Determining the scale at which variation in WUE traits changes population yields

Functional variation is known to influence population yield, but the scale at which this happens is still unknown. Relevant signals might only reach immediate neighbors of a phenotypically diverse plant (neighbor-scale) or conversely may distribute across the population (population-scale). We use Nicotiana attenuata silenced in mitogen-activated protein kinase 4 (irMPK4), plants with low water-use efficiency (WUE), to study the scale at which water-use traits alter intraspecific population yields. In the field and glasshouse, populations with low percentages of irMPK4 plants planted among isogenic control plants produced maximum overall growth and yield. Through paired-plant and local-plant-configuration analyses, we determined that this occurred at the population scale. However, we find that this effect was not due to irMPK4’s WUE phenotype. With micro-grafting, we additionally show that MPK4-deficiency may mediate the response at the population-scale: shoot-expressed MPK4 is required for N. attenuata to change yield in response to a neighbor.


Abstract 23
Functional variation is known to influence population yield, but the scale at which this 24 happens is still unknown. Relevant signals might only reach immediate neighbors of a 25 phenotypically diverse plant (neighbor-scale) or conversely may distribute across the population 26 (population-scale). We use Nicotiana attenuata silenced in mitogen-activated protein kinase 4 27 (irMPK4), plants with low water-use efficiency (WUE), to study the scale at which water-use traits 28 alter intraspecific population yields. In the field and glasshouse, populations with low percentages 29 of irMPK4 plants planted among isogenic control plants produced maximum overall growth and 30 yield. Through paired-plant and local-plant-configuration analyses, we determined that this 31 occurred at the population scale. However, we find that this effect was not due to irMPK4's WUE 32 phenotype. With micro-grafting, we additionally show that MPK4-deficiency may mediate the 33 response at the population-scale: shoot-expressed MPK4 is required for N. attenuata to change yield 34 in response to a neighbor. 35 Introduction moisture at any sampling depth ( Figure S4A; R 2 = 0.527, F(15, 44) = 5.374, p = 6.097e-06). From 245 these results we conclude that increasing the percentage of MPxCC plants in populations under 246 field conditions leads to a non-additive trend in population yield, unrelated to soil nutrient and 247 moisture availability, with maximum population yields occurring in low-irMPK4 populations. neighbors to the measured plant: for example, EV plants in 50% and 83% irMPK4 populations had 267 four immediate irMPK4 neighbors, and yet their growth and yields were not equivalent.

12
At the population level, rosette diameter, stalk height, and water content PATs and PTs were 269 additive (grey bars and insets, Figure 4A, B, D). However, shoot biomass and fitness correlate 270 PATs and PTs were non-additive, with both measurements reaching maximum values in 271 populations with 17% irMPK4 (grey bars and insets, Figure 4C, E). From these results, we conclude 272 that low-irMPK4 populations produce maximum yield both in the glasshouse and field, and these 273 differences are independent of potential differences in population water availability resulting from 274 irMPK4's WUE phenotype. These results also appear independent from immediate neighbor 275 configurations, indicating that involved responses likely occur at the population rather than the 276 neighbor scale. 277

