Dynamics and Control of Phloem Loading of Indoleacetic Acid in Seedling Cotyledons of Ricinus Communis During Germination

During seed germination sugars and auxin are produced from stored precursors or conjugates respectively and transported to the seedling axis. To elucidate the mode of travel of IAA into the phloem a solution of [3H]indoleacetic acid (IAA), together with [14C]sucrose, was injected in the endosperm cavity harboring the cotyledons of germinating seedlings of Ricinus communis. Phloem exudate from the cut hypocotyl was collected and the radioactivity recorded. Sucrose loading in the phloem was inhibited at higher IAA levels, and the rate of filling of the transient pool(s) was reduced by IAA. IAA was detected within 10 minutes, with the concentration increasing over 30 min and reaching a steady-state by 60 min. The kinetics indicated that phloem loading of IAA involving both an active, carrier-based, and a passive, diffusion-based component, with IAA traveling along a pathway containing an intermediary pool, possibly the protoplasts of mesophyll cells. Phloem loading of IAA was altered by sucrose, K+ and a range of non-specific and IAA-specific analogues and inhibitors in a manner that showed that IAA moves into the phloem from the extra cotyledonary solution by multiple pathways, with a carrier mediated pathway playing a principal role. Summary Statement Indoleacetic acid is transported from the peri-cotyledonary space into the phloem of germinating Ricinus seedlings by both trans-membrane carriers and diffusive pathways, with the cells of the cotyledons forming an intermediate reservoir.

These observations show that IAA transport in phloem can occur efficiently over a very wide 179 concentration range, reflecting perhaps the activity of a complex transport system with multiphasic 180 kinetics (Komor et al., 1977). More specifically, the results suggests the operation of a diffusive, 181 'linear' component at the higher levels (Komor et al., 1977;Lichtner and Spanswick, 1981), 182 analogous perhaps to the mode of sucrose transport at varying sucrose concentrations (Fig. 6B). 183 The fact that the relative IAA transport rate at the highest level of applied IAA (20 mM) was 184 somewhat less than linear ( Fig. 2A) suggests that IAA concentration is not the sole factor 185 controlling the loading process. 186 The [ 14 C]sucrose was injected into the endosperm cavity at a sufficiently low level (0.35 187 mM) that it would have a negligible effect on the native sucrose concentration, estimated to be 188 about 90 mM (Kriedemann and Beevers, 1967). Transported [ 14 C]sucrose concentration in the 189 phloem exudate at 1 h, expressed as the percentage of the injected sucrose concentration, was about 190 4% at the two lower applied IAA levels (Fig. 2B). The corresponding transported sucrose 191 concentration was only about 2.5% in the presence of 2.0 or 20.0 mM IAA, indicating that sucrose 192 loading in the phloem is slightly subject to inhibition at higher, non-physiological IAA levels. As 193 the time-dependent changes in transported sucrose concentration revealed, the trend was at or near 194 linearity for the lowest applied IAA level throughout the one hour run (the t0.5 value [the time (min) 195 required for the transported substance to attain the ½-steady-state level in the phloem exudate] is 196 marked with an asterisk to note that a steady-state has not been attained). However, the trend 197 became progressively more sigmoid with increasing IAA levels: at 20 mM IAA there was in the 198 first 20 min a much slower appearance of the labeled sucrose in the exudate indicating that the rate 199 of filling of the transient pool(s) was reduced by IAA. Perhaps as a consequence, sucrose transport 200 at 20 mM applied IAA came to a steady-state at a concentration much below that of the lowest 201 IAA level (Fig. 2B). 202

NAA competes with IAA transport in the phloem 204
The synthetic auxin α-naphthaleneacetic acid (NAA) is analogous to IAA in many of its 205 physiological properties, including the ability to serve as a competitive substrate for the auxin 206 efflux carrier. However, in contrast to IAA, NAA has only marginal affinity for the auxin influx 207 carrier (Delbarre et al., 1996;Marchant et al., 1999). Externally applied NAA enters cells by 208 diffusion. On the basis of these attributes, we selected NAA as a diagnostic probe to test whether 209 IAA loading into the phloem requires auxin efflux carrier activity. We measured the transport of 210 Basipetal transport of IAA, involving efflux carriers, may take place in files of parenchyma cells 217 that are closely associated with minor veins in developing leaves as described in the next section 218 (Aloni, 2010;Mattsson et al., 1999;Mattsson et al., 2003). 219 To test whether sucrose transport may also be altered in the presence of NAA, [ 14 C]sucrose 220 at 8.3 μM was included in the injection medium along with [ 3 H]IAA. The effect, compared to that 221 on IAA transport, was much less clear. There was some reduction in sucrose transport but only at 222 the lowest (0.1 mM) level of NAA (Fig. 3B). This is an interesting result as it seems to contradict 223 the inhibiting effect of IAA on sucrose transport (Fig. 2B). 224 225

