Increased exposure to UV-B radiation during early development leads to enhanced photoprotection and improved long-term performance in Lactuca sativa


J. J. Wargent. E-mail:


Plant responses to solar UV radiation are numerous and have often been considered from a perspective of negative outcomes for plant productivity. In this study, we used two experimental approaches consisting of: (1) field-based spectrally modifying filters in addition to (2) controlled indoor exposure to UV-B, to examine the effects of UV radiation on growth and photosynthetic performance of lettuce (Lactuca sativa L.) seedlings. Various aspects of growth were affected in plants grown under a UV-inclusive environment compared to a UV-depleted environment, including reductions in leaf expansion, increases in leaf thickness and the rate of net photosynthesis. After transplantation to a uniform field environment, lettuce plants initially propagated under the UV-inclusive environment exhibited higher harvestable yields than those from a UV-depleted environment. In controlled conditions, photosynthetic rates were higher in plants grown in the presence of UV-B radiation, and relative growth of plants pre-acclimatized to UV-B was also increased, in addition to higher maximum photochemical efficiency of photosystem II (PSII) (Fv/Fm) following subsequent exposure to high photosynthetically active radiation (PAR) and temperature stress. Our findings are discussed within the context of sustainability in agriculture and the paradigm shift in photobiology which such beneficial responses to UV radiation could represent.


There is a pressing need for new approaches to sustainable crop production in order to feed a growing population in the face of environmental change (Howden et al. 2007). New crop cultivation systems should arguably seek to exploit our improved understanding of fundamental plant biology to maximum effect, particularly plant responses to ubiquitous environmental stimuli. One such stimuli is the light environment, which has been the subject of considerable scientific attention in recent decades. It is clear that plants exhibit specific responses to a wide range of radiation wavelengths, including far red (710–740 nm), red (650–680 nm), blue (400–500 nm), ultraviolet-A (UV-A: 315–400 nm) and ultraviolet-B (UV-B: 280–315 nm) wavebands (e.g. Smith 1995; Eisinger, Bogomolni & Taiz 2003; Paul et al. 2005). Where crops are grown in protected environments such as glasshouses and polytunnels, there is now significant scope to exploit these fundamental light responses by manipulating the light environment reaching the crop. Production in protected environments accounted for around 7% of the total vegetable production area in developed countries in 1995 (Jensen & Malter 1995), with total area utilized for protected cropping increasing by approximately 20% per year on a global basis (Espíet al. 2006). Polyethylene cladding films are typically employed to protect horticultural crops from a range of biotic and abiotic stresses, and to extend cropping seasons (Lamont 2005). However, in recent years, advances in polymer chemistry and manufacturing methods have produced a range of novel films that selectively filter specific radiation wavelengths, and so provide a direct route to exploiting plant light responses for agronomic gain (Rajapakse & Kelly, 1995; Krizek, Clark & Mirecki 2005; Paul et al. 2005).

Previous studies have indicated that several basic aspects of plant development related to crop productivity can be regulated using wavelength-selective films. Physiological and ecological responses mediated by phytochromes are well characterized (Smith 1995; Ballare, Scopel & Sanchez 1997), and cladding materials that increase the ratio of red : far red light reaching a crop have been shown to reduce stem extension in several species (Rajapakse, Mcmahon & Kelly 1993; Runkle & Heins 2002; Paul et al. 2005). Plant responses to blue light include phototropism, control of stomatal opening and control of circadian rhythms (Briggs & Christie 2002; Eisinger et al. 2003), and because blue light is also inhibitory to the sporulation of many plant fungal pathogens (Ensminger 1993), spectral filters that modify transmission in the blue region of the spectrum may limit crop disease. Additionally, there is a growing understanding of the consequences of change in the UV environment for, for example, herbivorous insects (Foggo et al. 2007; Demkura et al. 2010), and thus UV manipulation may present valuable options for pest control. In terms of conventional horticultural glass and protective cladding films, there is often little or no transmission of UV-B, while transmission of UV-A decreases with decreasing wavelength, and thus claddings with modified transmission of UV radiation are now receiving increased attention (Krizek et al. 2005; Paul et al. 2005).

Plant responses to UV radiation have been largely characterized to date via studies driven by concerns regarding stratospheric ozone depletion and the resulting increases in UV-B radiation above current ambient UV fluxes (Caldwell et al. 2003). Initial concerns that such increases in UV-B posed a major threat to plant growth have been tempered both by the effective control of ozone depletion and the recognition that gross UV-B damage is largely a function of high UV-B doses, especially under conditions of unnaturally low photosynthetic radiation. However, this body of research has equally shown that variation in UV-B, within as well as above the ambient range, can induce a wide range of plant responses. These include changes in growth and morphology, such as inhibition of leaf growth and stem extension (Li, Yue & Wang 1998; Wargent et al. 2009a,b), and changes in leaf structure (Barnes et al. 1996). Changes in plant chemistry, including altered content and chemical composition of epicuticular waxes (Gonzalez et al. 1996; Kakani et al. 2003a), and the accumulation of UV-B-absorbing phenolic compounds such as flavonoids (reviewed by Bassman 2004), are generally interpreted as key elements in acclimation to UV-B exposure. It has also been demonstrated that exposure to UV-B can result in cross-protection against other environmental factors, for example, high and low temperature (Teklemariam & Blake 2003; L'Hirondelle & Binder 2005), and drought (Gitz & Liu-Gitz 2003).

