Role of light and jasmonic acid signaling in regulating foliar phloem cell wall ingrowth development


Author for correspondence: William W. Adams III Tel: +1 303 492 2880 Fax: +1 303 492 8699 Email:


  • • Phloem cells adjacent to sieve elements can possess wall invaginations. The role of light and jasmonic acid signaling in wall ingrowth development was examined in pea companion cells (CCs), Arabidopsis thaliana phloem parenchyma cells (PCs), and in Senecio vulgaris (with ingrowths in both cell types).
  • • Features characterized included wall ingrowths (from electron microscopic images), foliar vein density and photosynthetic capacity.
  • • In Arabidopsis, wall ingrowths were bulky compared with finger-like invaginations in pea and S. vulgaris. Relative to low light (LL), wall invagination in both CCs and PCs was greater in high light (HL). Treatment with methyl jasmonate in LL had no effect on CCs, but increased PC wall ingrowths. LL-to-HL transfer resulted in significantly less wall ingrowth in the fad7-1 fad8-1 (jasmonate-deficient) Arabidopsis mutant relative to the wild type.
  • • These results suggest that chloroplast oxidative status, via chloroplast-derived jasmonates, may modulate phloem structure and function. While CC wall ingrowths facilitate phloem loading by expanding the membrane area available for active uptake, one can speculate that phloem PC ingrowths may have two potential roles: to increase the efflux of sugars and/or protons into the apoplast to augment phloem loading; and/or to protect the phloem against pathogens and/or insects.


Strategically situated plant cells can develop invaginations or reinforcements of their cell walls, which may serve to enhance solute transfer, defense, strengthening of tissues, etc. Cells at key points of solute exchange (typically at locations where a bottleneck in the transfer of substances might potentially occur) often develop secondary wall invaginations coupled with a magnified plasma membrane surface area. These transfer cells are thought to facilitate an increase in solute transport through an enhancement of the area for exchange and increased levels of plasma membrane-localized transport proteins (for a review of their structure, distribution within and among plants, and function see Gunning & Pate, 1969, 1974; Pate & Gunning, 1969, 1972; Offler et al., 2003). H+/sucrose symporters and ATPases are colocalized to the plasma membrane in several types of transfer cell (Kühn, 2003; Offler et al., 2003), and the density of membrane transport proteins and ATPases is greater along walls with ingrowths (Bouche-Pillon et al., 1994; Harrington et al., 1997; Tegeder et al., 1999; Bagnall et al., 2000; Sondergaard et al., 2004).

Transfer cells form in specific tissues during the appropriate developmental phase, but their formation can also be induced in response to a variety of abiotic and biotic factors, including increased solute availability, specific nutrient deficiencies, during establishment of a symbiotic relationship with mycorrhizal fungi or nitrogen-fixing bacteria, during infection by pathogenic or parasitic organisms, and possibly as a result of wounding by nematodes and insects (for review see Offler et al., 2003). In pea leaves, the extent of cell wall ingrowths in companion cell (CC) transfer cells was found to be responsive to growth light (greater in high than in low light), and could be altered by transfer from low to high light and vice versa, even in fully mature leaves (Henry & Steer, 1980; Wimmers & Turgeon, 1991; Amiard et al., 2005).

High light not only increases sucrose production and thus carbon export from leaves, but also has the potential to enhance the formation of reactive oxygen species (ROS) that, in turn, oxidize macromolecules, particularly polyunsaturated lipids (Montillet et al., 2004; Ledford & Niyogi, 2005; Roberts & Paul, 2006). Many derivatives of the peroxidation of polyunsaturated fatty acids are messenger molecules, collectively termed oxylipins. For example, peroxidation of α-linolenic acid, found in high concentration in chloroplast membranes, generates precursors to the plant hormone jasmonic acid (JA; Howe & Schilmiller, 2002; Devoto & Turner, 2005). Jasmonic acid increases expression of defense genes (Berger, 2002; Halitschke & Baldwin, 2003; Thaler et al., 2004; Devoto & Turner, 2005) and augments antioxidation capacity (Sasaki-Sekimoto et al., 2005; Wolucka et al., 2005). Furthermore, JA modulates several aspects of cell wall structure and signaling (Ellis et al., 2002a; Uppalapati et al., 2005). Offler et al. (2003) hypothesized that JA might be responsible for inducing transfer-cell formation and cell wall invagination.