Phloem loading of IAA is stimulated by the auxin transport inhibitor triiodobenzoic acid 226
Phloem transport of [ 3 H]IAA from cotyledons of Ricinus seedlings was stimulated at both 20 and 227 100 μM 2,3,5-triiodobenzoic acid (TIBA) (Fig. 4A). The results suggest that a TIBA-enhanced 228 IAA accumulation in auxin-transporting tissues caused a diversion of IAA flow toward the 229 phloem, or an inhibition of lateral efflux from the final sieve tubes. This conclusion is supported 230 by published evidence indicating that TIBA and other auxin transport inhibitors cause auxin 231 accumulation in cells (Davies and Rubery, 1978); that auxin accumulation results in lateral 232 transport between neighboring cells or tissues (Nicolas et al., 2004); that lateral auxin transport 233 may be an integral component in auxin signaling pathways; and that the direction and rate of lateral 234 transport is determined by the prevailing auxin concentration gradient within the transport pathway 235 (ibid.). The role of TIBA as an inhibitor of the auxin efflux carrier PIN1 has been extensively 236 documented in studies on polar auxin transport (Morris et al., 2010). Applied TIBA inhibits the 237 basipetal release of auxin by cells in the polar transport pathway, thereby causing auxin 238 accumulation (Davies and Rubery, 1978). With rising auxin concentration, the lateral release of 239 auxin to neighboring tissues is enhanced thus altering the relative flux among different transport 240 pathways. Evidence for such a mechanism is contained in a study on vascular patterning in 241 Arabidopsis leaves showing that auxin transport provides the controlling signal for both the 242 initiation and the subsequent development of vascular strands in growing leaves. Basipetal auxin 243 transport originating in the tip of young leaf primordia will set the location of the primary vein by 244 inducing the formation of a line of procambial cells. Continuing basipetal auxin transport in the 245 fascicular cambium of the developing vein, together with lateral auxin flow from neighboring cells, 246 controls the ultimate size and composition of the vein (Aloni, 2010;Mattsson et al., 1999;247 Mattsson et al., 2003) (such a cambium would be restricted to the major veins as the minor veins 248 are too small consisting only of a very few cells, though parenchyma cell(s) may be included in 249 these smaller veins.) Cotyledons of Ricinus seedlings also possess an extended bundle sheath that 250 serves as a transport tissue and a temporal sink for assimilates (Rutten et al., 2003) and possibly 251 also auxin. The rate of lateral auxin flow varies with the concentration gradient, which is 252 maintained by the drainage capacity, or "sink effect", of the vein. 253 Several lines of evidence support the notion that auxin transport inhibitors can alter the rate 254 of lateral auxin efflux from cells. Results by Mattsson et al., (1999)

suggest that in developing 255
Arabidopsis leaves the lateral movement of auxin toward the vascular strands was enhanced 256 significantly by treatment with NPA or TIBA as shown by the increased width of the developing 257 veins. In transgenic Arabidopsis seedlings, subjected to gravity or light stimulation, there was a 258 tropic bending response of the hypocotyl which occurred concurrently with an elevated expression 259 of the synthetic DR5::GUS auxin reporter gene on the more elongated side of the hypocotyl. The Much of our knowledge about TIBA relates to its role in inhibiting polar auxin transport. 266 There is, however, accumulating evidence that its action is more broadly based through a general 267 influence on cellular protein trafficking (Geldner et al., 2001). TIBA and other auxin transport 268 inhibitors were shown to retard auxin transport by blocking PIN1 cycling, and also to interfere 269 with the trafficking of plasma membrane H + -ATPase and of other proteins. In the present work 270 we show that in plants treated with 100 μM TIBA [ 14 C]sucrose transport is enhanced, as is [ 3 H]IAA 271 transport (Fig. 4B). Conceivably, the localization of sucrose transporters could also be altered by 272 TIBA as these proteins are degraded and turned over. 273

