Growth, photosynthesis, and nutrient uptake in wheat are affected by differences in nitrogen levels and forms and potassium supply

Nitrogen (N) and potassium (K) are essential macronutrients for plants growth; however, the mechanism by which K mediates negative effects on ammonium-sensitive plants is still poorly understood. We hypothesized that K supplies may enhance antagonistic ammonium stress while improving nitrate nutrition function, which wheat seedlings were grown in sand culture in the presence of two N forms (ammonium; nitrate) supplied at two rates (2, 10 mmol L−1) and three K levels (0.5, 5, 15 mmol L−1). We found that a high N rate increased plant biomass under nitrate nutrition, while it had a negative effect under ammonium nutrition. Compared with nitrate, biomass was depressed by 54% or 85% for low or high N rate under ammonium. This resulted in a reduction in gas exchange parameters and a subsequent decrease in growth variables and nutrient uptake, whereas these parameters increased significantly with increasing K levels. Moreover, in principal components analysis, these variations were highly clustered under nitrate nutrition and highly separated under ammonium nutrition. Our study shows a clear positive interaction between K and N, suggesting that high K supply relieves ammonium stress while improving growth vigor under nitrate nutrition by enhancing nutrient uptake and assimilate production in wheat plants.

Nitrogen (N) is a key element required for plant growth, and is one of the most important yield-limiting nutrients in crop production in all agro-ecological regions of the world. N is commonly taken up from the soil in one of two inorganic forms: ammonium (NH 4 + ) and nitrate (NO 3 − ) [1][2][3] . Different N forms can affect the physiological and metabolic processes of plants, such as nutrient uptake, enzyme activity, photosynthesis rate, respiration rate, water balance, and signaling pathways, thus eventually influencing plant growth and crop yield [3][4][5][6] . Although NH 4 + is an intermediate in many metabolic reactions, it can result in toxicity symptoms in many higher plants when supplied as the sole N source 1,[7][8][9] .
Sole NH 4 + supply provokes negative effects on NH 4 + -sensitive plants. The effects of NH 4 + and NO 3 − nutrition on plant growth have been intensively studied; however, the results are not consistent and depend mainly on plant species. Wheat and maize 10 , sugar beet 11 , beans [12][13][14][15] , tobacco 16,17 and canola 18 , grow preferentially on NO 3 − nutrition; whereas, rice [19][20][21] , pine and larch 22 , grow preferentially on NH 4 + nutrition. The mechanisms responsible for NH 4 + toxicity have been the subject of much speculation, and proposals include proton extrusion associated with NH 4 + uptake, cytosolic pH disturbances, decreased water use efficiency, shifts in plant carbohydrate status, the uncoupling of photophosphorylation, high energy costs of membrane transport, and displacement of crucial cations, such as K 4,7,9,[12][13][14] . NH 4 + nutrition has been found to cause strong inhibition of potassium (K) uptake by plants 4,16,17 . K, which is an essential nutrient involved in many important plant physiological processes, can improve crop yield and quality and enhance stress tolerance 23,24 . Thus, N and K requirements and management of these essential nutrients for crop production have become a focus of research into the interactions between N and K in terms of factors such as form and rate. Currently, the imbalanced fertilizer use is common in field production in many developing countries. The practice often leads to an excess of soil N combined with a serious and continual depletion of soil K, mainly due to the application of excessive N and inadequate K 25 . Over-application of N is a serious problem in intensive agricultural production areas because this leads to enrichment of reactive N constituents in the environmental, soil acidification and also affects the transformation of soil N forms, with consequent impairment of ecosystems [26][27][28] . It is generally accepted that, regardless of the application of N fertilizer or not, the progress of N form transformation is inhibited by the interaction between NH 4 + and NO 3 − , with higher NH 4 + and lower NO 3 − status maintained in acidic soil 29,30 and low oxygen or waterlogged conditions 31,32 .