In an applied agronomic context, UV-induced growth regulation and cross-protection are especially pertinent for crops grown during early development in a protected environment, which are then subsequently transplanted to the field. The key aim in cultivating vegetable crops is to produce plants that can withstand both mechanical handling during transplanting and the effects of abiotic stress experienced during the more vulnerable, non-protected field phase. As a consequence of such factors, it is well established that the field transplantation phase of a cropping regime following the earlier protected or propagation phase can lead to deleterious consequences for growth and final yield (van Iersel 1998; Garner & Bjorkman 1999). Manipulations during the early-stage growth phase, for example, the use of chemical growth regulators, can have long-term effects on performance, as demonstrated in the husbandry of forestry transplants (Scagel, Linderman & Scagel 2000). In a series of experiments using commercial-scale polythene-clad structures, we showed previously (Paul et al. 2005) that growth under a wavelength-selective filter that was largely transparent to solar UV resulted in consistent and significant reductions in the leaf area and dry weight of seedlings of lettuce (Lactuca sativa L.), which is commonly cultivated in a protective environment before transplantation into the field. For example, averaged across six repeat experiments, a UV-transparent film reduced plant leaf area by 23% compared with a film that was opaque to solar UV (Paul et al. 2005). We hypothesized that these morphological changes in response to exposure to solar UV during the propagation period would be accompanied by physiological and biochemical changes, and that this suite of UV-induced changes would improve the long-term performance of transplanted crops in the field environment. We report here on an experimental investigation into a range of productivity-related responses to altered solar UV regimes in lettuce. Our primary experimental focus was on initial growth and on the transitional phase between the end of the protected propagation stage and transplantation into the field, but we also determined long-term performance in lettuce plants grown-on to final yield. To achieve this, we utilized both near-commercial-scale field experiments and more detailed mechanistic studies, including indoor controlled environment (CE) manipulations of supplementary UV radiation to characterize the underlying mechanisms driving the effects of early-stage UV modification regimes on both short- and long-term crop responses.


Experimental facilities and UV treatments

All crop-scale experiments were carried out at Stockbridge Technology Centre (North Yorkshire, UK; 53N 1W) using commercial multispan tunnel structures (Haygrove Tunnels Ltd, Ledbury, UK), which each covered 740 m2 over four individual bays, each measuring 3 m high × 6 m wide × 30 m long, with open sides to a height of 1 m. Three commercially produced polyethylene cladding films (supplied by bpi-Visqueen Ltd, Ardeer, UK) were used in these experiments, as detailed in our previous study (Paul et al. 2005). The control treatment (=‘standard’) was a film typical of commercial horticultural cladding used in protected cropping systems in the UK. The standard film had a total photosynthetically active radiation (PAR) transmission of 93%; transmission in the UV region of the spectrum decreased rapidly with decreasing wavelength from 90% at 400 nm to less than 10% below 350 nm (Fig. 1). Total transmission in the UV-A region was approximately 50%, and transmission of the UV-B waveband less than 5% and effectively zero below 300 nm. Two films with specifically modified UV transmission were used (Fig. 1). We used a ‘UV-transparent’ film (=‘UV-T’) that transmitted greater than 80% across the whole of the solar UV range from 290 to 400 nm. Total transmission in the PAR and the UV-A regions for the UV-T film was 94 and 90%, respectively. We also used a UV-opaque (‘UV-O’) film, with a PAR transmission of 95%, but a transmission in the UV-A region of only 10%, and zero UV-B transmission. The transmission properties of the films were shown to be relatively consistent over three growing seasons under UK conditions. In addition to transmission spectra, during the course of the experiment there were 2 d when we undertook campaigns of spectroradiometric measurements across all treatments using a double monochromator scanning spectroradiometer (model SR991-v7; Macam Photometrics, Livingston, UK) in order to provide indicative characterizations of local UV levels. It was not possible to undertake simultaneous measurements under different treatments, so the instrument was rotated between treatments over the course of the day so that each cycle through all tunnels took 30–40 min. These measurements not only confirmed that the effect of the films on the UV environment of the crop was consistent with the transmission spectra, but by integrating measurements through the day they also gave an indication of the incident solar UV. UV treatments are generally expressed as biologically weighted UV, and in plant research this has almost always been done using Caldwell's generalized plant action spectrum (Caldwell et al. 1986: UVCALD), and on this basis the daily weighted UV doses were 2.2, 0.2 and 0.0 kJ m−2 d−1 in UV-T, standard and UV-O treatments, respectively. However, we concluded from our previous studies (Paul et al. 2005), using the same experimental system, that the effects of treatments were more accurately described using the plant growth action spectrum of Flint & Caldwell (2003) (=UVF&C), which attributes far greater effectiveness to UV-A wavelengths. Weighted using this spectrum, the doses were 13.1, 5.0 and 0.1 kJ m−2 d−1 in UV-T, standard and UV-O treatments, respectively.

Figure 1.

Spectral transmissions for the three spectral filters used in the field experiments: standard horticultural filter (long dashed line), UV-transparent filter (no dash) and UV-opaque filter (short dash). Spectral transmissions were determined for triplicate samples (minimum 4 cm2) using a 75 W xenon arc lamp (LOT-Oriel, Leatherhead, UK) with a 10 cm integrating sphere, and a double-scanning spectroradiometer (Macam Photometrics, Livingston, UK). Variation between multiple samples of identical films was negligible, and SEs are too small to present.