Effect of potassium ion and sucrose concentration 275
The uptake and phloem loading of sucrose is known to be controlled by a diverse set of internal or 276 externally applied factors including inorganic ions, pH, substrate concentration, as well as reagents 277 for probing metabolic or transport activity (Komor, 1977;Maynard and Lucas, 1982;Schobert et 278 al., 1998;Williams et al., 1992). Given the complex role that sucrose and potassium ions seem to 279 play in phloem function, we examined the effect of these factors on IAA transport. Phloem input 280 and transport rates of IAA and of sucrose were measured together at varying sucrose 281 concentrations, with or without 20 mM K + present in the injection medium; in the latter case 20 282 mM Na + was substituted for K + . 283 With the inclusion of 20 mM K + in the injection medium the pattern of sucrose transport 284 was altered compared to that without K + . With 0.02 mM sucrose, the sucrose content of the 285 exudate was about 0.9% at the end of the 1 h run (Fig. 6B), a value less than half of that obtained 286 without K + (Fig. 5B). Therefore, 20 mM K + in the medium was inhibitory for sucrose transport, a 287 finding in agreement with published results (Van Bel and Koops, 1985). Also at this low applied 288 sucrose level, the presence of potassium caused a shift in the time course from a largely linear to 289 a strongly sigmoid shape, perhaps indicating a shift toward a longer loading pathway. With 290 potassium present there was no significant difference in the relative sucrose transport rates at the 291 three applied sucrose levels, so that the sucrose flux increased in proportion to the applied 292 concentration, suggesting that transport activity at the two higher levels was predominantly in its 293 linear, non-saturable phase (Fig. 6B). Also, with higher applied sucrose levels the value of t0.5 was 294 much increased, indicating a lengthening loading pathway and a strong upward trend in the 295 transient pool size (Fig. 6B); this may mean that a relatively greater portion of transported sucrose 296 was passing through the mesophyll on its way to the phloem. 297 In the absence of potassium ions, the effect of sucrose concentration on transport rates was 298 either insignificant, as in the case of IAA (Fig. 5A), or inconsistent, as in the case of sucrose (Fig.  299   5B). The inclusion of 20 mM K + in the injection medium evoked a set of correlated changes in 300 IAA transport (Fig. 6A) that provide a striking contrast to the results obtained in the absence of K + 301 ( Fig. 5A). At the lowest sucrose level, the amount of IAA nearly doubled after 1 h of transport 302 due to the presence of potassium, presumably resulting from an enhancement of the plasma 303 membrane H + -gradient with K + acting as a counter-ion. Whereas the steady-state concentration of 304 transported IAA in the phloem exudate in the presence of K + was about 0.7 % at 0.02 mM sucrose, 305 it was reduced to about 0.3 % and 0.15 % at 20 mM and 100 mM sucrose respectively ( Fig 6A). 306 In addition the t0.5 values in the presence of K + declined from 23 min at 0.02 mM sucrose to 12 307 and 6 min at 20 mM and 100 mM sucrose respectively ( Fig 6A). These responses are in agreement 308 with, and are explained by the combined effects of sucrose and K + on phloem loading previously 309 described. Therefore, the following conclusions may be drawn from the interactions of K+ and 310 sucrose on IAA loading into the phloem ( Fig. 5A and 6A): 1) The stimulation of IAA loading by 311 K + suggests that the IAA carrier was in its high affinity phase at the applied concentrations of 20 312 mM K + and 0.02 mM sucrose, and therefore the load-enhancing range of K + for the IAA carrier 313 must be wide enough to include the 20 mM level; 2) The degree of sensitivity of IAA loading to 314 the depolarization of the plasma membrane is correlated with sucrose concentration; 3) A t0.5 value 315 may be taken as a semi-quantitative measure of the collective size of the intermediary pools within 316 a given loading path. Because large pools would most likely be found outside the vascular tissues 317 --the latter being of relatively limited volume --it is assumed that their probable location is in the 318 mesophyll. Our results regarding t0.5 values therefore suggest that at the lowest applied sucrose 319 level IAA was being loaded primarily along a pathway passing through the protoplasts of 320 mesophyll cells. At higher sucrose levels, the loading path was drastically diminished in size, 321 suggesting that IAA loading was largely restricted to a direct transfer through the apoplast to the 322 phloem, without passage through the mesophyll. 323 324

