www.MaterialsViews.com
Gold Nanoparticles
In Planta Response of Arabidopsis to Photothermal Impact Mediated by Gold Nanoparticles Yeonjong Koo, Ekaterina Y. Lukianova-Hleb, Joann Pan, Sean M. Thompson, Dmitri O. Lapotko, and Janet Braam*
Biological responses to photothermal effects of gold nanoparticles (GNPs) have been demonstrated and employed for various applications in diverse systems except for one important class – plants. Here, the uptake of GNPs through Arabidopsis thaliana roots and translocation to leaves are reported. Successful plasmonic nanobubble generation and acoustic signal detection in planta is demonstrated. Furthermore, Arabidopsis leaves harboring GNPs and exposed to continuous laser or noncoherent light show elevated temperatures across the leaf surface and induced expression of heat-shock regulated genes. Overall, these results demonstrate that Arabidopsis can readily take up GNPs through the roots and translocate the particles to leaf tissues. Once within leaves, GNPs can act as photothermal agents for on-demand remote activation of localized biological processes in plants.
1. Introduction Metal plasmonic nanoparticles are robust photothermal converters acting through the mechanism of surface plasmon resonance.[1–5] This unique mechanism enables precise manipulation of thermal energy at nanoscale, including in biomedical applications.[6] Under stationary optical excitation, metal nanoparticles generate local heat via thermal diffusion and thus impose thermal impact to living systems.[4–7] Under nonstationary optical excitation with a short laser pulse, metal nanoparticles generate plasmonic nanobubbles,
Dr. Y. Koo, Dr. E. Y. Lukianova-Hleb, J. Pan, Prof. D. O. Lapotko, Prof. J. Braam Department of BioSciences Rice University Houston, TX 77005, USA E-mail: [email protected] Dr. S. M. Thompson Department of Horticultural Sciences Texas A&M University College Station, TX 77843, USA Prof. D. O. Lapotko Department of Physics and Astronomy Rice University Houston, TX 77005, USA DOI: 10.1002/smll.201502461 small 2016, 12, No. 5, 623–630
expanding and collapsing vapor bubbles,[8–10] which can deliver mechanical impact to biological tissues. Among metal nanoparticles, gold colloids[11] provide the best biocompatibility and therefore have been widely used in various biomedical studies, clinical trials, and the clinic.[12,13] Although a wide spectrum of biomedical applications exists for gold nanoparticle (GNP) photothermal phenomena, practical applications in plant biology have not yet been investigated. The safety of gold colloids coupled with the potential ease of their administration to plants via the root system opens many interesting opportunities for plant applications. We therefore investigated the potential of the stationary and nonstationary photothermal effects of gold colloids in the well-established plant model, Arabidopsis thaliana, and examined GNP uptake and photothermally activated biological responses.
2. Results 2.1. Gold Nanoparticles Can Be Detected Acoustically in Arabidopsis Leaves with Plasmonic Nanobubbles To determine the conditions to detect GNPs in plant tissues, one ppm (= 4.8 × 108 GNP mL−2) of 60 nm GNPs was injected directly into Arabidopsis leaves through the lower (abaxial) surface, and a single 532 nm laser pulse of 70 ps duration and 140 mJ cm−2 laser pulse fluence was applied to the same © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
623
full papers www.MaterialsViews.com
area. These initial conditions were chosen based on previous detection of plasmonic nanobubbles around individual 60 nm GNPs in water (Figure S1, Supporting Information).[14] Positive signals from plasmonic nanobubbles in leaf tissue were clearly distinguishable from background signals of control leaves without GNPs (Figure S1, Supporting Information).
acoustical signals correlates with bubble size, we conclude that single 60 nm GNPs are detectable in planta by this methodology. With this methodology, GNP detection data can be obtained quickly, avoiding the special protocols, reagents, and time of preparation typically required for trace metal analysis or electron microscopy.
