOVA and post-hoc comparisons. The letters indicate values with statistically significant (p<0.05) and non-significant
(p>0.05) differences, respectively.
Moreover, there likely exist reciprocal feedback regulations be-
tween the ER membrane expansion and the UPR signaling. In
other words, UPR plays positive regulatory roles in symbiosome
and nodule development.
Our findings uncovered ER expansion and UPR induction as
two key events in symbiosome development in M. truncatula nod-
ules, and further revealed the intrinsic coupling of ER morphol-
ogy with proper UPR function during root nodule symbiosis by
IRLC legumes. It should be noted that, anatomically, at least two
distinct forms of intracellular accommodation of symbiotic bacte-
ria in nitrogen-fixing legume nodules have been identified. One
type, known as SYM-type nodules, which involves rhizobia be-
ing released from infection threads and enclosed within sym-
biosome membrane, is observed in most genera of the legume
subfamily Papilionoideae, such as Pisum and Medicago,inall
nodulated genera of the tribe Mimosae in the legume subfamily
Caesalpinioideae, as well as in non-arboreal species of the Cae-
salpinioid genus Chamaecrista.[51,52]The other type, known as FT-
type nodules, in which rhizobia are retained within specialized,
thin-walled infection threads called fixation threads (FTs), are
found in a few early-branching Papilionoid legumes, most nodu-
lated non-Mimosoid Caesalpinioid legumes (the exception be-
ing arboreal Chamaecrista spp.),[52]and in nodules on Parasponia
(Cannabaceae), the only known non-legume capable of forming
nodules[1]. In the present study, we demonstrated that ER dy-
namic expansion and UPR activation are essential for symbio-
some accommodation and development in M. truncatula,which
has SYM-type nodules. Interestingly, however, recent TEM obser-
vations have shown that FTs in FT-type nodules on Caesalpinioid
nodules are also surrounded by a cell membrane arising from
ER, thus it is intriguing to investigate the interplay between ER
remodeling and UPR in a wider set of symbiotic nodule types in
future studies.[51]A deeper understanding of the cellular basis on
the root nodule symbiosis would unlock new strategies for opti-
mizing nitrogen fixation in crops and contribute to sustainable
farming practices in the future.
4. Experimental Section
Biological Materials and Growth Conditions:M. truncatula ecotypes
A17 and R108 were used in this study. M. truncatula seeds were scari-
fied with sulfuric acid (SCR, 7664-93-9, China) for 10 min, rinsed with ster-
ile water fifteen times, and surface-sterilized in 6% sodium hypochlorite
(SCR, 7681-52-9, China) for 5 min, and then washed fifteen times with ster-
ile water. Sterilized seeds were stratified on inverted agar plates for 24h at
4°C and germinated overnight at 22 °C before being transferred to sterile
perlite pots or used for hairy-root transformation. Plants were grown in
a greenhouse under controlled conditions with a 16-h light/8-h dark pho-
toperiod and a constant temperature of 22 °C. They were watered every
four days. For inoculation, the Sinorhizobium meliloti strain 2011 (Sm2011)
was used. Hairy-root transformation was performed using Agrobacterium
rhizogenes strain MSU440, following a previously described protocol.[53]
Figure 6. Dynamic ER expansion couples with enhanced UPR to achieve rhizobia endosymbiosis. The schematic diagram illustrates the model of ER
expansion coupled with UPR during root nodule symbiosis. The host ER surrounding symbiosomes undergoes dynamic expansion from tubules to
lamellae to support the accommodation and development of symbiosomes, which in turn induces host UPR activation. During UPR, IRE1B splices
the mRNA of bZIP60, and the resulting sbZIP60 protein translocates into the nucleus to activate downstream genes to further regulate ER expansion
surrounding symbiosomes and consequent nitrogen fixation. The figure was created with BioRender software.