Wheat is one of the most important cereal crops worldwide and grows preferentially under NO 3 − nutrition; however, its production is challenged by waterlogging of N forms, which is reported to cause yield losses between 15 and 20% 33 , and frequently occurs in regions with heavy rainfall and high ground water levels. Waterlogging causes significant reduction in gas diffusion and thereby, creates high NH 4 + conditions in soil 31,32 , leading to a substantial decrease in wheat growth and crop yield 33 . Most studies have shown that the application of NH 4 + as the sole N source decreased biomass in wheat compared with that achieved under NO 3 − nutrition 10,34,35 . In addition, changes in NH 4 + /NO 3 − ratios and K supply levels under soil culture conditions influence the yield and nutrient uptake of wheat plants 36 . Thus, we hypothesized that plant growth and crop yield are influenced not only by the amount of available N in the soil, but also by the N forms, and that the process can be regulated by K supply.
Although the individual effects of different N forms on plant growth have been widely studied, the combined effects of various levels of N forms and K supply on wheat growth and crop yield are largely unknown. Similarly, little is known about the effects of K supply relative to N forms on the photosynthetic process and nutrient uptake in wheat plants. In this study, we investigated the effects of different levels of N forms and K supply on growth in wheat plants, primarily by investigating biomass, growth, gas exchange, and N and K uptake.

Results
Wheat growth. Significant differences in plant biomass were observed between those supplied with NH 4 + and NO 3 − forms of N at two rates (Table 1). N supplied at the high rate in the form of NO 3 − significantly increased the biomass of different organs. In contrast, NH 4 + resulted in a reduction in plant biomass that was exacerbated by high NH 4 + . However, the plant biomass increased with K levels under both NH 4 + and NO 3 − nutrition. Compared with NO 3 − supply in the absence of K supply, the biomass of root, stem, leaf, and panicle in plants fed NH 4 + was reduced by 67%, 54%, 53% and 40% at the low N rate and by 87%, 88%, 88% and 77% at the high N rate, respectively. These results were also clearly supported by the images collected during the culture stage (Fig. 1). Regardless of the N form, the high N rate significantly decreased the root:shoot ratios. In contrast to the effects of NH 4 + , the root:shoot ratios of plants supplied NO 3 − increased with K levels, and the highest ratio was observed under low NO 3 − treatment. A clear positive interaction was observed among N forms and rates and K levels on the biomass of the different organs except the panicle; however, there were no significant differences in the root:shoot ratios among the different K levels.
In accordance with the effects on biomass, the other growth variables (root volume, number of tillers, flag leaf area, total leaf area and specific leaf weight) also changed with the N forms and rates in the presence of K, with the exception of the number of tillers at different K levels and specific leaf weight under treatment with the different N forms (Table 2). Overall, NO 3 − dramatically increased the growth variables, with greater increases under high NO 3 − treatment than those under low NO 3 − treatment. In contrast to the effects of NO 3 − alone, the high N rate had a negative effect on growth variables in plants fed NH 4 + , whereas no significant differences in root volume, number of tillers and specific leaf weight were observed between the N rates.
Gas exchange and relative chlorophyll content. There were significant differences in gas exchange parameters (P n , g s , C i , and T r ) and relative chlorophyll content (SPAD value) between plants under NH 4 + and NO 3 − nutrition at different N rates and K supply conditions ( Table 3). The P n , g s , and T r values increased significantly with the K levels, while C i decreased. However, under high NH 4 + nutrition, the high N rate restrained P n , g s , and T r by 6%, 12% and 7%, respectively, while position effects were observed under NO 3 − nutrition. Under NO 3 − nutrition, the SPAD of flag leaves was decreased by 11% under low N rate compared with that under the high N rate, while no significant differences in SPAD were observed between plants under NH 4 + nutrition and K levels.