The present study was undertaken to examine the role of light level and JA in modulating the level of transfer cell wall ingrowths in the phloem. Several plant species were chosen with distinctive transfer cell types within the phloem. Pea phloem is characterized by CC transfer cells (Henry & Steer, 1980; Wimmers & Turgeon, 1991; Amiard et al., 2005), Arabidopsis thaliana possesses phloem parenchyma cell (PC) transfer cells (Haritatos et al., 2000), and Senecio vulgaris possesses both CC and PC transfer cells in the phloem (Pate & Gunning, 1969; Gunning & Pate, 1974).

Materials and Methods

Plants and experimental treatments

Pisum sativum L. cv. Alaska (pea), Arabidopsis thaliana (L.) Heyn. (ecotype Columbia), and Senecio vulgaris L. were grown from seed in Fafard Canadian Growing Mix 2 (Conrad Fafard Inc., Agawam, MA, USA), receiving nutrients every other day and water regularly. In addition, Arabidopsis mutants deficient in oxylipin synthesis (the double mutant fad7-1 fad8-1; Falcone et al., 2004) or insensitive to JA signaling (jar1; Staswick et al., 1992) were also grown and characterized. All plants were grown either under low light (LL, 100 µmol photons m−2 s−1, fluorescent light bulbs, 10 h photoperiod for pea and S. vulgaris, 8 h photoperiod for A. thaliana to prevent flowering); or high light (HL, 1000 µmol photons m−2 s−1, 1000-W metal halide lamps, 14 h photoperiod for pea and S. vulgaris, 8 h photoperiod for A. thaliana to prevent flowering), all at 25°C day : 20°C night. Plants treated with methyl jasmonate (MeJA) were grown under LL and sprayed daily with a solution of 10 µm MeJA in water and 0.05% Tween 20 for 1 wk. Low-light control plants were sprayed daily with water and 0.05% Tween 20 for 1 wk. To check for any concentration dependency, another group of LL pea plants was subjected to the same treatment with 50 µm MeJA, yielding identical results (not shown).

Ultrastructural and anatomical characterization

Phloem companion cells and parenchyma cells were characterized from mature leaf samples fixed, sectioned, stained, and imaged as described by Amiard et al. (2005). The cross-sectional percentage increase in plasma membrane perimeter (Wimmers & Turgeon, 1991) or total area occupied by cell wall were determined using eeb viewer software (Amiard et al., 2005). The foliar density of minor loading veins was determined from leaves cleared, imaged, and quantified as described by Amiard et al. (2005).

Photosynthetic capacity

The capacity for photosynthetic oxygen evolution was determined in a leaf disc oxygen electrode (Model LD-2, equipped with an LS-2 halogen light source; Hansatech, King's Lynn, UK) under saturating CO2 (5%) and light (2425 µmol photons m−2 s−1) at 25°C (Delieu & Walker, 1981; Adams et al., 2002). Leaf area was determined using a portable leaf-area meter (AM100, Analytical Development Company Ltd, Hoddesdon, UK).

Statistical analyses

anova was performed followed by a Tukey–Kramer comparison for honestly significant differences among several means, or a Student's t-test was used to compare between two means (jmpin 3.2.1 software, SAS Institute Inc., Cary, NC, USA).


Companion cells

The minor vein CCs of pea and S. vulgaris leaves are transfer cells (Fig. 1a,c,d,f), with long, finger-like wall projections around the entire cell perimeter (Fig. 2b,h). These were classified as ‘type A’ transfer cells by Pate & Gunning (1969); Gunning & Pate (1974). When plants were grown in HL, wall invaginations were more extensive in both species compared with growth in LL (Figs 2a,b,g,h, 3a,c). Methyl jasmonate treatment in LL did not affect wall ingrowth development in the CCs of either species (Figs 2a,c,g,i, 3a,c).

