Plant defences mediate interactions between herbivory and the direct foliar uptake of atmospheric reactive nitrogen


Reactive nitrogen from human sources (e.g., nitrogen dioxide, NO2) is taken up by plant roots following deposition to soils, but can also be assimilated by leaves directly from the atmosphere. Leaf uptake should alter plant metabolism and overall nitrogen balance and indirectly influence plant consumers; however, these consequences remain poorly understood. Here we show that direct foliar assimilation of NO2 increases levels of nitrogen-based defensive metabolites in leaves and reduces herbivore consumption and growth. These results suggest that atmospheric reactive nitrogen could have cascading negative effects on communities of herbivorous insects. We further show that herbivory induces a decrease in foliar uptake, indicating that consumers could limit the ability of vegetation to act as a sink for nitrogen pollutants (e.g., smog from mobile emissions). Our study suggests that the interactions of foliar uptake, plant defence and herbivory could have significant implications for understanding the environmental consequences of reactive nitrogen.


As a result of human activity, inputs of reactive nitrogen (N) compounds into the atmosphere have tripled in the past 50 years, with deleterious consequences for urban and natural ecosystems1,2,3,4. For example, nitrogen oxides (NOx) have negative effects on human health2,5 and alter important biogeochemical processes, including carbon sequestration, N cycling and global warming6,7,8,9. Reactive N also poses a significant and growing threat to biodiversity10,11. In terrestrial ecosystems, numerous studies have demonstrated that reactive N deposition to soils is causing widespread declines in plant diversity by altering soil chemistry and disrupting competitive dynamics3,12,13,14. However, the cascading effects of reactive N on higher trophic levels remain poorly studied, and there are few data on the complex pathways by which reactive N could influence consumers15.

One mechanism by which reactive N could affect higher trophic levels is via foliar uptake. Most research has focussed on root uptake of reactive N from soil following deposition. However, NOx gases can also be taken up directly by plant leaves and used as a nutrient source7,16. Up to 15% of a plant’s N budget can be obtained via foliar uptake, making it a significant pathway for the cycling of reactive nitrogen17,18. NOx enters leaves via stomatal diffusion and undergoes apoplastic disproportionation and ascorbate scavenging before downstream metabolism. Variation in these processes within the plant and among species19 suggests that the environmental consequences of foliar uptake will be more complex than root uptake, particularly for consumers. NOx deposited onto soil is taken up by roots as NO3 and should simply increase plant %N and may benefit herbivores, a prediction supported by empirical studies15,20. In contrast, foliar NOx assimilation could have divergent effects on consumers depending on its metabolic fate: N derived from NOx uptake could be stored as free amino acids and could benefit leaf-feeding herbivores, similar to the effects of root fertilisation. Conversely, if NOx was incorporated into defensive metabolites or caused shifts in defensive secondary metabolism, then consumers would suffer reduced growth and reproduction. Negative effects of foliar uptake of reactive N for herbivores have not yet been demonstrated. The only prior studies used high-concentration, short-term fumigation and showed positive effects on aphid colony growth21,22. However, aphids feed on phloem, which is severely N-limited and contains few defensive compounds23. Thus the potential for negative impacts of foliar NOx assimilation may have been underestimated.

In this study, we test the effects of foliar NO2 uptake on plant metabolism and herbivorous insects and then test the reciprocal effects of herbivory on NO2 uptake, using tobacco, Nicotiana tabacum, and its natural leaf-feeding herbivore, the tobacco hornworm, Manduca sexta. We are able to accurately trace reactive N uptake in leaves and monitor how herbivory affected reactive N assimilation by using a stable isotope fumigation system with 15NO2. Plants exposed to NO2 assimilate significant quantities of atmospherically derived N, exhibit upregulation of alkaloid defensive metabolites and support lower herbivore growth. In turn, herbivore feeding causes plant-wide reductions in foliar NO2 assimilation. Our study demonstrates that, in addition to disrupting plant communities through soil deposition, anthropogenic reactive N may have extended negative consequences for higher trophic levels. Our results also indicate that insect herbivores could influence the capacity of leaves to absorb reactive N and act as a sink for these atmospheric pollutants. This feedback between foliar NO2 uptake and herbivory may have implications for predicting the fate of reactive N in terrestrial ecosystems.


Effects of foliar NO2 uptake on insect herbivores

Using sealed chambers, we grew plants from the seedling stage in enriched 15NO2 at a concentration (40 ppb) similar to current urban levels;24 control chambers were maintained at 0 ppb. Plant roots were suspended in a hydroponics system at either 50mM NO3 (low-N) or 500mM NO3 (moderate-N) to compare responses at two ecologically realistic root N levels; shoots and roots were segregated to allow accurate partitioning of plant N sources18. After 4 weeks, M. sexta larvae were applied to half the plants in each chamber and allowed to feed. Herbivores feeding on NO2-exposed plants showed a pronounced reduction in growth (a proxy for fitness in this species25) compared to those feeding on plants in control chambers (Fig. 1a, Supplementary Table 1), indicating that anthropogenic NO2 can have potent negative effects on plant consumers. There was also a significant negative effect of higher NO3 on larval performance, which was independent of NO2 level (Supplementary Table 1). The effect of NO2 was not due to toxicity from direct exposure: larvae feeding on artificial diet in the chambers showed no adverse effects (Fig. 1b), confirming that plant quality had been adversely affected by NO2. Analysis of δ15N values in larval tissues (excluding gut tissues and plant material) revealed that NO2-derived nitrogen was used by herbivores for growth (Fig. 1c). The negative effect of NO2 on herbivore performance suggests that, in contrast to the possible benefits to herbivores of soil N deposition15, atmospheric reactive N will increase plant resistance and have important negative consequences for plant consumers when taken up by leaves.