2.2. Spatial Distribution of Gold Nanoparticles in Leaves Taken Up through Vascular Tissue
2.3. Uptake of Gold Nanoparticles through Arabidopsis Roots and Translocation to Leaves
To test whether GNPs could be taken up into Arabidopsis leaves through the vascular tissues, mature leaves were detached from plants and the petioles were submerged into media with or without GNPs. After 5 h of exposure, leaves were analyzed by laser scanning. Laser pulses, with a beam diameter of 0.1 mm, were applied every 0.5 mm, resulting in ≈ 12.5% of leaf surface area surveyed. We optimized the conditions, using 532 nm laser pulses of 20 ps duration and laser pulse fluence 281 mJ cm−2, and found that the great majority (99%) of the background signal amplitudes fell below 90 mV, with a major background peak at 75 mV (Figure S2, Supporting Information). Therefore, only signals of 90 mV and above were used as above background data in our GNP plant detection analyses. To spatially resolve GNP distribution, we mapped the acoustic signals over photographic images of the analyzed leaves (Figure 1). As shown in Figure 1B, a representative detached leaf whose petiole was exposed to media containing 50 ppm of GNPs for 5 h, displayed robust (≥90 mV) acoustic signals over 15% of the scanned sites. A control leaf similarly treated, but without GNP exposure, showed only background signals (Figure 1A). Most (≥90%) of the signals detected in the GNP-treated leaf had amplitudes between 90 and 200 mV (Figure 2A). The plasmonic nanobubble-specific signals detected in the GNP-treated leaves were comparable to those generated by single GNPs in water and above background of GNPfree plant tissues. Because the amplitude of the detected
Although there is some limited evidence that intact plant roots can take up and translocate nanoparticles,[15–18] there remains some controversy[19–21] and GNP interaction with intact Arabidopsis has not yet been demonstrated. To assess GNP uptake and translocation, we transferred 4 week old hydroponically grown Arabidopsis to growth media harboring GNPs and, after 5 h of exposure, scanned the leaves with individual 532 nm laser pulses (20 ps, 281 mJ cm−2). Similar to the leaves whose petioles were directly immersed in media with GNPs (Figure 1B), leaves from plants whose roots were exposed to GNPs for 5 h also showed strong plasmonic nanobubble-specific acoustic signals across the entire leaf (Figure 1C), indicating that GNPs are readily taken up by the roots and translocated to leaves. Translocation likely occurred through similar mechanisms in detached leaves and nondetached leaves because the overall patterns of distribution and the signal strength distributions were similar whether uptake of the GNP was through a cut petiole or root system (Figure 1B,C). These results indicate that 60 nm GNPs can be absorbed through the Arabidopsis root and translocated through the vascular system, suggesting that particles of this size do not obstruct the xylem transport system. Understanding how the GNPs are transported across root cell layers will be of future interest. Plants have the ability to aggregate metal ions forming nanoparticles. To assess whether GNPs aggregate into clusters after uptake into Arabidopsis tissues, we compared leaf acoustic signal distributions from Arabidopsis leaves exposed to media with high (2.4 × 1010 NP mL−2) or low (4.8 × 108 NP mL−2) GNP concentrations. The high GNP concentration increased the percentage of leaf surface area that emitted plasmonic nanobubble-specific signals from 18% to 25% (Figure 2C), but regardless of concentration nearly all the signal remained in the 90–200 mV amplitude range (Figure 2A,C). Therefore, a greater GNP concentration led to enhanced GNP uptake, but not to an increase in frequency of signals detected with high amplitude. A lack of high-amplitude signals suggests that GNPs did not aggregate in planta. Similar results were obtained with leaves from whole plants whose roots were incubated with GNPs (Figure 2B,C and Figure S3, Supporting Information). The higher concentration of GNPs in growth media led to an increase in the percentage of leaf surface emitting plasmonic nanobubble-specific acoustic signals (from 7% to 14%) (Figure 2B,C), but the maximal amplitude of signal was not significantly affected (Figure 2B,C). A longer incubation time with the high concentration of GNP led to 24.5% of the leaf surface emitting signal, but
Figure 1. GNP distribution pattern in Arabidopsis leaves. A–C) Quantitative acoustic signals amplitudes (measured in mV) are displayed as contour lines overlaid on leaf photographic images. Representative leaves whose petioles were exposed to media A) lacking GNPs or B) with 50 ppm GNPs are shown. C) Leaf from intact Arabidopsis plant whose roots were exposed to 50 ppm GNP for 5 h is shown.
