By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Indole-3-acetic acid (IAA) is a primary phytohormone that regulates multiple aspects of plant development. Because polar transport of IAA is an essential determinant of organogenesis and dynamic tropic growth, methods to monitor IAA movement in vivo are in demand. A self-referencing electrochemical microsensor was optimized to non-invasively measure endogenous IAA flux near the surface of Zea mays roots without the addition of exogenous IAA. Enhanced sensor surface modification, decoupling of acquired signals, and integrated flux analyses were combined to provide direct, real time quantification of endogenous IAA movement in B73 maize inbred and brachytic2 (br2) auxin transport mutant roots. BR2 is localized in epidermal and hypodermal tissues at the root apex. br2 roots exhibit reduced shootward IAA transport at the root apex in radiotracer experiments and reduced gravitropic growth. IAA flux data indicates that maximal transport occurs in the distal elongation zone of maize roots, and net transport in/out of br2 roots was decreased compared to B73. Integration of short term real time flux data in this zone revealed oscillatory patterns, with B73 exhibiting shorter oscillatory periods and greater amplitude than br2. IAA efflux and influx were inhibited using 1-N-naphthylphthalamic acid (NPA), and 2-naphthoxyacetic acid (NOA), respectively. A simple harmonic oscillation model of these data produced a correlation between modeled and measured values of 0.70 for B73 and 0.69 for br2. These results indicate that this technique is useful for real-time IAA transport monitoring in surface tissues and that this approach can be performed simultaneously with current live imaging techniques.
Differential distribution of the phytohormone auxin regulates a variety of developmental processes in plants (Benjamins and Scheres, 2008), and polarized transport mechanisms mediate asymmetric auxin distributions in response to both developmental and environmental cues (Petrášek and Friml, 2009; Vanneste and Friml, 2009). The most abundant and physiologically relevant form of auxin, indole-3-acetic acid (IAA), is synthesized in young plant tissues (Ljung et al., 2005). Transport of IAA from sites of synthesis to target cells is mediated by directional uptake and efflux from cells and is driven by chemiosmotic gradients (Zazímalováet al., 2010).
Full length PIN proteins regulate polar auxin fluxes that are primary determinants of polar growth, organogenesis, and tropic responses. AUX/LAX uptake transporters form sinks that are important to redirection of auxin transport streams. ABCB auxin transporters function primarily in auxin loading and maintenance of long distance auxin transport streams. Arabidopsis and maize mutants bearing lesions in ABCB auxin transport genes have been particularly useful for studies of long distance auxin transport, as they exhibit reduced auxin transport and auxin-dependent elongation growth without the defects in organ formation associated with some pin mutations (Gälweiler et al., 1998; Noh et al., 2001; Multani et al., 2003).
Transport of IAA and other organic acids between the rhizosphere and roots has been investigated using various methods that include biosensors (Farrar et al., 2003), and a wide range of methods have been used to detect IAA levels and movements in roots. The most sensitive of these include nanoscale chromatography/mass spectrometry linked to fluorescent marker-based cell sorting, which detects IAA levels in all root tissues including inner tissues (Petersson et al., 2009), imaging of auxin-responsive reporters (Ulmasov et al., 1997), and nanoscale radiotracer experiments (Peer and Murphy, 2007). Mass spectral analyses of IAA content are now possible for single cells, but this destructive method cannot be applied to real time measurements. Auxin responsive reporters are useful, but are limited for time-resolved studies as there is a delay between the time IAA reaches a cell and the production of a visible transcription/translation reporter product (Peer and Murphy, 2007). Radiotracer assays have overcome many of their former limitations as they can now utilize femtomolar amounts of IAA applied to very small surfaces, but remain an invasive/destructive technique, as detectors that can monitor the movement of the isotope in discrete regions in real time have yet to be developed.
For measurement of IAA near plant tissue surfaces, in situ electrochemical microsensors overcome many of these problems. However, when used in a traditional concentration mode that measures the amount of a molecule of interest, microsensors suffer from low signal-to-noise ratio, temporal drift, and relatively high detection limits (Hernandez et al., 1994). In contrast, self-referencing (SR) microsensors (Porterfield, 2007) have the potential to overcome these limitations. SR is a sensing modality based on Fick’s first law of diffusion that allows real-time characterization of dynamic biophysical transport in plant physiology (Gilliham et al., 2006; Porterfield, 2002a,b) and other biological, biomedical, and environmental applications (Porterfield, 2007). Real-time flux measurements are based on oscillatory probe translation and the quantification of analyte concentration differentials (ΔC) between two spatial positions separated by a known excursion distance (ΔX). The SR technique ‘self-corrects’ for ambient drift and noise by phase-sensitive filtering facilitated by the probe oscillations (Kuhtreiber and Jaffe, 1990).
