J Plant Res (2015) 128:49–61 DOI 10.1007/s10265-014-0676-5

Plasmodesmata: Function and Diversity in Plant Intercellular Communication

JPR SYMPOSIUM

Diffusion or bulk flow: how plasmodesmata facilitate pre‑phloem transport of assimilates Alexander Schulz 

Received: 10 August 2014 / Accepted: 14 October 2014 / Published online: 17 December 2014 © The Botanical Society of Japan and Springer Japan 2014

Abstract  Assimilates synthesized in the mesophyll of mature leaves move along the pre-phloem transport pathway to the bundle sheath of the minor veins from which they are loaded into the phloem. The present review discusses the most probable driving force(s) for the prephloem pathway, diffusion down the concentration gradient or bulk flow along a pressure gradient. The driving force seems to depend on the mode of phloem loading. In a majority of plant species phloem loading is a thermodynamically active process, involving the activity of membrane transporters in the sieve-element companion cell complex. Since assimilate movement includes an apoplasmic step, this mode is called apoplasmic loading. Well established is also the polymer-trap loading mode, where the phloem-transport sugars are raffinose-family oligomers in herbaceous plants. Also this mode depends on the investment of energy, here for sugar oligomerization, and leads to a high sugar accumulation in the phloem, even though the phloem is not symplasmically isolated, but well coupled by plasmodesmata (PD). Hence the mode polymer-trap mode is also designated active symplasmic loading. For woody angiosperms and gymnosperms an alternate loading mode is currently matter of discussion, called passive symplasmic loading. Based on the limited material available, this review compares the different loading modes and suggests that diffusion is the driving force in apoplasmic loaders, while bulk flow plays an increasing role in plants having a continuous symplasmic pathway from mesophyll to sieve elements. Crucial for the driving force is the question

A. Schulz (*)  Department of Plant and Environmental Sciences, University of Copenhagen, Thorvaldsensvej 40, 1871 Frederiksberg C, Denmark e-mail: [email protected]

where water enters the pre-phloem pathway. Surprisingly, the role of PD in water movement has not been addressed so far appropriately. Modeling of assimilate and water fluxes indicates that in symplasmic loaders a considerable part of water flux happens through the PD between bundle sheath and phloem. Keywords  Active phloem loading · Bulk flow · Diffusion · Passive phloem loading · Plasmodesmata · Symplasmic transport

Introduction Plasmodesmata (PD) offer an energetically favorable transport pathway for assimilates and link assimilate-producing tissues, such as the photosynthetically active leaf mesophyll, with assimilate-consuming cells and tissues. The only alternative to PD-confined transport is the cell-to cell transport via efflux and uptake carriers in the plasma membrane of neighboring cells, involving two membrane steps and requiring expenditure of energy. For PD-confined transport, diffusion is efficient as driving force over short distances, since it is independent of energy-consuming processes. PD can accommodate bidirectional diffusional transport of solutes. PD can as well be a pathway for bulk flow. Bulk flow takes the solute (assimilates) together with its solvent (water). Here much higher transport rates can be achieved and long distances can be bridged, but transport gets unidirectional and energy is needed to fuel the development of a pressure gradient over the connected cell chain. The text book example for bulk flow is phloem transport, as already suggested by (Münch 1930), where the assimilates-conducting sieve elements are connected by modified PD, called sieve pores. In the majority of plants,

13

50

a pressure gradient is established in the phloem by thermodynamically active phloem loading in source leaves which leads to osmotic water uptake. The present review discusses the question, how PD facilitate the pre-phloem transport of assimilates, i.e. the transport from mesophyll to the phloem in source leaves, considering also the movement of water. The distribution of PD between the cells is not random, but predetermined by tissue patterning during primary growth. Therefore, the first chapter covers the role of tissue development for the formation of preferential symplasmic pathways within a given tissue and across tissue borders. Implications of development on PD distribution and symplasmic transport pathways In the plant embryo, plasmodesmata (PD) with a large size exclusion limit (SEL) link all cells with one another as evidenced by the work done in Zambryski’s lab. From heart stage onwards and again when the seedling germinates, growth is accompanied by tissue differentiation and concomitant expansion growth. Differentiation and expansion growth have two consequences for symplasmic transport: (1) The SEL is reduced as seen in GFP and tracer transport studies (Kim et al. 2005; Kim and Zambryski 2005), and (2) cell expansion leads to a reduction in PD per cell wall area (Kollmann and Glockmann 1999). Even if the total number of PD in the interface between two given cells stays constant, the reduction in the SEL means a reduction of symplasmic exchange rate between these cells. Both processes continue to happen in the shoot and root apical meristems, and can in a way also be observed in the vascular cambium in plants that undergo secondary growth (Ehlers and van Bel 2010; Fuchs et al. 2011). As a general result, the maxima of symplasmic transport rates should be highest in embryos and at the apical and vascular meristems, and PD transport should be most promiscuous with respect to the molecular size in just divided cells, but more and more controlled during cell differentiation. However, these general rules ignore the tissue polarity developing in expanding organs. During expansion growth directionality in symplasmic transport develops, since the cross walls of elongating cells experience much less stretch of the cell wall, and thus dilution of PD per area, than the lateral walls. Directionality of symplasmic transport Symplasmic transport gets a strong axial directionality, i.e. a preference for transport along the longitudinal axis of a cell, if the PD dilution is not compensated for. There is strong evidence that some compensation may happen in stretched cell walls by branching of existing PD with the

13

J Plant Res (2015) 128:49–61

effect that each original PD gives rise to two or multiple PD arms towards the protoplasts of the connected cells, while it is dilated to form a median cavity in the middle lamella of the cell wall (Burch-Smith et al. 2011; Faulkner et al. 2008a, b; Kollmann and Glockmann 1999; Krull 1960). In addition to the secondary modification of existing PD in stretched cell walls, it cannot be excluded that new PD develop across stretched cell wall parts, again indicated in TEM micrographs by PD branching (Kollmann and Glockmann 1999). However, both secondary modification and new development of PD seem to be tightly controlled by the physiological need in transport. This does not only include increase, but also reduction of the symplasmic transport capacity up to a total isolation as seen for mature guard cells (Palevitz and Hepler 1985). How this need is translated into activation of the genes necessary for either formation of PD or their blockage is as yet totally unknown. Directionality in symplasmic transport is pronounced in the radially symmetrical organs like shoot and root, where parallel cell tiers undergo the sequence division-differentiation-expansion asynchronously. This introduces shearing forces in the lateral walls where existing PD might be disrupted, while the end walls do not experience such forces to the same extend. Directionality is realised with maximal effect in the phloem sieve tube system of angiosperms where the end wall PD transform and dilate into sieve pores, while the lateral wall are virtually devoid of PD to other cells than the companion cells (CC). CC as sister cells keep a high density of primary PD that get secondarily modified under sieve element-companion cell complex (SECC complex) differentiation, involving multiple branching towards the CC and sieve pore formation on the SE side. Symplasmic transport pathways across particular tissue interfaces The establishment of epidermis from protoderm, ground parenchyma (i.e. cortex and pith in stems, mesophyll in leaves) from ground meristem and vascular tissue from procambial strands, each with different division patterns and expansion growth rates, marks two tissue interfaces in the ground parenchyma where expansion-growth caused PD dilution has to be compensated for: towards the epidermis and towards the vascular tissue. The ground parenchyma layer abutting the vascular tissue is called endodermis in roots, starch sheath in stems and bundle sheath in leaves. Since all three are initiated by SHR and SCR or homologous transcription factors, the endodermis, starch sheath and bundle sheath can be considered to be homologous tissue layers (Cui et al. 2014; Slewinski et al. 2012; WysockaDiller et al. 2000).