Fig. 1

NO2 indirectly reduces herbivore performance and is transferred across trophic levels. a Growth of Manduca sexta larvae on leaf tissue of Nicotiana plants exposed to 40 ppb NO2 (blue bars) and 0 ppb NO2 (white bars) under low and high concentrations of root-available NO3 (10 days). b Inset panel shows no direct effect of NO2 on larvae feeding on artificial diet in the experimental (40 ppb NO2) chambers. c δ15N values of tissue from larva feeding on plants in chambers. Within each plot, different combinations of letters denote significant differences among means (Tukey’s test, P < 0.05), where present; ns indicates not significant. Data represent N = 3 chamber averages with N = 4 plants per chamber, and error bars are ±1 SE, with full statistical results provided in Supplementary Table 1

Effect of foliar NO2 uptake on plant-defensive metabolites

The genus Nicotiana has a well-characterised suite of defence-related secondary metabolites26, allowing us to test whether NO2 affected insects by altering the expression of defensive compounds in leaves. Targeted metabolomic analyses revealed that increased herbivore resistance under NO2 was associated with significant increases in foliar alkaloids, a class of N-rich, toxic defensive compounds (Fig. 2, Supplementary Table 2). Levels of three alkaloids were all significantly higher in plants exposed to NO2, including the dominant defensive compound nicotine, levels of which were on average 45% higher under NO2 exposure. Total leaf alkaloid content did not differ significantly between low and high NO3 levels, indicating that the effect of NO2 uptake on defence was not based on overall N availability; instead, dissolved NO2 in the apoplast may represent a mobile N pool available for transport to the site of alkaloid biosynthesis (roots). In support of this interpretation, total alkaloid content of plants under 40 ppb NO2 was positively correlated with the amount of 15NO2-derived N, while remaining independent of total N (Supplementary Figure 1). It is possible that NO2 acted as a stress or stress signal27 causing upregulation of plant defences, but these interpretations are not supported by the lack of effect of NO2 either on plant growth (Supplementary Figure 2a, Supplementary Table 3) or on two other classes of defensive metabolites (phenolics and terpenoids) (Fig. 2, Supplementary Table 2). Instead, carbon-based phenolic and terpenoid compounds were strongly reduced under higher root nitrogen (NO3) (Fig. 2, Supplementary Table 2). This result is consistent with theories on plant defence, which predict reduced C-based defences in high-nutrient environments that allow plants to produce potent N-limited defences (e.g., alkaloids) and/or tolerate damage28. Lower herbivore performance on high NO3 plants was not explained by metabolite variation, suggesting that levels of an unmeasured defence trait (e.g., proteinase inhibitors) may have been higher in the high NO3 treatment. Our results demonstrate that plant defences will be sensitive to N deposition to soil but will also be uniquely sensitive to N derived from foliar uptake, with the potential to affect a wide range of organisms that interact with a plant’s secondary metabolome, including herbivores, pollinators and microbes.

Fig. 2

Analysis of the Nicotiana defensive metabolome under NO2. Foliar metabolite profiles of Nicotiana plants exposed to 40 ppb NO2 (blue bars) and 0 ppb NO2 (white bars) under low and high concentrations of root-available NO3, based on three classes of defence-related secondary metabolites (alkaloids, phenolics and diterpene glycosides). a Nicotine; b anatabine; c alkaloid 3; d chlorogenic acid; e caffeic acid 2; f caffeic acid 3; g terpenoid glycoside 1; h terpenoid glycoside 2; i terpenoid glycoside 3; j terpenoid glycoside 4. Alkaloid (ac) and phenolic (df) quantities are given as µg gFW−1; terpenoid glycosides (gj) are given as peak-area gFW−1. An asterisk (*) denotes a significant main or interactive effect of NO2 treatment; a dagger (†) denotes a significant main or interactive effect of NO3 treatment and ns indicates not significant (GLMM). Data represent N = 3 chamber averages with N = 4 plants per chamber, and error bars are +1 SE, with full statistical results provided in Supplementary Table 2