Adv. Sci. 2025,12, 2414519 2414519 (10 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
Scan Electronic Microscopy:Nodule samples were fixed with
2.5% (vol/vol) glutaraldehyde with 0.1 M phosphate buffer (PB,
Na2HPO4.12H2O, NaH2PO4.2H2O) (pH 7.4), and then washed four
times in PB. Then they were first immersed in 1% (wt/vol) OsO4(Ted
Pella, 20816-12-0, USA) and 1.5% (wt/vol) K3[Fe (CN)6] (Sigma-Aldrich,
13746-66-2, USA) aqueous solution at 4 °Cfor1h.Afterwashing,nodules
were incubated in filtered 1% thiocarbohydrazide aqueous solution
(Sigma-Aldrich, 2231-57-4, USA) at room temperature for 30 min, 1%
unbuffered OsO4aqueous solution at 4 °C for 1h and 1% Uranyl acetate
aqueous solution (ZXBR, 6159-44-0, China) at 4 °C overnight following
four rinses in double distilled water for 10 min each between each step.
Then tissues were dehydrated through graded alcohol (30, 50, 70, 80, 90,
100, and 100%, 10 min each at 4 °C) into pure acetone (Sigma–Aldrich,
67-64-1, USA). Samples were infiltrated in a graded mixtures (3:1, 1:1,
1:3) of acetone and SPI-PON812 resin (SPI Supplies, 905929-77-4, USA)
(including 16.2 mL SPI-PON812 monomer, 10 mL dodeceny succini-
canhydride (DDSA), and 8.9 mL N-Methylol acrylamide (NMA), 1.5%
N-dimthylbenzylamine (BDMA), then changed pure resin. Finally, tissues
were embedded in pure resin with 1.5% BDMA and polymerized for 12 h
at 45 °C, 48 h at 60 °C.
AutoCUTS-SEM and 3D Reconstruction:Automatic collector of ultra-
thin sections scanning electron microscopy (AutoCUTS-SEM) was per-
formed as previously described.[54]For the Large-scale 3D reconstruction
study, 200 sections were collected by the ultramicrotome (Leica, UC7, Ger-
many) with the AutoCUTS device for each sample. Next, high throughput
serial sections were finally automatically acquired by Scanning Electron
Microscope (FEI, Helios Nanolab 600i dual-beam SEM, USA) with au-
tomated software (AutoSEE), and an image reconstruction program was
conducted. The image parameters include an accelerating voltage of 2 kV,
beam current of 0.69 nA, CBS detector, pixel size of 58.6 nm, and dwell
time of 5 μs. For 3D reconstruction, manual segmentation of cells and their
associated processes were performed using commercial software Imaris
(Bitplane, version9.0, Switzerland). The stack of labeled images was ex-
ported and processed further for 3D rendering and data analysis.
Phylogenetic Analysis:The 19 protein sequences of 15 MtRTNLBs gene
family were obtained from Phytozome13 (http://phytozome.jgi.doe.gov/
pz/portal.html) and protein sequences were aligned using ClustalW, and
the phylogenetic tree was carried out using the neighbor-joining method of
MEGA11, the bootstrapping value was set at 1000 replications to evaluate
the consistency of the analysis. The motifs analysis was accomplished by
the MEME server (http://meme-suite.org/index.html).TheTBtoolswere
used to visualize combined results.[30]
Plasmid Construction:As described in the previous study, the 2 ×
Phanta Max Master Mix DNA polymerase with high-fidelity (Vazyme, P515-
01, China) was used to amplify the full-length cDNA of MtRTNLB4-1,
MtRTNLB4-2, MtbZIP60 and ≈2000-bp upstream gDNA (promoter) se-
quence of MtIRE1A (Medtr8g073190), MtIRE1B (Medtr5g024510), Mt-
bZIP60 (Medtr1g050502) from M. truncatula A17. To further obtain
the pCambia1391-ProIRE1A/IRE1B/bZIP60-GUS constructs, ClonExpress
II One Step Cloning Kit (Vazyme, 7E771E3, China) was used to ligate
the ProIRE1A/IRE1B/bZIP60 fragment, and Pst I- (NEB, R0140V, USA)
and BamH I- (NEB, R0136V, USA) linearized pCambia1391 binary vec-
tor. For the constitutive overexpression constructs ProLjUb::MtRTNLB4-1,
ProLjUb::MtRTNLB4-2, ClonExpress II One Step Cloning Kit to ligate target
fragments and Kpn I HF- (NEB, R3142S, USA) and Xba I-(NEB, R0145L,
USA) linearized ProLjUb-Pro35S::GFP binary vector was used. For the One
Step Constructs were confirmed by Sanger sequencing. (Table S1,Sup-
porting Information).