In addition to our outdoor experimental system, we also utilized an indoor CE system as described previously (Wargent et al. 2009a). Briefly, UV-A and PAR were supplied by 400 W metal halide lamps (Osram HQI-BT 400 W; Osram Ltd, St Helens, UK) and UV-A sources (Q Panel UVA-340 tubes; Q-Panel Laboratory Products, Bolton, UK). The PAR background was 500 µmol m−2 s−1 during the 14 h photoperiod. The UV-B treatment (‘UV+’) was provided for the middle 10 h of the photoperiod using 40 W UV-B tubes (Philips TL40/12–RS; Starna Ltd, Romford, UK) filtered with 0.13-mm-thick cellulose diacetate (Clarifoil, Courtaulds Ltd, Derby, UK) to remove wavelengths below approximately 290 nm. The control treatment (‘no UV’) was provided on the same bench as UV+, but used TL40 UV-B tubes wrapped in clear polyester (Lee Filters, Andover, UK) to remove wavelengths less than 320 nm. Filters were replaced on a regular basis, and plants and filters were rotated during experiments to account for any positional bias. UV treatments were measured with a double monochromator spectroradiometer (as above). As discussed earlier, we argued previously that plant responses to such manipulations of the UV environment were perhaps more consistent with the UVF&C function than the UVCALD biological spectral weighting functions, and the CE treatments reflected this. Thus, weighted using UVF&C, our UV+ treatment was 12.5 kJ m−2 d−1, which is broadly comparable to the UV measured in the UV-T treatment in the field, while our control treatment (no UV) was 0.32 kJ m−2 d−1. Weighted according to UVCALD, UV+ and no UV were 10.0 and 0.0 kJ m−2 d−1, respectively.

Plant material

For crop-scale studies, seeds of L. sativa (cv. Challenge; Syngenta Seeds Ltd, Lancashire, UK) were germinated using standard UK commercial practices (Crystal Heart Salads, Holme-on-Spalding-Moor, UK) in a dark germination room at ∼16 °C for 4 d before being randomly transferred to one of the three experimental treatments based at our experimental site. Plants were maintained in cell insert modules during the protected tunnel phase as opposed to being planted into the soil, which provided additional uniformity in planting conditions. Cell modules were placed in the central two bays of each tunnel system, and were located in the middle of each bay in order to avoid edge effects, with no differences in temperature or humidity between different UV regimes observed during the experimental period (data not shown). The plants then remained in their respective treatments for a period of 23 d (approximately the fourth true leaf emergence) before being transplanted by hand to the uncovered field site. The outdoor field experiment consisted of 12 mechanically cleared replicate plots measuring 12 × 4.5 m each arranged in a randomized block design and situated adjacent to the tunnel structure. Upon transplantation, the plants were randomly distributed across the field plots, and then managed according to standard commercial practice. The plants were harvested when judged ready according to commercial standards, approximately 7 weeks after transplantation.

For indoor CE experiments, seeds were sown in 15 cell tray inserts with Levington M3 compost (Henry Alty Ltd, Preston, UK) under the light conditions described earlier and a temperature of 25 ± 2 °C. Six days after initial sowing, when the first leaf had initiated, plants were exposed to supplementary UV treatments as detailed earlier. In our ‘cross-over’ experiments designed to mimic in the CE room the effects of transition from the propagation to field environment, plants were germinated and exposed to supplementary UV as above, but following 16 d of UV exposure, equivalent to approximate full expansion of leaf 2, a proportion of plants were switched from zero UV to UV+ (no UV/UV+) or held under UV+ conditions throughout (UV+/UV+). Final harvest of the cross-over plants took place 10 d following the cross-over time point.

Measurements of plant growth and yield

We selected the second true leaf as a focus for the majority of physiological measurements because of its suitable size during early growth, and also because the initiation and majority of expansion for leaf 2 occurred during the propagation phase in the field tunnels. Leaf area was measured following destructive harvests using an automatic leaf area meter (LI-3000 Li-Cor, Lincoln, NE, USA). Leaf thickness was measured at the central region of the lamina, adjacent to the mid-vein, using a 0–25 mm micrometer (RS Components, Corby, UK). Both leaf length and width measurements were taken at the widest point of the lamina using electronic digital callipers (Screwfix Direct, Yeovil, UK). Leaf and whole-plant dry weights were obtained by weighing the plant material after drying at 75 °C until constant mass was reached. In the CE experiments, absolute and relative growth rates were derived from these dry weights. As noted earlier, in the crop-scale studies, lettuce plants were harvested when they were judged ready according to commercial standards, approximately 7 weeks after transplantation. Fresh lettuce head harvests were performed according to standard commercial practice, with lettuce heads weighed using an SW Feihu Series flatbed electronic balance (Scalesworld, Milton Keynes, UK) to provide head fresh weight.

Physiological and biochemical measurements

In the crop-scale experiments, net photosynthesis was measured using a portable infrared gas analysis system (CIRAS-2; PP systems, Hitchin, UK) with cuvette conditions set to 1500 µmol m−2 s−1, 60% relative humidity and 380 ppm of CO2, with leaves left to equilibrate for 5 min prior to measurement. Measurements were made using leaf 2 and taken daily at noon for the 2 weeks prior to transplantation of study plants into the field. In the CE experiments, routine photosynthetic measurements were determined as outlined earlier, but with a cuvette PAR irradiance of 500 µmol m−2 s−1 to reflect the growing conditions within the CE room. Measurements commenced as soon as leaf 2 was of a suitable size to facilitate measurement (approximately 4 d from emergence), and continued every two days for 8 d in total. At day 8 of photosynthetic measurements, light response curves of net photosynthesis were generated using the same method, but with PAR increased at intervals of 250 µmol m−2 s−1 between measurements across a range of PAR irradiances (0–2000 µmol m−2 s−1). As an additional indicator of photosynthetic performance, maximum photochemical efficiency of photosystem II (PSII), represented as the ratio of variable to maximum chlorophyll fluorescence (Fv/Fm), was quantified in leaf 2 of lettuce plants using a plant efficiency analyser (Hansatech Instruments, King's Lynn, UK). Leaves were dark adapted for 30 min using standard leaf clips as supplied, and then measured with a saturating irradiance of 3000 µmol m−2 s−1 at 650 nm for 3 s. In the crop-scale study Fv/Fm was measured at three time points: (1) the final day of the propagation phase in the tunnel environments prior to transplantation into the field; (2) the first day following transplant into the field; and (3) the second day following transplant into the field. All measurements were taken at solar noon (1200–1300 h GMT). To further explore the consequences of UV-induced acclimation for response to other environmental stresses, Fv/Fm was also measured in the CE experiment in plants removed from the UV+ or zero UV treatments (i.e. following 16 d of exposure to UV treatments) which were then exposed for 1 d to elevated PAR (1000 µmol m−2 s−1) and daytime temperature (35 °C) in the absence of UV radiation. Fv/Fm in leaf 2 of lettuce plants was determined as above, with measurements taken at the following time points: (1) at the start of the photoperiod before exposure to elevated PAR and temperature; (2) the mid-point of the photoperiod of exposure to high PAR and temperature; (3) end of photoperiod of the high PAR and temperature treatment; and (4) at the start of the photoperiod following the day of exposure to elevated light and temperature treatments. For quantification of UV-B-absorbing compounds, the method followed closely to that of Gonzalez et al. (1998). Leaf material was then placed in liquid N2 for 30 s and stored at −20 °C. Foliar material was homogenized with acidified methanol (methanol : H2O : HCl; 79:20:1), with absorbance at 300 nm (UV-B-absorbing compounds) expressed on a unit leaf area basis.