J Plant Res (2015) 128:49–61

Branched PD, i.e. secondary and secondarily modified PD develop between mesophyll and epidermis (Ehlers and Kollmann 2001; Steinberg and Kollmann 1994). The epidermis depends on assimilates from the photosynthetically active mesophyll in plants where the pavement cells do not have chloroplasts; PD in that interface are obviously open for assimilates as indicated by formation of large starch grains in sucrose transporter antisense plants (Schulz et al. 1998). Branched PD are also observed in the interface between bundle sheath cells (BSC) and CC of symplasmic phloem loaders, and in a very specialized form in all interfaces between the endodermis-like bundle sheath of gymnosperm leaves and the bridge of living cells connecting to the sieve elements (Liesche et al. 2011). Branching of PD can thus be taken as indication for the need of a high symplasmic exchange rate across a given cell wall interface. In context of the present review this is obvious when considering the interface between companion cells and bundle sheath cells with virtually no PD in apoplasmic loaders like Vicia, and numerous PD between the same cell types in symplasmic loaders like Populus or Cucurbita; see also the diversity of minor veins in Asteridae (Batashev et al. 2013). The authors base their review on a comprehensive compilation of ultrastructure data, including the occurrence of PD in CC, and classify structural types and subtypes of minor veins with significance for phloem loading. Pre‑phloem transport of assimilates The pre-phloem transport of assimilates starts in the mesophyll and ends in cells bordering CC in angiosperms and Strasburger cells (Str) in gymnosperms. In most plant families the bundle sheath is the target cell of this transport, and the subsequent step of assimilate transport would be the loading of assimilates into the SECC complex. Whether or not phloem parenchyma cells (PP), where existing (Batashev et al. 2013), are interleaved in phloem loading is unclear as yet; localization of the SWEET efflux transporter in phloem parenchyma cells suggests such a role (Chen et al. 2012), as does the development of transfer-cell wall ingrowths towards the SECC complex (Amiard et al. 2007; Haritatos et al. 2000b). Accordingly the pre-phloem transport pathway consists of a few cells only. There are typically not more than three to six cells from a mesophyll cell (MC) to the nearest minor vein (Haritatos et al. 2000b). The pathway thus comprises typically four or five cells in angiosperms, i.e. MC– MC-MC-BSC, or MC–MC-MC-BSC-PP (but see below for gymnosperms) (Table 1). Leaf architecture and distribution of PD are compatible with the view that the pre-phloem transport of assimilates predominantly takes a symplasmic pathway (but see

51

discussion in (Schobert et al. 2000)). The mesophyll is characterized by a large, air-filled intercellular space which occupies 35–50 % of the leaf volume (Lohaus et al. 2001) and is continuous between the stomata, mesophyll cells (MC) and bundle sheath. Only the internal vein tissue is devoid of any intercellular space. The contact areas between MC are comparatively small and contain numerous PD, as do those between MC and BSC. An important prerequisite for symplasmic transport across these interfaces is that their PD are functional. Cell coupling was quantified with photo-activated fluorescein as tracer which has a comparable molecular size as sucrose (Liesche and Schulz 2012). Both, MC–MC and MC-BSC showed a high cell-coupling factor in different plant species (Fig. 1) supporting a purely symplasmic pre-phloem transport. Considering the small contact areas in the loosely arranged leaf tissue, apoplasmic steps in this transport, i.e. the release of assimilates from one MC into the apoplast and active uptake into the neighboring MC by plasma membrane transporters are improbable as being energetically inefficient. However, vacuolar membrane transporters might play a role in pre-phloem transport in adjusting the cytosolic assimilate concentration in the individual cells of the pathway (Fu et al. 2011). The driving force for pre‑phloem transport For the question of diffusion versus bulk flow or both, the concentration of assimilates along the pre-phloem pathway and the osmotic movement of water into the cells of that pathway are relevant. In a dynamic equilibrium between photosynthesis and export through the phloem, diffusion will tend to level out the difference in cytosolic sugar concentration between mesophyll and BSC. PD linking these cells would accordingly allow sugar diffusion towards the BSC. For bulk flow to occur, water, accessing the apoplast around the MC, has to be able to pass their plasma membranes to build up an osmotic pressure in the mesophyll. In consequence, sugars could reach the BSC not only by diffusion but also by bulk flow, when driven by an osmotic pressure differential between mesophyll and bundle sheath. The driving force for phloem transport is an osmotically generated pressure flow. The question to be treated in the following paragraphs is whether the driving force for the pre-phloem transport is pressure flow as well, diffusion, or a combination of both. Different modes of phloem loading are realized in angiosperms and typically discussed for sugars only (for amino acids see (Tegeder 2014)). In actively loading plant species, sugars are enriched in the SECC complex either by membrane transporters (apoplasmic loader) or by the polymer-trap mechanism (symplasmic loader). In both cases the cytosolic sugar concentration would be highest in the SECC complex (Fu et al. 2011; Liesche and Schulz 2013; Lucas et al. 2013; Slewinski et al. 2013). In

13

52

J Plant Res (2015) 128:49–61

Table 1  Key characters of the different phloem loading modes and derived driving force for pre-phloem transport Loading mode

Angiosperms Active apoplasmic (TC)

Minimal cell no of pre-phloem pathway Cell type inside BSC

Gymnosperms Active apoplasmic (CC)

4–5: MC–MC-MC-BSC (PP)

Active symplasmic (IC)

Passive symplasmic

Passive symplasmic

4–5: MC–MC-MCBSC (-PP) Intermediary-type CC (IC)

4–5: MC–MC-MCBSC (-PP) Ordinary CC

BSC-IC: asymmetrically branched PD act as filter

Apparently no: many branched PD

7–8: MC–MC- BSCTP-TP-Str-Stra Transfusion parenchyma (TP) Apparently no: many branched PD

Transfer-type CC (TC)

Ordinary CC

Symplasmic domain borderb

BSC-TC: virtually no PD

BSC-CC: few PD

Apoplasmic domain border

BSC can be suberized in monocotsc and/ or contain Casparian strips (Plantago)d

?

?

BSC: Casparian strip

MC–MC 0.372

MC–MC 0.492

MC–MC 0.990

N.D.

MC-BSC 0.239

MC-BSC 0.446

MC-BSC 0.995

BSC-TC 0.007

BSC-CC 0.017

BSC-IC 0.194

MC–MC 0.591 MC-BSC 0.346 BSC-TP 0.463 TP–TP 0.711 TP-Str 0.606 Str–Str 0.271

RFO BSCk

Sucroseg or polyolh ?

Sucrose TPl

?

?

