Saltbush seedlings (Atriplex spp.) shed border-like cells from closed-type root apical meristems

Liquid growth media

A half strength Hoagland Modified Basal Salt solution (815 mg L⁻¹, pH 5.4, Phyto Technology Laboratories, USA) buffered with 2 mM 2-(N-morpholino)ethanesulfonic acid (MES) (pH 6.0, Glentham Life Sciences, UK) was prepared. Salinity treatments were formulated by the addition of sodium chloride (NaCl; Chem-Supply, Australia) to final salt concentrations of 12, 70, 120, 230 and 310 mM NaCl, which encompasses the range where saltbush has been found to survive (Liddicoat and McFarlane 2007; Nedjimi 2014).

Plant growth

Seeds of three saltbush species (Atriplex nummularia L., Atriplex amnicola W., and Atriplex vesicaria Heward ex. Benth) (Succession Ecology, South Australia; Nindenthana Seeds, Western Australia) were surface sterilised in 4% (v/v) bleach, 95.9% (v/v) ethanol and 0.1% (v/v) Triton for 10 min with gentle shaking. Sterilised seeds were germinated under aseptic conditions in a sterile, semi-hydroponic system, which consisted of a plastic tissue culture container (46 × 92 mm) (ThermoFisher, Australia) filled with 1 mm glass beads (BioSpec Products, USA) up to 25 mm depth. Aluminium foil was placed around the bottom of the container (also to a depth of 25 mm), blocking most light to the glass beads to mimic soil conditions. The glass beads were saturated with 4 mL of liquid growth media. Seeds were placed on the bed of glass beads (15 seeds per container) and four containers for each species were incubated in a growth cabinet for 7 days (day/night cycle of 25°C/15°C, 16 h/8 h; light intensity of 410 μmol m⁻² s⁻¹). As seeds were placed on top of the glass beads for germination, shoots were exposed to light while roots grew into the dark, glass bead substrate (see Supplementary Fig. S1). To ensure that root elongation was not negatively affected by the growth method chosen (i.e. the saturated glass bead system), root length in glass beads was compared to root length on filter paper in Petri dishes, both with 4 mL liquid growth media. There were no significant differences in root length of the same species between the growth systems (data not shown).

Root morphology

Root exudation

Statistical analysis

We tested the effect of salinity concentration (12, 70, 120, 230 and 310 mM NaCl), saltbush species (A. nummularia, A. amnicola, and A. vesicaria), and their interaction on root morphology and root exudate traits using two-way ANOVAs with Tukey’s multiple comparisons test using GraphPad Prism (ver. 9). Graphs were plotted in GraphPad Prism (ver. 9).

Root apical meristem organisation

To characterise the RAM type of the saltbush species, root caps were probed with Direct Red 23, a fluorescent dye that is highly specific for crystalline cellulose microfibrils (Ropitaux et al. 2019). Direct Red 23 strongly stained saltbush cell walls but did not stain the surrounding mucilage. Cell lineages in the RAM were distinct and organised, indicating that saltbush have closed-type RAMs (Fig. 1). Saltbush released long, thin strands of cells that did not separate from each other after being released from the root surface and formed sheets at the end of the root tip, indicative of BLCs.

Fig. 1.

(a) Fluorescence microscopy of root caps of three species of saltbush (Atriplex nummularia, A. amnicola, A. vesicaria) grown in vitro at five salinity concentrations (12, 70, 120, 230, and 310 mM NaCl). Staining with Direct Red 23 reveals saltbush have border-like cells (BLCs). Images are representative of n = 3–5 replicates. (b) Representative image of A. amnicola grown at 12 mM NaCl shows that saltbush have closed-type root apical meristems with distinct cell lineages. Pre-BLCs, cells that will become border-like cells once detached from the root cap; Col, columella cells; QC, position of presumptive quiescent centre. Scale bars = 100 μm.

Root exudate quantification

Material from the four tissue culture containers for each treatment were combined into a bulk sample. The total dry weight of root exudates increased as salinity concentration increased in all saltbush species. At 12 mM NaCl, A. nummularia, A. amnicola, and A. vesicaria had very low total root exudate yields of 28.5 mg, 24.8 mg, and 36.4 mg, respectively (Fig. 2). This increased up to 59.5 mg, 74.4 mg, and 77.8 mg for A. nummularia, A. amnicola, and A. vesicaria, respectively, at 120 mM NaCl. The dry weight of root exudates was numerically greatest at 310 mM NaCl for all three species of saltbush, with A. nummularia showing the greatest increase to 302.3 mg (Fig. 2).