N and K content. The different N forms and rates supplied with K had a significant effect on both N and K concentration and accumulation in wheat at the harvest period (Figs 2 and 3). Independent of K levels, the N concentration of different organs (including root, stem, leaf, and panicle) were regulated by both N form and rate ( Fig. 2a-  under the low N rate, while a negative effect of N rate on organ K concentrations was observed under NH 4 + nutrition. Regardless of N forms and rates, the organ K concentrations increased with K levels ( Fig. 2e-h).
The accumulation of N and K was similar to pattern of changes in the K concentration (Fig. 3). Overall, compared with the low N rate, the average N accumulation in root, stem, leaf and panicle was decreased by 35%, 48%, 23% and 38%, respectively, by the high N rate under NH 4 + nutrition and without K. The average K accumulation was also decreased by 10%, 19%, 35% and 36%, respectively. However, the accumulation of both N and K in organs was increased by the high N rate under NO 3 − nutrition. The effects of the variations in NH 4 + , NO 3 − and NH 4 + plus NO 3 − among the different N rate with K supply treatments were further evaluated by PCA (Fig. 4). The results showed that the growth and physiological parameters were significantly separated under the different N forms, with the first two principal components accounting for 81.92% (69.02% for PC1 and 12.90% for PC2), 92.15% (81.59% for PC1 and 10.56% for PC2) and 88.05% (68.65% for PC1 and 19.40% for PC2) of the total variations under NH 4 + , NO 3 − and NH 4 + plus NO 3 − , respectively. According to the PCA, the total variations were highly clustered under NO 3 − nutrition. In contrast, the variations were less clustered under NH 4 + nutrition, especially the N content and gas exchange parameters. K + uptake rate. As shown in Fig. 5, there was a significant difference in K + uptake under the two N forms.
Compared with NH 4 + nutrition, the K + uptake of wheat seedlings was increased by different K + concentration under NO 3 − nutrition, while the K + uptake rate under NO 3 − nutrition (4.4494) was increased by 36% compared with that under NH 4 + nutrition (3.2818) based on regression analysis.

Discussion
Unlike plants such as rice 19,21 and pine 22 , wheat is a typical low NH 4 + tolerance plant, with toxicity symptoms manifested as a decline in growth and yield 1,2,7 . Generally speaking, wheat is not always confronted with a soil environment of high NH 4 + , except under special circumstances, such as heavy rainfall or irrigation and poor soil drainage combined with excess N fertilization 31,32 , which is especially common in wheat planting regions in the middle and lower of the Yangtze River Plain in southern China 37 . N forms and rates affecting growth and photosynthetic CO 2 fixation in wheat. Plant growth is clearly affected by the forms of N supplied as nutrient. In the present study, we showed that the biomass of both non-reproductive and reproductive organs was depressed by NH 4 + supply, and resulted in lower root:shoot ratios, indicating the negative effects of pure NH 4 + nutrition on wheat growth (Table 1). In accordance with this, Huang et al. 35 reported that root and shoot biomass of wheat plants was lower when cultured in solution with NH 4 + as a N source than that observed in the presence of NO 3 − . Furthermore, Wang et al. 38 demonstrated that wheat yield was higher when using NO 3 − fertilizer than that achieved using NH 4 + fertilizer under field conditions. Some studies showed that NH 4 + nutrition resulted in differences in shoot:root ratios compared with those obtained under NO 3 − nutrition in wheat 10 , bean 12,14,15 and canola 18 . These results were further confirmed in other studies of rice varieties 5,19,20,39,40 , which were found to be tolerant to pure NH 4 + nutrition. In contrast to these findings, Walch-Liu et al. 41 reported that the root:shoot ratios was higher under NH 4 + -fed than that under NO 3 − -fed in two tobacco cultivars. Moreover, the effect of N forms on root:shoot ratios is mediated by regulation of the N rate. Lu et al. 17 identified a more marked positive relationship between root:shoot ratios under low (2 mmol L −1 ) or high (15 mmol L −1 ) NH 4 + supply compared with those obtained under NO 3 − supply in tobacco plants. Zhou et al. 42 also observed that higher root:shoot ratios were obtained with higher NH 4 + supply (5 mmol L −1 ) than that using NO 3 − as a N supply, although lower N supply (1 mmol L −1 ) had no effect on the ratios in cucumber plants. Similar results were obtained in the present study, with consistently lower root:shoot ratios using NH 4 + than those with NO 3 − , while high N reduced the ratio regardless of the N form (Table 1). However, dynamic changes in root:shoot ratios at various sampling points have also been observed in wheat 35 and sugar beet 11 . These results indicate that the effects of N forms on root and shoot growth in different species and at different N rates and sample points are mediated primarily by altering the partitioning of fixed carbon (C) during photosynthesis in both root and shoot.
