Neighbour-detection causes shifts in allocation across multiple organs to prepare plants for light competition

1. To maximize their fitness, plants have to adjust their allocation strategy according to their abiotic and biotic

the availability of light and belowground resources both vary with ecological conditions, plants adapt their allocation strategies according to both resource availability and resource competition (Poorter et al., 2012).
The shade avoidance response is one way that plants adjust their allocation according to ecological conditions.The shade avoidance response is a suite of physiological and morphological responses that are triggered when plants are exposed to light with a low red to far-red ratio (R:FR).A low R:FR is a cue for the presence of aboveground competition that plants use to induce an increase in stem growth.Since leaves absorb more red light than far-red light, shade cast by plants has a lower R:FR than shade cast by neutral sources.Plants that are not directly shaded by neighbouring plants also have a reduced R:FR because neighbouring plants reflect more far-red than red light (Roig-Villanoca & Martínez-Garcia, 2016).This allows plants to use low R:FR not only as a cue for neighbours that are actively shading them, but also as a cue for neighbours that may shade them in the future (Ballaré & Pierik, 2017;Franklin & Whitelam, 2005).Thus, the shade avoidance response provides plants with a mechanism for adjusting their growth based on current competition as well as the threat of future competition (Fiorucci & Fankhause, 2017).
Since plants engage in height-structured competition for light, plants that do not invest in stem growth risk growing in the shade of neighbours (Anten, 2016;McNickle & Dybzinski, 2013).Increased stem growth in response to low R:FR is therefore an intuitive response to the presence of aboveground competitors.However, the resources used to elongate the stem must come at the expense of investment in other organs.Previous studies have found that the shade avoidance response is also associated with a reduction in root growth (Ruberti et al., 2012;van Gelderen, Kang, & Pierik, 2018).
A tradeoff between stem growth and root growth may reflect that investing in light acquisition is more valuable than water and nutrient uptake when a plant is in low R:FR.Nevertheless, a reduction in root growth also reduces a plant's ability to uptake nitrogen (Shi et al., 2013).Nitrogen availability can in turn affect primary production through its effect on both total leaf area and the photosynthetic rate per leaf area.Nitrogen limitation may lead to reduced leaf area because leaf growth is often limited by nitrogen availability (Palmer et al., 1996;Seufert et al., 2019;Vos et al., 2005).Nitrogen also plays a key role in the light-saturated photosynthetic rate because the photosynthetic enzymes require nitrogen.This explains why plants generally allocate more nitrogen to leaves in higher photosynthetically active radiation (PAR) and why the photosynthetic rate of leaves can be limited by nitrogen availability (Hikosaka & Terashima, 1995;Mu & Chen, 2021;Touwborst et al., 2011).As a result, the shade avoidance response may lead to a tradeoff between investment in access to light, via stem growth, and investment in the efficient harvesting of that light, via leaf area and photosynthetic rate.
If the shade avoidance response exacerbates nitrogen limitation, plants may need to adjust their nitrogen allocation to effectively compete with neighbours for light.This could explain why leaf nitrogen is not just responsive to PAR, but also R:FR (Kull & Krujit, 1998;Sakuraba, 2021).While the maximum photosynthetic rate increases with increasing PAR, R:FR may be a more reliable indicator of a leaf's expected future productivity than its current PAR.Thus, while PAR and R:FR are generally highly correlated, plants may be able to optimize their nitrogen allocation by decreasing leaf nitrogen in response to both low PAR and low R:FR (Rousseaux et al., 1996).
However, if a plant reduces nitrogen allocation to leaves when R:FR decreases, but PAR is still high, the accompanying reduction in photosynthetic rate could lead to a decrease in leaf productivity.For a drop in leaf productivity to be adaptive, the costs associated with the drop in leaf productivity must be outweighed by the benefits accrued by reallocating nitrogen from leaves in low R:FR.Since plants can translocate nitrogen from source leaves to sink leaves (Boonman et al., 2006;Brant & Chen, 2015;Havé et al., 2017;Masclaux-Daubresse et al., 2010;Rea et al., 2018), it is possible that nitrogen taken from leaves in low R:FR is used to meet the demand for nitrogen generated by leaves in higher R:FR, resulting in a net competitive advantage for the plant.
In this study, we experimentally investigated how exposure to low R:FR triggers shifts in biomass allocation across roots, stems, and leaves as part of a coordinated plant response to the presence of neighbours.We hypothesized that the shade avoidance response (1) exacerbates nitrogen limitation through a tradeoff between stem growth and root growth; (2) partially alleviates that nitrogen limitation by decreasing the nitrogen invested into leaves in low R:FR; and (3) prepares plants for light competition with neighbours by redistributing leaf nitrogen into higher canopy positions.To test these hypotheses, we tracked how biomass and nitrogen allocation changed in common sunflowers (Helianthus annuus L.) subjected to an artificially lowered R:FR treatment and a nitrogen fertilization treatment in a factorial experiment.This allowed us to explore how plants respond to the simultaneous pressures associated with aboveground competition and nitrogen limitation.Rows within a block were separated by 34 cm from stem to stem and plants within a row were separated by 40 cm from stem to stem.This experiment used a two-treatment factorial design, with a supplemental far-red light treatment and a nitrogen fertilization treatment.This yielded plants with no supplemental far-red light and no nitrogen fertilizer (FR−/N−), supplemental far-red light and no nitrogen fertilizer (FR+/N−), supplemental far-red light and nitrogen fertilizer (FR+/N+), and no supplemental far-red light and nitrogen fertilizer (FR−/N+).Each treatment group consisted of 30 plants, yielding a total of 120 plants.