Effect of herbivory on foliar NO2 uptake

The use of 15N-enriched NO2 allowed us to test whether herbivore-induced changes to plant physiology would influence foliar uptake of NO2 and its allocation among plant tissues. Plants exposed to herbivory contained 36% less NO2-derived N (Fig. 3). Larger total amounts of 15N were incorporated under high root NO3, consistent with the larger plant size in this treatment (Supplementary Figure 2a, Supplementary Table 3). Lower 15N in damaged leaf tissue likely resulted in part from differential efflux of NO2 metabolites from leaves following uptake, as indicated by the distribution of 15N: herbivory decreased the proportion of the plant’s total 15N in leaves by an average of 16% and caused a corresponding increase of 93% in roots, after accounting for the net reduction in total uptake (Fig. 3a, Supplementary Table 4). The effect of herbivory on 15N content did not appear to be due to changes in overall leaf %N (Supplementary Figure 2). The reduction in total 15N across all tissues suggested that herbivory also directly reduced foliar uptake rates as a result of induced changes in plant metabolism and/or stomatal conductance29. To independently test this hypothesis, we grew an additional set of N. tabacum plants at 50 mM NO3, damaged half with neonate M. sexta (5–7% leaf area removed) and then exposed them to 40 ppb 15NO2. Prior feeding by herbivores caused significant reductions in 15N assimilation in both damaged and undamaged leaves, confirming that herbivory reduced plant-wide foliar uptake (Fig. 3b, Supplementary Table 6). Consistent with results of our main experiment, NO2 exposure caused significant increases in overall foliar %N that were offset by reductions in damaged plants (Supplementary Figure 3, Supplementary Table 6). These results indicate that herbivory will strongly modulate the role of vegetation as a sink for atmospheric NOx compounds in at least two ways: first, by altering the physiological controls on NO2 uptake (i.e., metabolism and/or stomatal behaviour), and second, by reducing plant size through consumption (physical leaf area). The systemic reduction in 15N assimilation under relatively low damage (<10%) suggests that herbivores, which are ubiquitous even in urban environments, could represent a significant control on how readily plants assimilate and sequester reactive N pollutants.

Fig. 3

Root N availability and herbivory alter foliar uptake of NO2. a Distribution of NO2-derived N, as measured by average of the total mg of 15N in root (orange bars), stem (grey bars) and leaf (green bars) tissues under low (50 mM) and high (500 mM) levels of NO3 in the presence and absence of herbivore damage (“Dam”). Data represent N = 12 plants. b Foliar uptake under 40 ppb NO2 in plants growing at low (50 mM) NO3, following damage and induction by herbivores. Light green bars are data for control plants. Dark green bars represent data for herbivore-exposed plants (“Dam”); hatched bars are data for the site of herbivore feeding, while unhatched bars are values for systemically induced leaves that were not directly consumed. Data represent N = 6 plants, and error bars are ±1 SE. For each tissue type, an asterisk (*) denotes significant effect of NO2 treatment, a dagger (†) denotes a significant effect of NO3 treatment and a section sign (§) denotes a significant effect of herbivore exposure. Full statistical results provided in Supplementary Tables 4 and 6


We have identified a feedback between atmospheric reactive N, plants and consumer fitness that has implications for our understanding of plant responses to global change (Fig. 4). First, NO2 caused increases in N-based plant defences and reductions in herbivore performance, suggesting that increasing levels of NO2 could have profound consequences for consumers, particularly in urban ecosystems. In turn, herbivore-induced reductions in foliar uptake may impose a limit on the effects of NO2 on defensive chemistry. However, the stability of this feedback remains unclear: while low herbivory caused pronounced reductions in NO2 uptake, the dose–response relationship between NO2 and defence remains unknown. In one scenario, plants in high NOx environments would exhibit chronically elevated defences, with negative consequences for consumers, while the presence of even diminished herbivore populations could reduce the beneficial sequestration of NOx and foliar cycling of reactive N. Models of ecosystem functioning under global change rarely consider interactions with herbivores. However, in a recent study of CO2 enrichment, herbivory significantly reduced plant carbon sequestration30, which, together with our results, indicates a need to better understand how consumers will influence ecosystem and plant-level responses to different atmospheric changes.

Fig. 4

Summary of feedbacks between NO2/NOx, NO3, plants and herbivores. Direct effects are denoted by solid arrows and indirect interactions by dashed arrows. NO2 caused significant increases in total foliar alkaloids, e.g. nicotine (solid green arrow), and a corresponding decrease in herbivore performance. In turn, feeding by herbivores caused a decrease in foliar NO2 uptake. Plants in high NO3 conditions decreased investment in C-based defensive metabolites, but herbivores on these plants performed relatively poorly. NO2 exposure and NO3 level did not strongly influence C-based and N-based defensive secondary metabolite production, respectively

Our study also suggests that NOx emissions will interact with other global change drivers with synergistic negative effects on consumers. For example, elevated carbon-to-nitrogen (C:N) ratios in plants growing under rising CO2 are predicted to challenge herbivores to increase consumption rates in order to maintain a balance of limiting nitrogen31,32. However, herbivory will also depend critically on interactions among other anthropogenic changes, including the effects of NOx and CO2 on plant defences, and effects of N deposition on plant %N and responses to CO233,34,35. Moreover, in our experiments herbivore performance was not related to the effects of NOx on leaf C:N or %N (Supplementary Figure 2), highlighting the importance of understanding how reactive N, CO2 and other anthropogenic factors jointly influence defensive metabolites, in addition to specific nutritive indicators (e.g., soluble protein). Our data suggest that there will be significant consequences of these interactions for plant-feeding insects, including agricultural pests and rare species already vulnerable to the effects of soil N deposition on host plant abundance11. Recent demonstration of pronounced declines in arthropod diversity and abundance in regions with long histories of industrialisation36,37 further underscores the need to understand the wide-ranging environmental impacts of atmospheric reactive N on plant and insect communities.