Figure 1.

Arrangement of sieve elements (SE), companion cells (CC) and phloem parenchyma cells (PC) in pea (a,d); Arabidopsis thaliana (b,e); and Senecio vulgaris (c,f) minor veins. Transmission electron microscopic images of transverse sections of the phloem showing companion cells, phloem parenchyma cells and sieve elements. Positions of wall ingrowths are roughly indicated in white (d–f).

Figure 2.

Electron micrographic images of cell wall invaginations in transfer cells (companion cells (CC) in pea (a–c) and Senecio (g–i); phloem parenchyma cells (PC) in Arabidopsis (d–f) and Senecio (j–l)) from transverse sections of foliar minor veins of pea, Arabidopsis and Senecio grown in low light (LL; a,d,g,j); high light (HL; b,e,h,k); or in LL and treated with methyl jasmonate (MeJA; c,f,i,l) for 1 wk. SE, sieve element.

Figure 3.

Increases in plasma membrane length caused by wall ingrowths in minor vein companion cells (CC) of (a) pea; (c) Senecio and phloem parenchyma cells (PC) of (b) Arabidopsis; (d) Senecio from plants grown in low light (LL), high light (HL), or in LL and treated with methyl jasmonate (MeJA) for 1 wk. Values represent percentage of actual measured wall length relative to hypothetical wall length of cells without ingrowths, as assessed from electron micrographic images (Fig. 2). Means ± SE; significantly different means among each species/cell type are indicated by lower-case letters at P < 0.05. Cells from three to six leaves were selected from each treatment and species, and two to 20 cells were analysed from each leaf. Sample sizes thus ranged from n = 27–87 cells.

The CCs in the minor veins of Arabidopsis leaves are not transfer cells, the walls are smooth (Fig. 1b,e). Neither growth in HL nor MeJA treatment in LL induced wall ingrowths in Arabidopsis CCs (not shown).

Phloem parenchyma cells

Phloem PCs with wall ingrowths were classified as ‘type B’ transfer cells by Pate & Gunning (1969); Gunning & Pate (1974). Unlike the ingrowths of the CCs already described, these invaginations were restricted to a specific region of the walls. In Senecio, they are finger-like and abut both sieve elements and CCs (Figs 1c,f, 2j–l). In Arabidopsis they are bulky rather than finger-like, and are primarily adjacent to sieve elements (Figs 1b,e, 2d–f).

In both species, phloem PC wall invaginations were much more extensive in HL than in LL plants (Figs 2d,e,j,k, 3b,d). Treatment of LL plants with exogenous MeJA also increased PC wall invagination significantly, especially in Arabidopsis where MeJA induced ingrowths to the full extent seen in high light (Figs 2d–f,j,l, 3b,d). Considering the different appearance of wall ingrowths in Arabidopsis phloem PCs in comparison with the other transfer cells considered here, the ratio of the increase in cell wall area to the increase of plasma membrane perimeter length caused by wall ingrowths in transverse sections was determined for HL-grown and MeJA-treated, LL-grown leaves (Fig. 4). This analysis demonstrates that wall thickness increases in the PCs of Arabidopsis more than in the CCs of pea and Senecio and the phloem PCs of Senecio.

Figure 4.

Ratio of wall area to total inner wall perimeter length increase caused by wall ingrowths in minor vein companion cells (CC) of high light (HL) pea, and phloem parenchyma cells (PC) of Arabidopsis and Senecio, from plants grown in HL or in low light (LL) and treated with methyl jasmonate (MeJA) for 1 wk. Means ± SE; significantly different means among all species/cell types are indicated by lower-case letters at P < 0.05. For additional details, see Fig. 3 legend.

The phloem PCs in the veins of pea plants are not transfer cells (Fig. 1a). Neither HL nor exogenous MeJA induced wall projections or wall thickening in the phloem PCs of pea (data not shown).