| Experimental treatments
Two blocks were supplemented with far-red light from luminaires positioned in the centre of the rows and 45 cm above the soil surface, with one luminaire per 1.5 m 2 (Forever Green Indoors 730 nm Far-Red LED Grow Light, Washington, USA, 29.8 μmol m −2 s −1 ).The far-red luminaires were on from 06:00-21:00 every day, coinciding with the natural light in the greenhouse.By using supplemental farred light to lower the R:FR of some plants, we were able to study the effect of low R:FR without also limiting the PAR available to plants.
In nature, a low R:FR coincides with a low PAR, but in our experiment, the lower leaves of plants in the FR+ treatment actually received slightly more PAR than those of FR− plants.This is because additional FR light positively affects the photosynthetic rate (Zhen et al., 2019;Zhen & Bugbee, 2020).However, given that the plants were grown in ambient light, the supplemental FR light accounted for less than 2% of the estimated total PAR of FR+ plants.
All plants were fertilized twice per week with 200 mL of 50% nitrogen-free Hoagland's medium (Biotrend, Germany).Half of the plants in each block were assigned to the nitrogen treatment, which received 12 mM of calcium nitrate as part of their fertilization medium.Although the nitrogen treatment also increased the calcium available to N+ plants, none of the experimental plants were expected to be calcium-limited given that there is also calcium present in the nitrogen-free Hoagland's medium.In addition to Hoagland's medium, plants were watered throughout the week with tap water as necessary.Two plants, one in the FR−/N− treatment and one in the FR+/N+ treatment, failed to grow 50 cm tall by the end of the experiment and were excluded from analysis because they were deemed unhealthy.