BSC, TP, phloemn

High Low MC-BSC: Diffusion

Low High

Low High MC-BSC: Diffusion

Cell coupling factor (CCF) along the prephloem pathwayb Phloem transport sugar Plasma membrane transporters Plasma membrane aquaporins

Sucrose or polyolf i

TC

j

CC

BSC, CC, SEm

Hydraulic conductanceo High (woody species low) Low (woody species high) Leaf osmolalityo Derived driving force for Diffusion pre-phloem transport

Bulk flow

BSC-SE: Diffusion– bulk flow

TP-SE: Bulk flow

a

  Anatomical data (Liesche and Schulz 2012)

b

  PD structure and connectivity (Beebe and Russin 1999; Davidson et al. 2011)

c

  Specific structure of monocot BSC (Botha et al. 2007)

d

  Casparain strips in Plantago BSC (Thompson and Schulz 1999)

e

  CCF data (Liesche and Schulz 2012)

f

  Polyol in Plantago (Reidel et al. 2009)

g

  Salix: (Turgeon and Medville 1998)

h

  Malus: (Reidel et al. 2009)

i

  Localization of plasma membrane transporters (Andriunas et al. 2013; Tan et al. 2008)

j

  Localization of plasma membrane transporters (Pommerrenig et al. 2007; Schmitt et al. 2008; Wright et al. 2003)

k

  Suggested localization of amino acid transporters (Tegeder 2014)

l

  Localization of plasma membrane transporters discussed (Canny 1993)

m

  Aquaporins or their expression listed in (Fraysse et al. 2005; Heinen et al. 2009; Lee et al. 2009)

n

  Aquaporin localization (Laur and Hacke 2014)

o

  Data from (Becker et al. 1999; Fu et al. 2011; Liesche and Schulz 2012; Öner-Sieben and Lohaus 2014)

contrast, passive loaders have the highest sugar concentration in the mesophyll and comparatively lower concentrations in the SECC complex (Rennie and Turgeon 2009; Zhang et al. 2014). Since the mode of phloem loading has implications for the mechanism of the preceding assimilate transport from MC to BSC, the question of the driving force

13

in pre-phloem transport will be discussed for each of the established phloem loading modes separately. For a general overview of phloem loading, the reader is referred to (Lucas et al. 2013; Patrick 2012; Slewinski et al. 2013). Some plant species are capable of more than one loading mode; these special cases are left out in the present review.

J Plant Res (2015) 128:49–61

53

Fig. 1  Cell coupling along the pre-phloem pathway in leaves, derived from transport of photo-activated fluorescein. The colorcoded coupling factors (values see respective row in Table 1) are shown on schematic drawings of collection phloem in longitudinal (top) and cross-sectional (bottom) views. The color bar below depicts low to high coupling factors in spectral transitions from red to green, respectively. Whereas the minor vein phloem of the active apoplasmic

loaders Vicia faba (a) and Nicotiana tabacum (b) is symplasmically isolated (red interface), all cells along the pre-phloem pathway are well coupled in the active symplasmic loader Cucurbita maxima (c) and passive symplasmic loader Pinus sylvestris (d) (yellow-to-green interface). From (Liesche and Schulz 2012). © American Society of Plant Biologists

Pre‑phloem transport in apoplasmic loaders (Fig. 2a, b)

sugar concentration in this complex attracts water from the apoplast and build up the high osmotic pressure, necessary to drives phloem transport. In order to work out, this loading mode requires a sufficient number of aquaporins in the SECC complex plasma membrane, in addition to the sugar uptake transporters (Table 1). A contribution of bulk flow to pre-phloem transport in apoplasmic loaders seems improbable. The sugar concentration in MC would certainly allow osmotic water uptake and increase of the cell turgor. But also BSC would be target for some osmotic water uptake, provided that the distribution of aquaporins is similar in the plasma membrane of MC and BSC. Therefore, a pressure differential between MC and BSC cannot develop, even less so, since the scarcity of PD towards the SECC complex would not be compatible with a pressure release into the phloem. The situation seems to be different in Ricinus cotyledons that have been the model case of apoplasmic phloem loading, extensively studied in the 80s and 90s. Here, a moderate number of PD was seen between BSC, PP and CC. Despite inhibition of sucrose uptake with p-chloro–mercuri–benzene–sulfonic acid (PCMBS), phloem exudation of radioactive sucrose continued, contributing to approximately 50 % of the control (Orlich et al. 1998). The rate of exudation was indicative for a bulk flow developing in the mesophyll of the cotyledons and continuing through the phloem to the hypocotyl.

In apoplasmically loading plant genera, PD are virtually absent, or occur in low frequency only between BSC and SECC complex. In particular minor veins of plants that are characterized by transfer-type companion cells like Vicia faba (see Fig. 2a) (Schulz 1998) have an extremely low PD frequency in this interface [but see (Batashev et al. 2013)]. The uptake of sugars from the apoplast by membrane transporters would indeed be most efficient without any functional PD at that interface. However, PD functionality was tested in Vicia and Nicotiana with a fluorescent tracer of the molecular size of sucrose. Nicotiana that has a low PD density between BSC and CC (Fig. 2b) showed a low, but significant cell coupling factor (Table 1) (Liesche and Schulz 2012). Therefore, these PD seem to be important for other functions than loading, such as long-distance signaling (Gilbertson et al. 2007; van Bel et al. 2011). The possible backflow of assimilates into BSC and, thus, a futile sucrose loading-reflow cycle might be negligible in relation to the high uptake rates achieved by membrane transporters. The most probable driving force for the pre-phloem transport of assimilates in apoplasmic loaders is pure diffusion. Sugars reaching the BSC (and phloem parenchyma, where existing) will be released to the apoplast via SWEET transporters (Chen et al. 2012), from which uptake transporters will load the SECC complex. High

13

54

13

J Plant Res (2015) 128:49–61

J Plant Res (2015) 128:49–61

◂ Fig. 2  Pre-phloem pathways of active and passive phloem load-

ers with the suggested driving forces. a and b Driving force for the pre-phloem sucrose (S) transport in active apoplasmic loaders with transfer-type (TC) or ordinary companion cells (CC) is diffusion (stippled arrows). Bulk flow starts in the companion cells (dashed arrow), the plasma membrane of which contains sucrose transporters (red dots) and aquaporins (blue dots). Plasma membrane efflux carriers are shown as beige-shaded dots in the inner BSC membrane. Expected sucrose concentrations in the different cells are indicated by the font size of S. c Active symplasmic loaders translocate raffinose and stachyose (St), synthesized in their intermediary-type companion cells (IC). These oligomers are too large to pass the specific plasmodesmata (PD) between IC and BSC which have more branches on the IC side. High solute potential in the IC leads to water uptake which enters from the BSC through the specific plasmodesmata. Sucrose enters the IC by diffusion and/or bulk flow along with the water (dashed arrow). d Passive symplasmically loading woody angiosperms show a steadily decreasing cytosolic sucrose concentration (S). Sucrose goes by gradually increasing bulk flow together with water that enters the mesophyll cells through aquaporins (blue dots). e Also gymnosperms load sucrose passively. Two apoplasmic domains with high water potential (dark blue shaded apoplast inside the BSC) and low potential (light blue shaded apoplast outside the BSC) is separated by a Casparian strip (dark red shaded in BSC wall). Pre-phloem transport is by diffusion from MC to BSC (stippled arrow) and by bulk flow (dashed arrow) from the BSC via the 2–3 transfusion parenchyma (TP) and 3 Strasburger cells (Str) into the sieve elements (SE). TT = dead transfusion tracheids that are part of the xylem system

Pre‑phloem transport in active symplasmic loaders (Fig. 2c) In active symplasmic loaders, sucrose produced by photosynthesis in the mesophyll can enter the intermediarytype companion cells (IC) (Schulz 1998) symplasmically. The contact area between BSC and intermediary cell contains PD fields with numerous PD that are more strongly branched on the IC side than on the BSC side. The PD orifices on the intermediary-cell side are so small that they cannot be resolved by transmission electron microscopy. The polymer trap hypothesis, proposed and refined by Robert Turgeon in the early 1990s (Haritatos et al. 1996; Turgeon 1996; Turgeon and Beebe 1991; Turgeon et al. 1993; Turgeon and Gowan 1990) links the ultrastructure of these PD with the fact that the respective plant species mainly transport raffinose family oligosaccharides (RFO; i.e.. the tri-, tetra- and pentasaccharide raffinose, stachyose and/or verbascose, respectively) in the phloem (Fig. 2c). According to the hypothesis, the PD fields act as filters that allow the disaccharide sucrose to pass, but reject the larger RFO sugars which thus accumulate in the IC. In consequence, the high total sugar concentration in the SEIC complex leads to osmotic water uptake and high osmotic pressure in this complex, similar to that in apoplasmic loaders. Depending on the rate of RFO synthesis in IC, the concentration of its precursor sucrose will be low in this cell type, allowing the establishment of a sucrose downhill gradient