Wall ingrowths in Arabidopsis mutants deficient in oxylipin synthesis or insensitive to JA signaling

Because of the high level of variability, no significant differences were found in the level of phloem PC wall ingrowths among wild-type Arabidopsis, the jasmonic acid signaling-insensitive (jar1) mutant, and the oxylipin-deficient mutant (double mutant fad7-1 fad8-1) grown in HL (data not shown). However, 7 d following transfer from LL to HL (a condition resulting in greater oxidative stress than continuous growth under HL), the fad7-1 fad8-1, oxylipin-deficient mutant exhibited significantly fewer phloem PC wall ingrowths than wild-type Arabidopsis, with the jar1 mutant exhibiting an intermediate level of phloem PC wall ingrowths that was not significantly different from the wild type or the fad7-1 fad8-1 double mutant (Fig. 5).

Figure 5.

Increases in plasma membrane length caused by wall ingrowths in minor vein phloem parenchyma cells of wild-type (WT) Arabidopsis, the jar-1 mutant of Arabidopsis, and the fad7-1 fad8-1 double mutant of Arabidopsis from plants grown in low light (100 µmol photons m−2 s−1) then transferred to high light (1000 µmol photons m−2 s−1) for 1 wk. Values represent percentage of actual measured wall length relative to hypothetical wall length of cells without ingrowths, as assessed from electron micrographic images. Means ± SE. The fad7-1 fad8-1 double mutant exhibited a significantly lesser percentage increase in wall ingrowths compared with the wild type at P < 0.05 (Student's t-test), whereas the jar1-1 mutant was not significantly different from either the wild type or fad7-1 fad8-1 double mutant. Cells from five leaves were selected from each Arabidopsis line, and two to six cells were analysed from each leaf. Sample sizes thus ranged from n = 16–19 cells.

Photosynthetic capacity and phloem loading

The photosynthetic capacity of mature leaves grown under high light (n = 3) was approximately twofold higher in Senecio (74 ± 5 µmol O2 m−2 s−1) than in Arabidopsis (33 ± 3 µmol O2 m−2 s−1) and pea (44 ± 4 µmol O2 m−2 s−1). This suggests that Senecio leaves export considerably more carbohydrate per unit leaf area than the other two species. Phloem loading and export capacity are limited by structural features such as vein density and CCs per sieve element (Wimmers & Turgeon, 1991; Amiard et al., 2005). However, neither vein density (3.9 ± 0.3 mm vein length mm−2 leaf area in Senecio, compared with 3.1 ± 0.4 in Arabidopsis and 5.3 ± 0.3 in pea, n = 3), nor the number of CCs per sieve element (2.3 ± 0.7 in both Arabidopsis and Senecio vs 1.7 ± 0.5 in pea; n = 42, 80 and 38 sieve elements, respectively), explains the high apparent transport capacity of Senecio leaves. On the other hand, the sum of all Senecio phloem cells with cell wall ingrowths (CCs + PCs) per sieve element was about twofold higher (3.8 ± 0.8, n = 80 sieve elements) than that of CCs alone, and thus correlated with the higher photosynthetic capacity.


The dual nature of high light

Judging from (1) their different ultrastructural features; (2) their differential response to HL and exogenous MeJA; and (3) the different capacities for photosynthesis in HL-acclimated leaves among the three species, it is likely that the phloem transfer cells with high light-enhanced wall ingrowths in the species studied here are responsive to multiple signals and serve more than one function. High light can simultaneously be a stimulus for increased growth and upregulated photosynthesis, with a resulting enhanced need for carbon export (Amiard et al., 2005), while at the same time inducing oxidative stress with increased production of ROS and oxidative messengers (Montillet et al., 2004; Ledford & Niyogi, 2005; Roberts & Paul, 2006). Thus high light may be expected to have a positive impact on foliar carbon export through enhancement of the capacity for sugar transport, but might also be expected to provide signals that result in enhanced resistance to stress.