Statistical analysis

Field studies of early-stage growth responses were quantified during a single-season experiment as part of a larger series of field campaigns carried out across three growing seasons during the period 2004–2006 on which we have previously reported (Paul et al. 2005), and characterizations of final yield were also carried out during that single season. CE data presented are from a single experiment, with consistent responses observed in a second L. sativa cultivar (data not shown). Pairwise comparisons of physiological response between tunnel treatments and CE conditions were analysed by t-test, with multiple treatment comparisons between tunnel environments compared using one-way repeated measures analysis of variance (ANOVA) with post hoc comparisons using Tukey. Because leaves were measured repeatedly during the sampling period for measurements of net photosynthesis, analysis of those data was by one-way repeated measures ANOVA with post hoc comparisons using Tukey. Photosynthetic light response curves were analysed by t-test and two-way ANOVA, with four replicate light curves generated on each sampling occasion per treatment that were carried out twice in total, again in a second lettuce cultivar (data not shown). Light response curves were fitted according to Ye (2007) using non-linear regression in Prism GraphPad ver. 4 (GraphPad Software, San Diego, CA, USA). Pearson correlation was used to test linear associations between as well as between chlorophyll fluorescence yield and concentrations of UV-absorbing compounds. Unless otherwise stated, statistical analyses were performed using SPSS v 11.5 (Chicago, IL, USA).


Growth responses during propagation stage and at final harvest

The growth characteristics of second true leaves in L. sativa specimens under the three different UV regimes during the propagation stage in this experiment were consistent with certain observations from our previous experiments (Paul et al. 2005). For example, leaf 2 area was significantly decreased by 18% in UV-T plants compared to UV-O plants (P < 0.001; Table 1), whereas the mean reduction in previous experiments was 23%. Additionally, the second leaf area of plants propagated under the standard filter was not significantly smaller than those plants propagated under the UV-O film (P > 0.05; Table 1), yet the small decrease observed (14%) was very similar to those responses observed in a previous experiment; (mean = 13%: Paul et al. 2005). Reductions in leaf area were due largely to a progressive reduction in leaf 2 length with increasing UV. Leaf length was reduced by 15% in plants exposed to UV-T conditions when compared with UV-O plants (P < 0.001; Table 1), and by 12% in standard plants when compared with the UV-O regime (P < 0.05; Table 1). The width of leaf 2 was also significantly less in UV-T-grown plants compared to UV-O (P < 0.05; Table 1), but the difference was only 4%, while the differences between UV-T and standard, or standard and UV-O spectral filters were not significant (P > 0.05; Table 1). Leaf 2 of lettuce seedlings propagated under the UV-T film was 14% thicker than leaves of plants from both the standard and UV-O films (P < 0.01; Table 1). UV regime had no significant effect on leaf 2 fresh weight (P > 0.05; Table 1). Leaf dry weight was significantly higher in UV-T than standard (P < 0.05; Table 1), but the differences between UV-T and UV-O, or standard and UV-O spectral filters were not significant (P > 0.05; Table 1). There were no differences across our three treatments as a whole in terms of concentrations of UV-absorbing compounds (300 nm) (P > 0.05; Table 1). Finally, in order to quantify if differences in UV environment during the propagation phase influenced crop productivity, we measured final yield (fresh weight of lettuce heads) according to commercial practice. Plants that were propagated under UV-T conditions showed significantly increased head fresh weights at final harvest when compared to UV-O (69%; P < 0.001; Table 1) and when compared to plants propagated under the standard spectral filter treatment (31%; P < 0.001; Table 1).

Table 1.  Growth characteristics of leaf 2 of Lactuca sativa plants grown under three different UV environments at the end of the protected growth phase, in addition to harvestable shoot fresh weight at final harvest following a 7 week uniform field growth period
Parameter and treatmentUV-transparentStandardUV-opaqueP value
  1. Data are means of at least 17 replicates, and different letters indicate a significant difference between UV treatments for that particular experimental variable.