RNA Extraction and qPCR:Total RNA was extracted as the previously
described method.[55]Briefly, ground sample powder was homogenized
in TRIzol reagent (Thermo Fisher, 380 511, USA) by vortexing at high
speed for 1 min at room temperature. After a 5 min incubation, chlo-
roform was added to the mixture, followed by vortexing for 1 min and
centrifugation at 12000 rpm for 10 min to separate the phases. The up-
per aqueous phase was carefully transferred to a fresh RNase-free tube
and mixed with an equal volume of ice-cold isopropanol. The mixture
was centrifuged at 12000 rpm for 10 min at 4 °C to precipitate the RNA.
The resulting pellet was washed with 200 μL of 70% ethanol and cen-
trifuged at 7000 ×g for 5 min. The supernatant was decanted without
disturbing the pellet, and the RNA pellet was air-dried. Finally, the RNA
was resuspended in 30 μL of RNase-free deionized water. 2 μLoftotal
RNA for each sample was applied for reverse transcription using the Su-
perScript III First-Strand Synthesis System (Vazyme, 7E731J3, China) with
the mixture of oligo (dT) primers and random primers. The RT-qPCR as-
says were performed using the SYBR Green real-time PCR master mix
(Vazyme, 7E0813G4, China) with a Real-time fluorescence quantitative
PCR system (BioRad, CFX96, USA). The expression of plant genes and
bacterial genes were normalized to MtACTIN11 or rpoA expression, re-
spectively. All of the reactions were performed three times independently.
A list of the primers used for RT-qPCR is provided in Table S1 (Supporting
Information).
Histochemical Staining, Resin Embedding and Sectioning:The pro-
moter regions upstream of the start codon (ATG) of MtIRE1A (2262 bp),
MtIRE1B (2029 bp), and MtbZIP60 (2474 bp) were amplified from M. trun-
catula A17 genomic DNA. These promoter sequences were then cloned
into the pCAMBIA1391z vector. The resulting constructs were introduced
into Agrobacterium tumefaciens strain EHA105, which was subsequently
used to transform M. truncatula R108 via stable transformation. Trans-
genic seedlings were screened by PCR amplification and sequencing using
GUS-specific primers (Table S1, Supporting Information). Transformed
plants were inoculated with Sinorhizobium meliloti strain Sm2011. At 14
dpi, nodules were harvested and immersed in 5-Bromo-4-chloro-3-indoxyl-
beta-D-glucuronide (X-Gluc) (Lablead, 114162-64-0, China) staining buffer
for 8 h to detect GUS activity. For detailed analysis, 10-μm nodule sections
were prepared using a Leica RM2265 microtome, stained with 0.1% Ruthe-
nium Red (Sigma–Aldrich, 11103-72-3, USA), and visualized using a Leica
M205FA microscope equipped with a DFC450c camera. Additionally, 5-μm
nodule sections were stained with 0.05% toluidine blue (Amresco, 672–5,
USA) for nodule structure visualization.