Inhibition of phloem transport by sulfhydryl reagents 325
Photosynthates in leaves are generally loaded into the sieve element/companion cell complex 326 through the plasma membrane from the apoplast or, alternatively, pass from the mesophyll to the 327 phloem of minor veins through a symplastic pathway. Pathways may combine, run in parallel, or 328 include a diffusive component depending on the species and on the physiological conditions within 329 the tissue (Rennie and Turgeon, 2009). Evidence for the apoplastic loading of sugars and amino 330 acids into the phloem has been provided for many plant species by testing their sensitivity to 331 PCMBS, a membrane-impermeant inhibitor of proton-coupled transport (Lalonde et al., 2003). In 332 Ricinus cotyledons externally applied [ 14 C]sucrose was shown to move to the sieve elements in 333 two parallel pathways, directly from the apoplast and indirectly after transit through the mesophyll 334 cells (Orlich and Komor, 1992). One of the Ricinus sucrose carriers expressed in yeast can be 335 inhibited by PCMBS (Weig and Komor, 1996 Fig. 7A and B). The observed responses suggest that in the 343 loading pathway for IAA the active component is relatively smaller than that for sucrose. 344 Alternatively, the two carriers may differ in their sensitivity to the inhibitor. However PCMBS 345 also inhibits some aquaporins, which could upset water relations of the cells so altering the 346 observed responses. 347 With or without K + present in the injection medium, PCMB inhibited IAA transport to, or 348 nearly to, the same degree as did PCMBS (Figs. 7A, 7C; Na + replaced K + in 7C). In investigating 349 uptake and movement of IAA in pea stems, Davies and Rubery (1978) found that whereas PCMBS 350 decreased IAA accumulation in the stem segments, PCMB enhanced it. This was interpreted as 351 penetrant PCMB blocking the IAA-efflux carrier on the interior side of the lower plasma 352 membrane, so retaining more IAA in the transporting cells. That export into Ricinus phloem was 353 inhibited by both PCMB and PCMBS is, however, not surprising even though carriers are clearly 354 involved in IAA transport into the phloem: as the cut phloem where transport was measured 355 involves an open ended system, any build-up in the transporting cells due to carrier disruption 356 would simply remain in those cells and never reach the phloem. Nonetheless in the absence of K+ 357 PCMB was slightly less effective an inhibitor than PCMBS, matching the promotion of phloem 358 accumulation by TIBA. 359 When K + was excluded from the injection medium (with 20 mM Na + substituted for K + in 360 the buffer), the inhibitory effect of PCMBS on IAA entry into the phloem was about 54% (Fig.  361 7C), more than twice the effect obtained with K + present (Fig. 7A). Therefore, the presence of 20 362 mM K + was inhibitory for the active component in IAA loading. Interestingly, potassium ions had 363 the opposite effect on sucrose loading: in the absence of K + , PCMBS was wholly ineffective 364 against sucrose transport (Fig. 7D). Perhaps in the latter case the active component of sucrose 365 uptake was being disabled by the low proton motive force caused by the sharply reduced 366 availability of K + for charge compensation (Malek and Baker, 1978). However the active loading 367 of IAA not only continued, but actually doubled in rate when potassium ions were withheld from 368 the injection medium. This could be explained if the processes of IAA and sucrose loading are 369 driven by metabolic energy derived from two distinct sources. 370 While PCMBS was only effective in reducing sucrose transport into the phloem with K + , 371 PCMB was only effective in the absence of K + (Figs. 7B, 7D; Na + replaced K + in 7D). The 372 efficiency of each of the inhibitors may be differentially affected by the prevailing proton motive 373 force that is expected to vary with the applied K + concentration (see above). The observed effects 374 of K + on sucrose loading may involve the regulatory activity of K + channels located in phloem 375 cells together with H + pumps and sucrose carriers. The loss of AKT2/3 K + channel function in an 376 Arabidopsis mutant has been shown to result in impaired sucrose/H + symporter activity and 377 diminished phloem electric potential (Deeken et al., 2002). sucrose transport was enhanced by FC (Fig. 8B), whereas in the same plants IAA transport was 390 inhibited by FC (Fig. 8A). However, with 10.3 μM IAA present, FC failed to evoke these 391 responses (Fig. 8C, D) as though IAA at the