Leaf 2 area (mm2)2079.5 a2174.7 a2536.4 b<0.001
Leaf 2 length (mm)76.22 a79.45 b89.82 c<0.001
Leaf 2 width (mm)25.60 a26.12 ab26.76 b<0.05
Leaf 2 thickness (mm)0.43 a0.37 b0.37 b<0.01
Leaf 2 fresh weight (g)0.3730.3390.3570.115
Leaf 2 dry weight (g)0.028 a0.024 b0.026 ab0.052
Leaf 2 mean assimilation rate (µmol m−2 s−1)11.99 a11.25 ab10.17 b<0.05
Leaf 2 UV-absorbing compounds (abs 300 nm cm−2)0.1340.1370.1130.435
Harvestable fresh weight (g)413.5 a321.0 b244.3 c<0.001

To further characterize the effects of transfer from a zero UV environment to a UV+ environment (equivalent to UV-O/UV-T in the field experiments), we established a ‘cross-over’ experiment under controlled conditions, whereby lettuce plants were either switched from no UV to UV+ after 16 d, or were maintained under a UV+ environment throughout the experimental period. Prior to the cross-over, the areas of leaf 2 under no UV and UV+ treatments (area 22.1 ± 1.4 and 19.6 ± 2.2 cm2, respectively) were similar to those in the UV-O and UV-T treatments observed in the field (Table 1), and while the difference in area was not significant in the growth room, the magnitude of reduction was close to that observed in the field (12% compared with 18% between UV-T and UV-O). Total plant dry weight and total leaf area were both significantly lower in lettuce seedlings grown under the UV+ treatment compared to no UV under controlled conditions (P < 0.001; Fig. 2a,b). However, growth responses to UV-B during the post-cross-over period were very different between plants grown initially in different UV environments. The difference between UV+/UV+ and no UV/UV+ in relative growth rate during the post-cross-over period was substantial for both leaf area and dry weight growth (both P < 0.001: Fig. 2c,d). As a result, at the final harvest, 10 d following the cross-over time point, there were no significant differences in either total plant dry weight or leaf area between plants exposed to no UV/UV+ conditions and those that remained under the UV+ treatment throughout (P > 0.05; Fig. 2e,f), and control plants that had not been exposed to UV at all exhibited larger leaf area and plant dry weight than our other two treatments (P < 0.001: Fig. 2e,f).

Figure 2.

Plant growth in seedlings propagated and then transferred to differing UV+/− regimes in controlled environment (CE) room conditions; leaf area (a) and total plant dry weight (b) at the end of the initial period of exposure, UV+ = 10 kJ m−2 d−1, no UV = 0 kJ m−2 d−1; relative growth rates of leaf area (c) and plant dry weight (d) following the cross-over period, UV+/UV+, where plants remained under UV+ regime throughout, and no UV/UV+, where plants were moved from no UV to UV+ conditions after 16 d; and leaf area (e) and total plant dry weight (f) at the end of the cross-over period. Asterisks indicate significant differences at P < 0.001. n = 10 replicates ± 1 SE.

Photosynthetic performance during early-growth stage

To explore underlying mechanisms that may be driving changes in plant response to our UV modification regimes, we quantified photosynthetic rate on a daily basis for 13 d until the end of the propagation phase during our field experiment. Although there was notable day-to-day variability in photosynthesis, repeated measures anova showed a significant effect of treatment on photosynthetic rate (P < 0.05; Fig. 3), such that photosynthesis was significantly higher in plants propagated under the UV-T regime compared with UV-O plants (Tukey: P < 0.05; Fig. 3); differences between UV-T/standard plants and standard/UV-O were not significant (P > 0.05; Fig. 3). This is reflected in the mean photosynthetic rate across all measured days, which was 18% higher in plants from UV-T than those from UV-O (Table 1). There were no treatment-dependent significant differences in stomatal conductance, transpiration rate or internal CO2 concentrations (data not shown).

Figure 3.

Daily CO2 net photosynthetic rates of leaf 2 of Lactuca sativa plants grown under three different UV environments during the propagation stage of growth (UV-opaque = closed triangles q, standard = closed circles ●, UV-transparent = open circles and in bold ○). n = 4 replicates ± 1 SE.

Photosynthetic rate was also quantified under controlled conditions during the development of leaf 2. There was initially no significant difference between UV+ and no UV, but by the final measurement, 11 d since the emergence of leaf 2, the net photosynthetic rate was 21% higher in UV+ plants than in no UV plants (P < 0.01; Fig. 4a). This increase in net photosynthesis was mirrored by a concomitant increase in leaf thickness in UV+ plants compared to no UV plants by the same time point (P < 0.001; Fig. 4b). When changes in plant growth were analysed in terms of absolute growth rate (AGR), AGR was significantly higher in UV+ plants compared to no UV plants by the final time point (day 10; P < 0.01; Fig. 4c). As with our field observations, there were no significant differences in stomatal conductance, transpiration rate or internal CO2 concentrations according to UV treatment (data not shown). We also determined the photosynthetic light response curves of lettuce seedlings grown with and without UV radiation. Analysis of the light response curve fitted showed that maximum light saturated photosynthesis was significantly higher in UV+ plants compared to no UV plants (P < 0.05; Fig. 5), but there were no significant differences observed in quantum yield of photosynthesis or light compensation point according to UV treatment (data not shown).

Figure 4.

Photosynthetic performance and growth of leaf 2 of Lactuca sativa plants under controlled environment (CE) conditions; UV+ = 10 kJ m−2 d−1, no UV = 0 kJ m−2 d−1; (a) net photosynthesis, (b) leaf thickness, (c) absolute growth rate (AGR), according to days following leaf 2 emergence. Asterisks indicate significant differences at **P < 0.01, ***P < 0.001. n = 10 replicates ± 1 SE.

Figure 5.

Light response curve of net photosynthesis in Lactuca sativa after 8 d of exposure to supplementary UV radiation, UV+ = 10 kJ m−2 d−1, no UV = 0 kJ m−2 d−1. Light response curves were generated with photosynthetically active radiation (PAR) increased at intervals of 250 µmol m−2 s−1 across a range of PAR irradiances (0–2000 µmol m−2 s−1). n = 4 ± 1 SE. Curve fittings are made according to the method of Ye (2007).

Protective effects of early exposure to UV radiation

The relationship between physiological characteristics observed at the end of the propagation stage and subsequent performance in the field was further investigated by chlorophyll fluorescence analysis (Fv/Fm) combined with parallel quantification of foliar UV-absorbing compounds (300 nm) during the final day of the propagation phase and for two subsequent days following transplantation of lettuce plants into field conditions. At the end of the protected propagation stage, values of Fv/Fm were generally low (0.55, 0.66 and 0.65 for UV-O, standard and UV-T, respectively) with no indication that exposure to solar UV-B radiation under the UV-T had caused any significant decrease in Fv/Fm compared with other treatments (Fig. 6). In the 2 d following transplantation into the field, the values of Fv/Fm increased significantly in all treatments (P < 0.001; Fig. 6). In addition, non-treatment-dependent increases in Fv/Fm over time were closely matched by increases in the concentration of foliar UV-absorbing compounds across all three treatments, but entirely independent of UV treatment (r2 = 0.8; Fig. 6). At the end of propagation, A300 was somewhat lower in plants propagated under UV-O (0.114 compared with 0.133 and 0.139 in standard and UV-O, respectively; P > 0.05), and concentrations increased significantly in all three treatments post-transplantation (P < 0.001; Fig. 6). The increase in A300 in the first 2 d after the end of propagation was approximately twofold in plants propagated under UV-T, but threefold in plants from UV-O. There were highly significant correlations between changes in Fv/Fm and those in UV-absorbing pigments across all three treatments (r = 0.894; P < 0.001). In controlled conditions, there was no significant difference in Fv/Fm between no UV and UV+ plants prior to exposure to elevated PAR and temperature (Table 2), yet a steady decline in Fv/Fm during the first day of exposure to stressful conditions (Table 2). However, with increasing time under those elevated conditions, there was a slower recovery in Fv/Fm in plants from no UV compared to UV+, so that by dawn of the day following elevated PAR and temperature, Fv/Fm was significantly lower in no UV than UV+ plants (Tables 2, P < 0.01).

Figure 6.

Chlorophyll fluorescence ratio (Fv/Fm) as correlated with UV-B-absorbing compound content (expressed as absorbance units measured at 300 nm per unit leaf area) in leaf 2 of Lactuca sativa plants 24 h before field transplant and for 2 d thereafter according to spectral filter treatment; UV-transparent (○), standard (●), UV-opaque (q) spectral filter treatments. The first timed measurement (day 1) was taken while seedlings were under their respective UV treatments, and the subsequent two measurements were taken at 24 and 48 h after the seedlings were transferred to the field (days 2–3). n = 10 ± 1 SE.

Table 2. Fv/Fm in Lactuca sativa plants when subjected to high daytime temperature (35 °C) and light (1000 µmol m−2 s−1) stress following pre-exposure to supplementary UV radiation (UV+ = 10 kJ m−2 d−1)
 No UVSEUV+SEP value
  1. Data are means of 10 replicates ± SE.

Solar dawn (day 1)0.810.010.820.010.506
End of day0.640.030.710.030.120
Solar dawn (day 2)0.690.010.770.02<0.01


In this report, we present evidence for a role of UV-B exposure during the early growth phase of lettuce plants in improving long-term crop productivity in the field. Our findings suggest that earlier exposure to realistic levels of UV radiation leads to positive photosynthetic performance and other protective changes in leaf morphology, and when combined with enhanced photoprotection to high light and temperature as observed under controlled conditions, these early-stage effects driven by UV-B appear to enhance plant tolerance against generalized field transplantation stresses to a greater capacity, which can then lead to increased productivity in a longer-term field growth environment.

Early-stage effects of UV-B radiation on plant development can increase crop yields

The inhibitory effects of UV radiation upon stem elongation, leaf area and biomass accumulation are now well documented in a variety of plant species (Caldwell et al. 1995; Kakani et al. 2003b), and there have been some previous attempts to exploit crop responses to UV for growth regulation, using both wavelength-selective filters (Paul et al. 2005; Wargent et al. 2009a) and artificial UV sources (Giannini, Pardossi & Lercari 1996; Kobzar, Kreslavski & Muzafarov 1998). In this study, exposure of seedlings to solar UV-B radiation at a mid-northern latitude during summer (i.e. propagation under a UV-transparent filter) led to reductions in leaf area, length and width, but increases in leaf thickness and harvestable yield (Table 1). Although the mechanisms resulting in UV-mediated increases in leaf thickness and inhibition of leaf growth are still not well understood, certain processes regulating these responses are now being elucidated. In a previous investigation into the role of UV radiation in regulating leaf expansion of L. sativa plants using the same experimental field system, we provided evidence for the causal role of UV-mediated increases in cell wall-associated peroxidase in the inhibition of cell expansion (Wargent et al. 2009a). From an agricultural viewpoint, such cell wall ‘stiffening’ responses and other related biophysical properties may provide improved tolerance to stress factors, such as mechanical handling during vulnerable transplantation stages and other elements of crop cultivation which are ubiquitous in modern agriculture (Clarkson et al. 2003). Our data suggest that providing earlier exposure of protected crop plants to UV radiation may exploit such physiological responses, leading to improved crop robustness and final yield.