The feasibility of using a SR electrode to non-invasively measure IAA flux in maize roots was demonstrated when the electrode was modified with multi-walled carbon nanotubes (MWNT) to increase the surface area available for electron transfer and enhance catalysis (Mancuso et al., 2005). However, the temporal resolution and signal-to-noise ratio of this electrode were lower than desired, presumably because the experiments used an oscillation frequency below the values commonly specified (Kuhtreiber and Jaffe, 1990) and the data acquisition scheme utilized an AC coupling method, which suffers from differential signal capacitance decay (Porterfield, 2007). A further limitation of the approach was that it appeared to require external addition of IAA at concentrations sufficient to induce physiological changes in the root (Thimann, 1937).
In this study, a SR IAA microsensor was developed employing platinum black and carbon nanotube (CNT) surface modifications. This microsensor was then used to non-invasively characterize IAA flux in 3- to 5-day B73 inbred and brachytic2 roots. BRACHYTIC2 (BR2) is a maize paralog of the Arabidopsis ABCB1 auxin transport protein that functions in loading of auxin in apical and meristematic tissues (Multani et al., 2003; Geisler et al., 2005; Blakeslee et al., 2007; Knöller et al., 2010). Mutations in BR2 result in maize dwarfs with compacted lower internodes and expanded sub-epidermal cell layers. Plant root studies were performed both in the presence and absence of the IAA transport inhibitor 1-naphthylpthalamic acid (NPA), shown to bind and inhibit ABCB auxin transporters (Noh et al., 2001; Murphy et al., 2002; Geisler et al., 2005) as well as the influx inhibitor 1-naphthoxyacetic acid (NOA) (Parry et al., 2001; Rahman et al., 2002). A new technique for quantification of endogenous IAA transport was developed by integrating real-time SR data to measure the IAA anion, and these results showed harmonic oscillations in shootward auxin transport at the root apex. Therefore, this technique provides valuable insight into the fundamental mechanisms underlying auxin transport in root epidermal cells.
Results and discussion
Sensor construction/surface modification
First, to optimize sensor design, the relationship of sensitivity to electrode surface was analyzed using a macroelectrode format prior to microelectrode construction and validation. Based on macroelectrode characterization results (see Figure S1), a Pt black/MWNT electrode was selected as the optimal design, and microelectrodes were constructed using tapered parylene-insulated Pt wires (tip diameter of 2–4 μm). The Pt black/MWNT microsensor was calibrated between 0.1 and 40 μm IAA in ¼ Murashige and Skoog (MS) media (pH = 7.4); the lower limit of detection of the microelectrode was 0.4 μm IAA (Figure 1a). Following calibration, abiotic IAA flux was measured within the unstirred layer formed near a tapered glass pipette filled with 5 mm IAA and 1% agar to validate use of the microsensor in the SR modality (Figure 1b). The oscillation frequency was maintained at 0.33 Hz, as preliminary experiments indicated that sensors operating below 0.20 Hz are subject to extensive background noise and underestimation of IAA levels. The excursion distance was 30 μm. The enhanced response time and sensitivity of the sensor allowed operation within the oscillation frequency suggested in the literature (Kuhtreiber and Jaffe, 1990). The result of the modifications to the system was an acceptable dynamic efficiency (ε) for abiotic flux experiments of 0.98.
The performance of CNT-based IAA microsensors was previously shown to be unaffected (<3% change in output) by compounds found in plant growth media such as Ca(NO3)2, NaH2PO4, MgSO4, KCl, CuSO4, KH2PO4, KNO3, MnCl2, NaN3, sucrose and glucose; as well as NPA and NOA (Mancuso et al., 2005). To further investigate microsensor selectivity, additional tests were conducted in the presence of electroactive and non-electroactive compounds commonly found in plant exudates or growth media (citrate, oxalate, malate, ascorbate, nitric oxide, glucose, and 2,4-dichlorophenoxy acetic acid). Interferent tests for the sensor compared sensor responses for 10 μm IAA to those recorded for various interferents at 10 μm concentration levels. For common root exudates (malate, citrate, and oxalate), inorganic N ionic nutrients (NaNO3, NH4NO3) and the herbicide 2,4-D the sensor output was <1% of the IAA response and are not considered to be interferents. While the output of the sensor for ascorbate was 28% of that of IAA, it is not considered to be a viable interferent as ascorbate is not a root exudate, but was tested as part of common protocol. NO at 2% is also not a viable interferent, and NO levels in tissue are much smaller that IAA.
Induced IAA flux in Zea mays roots
Uptake of exogenously applied IAA across the root tissue surface has been attributed to a combination of diffusion and carrier mediated transport (Martin and Pilet, 1986; Bennett et al., 1996). A previously reported study using non-invasive SR microsensors used application of external IAA to investigate saturation of transport (Mancuso et al., 2005). Therefore, we examined the saturation of transport using Pt black/CNT microsensors and increasing concentrations of externally supplied IAA to non-invasively measure IAA flux near the surface of 3- to 5-day B73 inbred maize roots (Figure 2a). In B73, the maximum influx occurred in the distal elongation zone (DEZ), a region between the root apical meristem and elongation zone (Figure 2b,c; Ishikawa and Evans, 1995). The identity of this zone was determined by analysis of surface cell size and distance from the apex. This result is consistent with radiotracer assays of basipetal IAA movement in roots (Jones, 1990; Geisler et al., 2005; Peer and Murphy, 2007), analyses of root gravitropic bending in mutants and wild type plants (Mullen et al., 1998; Swarup et al., 2005), and subcellular localization studies of auxin transporters (reviewed in Zazimlova et al., 2010).