55

from mesophyll all way into the IC. There is in addition the possibility of adjustment of the cytosolic sucrose concentration through transfer into the small IC vacuoles, which are typical for this cell type (Batashev et al. 2013; Turgeon et al. 1975). Driving force for sucrose transport from MC to the phloem could accordingly be pure diffusion. The polymer-trap hypothesis is well supported by corroborative experiments: plasmolysis experiments and leaf disk uptake studies support the high total sugar concentration in the SEIC complex (Haritatos et al. 1996; Turgeon and Hepler 1989), galactinol synthase necessary for formation of RFO sugars is expressed and localised in IC (Haritatos et al. 2000a; Holthaus and Schmitz 1991; Volk et al. 2003), and the cell-coupling factor between mesophyll, BSC and the SEIC complex shows that the PD on this symplasmic route are functional (Fig. 1c) (Liesche and Schulz 2012). Moreover, modelling of the nano-architecture of PD and the hydrodynamic radius of the relevant sugars indicated that the permeability of intermediary-cell PD would be able to discriminate sucrose efficiently from RFO, if their cut-off is close to the hydrodynamic radius of stachyose (Liesche and Schulz 2013). However, the published estimates of sugar flux across the relevant interfaces (Schmitz et al. 1987), observed number of PD (Volk et al. 1996) and the calculated PD conductance (Liesche and Schulz 2013) did not fall into the same size order, casting doubts on the assumptions made and/or the compatibility of the data achieved in different symplasmic loader species (Cucumis melo, Coleus blumei and Cucurbita pepo, respectively). Therefore, we are presently working on measuring the PD conductance across the BSC-intermediary cell interface in Cucumis melo, where the most accurate data on PD number and sugar flux are available. When comparing the two thermodynamically active phloem loading modes, the unifying element is that the SECC/IC complex holds the highest sugar concentration. One might accordingly assume that osmotic water uptake for pressure driven phloem transport also in active symplasmic loaders occurs across the SECC complex plasma membrane [see also Fig. 13B2 in (Lucas et al. 2013)]. However, this assumption ignores the fact that the numerous PD in the BSC-IC interface do not only form a pathway for sucrose, but also for water. In a theoretical model we have tried to evaluate the sucrose, RFO and water fluxes via PD and across the plasma membranes of BSC, intermediary cells and sieve elements (Dölger et al. 2014). It appears that the PD-fields between BSC and IC can act similar to a semipermeable membrane, allowing water, together with sucrose, to enter the IC symplasmically, where a high osmotic pressure is formed. According to this modelling, there is no need for water to pass the SEIC complex plasma-membrane via aquaporins in active symplasmic loaders. Depending on different assumptions, the driving

13

56

J Plant Res (2015) 128:49–61

as the above described active loaders. In gymnosperms, the vascular tissue is composed differently and other vascular cell types are interleaved between BSC and SE, consisting of the living transfusion parenchyma cells, and outer, middle and inner Strasburger cells which flank the phloem (Table  1) (Liesche et al. 2011). Because of the considerably different vascular anatomy and water relations, the driving force for pre-phloem transport in passive symplasmic loaders will be treated separately for angiosperms and gymnosperms. Passive symplasmically loading angiosperms (Fig. 2d)

Fig. 3  Reproduction of Münch’s Figure 22, which corresponds to the passive loading mode in woody angiosperms (Fig. 2d). Assimilates in a photosynthetically active leaf (double-feathered arrows) flow from palisade parenchyma (Pal) through PD via spongy parenchyma (Schw), bundle sheath (Gsch), companion cell (Gel) finally into to a sieve tube (S). Water (simple arrows) moves the opposite way from the xylem (H) across cell walls into the bundle sheath and continues up to the palisade parenchyma where it evaporates: To a smaller part it enters also the phloem directly (thin arrows). From (Münch 1930) p. 153

force for sucrose transport from BSC into the IC might be a combination of diffusion and bulk flow, while that from the IC into the SE is bulk flow in any case (Dölger et al. 2014). In a careful morphometric analysis, Voitsekhoskaja and coworkers also arrived at the conclusion that symplasmic sucrose transport into the phloem could be driven by mass flow in two species with high PD coupling between BSC and IC (Voitsekhovskaja et al. 2006). Pre‑phloem transport in passive symplasmic loaders A majority of woody species seems to fall under yet another, alternate loading category, where the highest sugar concentrations are measured in the mesophyll and lower concentrations in the phloem (Fu et al. 2011). As the active symplasmic loaders, the passive ones have a high density of branched PD between BSC and vascular parenchymatous cells, but in contrast to those the branching pattern is symmetrical (Batashev et al. 2013; Liesche et al. 2011; Rennie and Turgeon 2009). In angiosperms, the best characterised family for this loading type are the Salicaceae, which have broad leaves with minor veins and direct contact between BSC and CC

13

Besides the open vein configuration, key character of passive symplasmic loading angiosperms seems to be the high global sugar concentration in the leaf, as measured as leaf sap osmolality (Davidson et al. 2011; Fu et al. 2011; Turgeon and Medville 1998). Leaf disk experiments with radioactive sucrose in the medium do not show the venation pattern in woody symplasmic loaders, including Quercus rubra, as is the case in all active loaders, apoplasmic and symplasmic ones (Rennie and Turgeon 2009). It would be appropriate to call this loading mode “Münch loading mode”, since Münch’s original hypothesis assigned the mesophyll to have the high sugar concentration able to drive the bulk flow in the phloem [see Fig. 3 reproduced from Fig. 22 in (Münch 1930)]. It is quite a coincidence that Münch’s understanding of the mechanism of long-distance transport seems to meet the situation in a majority of woody plants (Davidson et al. 2011), considering that he was professor of Forest Botany and motivated to his work by the challenge of long-distance assimilate transport in large trees. The passive mode of phloem loading was argued in a recent paper for Q. robur, based on a thorough analysis of the leaf sugar compartmentation and phloem sap, collected from severed aphid stylets. Here the phloem sap had a higher osmolality than the whole leaf sap (Öner-Sieben and Lohaus 2014). Moreover, the authors were able to isolate a SUT1-homologous sucrose transporter by RACE-PCR, otherwise typical for apoplasmic phloem loaders. They interpret their data as indicating that Q. robur is an apoplasmic phloem loader. Rennie and Turgeon’s data of Q. rubra lists that leaf discs did not show any accumulation of radioactive sucrose [Fig. 2 in (Rennie and Turgeon 2009)]. In order to solve the discrepancy in the two Quercus species it would be good to get phloem sap data from Q. rubra and leaf disc data from Q. robur. It appears improbable that the two species from the same genus follow a different loading strategy. If present, sucrose transporters in the plasma membrane of the phloem elements of passive symplasmic loaders do not have a significant role in loading, as recently was shown with Populus transformants: expression of an apoplasmic yeast invertase under the 35S ubiquitous or a