| Plant measurements
The plants were measured at the start of the experiment and then weekly until the end of the experiment.Each week, the height of the plant as well as the height of petiole insertion, length, width, and SPAD (SPAD-502 Plus, Konica Minolta, Japan) of each true leaf more than 3 cm long was measured.Each leaf was given a unique order to allow for re-measurement of the same leaf across weeks, with the leaf order increasing up the stem.In addition, we estimated the proportion of leaf area that was dead for each measured leaf, with leaves that were completely dead assigned a SPAD and leaf area of zero.We estimated the living leaf area of each leaf using an equation developed for transforming sunflower leaf measurements into area estimates (Leaf area cm 2 = 6.72 + 0.65 Leaf width (cm) 2 ) and then multiplying that area by the proportion of leaves that were alive (Rouphael et al., 2007).
After the third measurement, the far-red luminaires were raised an additional 10 cm so that the lower leaves of the plants continued to be under reduced R:FR.The experiment was stopped after the fifth measurement, when all plants had an immature flower bud 0.5-2 cm above the nearest leaf (R2 stage, Schneiter & Miller, 1981).
Since the FR+ plants eventually grew taller than the height of the far-red luminaires, we distinguished between the bottom 10 leaves, which spent on average more than 1 week under the supplemental far-red light, and the remaining top leaves, which may have never been exposed to the supplemental far-red light (Figure S2E,F).At the end of the experiment, the plants were harvested and separated into leaves (i.e., leaf blades), stem (including petioles), flower buds, and roots.The different plant organs were dried in an oven (65°C for 48 h) and weighed.We used the average of five SPAD readings per leaf to create a proxy for leaf nitrogen per area that was later validated with chemical analysis (Muñoz-Huerta et al., 2013;Netto et al., 2005).This method was chosen because it is non-destructive, allowing us to track the leaf nitrogen of individual leaves throughout the experiment.To estimate the leaf nitrogen per area (N a ) and total leaf nitrogen in a leaf (N t ), we created a validation curve by destructively sampling leaves from sunflowers (Helianthus annuus var.Solara F1) not used in the experiment.To confirm that SPAD measurements were a good proxy for N a across a large range of light intensity and R:FR we used 36 leaves from plants grown under three different light conditions: 12 leaves taken from plants grown in the ambient botanical garden conditions, 12 leaves grown with LED lights in low light (46 μmol m −2 s −1 PAR) and low R:FR conditions (0.26 R:FR), and 12 leaves grown with LED lights in moderate light (460 μmol m −2 s −1 PAR) and high R:FR conditions (2.6 R:FR).We measured the SPAD, leaf dimensions, dry weight, and percent nitrogen content (Elemental Analyser, Lifeasible, NY, USA) of the 36 leaves.We fit a power law relationship between SPAD and N a for all 36 leaves and found the typical relationship between the two variables and no difference across sample groups (Xiong et al., 2015).Given the good fit between SPAD and N a (R 2 = 0.86), we calculated N a by transforming SPAD using our validation curve (N a mmol m −2 = 27.87 + 0.017 SPAD 2.38 , Figure S1).To calculate N t , we multiplied the estimated N a by the living leaf area.
To estimate the nitrogen translocation out of the lower leaves, we calculated the per leaf translocation as Translocation = , where N tmax is the maximum total nitrogen measured in a leaf.
Since this experiment intentionally minimized any PAR gradients within a plant, and leaves were still maturing at the end of the experiment, nitrogen did not exhibit the typical exponential decrease through the plant canopy (Figure S2A,B).To characterize the leaf nitrogen distribution of plants in our experiment without assuming that leaf nitrogen followed a specific distribution, we designed three metrics that capture different aspects of leaf nitrogen distributions.First, we calculated the absolute centre of nitrogen, , where i is leaf order (counting from the bottom upward), n is the maximum leaf order of the plant, and H is the leaf height.This is the average height of leaves in a plant weighted by the relative nitrogen in each leaf and represents the average stem height at which nitrogen is distributed in a plant.To assess whether the treatments affected not just the average absolute height at which nitrogen was distributed, but also the relative height at which nitrogen was distributed, we also calculated the relative By normalizing the leaf heights, the relative centre of nitrogen represents the degree to which nitrogen is skewed towards the top of the stem.Finally, to assess whether the treatments affected the nitrogen distribution among leaves, we calculated the order-based centre of nitrogen This metric replaces leaf height with leaf order and represents the extent to which nitrogen is concentrated in newer leaves, regardless of their height.

| Statistical analysis
After confirming that our analyses did not violate the assumptions of the test, we used a two-factor ANOVA followed by a post hoc Tukey's HSD test to test for differences across treatments.When the assumptions of the ANOVA were violated, we performed a Kruskal-Wallis test followed by post hoc Mann-Whitney U tests.
To test the sensitivity of our analyses, we confirmed that the results were robust to error in the SPAD validation curve by repeating the analyses when estimating N a using the 95% confidence interval of the validation curve (SI).All statistical analysis was conducted with the statsmodels package (version 0.12.2) of Python (Seabold & Perktold, 2010).

| Biomass allocation
The treatments had different effects on biomass partitioning across stems, roots, flower buds, and leaves.The final stem mass of the plants was not affected by either the FR or nitrogen treatments (Figure 1a; Table S1), but FR+ plants were 30% taller than FR− plants at the end of the experiment (Figure 1b, Table S1).Root mass was affected by the FR treatment, but not the nitrogen treatment, with FR+ plants having 64% of the root mass of FR− plants (Figure 1c; Table S1).Flower bud mass was affected by both the FR and nitrogen treatments and their interaction, with FR−/N− plants having significantly lower flower bud mass than the other three groups (Figure 1d; Table S1).
The nitrogen and far-red treatments had a significant effect on the total number of leaves a plant had at the end of the experiment, with only the FR−/N+ treatment having significantly more leaves than the other groups (Figure 2a; Table S1).Both treatments also had an effect on the final leaf mass, leaf area, and total leaf nitro-  S1).

| Leaf and internode development
Leaf development followed a general pattern across the treatments.
At the end of the experiment, total leaf nitrogen (N t ) was highest for leaves of intermediate order.These leaves had both a high area and a high nitrogen content per area (N a ), whereas leaves lower on the stem had low N a due to the onset of senescence and leaves higher on the stem still had a small area (Figure S2A,B).For leaves lower on the plant, N t tended to increase for the first 2-3 weeks, before decreasing at the end of the experiment (Figure S2C).This suggests that regardless of light environment and nitrogen availability, older leaves eventually became nitrogen sources as part of their normal development.In contrast, leaves higher on the plant were still increasing in N t at the end of the experiment (Figure S2D).This value represents the maximum N a allocated to a leaf at maturity in our greenhouse conditions.

| Leaf nitrogen: Lower leaves
When analysing the bottom 10 leaves, the FR+ groups had significantly lower average maximum leaf area, average maximum N a , and maximum total leaf nitrogen than the FR−groups (Figure 3a,c; Table S2).This amounted to FR+ plants investing c. 13 mg, or 35%, less nitrogen in their bottom 10 leaves than FR− plants at maturity.In contrast, only the nitrogen treatment affected the average nitrogen translocation out of the lower ten leaves (Table S2), with the N− leaves losing around 48% of their nitrogen and N+ leaves losing around 34% of their nitrogen by the end of the experiment.
Both the far-red and nitrogen treatments had a significant impact on the average N a of the bottom leaves at the end of the experiment (Figure 3b; Table S2).Both the far-red and nitrogen treatments had a significant impact on the N t of the bottom leaves at the end of the experiment (Figure 3d; Table S2).

| Leaf nitrogen: Upper leaves
The upper leaves of the plants responded to the treatments differently than the bottom 10 leaves.The average N a of the top leaves was affected by the nitrogen treatment and the interaction between the nitrogen and far-red treatments (Figure 4b; Table S2).
Post hoc comparisons found that N− plants had lower average upper leaf N a than N+ plants, but there were no significant differences within each nitrogen treatment.Both treatments and their interaction had a significant effect on the total leaf area and the total N t of the top leaves at the end of the experiment (Table S2),

| Leaf nitrogen distribution
The average maximum N a of upper and lower leaves did not differ significantly between the FR+/N− and FR−/N+ treatments (Figure 4d).
However, the lower leaves in the FR−/N− treatment had a significantly higher average maximum N a than their upper leaves, while the lower leaves in the FR+/N+ treatment had a significantly lower average maximum N a than their upper leaves.This divided the leaves  S1), with the FR+ plants having significantly higher centres of nitrogen than FR− plants.The higher CN A in FR+ plants relative to FR− plants corresponds to FR+ plants centering their nitrogen roughly 20 cm higher than FR− plants due to their increased height growth.However, the higher CN R in FR+ plants means that even after controlling for the increased height of FR+ plants, the leaf nitrogen in FR+ plants was still centred proportionally higher on the stem than in FR− plants.This is because the increased internode elongation of the bottom 10 internodes in FR+ plants resulted in their upper leaves being concentrated relatively higher on the stem compared with FR− plants.Finally, the higher CN O in FR+ plants shows that FR+ plants shift their N t distribution more towards their higher order leaves than FR− plants (Figure S2A).
This means that leaf nitrogen was allocated to relatively younger leaves in FR+ plants relative to FR− plants.

| DISCUSS ION
Plants must allocate resources efficiently to maximize their fitness, but these allocation decisions come with tradeoffs.The shade avoidance response allows a plant to increase in height when there is competition for light or there is likely to be competition for light

| Low R:FR drives a tradeoff between stem elongation and leaf growth via nitrogen limitation
As expected, plants exposed to supplemental far-red light shifted their allocation on the whole-plant scale.The internodes under farred light grew faster and longer than those without the far-red treatment, while the internodes above the far-red light were not different from controls (Figure S2E,F).This supports previous evidence of internode extension being driven by local perception of R:FR (Ma & Li, 2019;Morgan & Smith, 1976;Rousseaux et al., 1997).The use of local information prevents plants from unnecessarily elongating the internodes of leaves that are not at risk of being shaded by competitors.Although FR+ plants did not have significantly more stem mass than FR− plants at the end of the experiment, their stem extension likely required other metabolic costs associated with the extension process itself (Mazzella et al., 2008).
Our results indicate that the increased stem height of FR+ plants coincided with a decrease in root mass, suggesting that internode extension came at the expense of root growth.This follows the results of other studies, which found that the shade avoidance response leads to decreased root growth through shoot-root communication (Salisbury et al., 2007;van Gelderen, Kang, Paalman, et al., 2018;van Gelderen, Kang, & Pierik, 2018).
We can rank the nitrogen status of the treatment groups using their root mass and their access to nitrogen fertilizer.In this framework, we assumed that the FR+/N− group had the lowest nitrogen supply due to a low root mass and low nitrogen availability, the

| Nitrogen limitation is mitigated by the direct effects of R:FR on leaf development
Leaves that primarily developed in low R:FR conditions had smaller leaf areas, less peak nitrogen per leaf area, and less peak total nitrogen than comparable leaves that developed without supplemental far-red light regardless of whether they received nitrogen fertilizer (Figure 4a-c).The fact that far-red light had an effect on the development of lower leaves, but not the leaves on the same plant that were above the far-red lights suggests that R:FR has a direct and local effect on leaf development.This is supported by other experiments that have found evidence that R:FR perception at the leaflevel affects leaf expansion and local nitrogen allocation (Carabelli et al., 2007;Kozuka et al., 2005;Pons & de Jong-Van Berkel, 2004;Rousseaux et al., 1997).Thus, our results suggest that while low R:FR results in a decrease in nitrogen supply, this is mitigated by a decrease in nitrogen demand from the leaves that are most likely to be shaded in the future.
While low R:FR decreased the maximum investment into lower leaves, the nitrogen translocated out of leaves was controlled by whether the plant received nitrogen fertilizer, suggesting that nitrogen resorption is a function of the nitrogen status of the plant (Agüera et al., 2010;Oikawa et al., 2006;Ono et al., 1996Ono et al., , 2001;;Yasumura et al., 2007).Since nitrogen translocation takes both time and resources, limiting the initial investment in low R:FR leaves rather than increasing translocation from those leaves may minimize over-investment into leaves in poor competitive positions.Additionally, by only translocating nitrogen according to the unmet demand of other nitrogen sinks, a plant can keep productive lower leaves for longer when the opportunity cost of translocating leaf nitrogen is low, allowing them to maintain high net productivity.
In our experiment, leaves in the low R:FR environment actually had slightly higher incident PAR than controls because far-red photons can increase photosynthesis (Zhen et al., 2019;Zhen & Bugbee, 2020).Based on PAR, the far-red treated leaves, therefore, had a higher maximum potential photosynthetic rate than controls, further showing that their reduced leaf growth and leaf nitrogen were due to neighbour-detection rather than decreased PAR (Anten et al., 1998;Rousseaux et al., 1996Rousseaux et al., , 1999Rousseaux et al., , 2000)).This shows that when nitrogen limitation is low, the shade avoidance response allows plants to preferentially allocate their nitrogen to their high R:FR leaves.
The combination of increased internode elongation and decreased nitrogen investment in leaves in response to low R:FR allowed FR+ plants to skew their leaf nitrogen distributions towards higher, younger leaves concentrated near the top of the stem (Figure 5).By deprioritizing lower leaves, FR+ plants were able to distribute their leaf nitrogen in a way that focuses nitrogen into leaves that are the least likely to be shaded in the future.This further suggests that the shade avoidance response allows plants to preemptively abandon leaves before they become unproductive so that those resources can be diverted to leaves with better long-term access to high PAR, even if this comes at the expense of current productivity.Our results provide additional evidence that plant growth and allocation are unlikely to optimize current productivity and that plants engaged in ongoing or future competition for light will sacrifice leaf area, as well as current carbon and nitrogen gain, to position their leaf area in more competitive positions (Ackerly, 1999;Anten & During, 2011;Franklin & Agren, 2002;Hirose & Werger, 1987;McNickle & Dybzinski, 2013).

| CON CLUS IONS
Our results suggest that R:FR has multiple direct and indirect effects on organ growth and nitrogen allocation.First, internodes in low R:FR elongate, but this growth comes at the expense of root growth.This decreased root growth leads to a reduction in nitrogen uptake, which in turn limits leaf growth.Thus, low R:FR forces plants

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors have no conflicting interests.
Common sunflowers (Helianthus annuus var.Solara F1) were germinated and grown at the Botanical Garden of the University of Konstanz, Germany (N: 47°69′19.56″,E: 9°17′78.42″)from July 2022 to August 2022.After germination, plants were transplanted into 2 L pots filled with potting soil (Einheitserde, Gebr.Patzer GmbH & Co. KG, Sinntal, Germany) and watered with tap water as necessary.After the plants had at least two true leaves over 3 cm long, we selected 120 of the healthiest plants to be used in the experiment.They were arranged in four blocks, with each block containing two rows of 15 plants and each block separated by at least 1 m.
Individuals FR−/N− treatment: 29 FR+/N− treatment: 30 FR+/N+ treatment: 29 FR−/N+ treatment: 30 gen (Kruskal-Wallis; H(3) = 78.71,p = 5.79 × 10 −17 ), with post hoc tests finding the same statistical pattern across treatments for leaf masses, leaf areas, and total leaf nitrogen.FR−/N+ plants had the highest values, FR−/N− and FR+/N+ treatments had statistically similar and intermediate values, and FR+/N− plants had the lowest values (Figure 2b-d; Table It generally took around 2 weeks for leaves to reach their final height, with only the first eight internodes reaching their final length before the end of the experiment and the lower internodes of far-red treated plants growing longer and more quickly than those of non-far-red treated plants (Figure S2E,F).Regardless of leaf position or treatment, plants did not allocate more than an average of c. 125 mmol m −2 of nitrogen to their leaves.
Post hoc tests revealed that while FR+ plants had the lowest maximum N a , N− plants lost the highest proportion of their nitrogen.This resulted in FR+/N− plants having the lowest average final N a in their bottom leaves, followed by FR−/N− plants, FR+/N+ plants, and finally FR−/N+ plants.In this case, the high translocation out of the bottom 10 leaves of FR−/ N− plants outweighed their higher maximum N a relative to FR+/ N+ plants.
Post hoc tests revealed that the FR−/N+ treatment had the highest values of final lower leaf N t , the FR−/ N− and FR+/N+ treatments had statistically similar and intermediate values, and the FR+/N− treatment had the lowest values.The pattern of N t was similar to N a , but since the bottom leaves of FR+/N+ plants were smaller than those in FR−/N− plants, the lower N a of the bottom leaves of FR−/N− plants was balanced by their larger area relative to FR+/N+ plants, leading to similar final N t values.Thus, FR+/N+ plants invested less total nitrogen in their bottom 10 leaves than FR−/N− plants, but translocated a lower proportion of their nitrogen from those leaves, resulting in comparable final N t values.In all, FR+/N− plants had around 15 mg of nitrogen in their bottom 10 leaves at the end of the experiment, as compared with an average of 23 mg for FR−/N− and FR+/N+ plants, and 30 mg for FR−/N+ plants.
with post hoc tests finding that the FR−/N+ treatment had the highest upper leaf area and N t , the FR+/N+ treatment had intermediate values, and the lowest values were found in the statistically similar FR−/N− and FR+/N− treatments (Figure4a,c).Unlike whole-plant N t and bottom leaf N t , the N− groups had the lowest upper leaf N t , with on average 32 mg of nitrogen in their upper leaves, followed by 43 mg for FR+/N+ plants, and 58 mg for FR−/ N+ plants.F I G U R E 1 The final (a) stem mass, (b) height, (c) root mass, and (d) flower bud mass of each treatment at the end of the experiment.Boxes indicate the interquartile range, bars indicate the rest of the distribution excluding outliers, and the significance post hoc comparisons are indicated with * (0.01 < p ≤ 0.05), ** (0.001 < p ≤ 0.01), *** (1 × 10 −4 < p ≤ 0.001), and **** (p < 1 × 10 −4 ).
into three categories based on their maximum average N a .The highest average maximum N a was found in the lower leaves of FR−/N− plants, the upper leaves of FR+/N+ plants, and both the lower and upper leaves of FR−/N+, which all reached the maximum N a in our experiment of around 125 mmol m −2 .The lower leaves of FR+/N+ plants and both the upper and lower leaves of FR+/N− plants made up an intermediate group of leaves with similar average maximum N a values of 108 mmol m −2 .Lastly, the upper leaves of FR−/N− plants had the lowest average maximum N a of 100 mmol m −2 , with only the upper FR+/N− leaves having statistically similar values across all pairwise comparisons.Hence, only the FR+/N− plants, which are assumed to have the lowest nitrogen uptake, were unable to produce leaves with a high N a .All other plants were able to produce high N a leaves, with FR−/N− plants allocating the maximum N a to their bottom leaves, FR+/N+ allocating the maximum N a to their upper leaves, and FR−/N+ plants able to allocate the maximum N a to all of their leaves.The absolute centre of nitrogen (CN A ), relative centre of nitrogen (CN R ), and order-based centre of nitrogen (CN O ) were all significantly affected by the far-red treatment (Table

F
The final (a) number of leaves, (b) total leaf mass, (c) total leaf area, and (d) total leaf nitrogen for each treatment at the end of the experiment.Boxes indicate the interquartile range, bars indicate the rest of the distribution excluding outliers, and the significance post hoc comparisons are indicated with * (0.01 < p ≤ 0.05), ** (0.001 < p ≤ 0.01), *** (1 × 10 −4 < p ≤ 0.001), and **** (p < 1 × 10 −4 ). .Although investing in longer internodes may allow a plant to maintain access to high PAR, it also precludes investment in other areas.To mitigate the negative consequences of this shift in allocation, plants may need to make further adjustments to optimally respond to competition.In this experiment, we examined how the different components of the shade avoidance response interact with nitrogen limitation to study how the distribution of leaf nitrogen changes when plants are threatened by future light competition.

FR− /
N+ group had the highest nitrogen supply due to a high root mass and high nitrogen availability, and both the FR−/N− and FR+/ N+ groups had intermediate nitrogen because they had either high root mass or high nitrogen availability.The correspondence between the leaf mass, leaf area, and total leaf nitrogen measurements and the presumed nitrogen status of the plants suggests that leaf growth was limited by nitrogen, with the nitrogen treatment offsetting the effect of the far-red treatment for the FR−/N− and FR+/N+ treatments (Figure2b-d).Consequently, the shade avoidance response prioritized internode growth at the expense of root growth and the resulting nitrogen limitation then restricted the total leaf area that the plants could produce, providing an explanation for previous findings that low R:FR reduced leaf allocation and leaf chlorophyll(Cowan & Reekie, 2008).This suggests that plants responding to future light competition exacerbate their F I G U R E 3 The (a) average maximum leaf area of the bottom 10 leaves of each treatment group, (b) average leaf nitrogen per area of the bottom leaves at the end of the experiment, (c) the maximum total leaf nitrogen for the bottom leaves, and (d) the total leaf nitrogen for the bottom leaves at the end of the experiment.Boxes indicate the interquartile range, bars indicate the rest of the distribution excluding outliers, and the significance post hoc comparisons are indicated with * (0.01 < p ≤ 0.05), ** (0.001 < p ≤ 0.01), *** (1 × 10 −4 < p ≤ 0.001), and **** (p < 1 × 10 −4 ).creating an indirect tradeoff between internode extension and leaf growth.