Characterization of br2 roots
Although auxin transport and growth defects have been extensively characterized in ‘br2’ shoots (Multani et al., 2003; Knöller et al., 2010), no root phenotypes have been described in the mutant. In Arabidopsis, the BR2 ortholog ABCB1 exports auxin from root cells in the apex and tissues immediately above (Geisler et al., 2005; Mravec et al., 2008). Loss of ABCB1 function in Arabidopsis results in a decreased auxin export from the root apex, increased root waving and slightly agravitropic growth (Noh et al., 2003; Geisler et al., 2005). A characteristic of mutants with altered shootward auxin movement at the root apex is altered gravitropic growth (Bennett et al., 1996; Chen et al., 1998; Mullen et al., 1998). An examination of br2 seedlings grown on vertical media indicated that the roots exhibited some agravitropic growth and tended to curl instead of growing in line with the gravity vector (Figure 3a,b).
Nanoscale radiotracer assays used in Arabidopsis (Peer and Murphy, 2007) were adapted to measure shootward auxin transport in maize roots. In these assays, 3H-IAA was transported from the root apex to the elongation zone in B73, but movement from the apex to the elongation zone was reduced in br2, with the greatest reductions observed at 0.5 mm above the root apex (Figure 3c). This transport was further reduced to background levels by addition of the auxin efflux inhibitor 1-naphthylphthalamic acid (NPA, 10 μm). These results indicate that BR2/ABCB1 functions in conjunction with other auxin transporters to mediate shootward movement of auxin in maize roots. The results obtained were consistent with reported radiotracer assays of shootward IAA movement from root apices (Geisler et al., 2005; Peer and Murphy, 2007).
To determine if BR2 is expressed in this region of the root, flash frozen roots were scraped to remove surface cells or sectioned (0.25 mm) in a cylindrical microtome and assayed for BR2 expression by quantitative RT-PCR. BR2 expression was detected in root surface scrapings, and maximal BR2 expression was found in the sections corresponding to the root apex (Figure 3d). BR2 was also localized to the root surface using histochemical staining. NPA has been shown to bind to ABCB1 and, to a lesser extent, ABCB19, in vivo and in heterologous systems (Murphy and Taiz, 1999; Noh et al., 2001; Murphy et al., 2002; Rojas-Pierce et al., 2007). Previously, this binding was histochemically detected by oxidative or azo coupling of bound NPA and was confirmed by radiolabelling assays (Murphy and Taiz, 1999). This method was used with B73 and br2 seedlings. Binding/immobilization of the coupled NPA product was seen at the apical surface of B73 roots, and was greatly reduced in br2 (Figure 3e,f). The absence of NPA staining in br2 suggests that BR2 is an immobilization component in this histochemical reaction. Taken together, these results indicate that BR2 functions as an active auxin transporter in apical surface tissues in maize roots in a manner similar to Arabidopsis ABCB1 in addition to its role in delivering shoot-derived auxin to the root.
Induced IAA influx occurs at the distal elongation zone
To further validate the Pt/CNT microsensor, we compared the induced IAA uptake near the DEZ of B73 and br2 roots, whereby IAA flux was continuously measured during exogenous addition of IAA. Dose–response curves were prepared by continuously measuring IAA flux at the elongation zone of B73 and br2 roots. For each root, the maximum induced IAA influx (saturation) occurred following addition of 15 μm IAA. As expected, transport in B73 roots was significantly higher than in mutant br2 roots for all external additions of IAA (Figure 4). In both B73 and br2 roots, the induced flux remained at this level for up to 6 h in repetitions of this experiment (B73 h 1 influx = 62.5 ± 3.1, h 6 influx = 59.9 ± 2.8, P = 0.09; br2 h 1 influx = 0.58 ± 0.09, h 6 influx = 0.51 ± 0.02, P = 0.07). The exponential rate constant (0.16) and maximum induced IAA influx (70 fmol cm−2 sec−1) from an empirical model for B73 roots was significantly higher than the modeled br2 constant (0.07) and maximum influx (0.6 fmol cm−2 sec−1). These results highlight a limitation of selective electrode methods that employ addition of exogenous IAA to monitor auxin transport in intact tissues. Reduced induced IAA influx was measured in the br2 mutants despite the fact that the br2 mutant is defective for plasma membrane IAA efflux, not influx. However, while counterintuitive, these measurements of artificially induced influx do correlate indirectly with the overall reduced capacity for polar auxin transport in the br2 mutant. The low IAA clearing capacity of br2 increases the levels of IAA in the tissue (2.4 ± 0.37 times greater than B73 levels as determined by GC-MS with 13C-IAA as an internal standard), which in turn decreases the concentration gradient between the root and the artificially augmented external media. The decreased concentration gradient results in decreased induced flux measurements by br2.