Plant and insect material

We used a model plant–herbivore system consisting of tobacco, N. tabacum L., (Solanaceae) and its natural leaf feeding herbivore, the tobacco hornworm, M. sexta L. (Lepidoptera: Sphingidae). Tobacco was selected because it is a fast-growing herbaceous species that grows well in hydroponic systems, exhibits significant N reduction in leaf tissue38 and has a well-characterised suite of defence-related secondary metabolites39. Tobacco hornworm was chosen because it is a native Solanaceae specialist that readily feeds on tobacco and rarely leaves a tobacco host plant with sufficient leaf material; experimental larvae were obtained from a large in-house colony. Seeds (N. tabacum x. sandera) were purchased from a commercial producer (Paramount Seeds, Inc., Palm City, FL, USA) and germinated and grown in perlite (Sun-Gro Horticulture, Bellevue, WA, USA) in a climate-controlled growth chamber (EGC, Chagrin Falls, OH, USA) at day and night temperatures of 27 and 21 °C, respectively, under moderate light (700 μmol m−2 s−1) and a 16-h photoperiod. Plants were watered daily to saturation for 2 weeks and fertilised with a Hoagland’s solution containing NO3 of known δ15N. Forty-eight 2-week-old plants were transplanted into the hydroponics–fumigation system and used for the experiment. The remaining 12 plants were harvested and measured for biomass, leaf area and isotopic composition to provide baseline data for the experimental plants.

Hydroponics–fumigation system and experimental design

We used a factorial design in which we manipulated foliar exposure to atmospheric reactive N (simulating pristine and urban environments), together with the availability of root N (low vs. moderate), using a coupled hydroponics–fumigation system. The system consisted of 4, 50-L polyethylene nutrient tanks (120 × 58 × 16 cm3) each fitted with three airtight Plexiglas fumigation chambers (36 × 25 × 43 cm3) with an opaque base. The system was located in a greenhouse with day and night temperatures of 27 and 21 °C, respectively, relative humidity of 60–70% and moderate light conditions (800 ± 75 μmol m−2 s−1) using natural and supplemental metal halide lighting (400 MH Econo Cool Grow Light, Sunlight Supply, Vancouver, WA, USA) under a 16-h photoperiod. Roots were suspended in nutrient solution (20 °C) in the hydroponics tank via small holes in the chamber base, and shoots were enclosed in fumigation chambers (ambient temperature). Plants were stabilised and fitted with modelling clay (Loctite, Henkel Consumer Adhesives, Avon, OH, USA) at the root–shoot junction at the base of each chamber to ensure an impermeable seal between the fumigation and nutrient solution system components.

Fumigation chambers were supplied with activated charcoal-filtered, ambient air using a reciprocating air compressor (Model C403L, Gardner Denver, Quincy, IL, USA) at a flow rate of 15 L min−1. Half the chambers served as controls, and half the chambers received enriched 15NO2 from a compressed tank (1% NO2: 99% N2, Scott Marrin, Inc., Riverside, CA, USA), which was diluted into the filtered air of randomly selected chambers using high-precision rotometers and mass flow controllers (Models 03216–34 and 32044–00, Cole-Parmer, Vernon Hills, IL, USA) at a fixed partial pressure (40 ppb). NO2 was selected as the atmospheric N source because it is a common atmospheric reactive N compound. The two fumigation treatments were selected to simulate pristine (0 ppb) and realistic 1-h urban (40 ppb) atmospheric NO2 concentrations24,40,41,42,43,44. The δ15N of the enriched 15NO2 was 1720 ± 17‰, providing a large signal separation from the NO3 source. NO2 and nitric oxide (NO) concentrations were monitored using a chemiluminescence NO-NO2-NOx analyser (TECO Model 42, Thermo Environmental Instruments, Inc., Franklin, MA, USA). Exhaust air from each chamber was filtered using activated charcoal and exited the system through an output line extending outside the greenhouse. Temperature and relative humidity within the chambers were monitored using humidity and temperature probes (Model HMP45A, Vaisala, Inc., Boulder, Colorado, USA) connected to a datalogger (Model CR10x, Campbell Scientific, Inc., Logan, Utah, USA).

Two NO3 regimes, simulating low (50 μM) and moderate (500 μM) N availability, were applied to roots using nutrient solutions with fixed concentrations of NO3 as the sole root N source in a modified quarter-strength Hoagland’s solution45. The δ15N value of the NO3 nutrient solutions was −0.56 ± 0.1‰. The nutrient solutions were vigorously aerated at all times to provide adequate oxygenation and ensure complete mixing and pH was maintained at 5.8–6.2 by addition of either KOH or H2SO4. A 130-L reservoir of stock solution was used for each experimental N treatment. Nutrient solutions were replaced weekly to minimise microbial activity and prevent N depletion, and NO3 concentrations were measured weekly using an auto-analyser (Astoria Pacific, Inc., Clackamas, OR, USA).

Each of the four treatment combinations ([1] 0 ppb NO2 and 50 μM NO3, [2] 40 ppb NO2 and 50 μM NO3, [3] 0 ppb NO2 and 500 μM NO3, and [4] 40 ppb NO2 and 500 μM NO3) was replicated in three chambers (N = 12) with four plants/chamber. After 4 weeks of growth under these treatment conditions, 3 neonate (freshly hatched) M. sexta larvae were placed on two plants in each chamber (6 larvae per chamber). We began the herbivory treatments at 4 weeks to maximise the period of NO2 exposure while ensuring that plants had sufficient space within the chambers and did not initiate reproduction (which can alter defence trait expression). To minimise herbivore movement and ensure that feeding was confined to the selected plants, we used the pair of plants at a randomly selected end of the rectangular chamber and visually inspected plants several times per day. Two small cups filled with a wheat-germ diet46 and a single neonate larva were also placed in each chamber to test for effects of unforeseen environmental variation and any direct toxic effects of NO2 exposure. Larvae on both plants and diet were allowed to feed for 10 days and were then removed, allowed to purge their gut contents and peritrophic membranes (gut lining) and weighed as a proxy for performance/fitness25. Individual larvae were then killed by freezing and freeze-dried for subsequent stable isotope analysis; allowing natural purging of the gut lining allowed us to infer insect 15N assimilation without plant contaminants (see next). Plants were harvested for analysis following removal of the herbivores.