624 www.small-journal.com
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2016, 12, No. 5, 623–630
www.MaterialsViews.com
Figure 2. Statistical analysis of acoustic signals detected from GNPs in Arabidopsis leaves. A,B) The frequency of leaf signal amplitudes are compared between A) high and low GNP concentration exposure to detached leaf petioles and B) high and low GNP concentration exposure to whole plants for two different durations. Signal amplitudes below 200 mV and above 200 mV are indicated on upper side of each graph. C) Percentage of leaf surface that emitted detectable signal (% surface with signal, x-axis) and acoustic signal amplitude (average signal amplitude over 90 mV—average signal amplitude below 90 mV, y-axis) from (A) and (B) are plotted. Detached leaf data are shown in green; whole plant exposure data are shown in orange.
the signal amplitudes again were not significantly affected (Figure S3B, Supporting Information). Under all experimental conditions, including direct GNP absorption through the leaf petiole, strong signals over 200 mV were fewer than 23% of total positive signals (Figure 2A,B). The detection sites were primarily localized near the leaf vein or close to the leaf margins (Figure 1B,C). Overall, these results suggest that the GNPs travel through the Arabidopsis leaf veins, which have xylem vessel diameters of 5–25 µm,[22] and tend to remain unaggregated. Similar results were reported for plant uptake and translocation of fluorescence nanoparticles.[16] The similarity in patterns of GNP localization in leaves regardless of whether the GNPs were introduced through either the roots or cut petioles is consistent with the interpretation that particles translocated from the roots move through the vascular system and not symplastically from cell to cell. Furthermore, this work suggests that GNPs are relatively stable in the plant tissue and do not have aggregation tendencies because nearly 90% of all GNPs detected had signals consistent with single particles. small 2016, 12, No. 5, 623–630
2.4. Stationary Photothermal Impact—Thermal Imaging of Leaves Because GNP interaction with light can create heat, we examined whether light could be used to trigger localized heat production through GNP interaction. First, using thermosensitive imaging, we compared temperatures of leaves whose petioles were incubated in media with or without GNPs and exposed to transient (60 s) 532 nm laser light at three different intensities of laser radiation. As shown in Figure 3A, the leaves pretreated with GNPs displayed elevated temperatures compared to leaves that lacked GNPs. The temperature change was distributed broadly across the leaf surface, with the greatest intensity near the leaf middle. These results indicate that GNPs incorporated into leaves and excited by continuous laser light for 60 s can generate detectable photothermal effects by increasing tissue temperature via conversion of the absorbed laser energy into the heat. To quantitatively compare laser-induced temperature changes and assess temperature variation as a function of intensity of laser radiation, we assayed maximal
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
625
full papers www.MaterialsViews.com
Figure 3. Detection of GNP-dependent laser-induced temperature. A) Representative false-color images of leaf pairs whose petioles were placed in media either with or without GNPs, obtained with a thermosensitive camera after 60 s of 34, 23, or 11 mW cm−2 of laser light. B) Maximal leaf temperatures attained as a function of laser light intensity are graphed. Solid (red, orange, and yellow) lines and dashed (blue) lines indicate GNP-treated and untreated leaves, respectively (n = 6). Significant differences in maximal leaf temperature dependent upon GNP exposure were detected at 23 and 34 mW cm−2 light intensity after 30 s of laser exposure (p < 0.05). No significant differences dependent on GNP exposure were found at 11 mW cm−2 laser light intensity. C) Maximal leaf temperatures detected over time after exposure to different GNP concentrations. D) The percent leaf area with temperatures over 24 °C at 30 s laser treatment dependent on GNP concentration. E) GNP-dependent heat production in leaves exposed to noncoherent light sources, fluorescence bulb, sunlight, or metal halide lamp. Left panel illustrates the experimental setup, with detached leaves arranged with petioles submerged in 1/16 strength Hoagland solution to prevent dehydration; four leaves served as non-GNPtreated controls and four leaves were preexposed to 10 ppm GNPs for 24 h. False-color thermo leaf images at 2.2 mW cm−2 of fluorescent light intensities (second panel), 15 mW cm−2 of sun light intensity (third panel), and 2.2 mW cm−2 of metal halide lamp light intensity.
laser-induced temperature obtained over time (Figure 3B) in GNP-treated and GNP-untreated leaves and under different conditions of laser excitation. The higher the light intensity, the greater the heat emission detected in both control and GNP-treated leaves (Figure 3B). Laser light in the absence of GNPs was also sufficient to increase leaf temperature indicating that the light energy alone produced heat over the threshold capacity of transpirational leaf cooling, with the greatest temperature change occurring over the first 25 s of laser light exposure. While the temperature of leaves with GNPs reached highest at 34 mW cm−2, the greatest temperature difference attributed to GNP was seen at 23 mW cm−2 (Figure 3B).
