Symbiotic nitrogen fixation in the root nodules of legumes is one of the important sources of fixed nitrogen for the biosphere. Nitrogen-fixing bacteria temporary reside in the infected cells of root nodule symplast (Oldroyd et al., 2011) forming a colony, putatively with some extracellular property (Luo et al., 2022). Simultaneously with the growth of the root nodule, intracellular bacteria gain access to host cell resources, colonize the host cells, and begin to fix atmospheric nitrogen through the induction of the enzyme nitrogenase (Beringer, 1974; Parniske, 2018; Roy et al., 2020; Coba de la Peña et al., 2018).

The termination of symbiosis, known as “root nodule senescence”, is manifested as a lysis of the infected cells (Berrabah et al., 2023). At the cellular level, the life span of infected cells is quite short, around 3–6 weeks, and the root nodule is highly sensitive to suboptimal environmental conditions (Chakraborty and Harris, 2022). The reasons for the rapid termination of symbiosis have not yet been elucidated in detail, but the short lifespan of infected cells compared to the lifespan of root cells from which the nodules are formed may indicate the suboptimal conditions of the infected cells and the induction of the termination processes specific for the root nodule (Wojciechowska et al., 2018; Sauviac et al., 2022; Berrabah et al., 2023). A high vulnerability of root nodule tissue may be caused by some pre-existing defects in the ion balance of the root nodule (Fedorova et al., 2021; Trifonova et al., 2022). Therefore, the study of factors involved in root nodule homeostasis requires further research. Potassium is one of the major ions involved in the adaptive responses of plants to abiotic stresses (Rubio et al., 2020; Zhang et al., 2022; Wakeel and Ishfaq, 2021). It is essential for controlling the osmotic status as well as the cation–anion balance of the cell. The availability of K⁺ to the cell depends on the transport and concentration of K⁺ on both sides of the membrane. Potassium transporters in plants are involved in K⁺ uptake or release as well as storage in vacuoles and translocation of ions between tissues and organs (Ragel et al., 2019; Fedorova et al., 2021, Luo et al., 2022).

With the aim to elucidate the interdependence between potassium transport mediated by host plant ion transporters in nodule cells (Domínguez-Ferreras et al., 2009; Zhang et al., 2022) and potassium transport in bacteroids, we have created and analyzed a bacterial null mutant of bacterial potassium ion transporter Smkup1. To elucidate the possible reaction of symbiotic cells, the expression of genes induced during root nodule senescence and autophagy induction was analyzed since the termination of symbiosis involves the lysis of the nodule cells and the bacteria within them (Berrabah et al., 2023; Wleklik and Borek, 2023). To identify the putative mechanisms of symbiosis termination, we conducted an in silico analysis of protein/protein interactions of several regulators of autophagy and senescence.

Medicago truncatula Gaertn. cv. Jemalong A17 plants were grown according to Limpens et al. (2004). Sinorhizobium meliloti strain Sm2011 was used for nodulation. M. truncatula seeds were sterilized and germinated according to Limpens et al. (2004). Seedlings were sown in sand/vermiculite (1:1) mixture supplemented with Fahraeus medium and inoculated with S. meliloti 2011–3 days after sowing. M. truncatula plants were grown in a growth chamber (20°C, 16 h day light, 60%–70% relative humidity), supplemented by halogen lamps (PPF 1580 мкмоль/м²/с). The light intensity on the level of plants is 140–160 μ mo. Nodules were harvested after 8 weeks.

To analyze the growth of control (Sm2011) and mutant (Smkup1) strains of rhizobia, the density of the bacteria culture was measured over a period of days of growth in LB liquid culture; the suspension density was determined using a GENESYS™ 40/50 vis/UV– vis spectrophotometer. The samples have been measured in triplicates from the set of nine analytical 15- mL tubes and grown on a shaker for 3 days. On each day, the three tubes have been measured.

The nodule weight per plant was measured after 8 weeks of cultivation. The sample size in both experiments was 20 plants per analysis.

The analysis of gene expression was performed by using real-time quantitative PCR (qPCR) with a 7300 Real Time PCR System (PE Applied Biosystems) with SYBR 455 Green Supermix (Evrogen, Russia). The genes Mtc27, MtGAPDH, and SMc00128 were used as endogenous controls for the plant host and for bacteria, respectively (Coba de la Peña et al., 2008). The ΔΔC_T method (Pfaffl, 2001) was applied for PCR relative quantification. For PCR analysis, four to six biological replications were used.