J Plant Res (2015) 128:49–61

CC- specific promoter does not seem to have any impact on phloem transport, since the transformants do not show a growth phenotype, in contrast to the active apoplasmic loader they compared it with (Zhang et al. 2014). Yeast invertase splits sucrose into fructose and glucose, which cannot be loaded since they are not substrates for phloem sucrose transporters. The driving force for the sucrose transport from mesophyll into the SECC complex was discussed to be diffusion, supported by the high sugar concentration in the mesophyll, a continuous symplasmic pathway and a lower concentration in the phloem sap (Fu et al. 2011), provided that the MC vacuoles contain other osmotic active compounds which are able to balance the cytosolic assimilate concentration. In a recent paper from Turgeon’s group, the question of the driving force for pre-phloem sucrose transport is formulated a bit more cautious: “If poplar loads symplastically, but does not have a trapping mechanism, it is reasonable to postulate that the process is passive in the sense that sucrose diffuses or flows through plasmodesmata into the phloem without a concentration step” (italics by me) (Zhang et al. 2014). It is obvious that the question of the driving force for this transport, bulk flow or diffusion, is not finally solved for symplasmic loaders. Already Münch discussed this problem thoroughly in his famous monograph without coming to a conclusion. His conclusion was that this can only be solved when the distribution of turgor in the different leaf cell types is assessed [p. 152 (Münch 1930)], a challenge which is not resolved yet, 85 years later. Assuming that bulk flow, instead of diffusion, is the driving force, osmotic water uptake would be strongest in the mesophyll cells and decline in the cells along the prephloem route, considering the falling sucrose concentration from mesophyll to the SECC complex. This can only happen when the respective cells contain plasma-membrane aquaporins. Unfortunately data about the cellular localization of aquaporins in leaves seem to be limited to active phloem loaders (see Table 1). In poplar, expression of aquaporin genes was shown in the stem cambial regions, including young phloem and xylem; in leaves a subset of aquaporins showed a light-dependent, circadian expression pattern concomitant with an increase in hydraulic conductance (Ben Baaziz et al. 2012; Lopez et al. 2013). Since mesophyll needs water for photosynthetic water splitting, it is reasonable to assume that MC plasma membranes do contain aquaporins. Provided this is the case, a pressure differential between mesophyll and SECC, necessary for bulk flow, could develop, if the PD along the pathway show a constant or increasing conductance (Fig. 2d). The bulk flow will in any case experience increasing conductance across the last cell of this pathway, the CC, which on the import side from the BSC have branched PD, and PD-sieve pore contacts on the export side, i.e. towards the

57

SE [comparable to the situation in active symplasmic loaders; cf. (Dölger et al. 2014)]. Passive symplasmically loading gymnosperms (Fig. 2e) Most research on the structure and function of the phloem in gymnosperms has been done with conifers in contrast to the living fossils Cycads, Gnetophytes and Gingko (Behnke 1990): However, the unifying composition of the gymnosperm phloem with sieve cells and Strasburger cells (Str) and presence of a transfusion tissue inside the bundle sheath (Liesche et al. 2011) might allow a generalization from conifers to all gymnosperms. The conifers have needle-like leaves that indicate their xerophytic adaptation. Not only is the surface available for transpiration much smaller than that of a blade-shaped leaf, but also are the stomatal complexes sunken and have their guard cells at the subepidermal level. There are no vein classes, but the vascular tissue runs centrally in the needle forming one or two strands and is surrounded by a endodermis-like bundle sheath, which forms an apoplasmic barrier by Casparian strips in the radial cell walls (Liesche et al. 2011). The most dramatic difference to the angiosperm-leaf vasculature is the presence of an elaborate system of dead transfusion tracheids (TT) linking the xylem to the bundle sheath and, in between these tracheids, living transfusion parenchyma cells (TP) that bridge the distance from the BSC to the phloem. Strasburger cells have a similar function as the angiosperm companion cells, but contrast in not being sister cells to the sieve elements, i.e. they derive from unrelated cell tiers (Schulz 1990). In the needle, they flank the axial phloem in three layers, where the outer one contacts the TP and the inner one the sieve elements (Carde 1973). All the living cells are well-connected with branched PD (Liesche et al. 2011) and show a high cell coupling factor (Table 1). Multiply branched PD occur in conspicuous, dome-shaped wall thickenings between the different Strasburger cells and indicate a high physiological demand of connectivity during needle development (Glockmann and Kollmann 1996). Even though the Gneotphyte Gnetum gnemon does not have needles, but broad leaves with several vein orders, the composition of each of the vein classes is similar and contains—in addition to phloem and xylem sensu strictu—TT and TP within an endodermis-like bundle sheath (Liesche et al. 2011). With at least seven cells the pre-phloem pathway for assimilates spans a much wider distance in terms of cell number than that of angiosperms. As the passively loading angiosperm trees, gymnosperms leaves have a high leaf osmolality and low hydraulic conductance (Table 1) (Liesche and Schulz 2012; Öner-Sieben and Lohaus 2014). The particular architecture of conifer needles and the localization of aquaporins in BSC and TP (Laur and Hacke

13

58

2014) are suggestive for a pre-phloem transport that is divided in two sections: (1) diffusional transport from MC to BSC and (2) bulk flow from TP to Str and further into the sieve element system (Fig. 2e). The bundle sheath has a pivotal role for both, water transport out of the vascular tissue and assimilates transport into the phloem, since it controls what may enter and leave the vascular tissue as the root endodermis does. The apoplasmic barrier by the Casparian strip and properties of the radial BSC cell walls (Liesche et al. 2011) imply that mineral nutrients arriving from the axial xylem via TT have to cross the inner BSC plasma membrane to be delivered to MC and epidermis. At the bundle sheath, the needle apoplast is accordingly divided into two domains: the vascular apoplast with a high water potential and the mesophyll apoplast with a low water potential. The endodermis-like character of the bundle sheath can be considered as one of the xerophytic adaptation of gymnosperms. Dye tracer studies done by Canny in pine needles confirm the barrier function of the bundle sheath and the close connection of the TT with the xylem (Canny 1993). Assimilates will obviously move in opposite direction to xylem-derived mineral nutrients. Considering the low water potential in the mesophyll apoplast and low abundance of aquaporins (Laur and Hacke 2014), the high sucrose concentration in the photosynthetically active MC is not expected to lead to a strong osmotic water uptake and concomitant development of a pressure potential. Sucrose will accordingly diffuse into the BSC, rather than going by bulk flow. Mineral ions taken up by the inner side of BSC can diffuse through the MC-BSC in the opposite direction. An inwards bulk flow of assimilates would render this impossible. Inside the bundle sheath, the situation changes dramatically: now a high sucrose concentration in the TP cytosol meets a high water potential in the apoplast, where xylemdelivered water arrives via TT. Transfusion tracheids are dead at maturity and belong to the apoplast, as demonstrated by Canny’s tracer experiments. The steep gradient across the TP plasma membrane will lead to strong osmotic water uptake which is facilitated by a high abundance of aquaporins in the TP and BSC plasma membranes (Laur and Hacke 2014). From TP to Str and into the axial sieve elements a pressure differential, necessary for driving a bulk flow, can develop because the sucrose concentration is decreasing by phloem export, and the downstream PD coupling is high (Table 1). Interestingly, TP and BSC are able to take up radioactive aspartate from the apoplast, which then accumulate in Str of the phloem, showing the TP have a function in xylem-phloem transfer of solutes (Canny 1993). In conclusion, the osmotically generated pressure flow of sucrose seems to start already in the first TP inside the

13

J Plant Res (2015) 128:49–61

bundle sheath of needles to continue in the axial phloem. In the intact phloem system, the tubular ER present in the prephloem PD, and the ER complexes found in sieve areas of all gymnosperms do not seem to increase resistance to bulk flow of sucrose and water decisively (Schulz 1992, 2005), but lead to a phloem transport rate that in general is lower than in angiosperm trees (Jensen et al. 2012).

Concluding remarks According to the present review, the pre-phloem transport of assimilates is driven by diffusion in apoplasmic loaders. The main route of osmotic water entry to build up pressure in the phloem is across the plasma membrane of the SETC or SECC complex, as also evidenced by the localization of aquaporins in that interface (Table 1). In minor veins, CC are generally larger than SE and well connected with PD-sieve pore contacts. Considering the membrane surface available for transporter proteins and aquaporins in TC—but also ordinary CC—the osmotically-driven pressure flow might already start already in CC (Fig. 2a, b). There is some uncertainty with respect to the role of the few PD found between BSC (or PC) and CC. The expected large turgor difference between these cells would be sufficient to lead to a pressure-generated PD closure as discussed by Oparka and coworker (Oparka and Prior 1992). Only if sucrose membrane transporter activity, and thus apoplasmic loading, are downregulated, and the turgor in the SECC complex declines, assimilates might be able to enter the phloem symplasmically. PCMBS experiments in Ricinus seedlings have shown that this can happen as bulk flow (Orlich et al. 1998). In contrast to apoplasmic loading, diffusion is not the only component in pre-phloem transport in active symplasmic loaders. Bulk flow seems to contribute to a certain degree already from BSC onwards. According to the polymer-trap hypothesis, the specific PD connections between BSC and IC filter sucrose in favor for larger oligosaccharides. Neglected up to now was however the fact that these PD offer a predominant route for water into the IC. Previously, water was assumed to be taken up across the plasma membrane of the ICSE complex. In contrast, our calculations strongly indicate that there is no need for water uptake across this membrane and that aquaporins, if present here at all, do not play a major role in phloem transport (Dölger et al. 2014). PD transport from BSC to IC can be speculated to be a mixture of bulk-flow and water-sucrose co-diffusion (Fig. 2c). The passive symplasmic phloem loading mode is least understood. A majority of woody plants have been assigned to this mode, based on convincing leaf disc experiments, the distribution of PD and a combination of high leaf sap

J Plant Res (2015) 128:49–61

osmolality and low hydraulic conductance (Table 1). Based on the limited data available, our working hypothesis for passive angiosperm loaders is that the driving force for prephloem transport is bulk flow. Accordingly, osmotic water uptake happens already in the mesophyll and along the prephloem pathway. Connectivity between the cells involved seems high; fluorescence-loss in photoactivation experiments indicated that the PD seen in electron microscopy are indeed functional (personal communication, Helle J. Martens, University of Copenhagen). A high turgor pressure in the mesophyll can gradually be released towards the minor veins via these PD. When entering CC, the PD-pore contacts, and subsequently the phloem sieve plates, offer high conductance, so that bulk flow from its slow start will be accelerated to the typical speed of phloem translocation (Fig. 2d). The situation is different in the gymnosperms which also are defined as passive symplasmic loaders. Two domains in the apoplast, the vascular one with a high water potential, and the mesophyll one with a low water potential dictate the driving force in pre-phloem transport to be biphasic: first diffusion from mesophyll to bundle sheath, and then bulk flow from TP via Str to SE (Fig. 2e). The localization of aquaporins to BSC and TP support this hypothesis. Interestingly, expression of aquaporins is regulated allowing e.g., embolism repair of stem xylem (Laur and Hacke 2014). It is conceivable that phloem transport in passive symplasmic loaders is regulated not only by the gating but also by the expression level of aquaporins. The above working hypotheses have to be tested with complementary molecular and physiological approaches. In the last two decades the localization and regulation of assimilate transporters and efflux carriers in plasma membrane and tonoplast have been in focus. In order to solve the questions of passive phloem loading the localization and regulation of aquaporins in plasma membrane and tonoplast of MC, BSC and PC are now highly relevant. For the specific turgor in individual cells, sucrose-splitting enzyme activities have also to be accounted for, since they can fine-tune the specific cell turgor. We are aiming at achieving direct evidence for the suggested bulk flow in the prephloem pathway of passive angiosperm loaders by pursuing photoactivation and bleaching experiments [see (Liesche and Schulz 2012; Pina et al. 2009)]. In whole-leaf experiments, the xenobiotic tracer would move from a BSC both upstream and downstream, if diffusion is the only driving force. If there is a bulk-flow component in addition, spread of the tracer will be preferentially downstream, i.e. towards the SECC complex. Acknowledgments  This work was supported by the Danish Council for Independent Research—Natural Sciences Grant 12-126055. AS thanks Tomas Bohr (Technical University of Denmark), Johannes

59 Liesche and Helle J. Martens (University of Copenhagen) for fruitful discussions and inspiring collaboration.

References Amiard V, Demmig-Adams B, Mueh KE, Turgeon R, Combs AF, Adams WW (2007) Role of light and jasmonic acid signaling in regulating foliar phloem cell wall ingrowth development. New Phytol 173:722–731. doi:10.1111/j.1469-8137.2006.01954.x Andriunas FA, Zhang HM, Xia X, Patrick JW, Offler CE (2013) Intersection of transfer cells with phloem biology-broad evolutionary trends, function, and induction. Frontiers in plant science 4. doi: 10.3389/Fpls.2013.00221 Batashev DR, Pakhomova MV, Razumovskaya AV, Voitsekhovskaja OV, Gamalei YV (2013) Cytology of the minor-vein phloem in 320 species from the subclass Asteridae suggests a high diversity of phloem-loading modes. Frontiers in plant science 4. doi:10. 3389/Fpls.2013.00312 Becker P, Tyree MT, Tsuda M (1999) Hydraulic conductances of angiosperms versus conifers: similar transport sufficiency at the whole-plant level. Tree Physiol 19:445–452 Beebe DU, Russin WA (1999) Plasmodesmata in the phloem-loading pathway. In: van Bel AJE, Van Kesteren WP (eds) Plasmodesmata. Springer Berlin, pp 261–293. doi:10.1007/978-3-642-60035-7_15 Behnke HD (1990) Cycads and gnetophytes. In: Behnke HD, Sjolund RD (eds) sieve elements. Springer Berlin, pp 89–101. doi:10.1007/978-3-642-74445-7_5 Ben Baaziz K, Lopez D, Rabot A, Combes D, Gousset A, Bouzid S, Cochard H, Sakr S, Venisse JS (2012) Light-mediated K-leaf induction and contribution of both the PIP1 s and PIP2 s aquaporins in five tree species: walnut (Juglans regia) case study. Tree Physiol 32:423–434. doi:10.1093/treephys/tps022 Botha CEJ, Cross RHM, Liu L (2007) Comparative Structure of Specialised Monocotyledonous Leaf Blade Plasmodesmata. In: Plasmodesmata. Blackwell, pp 73–89. doi:10.1002/9780470988572.ch4 Burch-Smith TM, Stonebloom S, Xu M, Zambryski PC (2011) Plasmodesmata during development: re-examination of the importance of primary, secondary, and branched plasmodesmata structure versus function. Protoplasma 248:61–74. doi:10.1007/s00709-010-0252-3 Canny MJ (1993) Transfusion tissue of pine needles as a site of retrieval of solutes from the transpiration stream. New Phytol 123:227–232 Carde JP (1973) Transfer tissue (strasburger cells) in needles of maritime pine (Pinus pinaster Ait).1. Histology and ultrastructural study of grown up tissue. J Microsc Oxf 17:65–88 Chen LQ, Qu XQ, Hou BH, Sosso D, Osorio S, Fernie AR, Frommer WB (2012) Sucrose efflux mediated by sweet proteins as a key step for phloem transport. Science 335:207–211. doi:10.1126/ science.1213351 Cui H, Kong D, Liu X, Hao Y (2014) SCARECROW, SCR-LIKE 23 and SHORT-ROOT control bundle sheath cell fate and function in Arabidopsis thaliana. Plant J 78:319–327. doi:10.1111/tpj.12470 Davidson A, Keller F, Turgeon R (2011) Phloem loading, plant growth form, and climate. Protoplasma 248:153–163. doi:10.1007/ s00709-010-0240-7 Dölger J, Rademaker H, Liesche J, Schulz A, Bohr T (2014) Diffusion and bulk flow in phloem loading: a theoretical analysis of the polymer trap mechanism for sugar transport in plants. Phys Rev E 90:042704 Ehlers K, Kollmann R (2001) Primary and secondary plasmodesmata: structure, origin, and functioning. Protoplasma 216:1–30. doi:10. 1007/Bf02680127 Ehlers K, van Bel AJE (2010) Dynamics of plasmodesmal connectivity in successive interfaces of the cambial zone. Planta 231:371– 385. doi:10.1007/s00425-009-1046-8