626 www.small-journal.com
We next compared the effect of increasing GNP concentration at constant intensity of the laser radiation (23 mW cm−2) on maximal laser-induced leaf temperature. At low GNP concentration (1 ppm), heat emission upon laser light exposure had no significant effect, whereas 10 ppm GNP was sufficient to induce a significant laserinduced temperature change, with no further increase at higher (25 ppm) GNPs (Figure 3C). The overall proportion of leaf tissue reporting elevated temperature (i.e., greater than 24 °C) was also significantly higher for leaves exposed to 10 or 25 ppm GNPs than control leaves without GNP incorporation (Figure 3D). Therefore, both the laserinduced temperature peak reached and the extent of
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2016, 12, No. 5, 623–630
www.MaterialsViews.com
temperature change across the leaf area were enhanced by the presence of GNPs.
2.5. Response to the Stationary Photothermal Effects of Noncoherent Optical Irradiation in GNP-Treated Plant Leaves We sought to determine whether GNP-treated plants exposed to ambient light sources, such as natural sunlight or artificial light sources, also experience and respond to elevated temperature due to GNP-dependent light energy conversion to heat. Therefore, we compared temperature readings from control leaves to GNP-treated leaves after exposure to fluorescent light, sunlight, or metal halide lamps (Figure 3E and Figure S4, Supporting Information). Detached leaf petioles were placed in 1/16 strength Hoagland solution droplets to maintain transpiration and prevent tissue desiccation; the petioles and proximal leaf therefore remained relatively cool and are false-colored blue in the thermosensitive camera images (Figure 3E and Figure S4, Supporting Information). Under either low (0.22 mW cm−2) or high (1.1–2.2 mW cm−2) intensity fluorescent light, no temperature differentials were apparent for the GNP-treated versus control leaves (Figure 3E). The lack of a GNP effect is likely because fluorescence bulbs emit only weakly in the green wavelengths, which excite the GNPs (Figure S5, Supporting Information). To test the effects of sunlight on leaf temperature while maintaining cool ambient room temperature of 23 °C, detached leaves were exposed to sunlight at midday through glass. Under these conditions, one-minute exposure to sunlight with moderate cloud cover (6.6 mW cm−2) or without cloud cover (15 mW cm−2) resulted in maximal temperature readings up to 0.5 °C higher in the distal portions of the GNPtreated leaves than the leaves lacking GNPs (Figure 3E and Figure S4 (Supporting Information), and Table 1). We also tested metal halide light exposure because these lamps emit strongly in the green spectrum (Figure S5, Supporting Information). Metal halide lamp light exposure at 23 °C resulted in significant differential temperature increases in the GNPtreated leaf over control leaves at high intensity (over 2 mW cm−2) but not lower light intensity (below 1.7 mW cm−2) (Figure 3E and Figure S4D, Supporting Information). Therefore, nonlaser light is also sufficient to result in GNPmediated heat production.
Figure 4. Expression of heat-responsive Arabidopsis genes HSP17.4 and HSP90. A) Transcript accumulation of HSP17.4 and HSP90 were quantified by qPCR in leaves left untreated (NT), exposed to 10 ppm GNPs for 24 h, treated with 1 h of 23 mW cm−2 laser light, or exposed to GNP and 1 h of 23 mW cm−2 laser light (83 J cm−2) (n ≥ 3). B) Left panel: Beta-glucuronidase activity in representative PHSP17.4-GUS transgenic leaves left untreated (NT), exposed to 37 °C (heat) for 1 h, exposed to 1.5 h of 23 mW cm−2 laser light alone (120 J cm−2), or exposed to 1.5 h of 23 mW cm−2 laser light after treatment with 10 ppm GNPs for 24 h. Right panel: False-color images of the pair of leaves treated with laser light without or with GNPs as show in left panel.