Total RNA was isolated from 8-week-old nodules of control plants (inoculated with wt Sm2011) and of plants inoculated with the Smkup1 mutant using the RNeasy Plant Minikit (Qiagen, USA) and retrotranscribed into cDNA using SKO22L kit (Evrogen, Russia). A Thermo Scientific NanoDrop microvolume spectrophotometer was used for RNA concentration measurements. The RNA purity and integrity quality have been estimated by using the A260/A280 ratio.

Gene-specific primers were designed using the PRIMER 3-PLUS software (Untergasser et al., 2007). Light and electron microscopy, respectively, have been used to analyze the anatomy and ultrastructure of the nodules. The nodules were fixed and embedded in LR White as described (Fedorova et al., 2021), observed by using a light microscope, and analyzed using a LIBRA120 electron microscope.

The mutation Smkup1 was verified using the quantitative PCR method (Figure 1), which demonstrated the absence of kup1 gene expression on the mutant background.

To elucidate the biological characteristics of the mutant Smkup1 in symbiosis, an analysis of its growth dynamics in bacterial culture, an analysis of nitrogen fixation ability of root nodules, and an analysis of the expression of macro- and microsymbiont genes that respond to stress and the genes of autophagy pathway were conducted. The effect of mutation on the growth of rhizobia was analyzed in a bacterial culture grown for 3 days on LB medium (Table 1).

The dynamics of wild-type (Sm2011) and mutant bacteria (Smkup1) growth in liquid culture. Asterisks (*) indicate significant statistical differences between the two strains of rhizobia. The analysis of bacterial culture growth has shown a significant difference between control strain Sm2011 and Smkup1 mutant after 48 and 72 h of growth (*) (Table 1). The mutant strain growing in culture significantly lags behind the control strain by growth rate.

The impact of Smkup1 mutation on the growth of root nodules was assessed based on the mass of root nodules per plant after cultivation for 8 weeks post-sowing (Table 2).

Based on the analysis of nodule growth, no significant differences in root nodule mass between the wild type and the mutant was observed; that means that the root nodule growth per se was not negatively affected by Kup1 loss of function.

Phenotypically, the root nodules also did not show signs of rapid senescence such as the formation of green-colored zones as a result of the breakdown of leghemoglobin.

In the process of legume–rhizobia symbiosis, a colony of living bacteria is maintained for a long time (up to 4 to 5 weeks) in the symplast of living root nodule cells, which is an exceptionally rare phenomenon for any eukaryote. The maintenance of the colony leads to changes in the biological characteristics of the infected cell and may be causal for the infected cell’s short lifespan (Chakraborty and Harris, 2022)— for example, the potassium content diminishes in infected cells of Medicago truncatula nodules due to the mislocation of host plant channels MtAKT1 and MtSKOR/GORK involved in intracellular potassium transport (Fedorova et al., 2021).

The role of each partner in potassium transport in root nodules is not fully understood (Domínguez-Ferreras et al., 2009; Fedorova et al., 2021; Li et al., 2024), so we have analyzed the involvement of the bacterial partner of the symbiosis using the mutant of a bacterial potassium transporter Smkup1. The mutation of one from several bacterial transporters has not caused a structural damage of the root nodule or rapid senescence of the symbiosomes. However, despite the absence of visible aberrations, root nodules have a high level of expression of genes involved in stress response (Figures 4, 5). At the same time, the expression of genes from ATG group responsible for the induction of autophagy was downregulated in comparison with wild-type-induced root nodules (Figures 6, 7).

Autophagy is a housekeeping process that is induced in the cells of eukaryotic organisms under unfavorable conditions, such as lack of energy, limited access to food, or the presence of microorganisms in the cells. It is assumed that the introduction of bacteria into any eukaryotic cell triggers a protective response that manifests as the induction of autophagy processes. Autophagosomes, the lytic organelles induced in the cell for digestions of parts of the cytoplasm or cell organelles, are formed under stress conditions or defense against bacteria that have entered the cell (Cadwell et al., 2025). In light of this, the role of autophagy is recognized as essential housekeeping processes (Agbemafle et al., 2023; Estrada-Navarrete et al., 2016; Liu et al., 2020; Zheng et al., 2025).

The studies investigating autophagy processes induced by microorganisms mainly are focused in Gram-negative mammalian pathogens, such as Salmonella and Coxiella (Cadwell et al., 2025). Animal pathogens are able to form so-called bacteria-containing vacuoles, which consist of a host cell membrane containing one or more bacteria, that allows pathogens to survive within a cell for several hours and avoid immediate elimination (Cadwell et al., 2025). The membrane of bacteria-containing vacuoles is marked by small GTPases of the RAB family (RAB7 and RAB24) and specific polyphosphoinositides, that determine membrane/organelle identity and the membrane fusion (Liu et al., 2020; Cadwell et al., 2025). The existence of pathogen-containing vacuoles is terminated by the fusion with autophagosomes, vacuole-like structures with lytic properties. The process of fusion is coordinated by RAB GTPase RAB7 and effectors of homotypic fusion and protein sorting (HOPS) complex VPS11 and VPS15.