|
Forgoing high investment into leaves in high PAR is counterintuitive because increased leaf expansion and leaf nitrogen would have increased the canopy-level photosynthesis.However, investing in a leaf with a low future PAR may have a high opportunity cost, as it may reduce the long-term productivity of the plant once it must contend with shade from neighbours.F I G U R E 4 The (a) total leaf area of the upper leaves (leaves of order ≥10), (b) average leaf nitrogen per area of the upper leaves, (c) total leaf nitrogen of the upper leaves, and (d) the maximum nitrogen per leaf area across treatments for the lower leaves (grey) and upper leaves (yellow).Boxes indicate the interquartile range, bars indicate the rest of the distribution excluding outliers, and the significance post hoc comparisons are indicated with * (0.01 < p ≤ 0.05), ** (0.001 < p ≤ 0.01), *** (1 × 10 −4 < p ≤ 0.001), and **** (p < 1 × 10 −4 ).In (d), bars sharing a letter are not significantly different in pairwise post hoc Tukey's tests, while all other combinations were significantly different (p < 0.05).The shade avoidance response shifts leaf nitrogen distributions higher in the canopy The reduced nitrogen investment into leaves with supplemental farred light allowed FR+ plants to instead focus their nitrogen into their upper leaves.Despite having the lowest total leaf nitrogen, FR+/N− plants were able to maintain similar upper canopy leaf area and leaf nitrogen as FR−/N− plants (Figure 4).This is because FR−/N− plants held more nitrogen in their lower leaves than FR+/N− plants, which limited the nitrogen available for the growth of their upper leaves relative to FR+/N− plants.Although FR−/N− and FR+/N+ plants had similar overall nitrogen uptake, they showed opposite investment strategies.FR−/N− plants fully invested into lower leaves, which eventually limited the leaf nitrogen per area and leaf area of upper leaves, while FR+/N+ plants avoided investment into lower leaves, allowing them to invest the maximum leaf nitrogen per area into upper leaves and grow a relatively large upper leaf area (Figure 4a,d).
into a height versus leaf growth tradeoff, whereby increased height growth raises leaves into environments that are more likely to retain a high PAR, but this comes at the expense of leaf area, which is needed to exploit that high PAR.However, this tradeoff is weakened because low R:FR not only increases local internode growth, it also decreases local nitrogen demand in leaves.By preemptively limiting nitrogen investment into leaves that develop near neighbours, plants can shift their leaf nitrogen towards younger leaves at the top of their stems, all while also physically raising their leaves above the threat of competitors.In all, plants preparing for future light competition also have to cope with present nitrogen limitation, but they are able to manage their limited resources by trading the growth of leaves that are likely to be shaded by competitors for the growth of leaves that are more likely to shade competitors.AUTH O R CO NTR I B UTI O N S I.K.U conceived the ideas and designed the methodology, I.K.U and T.B. collected the data, I.K.U analysed the data, I.K.U led the writing of the manuscript.All authors contributed to the drafts and gave final approval for publication.ACK N OWLED G EM ENTS We thank Heinz Vahlenkamp and Otmar Ficht for assistance in the botanical garden.This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy-EXC 2117-422037984 and the Human Frontier Science Program, RGY0078/2019.Open Access funding enabled and organized by Projekt DEAL.

F
I G U R E 5 A conceptual representation of the multiple effects of neighbour detection on plant growth.Plants grown under ambient conditions (a) have higher root growth, but lower stem growth than plants grown with low R:FR (b).The nitrogen limitation caused by the reduced root growth in plants experiencing low R:FR leads to less nitrogen available for leaf growth.However, leaves developing in low R:FR are smaller and are allocated less nitrogen, allowing plants in low R:FR to concentrate their leaf nitrogen into their upper canopy.This results in low R:FR plants having comparable upper canopy leaf areas and total leaf nitrogen as high R:FR plants, but with a significantly higher canopy height.