Subcellular Localization and Confocal Microscopy:Live-cell imaging was
performed using a spinning-disk confocal microscope (PerkinElmer, Ul-
traView VoX, USA) equipped with a Yokogawa Nipkow CSU-X1 spinning-
disk scanner, Hamamatsu EMCCD 9100–13, Nikon TiE inverted micro-
scope with the Perfect Focus System45. For fluorescence imaging, the
Pro35S::HDEL-RFP and Pro35S::RFP-MtbZIP60 vectors were transformed
into M. truncatula A17 by Agrobacterium rhizogenes transformation. The
transformed plants were infected with Rhizobium Sm2011. 21-day-post-
inoculation nodules were hand-sectioned using a double-edged razor
blade and immediately immersed at 2 μgmL
−1solution of 4′,6-diamidino-
2-phenylindole (DAPI) (Roche, 236 276, Switzerland) for 5 min. To visu-
alize the interaction between symbiosomes and the ER in nodule cells,
hand-sectioned nodules expressing HDEL-RFP were submerged in SYTO9
(Thermo Fisher, S34854, USA), stained for 5 min, and briefly washed with
distilled water. Acquired images were processed and analyzed using the
software of Volocity (Perkin Elmer, version 6.3, USA) and Image J software
as described previously.[55]
CRISPR–Cas9-Mediated Gene Editing:CRISPR–Cas9-mediated gene
editing was performed as described previously.[55,56]gRNA sequences
were used to guide Cas9 to the exons of MtIRE1A, MtIRE1B, MtbZIP60
(Table S1, Supporting Information). gRNA was cloned into the pAtU6-
26-sgRNA-SK vector. The cassette containing the sgRNA was then cloned
into the pCAMBIA1300-ProAtUBQ10::Cas9 binary vector and used to trans-
form Agrobacterium EHA105, which was then transformed into M. truncat-
ula R108. The mutants were screened by sequencing using gene-specific
primers (Table S1, Supporting Information).
UPR-Defective Mutant Phenotyping:For normal growth analysis,
plants were grown in vermiculite supplemented with FM medium and
sampled at 12 and 20 days for phenotypic evaluation. At least 16 plants
per genotype were analyzed. For infection event analysis, five-day-old
seedlings of Medicago wild-type R108 and UPR-defective mutants, grown
in vermiculite with FM medium, were inoculated with Sm2011-GFP
(OD600 =0.02). Plants were harvested at 3, 5, and 7 dpi to examine in-
fection events using a Nikon confocal microscope equipped with pho-
tomultiplier tubes (PMTs). At least 8 plants from each genotype were
used for quantification of infection events at each time point. For nodula-
tion phenotype analysis, nodules from wild-type (R108) and UPR-defective
Adv. Sci. 2025,12, 2414519 2414519 (11 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
mutants were harvested at 21 dpi. At least 36 plants per genotype were
used for this analysis.
Acetylene Reduction Assay:The M. truncatula seedlings were grown
on vermiculite and inoculated with rhizobium Sm2011. Nodules at 21 dpi
were collected and put into a closed 20 mL vial containing 2mL acetylene
(C2H2)at28°C for 3 h. For each sample, three biological replicates were
performed for analysis. Acetylene was measured using a gas chromato-
graph (EWAI, GC-4000A, China).
ER Stress Treatment Assay:For ER stress assay, 5-day-old seedlings
were transferred to FM culture medium containing 0.5 μgmL
−1TM (Al-
addin, 11089-65-9, China) with DMSO as the solvent or DMSO-only as
mock, 20 seedlings per replicate. Then, the seedlings grew vertically un-
der the normal growth condition. After 7 days, the seedlings primary root
length was measured. The length of primary roots was photographed and
further measured using ImageJ software. The above experiments were per-
formed three times independently with consistent results.
Bioinformation Prediction:The UNAFold Web Server(http://www.
unafold.org/mfold)was used to predict the RNA secondary structure of
MtbZIP60. The full structure information was shown in the source data
(Figure S4, Supporting Information). The cNLS Mapper Web Server (https:
//nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi) was used to
predict the nuclear localization signal. The TMHMM (http://www.cbs.dtu.
dk/services/) TMHMM/Web Server was used to predict the transmem-
brane domain of MtbZIP60.