Structurally, bacteria-containing vacuoles have much in common with symbiosomes in root nodule cells, symbiosome membranes are also recruiting small GTPase RAB7, and SNARE proteins (Gavrin et al., 2014). However, HOPS effectors VPS11 and VPS15 recruitment to symbiosome membranes is inhibited due to the strong downregulation of the expression of these effectors in infected root nodule cells (Gavrin et al., 2014), making symbiosomes temporarily immune to the autolysis.

In this regard the conditions that prevent or temporarily inhibit the autophagic clearance of bacteria in root nodule infected cell may be of a quite important for the symbiotic relationships.

According to in situ expression, the genes involved in the autophagy pathway are expressed in all zones of the root nodule (Semenova et al., 2024); however, autophagosomes in their “classical form” (similar to structures found in animal cells infected with bacteria) were not diagnosed in root nodule cells apart of the mutant DNF1 nodules, that demonstrates that the presence of autophagosomes in infected cells is an extremely rare phenomenon (Wang et al., 2010). The autophagy clearance in root nodule infected cells elicited by mutant strain Smkup1 is rather inhibited than upregulated, that hints at the opportunities for maintaining the existence of a bacterial colony even in unfavorable conditions. Although infected cell experiences suboptimal sugar levels, insufficient levels of essential ions like potassium, a low oxygen level, it is able to support a colony of nitrogen-fixing bacteria up to 6–8 weeks (Belfquih et al., 2022; Ayala-García et al., 2025; Zhou et al., 2021) albeit at the cost of its own lifespan.

The high expression of such genes as Beclin1, SnF and Dhh1, according to in situ analysis (Table 3) key regulators of the autophagy process hints the starvation environment in the nitrogen fixation zone of the nodule as well as an induction of autophagy. Quite interesting is the expression in the root nodule cells of WRKY20, the gene involved in autophagosome maturation, taking into account that in these cells the autophagosomes have never been found. The autophagosomes’ membrane formation depends on the host membranes, mainly those of the endoplasmic reticulum. It should be remembered that the membranes of symbiosomes are also formed from the host cell endoplasmic reticulum (Bellucci et al., 2017; Behnia and Munro, 2005, Fedorova, 2023). How to reconcile and explain the relatively high level of autophagy gene expression in the root nodule and the absence of “classical “ autophagosomes in the cells remains unclear. The mechanisms involved in preventing the autophagic removal of symbiosomes and death of infected cells, which are induced in the root nodule, are also not yet understood.

We can hypothesize that the regulation of autophagy in root nodules, at least in part, may depend on the conformation of proteins rather than genes. To identify potential pathways for autophagy regulation in root nodules, we decided to conduct an in silico analysis of protein–protein interactions based on the genes of autophagy pathways that we have already identified.

Quite interesting is the role of a positive regulator of autophagy Beclin1 that, according to the analysis of protein/protein interactions, is associated with the ubiquitin-like protein ATG12 (UniProt identifier AOAO72TLU6), which belongs to an autophagy pathway and is involved in cytoplasm- to-vacuole transport (Figure 8), the processes that are partly inhibited in infected cells (Gavrin et al., 2014).

The analysis of the interaction network for M. truncatula NAC1, a putative transcription factor of symbiotic interaction, has shown interacting proteins involved in programmed cell death (PCD) and stress response as well as a STAY-GREEN protein (UniProt identifier G7KES8_MEDTR) that regulates nodule development and senescence (Zhou et al., 2021) (Figure 9).

The analysis of BRASSINOSTEROID INSENSITIVE 1-associated receptor kinase 1 (BAK1) has shown interacting proteins involved in the negative regulation of defense response to bacteria and negative regulation of cell death (Figure 10). The role of BAK1 in the prevention of bacteria elimination in symbiotic cells may be of major importance and needs further research.

Data concerning protein–protein interactions in infected cells of root nodules and certain similarities between the biogenesis of the symbiosome and the formation of the autophagosome permit us to hypothesize that the autophagy pathway may be co-opted into the processes of symbiosis formation and that the regulation of this process may be quite important in stress conditions.

This could indicate putative neofunctionalization of autophagy genes in a symbiotic process that may be involved at a later stage of intracellular accommodation, after the formation of nitrogen-fixing symbiosomes, and be a useful tool for symbiosome maintenance regulation.