13

60 Faulkner C, Akman O, Bell K, Jeffree C, Oparka K (2008a) Peeking into pit fields - A new model of secondary plasmodesmata formation. Comp Biochem Phys A 150:S140–S141. doi:10.1016/j.cbpa. 2008.04.347 Faulkner C, Akman OE, Bell K, Jeffree C, Oparka K (2008b) Peeking into pit fields: a multiple twinning model of secondary plasmodesmata formation in tobacco. Plant Cell 20:1504–1518. doi: 10.1105/tpc.107.056903 Fraysse LC, Wells B, McCann MC, Kjellbom P (2005) Specific plasma membrane aquaporins of the PIP1 subfamily are expressed in sieve elements and guard cells. Biol Cell 97:519–534 Fu QS, Cheng LL, Guo YD, Turgeon R (2011) Phloem loading strategies and water relations in trees and herbaceous plants. Plant Physiol 157:1518–1527. doi:10.1104/pp.111.184820 Fuchs M, van Bel AJ, Ehlers K (2011) Do symplasmic networks in cambial zones correspond with secondary growth patterns? Protoplasma 248:141–151. doi:10.1007/s00709-010-0208-7 Gilbertson RL, Rojas MR, Lucas WJ (2007) Plasmodesmata and phloem: conduits for local and long-distance signaling. In: Plasmodesmata. Blackwell, pp 162–187. doi:10.1002/9780470988572.ch8 Glockmann C, Kollmann R (1996) Structure and development of cell connections in the phloem of Metasequoia glyptostroboides needles.1. Ultrastructural aspects of modified primary plasmodesmata in Strasburger cells. Protoplasma 193:191–203. doi:10. 1007/Bf01276645 Haritatos E, Keller F, Turgeon R (1996) Raffinose oligosaccharide concentrations measured in individual cell and tissue types in Cucumis melo L leaves: implications for phloem loading. Planta 198:614–622 Haritatos E, Ayre BG, Turgeon R (2000a) Identification of phloem involved in assimilate loading in leaves by the activity of the galactinol synthase promoter. Plant Physiol 123:929–937. doi:10.1104/Pp.123.3.929 Haritatos E, Medville R, Turgeon R (2000b) Minor vein structure and sugar transport in Arabidopsis thaliana. Planta 211:105–111. doi:10.1007/s004250000268 Heinen RB, Ye Q, Chaumont F (2009) Role of aquaporins in leaf physiology. J Exp Bot 60:2971–2985. doi:10.1093/Jxb/Erp171 Holthaus U, Schmitz K (1991) Distribution and immunolocalization of stachyose synthase in Cucumis melo L. Planta 185:479–486 Jensen KH, Liesche J, Bohr T, Schulz A (2012) Universality of phloem transport in seed plants. Plant Cell Environ 35:1065– 1076. doi:10.1111/j.1365-3040.2011.02472.x Kim I, Zambryski PC (2005) Cell-to-cell communication via plasmodesmata during arabidopsis embryogenesis. Curr Opin Plant Biol 8:593–599. doi:10.1016/j.pbi.2005.09.013 Kim I, Kobayashi K, Cho E, Zambryski PC (2005) Subdomains for transport via plasmodesmata corresponding to the apical-basal axis are established during Arabidopsis embryogenesis. P Natl Acad Sci USA 102:11945–11950. doi:10.1073/pnas.0505622102 Kollmann R, Glockmann C (1999) Multimorphology and nomenclature of plasmodesmata in higher plants. In: van Bel AJE, Van Kesteren WP (eds) Plasmodesmata. Springer, Berlin, pp 149– 172. doi:10.1007/978-3-642-60035-7_10 Krull R (1960) Untersuchungen über den Bau und die Entwicklung der Plasmodesmen im Rindenparenchym von Viscum album. Planta 55:598–629. doi:10.1007/Bf01884804 Laur J, Hacke UG (2014) Exploring Picea glauca aquaporins in the context of needle water uptake and xylem refilling. New Phytol 203:388–400. doi:10.1111/Nph.12806 Lee SH, Chung GC, Zwiazek JJ (2009) Effects of irradiance on cell water relations in leaf bundle sheath cells of wild-type and transgenic tobacco (Nicotiana tabacum) plants overexpressing aquaporins. Plant Sci 176:248–255. doi:10.1016/j.plantsci.2008. 10.013