These results was consistent with those reported by Guo et al. 3 and Gao et al. 18 , who suggested that the conflicting effects of N forms on plant growth are related to the gas exchange parameter P n , which is associated with g s and also C i . The lower biomass of different organs in wheat plants under NH 4 + nutrition compared to those under NO 3 − nutrition was consistent with the observed differences in the gas exchange parameters (Table 3). Thus, it can be concluded that the reduced carbon assimilation in NH 4 + -fed wheat plants may have been important in contributing to reduced growth. The higher P n and higher root:shoot ratios of the wheat plants under NO 3 − nutrition ensure greater availability and allocation of carbon to the root than to the shoot, thus improving plant growth. In addition, the inhibition of wheat growth in response to NH 4 + application could be attributed mainly to a reduction in growth variables including root volume, number of tillers, and leaf area (flag leaf and total leaf), with the response intensified under high N conditions (Table 2). Similar results have been reported for sugar beet 11 and tobacco 41 , in which the negative effect of NH 4 + nutrition on leaf growth was explained by either increased chlorophyll content and chloroplast number and volume, reduced cell number and cell expansion, or by osmotic regulation 43 and water uptake in bean plants [12][13][14] . However, differences in biomass production, gas exchange, root and leaf morphological and physiology observed under conditions of different N forms 5,6,21,40,41,44 , indicate that the N forms available affect plant growth and photosynthesis. N forms and rates affecting nutrient uptake in wheat. N form and rate affect plant growth by regulating photosynthetic carbon fixation and distribution as well as plant nutrient uptake. Plant N concentrations in different organs (root, stem, leaves and panicle) were increased in response to NH 4 + application with two N rates (Fig. 2). Brück and Guo 15 reported that N concentrations in young, fully developed leaves under NH 4 + -fed were significantly increased by 59% compared to leaves under NO 3 − -fed in beans, and reduced in wheat, maize 10 and rice 19 Table 3. Effects of different levels of N forms and K supply on leaf photosynthetic rate (P n ), stomatal conductance (g s ), intercellular CO 2 concentration (C i ), transpiration rate (T r ), and relative chlorophyll content (SPAD value, n = 6) in wheat plants. Data presented as mean ± standard deviation of three replicates (Mean ± SD, n = 3). Different lowercase letters in the same column indicate significant differences among treatments at P < 0.05 level. was not limited by N availability. In PCA results, the growth and physiological parameters were significantly separated by the different N forms, with higher levels of variation clusters under NO 3 − nutrition than those under NH 4 + nutrition, especially in terms of N content and gas exchange parameters (Fig. 4). These findings further indicated the influence of N uptake and assimilation on the responses of plants to N forms and rates in C fixation or biomass production. In addition, Guo et al. 13,14 found that N uptake under NH 4 + -fed was significantly higher than that under NO 3 − -fed during the dark period, whereas N uptake under NH 4 + -fed was significantly lower compared to that under NO 3 − -fed during the light period with a split root system in bean plants. Thus, we suggest ). Data represent the mean ± standard deviation of three replicates. Significant differences (P < 0.05) were determined by different lowercase letters according to a one-way or multi-way ANOVA followed by Duncan's multiple range test. NS, no significant difference. that plant N uptake may be influenced not only by N forms, but also by the environmental condition, especially changes in light/dark conditions. Furthermore, as an important osmoticum, K + was significantly decreased in wheat plants under NH 4 + nutrition (Fig. 2). Similar results have been reported for tobacco 17 , in which K + uptake was inhibited under NH 4 + nutrition, while K + transport in the xylem and K + translocation in the phloem was still higher than that