Statistical Analysis:Statistical analysis was performed using Graph-
Pad Prism (version 8.0.2). Results were expressed as means ±stan-
dard deviation (SD). To determine statistical significance, an unpaired
two-sided Student’s t-test was used for comparisons between the two
groups, statistical significance was denoted as follows: *p<0.05, **p<0.01,
***p<0.001, **** p<0.0001, and ns indicates not significant. Then multi-
ple t-tests and one-way ANOVA were employed for comparisons involv-
ing multiple groups. Different and same letters indicate values with statis-
tically significant (p<0.05) and nonsignificant (p>0.05) differences, re-
spectively.
Supporting Information
Supporting Information is available from the Wiley Online Library or from
the author.
Acknowledgements
The authors are grateful to Haiyun Wang, Lei Su, and Yao Wu for help-
ing with microscopic observation at the State Key Laboratory of Plant Ge-
nomics, Institute of Microbiology, Chinese Academy of Sciences. And the
authors are grateful to Yun Feng, Xixia Li, Zhongshuang Lv, Chunliu Liu,
and Xiaoyun Zhang for helping with sample preparation and taking TEM
images at the Center for Biological Imaging (CBI), Institute of Biophysics,
Chinese Academy of Sciences. The authors also thank Dr. Guixian Xia for
the manuscript revision. This study was supported by the National Natural
Science Foundation of China (31925003 and 32230007), the CAS Project
for Young Scientists in Basic Research (YSBR-011).
Conflict of Interest
The authors declare no conflict of interest.
Author Contributions
J.R. and Q.W. contributed equally to this work. J.R. and Q.W. designed
the research plan and experiments. J.R. performed experiments, prepared
figures and videos, and wrote the original draft. Q.W. performed experi-
mental observation and interpreted the data and revised the manuscript.
X.Z., Y.C., J.W., Y.Y., and J.T. provided essential technical assistance. G.Q.
reviewed and edited the manuscript. Z.K. conceived the project, inter-
preted the data, wrote and revised the article.
Data Availability Statement
The data that support the findings of this study are available from the cor-
responding author upon reasonable request.
Keywords
compartmentalization, endosymbiosis, ER, symbiosomes, UPR
Received: November 7, 2024
Revised: February 12, 2025
Published online: February 22, 2025
[1] J. I. Sprent, New Phytol. 2007,174, 11.
[2] G. E. D. Oldroyd, Nat. Rev. Microbiol. 2013,11, 252.
[3] F. M. Martin, S. Uroz, D. G. Barker, Science 2017,356, eaad4501.
[4] C. Stonoha-Arther, D. Wang, Curr. Opin. Plant Biol. 2018,44, 155.
[5] J. I. Sprent, J. Ardley, E. K. James, New Phytol. 2017,215, 40.
[6] P. Mergaert, A. Kereszt, E. Kondorosi, Plant Cell 2020,32,42.
[7] M. Parniske, Curr. Opin. Plant Biol. 2000,3, 320.
[8] S. Roy, W. Liu, R. S. Nandety, A. Crook, K. S. Mysore, C. I. Pislariu, J.
Frugoli, R. Dickstein, M. K. Udvardi, Plant Cell 2020,32, 15.
[9] J. Fournier, A. Teillet, M. Chabaud, S. Ivanov, A. Genre, E. Limpens,
F. De Carvalho-Niebel, D. G. Barker, Plant Physiol. 2015,167, 1233.
[10] C.-W. Liu, A. Breakspear, N. Stacey, K. Findlay, J. Nakashima, K.
Ramakrishnan, M. Liu, F. Xie, G. Endre, F. de Carvalho-Niebel, G. E.
D. Oldroyd, M. K. Udvardi, J. Fournier, J. D. Murray, Nat. Commun.
2019,10, 2848.
[11] D. J. Gage, Microbiol. Mol. Biol. Rev. 2004,68, 280.
[12] S. Ivanov, E. Fedorova, T. Bisseling, Curr. Opin. Plant Biol. 2010,13,
372.
[13] F. M. Perrine-Walker, H. Kouchi, R. W. Ridge, Protoplasma 2014,251,
817.
[14] H. Zhang, J. Hu, Trends Cell Biol. 2016,26, 934.
[15] E. E. Fedorova, Int. J. Mol. Sci. 2023,24, 4647.