Effects of exogenous IAA on oxygen flux
However, a drawback of the induced IAA experimental approach is that application of 5–15 μm IAA inhibits root growth (Pilet and Saugy, 1985; Martin and Pilet, 1986). To study the effect of exogenous IAA on root metabolism, oxygen influx was measured in 4-day B73 and br2 roots during and after IAA addition using a fiber optic SR oxygen sensor (Chatni and Porterfield, 2009). Prior to addition of IAA, average oxygen influx at the distal elongation zone of B73 roots (26.1 ± 2.3 pmol cm−2 sec−1, Figure 5a,b) was similar to previous reports (Porterfield, 2002a,b, 2007; Mancuso et al., 2005; Chatni and Porterfield, 2009), and average oxygen influx was significantly higher in B73 than in br2 roots (21.7 ± 1.6 pmol cm−2 sec−1, Figure 5c,d). This difference in oxygen influx is consistent with inhibitory effects of endogenous IAA accumulations observed in br2 roots (2.4 ± 0.37 times greater than B73 levels as determined by GC-MS with 13C-IAA as an internal standard). Therefore, the addition of external IAA significantly decreased aerobic metabolism. A dose–response curve for each root is shown in Figure 5.
Exogenous application of IAA at concentrations as low as 1.5 μm caused a significant reduction in root oxygen flux in both B73 and br2 after approximately 10 min, although oxygen flux in the DEZ recovered to basal levels for additions <1 μm IAA. Fifteen micromolar IAA, the optimal concentration for induced IAA influx, reduced oxygen influx after 2 h by 50 ± 5% and 89 ± 2% for B73 and br2 roots, respectively. After 6 h, oxygen influx was completely inhibited in both B73 and br2 roots (B73 0.11 ± 0.06 pmol cm−2 sec−1; br2 0.07 ± 0.01 pmol cm−2 sec−1, n = 5), demonstrating the short- and long-term effects of exogenous IAA addition on root development. These results were confirmed by trypan blue staining, which indicated decreased cell viability, and carboxy-H2DCFDA staining, which indicated increased accumulation of reactive oxygen species, after 2 h of 15 μm IAA treatment (Figure S2).
Root elongation rate (RER) was measured in 4-day maize roots before and after external addition of 15 μm IAA using an image subtraction technique (McLamore et al., 2010a). In the absence of external IAA, B73 and br2 roots elongated at a rate of 2.6 ± 0.3 mm h−1, and 0.8 ± 0.5 mm h−1, respectively, which was similar to previously reported results for B73 (Verslues et al., 1998). RER decreased within 30 min of 15 μm IAA addition, and average RER values for B73 and br2 were 0.5 ± 0.1 and 0.1 ± 0.1 mm h−1, respectively, confirming that use of exogenous IAA with SR electrode assays can bias assay results (Figure S2) as well as produce that are counterintuitive.
Endogenous IAA flux in maize roots
Next, the efficacy of using the SR IAA microsensor to measure endogenous IAA flux in maize roots was examined without addition of exogenous IAA was tested. No significant net IAA flux was detected at the DEZ of roots (B73 0.3 ± 6.3 fmol cm−2 sec−1; br2 0.2 ± 0.8 fmol cm−2 sec−1) relative to a reference measurement taken 2 mm from the root surface (0.1 ± 0.1 fmol cm−2 sec−1) roots (Figure 6a,b). However, closer examination of what appeared to be noise in the system (indicated by the large comparative SD) suggested that this ‘noise’ actually represented a transient signal in the rhizosphere reflecting apoplastic/cellular exchange in the root epidermis. In order to resolve transient and dynamic influx/efflux activities of IAA, we integrated the real-time IAA flux (iFlux) based on an approach recently developed for measurement of net surface uptake of neurotransmitters (glutamate) from cultured neurons (McLamore et al., 2010b) and this revealed regulated flux in both roots (Figure 6c,d). iFlux profiles were calculated from integrated efflux (positive values) and integrated influx (negative values) and were averaged over a 12-min time period. An example of iFlux calculations is provided in the supplemental data (Figure S3). iFlux profiles were consistent with results from radiotracer assays and indicated that endogenous IAA transport was maximal near the DEZ (Figure 6).
Average absolute integrated flux near the DEZ of B73 (7440 ± 600 fmol cm−2 sec−1) was significantly greater than flux observed near the distal elongation zone of br2 (24.1 ± 2.1 fmol cm−2 sec−1) (n = 5). Mean integrated IAA efflux (4260 ± 24 fmol cm−2) at the DEZ of B73 roots was greater than influx (3420 ± 18 fmol cm−2), indicating a net IAA loss in this region (n = 5, P = 0.01). As would be expected from the decreased availability of shoot-derived auxin in br2 (Multani et al., 2003), IAA fluxes in the br2 root were two orders of magnitude lower than in B73, but exhibited a similar profile. In both B73 and br2, auxin release was predominant (144 ± 2.0 fmol cm−2, and 102 ± 5.1 fmol cm−2, respectively) within the DEZ. These results are consistent with reports of ABCB1-mediated auxin retention in the root cells in Arabidopsis (Murphy et al., 2000; Geisler et al., 2005).