2.6. Biological Response of Arabidopsis Leaf Tissues to the Stationary Photothermal Impact To test whether the light-induced temperatures detected in GNP-treated leaves are biologically active, we examined the alteration in expression levels of plant genes known to be induced by high temperature. Both HSP17.4 and HSP90 transcript levels were significantly higher in leaves subjected to both GNP and laser light than control leaves, GNP-treated leaves, or laser light-treated leaves (Figure 4A). HSP17.4 showed the most robust response with over a threefold
Table 1. Leaf maximum temperature comparison between GNP and no GNP absorbing Arabidopsis leaves. * indicates significance. Light intensity [mW cm−2]
Low light
0.22
Temperatures [°C] GNP −
GNP +
24 ± 0.1
24 ± 0.1
Difference [°C]
t-test
0
0.5 0.32
1.1
22.3± 0.2
22.5 ± 0.2
0.2
2.2
23.1 ± 0.2
23.1 ± 0.1
0
1
Sun
6.6
34.0 ± 0.2
34.6 ± 0.1
0.6*
Gold Nanoparticles
In Planta Response of Arabidopsis to Photothermal Impact Mediated by Gold Nanoparticles Yeonjong Koo, Ekaterina Y. Lukianova-Hleb, Joann Pan, Sean M. Thompson, Dmitri O. Lapotko, and Janet Braam*
Biological responses to photothermal effects of gold nanoparticles (GNPs) have been demonstrated and employed for various applications in diverse systems except for one important class – plants. Here, the uptake of GNPs through Arabidopsis thaliana roots and translocation to leaves are reported. Successful plasmonic nanobubble generation and acoustic signal detection in planta is demonstrated. Furthermore, Arabidopsis leaves harboring GNPs and exposed to continuous laser or noncoherent light show elevated temperatures across the leaf surface and induced expression of heat-shock regulated genes. Overall, these results demonstrate that Arabidopsis can readily take up GNPs through the roots and translocate the particles to leaf tissues. Once within leaves, GNPs can act as photothermal agents for on-demand remote activation of localized biological processes in plants.
1. Introduction Metal plasmonic nanoparticles are robust photothermal converters acting through the mechanism of surface plasmon resonance.[1–5] This unique mechanism enables precise manipulation of thermal energy at nanoscale, including in biomedical applications.[6] Under stationary optical excitation, metal nanoparticles generate local heat via thermal diffusion and thus impose thermal impact to living systems.[4–7] Under nonstationary optical excitation with a short laser pulse, metal nanoparticles generate plasmonic nanobubbles,
Dr. Y. Koo, Dr. E. Y. Lukianova-Hleb, J. Pan, Prof. D. O. Lapotko, Prof. J. Braam Department of BioSciences Rice University Houston, TX 77005, USA E-mail: [email protected] Dr. S. M. Thompson Department of Horticultural Sciences Texas A&M University College Station, TX 77843, USA Prof. D. O. Lapotko Department of Physics and Astronomy Rice University Houston, TX 77005, USA DOI: 10.1002/smll.201502461 small 2016, 12, No. 5, 623–630
expanding and collapsing vapor bubbles,[8–10] which can deliver mechanical impact to biological tissues. Among metal nanoparticles, gold colloids[11] provide the best biocompatibility and therefore have been widely used in various biomedical studies, clinical trials, and the clinic.[12,13] Although a wide spectrum of biomedical applications exists for gold nanoparticle (GNP) photothermal phenomena, practical applications in plant biology have not yet been investigated. The safety of gold colloids coupled with the potential ease of their administration to plants via the root system opens many interesting opportunities for plant applications. We therefore investigated the potential of the stationary and nonstationary photothermal effects of gold colloids in the well-established plant model, Arabidopsis thaliana, and examined GNP uptake and photothermally activated biological responses.
2. Results 2.1. Gold Nanoparticles Can Be Detected Acoustically in Arabidopsis Leaves with Plasmonic Nanobubbles To determine the conditions to detect GNPs in plant tissues, one ppm (= 4.8 × 108 GNP mL−2) of 60 nm GNPs was injected directly into Arabidopsis leaves through the lower (abaxial) surface, and a single 532 nm laser pulse of 70 ps duration and 140 mJ cm−2 laser pulse fluence was applied to the same © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
623
full papers www.MaterialsViews.com
area. These initial conditions were chosen based on previous detection of plasmonic nanobubbles around individual 60 nm GNPs in water (Figure S1, Supporting Information).[14] Positive signals from plasmonic nanobubbles in leaf tissue were clearly distinguishable from background signals of control leaves without GNPs (Figure S1, Supporting Information).
acoustical signals correlates with bubble size, we conclude that single 60 nm GNPs are detectable in planta by this methodology. With this methodology, GNP detection data can be obtained quickly, avoiding the special protocols, reagents, and time of preparation typically required for trace metal analysis or electron microscopy.