[16] D. S. Schwarz, M. D. Blower, Cell. Mol. Life Sci. 2016,73, 79.
[17] H. Chino, N. Mizushima, Trends Cell Biol. 2020,30, 384.
[18] P. Walter, D. Ron, Science 2011,334, 1081.
[19] C. Hetz, F. R. Papa, Mol. Cell 2018,69, 169.
[20] J. Celli, R. M. Tsolis, Nat. Rev. Microbiol. 2015,13, 71.
[21] D. Ron, P. Walter, Nat. Rev. Mol. Cell Biol. 2007,8, 519.
[22] Y. Wu, C. Whiteus, C. S. Xu, K. J. Hayworth, R. J. Weinberg, H. F. Hess,
P. De Camilli, Proc. Natl. Acad. Sci. U. S. A. 2017,114, E4859.
[23] W. Shen, L. Ma, X. Zhang, X. Li, Y. Zhao, Y. Jing, Y. Feng, X. Tan, F.
Sun, J. Lin, Sci. China:Life Sci. 2020,63, 171.
[24] R. Yang, Y. K. Dzowo, C. E. Wilson, R. L. Russell, G. J. Kidd, E. Salcedo,
R. S. Lasher, J. C. Kinnamon, T. E. Finger, J. Comp. Neurol. 2020,528,
756.
[25] N. J. Brewin, Annu. Rev. Cell Dev. Biol. 1991,7, 191.
[26] Y. S. Yang, S. M. Strittmatter, Genome Biol. 2007,8, 234.
[27] H. Nziengui, K. Bouhidel, D. Pillon, C. Der, F. Marty, B. Schoefs, FEBS
Lett. 2007,581, 3356.
[28] G. K. Voeltz, W. A. Prinz, Y. Shibata, J. M. Rist, T. A. Rapoport, Cell
2006,124, 573.
[29] M.West,N.Zurek,A.Hoenger,G.K.Voeltz,J. Cell Biol. 2011,193,
333.
[30] C. Chen, Y. Wu, J. Li, X. Wang, Z. Zeng, J. Xu, Y. Liu, J. Feng, H. Chen,
Y. He, R. Xia, Mol. Plant 2023,16, 1733.
[31] D. L. Wiest, J. K. Burkhardt, S. Hester, M. Hortsch, D. I. Meyer, Y.
Argon, J. Cell Biol. 1990,110, 1501.
[32] J. S. Cox, R. E. Chapman, P. Walter, Mol. Biol. Cell 1997,8, 1805.
Adv. Sci. 2025,12, 2414519 2414519 (12 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH

www.advancedsciencenews.com www.advancedscience.com
[33] H. Zhu, B. Bhatt, S. Sivaprakasam, Y. Cai, S. Liu, S. K. Kodeboyina, N.
Patel, N. M. Savage, A. Sharma, R. J. Kaufman, H. Li, N. Singh, Nat.
Commun. 2019,10, 1084.
[34] C. Ruberti, S.-J. Kim, G. Stefano, F. Brandizzi, Curr. Opin. Plant Biol.
2015,27, 59.
[35] C.-Y. Yu, Y. Cho, O. Sharma, K. Kanehara, J. Exp. Bot. 2022,73, 1268.
[36] T. Imamichi, N. Kusumoto, H. Aoyama, S. Takamatsu, Y. Honda, S.
Muraoka, Y. Hagiwara-Komoda, Y. Chiba, H. Onouchi, Y. Yamashita,
S. Naito, Nucleic Acids Res. 2024,52, 4276.
[37] A. A. Moreno, M. S. Mukhtar, F. Blanco, J. L. Boatwright, I. Moreno,
M. R. Jordan, Y. Chen, F. Brandizzi, X. Dong, A. Orellana, K. M.
Pajerowska-Mukhtar, PLoS One 2012,7, e31944.
[38] S. H. Howell, Plant Cell Environ. 2021,44, 2625.