Fig. 2.

Total dry weight of root exudates in three saltbush species at five salinity concentrations (12, 70, 120, 230, and 310 mM NaCl). Root exudates from each treatment (four tissue culture containers) were pooled into a bulk sample, so statistical analysis is not presented. Anum, Atriplex nummularia; Aamn, Atriplex amnicola; Aves, Atriplex vesicaria.

Root exudate analysis for monosaccharide identification

To define the different components of root exudates produced by saltbush, water-soluble and water-insoluble fractions were separated, freeze-dried, and weighed. The water-soluble fraction of root exudates, referred to hereafter as root mucilage, increased from the lowest salinity concentrations up to a maximum at 230 mM NaCl (Fig. 3a). The water-soluble fraction comprised 51 to 60% of the total weight for the saltbush species at 12 mM NaCl. This increased with salinity levels such that the water-soluble fraction was 95 to 99% at 230 mM NaCl and 90 to 93% at 310 mM NaCl (Fig. 3a).

Fig. 3.

Comparison of the (a) water-insoluble to water-soluble (root mucilage) fractions of root exudates and (b) the relative quantities of salt, monosaccharides and unknown compounds in the root mucilage of saltbush at five salinity concentrations. Anum, Atriplex nummularia; Aamn, Atriplex amnicola; Aves, Atriplex vesicaria.

The greatest percentage of known compounds in the root mucilage of saltbush was attributed to salt (Fig. 3b). Salt concentration was greatest at medium external salinity concentrations of 70 mM NaCl for A. amnicola and A. vesicaria, and 120 mM NaCl for A. nummularia. At 70 mM NaCl, the mucilage of A. amnicola and A. vesicaria was comprised almost completely of salt with very low amounts of monosaccharides. However, at the higher external salt concentrations of 230 mM and 310 mM NaCl, the percentage of salt in the mucilage dropped and the percentage of unknown compounds was greatest. A. nummularia root mucilage had the greatest percentage of unknowns compared to the other species, but the lowest percentage of salt.

Monosaccharide analysis of saltbush root mucilage showed the most prominent peaks in the low-molecular weight range between 0 and 6 min, with only very minor peaks observed in the high molecular weight range (Fig. S2). In the low molecular weight range, all three saltbush species showed peaks at 3.5 min and 3.8 min, indicative of glucose and fructose, respectively (Fig. S2). Trace amounts of mannose, arabinose and galactose, were also present at some, but not all, salinity levels (Table 1). Mannose only appeared in A. vesicaria at 12 mM NaCl. Overall, A. vesicaria contained the greatest range and quantity of monosaccharides, predominantly at low to moderate salinity concentrations (Table 1).

The water-insoluble fraction of the exudates was determined to mainly comprise damaged and broken BLCs that had become enmeshed in mucilage (Fig. S5). This fraction was not analysed further.

Root exudate microscopy for polysaccharide identification

Initially, general polysaccharide stains were applied to the extruded root mucilage. Calcofluor white (CFW) staining indicated the presence of β-glucans, likely callose, cellulose and xyloglucan (Harrington and Hageage 2003). CFW staining was also present in the cell walls of saltbush roots, suggesting that they contain β-glucans (Fig. S3a). Staining with ruthenium red and crystal violet indicated the presence of acidic polysaccharides, most likely pectins, in the root mucilage and BLCs (Fig. S3b).

Immersion immunofluorescence labelling was used in combination with tissue clearing to investigate polysaccharide distribution within root exudates of saltbush. Staining with mAb LM19, which binds to un-esterified and partially esterified homogalacturonan, was detectable on the surface of BLCs produced by all three saltbush species at all salinity levels, but was not present inside cells (Fig. 4) (Verhertbruggen et al. 2009). Secondary antibody-only controls did not show any labelling with LM19 (Fig. S4). LM20, which labels esterified homogalacturonan, did not bind to cell surfaces (data not shown).

Fig. 4.

Immunofluorescent labelling of saltbush root caps with LM19 (green) indicates the presence of un-esterified homogalacturonan on the surface of border-like cells. Roots are stained with Direct Red 23 (red). Images are representative of n = 3–5 replicates. Scale bars = 100 μm.