IAA flux oscillations occur at the DEZ
To further investigate the temporally dynamic endogenous transport of IAA, endogenous IAA flux was continuously recorded at the DEZ of B73 and br2 maize roots, and integrated flux data revealed an oscillatory pattern (rolling average, n = 7). Representative plots demonstrating four consecutive oscillations in B73 and br2 are presented in Figure 7(a–g) (n = 3, see Figure S4). As apoplastic auxin has been previously shown to diffuse into external media (Blakeslee et al., 2007), these oscillations are a direct result of changes in rhizosphere IAA levels that are derived from apoplastic pools. Changes in apoplastic pools are, in turn, regulated by uptake transport proteins (e.g. AUX1), and efflux transporters (e.g. PIN/ABCB) (Figure 7g). For the B73 root, the integrated efflux oscillation period (3.6 ± 0.3 min) was shorter than the influx oscillation period (5.7 ± 0.1 min), and oscillations were out of phase by approximately 109 ± 28°. Efflux (7.9 ± 0.8 min) and influx (6.3 ± 0.9 min) oscillation periods for br2 roots were significantly different than B73 roots (P = 0.20, α = 0.05), and the efflux period was slightly longer than the influx period (phase angle approximately 148 ± 23°).
The noted oscillations were modeled using a simple harmonic oscillation model (see experimental procedures), and the correlation between the measured and modeled data for B73 roots (0.70) was similar to the value for br2 roots (0.69). Although the amplitude of the measured oscillations varied amongst roots (Figure S4), the amplitude of IAA iFlux oscillations in B73 roots (136 ± 6 fmol cm−2) was significantly larger than the amplitude in br2 roots (0.8 ± 0.1 fmol cm−2; Table S1). These oscillatory patterns of IAA flux indicate the transient activity of IAA transport at the DEZ, and provide insight into the transport governing net efflux of IAA at the DEZ. Relatively long influx periods lead to the measured net IAA leakage at the DEZ, since the amplitude of oscillations was not significantly different for individual roots. Although the magnitude of integrated flux values was significantly lower for br2 roots than B73 roots, similar oscillation patterns were noted, where transient efflux and influx were out of phase, but net transport indicated a slight loss of IAA due to relatively long-propagating influx oscillations.
Endogenous B73 IAA flux was measured continuously for 20 min, and integrated influx and efflux oscillated as reported in Figure 6. After monitoring IAA flux for 20 min, 5 μm NPA was applied. After approximately 30 min, IAA efflux was not detected, and influx was reduced by 25 ± 3%, indicating that apoplast auxin pools were depleted due to inhibition of auxin efflux (Figure 8). In a parallel experiment, auxin uptake was inhibited in B73 roots using 5 μm NOA after 20 min of continuous measurement. After 30 min of NOA treatment, influx was not detected, while efflux increased by an average of 170 ± 16%, over the time scale that we did our measurements, validating previous data implicating NOA as a carrier influx inhibitor (Parry et al., 2001; Rahman et al., 2002). Increased efflux activity observed in the time frame of these experiments appears to be a result of depletion of cellular IAA pools as NOA prevents auxin reentry from the apoplast and consequent increased IAA diffusion to the rhizosphere. It is also worthwhile to point out that it takes much longer for NPA to affect the efflux than for NOA to affect the influx. This situation suggests differences in the mode of action of these compounds and/or differences in rates of movement into the root tissue. This may be a result of the competitive nature of NOA inhibition versus the non-competitive inhibition of efflux by NPA. These inhibition experiments validated that the measured oscillations were due to endogenous auxin transport, and demonstrate the usefulness of the non-invasive sensor technique for dissecting the biochemical mechanisms involved in root development.
Using enhanced sensor surface modifications, signal processing, and after measurement data integration, SR microsensors provide for non-invasive quantification of in vivo IAA transport dynamics in the mass boundary layer near the root/fluid interface. Oscillating IAA influx and efflux can be detected within this boundary layer surface with the majority of the transport occurring near the distal elongation zone. As expected, endogenous flux in IAA transport mutants of maize was significantly less than controls. Analysis of integrated IAA flux indicates that IAA transport oscillates near the distal elongation zone of intact roots. Relatively long influx oscillation periods and short efflux oscillations lead to measured net IAA leakage at the elongation zone. Loss of ABCB1 has been shown to result in net leakage of IAA from Arabidopsis root tips (Geisler et al., 2005). The data shown here indicate that loss of the ABCB1 ortholog BR2 has a similar effect in maize roots. Further, as the oscillations correlated well with a simple harmonic oscillation equation model (r approximately 0.70) and these oscillations were inhibited predictably by NPA and NOA treatment, the utility of this non-invasive sensor technique for understanding the biochemical mechanisms involved in root development is demonstrated. Further experiments in Arabidopsis mutants monitoring IAA flux, oxygen consumption, and pH oscillations during gravitropic bending are expected to characterize the contributions of auxin efflux and influx transporters to epidermal shootward flows near the root apex with high resolution.