MPK4 is necessary for N. attenuata's growth and yield responses to neighbors 279
To test if the increased yields in low-irMPK4 glasshouse ( Figure 4E) and field ( Figure 3D this reduction was independent of the neighbor's genotype ( Figure 5B). Shoot biomass was only 288 significantly reduced in EV plants with an irMPK4 neighbor in comparison to individually-grown 289 EV plants (left panel, Figure 5D Figure 5E). irMPK4 plants showed no 292 13 differences in their rosette growth, water content, or yield when planted in pairs as compared to 293 being grown alone ( Figure 5B; Figure 5D-F). 294 Rosette diameters, shoot biomass, water contents, and fitness correlates of individuals in 295 each type of pair were summed for total outputs per pair type, but no differences were found among 296 pairs for any measurement ( Figure 5C; right panels, Figure 5D-F). From these results we conclude 297 that MPK4 is required for N. attenuata growth and yield responses to a neighbor. However, 298 neighbor-scale yield responses of EV and irMPK4 plants in pairs did not have consequences for the 299 total growth and yield of pairs. 300 301 EV and irMPK4 achieve peak WUE in low-irMPK4 populations in the glasshouse, but 302 photosynthetic parameters do not differ in the field 303 To determine if the WUE phenotypes of EV and irMPK4 plants in glasshouse or field 304 populations change with the percentage of MPK4-deficient plants, potentially causing neighbor-or 305 population-scale effects that increase yield (Figure 1), we measured leaf photosynthetic parameters 306 (assimilation rate, transpiration rate, stomatal conductance) and calculated the WUE of all 307 individuals in both glasshouse and field experiments. 308 In the glasshouse paired experiment, all measured leaf photosynthetic parameters of EV and 309 irMPK4 plants in single pots were as previously reported (Figure 2A), with irMPK4 plants having 310 significantly higher assimilation rates, transpiration rates, and stomatal conductance than EV plants, 311 and significantly lower WUE ( Figure 6A   To test the tissue-specific function of MPK4 expression in plant yield responses to a 356 neighbor, we created chimeric plants by micro-grafting irMPK4 roots to EV shoots (heterografts), 357 EV roots to EV shoots (EV homografts) and irMPK4 roots to irMPK4 shoots (irMPK4 homografts; 358 Figure 7A). We grew the grafts under conditions of equal water availability, with or without an 359 ungrafted EV neighbor. Photosynthetic parameter profiling of these grafts revealed that the 360 heterografts were similar to EV homografts in assimilation, transpiration, stomatal conductance, 361 and WUE within planting type ( Figure 7B), while the irMPK4 homografts showed significantly 362 higher transpiration rates and stomatal conductance and lower WUE ( Figure 7B, Table 1). Hetero-363 and homo-irMPK4 grafts retained similar levels of MPK4 silencing in roots or roots and shoots, 364 respectively ( Figure S1). 365 Table 1. Statistical emmeans contrasts within planting treatments for Figure 7B 366  18 attenuata population yields both in the glasshouse and the field ( Figures 3D, 4C, 4E). This was not 417 due to differences in soil water availability, which was controlled for in the glasshouse (Figure 4; 418 Figure S5B), nor irMPK4's WUE phenotype, which was not observed in the field ( Figure 6). 419 Interestingly, we find that the signal which was responsible for the yield-increasing responses in 420 low-irMPK4 populations likely occurred at the population scale. 421 Based on results from previous neighbor-effect studies using low-WUE phenotypes, we 422 initially hypothesized that the neighbor scale (Figure 1 However, we interpreted these responses to be independent of the responses causing the low-458 irMPK4 population yield increase, given the lack of the irMPK4 WUE phenotype in the field 459 ( Figure 6C). We concluded that the population-scale factor responsible for the increased low-460 irMPK4 population yield originates from an irMPK4 phenotype that is independent of its WUE 461 phenotype. 462 We considered other types of factors that could alter yields at the population scale.  counted simultaneously for all plants, immediately before harvesting for shoot and root biomass. 554 Due to APHIS regulations, ripening seed capsules were counted and subsequently removed to 555 prevent opening and releasing seeds into the field; the total ripe capsules collected is presented 556 ( Figure 3D). 557 For all glasshouse experiments, the planting substrate consisted of a bottom layer of large 558 clay aggregate (Lecaton, 816mm diameter, approx. 10% of pot volume), a central layer of small 559 clay aggregate (Lecaton, 24mm diameter, approx. 80% of pot volume) and a top layer of fine sand 560 (approx. 10% of pot volume). This substrate provides optimal drainage in the pots for the purposes 561 of water control, and conditions similar to the sandy, clay nature of the natural habitat of N. 562 attenuata. Rosette diameter was measured directly on the plant. Plant stalk height was measured as 563 in the field. Shoot biomass consisted of all above ground matter (severed below the rosette), placed 564 inside a bag for drying at 80°C for 2 h, after which the plant matter was removed from the bag and 565 weighed. The shoot biomass was also weighed for fresh mass, and the water content of the plant at 566 harvest was reported as the difference between the fresh and dry shoot biomasses. All fitness 567 correlates were counted at harvest, including buds (larger than 1mm), flowers (counted as flowers 568 when the corolla became visible by pushing through the sepals), unripe and ripe seed capsules, and 569 the total of all of these together was reported ( Figure 4E). 570 In the grafted experiment, predicted paired root biomass (PPRB) is presented in Figure 7D. 571 This was calculated as the sum of the means of the single graft root biomasses present in the pair 572 (e.g. EV/EV PPRB = 2*(EV/EV single root biomass mean); EV/irMPK4 PPRB = (EV/irMPK4 573 single root biomass mean) + (EV/EV single root biomass mean)). 574 575