[39] B. Roux, N. Rodde, M.-F. Jardinaud, T. Timmers, L. Sauviac, L. Cottret,
S. Carrère, E. Sallet, E. Courcelle, S. Moreau, F. Debellé, D. Capela, F.
de Carvalho-Niebel, J. Gouzy, C. Bruand, P. Gamas, Plant J. 2014,77,
817.
[40] H. Yoshida, M. Oku, M. Suzuki, K. Mori, J. Cell Biol. 2006,172, 565.
[41] Y. Deng, R. Srivastava, S. H. Howell, Proc. Natl. Acad. Sci. U. S. A.
2013,110, 19633.
[42] D. K. Ko, F. Brandizzi, Nat. Plants 2022,8, 481.
[43] Q. Ye, F. Zhu, F. Sun, T.-C. Wang, J. Wu, P. Liu, C. Shen, J. Dong, T.
Wan g, Mol. Plant 2022,15, 1852.
[44] J. J. Terpolilli, G. A. Hood, P. S. Poole, Adv. Microb. Physiol. 2012,60,
325.
[45] B. Li, F. Niu, Y. Zeng, M. K. Tse, C. Deng, L. Hong, S. Gao, S. W. Lo,
W. Cao, S. Huang, Y. Dagdas, L. Jiang, Proc. Natl. Acad. Sci. U. S. A.
2023,120, e2208351120.
[46] S. Schuck, Nat. Cell Biol. 2016,18, 1118.
[47] G. Stefano, C. Hawes, F. Brandizzi, Curr. Opin. Plant Biol. 2014,22,
30.
[48] S. Sun, G. Zhao, M. Jia, Q. Jiang, S. Li, H. Wang, W. Li, Y. Wang, X.
Bian, Y. G. Zhao, X. Huang, G. Yang, H. Cai, J. C. Pastor-Pareja, L. Ge,
C. Zhang, J. Hu, Sci. China:Life Sci. 2024,67, 230.
[49] S. Schuck, W. A. Prinz, K. S. Thorn, C. Voss, P. Walter, J. Cell Biol. 2009,
187, 525.
[50] R. Wright, M. Basson, L. D’Ari, J. Rine, J. Cell Biol. 1988,107,
101.
[51] S. M. de Faria, J. J. Ringelberg, E. Gross, E. J. M. Koenen, D. Cardoso,
G. K. D. Ametsitsi, J. Akomatey, M. Maluk, N. Tak, H. S. Gehlot, K.
M. Wright, N. Teaumroong, P. Songwattana, H. C. de Lima, Y. Prin,
C. E. Zartman, J. I. Sprent, J. Ardley, C. E. Hughes, E. K. James, New
Phytol. 2022,235, 2365.
[52] P. A. Casaes, J. M. Ferreira Dos Santos, V. C. Silva, M. F. K. Rhem, M.
M.TeixeiraCota,S.M.deFaria,J.G.Rando,E.K.James,E.Gross,J.
Exp. Bot. 2024,75, 3643.
[53] E. Limpens, J. Ramos, C. Franken, V. Raz, B. Compaan, H. Franssen,
T. Bisseling, R. Geurts, J. Exp. Bot. 2004,55, 983.
[54] X. Li, G. Ji, X. Chen, W. Ding, L. Sun, W. Xu, H. Han, F. Sun, J. Struct.
Biol. 2017,200, 87.
[55] Q. Wang, Y. Huang, Z. Ren, X. Zhang, J. Ren, J. Su, C. Zhang,
J. Tian, Y. Yu, G. F. Gao, L. Li, Z. Kong, Nat. Plants 2020,6,
800.
[56] X. Zhang, Q. Wang, J. Wu, M. Qi, C. Zhang, Y. Huang, G. Wang, H.
Wang, J. Tian, Y. Yu, D. Chen, Y. Li, D. Wang, Y. Zhang, Y. Xue, Z. Kong,
Nat. Plants 2022,8, 1275.
Adv. Sci. 2025,12, 2414519 2414519 (13 of 13) © 2025 The Author(s). Advanced Science published by Wiley-VCH GmbH