The chemical reagents potassium chloride, 3-mercapto-propyl trimethoxysilane (MPS), dimethyl formamide, indole acetic acid (cell culture grade in crystalline form), hexachloroplatinate, lead acetate, hexachloroplatinate, 2,4-dichlorophenoxy acetic acid (2,4-D), 1-N-naphthylpthalamic acid (NPA), 1-naphthoxyacetic acid (NOA), and the Plant Cell Viability Assay kit were purchased from Sigma Aldrich (http://www.sigmaaldrich.com). Dibasic sodium phosphate and sodium chloride were purchased from Mallinkrodt (http://www.acros.com). Monobasic potassium phosphate was purchased from Acros Organics. Multi-walled carbon nanotubes (outer diameter ≤10 nm, inner diameter ≤2–5 nm) were obtained from Cheap Tubes, Inc. (http://www.cheaptubesinc.com).
Sensor construction and surface modification
Sensor design was optimized by first modifying the surface of insulated platinum discs (1.6 mm outside diameter) and analyzing for sensitivity towards IAA in ¼ MS media. The surface modifications included: (i) addition of a nano-patterned platinization (or platinum black) layer; (ii) immobilization of MWNTs on platinum; and (iii) addition of a hybrid Pt black/MWNT layer. The modified surface area (SA) of each electrode is a major parameter governing the sensor performance, and the average SA of the three designs was 3.9 ± 0.1 μm2, 12.8 ± 0.2 μm2, and 7.9 ± 0.3 μm2, respectively (Figure S1). Platinum black electrode sensitivity was dependent on the amount of platinization time, and the optimum platinization time was determined to be 40 sec. The average sensitivity for each electrodes design was 22.3 ± 2.3 nA μm−1 μm−2, 96.6 ± 3.2 nA μm−1 μm−2, and 312.1 ± 7.0 nA μm−1 μm−2 (Figure S1). Although the electrically active surface area of platinum black-CNT electrodes was slightly smaller than CNT modified electrodes, this hybrid design was found to be more sensitive per active surface area. Although the hybrid platinum nanoparticle CNT microelectrode had significantly less active surface area than the CNT microelectrode used by Mancuso et al. (2005), data indicate enhanced response time (3.0 ± 0.2 sec), reproducibility, and sensitivity per active surface area (312.1 ± 7.0 nA μm−1 μm−2).
Macroelectrodes (1.6 mm outside diameter) were obtained from BASi (http://www.basinc.com) and microelectrodes (2–4 μm outside diameter) were obtained from Microprobe Inc. Electrodes were modified using various surface modification approaches to optimize the fabrication scheme. These modifications included addition of a platinum black layer and immobilization of MWNTs following methods common to the literature. To optimize sensor fabrication scheme, various sensor designs were analyzed in a macroelectrode format, compared, and the optimum design was then miniaturized.
Electrodes were prepared by polishing with a 1 μm diamond slurry on a vinyl polishing pad (BASi) rinsed with methanol, and polished with a 1 μm aluminum slurry from Buehler (http://www.buehler.com) on a microcloth pad (BASi), and rinsed with deionized distilled water (ddH2O). Electrodes were then sonicated for 5 sec in ethanol and then sonicated for 5 sec in ddH2O. Prepared platinum electrode surfaces were modified by the application of a layer of platinum black (amorphous nanopatterned platinum) using an electroplating technique (Jaffe and Nuccitelli, 1974). The bare Pt electrodes were first polished using the protocol mentioned below, and placed in a solution of 0.72% (v/v) hexachloroplatinate and 0.0001% (v/v) lead acetate. A bare platinum wire was used as a counter electrode during the electrodeposition process at 10 volts for 40 sec. Platinum black electrodes were then silanized by immersing in 3-mercaptopropyltrimethoxysilane for a 6–10 h period and subsequently washed with DI water.
Multi-walled carbon nanotubes (MNWTs) (2–4 nm inner diameter, <8 nm outer diameter, and 30 μm in length) were refluxed in concentrated HNO3 for 6–10 h, and immediately washed three times with distilled water. Next, 2 mg of MWNTs were suspended in 2 ml of dimethyl formamide (DMF), and the DMF/MWNT suspension was sonicated for 5 min according to Zeng and Huang (2004). Platinium black, silanized electrodes were immersed into the DMF/MWNT solution for 6 sec, and left to dry for 6 h.
A BASi cell stand 3 was used for electroanalytical characterization of sensor performance during surface modification using a three electrode scheme (auxiliary, reference, and sensing). Reference electrodes (Ag/AgCl) and platinum auxiliary electrodes were obtained from BASi. The acquisition rate was 0.1 kHz for all macro and microelectrode analysis on the BASi cell stand. Cyclic voltammograms (CV) for all electrode designs were run using a Cell Stand 3 (BASi) in potassium ferricyanide and potassium chloride solution from 0 to 650 mV at a scan rate of 20 mV/sec in order to determine effective (electrically active) surface area. Effective surface area was determined using the Randles–Sevick equation: , where: ip = reduction peak, n = number of electrons transferred, D = molecular diffusion coefficient, C = molar analyte concentration, A = effective surface area, and v = scan rate. The value of A was calculated by holding v constant in all experiments, and using the constants n, D, and C in the Randles–Sevick equation. CV scans were replicated 10 times for each electrode design and the average peak current values reported.