Morphological and stable isotope analyses

Whole-plant samples were separated into leaf, shoot and root tissue. Shoot length and leaf number were measured for each individual and leaf area was estimated using a leaf area meter (LI-3100 Area Meter, LI-COR, Inc., Lincoln, NE, USA). Plant samples were then dried, weighed and sub-samples analysed for tissue N and C content and δ15N and δ13C. Previously separated tissue samples were rinsed with deionised water to remove any NO2 deposited to the leaf surface and dried for 3 days at 55 °C in a drying oven. Dried plant and insect tissue samples were weighed, ground to a fine powder with a mortar and pestle and sub-samples of 2.55–3.15 mg were weighed using a microbalance (Model 4504MP8; Sartorius Corp. Edgewood, NY, USA). Tissue N and C contents were measured using a CHN elemental analyser (Model Carlo Erba NC2500; Thermo Finnigan, San Jose, CA, USA). Tissue δ15N and δ13C were measured using a continuous flow isotope ratio mass spectrometer (Model Delta Plus; Thermo Finnigan, San Jose, CA, USA). All analyses were conducted at the Cornell Stable Isotope Laboratory.

Calculation of N source partitioning

Partitioning of plant N among sources (gaseous NO2 and nutrient solution NO3) was calculated using a two-ended linear mixing model47 and published fractionation factors for root NH4+/NO3 assimilation48. Because we used an artificially high enrichment of 15N in the NO2 fumigation, fractionation events associated with foliar uptake were likely not detectable (i.e., the signal separation generated by the tracer greatly exceeded natural fractionations). Using this model, we estimated the total amount of NO2 incorporated via direct foliar uptake during the fumigation period for leaf, stem and root tissues in each experiment. In testing for larval uptake of NO2-derived N, we used the δ15N values directly due to unknown fractionation factors associated with sequential larval instars49 and since larvae differed in their developmental stage due to the treatments.

Plant chemistry

We estimated the levels of defence-related secondary metabolites in control (undamaged) plants using protocols developed for the major classes of tobacco defensive compounds, specifically pyridine alkaloids (e.g., nicotine, anatabine) phenolic compounds (caffeic acids and flavonoids) and terpenoid glycosides50. Considerable prior research has established these compounds as defence-related metabolites in tobacco, N. tabacum, and other Nicotiana species, e.g. refs. 39,51, and defence expression in Nicotiana can be sensitive to root N availability52. We used control plants to measure the effects of NO2 and NO3 treatments on plant secondary metabolism; we could not assess herbivore-induced plant responses in leaf chemistry, since herbivory was variable among treatments (see Results), and variation in herbivory leads to differentially induced metabolite expression. Fresh tissue from a fully expanded leaf was excised (avoiding the midvein), weighed (ca. 100 mg), flash frozen, ground to a fine powder in liquid N2 using a pestle and stored at –80 °C. We homogenised samples using a FastPrep® tissue homogeniser (MP Biomedicals® LLC, Santa Ana, CA, USA) at 6 m s−1 for 90 s using 0.9 g of grinding beads (Zirconia/Silica 2.3 mm, Biospec® Products Inc., Bartlesville, OK, USA) with 1 mL of ice-cold 40% methanol and 0.5% acetic acid solvent. Samples were centrifuged and a 15-µL aliquot of supernatant was analysed by high-performance liquid chromatography (HPLC) using an Agilent® 1100 series HPLC-DAD equipped with a Gemini C18 reverse-phase column (3 µm, 150 × 4.6 mm2, Phenomenex Inc., Torrance, CA, USA) and a standard method50. Alkaloid, phenolic and diterpene glycoside analytes with identifiable ultraviolet spectra were selected and initially quantitated by peak area. Individual compound identification of nicotine, anatabine and chlorogenic acid was based on comparison with authentic standards, and peak areas were converted to µg gFW−1 using standard curves. A third, unidentified pyridine alkaloid (Alkaloid 3) and two unidentified caffeic acid derivatives (Caffeic acid 2 and Caffeic acid 3) were converted to mass equivalents of nicotine and chlorogenic acid, respectively. All analyte quantities were normalised by the fresh sample masses prior to statistical analysis.

Test for herbivore-induced changes to foliar uptake

We observed indications of reduced incorporation of 15N under herbivory in our main experiment, suggesting the potential for herbivores to induce changes to foliar NO2 assimilation. To test this hypothesis and distinguish effects of herbivore-induced plant responses on uptake from effects of induced responses on N allocation (see Results), we conducted a second experiment: We grew a new group of 24 N. tabacum plants in the hydroponics–fumigation system at 50 µM NO3 and 0 ppb 15NO2 for 2 weeks, at which point we applied four neonate M. sexta on each of two plants per chamber, as in the main experiment. Larvae fed for 2 days, at which point damage levels were ca. 5–7%, which is sufficient to cause an induced response in Nicotiana26. Half the chambers then received 15NO2 fumigation at 40 ppb for a subsequent 5 days, and half served as controls (0 ppb). During this period, plants were inspected and 5–6 additional larvae were added to a few plants to maintain similar total damage levels (ca. 10% on each of four leaves) across treatments. Leaf tissue was harvested from control plants and from the damaged and undamaged leaves of the herbivore-treated plants and analysed for incorporation of 15N from NO2 uptake in the four treatments (0 ppb NO2/Control; 0 ppb NO2/Herbivory; 40 ppb NO2/Control; 40 ppb NO2/Herbivory). Analysis of the damaged (locally induced) and undamaged leaves on damaged plants allowed us to test whether herbivory had caused a systemic (plant-wide) induced change in foliar NO2 uptake.