in plants under NO 3 − nutrition, although the process was improved more under high nutrient levels (6 mmol L −1 K and 15 mmol L −1 N) than under low levels (2.5 mmol L −1 K and 2 mmol L −1 N). Walch-Liu et al. 41 found that the leaves ). Data represent the mean ± standard deviation of three replicates. Significant differences (P < 0.05) were determined by different lowercase letters according to a one-way or multi-way ANOVA followed by Duncan's multiple range test. NS, no significant difference. of young, expanding and old tobacco plants under NH 4 + -fed showed 20%, 22% and 60% decreases in K concentrations, respectively, which indicated that K may mediate the effect of N forms on leaf morphogenesis. Most studies have revealed a negative effect of NH 4 + nutrition on the osmotic regulation of leaves due to reduced K absorption 41,45,46 . These results suggest that this effect might be responsible for smaller leaf area and lower specific leaf weight as well as the reduction in root cell length and root morphological parameters frequently observed in rice plants under NH 4 + nutrition 6,32,39 . Also, the K uptake by NH 4 + -fed vessels was significantly lower than that by NO 3 -fed vessels, while the K uptake rate was reduced by 466% 13 and 1231% 14 , respectively, by the application of a split root system in bean plants. Similarly, in the present study, NH 4 + -fed resulted in a higher K + uptake rate compared to that of NO 3 − -fed plants (Fig. 5). Thus, we suggest that plant K uptake may be influenced not only by the amount of available N in the environment, but also by the N forms.
Interaction between N and K supply in wheat. The interaction between K and N, especially K and N forms (both NH 4 + and NO 3 − ) on plant growth and development, has become a focus of research 4,16,17,23,47,48 . Our studies in wheat showed that N forms affect plant growth and the uptake of N and K nutrients; however, the supply level of K also has a significant influence on the regulation of plant growth, photosynthesis and nutrient absorption, with a positive interaction identified between N and K. This is consistent with earlier results of studies in wheat 49 , rice 4 , tobacco 16,17 and beans 13,14 , in which NH 4 + nutrition not only strong inhibited K uptake, but also had a marked influence on the flow and partitioning within plants, resulting in reducted water uptake and T r , apparently due to reduced g s . This is consistent with the proposed theory that cycling of K in plants can act as an important signal for feedback control of nutrient uptake 50 . However, in this study, the growth variables, gas exchange parameters, and nutrient content of wheat plants were increased with increasing K supply, and the beneficial effects were observed under both NH 4 + and NO 3 − nutrition. These results further implied that optimal K management may alleviate NH 4 + stress or toxicity and improve the nutritional function of NO 3 − in wheat plants. K is an essential macro-element of nutrition in plants and its uptake is strongly influenced by other elements, such as N. Elevating the K supply results in a significant reduction of NH 4 + influx in rice plants 4 . Similarly, we observed that the growth of wheat plants under pure NH 4 + nutrition was improved by K supply (Fig. 1), a phenomenon that demonstrates the influence of K supply on the tolerance of wheat to NH 4 + stress or toxicity 49 . Based on the proposals of Szczerba et al. 4 and Kong et al. 49 , we suggested that adequate K application is essential for plant growth, especially under conditions of high environmental NH 4 + stress. The results of the present study also demonstrated that N and K interactions not only affect plant growth and nutrient uptake, but also significantly  affect crop yield parameters, such as panicle biomass (Table 1). These results provide evidence that assimilate partitioning in plants can be improved by changes in K supply. Some studies suggest that the controversial effects of N forms and K supply on plant growth are related to the availability of photo-assimilates for production, transportation and distribution, further indicating that K plays an important role in yield formation 23,47,48 .
Although the plants did not show any visible symptoms of mineral nutrient deficiency under different N and K treatments, there were marked morphological differences between wheat plants, especially the shoots, for which the erecting and tilting degree were improved under high N and low K conditions under NO 3 − nutrition compared with NH 4 + nutrition (Fig. 1). Previous studies demonstrated that higher N rates reduced plant morphology traits, culm physical strength and lodging resistance in wheat 51 and rice 52 , with every 2% increase in lodging resistance causing a 1% decrease in grain yield 53 . However, Zaman et al. 54 found that optimizing K fertilizer improved stem strength and yield to alleviate the negative effects of higher N application. Kong et al. 49 also reported that the additional provision of K + considerably alleviated the negative effects of high NH 4 + , resulting in a 23% increase in culm mechanical strength and a 35% increase in the N remobilization efficiency in wheat plants. It has also been reported that the consequences of N metabolism 55 , N use efficiency 48 and C-N balance 56 are improved by K fertilizer application. Similarly, the results of the present study suggest that a positive synergistic interaction between K and N on assimilate production, nutrient uptake, yield formation and stress tolerance 47,57 . Thus, an understanding of the roles of the interaction between N (especially in NH 4 + ) and K in the regulation of physiological and biochemical mechanical and yield formation is required to improve sustainable productivity in wheat plants.
In conclusion, wheat grown in anaerobic soils showed growth inhibition when the predominant form of nitrogen is NH 4 + or a mixture of NH 4 + and NO 3 − , and under conditions of limited K availability. NO 3 − treatment increased root growth, N uptake and photosynthetic productivity in wheat plants in comparison with NH 4 + treated plants, and the effects were enhanced by K supply, especially under high N conditions. Thus, our results indicate that increased K supply mediates nutrient balance between N and K uptake, and enhances NH 4 + tolerance in wheat plants supplied with pure NH 4 + nutrition. Our results provide the basis for the development of new nitrogen fertilizer utilization schemes for wetland wheat production.

Materials and Methods
Plant materials and growing conditions. A split-unit randomized complete block design, with N forms as the main unit and factorial combinations of N rates and K levels as subunits, was used in this study. The treatments were: two N forms (ammonium, as NH 4 + [AN]; nitrate, as NO 3 − [NN]), two N rates (2 and 10 mmol L −1 ), and three K levels (0.5, 5 and 15 mmol L −1 [K0.5, K5 and K15]) replicated three times. The 'Yangmai 16' hard red winter wheat (Triticum aestivum L.) was used as a model. Briefly, after germination on moist filter paper, wheat seeds were disinfected with 10% H 2 O 2 for 30 min then transferred to a 2 mmol L −1 CaSO 4 solution for germination at 25 ± 5 °C. When the seedlings had an average of 2.5 visible leaves (0.5 g fresh weight, 17 cm plant height), they were transplanted to 0.75 L (top diameter 9.5 cm × bottom diameter 6.5 cm × 16 cm high) plastic pots (two seedlings per pot) containing clean quartz sand and transferred into quarter-strength Hoagland's nutrient solution (for composition, see below) with 1 cm depth in tray. Seedlings were grown in a greenhouse under a natural photoperiod. Four days later, the seedlings were transferred to half-strength nutrient solution. After an additional four days, the seedlings were treated with full-strength nutrient solution containing 12 different treatments (AN2K0.5, AN2K5, AN2K15, AN10K0.5, AN10K5, AN10K0.5, NN2K0.5, NN2K5, NN2K15, NN10K0.5, NN10K5, and NN10K15). The composition of the other nutrients in the solutions was as follows: macronutrients (mmol L −1 ): 2 or 10 N as (NH 4 ) 2 SO 4 or Ca(NO 3 ) 2 , 0.5, 5 or 15 K as K 2 SO 4 and KH 2 PO 4 , 1 P as KH 2  A nitrification inhibitor (dicyandiamide, DCD) was added to prevent oxidation of NH 4 + at a dose of 5% of total nitrogen. Nutrient solutions were changed every 4 d, after drip washing the sand surface with 0.5 L tap water, and the pH was monitored daily and maintained at 6.00 ± 0.05 by adding either 0.1 mmol L −1 NaOH or HCl until the final harvest. Each treatment group consisted of six plants in a completely randomized design to minimize edge effects.