Immersion immunofluorescence labelling was also used to investigate the occurrence of xyloglucan and AGPs in the root tissue using mAbs LM15 and JIM8 respectively. LM15 and JIM8 both bound strongly to the surfaces of BLCs (Fig. 5), as seen for LM19.

Fig. 5.

Immunofluorescent labelling of saltbush root caps (green) indicates the presence of xyloglucan and AGPs on the surface of border-like cells (BLCs). Roots are stained with Direct Red 23 (red). Images are representative. JIM8 recognised AGPs in (a) Atriplex vesicaria at 120 mM NaCl and (c) Atriplex nummularia at 70 mM NaCl. LM15 recognised xyloglucan in (b) Atriplex amnicola at 120 mM NaCl and (d) A. amnicola at 120 mM NaCl. Scale bars = 100 μm.

Root growth

Roots grown on glass beads were measured 7 days after sowing to determine whether the root length of young saltbush seedlings is affected by salinity. The only significant differences between root length at different salinity concentrations occurred in A. vesicaria (Fig. 6). A. vesicaria showed the greatest average root length at all salinity concentrations. At 230 mM NaCl, A. vesicaria had an average root length of 32.4 mm while A. amnicola and A. nummularia had average root lengths of 9.6 mm and 15.6 mm, respectively (P < 0.0001, Fig. 6a). A. amnicola and A. nummularia had consistent average root lengths across the range of salinity concentrations, while A. vesicaria peaked at 230 mM NaCl (Fig. 6a). Time to germination was not significantly different between species, with germination on glass beads within 7 days varying between 67% and 100% (data not shown).

Fig. 6.

(a) Root length and (b) root:shoot ratio in three saltbush species at five salinity concentrations (12, 70, 120, 230, and 310 mM NaCl) measured seven days after sowing. Lines are guide to sight only. Anum, Atriplex nummularia; Aamn, Atriplex amnicola; Aves, Atriplex vesicaria. Shown are observed means ± s.e. (n = 2–10). Significant differences in root length occur only in A. vesicaria at different salinity concentrations (in green). *P < 0.05; **P < 0.01.

Root:shoot ratio was not significantly different between salinity concentrations within each species (Fig. 6b). A. amnicola root:shoot ratio was significantly lower than the other two species (P < 0.0001, Fig. 6b). A. nummularia and A. vesicaria showed identical root:shoot ratios of 0.56 at 310 mM NaCl (Fig. 6b).

Saltbush species have closed-type root apical meristems that shed border-like cells

Here, we report that saltbush species have closed-type RAMs, as shown by cell organisation in multiple, well-defined layers (Fig. 1). Closed-type RAMs are characterised by clonally-distinct tiers of initials where cells have a defined lineage and can be traced back to their origin (Hamamoto et al. 2006). Kumpf and Nowack (2015) report two distinct different tissue types: (1) the central columella root cap; and (2) the peripheral lateral root cap. Columella cells are distal to the quiescent centre while lateral root cap cells are long, thin cells located on the periphery of the columella cells. As the columella and lateral root caps differentiate, they are displaced towards the root periphery. When they reach the root periphery, the oldest cells are exuviated into the rhizosphere as border cells or BLCs (del Campillo et al. 2004).

Closed-type RAMs have been observed in many families, including Brassicaceae with Arabidopsis thaliana (Hamamoto et al. 2006), Linaceae, Asteraceae, Lamiaceae and Poaceae (Groot et al. 2004; Rost 2011; Saito et al. 2019). Here, we extend the occurrence of closed-type RAMs to saltbush in the Amaranthaceae family and to the first halophyte. Notably, we report that Atriplex species release BLCs into the rhizosphere rather than border cells. Cells sloughed off from saltbush roots into the rhizosphere are arranged in organised layers that remain attached to each other when immersed in water and are therefore defined as border-like. These cells do not become dispersed individually, and thus cannot be classified as border cells, which separate once in contact with water (Driouich et al. 2010).

Saltbush roots are surrounded both by long, thin BLCs that detach from the epidermal layer in the maturation region of the root and short, fat BLCs derived from columella cells that form clumps around the root cap (Fig. 1). The distribution of BLCs around the root cap may play a role in stress tolerance by offering physical protection to the growing root region for the greatest period of time. The distribution pattern is different to Arabidopsis, where BLCs exist only as clumps of fat cells around the root cap (Durand et al. 2009).