Statistical analysis

Data were confirmed for normality and analysed using standard general linear models in a maximum likelihood framework (JMP® v.13). Larval performance data, larval 15N uptake and plant metabolite data were analysed by two factor models, with nutrient solution (NO3 level) and NO2 fumigation level as fixed factors and their interaction. For consistency, the data for larval performance on diet were analysed in the same way, though there was no predicted effect of NO3. Morphological data were analysed by three factor models with NO3, NO2 and herbivory as fixed factors and their interactions. Data were averaged within individual chambers; analyses with plant average as the observational unit did not qualitatively change the results, indicating that our experiment had sufficient statistical power for the observed effect sizes for these response variables. Plant 15N uptake, %N and C:N data were analysed by three-factor models that included NO3, NO2 and herbivory levels. We used plant averages for these analyses to compensate for the reduced power available in higher-order models; however, plant-level and chamber-level averages again gave similar results. We analysed foliar uptake data using separate models for (a) damaged leaves, (b) undamaged leaves and (c) pooled leaf data from plants exposed to herbivores, in order to verify that plant responses were consistent at the local and systemic (plant-wide) scale. Means and standard errors for uptake rates for all treatments and tissue types are provided in Supplementary Table 7. Finally, the effect of NO2 exposure on secondary metabolites prompted us to determine whether there was a quantitative relationship between e.g. alkaloid content and the amount of NO2-derived N taken up by leaves. For this, we used plants in the 40 ppb NO2 treatment and tested for correlations (restricted maximum likelihood estimation) between total amounts of each metabolite class in each plant and the amount of 15NO2-derived N and total percentage of nitrogen in leaves.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.


  1. 1.

    Galloway, J. N. et al. Transformation of the nitrogen cycle: recent trends, questions, and potential solutions. Science 320, 889–892 (2008).

    ADS  CAS  Article  Google Scholar 

  2. 2.

    Townsend, A. R. et al. Human health effects of a changing global nitrogen cycle. Front. Ecol. Environ. 1, 240–246 (2003).

    Article  Google Scholar 

  3. 3.

    Basto, S. et al. Long-term nitrogen deposition depletes grassland seed banks. Nat. Commun. 6, 6185 (2015).

    CAS  Article  Google Scholar 

  4. 4.

    Woodward, G. et al. Continental-scale effects of nutrient pollution on stream ecosystem functioning. Science 336, 1438–1440 (2012).

    ADS  CAS  Article  Google Scholar 

  5. 5.

    European Commission. Nitrogen Pollution and the European Environment: Implications for Air Quality Policy. (accessed 14 Nov 2017) (Science Communication Unit, University of the West of England, Bristol, UK, 2013).

  6. 6.

    Thomas, R. Q., Canham, C. D., Weathers, K. C. & Goodale, C. L. Increased tree carbon storage in response to nitrogen deposition in the US. Nat. Geosci. 3, 13–17 (2010).

    ADS  Article  Google Scholar 

  7. 7.

    Sparks, J. P., Roberts, J. M. & Monson, R. K. The uptake of gaseous organic nitrogen by leaves: a significant global nitrogen transfer process. Geophys. Res. Lett. 30, 2189 (2003).

    ADS  Article  Google Scholar 

  8. 8.

    Canfield, D. E., Glazer, A. N. & Falkowski, P. G. The evolution and future of Earth's nitrogen cycle. Science 330, 192–196 (2010).

    ADS  CAS  Article  Google Scholar 

  9. 9.

    Isaksen, I. S. A. et al. Atmospheric composition change: climate-chemistry interactions. Atmos. Environ. 43, 5138–5192 (2009).

    ADS  CAS  Article  Google Scholar 

  10. 10.

    Payne, R. J., Dise, N. B., Stevens, C. J., Gowing, D. J. & Partners, B. Impact of nitrogen deposition at the species level. PNAS 110, 984–987 (2013).

    ADS  CAS  Article  Google Scholar 

  11. 11.

    Hernandez, D. L. et al. Nitrogen pollution is linked to US listed species declines. Bioscience 66, 213–222 (2016).

    Article  Google Scholar 

  12. 12.

    Field, C. D. et al. The role of nitrogen deposition in widespread plant community change across semi-natural habitats. Ecosystems 17, 864–877 (2014).

    CAS  Article  Google Scholar 

  13. 13.

    Phoenix, G. K. et al. Impacts of atmospheric nitrogen deposition: responses of multiple plant and soil parameters across contrasting ecosystems in long-term field experiments. Glob. Change Biol. 18, 1197–1215 (2012).

    ADS  Article  Google Scholar 

  14. 14.

    Simkin, S. M. et al. Conditional vulnerability of plant diversity to atmospheric nitrogen deposition across the United States. PNAS 113, 4086–4091 (2016).