2.2. Spatial Distribution of Gold Nanoparticles in Leaves Taken Up through Vascular Tissue
2.3. Uptake of Gold Nanoparticles through Arabidopsis Roots and Translocation to Leaves
To test whether GNPs could be taken up into Arabidopsis leaves through the vascular tissues, mature leaves were detached from plants and the petioles were submerged into media with or without GNPs. After 5 h of exposure, leaves were analyzed by laser scanning. Laser pulses, with a beam diameter of 0.1 mm, were applied every 0.5 mm, resulting in ≈ 12.5% of leaf surface area surveyed. We optimized the conditions, using 532 nm laser pulses of 20 ps duration and laser pulse fluence 281 mJ cm−2, and found that the great majority (99%) of the background signal amplitudes fell below 90 mV, with a major background peak at 75 mV (Figure S2, Supporting Information). Therefore, only signals of 90 mV and above were used as above background data in our GNP plant detection analyses. To spatially resolve GNP distribution, we mapped the acoustic signals over photographic images of the analyzed leaves (Figure 1). As shown in Figure 1B, a representative detached leaf whose petiole was exposed to media containing 50 ppm of GNPs for 5 h, displayed robust (≥90 mV) acoustic signals over 15% of the scanned sites. A control leaf similarly treated, but without GNP exposure, showed only background signals (Figure 1A). Most (≥90%) of the signals detected in the GNP-treated leaf had amplitudes between 90 and 200 mV (Figure 2A). The plasmonic nanobubble-specific signals detected in the GNP-treated leaves were comparable to those generated by single GNPs in water and above background of GNPfree plant tissues. Because the amplitude of the detected
Although there is some limited evidence that intact plant roots can take up and translocate nanoparticles,[15–18] there remains some controversy[19–21] and GNP interaction with intact Arabidopsis has not yet been demonstrated. To assess GNP uptake and translocation, we transferred 4 week old hydroponically grown Arabidopsis to growth media harboring GNPs and, after 5 h of exposure, scanned the leaves with individual 532 nm laser pulses (20 ps, 281 mJ cm−2). Similar to the leaves whose petioles were directly immersed in media with GNPs (Figure 1B), leaves from plants whose roots were exposed to GNPs for 5 h also showed strong plasmonic nanobubble-specific acoustic signals across the entire leaf (Figure 1C), indicating that GNPs are readily taken up by the roots and translocated to leaves. Translocation likely occurred through similar mechanisms in detached leaves and nondetached leaves because the overall patterns of distribution and the signal strength distributions were similar whether uptake of the GNP was through a cut petiole or root system (Figure 1B,C). These results indicate that 60 nm GNPs can be absorbed through the Arabidopsis root and translocated through the vascular system, suggesting that particles of this size do not obstruct the xylem transport system. Understanding how the GNPs are transported across root cell layers will be of future interest. Plants have the ability to aggregate metal ions forming nanoparticles. To assess whether GNPs aggregate into clusters after uptake into Arabidopsis tissues, we compared leaf acoustic signal distributions from Arabidopsis leaves exposed to media with high (2.4 × 1010 NP mL−2) or low (4.8 × 108 NP mL−2) GNP concentrations. The high GNP concentration increased the percentage of leaf surface area that emitted plasmonic nanobubble-specific signals from 18% to 25% (Figure 2C), but regardless of concentration nearly all the signal remained in the 90–200 mV amplitude range (Figure 2A,C). Therefore, a greater GNP concentration led to enhanced GNP uptake, but not to an increase in frequency of signals detected with high amplitude. A lack of high-amplitude signals suggests that GNPs did not aggregate in planta. Similar results were obtained with leaves from whole plants whose roots were incubated with GNPs (Figure 2B,C and Figure S3, Supporting Information). The higher concentration of GNPs in growth media led to an increase in the percentage of leaf surface emitting plasmonic nanobubble-specific acoustic signals (from 7% to 14%) (Figure 2B,C), but the maximal amplitude of signal was not significantly affected (Figure 2B,C). A longer incubation time with the high concentration of GNP led to 24.5% of the leaf surface emitting signal, but
Figure 1. GNP distribution pattern in Arabidopsis leaves. A–C) Quantitative acoustic signals amplitudes (measured in mV) are displayed as contour lines overlaid on leaf photographic images. Representative leaves whose petioles were exposed to media A) lacking GNPs or B) with 50 ppm GNPs are shown. C) Leaf from intact Arabidopsis plant whose roots were exposed to 50 ppm GNP for 5 h is shown.