Exposure to solar UV-B can mediate increased photosynthetic output and enhanced photoprotection in crop plants

Net photosynthesis was significantly increased in plants propagated under the UV-transparent film compared with the UV-opaque filter (Fig. 3), and the same result was observed under CE conditions (Fig. 4). Despite some past evidence that elevated exposure to UV-B above ambient levels leads to inhibition of photosynthetic performance in higher plants (Tevini & Teramura 1989; Teramura & Sullivan 1994), Allen, Nogues & Baker 1998 concluded in a review of a range of studies that the majority of more pronounced inhibitory effects on photosynthesis reported have usually been observed either in CEs where low PAR fluxes can exaggerate UV-B responses, or from square-wave UV supplementation studies, where variations in season and weather conditions can lead to overestimations of the UV-B flux applied. As with growth responses, the magnitude of the increase in photosynthesis was broadly comparable in the field (the average photosynthetic rate during propagation was 18% higher in UV-T than in UV-O) and in the CE room (net photosynthesis measured under growth conditions was 21% higher in UV+ plants than no UV; Fig. 4). A recent meta-analysis of more than 140 studies in both CE room and field conditions concluded that UV-B radiation reduced net photosynthesis in both woody and herbaceous plants (Li et al. 2010). This contrasts with an earlier meta-analysis using data only from field studies mimicking the effects of stratospheric ozone depletion (Searles, Flint & Caldwell 2001), which concluded that there was no overall significant effect of treatments on either CO2 fixation or chlorophyll fluorescence. This contrast may be caused by the risk of low PAR irradiance in CEs exaggerating the negative effects of UV-B (e.g. Fiscus & Booker 1995). The analysis of Searles et al. (2001) also highlighted that while most field studies found some degree of reduction in photosynthetic CO2 fixation and/or Fv/Fm, there are some examples where these parameters are increased by exposure to UV-B in the field. Reports of increased rates of photosynthesis appear to derive mostly from long-term studies of deciduous tree species (e.g. Sprtova et al. 2003, Sullivan et al., 2003, Yang & Yao 2008), but appear to have parallels with our data for lettuce. For example, Sprtova et al. (2003) reported that light-saturated photosynthesis (Amax) measured in the third season of a UV-B supplementation treatment of field-grown beech was significantly increased by UV-B supplementation, and attributed this increase partly to increased leaf thickness and partly to improved biochemical photoprotection against photoinhibition. Similarly, we showed that increased photosynthesis was associated with increased leaf thickness in the field (Table 1), a finding that was mirrored under controlled conditions (Fig. 4). However, changes in thickness do not fully explain the effect of UV-B on photosynthesis, because if photosynthetic rate was corrected for leaf thickness and expressed as CO2 fixation per unit volume, photosynthetic rates were still 14% higher in plants grown in the presence of UV-B, indicating changes in as-yet uncharacterized components of UV response. Our findings demonstrate that that beneficial up-regulation of photosynthetic performance can result from carefully managed UV-B exposure in food crops, yet the pattern of photosynthetic photochemical responses to moderate fluxes of UV-B, and up-stream regulatory mechanisms, remain poorly defined.

One archetypal component of UV response that may contribute to protection against photoinhibition could be increased concentrations of UV-absorbing secondary metabolites. Such increases are very widely observed in the field and within CE studies, including our own previous studies carried out under the same controlled conditions [e.g. UV-absorbing pigments (A300) were 46% higher than zero UV plants at a similar UV dose to that which we used in the present study (Wargent, Taylor & Paul 2006)], but our field data are inconclusive regarding this mechanism. There is some evidence to support the hypothesis that elevated phenolic pigmentation can provide photoprotection (Zhang et al. 2010), and that UV-absorbing compounds may actually serve a wider function in photoprotection, possibly through heightened antioxidant effects (Agati & Tattini 2010). Further work is required to establish the relative role of morphological and biochemical changes in UV-B-induced increases in photosynthesis, and the extent to which their expression depends on environmental conditions. In accordance with past studies that have focused on the direct effects of UV on photosynthetic competence, the vast majority of previous evidence suggests that UV has a deleterious effect on maximum photochemical efficiency of PSII, quantified as Fv/Fm (Teramura & Sullivan 1994; Fiscus & Booker 1995; Krause et al. 2003). Yet, our findings support some of the emerging evidence that UV radiation may mediate cross-protective benefits to a variety of stress factors. For example, excised leaf discs of Pisum sativum and Phaseolus vulgaris pre-treated with UV-B radiation were subsequently more tolerant to high light stress (Bolink et al. 2001), and in accordance with our findings, it was further suggested that an observed increase in leaf thickness, together with an increase in antioxidant capacity, could have contributed to the higher protection against photoinhibition by visible wavelengths, an observation that may be pertinent to our findings. Other authors have shown that pre-exposure to UV-B wavelengths can lead to reduced sensitivity to drought stress (Manetas et al. 1996; Nogues et al. 1998). Some studies have attributed UV-mediated delimiting of photosynthetic capacity to reductions in stomatal conductance and subsequent increased water-use efficiency (Poulson et al. 2002; Poulson, Boeger & Donahue 2006), and although we observed no changes in stomatal gas exchange in our study, other authors have observed direct effects of solar UV-B on the up-regulation of photosynthetic genes (Izaguirre et al. 2003), and thus the relationship between leaf growth, PSII performance, induction of secondary phenolics and subsequent resistance to transplant shock warrants further investigation.