    ADS  CAS  Article  Google Scholar 

  15. 15.

    Throop, H. L. & Lerdau, M. T. Effects of nitrogen deposition on insect herbivory: implications for community and ecosystem processes. Ecosystems 7, 109–133 (2004).

    CAS  Article  Google Scholar 

  16. 16.

    Rogers, H. H., Campbell, J. C. & Volk, R. J. Nitrogen-15 dioxide uptake and incorporation by Phaseolus vulgaris (L). Science 206, 333–335 (1979).

    ADS  CAS  Article  Google Scholar 

  17. 17.

    Sparks, J. P., Monson, R. K., Sparks, K. L. & Lerdau, M. Leaf uptake of nitrogen dioxide (NO2) in a tropical wet forest: implications for tropospheric chemistry. Oecologia 127, 214–221 (2001).

    ADS  Article  Google Scholar 

  18. 18.

    Vallano, D. & Sparks, J. Quantifying foliar uptake of gaseous nitrogen dioxide using enriched foliar delta N-15 values. New Phytol. 177, 946–955 (2008).

    CAS  Article  Google Scholar 

  19. 19.

    Teklemariam, T. A. & Sparks, J. P. Leaf fluxes of NO and NO2 in four herbaceous plant species: the role of ascorbic acid. Atmos. Environ. 40, 2235–2244 (2006).

    ADS  CAS  Article  Google Scholar 

  20. 20.

    Zehnder, C. B. & Hunter, M. D. Effects of nitrogen deposition on the interaction between an aphid and its host plant. Ecol. Entomol. 33, 24–30 (2008).

    Article  Google Scholar 

  21. 21.

    Dohmen, G. P., McNeill, S. & Bell, J. N. B. Air-pollution increases Aphis fabae pest potential. Nature 307, 52–53 (1984).

    ADS  CAS  Article  Google Scholar 

  22. 22.

    Masters, G. J. & McNeill, S. Evidence that plant varieties respond differently to NO2 pollution as indicated by resistance to insect herbivores. Environ. Pollut. 91, 351–354 (1996).

    CAS  Article  Google Scholar 

  23. 23.

    Mattson, W. J. Herbivory in relation to plant nitrogen content. Ann. Rev. Ecol. Syst. 11, 119–161 (1980).

    Article  Google Scholar 

  24. 24.

    E.P.A. (U.S.). Nitrogen Dioxide Trends (U.S. Environmental Protection Agency, Washington D.C., 2017). Accessed 05 Feb 2018.

  25. 25.

    Diamond, S. E. & Kingsolver, J. G. Fitness consequences of host plant choice: a field experiment. Oikos 119, 542–550 (2010).

    Article  Google Scholar 

  26. 26.

    Kessler, A. & Baldwin, I. T. Plant responses to insect herbivory: the emerging molecular analysis. Ann. Rev. Plant Biol. 53, 299–328 (2002).

    CAS  Article  Google Scholar 

  27. 27.

    Neill, S. J., Desikan, R. & Hancock, J. T. Nitric oxide signalling in plants. New Phytol. 159, 11–35 (2003).

    CAS  Article  Google Scholar 

  28. 28.

    Stamp, N. Out of the quagmire of plant defense hypotheses. Q. Rev. Biol. 78, 23–55 (2003).

    Article  Google Scholar 

  29. 29.

    Chaparro-Suarez, I. G., Meixner, F. X. & Kesselmeier, J. Nitrogen dioxide (NO2) uptake by vegetation controlled by atmospheric concentrations and plant stomatal aperture. Atmos. Environ. 45, 5742–5750 (2011).

    ADS  CAS  Article  Google Scholar 

  30. 30.

    Couture, J. J., Meehan, T. D., Kruger, E. L. & Lindroth, R. L. Insect herbivory alters impact of atmospheric change on northern temperate forests. Nat. Plants 1, 15016 (2015).

    CAS  Article  Google Scholar 

  31. 31.

    Stiling, P. & Cornelissen, T. How does elevated carbon dioxide (CO2) affect plant-herbivore interactions? A field experiment and meta-analysis of CO2-mediated changes on plant chemistry and herbivore performance. Glob. Change Biol. 13, 1823–1842 (2007).

    ADS  Article  Google Scholar 

  32. 32.

    Percy, K. E. et al. Altered performance of forest pests under atmospheres enriched by CO2 and O3. Nature 420, 403–407 (2007).

    ADS  Article  Google Scholar 

  33. 33.

    Reich, P. B. et al. Nitrogen limitation constrains sustainability of ecosystem response to CO2. Nature 440, 922–925 (2006).

    ADS  CAS  Article  Google Scholar 

  34. 34.

    Throop, H. L., Holland, E. A., Parton, W. J., Ojima, D. S. & Keough, C. A. Effects of nitrogen deposition and insect herbivory on patterns of ecosystem-level carbon and nitrogen dynamics: results from the CENTURY model. Glob. Change Biol. 10, 1092–1105 (2004).

    ADS  Article  Google Scholar 

  35. 35.

    Greaver, T. L. et al. Key ecological responses to nitrogen are altered by climate change. Nat. Clim. Change 6, 836–843 (2016).

    ADS  CAS  Article  Google Scholar 

  36. 36.

    Hallmann, C. A. et al. More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLoS ONE 12, e0185809 (2017).

    Article  Google Scholar 

  37. 37.