624 www.small-journal.com
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2016, 12, No. 5, 623–630
www.MaterialsViews.com
Figure 2. Statistical analysis of acoustic signals detected from GNPs in Arabidopsis leaves. A,B) The frequency of leaf signal amplitudes are compared between A) high and low GNP concentration exposure to detached leaf petioles and B) high and low GNP concentration exposure to whole plants for two different durations. Signal amplitudes below 200 mV and above 200 mV are indicated on upper side of each graph. C) Percentage of leaf surface that emitted detectable signal (% surface with signal, x-axis) and acoustic signal amplitude (average signal amplitude over 90 mV—average signal amplitude below 90 mV, y-axis) from (A) and (B) are plotted. Detached leaf data are shown in green; whole plant exposure data are shown in orange.
the signal amplitudes again were not significantly affected (Figure S3B, Supporting Information). Under all experimental conditions, including direct GNP absorption through the leaf petiole, strong signals over 200 mV were fewer than 23% of total positive signals (Figure 2A,B). The detection sites were primarily localized near the leaf vein or close to the leaf margins (Figure 1B,C). Overall, these results suggest that the GNPs travel through the Arabidopsis leaf veins, which have xylem vessel diameters of 5–25 µm,[22] and tend to remain unaggregated. Similar results were reported for plant uptake and translocation of fluorescence nanoparticles.[16] The similarity in patterns of GNP localization in leaves regardless of whether the GNPs were introduced through either the roots or cut petioles is consistent with the interpretation that particles translocated from the roots move through the vascular system and not symplastically from cell to cell. Furthermore, this work suggests that GNPs are relatively stable in the plant tissue and do not have aggregation tendencies because nearly 90% of all GNPs detected had signals consistent with single particles. small 2016, 12, No. 5, 623–630
2.4. Stationary Photothermal Impact—Thermal Imaging of Leaves Because GNP interaction with light can create heat, we examined whether light could be used to trigger localized heat production through GNP interaction. First, using thermosensitive imaging, we compared temperatures of leaves whose petioles were incubated in media with or without GNPs and exposed to transient (60 s) 532 nm laser light at three different intensities of laser radiation. As shown in Figure 3A, the leaves pretreated with GNPs displayed elevated temperatures compared to leaves that lacked GNPs. The temperature change was distributed broadly across the leaf surface, with the greatest intensity near the leaf middle. These results indicate that GNPs incorporated into leaves and excited by continuous laser light for 60 s can generate detectable photothermal effects by increasing tissue temperature via conversion of the absorbed laser energy into the heat. To quantitatively compare laser-induced temperature changes and assess temperature variation as a function of intensity of laser radiation, we assayed maximal
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
625
full papers www.MaterialsViews.com
Figure 3. Detection of GNP-dependent laser-induced temperature. A) Representative false-color images of leaf pairs whose petioles were placed in media either with or without GNPs, obtained with a thermosensitive camera after 60 s of 34, 23, or 11 mW cm−2 of laser light. B) Maximal leaf temperatures attained as a function of laser light intensity are graphed. Solid (red, orange, and yellow) lines and dashed (blue) lines indicate GNP-treated and untreated leaves, respectively (n = 6). Significant differences in maximal leaf temperature dependent upon GNP exposure were detected at 23 and 34 mW cm−2 light intensity after 30 s of laser exposure (p < 0.05). No significant differences dependent on GNP exposure were found at 11 mW cm−2 laser light intensity. C) Maximal leaf temperatures detected over time after exposure to different GNP concentrations. D) The percent leaf area with temperatures over 24 °C at 30 s laser treatment dependent on GNP concentration. E) GNP-dependent heat production in leaves exposed to noncoherent light sources, fluorescence bulb, sunlight, or metal halide lamp. Left panel illustrates the experimental setup, with detached leaves arranged with petioles submerged in 1/16 strength Hoagland solution to prevent dehydration; four leaves served as non-GNPtreated controls and four leaves were preexposed to 10 ppm GNPs for 24 h. False-color thermo leaf images at 2.2 mW cm−2 of fluorescent light intensities (second panel), 15 mW cm−2 of sun light intensity (third panel), and 2.2 mW cm−2 of metal halide lamp light intensity.
laser-induced temperature obtained over time (Figure 3B) in GNP-treated and GNP-untreated leaves and under different conditions of laser excitation. The higher the light intensity, the greater the heat emission detected in both control and GNP-treated leaves (Figure 3B). Laser light in the absence of GNPs was also sufficient to increase leaf temperature indicating that the light energy alone produced heat over the threshold capacity of transpirational leaf cooling, with the greatest temperature change occurring over the first 25 s of laser light exposure. While the temperature of leaves with GNPs reached highest at 34 mW cm−2, the greatest temperature difference attributed to GNP was seen at 23 mW cm−2 (Figure 3B).