Saltbush root mucilage contains minimal monosaccharides

In this experiment, we used two chromatography techniques to ascertain the monosaccharides present in saltbush root mucilage, as well as microscopy techniques to visualise polysaccharides in and around the root cap. Monosaccharide analysis using RP-HPLC and HPAEC-PAD both indicated that low molecular weight species are present in the saltbush root mucilage. Glucose is present in all three species in the greatest concentrations at low to medium salinity, as well as low concentrations of arabinose and galactose (Table 1). Fructose was also determined to be present in saltbush root mucilage (Fig. S2).

There are limited reports of root mucilage in other halophytic species. In one study with seedlings of Kosteletzkya virginica, rhamnose, galactose, and glucose were found in the greatest quantities compared to fucose, arabinose, xylose, and mannose (Ghanem et al. 2010). In this study, 100 mM NaCl salinity slightly decreased the glucose content in root mucilage. In our study, A. nummularia showed a similar result for glucose, with a decrease in glucose at 120 mM NaCl compared to 12 mM NaCl.

However, only around 1.5% of the total mucilage extract was identified as monosaccharides (Fig. 3), which is exceptionally low compared to cereal crops such as maize and wheat, where root mucilage is comprised of 80–100% monosaccharides (Ray et al. 1988; Knee et al. 2001; Amicucci et al. 2019). This unexpected difference may represent a tactic for the conservation of carbon output via limiting polysaccharides within root mucilage, assisting plants in surviving in harsh environments (Zickenrott et al. 2016). Salinity concentrations between 70 and 230 mM NaCl, when monosaccharides were present in the root mucilage in the greatest quantities, might potentially represent the most optimal range for the growth of saltbush, allowing carbon release into the soil for advantageous but not essential processes such as attracting beneficial soil microorganisms, repelling pathogenic microorganisms, or root lubrication (McCully 1999; Minz et al. 2013; Baetz and Martinoia 2014; Sasse et al. 2018). Conversely, stress responses may be triggered in saltbush grown at very high external salt concentrations, resulting in the allocation of carbon elsewhere in the plant, such as for the production of osmoprotectants or salt bladders (Jamil et al. 2011).

As monosaccharides are present in minimal quantities in the root mucilage, there must be a substantial quantity of other compounds present that are not detectable using RP-HPLC and HPAEC-PAD methods (Fig. 3b). These could include salt as well as a range of proteins, organic acids, phenolics, and AGPs (Bais et al. 2006; Jones et al. 2009; Kawasaki et al. 2018). The nature and roles of this collection of compounds produced by plant roots as a measure to mitigate a number of stresses has recently been reviewed (Chai and Schachtman 2022). This review highlights the complexity of root exudates and indicates that, as discriminatory analytical methods become better, more compounds are likely to be identified. We measured salt in the root mucilage using flame photometry and found that large quantities are present, although the increased root mucilage quantities produced under high salinity are not attributed to greater amounts of salt (Figs 2, 3). The greatest quantity of salt in the root mucilage occurs at 12–120 mM NaCl, and is reduced at higher salinity levels. Importantly, this shows that the increased exudate weights at higher salinity (Fig. 2) cannot be attributed only to higher salt contents, but must also include other compounds. At increased external NaCl concentrations, salt may be transported to epidermal bladder cells on leaves for storage, which reduces the salinity effects on the plant and decreases NaCl in root mucilage (Kiani-Pouya et al. 2017). This supports findings from Aslam et al. (1986) that suggest an accumulation of ions in the tissues of A. amnicola at high external NaCl concentrations. Additional studies that include a de-salting step prior to monosaccharide analysis would provide further clarity on the levels of salt and quantities of other as yet unidentified compounds present in the root mucilage, which are likely to belong to the list identified by previous researchers (Bais et al. 2006; Chai and Schachtman 2022).

Un-esterified homogalacturonan, xyloglucan, and arabinogalactan proteins are present on the surface of border-like cells

Overall, saltbush root mucilage contains only small amounts of monosaccharides, differing quantities of salt, and other unknown soluble components. To further investigate the unknown components present in root mucilage and around the root cap, we used microscopy techniques. In particular, we probed for pectic compounds, particularly homogalacturonan, because they have previously been found to increase in root mucilage under high salt treatments and play a role in cell adhesion around the root cap (Durand et al. 2009; Hayashi and Kaida 2011; Tenhaken 2015; Lutts et al. 2016). We discovered that un-esterified, or partially esterified, homogalacturonan is present on the surface of BLCs (Fig. 4) as well as xyloglucan and AGPs (Fig. 5). These compounds may glue BLCs to each other, as in Arabidopsis roots (Durand et al. 2009), since similar polysaccharides have previously been found on the surface of BLCs, including homogalacturonan and AGPs (Driouich et al. 2010; Kumar and Iyer-Pascuzzi 2020) with pectins surrounding border cells in pea (P. sativum) (Stephenson and Hawes 1994).