    Vogel, G. Where have all the insects gone? Science 356, 575–578 (2017).

    ADS  Article  Google Scholar 

  38. 38.

    Wakhloo, J. L. & Staudt, A. Development of nitrate reductase activity in expanding leaves of Nicotiana tabacum in relation to the concentration of nitrate and potassium. Plant Physiol. 87, 258–263 (1988).

    CAS  Article  Google Scholar 

  39. 39.

    Kaplan, I., Halitschke, R., Kessler, A., Sardanelli, S. & Denno, R. F. Constitutive and induced defenses to herbivory in above- and belowground plant tissues. Ecology 89, 392–406 (2008).

    Article  Google Scholar 

  40. 40.

    Tong, D. Q. et al. Long-term NOx trends over large cities in the United States during the great recession: comparison of satellite retrievals, ground observations, and emission inventories. Atmos. Environ. 107, 70–84 (2015).

    ADS  CAS  Article  Google Scholar 

  41. 41.

    Guerreiro, C. B. B., Foltescu, V. & de Leeuw, F. Air quality status and trends in Europe. Atmos. Environ. 98, 376–384 (2014).

    ADS  CAS  Article  Google Scholar 

  42. 42.

    E.P.A. (U.S.). Report on the Environment (Nitrogen Dioxide) (U.S. Environmental Protection Agency, Washington, D.C., 2018).

  43. 43.

    Pike, R. C. et al. NOx and O-3 above a tropical rainforest: an analysis with a global and box model. Atmos. Chem. Phys. 10, 10607–10620 (2010).

    ADS  CAS  Article  Google Scholar 

  44. 44.

    Lee, D. S. et al. Estimations of global NOx emissions and their uncertainties. Atmos. Environ. 31, 1735–1749 (1997).

    ADS  CAS  Article  Google Scholar 

  45. 45.

    Smart, D. R. & Bloom, A. J. The relationship between kinetics of NH4+ and NO3- absorption and growth in cultivated tomato (Lycopersicon esculentum Mill cv. T5). Plant Cell Environ. 16, 259–267 (1993).

    CAS  Article  Google Scholar 

  46. 46.

    Yamamoto, R. T. Mass rearing of tobacco hornworm 2. Larval rearing and pupation. J. Econ. Entomol. 62, 1427–1431 (1969).

    Article  Google Scholar 

  47. 47.

    Dawson, T. E., Mambelli, S., Plamboeck, A. H., Templer, P. H. & Tu, K. P. Stable isotopes in plant ecology. Ann. Rev. Ecol. Syst. 33, 507–559 (2002).

    Article  Google Scholar 

  48. 48.

    Kolb, K. J. & Evans, R. D. Influence of nitrogen source and concentration on nitrogen isotopic discrimination in two barley genotypes (Hordeum vulgare L.). Plant Cell Environ. 26, 1431–1440 (2003).

    CAS  Article  Google Scholar 

  49. 49.

    Tibbets, T. M., Wheeless, L. A. & del Rio, C. M. Isotopic enrichment without change in diet: an ontogenetic shift in delta N-15 during insect metamorphosis. Funct. Ecol. 22, 109–113 (2008).

    Google Scholar 

  50. 50.

    Keinanen, M., Oldham, N. J. & Baldwin, I. T. Rapid HPLC screening of jasmonate-induced increases in tobacco alkaloids, phenolics, and diterpene glycosides in Nicotiana attenuata. J. Agric. Food Chem. 49, 3553–3558 (2001).

    CAS  Article  Google Scholar 

  51. 51.

    Kessler, A., Halitschke, R. & Baldwin, I. T. Silencing the jasmonate cascade: induced plant defenses and insect populations. Science 305, 665–668 (2004).

    ADS  CAS  Article  Google Scholar 

  52. 52.

    Lou, Y. G. & Baldwin, I. T. Nitrogen supply influences herbivore-induced direct and indirect defenses and transcriptional responses to Nicotiana attenuata. Plant Physiol. 135, 496–506 (2004).

    CAS  Article  Google Scholar 

Download references


We thank A. Kessler (Cornell University) for generous use of the HPLC instrument and advice; K.E. Sparks, J.P. Sparks and A. Kasson (Cornell Stable Isotope Laboratory) and R. Halitschke (Cornell University; Max Planck Institute for Chemical Ecology) for technical assistance and advice; M. del Campo for assistance with Manduca sexta; and P. Cooper for assistance with greenhouse management. This study was supported by a Biogeochemistry and Environmental Biocomplexity Initiative Small Grant awarded to S.A.C. and D.M.V. through a National Science Foundation (U.S.A.) Integrative Graduate Education and Research Traineeship (NSF-IGERT DGE-0221658); S.A.C. was supported by a Natural Sciences and Engineering Research Council (NSERC Canada) PGS-D fellowship and an Independent Research Fellowship from the University of Sheffield P3 Centre for Translational Plant Science (BB/IAA/Sheffield/15). M. Stastny, A. Kessler and J. Sparks provided feedback on the project and/or drafts of this manuscript.

Author information




S.A.C. and D.M.V. conceived and designed the study, collected and analysed the data and wrote the paper.

Corresponding author

Correspondence to Stuart A. Campbell.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Campbell, S.A., Vallano, D.M. Plant defences mediate interactions between herbivory and the direct foliar uptake of atmospheric reactive nitrogen. Nat Commun 9, 4743 (2018).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing