Molecular determinants of a symbiotic chronic infection

Abstract

Rhizobial bacteria colonize legume roots for the purpose of biological nitrogen fixation. A complex series of events, coordinated by host and bacterial signal molecules, underlie the development of this symbiotic interaction. Rhizobia elicit de novo formation of a novel root organ within which they establish a chronic intracellular infection. Legumes permit rhizobia to invade these root tissues while exerting control over the infection process. Once rhizobia gain intracellular access to their host, legumes also strongly influence the process of bacterial differentiation that is required for nitrogen fixation. Even so, symbiotic rhizobia play an active role in promoting their goal of host invasion and chronic persistence by producing a variety of signal molecules that elicit changes in host gene expression. In particular, rhizobia appear to advocate for their access to the host by producing a variety of signal molecules capable of suppressing a general pathogen defense response.

Figure 1

Schematic model of nodule development. (a–b) Host flavonoids exuded into the soil trigger bacterial Nod Factor production. Nod factor is perceived by host receptors and elicits various host responses, such as root hair curling and root hair invasion. Root hair invasion also requires bacteria EPS and host ROS production. Nod factors induce mitotic cell division in the root cortex (represented in blue), leading to formation of the nodule meristem. An indeterminate nodule originates from the root inner cortex and has a persistent meristem (Zone I). The nodule also contains an invasion zone (Zone II) and a nitrogen-fixing zone (Zone III). In older nodules, a senescent zone (Zone IV) develops in which both plant and bacterial cells degenerate. (c) Bacteria enter the nodule through root hairs in a structure called the infection thread (IT) that elongates toward the nodule meristem; nucleus (n), vacuole (v). (d ) At the tip of the growing IT, bacteria are endocytosed into the cytoplasm of postmitotic endoploid cells. Each bacterium is surrounded by a host-derived peribacteroid membrane (PBM) and proceeds to differentiate into the specialized symbiotic form called a bacteroid. Bacteroids establish a chronic infection of the host cytoplasm and enzymatically reduce dinitrogen to provide a source of biologically usable nitrogen to the host (Zone III). (e) In contrast to an indeterminate nodule, a determinate nodule lacks a persistent meristem and all developmental stages proceed synchronously. (f) Infected cells of determinate nodules typically lack vacuoles (v).

Figure 2

Representative signaling molecules critical for symbiosis. (a) Host flavonoids: luteolin, daidzein, and genestin. (b) The Nod factor produced by S. meliloti and biosynthetic enzymes (Nod proteins) involved in its synthesis. (c) The S. meliloti exopolysaccharide succinoglycan. ExoH is responsible for succinyl modification; succinoglycan molecular weight is controlled by ExoPTQ and two extracellular glycosylases, ExsH and ExoK. (d ) Schematic representation of S. meliloti lipopolysaccharide (LPS). LpsB is a glycosyltransferase with broad substrate specificity involved in synthesis of the LPS core. AcpXL, LpsXL, and BacA are required for the Very-Long-Chain Fatty Acid (C28) modification of lipid A.

Figure 3

Schematic representation of the rhizobium cell cycle at different stages of symbiosis. (a) The S. meliloti cell cycle is modeled after that of the alphaproteobacterium Caulobacter crescentus. A cell division cycle is comprised of three distinct phases: G₁, S, and G₂. In C. crescentus, DNA replication is initiated only once per cell cycle, in S phase; however, chromosome segregation begins during S phase and continues in G₂ phase. Cell division begins in G₂ phase and is completed before the next DNA replication initiation event. During free-living growth, S. meliloti is thought to initiate DNA replication only once per cell cycle and divides asymmetrically to produce daughter cells of different size. In analogy to C. crescentus, the small daughter cell likely proceeds into G₁ phase while the larger daughter cell directly re-enters S phase. (b) S. meliloti proliferating in the IT originate from a clonal expansion of founder cells entrapped in the tip of the root hair curl. Cells appear to lack flagella and are loosely associated with one another in a pole-to-pole manner, typically forming two or three columns with a braided appearance. Active propagation of bacteria is observed only in a limited area called the growth zone near the tip of the IT, while bacteria outside of the growth zone do not grow or divide. It seems likely that the restricted growth of bacteria enables synchronization of bacterial growth with extension of the IT. (c) Bacteria colonize the cytoplasm of plant cells located in the invasion zone (see Figure 1d ). Bacteria are surrounded by a plant-derived membrane and differentiate into a bacteroid. In S. meliloti, DNA endoreduplication occurs during bacteroid differentiation and results in dramatic cell branching and enlargement to a length of 5–10 μm; in comparison, free-living counterparts are rod-shaped cells of 1–2 μm. These terminally differentiated (G₀ phase) bacteroids are unable to resume future growth. Orange lines, host plasma membrane; green lines, host cell wall. (d ) A model of the S. meliloti cell cycle in planta has three possible exits from S phase, two of which (in blue) represent an exit from the typical free-living cycle (in red ). Bacteria within the infection thread are thought to progress through the cell cycle in the same manner as free-living cells, and in particular transition from S phase into G₂ phase (represented by arrow 1). Bacteria that undergo bacteroid differentiation undertake the process of endoreduplication and therefore re-enter G₁ phase after the completion of S phase (represented by arrow 2); the bacteria may cycle from S to G₁ multiple times during endoreduplication. Once endoreduplication is complete, the bacteroid enters a terminally differentiated state (G₀) and is no longer able to initiate cellular growth or DNA replication (represented by arrow 3).

Figure 4

Schematic representation of cell-cycle progression in C. crescentus and S. meliloti. (a) The cell cycle is comprised of three distinct phases: G₁, S, and G₂ (shown in red ). The timing of cell cycle-related events is shown in blue. (b) During G₁ phase, C. crescentus is a swarmer cell, which is motile and has a polar flagellum and pili. When entering S phase, the swarmer cell ejects the flagellum, retracts the pili, and differentiates into a stalked cell. The stalked cell is uniquely competent to initiate DNA replication. After chromosome segregation in G₂ phase, the cell divides asymmetrically to produce two different daughter cells: a swarmer and a stalked cell. During cell-cycle progression, cellular localization of DivK is controlled by phosphorylation: nonphosphorylated DivK is uniformly distributed in the cytoplasm and DivK~P is localized to the cell poles. (c) Recently it was shown that S. meliloti also divides asymmetrically to form a “small” cell and a “large” cell. Localization of the DivK homolog indicates that the “small” and “large” cells are counterparts of the C. crescentus swarmer and stalked cells, respectively. Flagella are omitted from this scheme because their localization has not been examined during cell cycle progression in S. meliloti.