A role of transfer cells in enhanced phloem loading and photosynthetic acclimation

Based on studies of plants with different loading mechanisms (Wimmers & Turgeon, 1991; Amiard et al., 2005), phloem-loading capacity can be a limiting factor in the amount of carbon exported from leaves. When grown in HL, Arabidopsis possessed the lowest photosynthetic capacity of the three species studied. This relatively low photosynthetic capacity may be related to the fact that Arabidopsis also had the lowest vein density among the three species and, in contrast to the other two species, its relatively smaller CCs (Fig. 1) did not possess membrane surface-enhancing wall ingrowths. Pea grown in HL had an intermediate capacity for photosynthetic oxygen evolution, the highest vein density, and its CCs had extensive wall invaginations, the length of which correlates well with transport capacity (Wimmers & Turgeon, 1991). If the sole function of the CC wall ingrowths is to facilitate sucrose transport, then one might expect them to respond to messengers that signal increased carbon flow from photosynthesis, and not to messengers (such as jasmonates formed via lipid peroxidation) that signal oxidative stress under excess light.

In comparison with the other two species, Senecio grown in HL had a much higher photosynthetic capacity than pea. While its vein density was lower than that of pea, its wall ingrowths in the CCs appear to be more extensive (Fig. 2b,h). In addition, it is possible that the elaborate wall ingrowths in the phloem parenchyma cells of Senecio assist in loading. Presuming that sucrose effluxes from phloem parenchyma cells into the apoplast as a first step in loading (Turgeon & Ayre, 2005), these wall ingrowths could potentially overcome a rate limitation in the efflux step. The orientation of cell wall invaginations in the phloem PCs of Senecio suggests such a function. Furthermore, the active loading of sucrose into CCs and/or sieve elements by H+/sucrose symporters requires the pumping of protons into the apoplast, and the increased plasma-membrane area afforded by cell wall ingrowths is often found to have increased levels of ATPases for proton pumping (Bouche-Pillon et al., 1994; Bagnall et al., 2000; Offler et al., 2003; Sondergaard et al., 2004). The phloem PC wall ingrowths of Senecio, all facing towards the adjacent CC transfer cells and sieve elements (Figs 1c,f, 2j–l), are thus also ideally positioned for contributing to the proton gradient driving the active loading of sucrose into the CC transfer cells and/or sieve elements of Senecio. If one assumes that Senecio phloem PCs, with their extensive wall ingrowths, also have a role in the facilitation of phloem loading, the relationship between total plasma-membrane area and photosynthetic capacity among the three species is greatly improved. More information is needed on membrane areas and transporter densities to evaluate this hypothesis.

Responses of phloem PC wall ingrowths to light and exogenous MeJA, and their possible role in defense

High or excess light can stimulate the transfer of excitation energy from chlorophyll to oxygen, resulting in the generation of singlet excited oxygen, as well as electron transfer to oxygen and the production of superoxide (Adams et al., 2004; Montillet et al., 2004; Ledford & Niyogi, 2005). Such an increased production of ROS presumably activates lipoxygenase (Maccarrone et al., 1997), thus inducing the peroxidation of α-linolenic acid as an early step in the biosynthesis of JA. A series of additional steps taking place in the chloroplast envelope and then in peroxisomes (Chrispeels et al., 1999; Devoto & Turner, 2005) ultimately leads to the formation of JA. A chloroplast-localized lipoxygenase has been shown to be responsible for the accumulation of JA in response to wounding (Bell et al., 1995).

In fact, JA synthesis takes place in the vascular bundles (Stenzel et al., 2003; Howe, 2004) and the enzymes for synthesis of JA are localized to the plastids of the CCs and sieve elements (Hause et al., 2003). Synthesis of cell wall components (Takahashi et al., 1995) and cell walls (Montague, 1997) can be stimulated by JA, and incorporation of additional cell wall components into phloem PCs could increase the degree of wall invagination. The significantly lower level of phloem PC wall ingrowths in an oxylipin-deficient mutant of Arabidopsis (fad7-1 fad8-1) compared with the wild type, in response to transfer from LL to HL, provides evidence in support of a role of oxylipins in stimulating such cell wall invaginations. Although not significant, the lower level of phloem PC wall ingrowths in the JA-insensitive mutant (jar1-1) compared with wild-type Arabidopsis in an identical transfer experiment is also noteworthy. However, future research should address the possibility that a jasmonate other than JA may be involved.