higher level was able to supersede or mimic FC's 392 action by evoking a parallel or identical effect. At 10.3 μM IAA, with no FC added, the relative 393 rate of sucrose transport was doubled compared to the rate at 1.3 μM IAA (Fig. 8B, D). The 394 enhancement was equal to that with 10 μM FC at the lower IAA level (Fig. 8B). Some of the 395 responses to FC described here could have been affected by IAA in complex ways. The IAA-396 accelerated acidification of the apoplasm, at least in cells undergoing expansion, appears to be 397 mediated by 14-3-3 receptors, transduction proteins not unlike the receptor for FC on the H + -398 ATPase enzyme (Sanders and Bethke, 2000;Trewavas, 2000). Alternatively, the amount of 399 plasma membrane H + -ATPase may be increased by IAA (Hager et al., 1991). reduced uptake of both IAA and sucrose by about 25% and 40%, respectively, indicating that 410 carrier-mediated uptake into cells, not surprisingly, is involved at some point en route, and was 411 more important for sucrose than for membrane-permeant IAA. As the IAA efflux-carrier inhibitor 412 TIBA enhanced IAA accumulation in the phloem it would appear that the blocking of cell to cell 413 IAA transport may force more IAA into the phloem, or that there is an efflux carrier sieve tubes 414 themselves preventing diversion to other cells en route. The presence of K + at low sucrose 415 concentrations doubled IAA loading into the phloem, whereas at 100mM sucrose the loading of 416 IAA was severely diminished in the presence of K+ even though sucrose without K+ had no effect. 417 Thus the degree of sensitivity of IAA loading to the depolarization of the plasma membrane by 418 K+ is correlated with sucrose concentration (the saturable influx via the proton cotransport system 419 has a Km around 25 mM in Ricinus cotyledons though the value for the outer layer is about 5 mM 420 (Komor, 1977)). At the lowest applied sucrose level, IAA was being loaded primarily along a 421 pathway passing through the protoplasts of mesophyll cells, but at higher sucrose levels IAA 422 loading appeared to be restricted to a direct transfer through the apoplast to the phloem, without 423 passage through the mesophyll. We conclude that the transport of IAA into the phloem is multi-424 faceted with a carrier-mediated pathway playing a significant role. In preparing the seedlings for injection, the hypocotyl was cut with a sharp razor blade to remove 462 the roots and lower hypocotyl, thus leaving a hypocotyl stump, seven to ten mm in length, attached 463 to the cotyledons enclosed within the endosperm. Using a microsyringe, five μL IM was injected 464 into the endosperm cavity between the cotyledons, thus exposing the two enclosed cotyledons to 465 the radiolabeled substances being tested (Fig. 1). Then, the endosperm was placed in a small 466 beaker, between layers of absorbent paper moistened with IB, ensuring that the endosperm with 467 the emerging hypocotyl stump was held in an upright position. Freely exuding phloem sap from 468 the cut surface of the hypocotyl was collected during ten minute intervals, starting at 0-10 minutes, 469 generally for an hour, with a graduated microcapillary tube resting on the hypocotyl stump. High 470 relative humidity was maintained throughout the transport period. The volume of the collected 471 exudate was recorded, and the sample was transferred with 95% ethanol as a rinse into a liquid 472 scintillation vial for analysis. 473 474

Radioactivity Counting and Data Presentation 475
The 3 H-and 14 C-activity in each exudate sample was determined simultaneously using Ecoscint 476 was cut off leaving a 1cm stump, which was placed against a calibrated 10µL capillary tube. Five 526 μL of injection medium containing the radiolabeled substances to be transported was injected into 527 the endosperm cavity between the cotyledons. The capillary tube with exuded phloem sap was 528 replaced every 10 min, the volume recorded, and the contents counted for radioactivity. 529 Ricinus seedlings was cut off leaving a 1cm stump, which was placed against a calibrated 10µL capillary tube. Five μL of injection medium containing the radiolabeled substances to be transported was injected into the endosperm cavity between the cotyledons. The capillary tube with exuded phloem sap was replaced every 10 min, the volume recorded, and the contents counted for radioactivity. at ½-equilibrium (or at ½-maximum) concentration attained within 1 h of transport. If no equilibrium is apparent, the t 0.5 value is marked with an asterisk. The data are the combined results of three independent experiments, and represent the means of at least ten replicate measurements.