BLCs in saltbush remain attached to each other once released from the root surface, potentially due to the un-esterified homogalacturonan, which has been found to form gels in the presence of multivalent ions such as Ca²⁺. These Ca²⁺/pectate crosslinks hold adjacent pectate polymers together and may contribute to adhesion between cells (Willats et al. 2001; Proseus and Boyer 2012). Importantly, Arabidopsis mutants defective in homogalacturonan synthesis had altered BLC organisation (Durand et al. 2009), suggesting that homogalacturonan plays a key role in attachment between BLCs. Additionally, AGPs likely play a role in BLC adhesion, as the alteration of AGP biosynthesis in Arabidopsis results in a significant reduction in binding of BLCs to the root cap (Driouich et al. 2010).

As immersion immunofluorescence labelling only allows detection of epitopes at the surface of the tissue (Durand et al. 2009), the labelled compounds may be present either in the mucilage layer or be part of the cell walls, so further studies are needed to elucidate their location within or between cells, as well as their roles in adhesion. If they comprise an intrinsic part of the cell walls, they may not readily detach into the soluble fraction of the root mucilage that has been analysed here. The distribution of polysaccharides deserves further investigation, which may be particularly illuminating at the transmission electron microscope level. The expression patterns of the encoding genes could also be defined, particularly as roots elongate or are exposed to environmental stresses. Finally, while mucilage is secreted primarily by the root cap (Knee et al. 2001), it can also be released from root hairs and older root regions away from the root cap, so future studies should investigate polysaccharide labelling beyond the root cap alone. While the composition of polysaccharides in root mucilage is not yet well understood, different polysaccharides may play key roles in salinity tolerance, warranting further research in the future.

Increased salinity alters saltbush seedling root cap morphology but not root:shoot ratio

At 7 days after sowing, the root:shoot ratio of the three saltbush species did not differ significantly as salinity concentration increased (Fig. 6). This suggests that saltbush seedlings can survive without significant growth penalties at very high concentrations of salt, since 310 mM NaCl is 60% the salt concentration of seawater and up to seven times the amount required to seriously disrupt roots in cereal crops (Byrt et al. 2018). Root length was greater at 230 mM NaCl than 12 or 120 mM NaCl in A. vesicaria, which further supports this hypothesis. Germination time between species was not significantly different, but the number of roots measured (n = 2–10) depended on successful germination, which was variable. Future studies should investigate older saltbush seedling root lengths, ideally standardising root length measurements by time from germination, rather than time from sowing.

However, there was a noticeable effect of salinity on the saltbush root cap morphology. We found that in A. nummularia and A. vesicaria, root cap cell organisation becomes disorganised at high salinity levels of 310 mM NaCl (Figs 1 and 4). As the root cap is one of the first plant organs to come into physical contact with biotic or abiotic stresses (Kumar and Iyer-Pascuzzi 2020), these cells can undergo significant changes (Baranova et al. 2019), but research into the effect of salinity on saltbush root morphology has been limited. The reduction in cell organisation may result from a loss of control of cell division, potentially due to osmotic stress and inadequate rates of accumulation of osmotic solutes (Aslam et al. 1986). Alternatively, saltbush may have reduced root cap cell organisation at high salinity concentrations due to changes in pectin methylesterase (PME) activity (Pelloux et al. 2007). Wolf et al. (2012) suggest that a change in PME activity, which can occur due to an excess of external cations present in high salinity conditions, results in the loss of cell wall integrity and extensibility. PME may degrade homogalacturonan and other binding pectic compounds, resulting in reduced cell binding in the root cap. To test whether PME activity affects adhesion between BLCs, as well as the impact of salinity on pectin content in saltbush cells, we suggest future studies that inhibit PME in saltbush at different salinity concentrations. This has previously been investigated in Arabidopsis using catechin-derivative (-)-epi-gallocatechin gallate (EGCG) (Wolf et al. 2012).