Moreover, this is the first report that treatment of LL plants with exogenous MeJA increased cell wall invaginations exclusively in the phloem PCs, with no effect on CCs. This differential effect is consistent with the role of these two cell types during pathogen attack. While CC transfer cells in species such as pea and bean are typically avoided by pathogenic attack, phloem PCs are often the primary target of plant pathogens in the phloem (Ding et al., 1995, 1998; Heller & Gierth, 2001; Zhou et al., 2002), and are often the first cells to respond to such attacks with defense signals and wall strengthening (Rioux & Quellette, 1991; Kpemoua et al., 1996; Narváez-Vásquez & Ryan, 2004; Narváez-Vásquez et al., 2005). Enhanced cell wall growth has been interpreted as providing a physical barrier to the spread of pathogens (Rioux & Quellette, 1991; Kpemoua et al., 1996; Vogel et al., 2004), but cell wall modification clearly also functions in defense signaling (Nishimura et al., 2003). Furthermore, aphids, which often transmit pathogens while feeding from phloem (Taliansky et al., 2003), activate the JA-signaling pathway (Walling, 2000; Moran & Thompson, 2001; de Ilarduya et al., 2003; Zhu-Salzman et al., 2004) and stimulate cell wall modification (Shinoda, 1993; Divol et al., 2005). Other phloem-feeding insects also stimulate plants to modify their cell walls, including fortification (Park et al., 2006; Thompson & Goggin, 2006). Given that the phloem is a prime target for pathogenic attack, and can be utilized by pathogens as a pathway for distribution throughout the plant (Gilbertson & Lucas, 1996; Lucas & Wolf, 1999; Waigman et al., 2004; Scholthof, 2005), a strong defensive response of the phloem to pathogens and aphids would not be unexpected.

The localization of Arabidopsis phloem PC ingrowths to those portions of the wall located adjacent to sieve elements is thus ideal as a potential barrier to the movement of a pathogen into sieve elements, thereby denying the pathogen access to the rest of the plant (Gilbertson & Lucas, 1996; Waigman et al., 2004; Scholthof, 2005). Such localization, coupled with the bulky nature of PC wall ingrowths in Arabidopsis (maximizing wall thickness with a lesser impact on plasma-membrane area), as well as their responsiveness to MeJA treatment, are all consistent with a role in defense. Wall ingrowths in Senecio phloem PCs may also help protect against pathogen spread, given that they respond to exogenous MeJA in LL plants, although to a lesser extent than they do to HL.

Further tests of signaling responsible for phloem PC wall ingrowths and the latter's possible role in defense

The results presented here suggest at least two roles for phloem PC wall ingrowths in the minor vein phloem of leaves. While CC ingrowths (which responded only to growth light environment) have long been thought to increase carbohydrate export capacity, phloem PCs have been viewed simply as passage cells for the movement of sugars from the mesophyll tissue to the CCs and sieve elements (Kühn, 2003; Turgeon & Ayre, 2005). However, the highly elaborate cell wall invaginations in the phloem PCs of species such as Senecio may actually augment the loading of phloem through increased sucrose efflux into the apoplast and/or increased pumping of protons into the apoplast to stimulate the activity of the H+/sucrose symporters. The latter could be evaluated further through immunolocalization using an antibody against ATPase. On the other hand, physically bulky cell wall ingrowths in phloem PCs, with a lesser impact on membrane length, such as those in Arabidopsis, may play an important role in impeding systemic pathogen spread. The induction of phloem PC wall ingrowths by high light or the oxylipin MeJA suggest that moderate shifts in chloroplast redox state (particularly chloroplast lipid peroxidation level) can modulate the production of oxylipin messengers and, in turn, phloem structure and function. This could be evaluated further using various mutants with an altered chloroplast redox state (e.g. zeaxanthin-deficient mutants, npq1; zeaxanthin-overproducing mutants, npq2; or PsbS-deficient mutants, npq4; Niyogi et al., 1998; Li et al., 2000). Furthermore, it will be interesting to characterize the responsiveness of CC transfer cells and phloem PCs with wall ingrowths to light and JA. Cell wall ingrowths with unusual morphology (Pate & Gunning, 1969), and species that possess both CC and phloem PC transfer cells (Pate & Gunning, 1969; Estelita & Marinho, 1995), should prove especially interesting in this regard.