Calculations of population additive totals and population trends 576
In Figures 3 and 4, data from individuals are represented by red and blue bars. Population 577 additive totals (PAT) are presented as gray bars and population trends (PT) as line graphs. 578 PATs for each population type were estimated by multiplying the mean of each individual 579 genotype in each type of population by the number of that genotype in that population type. Two 580 example calculations are shown in Figure 3B Soil cores were taken from the field by driving a split tube core borer (53 mm, Eijkelkamp, 590 Giesbeek, Netherlands) 30 cm into the ground, and carefully removing it with the core intact. 5 cm 591 pieces of field soil were cut from the core from 0 to 5, 10 to 15, and 25 to 30 cm below ground. 592 Each of these 5 cm thick pieces were weighed, left to dry in the sun in UV-excluding boxes similar 593 to those used for the drying of shoot biomass (see Plant growth and yield measurements), and 594 25 weighed again when dry (determined to be when the mass fluctuated <0.1 g between days). Soil 595 moisture was calculated from each sample (% soil moisture = (fresh soil mass -dry soil mass / fresh 596 soil mass) * 100), taken from 21 to 30 dpp ( Figure S4, n = 1 per population). 597 Soil cores were obtained using the same method at 54 dpp with replication (n = 2-9) to 598 determine the soil content of total, inorganic and organic carbon (Ctotal, Cinorg, Corg, respectively), 599 nitrogen (N), copper (Cu), iron (Fe), potassium (K), phosphorus (P), and zinc (Zn) in each type of 600 population at the end of the season. These samples were dried at 80°C for 6 h in a drying oven 601 ( Figure 3E; Figure  reported together as one mean (Figure 3). 620 In the glasshouse, all populations and pairs (grafted and ungrafted) underwent the following 621 regimented watering to control for water availability: after potting, they were given establishment 622 watering (soil moisture maintained around 20%) for three weeks, allowing root development to the 623 bottom of the pot for a transition from top watering to bottom watering. After three weeks, pots 624 began to show detectable differences in water loss by population type and consumption-based 625 watering began. For the population and homozygous pair experiment, pots were individually 626 watered daily to a two-day water supply, calculated as: 627 WM = 2*mean(WL-1, WL-2) + DP 628 WM = pot mass (g) to which the pot needed to be watered 629 WL-1 = water loss (g) from the previous to the current day 630 WL-2 = water loss (g) from two days to one day prior 631 DP = dry pot mass (g of pot with dry substrate, before planting) 632 The two-day water supply was chosen as it reflected ecologically-relevant soil moisture at the field 633 site in the natural habitat of N. attenuata (Valim et al., 2019). To allow larger growth and thus 634 accentuate growth differences in plants in the grafted pair experiment, the water supply was raised 635 to five days (WM = 5*mean(WL-1, WL-2) + DP), bringing soil moisture percentages up to 20-30%. 636 The higher soil moisture did not affect the differences in photosynthetic parameters of EV and 637 irMPK4 homozygous grafts compared to those reported for the homozygous EV and irMPK4 plants 638 in the paired experiment ( Figure 6A, 7B). There was no significant correlation between the amount 639 of water added in our watering regimes and the amount of water lost (demonstrated two times -640 Figure S7B, C -during watering regime of the grafted experiment, Figure S7A). 641

642
Gas exchange measurements and water-use efficiency calculations 643 Gas exchange measurements including photosynthesis and transpiration rates, and stomatal 644 conductance (via calculation), were obtained using a LI-COR 6400XT infrared gas analyzer 645 (Lincoln, Nebraska, USA), both in the field and the glasshouse between 12:00 and 14:00 ('PM'; 646 Figure 6). 647 The LI-6400XT was combined with a Leaf Chamber Fluorometer in the glasshouse to 648 obtain chlorophyll fluorescence measurements after 6 h of dark adaptation (lights off at 22:00, 649 measurements from 4:00 to 6:00; 'AM'; Figure 6B Water-use efficiency (WUE) was calculated as the ratio of photosynthetic rate (µmol 656 CO2/m2s) to transpiration rate (mmol H2O/m2s), thus resulting in units of carbon dioxide molecules 657 used per 1000 water molecules ( Figure 6A, B, C). 658 659

Micro-grafting 660
Seven-day-old seedlings were micro-grafted as described previously (Fragoso et al., 2011), 661 with EV scions grafted to both EV (EV/EV) and irMPK4 (EV/irMPK4) rootstocks, and irMPK4 662 grafted only to irMPK4 (irMPK4/irMPK4) rootstocks. The average grafting success was 90% (p > 663 0.05 between genotypes, ANOVA, Tukey HSD post hoc). 664 665 Transcript abundance 666 RNA was extracted with TRIzol reagent (Invitrogen) according to the manufacturer's 667 instructions. cDNA was synthesized from 500 ng of total RNA using RevertAid H Minus reverse 668 transcriptase (Fermentas) and oligo (dT) primer (Fermentas). qPCR was performed in a Mx3005P 669 28 PCR cycler (Stratagene) using SYBR GREEN1 kit (Eurogentec) using TaqMan primer pairs and 670 double fluorescent dye-labeled probe. An N. attenuata sulfite reductase (ECI) was used as a 671 standard housekeeping gene for normalization, and its primer sequences and probe, as well as the 672 MPK4 primer sequences and probes, are as published previously (Wu et al., 2007). Ct values were 673 converted to relative transcript abundances using the standard curve method ( Figure 2C, Figure S1). 674 Silencing efficiency was calculated using the ΔΔCt method.  Pseudoreplication was accounted for in our statistical analysis: these datasets were fit to 688 linear mixed effect (LME) models with the population or pair they originated from indicated as a 689 random effect. These models were checked for outliers, homoscedasticity and normality. Pairwise 690 post hoc comparisons were made using the R package emmeans (Lenth et al., 2019). 691 Datasets without pseudoreplication were fit to the best suited of either AOV, LM, GLS or 692 LME models, and were also checked for outliers, homoscedasticity and normality. Pairwise post 693 hoc comparisons were extracted as above, or else using Tukey HSD tests following significant main 694 effects in ANOVA.   EVxEV