IAA macro and microelectrodes were calibrated by external addition of IAA (0.1–20 μm) to a bath solution containing 23 ml of PBS (0.14 m NaCl, 0.0027 m KCl, 0.0015 KH2PO4, and 0.008 m Na2PO4; 6.54 < pH < 7.24) at a polarization potential of 700 mV. Sensor response time was calculated by adding IAA in 1 μm intervals to a stirred, 25 ml solution of ¼ MS media, and calculating the time to reach 95% of the measured steady state current at a fixed polarization potential.
For each surface modification, sensor selectivity was characterized using the interferents citrate, malate, oxalate, ascorbate, nitric oxide, glucose, and 2,4-dichlorophenoxy acetic acid (2,4-D). Interference was analyzed by recording a baseline current value in the presence of 10 μm IAA until steady state (<3% change in current) was observed. The interferent compound was then added to the solution, and changes in baseline current below 3% were deemed insignificant.
Oxygen optrodes were fabricated using previously published techniques (Kühl and Jørgensen, 1992; Lee and Okura, 1997; Chatni and Porterfield, 2009). Any temperature effects were corrected through the use of an integrated thermocouple that provided input for a digital signal processor (DSP) fluorometer (Presens). Illumination utilized un-modulated white light via a fiber optic illuminator with standard light source (World Precision Instruments Inc., http://www.presens.dc). Optrodes were calibrated in de-aerated ¼ MS solution (nitrogen purged) and O2 saturated ¼ MS (21% O2).
SR hardware included a vibration isolation table with Faraday cage (Technical Manufacturing Co., http://www.techmfg.com), camera/zoomscope, microsensor, and reference sensor (IAA microsensor only) mounted on a head stage controlled by a motion control system (see Figure S5). Automated Scanning Electrode Technique (ASET) software was used for data acquisition (A/D) and control functions (D/A) (Science Wares, http://www.sciencewares.com). The A/D board with DC-coupled differential amplifier, low/high pass filters, and video/data acquisition system were obtained through Applicable Electronics, Inc. To avoid capacitive decay in the 0.1–0.5 Hz time domain (leading to an underestimation of the true flux values), baseline signals were discretely subtracted (DC coupling) prior to amplification (Porterfield, 2007). DC-coupling allowed for baseline subtraction without differential signal degradation due to capacitive discharge, allowing calculation of concentration differentials without making assumptions regarding background signals and measurement efficiency (Porterfield, 2007). The sample rate for all SR experiments was 1 kHz. Use of an oscillation frequency (0.2–0.5 Hz) and excursion distance (20–60 μm) within the range suggested in the literature (Kuhtreiber and Jaffe, 1990) ensured that background noise and drift were common between the two measurements, and mechanical motion of the electrode did not significantly disrupt the concentration gradient.
Use of the IAA microelectrode in the SR modality was validated using step-back experiments (SBE) designed to optimize the oscillation frequency according to Porterfield (2007). This optimization is a function of both the oscillation time, and the excursion distance (ΔX) over which the sensor is translated. SBE involves creating a gradient within a bath solution simulating that to be used in the biological experiment. A solution containing 5 mm IAA and 2% agar was injected into a tapered borosilicate glass pipette with a tip diameter of 100 μm. The solution was allowed to gel, and the filled pipette was then placed in a Petri dish containing ¼ MS solution and allowed to stabilize for 30 min (creating a radial concentration gradient at the exposed pipette tip). A SR IAA sensor was placed at the surface of the filled pipette with the aid of a stepper motor-video zoomscope apparatus. The SR sensor was then operated at a frequency of 0.33 Hz (ΔX = 30 μm), and abiotic diffusive flux in the radial direction (perpendicular to the tangent of the pipette tip) was recorded. The sensor was then positioned farther away from the surface, and diffusive flux recorded. This ‘step-back’ protocol was repeated until there was no significant flux of IAA (less than ±2% change in flux). An empirical model was developed based on the concentration profile in the bath solution, and predicted flux was compared with measured flux during the SBE (McLamore et al., 2009).
3H-IAA transport, cell viability, qRT-PCR, and NPA histochemical staining assays
3H-IAA transport assays were conducted as described previously (Multani et al., 2003; Peer and Murphy, 2007), except that maize seedlings were secured in place by attachment of the seed with 1% agarose. Cell viability assays were performed per the manufacturer’s protocol and imaged utilizing a Nikon E800 and Zeiss LSM710 confocal laser scanning microscope. qRT-PCR analyses of BR2/ABCB1 gene expression was performed as described previously (Multani et al., 2003; Peer et al., 2004) using the primers 5′-GCCGCGTAGGACGGAATG-3′; 5′-TGCGATCATGGAGTACCACC-3′ and UBI1 (5′-AGCTGCCGATGTGCCTGCGTCG-3′; 5′-GAAAGA-CAGAACATAATGAGCACAG-3′) as a normalization standard. NPA histochemical staining was performed as described (Murphy and Taiz, 1999).