Other Arabidopsis mutants may also prove useful in dissecting the signaling events responsible for increased levels of phloem PC wall ingrowths. While it is well established that signaling by JA enhances resistance to pathogens and insect pests (Berger, 2002; Halitschke & Baldwin, 2003; Thaler et al., 2004; Devoto & Turner, 2005), experiments utilizing an Arabidopsis mutant (opr3) lacking JA suggest that 12-oxo-phytodienoic acid (OPDA, an intermediate in the synthesis of JA) also acts as a signal to upregulate defense in response to such attack (Stintzi et al., 2001). Can OPDA provide the signal to increase phloem PC wall ingrowths, or will such a mutant lacking JA fail to respond to high light with increased phloem PC wall ingrowths?

In addition to mutants that have low or no JA, there are some that possess elevated levels. The level of JA was 60 times greater in the tissues of the cev1 mutant of Arabidopsis compared with the wild type (Ellis et al., 2002a). Compared with wild-type Arabidopsis, would such a mutant be expected to have enhanced cell wall ingrowths in foliar phloem PCs? This is difficult to predict, and might not be the case, for several reasons. (1) The gene that is altered in the cev1 mutant codes for a cellulose synthase (Ellis et al., 2002a), that is, the cell walls of this mutant accumulate less cellulose and one might therefore hypothesize that it would have a lower level of cell wall ingrowths. (2) The cev1 mutation was expressed strongly in flower and root tissues but much less in stems and leaves (Ellis et al., 2002a), and one might predict a relatively minor impact on the phloem PCs of the leaves. (3) The ‘inhibition of cellulose synthesis activates (both) JA- and ethylene-dependent stress responses’ (Ellis et al., 2002a). It is interesting that treatment of tomato roots with the precursor to ethylene, 1-aminocyclopropane-1-carboxylic acid, increased the number of rhizodermal cells with wall ingrowths (a feature important to the acquisition of nutrients from the soil when they are in short supply; Schikora & Schmidt, 2002). On the other hand, inhibitors of ethylene synthesis and of ethylene action did not affect such wall ingrowths, and the ethylene-insensitive Never-ripe mutant of tomato actually possessed a higher number of rhizodermal cells with wall ingrowths, leading Schikora & Schmidt (2002) to conclude that ‘ethylene responsiveness played no critical role in the differentiation of transfer cells and that the transduction of signals ultimately leading to their formation was independent of the ethylene signaling cascade.’

Defense against pathogen and insect attack is enhanced by multiple means through JA signaling (Berger, 2002; Halitschke & Baldwin, 2003; Thaler et al., 2004; Devoto & Turner, 2005). JA-mediated resistance has already been well established for Arabidopsis via JA mutants (Berger, 2002; Ellis et al., 2002b; Bostock, 2005). For instance, the JA signaling-insensitive coi1 mutants have been shown to be more susceptible to infection by fungi and bacteria, as well as to herbivory by aphids (Thomma et al., 1998; Norman-Setterblad et al., 2000; Ellis et al., 2002b). On the other hand, the JA- and ethylene-overproducing mutant cev1 was much less susceptible to pathogenic fungi and aphids (Ellis et al., 2002b). The results shown here are consistent with a role of PC wall ingrowths in this defense, but it will not be a trivial matter to differentiate the role of the simple physical barrier provided by cell wall ingrowths from wall-associated signaling events and upregulation of biochemical defense.


We thank Dr Tom Giddings for his support and guidance with electron microscopy, and Dr Volker Ebbert for technical support. The financial support of the National Science Foundation (awards IBN-0235351 to W.W.A. and B.D.-A., and IBN-0235709 to R.T.) is gratefully acknowledged.