The role of ‘UV-B shock’ relative to other stresses at transplanting remains unclear, but our CE data show that plants moved from no UV to UV+ grew very little in the 10 d following cross-over (Fig. 2). So far, as we are aware, plant responses to transfer from one UV regime to another have rarely be defined in detail, but Gonzalez et al. (1998) also found a similar significant reduction in RGR in P. sativum on transfer from zero to 10 kJ m−2 d−1 plant weighted UV-B. This marked growth check in response to transfer from a UV-B-deficient environment to field solar UV may be in part caused by acute responses of non-acclimated plants to acute UV-B damage and in part caused by the rapid induction of protection to limit such damage in non-acclimated plants. That rapid acclimation is demonstrated by the positive correlation between induction of UV-B-absorbing compounds and maximum photochemical efficiency of PSII (Fv/Fm) in the first 2 d after seedlings were transferred from protected to field conditions (Fig. 6). Such possibilities relating to generalized cross-protection of plants to field conditions during the propagation stage are further underlined by the fact that although no UV-dependent differences in foliar pigmentation were observed in lettuce plants upon initial transplantation into the uniform field environment, plants propagated under the UV-opaque regime exhibited an increase of 99% in induction of UV-absorbing compounds compared to UV-T plants from the pre-transplant time point to the final post-transplant time point (Fig. 6). Such differences in induction indicate a ‘catch-up’ effect in phenolic metabolism in order for UV-opaque plants to acclimate to field conditions, a process which may utilize resources typically vital in acclimating to the other significant stresses associated with field transfer, such as drought stress, and wounding effects via mechanical handling. Such catch-up effects could have certain knock-on consequences for growth at the transplant stage in UV-opaque plants, particularly via sudden and increased carbon partitioning to phenolic metabolism at the expense of growth (Bryant & Julkunentiitto 1995). It is also intriguing that Fv/Fm values for those plants initially grown in the absence of UV were somewhat lower than our other two treatments while in the protected tunnel phase of field assessments (Fig. 6), although these differences were not quite significant at P > 0.05; for example, at day 1, when plants were still under UV filter treatments, Fv/Fm values were higher than those of the UV-O treatment by 0.10 and 0.12 in UV-T and standard treatments, respectively (P = 0.09; Fig. 6). In addition, there were indications of a greater increase in Fv/Fm in plants propagated under UV-O compared to UV-T plants following field transplantation; for example, the average increase in Fv/Fm from day 1 to day 3 was 0.14 and 0.24 in UV-T and UV-O plants, respectively (P = 0.07), although this current study did not clearly indicate the UV-mediated causes of any changes in Fv/Fm. Indeed, given the growing evidence that exposure to UV may regulate protective responses to high PAR irradiances, it may not be entirely surprising that our UV-opaque plants experienced somewhat lower Fv/Fm values as compared to our standard and UV-inclusive treatments, although the standard filter treatment partially transmits within the longer-wavelength UV-A region only. Our findings nonetheless suggest that plants raised in the absence of UV may be subject to higher levels of PAR-induced photoinhibition under typical sunlight conditions.

Implications of study findings

While it is clear that initial signalling events regulated by UV-B radiation in particular are often highly specific in nature (Brown et al. 2005; Cloix & Jenkins 2008), it is now also evident that UV can mediate overlapping patterns of response to a range of stimuli including high light, wounding and drought (Mackerness et al. 1999; Stratmann 2003). The concept of plant priming to stresses encountered during later growth is an area receiving increased attention (Luo et al. 2009; Macarisin et al. 2009; Verhagen et al. 2010), and is certainly pertinent to this study. With the expansion of protected cropping set to increase substantially in the future, largely as a consequence of increasingly dynamic and challenging growing conditions in a changing environment, such indicators of crop performance could act as valuable contributors to sustainable programmes of crop production. Overall, it seems likely that exposure to solar UV-B radiation confers protection against transplantation stress through a range of mechanisms, both morphological and biochemical, yet certain elements remain elusive. As well as UV-induced photoprotection, the capacity of UV-B-specific signalling to induce responses that overlap with a range of environmental stimuli (Stratmann 2003) and reduce sensitivity to drought (Gitz & Liu-Gitz 2003) or low temperature (Teklemariam & Blake 2003) may also be relevant to transplant protection. However, our data also highlight that biochemical mechanisms of cross-protection should be seen alongside morphological changes, reflecting the role of morphology in both photoprotection and in alterations within the mechanical properties of leaves that may improve tolerance to mechanical handling during stages of cultivation.


Our findings reflect a building paradigm shift in plant photobiology, whereby there is now a mounting body of evidence that instead of eliciting damaging responses in plants under ambient conditions, UV radiation can mediate beneficial end points in crop production, not only in terms of plant growth regulation (Moore, Paul & Jacobson 2006; Wargent et al. 2009a), but also in terms of phytochemistry and possible nutritional value (Krizek, Britz & Mirecki 1998; Jansen et al. 2008). When considered alongside the regulatory roles now becoming established for UV radiation in terms of pest and pathogen defence (Paul et al. 2005; Foggo et al. 2007), such considerations could present interesting possibilities for future exploitation in a range of agro-ecosystems. Our data demonstrate clearly that ambient UV-B at our chosen mid-latitude location does limit plant growth; for example, comparing our UV-transparent with UV-opaque treatment indicates that ambient UV radiation in this context reduces leaf area by around 23%, which is broadly comparable to the recent suggestion of a 3% reduction in biomass for each 10% change in solar UV-B (Newsham & Robinson 2009). But, most importantly, evidence for UV-induced photoprotection and cross-protection, and their long-term beneficial effects on yield observed here, may also suggest that changes induced by ambient UV may play a wider positive role in regulating plant responses to their environment than previously recognized. In conclusion, this study suggests that the informed manipulation of ambient light in protected cropping systems may provide profound benefits for production of food crops. Further exploration of underlying mechanisms of response and other such routes to achieving crop growth regulation with reduced agrochemical inputs is currently of vital importance to global food sustainability, and such priorities are likely to remain significant for some time into the future.


We are grateful to the Horticultural Development Council (UK) for funding this work with a Council Fellowship to J.P.M. (PC 221) and a studentship to J.J.W. (CP 26). We also acknowledge bpi visqueen for providing a studentship to E.M.E., and for the supply of spectral filters. In addition, we thank Haygrove Ltd for the supply of tunnel structures, James Bean at Crystal Heart Salads for supplying experimental plant material, Rob Jacobsen for helpful advice and discussion, and staff at Stockbridge Technology Centre for their technical support. This work has also been supported by the European Union COST Action FA0906 (‘UV4growth’).