626 www.small-journal.com
We next compared the effect of increasing GNP concentration at constant intensity of the laser radiation (23 mW cm−2) on maximal laser-induced leaf temperature. At low GNP concentration (1 ppm), heat emission upon laser light exposure had no significant effect, whereas 10 ppm GNP was sufficient to induce a significant laserinduced temperature change, with no further increase at higher (25 ppm) GNPs (Figure 3C). The overall proportion of leaf tissue reporting elevated temperature (i.e., greater than 24 °C) was also significantly higher for leaves exposed to 10 or 25 ppm GNPs than control leaves without GNP incorporation (Figure 3D). Therefore, both the laserinduced temperature peak reached and the extent of
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2016, 12, No. 5, 623–630
www.MaterialsViews.com
temperature change across the leaf area were enhanced by the presence of GNPs.
2.5. Response to the Stationary Photothermal Effects of Noncoherent Optical Irradiation in GNP-Treated Plant Leaves We sought to determine whether GNP-treated plants exposed to ambient light sources, such as natural sunlight or artificial light sources, also experience and respond to elevated temperature due to GNP-dependent light energy conversion to heat. Therefore, we compared temperature readings from control leaves to GNP-treated leaves after exposure to fluorescent light, sunlight, or metal halide lamps (Figure 3E and Figure S4, Supporting Information). Detached leaf petioles were placed in 1/16 strength Hoagland solution droplets to maintain transpiration and prevent tissue desiccation; the petioles and proximal leaf therefore remained relatively cool and are false-colored blue in the thermosensitive camera images (Figure 3E and Figure S4, Supporting Information). Under either low (0.22 mW cm−2) or high (1.1–2.2 mW cm−2) intensity fluorescent light, no temperature differentials were apparent for the GNP-treated versus control leaves (Figure 3E). The lack of a GNP effect is likely because fluorescence bulbs emit only weakly in the green wavelengths, which excite the GNPs (Figure S5, Supporting Information). To test the effects of sunlight on leaf temperature while maintaining cool ambient room temperature of 23 °C, detached leaves were exposed to sunlight at midday through glass. Under these conditions, one-minute exposure to sunlight with moderate cloud cover (6.6 mW cm−2) or without cloud cover (15 mW cm−2) resulted in maximal temperature readings up to 0.5 °C higher in the distal portions of the GNPtreated leaves than the leaves lacking GNPs (Figure 3E and Figure S4 (Supporting Information), and Table 1). We also tested metal halide light exposure because these lamps emit strongly in the green spectrum (Figure S5, Supporting Information). Metal halide lamp light exposure at 23 °C resulted in significant differential temperature increases in the GNPtreated leaf over control leaves at high intensity (over 2 mW cm−2) but not lower light intensity (below 1.7 mW cm−2) (Figure 3E and Figure S4D, Supporting Information). Therefore, nonlaser light is also sufficient to result in GNPmediated heat production.
Figure 4. Expression of heat-responsive Arabidopsis genes HSP17.4 and HSP90. A) Transcript accumulation of HSP17.4 and HSP90 were quantified by qPCR in leaves left untreated (NT), exposed to 10 ppm GNPs for 24 h, treated with 1 h of 23 mW cm−2 laser light, or exposed to GNP and 1 h of 23 mW cm−2 laser light (83 J cm−2) (n ≥ 3). B) Left panel: Beta-glucuronidase activity in representative PHSP17.4-GUS transgenic leaves left untreated (NT), exposed to 37 °C (heat) for 1 h, exposed to 1.5 h of 23 mW cm−2 laser light alone (120 J cm−2), or exposed to 1.5 h of 23 mW cm−2 laser light after treatment with 10 ppm GNPs for 24 h. Right panel: False-color images of the pair of leaves treated with laser light without or with GNPs as show in left panel.
2.6. Biological Response of Arabidopsis Leaf Tissues to the Stationary Photothermal Impact To test whether the light-induced temperatures detected in GNP-treated leaves are biologically active, we examined the alteration in expression levels of plant genes known to be induced by high temperature. Both HSP17.4 and HSP90 transcript levels were significantly higher in leaves subjected to both GNP and laser light than control leaves, GNP-treated leaves, or laser light-treated leaves (Figure 4A). HSP17.4 showed the most robust response with over a threefold
Table 1. Leaf maximum temperature comparison between GNP and no GNP absorbing Arabidopsis leaves. * indicates significance. Light intensity [mW cm−2]
Low light
0.22
Temperatures [°C] GNP −
GNP +
24 ± 0.1
24 ± 0.1
Difference [°C]
t-test
0
0.5 0.32
1.1
22.3± 0.2
22.5 ± 0.2
0.2
2.2
23.1 ± 0.2
23.1 ± 0.1
0
1
Sun
6.6
34.0 ± 0.2
34.6 ± 0.1
0.6*