K + uptake rate measurement. The examine the effect of N forms on K + uptake rate, K + influx of intact wheat plants was determined by the depletion of nutrient solution directly using a K + concentration gradient method with N forms supplied under hydroponic solution conditions. Wheat seedlings grown in 2 mmol L −1 CaSO 4 solution were pre-equilibrated for 2 h, then immersed in labelling solution between 8:30 to 16:30 (8 h). The solution was identical to the growth solution, except that it contained different K + concentrations (0, 0.05, 0.1, 0.2, 0.4, 0.5, 1, 2, 4, 5, 10 mmol L −1 ) with the addition of 2 mmol L −1 AN or NN. Each treatment group consisted of three plants and was replicated three times in a completely randomized design. The seedlings were grown under greenhouse conditions (air temperature 30 °C; relative humidity 50%) under a photosynthetic photon flux density (PPFD) of 1000 μmol photons m −2 s −1 at the leaf level.
Gas exchange measurements. At the heading stage, 60 days after treatments initiation, the light-saturated photosynthetic rates of newly expanded leaves (flag leaf) were measured simultaneously between 09:00 and 15:00 with an infrared gas analyzer (6400XT, Li-Cor, Lincoln, NE, USA). Leaf temperature during the measurements was maintained at 28 °C and a relative humidity of 50% under a PPFD of 1000 μmol photons m −2 s −1 . Data were recorded after equilibration to a steady state. Relative chlorophyll content measurement. The one-dimensional (1 D) chlorophyll index of the labeled leaf segments was determined using a portable relative chlorophyll meter (SPAD-502, Minolta Camera, Osaka, Japan) to measure absorbance at 650 nm. The chlorophyll index was determined as the mean of six SPAD-502 readings from the same leaf at the heading stage.
Root volume, dry weight and specific leaf weight measurements. Wheat plants were harvested and separated into root, stem (including sheath and culm), leaf and panicle sections. Sand was washed from the roots before the root volume was measured using the displacement method described by Sattelmacher et al. 58 . The dry weight was measured after all samples were oven-dried at 105 °C for 30 min and then at 70 °C to constant weight. The leaf area was determined using a photocopy of the leaf and calculated according to the paper area. The specific leaf weight was then calculated as the ratio of leaf weight to leaf area.

N and K concentration and accumulation measurements.
To determine the total concentrations of N and K in different organs, dried and ground samples were digested with H 2 SO 4 -H 2 O 2 at 260-270 °C. N concentrations were measured using an Auto-analyzer 3 digital colorimeter (AA3, Bran + Luebbe, Hamburg, Germany) and K concentrations were measured by Flame Photometry (FP6400, Shanghai Precision Scientific Instrument, Shanghai, China). The measurements were validated using certified standard reference materials obtained from the Institute for Environmental Reference Materials of the Ministry of Environmental Protection (Beijing, China). The total accumulation of N and K was calculated from the sum of organ N and K content (element concentration × dry weight) at harvest.
Statistical analysis. Samples were analyzed in triplicate and mean values were used in comparisons analysis. Variance analysis (ANOVA) was performed using the SAS 9.3 statistical software package (SAS Institute, Cary, NC, USA). Means were compared among treatments by the least significant difference (LSD) test with P < 0.05 considered to indicate statistical significance. Principal components analysis (PCA) was used to analyze the growth and physiological variations of NH 4 + , NO 3 − and NH 4 + plus NO 3 − under treatment with different N rates and K supplies using SPSS Statistics 17.0 (IBM, Armonk, NY, USA) and plotted using SigmaPlot 12.5 (Systat Software Inc., San Jose, CA, USA).