IAA flux in Zea mays roots
B73 and br2 caryopses were sterilized with 0.5% NaOCl and rinsed three times with ddH2O and placed in paper towels soaked in tap water, and then sealed in plastic pouches for 3 days to germinate. The head space within the storage bag was replaced 3–4 times per day with fresh air. Seeds were then removed and placed upright (root down) in hydroponic solutions containing ¼ MS media (20.6 mm NH4NO3, 0.1 mm H3BO3, 4.0 mm CaCl2·H2O, 0.1 μm CoCl2·6 H2O, 1.5 mm MgSO4·7 H2O, 0.1 μm CuSO4·5 H2O, 1.2 mm KH2PO4, 0.1 mm FeSO4·7 H2O, 18.8 mm KNO3, 0.1 mm MnSO4·4 H2O, 5.0 μm KI, 1.0 μm Na2MoO4·2 H2O, 30 μm ZnSO4·7 H2O). Four days after germination, roots (1.5–2.0 cm long) were transferred to 7 ml Petri dishes containing ¼ MS media for quantification of physiological IAA flux.
The root surface was scanned in the direction perpendicular to the tangent of the surface, and ΔC/ΔX values used to calculate diffusive flux using Fick’s first law of diffusion. All experiments were conducted at 25 ± 0.2°C with an oxygen concentration of 150 ± 5 μm O2 in the dark. The diffusion coefficient used for IAA data analysis was 7 × 10−6 cm2 sec−1 (Sussman and Goldsmith, 1981), and 2.1 × 104 cm2 sec−1 was used for all O2 optrode data analysis. During profiles of root IAA flux, the light was turned on (<10 μmol m−2 sec−1 photon flux) only briefly to adjust the spatial location for no longer than 20 sec, and immediately turned off to avoid IAA degradation. For induced IAA flux studies, root profiles were spatially mapped continuously as external IAA was added (1 mm stock solution in 0.1% DMSO w/w). For inhibitor studies, endogenous IAA flux was continuously measured at the elongation zone for 20 min, and then 5 μm NPA or NOA was added to inhibit IAA transport. For all IAA flux data, error bars represent ±2 standard errors of the arithmetic mean, and n-values are reported where applicable.
Integrated flux was determined by calculating the area under real time IAA flux curves over a time period of 6 sec. This was calculated by summing the integrated flux in the positive limit (efflux) and the negative limit (influx). Calculation of absolute integrated IAA flux was calculated by summing the integrated IAA flux in the positive limit and the negative limit (absolute value of integrated influx). For an example of integration of real time flux at the elongation zone of a B73 root, see Figure S3.
where J = Measured IAA flux [fmol cm−2 sec−1], Δt = Temporal resolution of SR sensor [min], t1 = x-intercept at leading edge of oscillation, t1 = x-intercept at leading edge of oscillation, t2 = x-intercept at key point (center) of oscillation, t3 = x-intercept at lagging edge of oscillation.
Endogenous IAA integrated flux (iFlux) was empirically modeled using a simple harmonic oscillation model for continuous recordings from three roots (Equation 2).
where iFluxave= Average integrated IAA flux prior to acrolein exposure [fmol cm−2], a = amplitude of oscillation [fmol cm−2], T = oscillation period [min], and t = time [min]. Where applicable, phase angle was calculated using Equation 3.
where φ = phase angle between integrated efflux/influx [°], and Δtpeak = shift in time at peak of oscillation [min].
Root growth rate measurements
Imaging was conducted using a Pulnix progressive scan camera/zoomscope (Pulnix, http://www.jai.com) for on a stage controlled by three-dimensional linear stepper motors mounted on a vibration isolation table with a Faraday cage (Technical Manufacturing Co.) mounted on a head stage controlled by a motion control system. The framegrabber card (PCI-1428) was obtained from National Instruments (http://www.ni.com). Bright field and stereo lighting were used, and evenly distributed, non-glaring illumination was verified using a Quantum meter (Apogee Instruments). Difference imaging video technique (DVIT) software was supplied by Science Wares. DVIT software was used to control video imaging hardware and image acquisition. Pixel changes were recorded at a 1 kHz acquisition rate in the visible spectra, and image subtraction was conducted against a reference image according to McLamore et al. (2010a). Processed images were displayed against a grey background, and no convolution kernels or image alteration were used during post-processing.
Root elongation rate was measured using a technique based on Verslues et al. (1998). Perforated root guides were constructed (6 mm inside diameter with 0.5 mm perforations), and root guides were placed 5 mm above a bath containing ¼ MS media which was vigorously aerated with a Rena Air 200 air pump and air stones (Rena, http://www.plantrena.com). Root elongation rate was quantified by marking the position of the root tip on the root guide over time, and later measuring the distance between markings.
Funding for this work was provided by the National Science Foundation and National Aeronautics and Space Administration to DMP and by the US Department of Energy DE-FG02–06ER15804 to ASM.