13

J Plant Res (2015) 128:49–61 Liesche J, Schulz A (2012) In vivo quantification of cell coupling in plants with different phloem-loading strategies. Plant Physiol 159:355–365. doi:10.1104/pp.112.195115 Liesche J, Schulz A (2013) Modeling the parameters for plasmodesmal sugar filtering in active symplasmic phloem loaders. Frontiers in plant science 4. doi:10.3389/Fpls.2013.00207 Liesche J, Martens HJ, Schulz A (2011) Symplasmic transport and phloem loading in gymnosperm leaves. Protoplasma 248:181– 190. doi:10.1007/s00709-010-0239-0 Lohaus G, Pennewiss K, Sattelmacher B, Hussmann M, Muehling KH (2001) Is the infiltration-centrifugation technique appropriate for the isolation of apoplastic fluid? A critical evaluation with different plant species. Physiol Plant 111:457–465. doi:10.1034/j.1399-3054.2001.1110405.x Lopez D, Venisse JS, Fumanal B, Chaumont F, Guillot E, Daniels MJ, Cochard H, Julien JL, Gousset-Dupont A (2013) Aquaporins and leaf hydraulics: poplar sheds new light. Plant Cell Physiol 54:1963–1975. doi:10.1093/Pcp/Pct135 Lucas WJ, Groover A, Lichtenberger R, Furuta K, Yadav SR, Helariutta Y et al (2013) The plant vascular system: evolution, development and functions. J Integr Plant Biol 55:294–388. doi:10.1111/ Jipb.12041 Münch E (1930) Die Stoffbewegungen in der Pflanze. Gustav Fischer, Jena Öner-Sieben S, Lohaus G (2014) Apoplastic and symplastic phloem loading in Quercus robur and Fraxinus excelsior. J Exp Bot 65:1905–1916. doi:10.1093/Jxb/Eru066 Oparka KJ, Prior DAM (1992) Direct evidence for pressure-generated closure of plasmodesmata. Plant J 2:741–750. doi:10.1111/ j.1365-313X.1992.tb00143.x Orlich G, Hofbrückl M, Schulz A (1998) A symplasmic flow of sucrose contributes to phloem loading in Ricinus cotyledons. Planta 206:108–116. doi:10.1007/s004250050380 Palevitz BA, Hepler PK (1985) Changes in dye coupling of stomatal cells of Allium and Commelina demonstrated by microinjection of lucifer yellow. Planta 164:473–479. doi:10.1007/Bf00395962 Patrick JW (2012) Fundamentals of phloem transport physiology. In: Thompson GA, van Bel AJE (eds) Phloem: molecular cell biology, systemic communication, biotic interactions. Wiley, pp 30– 59. doi:10.1002/9781118382806.ch3 Pina A, Errea P, Schulz A, Martens HJ (2009) Cell-to-cell transport through plasmodesmata in tree callus cultures. Tree Physiol 29:809–818. doi:10.1093/treephys/tpp025 Pommerrenig B, Papini-Terzi FS, Sauer N (2007) Differential regulation of sorbitol and sucrose loading into the phloem of plantago major in response to salt stress. Plant Physiol 144:1029–1038. doi:10.1104/pp.106.089151 Reidel EJ, Rennie EA, Amiard V, Cheng L, Turgeon R (2009) Phloem loading strategies in three plant species that transport sugar alcohols. Plant Physiol 149:1601–1608. doi:10.1104/pp.108.134791 Rennie EA, Turgeon R (2009) A comprehensive picture of phloem loading strategies. P Natl Acad Sci USA 106:14162–14167. doi: 10.1073/pnas.0902279106 Schmitt B, Stadler R, Sauer N (2008) Immunolocalization of solanaceous SUT1 proteins in companion cells and xylem parenchyma: new perspectives for phloem loading and transport. Plant Physiol 148:187–199. doi:10.1104/pp.108.120410 Schmitz K, Cuypers B, Moll M (1987) Pathway of assimilate transfer between mesophyll-cells and minor veins in leaves of cucumismelo L. Planta 171:19–29. doi:10.1007/Bf00395064 Schobert C, Lucas W, Franceschi V, Frommer W (2000) Intercellular transport and phloem loading of sucrose, oligosaccharides and amino acids. In: Leegood R, Sharkey T, von Caemmerer S (eds) Photosynthesis, vol 9. Advances in photosynthesis and respiration. Springer, Netherlands, pp 249–274. doi:10.1007/0-306-48137-5_11

J Plant Res (2015) 128:49–61 Schulz A (1990) Conifers. In: Behnke HD, Sjolund RD (eds) Sieve elements. Springer, Berlin, pp 63–88. doi:10.1007/978-3-64274445-7_4 Schulz A (1992) Living sieve cells of conifers as visualized by confocal, laser-scanning fluorescence microscopy. Protoplasma 166:153–164. doi:10.1007/bf01322778 Schulz A (1998) Phloem. Structure Related to Function. In: Behnke HD, Esser K, Kadereit JW, Lüttge U, Runge M (eds) Progress in Botany, vol 59. Progress in botany. Springer, Berlin, pp 429–475. doi:10.1007/978-3-642-80446-5_16 Schulz A (2005) Role of Plasmodesmata in solute loading and unloading. In: Oparka KJ (ed) Plasmodesmata. Annual plant reviews, vol 18. Blackwell, Oxford, pp 135–161 Schulz A, Kühn C, Riesmeier JW, Frommer WR (1998) Ultrastructural effects in potato leaves due to antisense-inhibition of the sucrose transporter indicate an apoplasmic mode of phloem loading. Planta 206:533–543. doi:10.1007/s004250050430 Slewinski TL, Anderson AA, Zhang CK, Turgeon R (2012) Scarecrow plays a role in establishing kranz anatomy in maize leaves. Plant Cell Physiol 53:2030–2037. doi:10.1093/Pcp/Pcs147 Slewinski TL, Zhang CK, Turgeon R (2013) Structural and functional heterogeneity in phloem loading and transport. Frontiers in plant science 4. doi:10.3389/Fpls.2013.00244 Steinberg G, Kollmann R (1994) A quantitative-analysis of the interspecific plasmodesmata in the non-division walls of the plant chimera Laburnocytisus adamii (Poit) schneid. Planta 192:75–83 Tan QM, Grennan AK, Pelissier HC, Rentsch D, Tegeder M (2008) Characterization and expression of French bean amino acid transporter PvAAP1. Plant Sci 174:348–356. doi:10.1016/j.plantsci.2007.12.008 Tegeder M (2014) Transporters involved in source to sink partitioning of amino acids and ureides: opportunities for crop improvement. J Exp Bot 65:1865–1878. doi:10.1093/Jxb/Eru012 Thompson GA, Schulz A (1999) Macromolecular trafficking in the phloem. Trends Plant Sci 4:354–360. doi:10.1016/s1360-1385(99) 01463-6 Turgeon R (1996) Phloem loading and plasmodesmata. Trends Plant Sci 1:418–423. doi:10.1016/S1360-1385(96)10045-5 Turgeon R, Beebe DU (1991) The evidence for symplastic Phloem loading. Plant Physiol 96:349–354

61 Turgeon R, Gowan E (1990) Phloem loading in coleus blumei in the absence of carrier-mediated uptake of export sugar from the apoplast. Plant Physiol 94:1244–1249. doi:10.1104/Pp.94.3.1244 Turgeon R, Hepler PK (1989) Symplastic continuity between mesophyll and companion cells in minor veins of mature Cucurbita pepo L. leaves. Planta 179:24–31. doi:10.1007/BF00395767 Turgeon R, Medville R (1998) The absence of phloem loading in willow leaves. P Natl Acad Sci USA 95:12055–12060. doi:10.1073/ pnas.95.20.12055 Turgeon R, Webb JA, Evert RF (1975) Ultrastructure of minor veins in cucurbita-pepo leaves. Protoplasma 83:217–232. doi:10.1007/ Bf01282555 Turgeon R, Beebe DU, Gowan E (1993) The intermediary cell— minor-vein anatomy and raffinose oligosaccharide synthesis in the scrophulariaceae. Planta 191:446–456 van Bel AJE, Furch ACU, Hafke JB, Knoblauch M, Patrick JW (2011) (Questions) m(n) on phloem biology. 2. Mass flow, molecular hopping, distribution patterns and macromolecular signalling. Plant Sci 181:325–330. doi:10.1016/j.plantsci.2011.05.008 Voitsekhovskaja OV, Koroleva OA, Batashev DR, Knop C, Tomos AD, Gamalei YV, Heldt HW, Lohaus G (2006) Phloem loading in two Scrophulariaceae species. What can drive symplastic flow via plasmodesmata? Plant Physiol 140:383–395 Volk GM, Turgeon R, Beebe DU (1996) Secondary plasmodesmata formation in the minor-vein phloem of Cucumis melo L and Cucurbita pepo L. Planta 199:425–432 Volk GM, Haritatos EE, Turgeon R (2003) Galactinol synthase gene expression in melon. J Am Soc Hortic Sci 128:8–15 Wright KM, Roberts AG, Martens HJ, Sauer N, Oparka KJ (2003) Structural and functional vein maturation in developing tobacco leaves in relation to AtSUC2 promoter activity. Plant Physiol 131:1555–1565. doi:10.1104/pp.102.016022 Wysocka-Diller JW, Helariutta Y, Fukaki H, Malamy JE, Benfey PN (2000) Molecular analysis of SCARECROW function reveals a radial patterning mechanism common to root and shoot. Development 127:595–603 Zhang C, Han L, Slewinski TL, Sun J, Zhang J, Wang ZY, Turgeon R (2014) Symplastic phloem loading in poplar. Plant Physiol. doi:10.1104/pp.114.245845

13